[go: up one dir, main page]

WO2025042943A2 - Ultrastructural membrane expansion microscopy - Google Patents

Ultrastructural membrane expansion microscopy Download PDF

Info

Publication number
WO2025042943A2
WO2025042943A2 PCT/US2024/043153 US2024043153W WO2025042943A2 WO 2025042943 A2 WO2025042943 A2 WO 2025042943A2 US 2024043153 W US2024043153 W US 2024043153W WO 2025042943 A2 WO2025042943 A2 WO 2025042943A2
Authority
WO
WIPO (PCT)
Prior art keywords
umexm
lysine
tissue
membrane
biological sample
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/US2024/043153
Other languages
French (fr)
Other versions
WO2025042943A3 (en
Inventor
Tay SHIN
Edward Stuart Boyden
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Massachusetts Institute of Technology
Original Assignee
Massachusetts Institute of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Massachusetts Institute of Technology filed Critical Massachusetts Institute of Technology
Publication of WO2025042943A2 publication Critical patent/WO2025042943A2/en
Publication of WO2025042943A3 publication Critical patent/WO2025042943A3/en
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/58—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving labelled substances
    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K7/00—Peptides having 5 to 20 amino acids in a fully defined sequence; Derivatives thereof
    • C07K7/04—Linear peptides containing only normal peptide links
    • C07K7/06—Linear peptides containing only normal peptide links having 5 to 11 amino acids
    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K7/00—Peptides having 5 to 20 amino acids in a fully defined sequence; Derivatives thereof
    • C07K7/04—Linear peptides containing only normal peptide links
    • C07K7/08—Linear peptides containing only normal peptide links having 12 to 20 amino acids
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N1/00—Sampling; Preparing specimens for investigation
    • G01N1/28—Preparing specimens for investigation including physical details of (bio-)chemical methods covered elsewhere, e.g. G01N33/50, C12Q
    • G01N1/36—Embedding or analogous mounting of samples
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N1/00—Sampling; Preparing specimens for investigation
    • G01N1/28—Preparing specimens for investigation including physical details of (bio-)chemical methods covered elsewhere, e.g. G01N33/50, C12Q
    • G01N1/40—Concentrating samples
    • G01N1/4044—Concentrating samples by chemical techniques; Digestion; Chemical decomposition
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N1/00—Sampling; Preparing specimens for investigation
    • G01N1/28—Preparing specimens for investigation including physical details of (bio-)chemical methods covered elsewhere, e.g. G01N33/50, C12Q
    • G01N1/44—Sample treatment involving radiation, e.g. heat
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2474/00—Immunochemical assays or immunoassays characterised by detection mode or means of detection
    • G01N2474/20—Immunohistochemistry assay

Definitions

  • the invention relates, in part, methods and compounds for ultrastructure membrane expansion microscopy (umExM), which permits high-resolution visualization of membrane ultrastructure using light microscopy.
  • umExM ultrastructure membrane expansion microscopy
  • BACKGROUND OF THE INVENTION Neuroscientists have long studied complex neural circuits by examining their molecular features and the ultrastructure context of the neural circuits. Fluorescence microscopy (FM), including super-resolution microscopy, allowed neuroscientists to identify specific biomolecules, while electron microscopy (EM) allowed neuroscientists to reveal ultrastructural context through dense labeling of membranous structures. Both FM and EM greatly advanced fundamental understandings of neurobiology.
  • FM lacks the ability to provide the biomolecule localization in the ultrastructural context
  • EM cannot identify specific biomolecules in the ultrastructural context.
  • CLEM correlative EM
  • the anchored biomolecules and/or their labels are pulled apart from each other isotropically, typically to a physical magnification of ⁇ 4-10x in each linear dimension.
  • ExM ExM
  • the expanded sample can be expanded a second time, resulting in an overall physical magnification of beyond 10x in each linear dimension.
  • a probe molecule including: (a) a chain including 2, 3, 4, 5, 6, 7, 8,9, 10, 11, 12, 13, 14, 15 or more lysines and/or cystines; (b) a chemical handle; and (c) a polymer-anchorable handle.
  • the probe molecule includes (a) a chain including 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more lysines; (b) a chemical handle; and (c) a polymer-anchorable handle.
  • one or more of the lysines are D-lysines.
  • one or more of the cystines are D-cystines.
  • the chain includes acrylic acid.
  • the chemical handle includes azide.
  • the chemical handle includes biotin.
  • the polymer-anchorable handle is attached to a primary amine of the chain.
  • the polymer-anchorable handle includes acryloyl–X (AcX).
  • the chain includes a lipid tail on the amine terminus of the chain. In some embodiments, the lipid tail includes palmitoyl.
  • the probe molecule includes: palmitoyl-glycine-acrylic-lysine-lysine- lysine-lysine-lysines-lysine-lysine-lysine-lysine-lysine-azide (SEQ ID NO: 1).
  • a glycine linker molecule is positioned between the lipid tail and the amine terminus of the lysine chain.
  • the lysine chain includes 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, or more lysines.
  • the lysine chain includes five lysines.
  • the probe molecule includes a glycine and penta-lysine D-peptidic chain, with a palmitoyl lipid group on the amine- terminus and a biotin on the carboxy-terminus, wherein the probe molecule includes: palmitoyl-glycine-lysine-lysine-lysine-lysine-lysine-biotin (SEQ ID NO: 2).
  • the probe molecule includes pGk5b, pGk5a, pGk13a, or pGk13b.
  • the probe molecule is one of SEQ ID NOs: 1-16.
  • a method for preparing a biological sample for detecting ultrastructural features of a membrane in the biological sample including: (a) contacting the biological sample with a probe of any embodiment the aforementioned probes.; (b) embedding the contacted biological sample within a polymer material; (c) homogenizing the embedded biological sample; and (d) physically expanding the homogenized embedded sample.
  • 3 #17358430v1 embedding the biological sample in the polymer material includes incubating the biological sample in a polymer monomer and polymerizing the monomer.
  • the polymer material includes a swellable polymer material.
  • the swellable polymer material includes an acrylamide-co-acrylate copolymer.
  • the method also includes polymerizing the polymer material.
  • the polymerization includes polymerization at approximately 40C for a time period of at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more hours.
  • the polymer material includes a non-swellable polymer material capable of conversion to a swellable polymer material, and the method further includes converting the polymerized non-swellable polymer material into a swellable polymer material.
  • a means for the physical expansion of the biological sample includes contacting the homogenized embedded biological sample with a solvent or liquid that swells the swellable polymer.
  • the liquid includes water.
  • homogenizing the embedded biological sample includes contacting the polymer material in which the biological sample is embedded with one or more enzymes.
  • the contacting enzyme is an endoproteinase, optionally wherein the endoproteinase is LysC or Trypsin.
  • the homogenization includes contacting the embedded biological sample with one or both of LysC and Trypsin.
  • the physically expanding of the biological sample includes expanding the polymer material in which the biological sample is embedded, wherein the expansion of the polymer material expands the homogenized biological sample isotropically in at least a linear manner within the polymer material.
  • the polymer material includes a hydrogel and a means of expanding the hydrogel includes contacting the hydrogel with an aqueous solution, optionally wherein the aqueous solution includes water.
  • the method also includes attaching a detectable label to the probe molecule at one or both of before and after the physical expansion step.
  • the method also includes re-embedding the physically expanded polymer and homogenized biological sample in either a swellable or a non-swellable polymer.
  • the method also includes expanding the embedded sample 1, 2, 3, 4, 5, or more times. In certain embodiments, the method also includes detecting one or more of a spatial position, a structure, a component of, and an identity of the expanded biological sample. In certain embodiments, the method also includes imaging the biological sample after the physical expansion. In some embodiments, the imaging includes optical microscopy. In some embodiments, the microscopy is light microscopy, fluorescence microscopy or electron 4 #17358430v1 microscopy. In some embodiments, the imagining includes optical fluctuation imaging. In certain embodiments, the method also includes detecting one or more proteins and/or polynucleotide molecules in the embedded, expanded sample.
  • the polynucleotide molecule comprises an RNA molecule.
  • the method also includes antibody staining the expanded sample.
  • biological sample includes a cell or tissue.
  • the cell or tissue is a fixed cell or tissue.
  • the cell or tissue is obtained from a subject.
  • the subject is a mammal, optionally is a human or a rodent.
  • the cell or tissue is a cultured cell or cultured tissue.
  • the biological sample includes a cell membrane.
  • the tissue is a tissue slice.
  • the tissue includes CNS tissue.
  • the tissue is brain tissue.
  • the tissue slice is between about 50 and 100 microns thick. In some embodiments, the tissue slice is less than 50 microns thick or is more than 100 microns thick. In certain embodiments, the method also includes segmenting cell compartments of the expanded sample, optionally wherein the compartments are cell bodies, dendrites and/or axons in the expanded sample. In certain embodiments, the method also includes tracing cell processes in the expanded sample, optionally wherein the cell processes are myelinated axons and/or unmyelinated axons in the expanded sample. According to yet another aspect of the invention, a method is provided, the method including use of a probe of any embodiment of an aforementioned aspect of the invention in an imaging method.
  • the imaging method includes imaging an ExM- processed biological sample.
  • the imaging method comprises light microscopy.
  • the imaging method comprises confocal microscopy.
  • FIG 1A-B provides schematic diagrams illustrating ultrastructural membrane expansion microscopy (umExM) concept and workflow.
  • umExM is a modified form of expansion microscopy with a custom-designed amphiphilic membrane labeling probe (termed pGk13a).
  • Fig.1A shows chemical structure of pGk13a. The probe does not contain any fluorophore but has an azide to bind a fluorophore later.
  • Fig.1B is a schematic diagram of umExM workflow.
  • Fig.1B panel (ii) the specimen is treated with pGk13a (structure is depicted in (Fig.1A) at 4 0C overnight (unless otherwise noted, overnight means >16 hours).
  • Fig.1b panel (iii) the specimen is treated with acrylic acid N-hydroxysuccinimide ester (AX) at 4 0C overnight.
  • AX acrylic acid N-hydroxysuccinimide ester
  • Fig.1 B panel (iv) the specimen is embedded in an expandable hydrogel (made with N,N'-Diallyl-L-tartardiamide (DATD) crosslinker [Yu, C. C. et al. Elife 9, 1–78 (2020)]) at 4 0C for at least 24 hours.
  • DATD N,N'-Diallyl-L-tartardiamide
  • Fig.1B panel (v) the sample (specimen- embedded hydrogel) is chemically softened with enzymatic cleavage of proteins (i.e., non-specific cleavage with proteinase K) at room temperature ( ⁇ 24 0C) overnight.
  • the probe is not digested during proteinase K treatment because it is composed of D-amino acids.
  • Fig. 1B panel (vi) the sample is treated with 1x phosphate-buffered saline (PBS) to partially expand it.
  • PBS 1x phosphate-buffered saline
  • the probe, pGk13a that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight.
  • FIG.1B panel (vii) the sample is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes).
  • Figure 2A-V provides photomicrographic images and graphs illustrating embodiments of resolution and distortion of umExM.
  • Fig.2B provides a single z-plane confocal image of the same HEK293 cell as in Fig. 2A, after undergoing the umExM protocol, showing expression of mitochondrial matrix- targeted GFP in the same field of view as shown in Fig.2A.
  • SIM z-plane structured illumination microscopy
  • FIG.2C provides a single z-plane confocal image of the same umExM-expanded fixed HEK293 cell as in Fig. 2A, showing pGk13a staining of the membrane in the same field of view as shown in Fig. 2A. pGk13a staining shown.
  • Fig.2E as in Fig.2D, but with post-expansion images showing pGk13a staining of the membrane.
  • Fig.2I shows magnified view of black boxed region in Fig.2H. Fig.2J, as in Fig.2H, but imaging of the third ventricle.
  • Fig.2K as in Fig.2H but imaging of mouse somatosensory cortex layer 6 (L6).
  • Fig.2L shows magnified view of black boxed region in Fig.2K.
  • Fig.2O as in Fig.2M, but imaging of somatosensory cortex L6 that was used for measuring the diameter of myelinated axons.
  • Fig.2Q as in Fig.2M but imaging of the third ventricle that was used for measuring the diameter of cilia.
  • Fig.2S shows magnified view of boxed region in left panel. Arrows indicate putative extracellular vesicles. Serial image sections that were used for the 3D rendering are in Fig.23.
  • Fig.2T shows single z-plane confocal image of expanded mouse brain tissue (third ventricle) processed by umExM, showing pGk5b staining (gray), focusing on the plasma membrane of cilia (i.e., ciliary membrane).
  • Fig.2U shows transverse profile of cilia in the dotted boxed region in Fig.2U after averaging down the long axis of the box and then normalizing to the peak of pGk13a signal.
  • Scale bars are provided in 7 #17358430v1 biological units throughout all figures (i.e., physical size divided by expansion factor): Fig. 2A, B, & D 5 ⁇ m; Fig.2H 5 ⁇ m; Fig.2I 2 ⁇ m; Fig.2J 5 ⁇ m; Fig.2K 5 ⁇ m; Fig.2L 5 ⁇ m; fig. 2M 0.25 ⁇ m; Fig.2O & Q 1 ⁇ m; Fig.2S left panel (x); 13.57 ⁇ m (y); and 7.5 ⁇ m (z); Fig.2S right panel 3.76 ⁇ m (x); 3.76 ⁇ m (y); 1.5 ⁇ m (z; Fig.2T 2 ⁇ m.
  • Figure 3A-O shows photomicrographic images of umExM with antibody staining and RNA fluorescence in situ hybridization (FISH).
  • Fig.3B provides magnified view of the box in Fig.3A.
  • Fig.3C shows single z-plane confocal image of the specimen of Fig.3A, showing pGk13a staining of the same field of view as in Fig.3A.
  • FIG.3D shows magnified view of the box in Fig.3C.
  • Fig.3E shows an overlay of Fig.3A and Fig.3C.
  • Fig.3F shows a magnified view of the box in Fig. 3E.
  • Fig.3H shows magnified view of the box in Fig.3G.
  • Fig.3I shows single z-plane confocal image of the specimen of Fig.3G, showing pGk13a staining of the same field of view as in Fig.3G.
  • Fig.3J shows magnified view of the box in Fig.3I.
  • Fig.3K shows overlay of Fig.3G and Fig.3I.
  • Fig.3L shows magnified view of the box in Fig.3C.
  • the examples of PSD95 signals that were aligned with pGk13a signals were pinpointed with arrows.
  • Fig.3N shows single z-plane confocal image of the specimen of Fig.3J, showing pGk13a staining of the same field of view as in Fig.3J.
  • Fig.3O shows an overlay of Fig.3M and Fig.3N. Scale bars: (Fig.3A, B, C, G, H, & J) 5 ⁇ m; Fig.
  • Figure 4A-L provides photomicrographic images and graphs showing results of certain embodiments of methods of the invention.
  • Fig.4D shows single z-plane confocal image of the same U2OS cell as in Fig.4C, after undergoing the mExM protocol, showing expression of mitochondrial matrix-targeted GFP in the same field of view as shown in Fig.
  • Fig.4E shows non-rigidly registered and overlaid pre-expansion SIM image of the U2OS cell expressing mitochondrial matrix-targeted GFP in Fig.4C, and post-expansion confocal image of the same fixed U2OS cell after mExM processing, showing the mitochondrial matrix-targeted GFP channel in Fig.4D.
  • Fig.4G shows single z-plane confocal image of the same mExM-expanded fixed U2OS cell as in Fig.4D, showing pGk5b staining of the membrane in the same field of view as shown in Fig.4C.
  • Fig.4H shows non-rigidly registered and overlaid pre-expansion SIM image of the U2OS cell expressing mitochondrial matrix-targeted GFP in Fig.4C and post-expansion confocal image of the same U2OS cell in Fig.4E after mExM processing, showing pGk5b staining.
  • Fig.4J-K shows results of expansion factor analysis on HEK293 cells, which underwent mExM, after expressing mitochondrial matrix-targeted GFP. Two landmark points in pre-expansion images were randomly selected and the corresponding landmarks were found in expanded sample images.
  • Fig.4J shows representative (out of 10 cells from two cultures) single z-plane confocal image of pre-expanded HEK293 cell.
  • Fig.4K as in Fig.4J, but post-expansion, for the same field of view shown in Fig.4J.
  • Figure 5A-C provides photomicrographic images and a graph.
  • Fig.5A panel (ii) shows single z-plane confocal image of the same expanded HEK293 cell as in Fig.
  • FIG. 5 panel (i) showing pGk5b staining in the same field of view as in Fig.5 panel (i).
  • Fig.5 panel (iii) provides an overlay of Fig.5 panel (i) and panel (ii).
  • Figure 6 provides photomicrographic images showing effect of the glycine linker attached to the palmitoyl group on the efficacy of membrane labeling in fixed brain tissue. Image visualized in inverted gray color. Two versions of the palmitoylated 5 lysine biotin membrane probe were tested.
  • Fig.6 Left panel shows results with a probe that contained a glycine linker attached to the palmitoyl group enabling flexibility of the lipid relative to the peptide carrier.
  • Fig.6 Right panel shows results with a probe that did not contain a glycine but in which the lipid was directly attached to the lysine backbone.
  • Fig.7 panel (iii) as in Fig.7 panel (i) but focused on one cell body.
  • Fig.7 panel (iv) as in Fig.7 panel (ii) but focused on one cell body.
  • the pre-expansion images of mouse brain tissue in Fig.7 contain native lipids, which were not removed before imaging. Accordingly, the pre- expansion sample exhibits substantial light scattering and a mismatch in refractive index, which significantly impacts image contrast20.
  • Fig.8B shows average z-projection of images in Fig.8A.
  • Fig.8C shows fluctuations in the acquired frames (as in Fig.8A) were resolved with the ‘super-resolution imaging based on autocorrelation with a two-step deconvolution’ (SACD) algorithm [Zhao, W. et al. Nat. Photonics 1–8 (2023)].
  • Scale bars Fig. 8B-C 10 ⁇ m; Fig.8E-F 1.5 ⁇ m.
  • Figure 9 shows photomicrographic images of five examples of single z-plane confocal images of post-expansion mouse brain tissue (somatosensory cortex, L4) after the iterative form of umExM processing, showing pGk13a staining of the mitochondrial cristae. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): 0.2 ⁇ m.
  • Figure 10 provides a workflow schematic diagram of umExM with pre-expansion antibody staining. Lighter captions highlight the key differences from ExM [M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022)] and proExM [Klimas, A. et al.
  • Fig.10 panel (i) a specimen is chemically fixed with 4% PFA + 0.5% CaCl 2 at 4 0C for 24 hours. The brain is sliced on a vibratome at 100 ⁇ m thickness at 4 0C. In Fig.10 panel (ii) the specimen is treated with either 0.005-0.01% detergent (i.e., saponin or triton) at 4 0C overnight (unless otherwise noted, overnight means >16 hours throughout the figure). Then specimen is incubated with a primary antibody. In Fig.10 panel (iii) the specimen is treated with pGk13a at 4 0C overnight.
  • detergent i.e., saponin or triton
  • Fig.10 panel (iv) the specimen is treated with AX at 4 0C overnight.
  • the specimen is embedded in an expandable hydrogel (made with DATD crosslinker [Boutin, J. A. Cell. Signal.9, 15–35 (1997)]) at 4 0C overnight.
  • the sample (specimen-embedded hydrogel) is chemically softened with enzymatic cleavage of proteins (i.e., non-specific cleavage with proteinase K) at room temperature ( ⁇ 24 0C), overnight.
  • the sample is then treated with PBS to partially expand it.
  • the pGk13a that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight.
  • click-chemistry i.e., DBCO-fluorophore
  • the sample is then incubated with a secondary antibody at 4 0C for 2-3 days.
  • Fig.10 panel (ix) the sample is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes).
  • Figure 11 provides a workflow schematic diagram of umExM with post-expansion antibody staining. Light-colored captions highlight the key differences from ExM [M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022)] and proExM [Klimas, A.
  • Fig.11 panel (i) a specimen is chemically fixed with 4% PFA + 0.5% CaCl 2 at 4 0C for 24 hours. The brain is sliced on a vibratome at 100 ⁇ m thickness at 4 0C. Fig.11 panel (ii) the specimen is treated with pGk13a at 4 0C overnight (unless otherwise noted, overnight means >16 hours throughout the figure). In Fig.11 panel (iii) the specimen 12 #17358430v1 is treated with AX at 4 0C overnight.
  • Fig.11 panel (iv) the specimen is embedded in an expandable hydrogel (made with DATD crosslinker [Boutin, J. A. Cell. Signal.9, 15–35 (1997)]) at 4 0C overnight.
  • the sample (specimen-embedded hydrogel) is mechanically softened with enzymatic cleavage of proteins (i.e., specific cleavage with Trypsin and LysC) at room temperature ( ⁇ 24 0C), overnight.
  • the sample is then treated with PBS to partially expand it.
  • the pGk13a that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight.
  • click-chemistry i.e., DBCO-fluorophore
  • the sample is then incubated with a primary antibody at ⁇ 4 0C, for 48-72 hours.
  • Fig.11 panel (vii) the sample is then incubated with a secondary antibody at ⁇ 4 0C, for 48-48 hours.
  • Figure 12 provides a workflow schematic diagram of umExM with FISH staining. Light- colored captions highlight the key differences from ExM [M’Saad, O.
  • Fig.12 panel (ii) the specimen is treated with the pGk13a at 4 0C overnight (unless otherwise noted, overnight means >16 hours throughout the figure).
  • the specimen is treated with acrylic acid N-hydroxysuccinimide ester (AX) at 4 0C overnight.
  • Fig.12 panel (iv) the specimen is embedded in an expandable hydrogel (made with DATD crosslinker [Wen, G. et al. ACS Nano 14, 7860–7867 (2020)]) at 4 0C for at least 24 hours.
  • Fig.12 panel (v) the specimen is mechanically softened with enzymatic cleavage of proteins (i.e., specific cleavage with Proteinase k) at room temperature ( ⁇ 24 0C), overnight.
  • Fig.12 panel (vi) the specimen-embedded hydrogel is treated with PBS to partially expand it.
  • the pGk13a, that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight.
  • Fig. 12 (vii) the specimen-embedded hydrogel is incubated with HCR-FISH probe at 37 0C, overnight.
  • FIG.12 panel (vii) the specimen-embedded hydrogel is incubated with fluorescently labeled HCR-hairpin amplifiers at ⁇ 24 0C, overnight.
  • Fig.12 panel (viii) the specimen-embedded hydrogel is expanded with 0.05x SSCT at room temperature for 1.5 hours (exchanging water every 30 minutes).
  • 13 #17358430v1 Figure 13A-C provides photomicrographic images and a graph illustrating the performance of a biotin handle (pGk13b) vs. azide handle (pGk13a) with fixed mouse brain tissues in the context of ExM. Mouse brain tissues were fixed with ice-cold 4% PFA.
  • pGk13b or pGk13a was applied overnight, and the standard ExM protocol [Yu, C. C. et al. Elife 9, 1–78 (2020)] was then followed. In short, the tissues were processed with AcX, and the ExM gel was formed.
  • the probe was fluorescently labeled with fluorescent streptavidin (i.e., cy3-streptavidin, >1 fluorophore per streptavidin), expanded, and imaged (results show in Fig.13A) or with fluorescent DBCO (i.e., cy3-DBCO, 1 fluorophore per DBCO), expanded and imaged (results shown in Fig.13B).
  • Fig.13B is as in Fig.13A but with pGk13a, showing the pGk13a staining of the membrane. Images in Fig.13A-B were visualized with the same brightness and contrast with ImageJ software to highlight the difference between the two images.
  • the specimen is treated with acrylic acid N- hydroxysuccinimide ester (AX) at 4 0C overnight.
  • the specimen is embedded in an expandable hydrogel (made with cleavable crosslinker DATD) at 4 0C for at 14 #17358430v1 least 24 hours.
  • the specimen is mechanically softened with enzymatic cleavage of proteins (i.e., specific cleavage with proteinase-k) at room temperature ( ⁇ 24 0C), overnight.
  • the specimen-embedded hydrogel is treated with PBS to partially expand it.
  • the sample is treated with AX as was done in Fig.14 panel (iii). Subsequently, the sample was gelled again but with a monomer solution that contains the non-cleavable crosslinker N,N -Methylenebis(acrylamide) (BIS) at room temperature ( ⁇ 24 0C), overnight.
  • the specimen-embedded hydrogel is incubated in a gel cleaving solution (containing sodium metaperiodate) at room temperature ( ⁇ 24 0C) for 1 hour. This step cleaves the initial gel that was formed in Fig.14 panel (iv).
  • the pGk13a that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight.
  • the specimen-embedded hydrogel is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes).
  • the modified protocol requires additional time and resources (i.e., necessitating two gelations) compared to the unmodified protocol, which is suitable for all these regions except the corpus callosum. Hence, this modified protocol is suggested specifically for the corpus callosum region.
  • Fig.16B shows magnified view of boxed region in Fig.16A. Only a subset of axons can be identified in the images.
  • Fig.16D shows magnified view of boxed region in Fig. 16C. The modification of the protocol drastically improved the visualization of axons in the 15 #17358430v1 corpus callosum.
  • Figure 17A-B provides electron microscopy images of membrane label (pGk5a)-stained mouse brain slices (hippocampus region).
  • pGk5a membrane label
  • Mouse brain tissue was preserved in 4% PFA and 0.1% glutaraldehyde at 4 0C, and labeled with pGk5a for >16 hours at 4 0C.
  • the tissue was post-fixed in 2% PFA and 2% glutaraldehyde and labeled with 1.8nm undecagold gold nanoparticles, conjugated to dibenzocyclooctyne to attach to the azide handle on pGk5a.
  • the tissue was counter-labeled with uranyl acetate, embedded in resin, sliced, and imaged on a TEM scope.
  • Fig.18 panel (ii) shows single z-plane confocal image of the specimen of Fig.18 panel (i), showing pGk5b staining of the same field of view as in Fig.18 panel (i).
  • Fig.18 panel (iii) shows an overlay of Fig 18 panel (i) and panel (ii).
  • Fig.18 panel (iv) shows magnified views of white boxed regions in Fig.18 panel (iii) showing TOM20 signal.
  • Fig.18 panel (vi) as in Fig.18 panel (iv) but showing pGk5b signal.
  • Fig.18 panel (vii) shows overlay of Fig.18 panel (iv) and panel (v).
  • Fig.18B as in Fig.18A, but for the nuclear pore protein NUP98.
  • Fig.18C as in Fig.18A, but for myelin basic protein (MBP), using antibodies from three separate vendors (see Table 4 for details).
  • Scale bars Fig.18 panels (i) – (iii) 10 ⁇ m, Fig.18 panel (iv) –(vi) 1 ⁇ m.
  • Figure 19A-D provides photomicrographic images of results from tests of two versions of the membrane labeling probe with ExM:
  • Fig.19A shows results using palmitoyl-glycine-(D- lysine) 5 -biotin (pGk5b)
  • Fig.19B shows result using farnesyl-glycine-(D-lysine) 5 -biotin (i.e., replacing palmitoyl in pGk5b with farnesyl, fGk5b), as well as (Fig.19C-D) a mixture of pGk5b+fGk5b at varying concentrations .
  • Images Fig.19A-D are visualized with the same brightness and contrast with ImageJ software.
  • Figure 20A-G provides photomicrographic images of results of testing varying numbers of lysines (i.e., 3, 5, 7, 9, 11, 13, and 19 lysines) in the backbone of the membrane labeling probe while holding other moieties (palmitoyl tail, glycine, and biotin) constant.
  • Fig.20B-G as in Fig.20A but with membrane labeling probes containing 5, 7, 9, 11, 13, and 19 lysines, respectively.
  • Scale bars are provided in biological units (i.e., physical size divided by expansion factor):
  • Figure 21A-F provides photomicrographic images of certain experimental results.
  • Fig.21B shows single z-plane confocal image of the specimen of Fig.21A showing anti-GFP staining of the same field of view as in Fig.21A.
  • Fig.21C is an overlay of Fig.21A and Fig.21B.
  • Fig.21D shows magnified views of boxed region in Fig.21A.
  • Fig.21E shows magnified views of boxed region in Fig.21B.
  • Fig.21F shows magnified views of boxed region in Fig.21C.
  • Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig. 21A-C 5 ⁇ m, Fig.21D-F 0.5 ⁇ m.
  • Figure 22A-C provides photomicrographic images and a graph of results of pGk13a staining of the membrane of mouse brain tissue fixed with 4%PFA and 0.5% CaCl2 at 4 0C and proceeded with the standard ExM protocol (37 0C gelation) [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)] vs. modified ExM protocol (i.e., 4 0C gelation).
  • Mouse brain tissues were fixed with ice-cold 4% PFA and 0.5% CaCl 2 fixatives.
  • the pGk13a probe was applied overnight at 4 0C.
  • the standard ExM protocol or modified ExM protocol i.e., 4 0C gelation instead of 37 0C gelation
  • the tissues were processed with AcX, and ExM gel was formed at 37 0C (see Fig.22A) or 4 0C (see Fig.22B).
  • the pGk13a was fluorescently labeled with fluorescent DBCO (i.e., cy3-DBCO, 1 fluorophore per DBCO), expanded, and imaged.
  • fluorescent DBCO i.e., cy3-DBCO, 1 fluorophore per DBCO
  • Fig.22B as in Fig.22A but with 4 0C gelation. Images Fig.22A-B were visualized with the same brightness and contrast with ImageJ software to highlight the difference between the two images.
  • Black points individual measured average intensity of each image; median, middle line; 1 st quartile, lower box boundary; 3 rd quartile, upper box boundary; error bars are the 95% confidence interval; p-value, unpaired two-sided t-test between signals from the 37 0C gelation (left bar), and 4 0C gelation (right bar). Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig.22A-B 20 ⁇ m.
  • Fig.23B an in Fig.23A but with pGk13a staining of the membrane.
  • Images Fig.23A-B were taken under identical optical conditions and visualized with the same brightness and contrast with ImageJ software. Scale bars are provided in biological units: Fig.23A-B 40 ⁇ m.
  • Figure 24 shows images of fifteen serial sections from the 3D volume rendering in Fig.2S, right.
  • the arrows indicate membrane vesicles.
  • Scale bar in biological units i.e., physical size divided by expansion factor: 1 ⁇ m.
  • Scale bar in biological units i.e., physical size divided by expansion factor
  • Figure 26A-H provides photomicrographic images and graphs of results of certain studies.
  • Fig.26A shows photograph of a fixed 100 ⁇ m thick adult mouse coronal slice that underwent the umExM protocol.
  • Fig.26B shows single z-plane confocal image of green boxed region in Fig.26A. Images were taken with a 4x objective at 30ms laser exposure time, and they were stitched with shading correction function via default setting from Nikon Element software version 4.30. pGk13a staining of the membrane visualized in inverted gray color throughout this figure (dark signals on light background). No image processing (e.g., denoising or deconvolution) other than stitching was performed for images presented throughout this figure.
  • image processing e.g., denoising or deconvolution
  • Fig.26C is a volume rendering of the box (i) in Fig.26B. Images were taken with a 4x objective at 50ms laser exposure time with a z step size of 0.375 ⁇ m (in biological unit). Unless otherwise noted, clipping planes that are red colored indicate the portion that has been clipped out to expose the inside of the volume for 3D images presented throughout this figure.
  • Fig.26D shows profile of mean pGk13a signal intensity of XY planes taken along the depth of the volume in Fig.26C.
  • Fig.26E is a volume rendering of the white box (ii) in Fig. 26B. Images were taken with a 60x objective at 100ms laser exposure time with a z step size 19 #17358430v1 of 0.125 ⁇ m.
  • Fig.26F shows magnified view of green boxed region in Fig.26E.
  • Fig.26G shows profile of mean pGk13a signal intensity of XY planes taken along the depth of the volume in Fig.26E.
  • Fig.26H shows cross-sectional images of dentate gyrus region in 100- ⁇ m thick mouse coronal slices that underwent the umExM protocol, showing pGk13a staining of the membrane. Images were taken with a 60x objective at 100ms laser exposure time with a z step size of 0.075 ⁇ m. Lines indicate the cross-sectional views in y-z and x-z images.
  • FIG.26B 500 ⁇ m; Fig.26C 340 ⁇ m (x); 340 ⁇ m (y); and 100 ⁇ m (z); Fig.26E 62 ⁇ m (x); 62 ⁇ m (y); and 20 ⁇ m (z); Fig.26H 5 ⁇ m (x-y); 1 ⁇ m (y-z); 1 ⁇ m (x-z).
  • Fig.27B shows magnified view of boxed region in Fig.27A.
  • the image is visualized in inverted gray color.
  • Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig.27A 10 ⁇ m; Fig.27B 2 ⁇ m.
  • Fig.28B as in Fig.28A but with the BODIPY staining protocol established in ref [Schmidt, R. et al. Nano Lett.9, 2508–2510 (2009)] and followed by proExM [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)], showing BODIPY staining of the membrane.
  • Fig.28C as in Fig.28A but with the mCling staining protocol established in ref [Linder, M. E. & Deschenes, R. J. Nat. Rev. Mol. Cell Biol.8, 74–84 (2007)] and followed by proExM, showing mCling staining of the membrane.
  • Fig.28D as in Fig.28A but with the umExM protocol, showing pGk13a staining of the membrane.
  • Fig.28F as in Fig.28C, but with imaging of the 3rd ventricle. Image visualized in inverted gray color (dark signals on light background).
  • Fig.28G shows magnified view of boxed region in Fig.28F.
  • FIG.28A-D 100 ⁇ m
  • Fig.28F 10 ⁇ m
  • Figure 29A-E provides photomicrographic images and a graph illustrating segmentation ability of umExM.
  • Fig.29A panel (i) shows single z-plane confocal image of expanded Thy1-YFP mouse brain tissue after umExM processing, showing pGk13a staining of the membrane.
  • Fig.29A, panel (ii) shows single z-plane image showing manual segmentation of the cell body in Fig.29A, panel (i).
  • Fig.29A, panel (iii) shows overlay of Fig.29A, panel (i) and panel (ii).
  • Fig.29A panel (iv) shows single z-plane confocal image of the specimen of Fig.29A, panel (i) showing GFP signal of the same field of view as in Fig.29A, panel (i).
  • Fig.29A panel (v shows single z-plane image showing manual segmentation of the cell body from Fig.29A, panel (iv).
  • Fig.29A panel (vi) shows an overlay of Fig.29A, panel (iv) and panel (v).
  • Fig.29B as in Fig.29B, but for segmenting dendrites.
  • Fig.29C left panel shows single z-plane confocal image of expanded Thy1-YFP mouse brain tissue showing pGk13a staining of the membrane.
  • Fig.29C panel (i) shows magnified view of the box on the left.
  • Fig.29C panel (ii) shows single z-plane image showing manual segmentation of the myelinated axon in Fig.29C panel (i).
  • Fig.29C panel (iii) shows overlay of Fig.29C panel (i) and panel (ii).
  • Fig.29C panel (iv shows single z-plane confocal image of the specimen of Fig.29C panel (i), showing GFP signal of the same field of view as in Fig.29C panel (i).
  • Fig. 29C panel (v) shows single z-plane image showing manual segmentation of the myelinated axon in Fig.29C panel (iv).
  • Fig.29C panel (vi) shows overlay of Fig.29C panel (iv) and panel (v).
  • Fig.29D as in Fig.29C, but for segmenting unmyelinated axons.
  • Fig.29E shows Rand score of pGk13a signal-guided segmentation of cell body, dendrites, myelinated axon and unmyelinated axons, using anti-GFP signal-guided segmentation as a “ground truth.”
  • FIG.29A shows results demonstrating traceability of umExM.
  • Fig.30A shows serial confocal images of expanded Thy1-YFP mouse brain tissue after umExM processing, showing pGk13a staining of the membrane.
  • GFP column anti-GFP 21 #17358430v1 signal of the same sample in the same field of view.
  • Fig.30B left shows pGk13a signal- guided manually traced and reconstructed myelinated axon from (Fig.30A, pGk13a column).
  • Fig.30B right, as in Fig.30B left, but with anti-GFP signals.
  • Fig.30D as in Fig.30A, but with an unmyelinated axon.
  • Fig.30E shows in Fig.30B but for Fig.30D.
  • Fig.30H shows magnified view of the box in Fig.30G.
  • Fig.30I shows 3D rendering of 20 manually traced and reconstructed myelinated axons in the corpus callosum. Planes were visualized from raw umExM images that were used for tracing. Scale bars: Fig.30A 0.5 ⁇ m; Fig.30D 0.2 ⁇ m; Fig.30G 18 ⁇ m; Fig.30I 39.25 ⁇ m (x); 39.25 ⁇ m (y); and 20 ⁇ m (z).
  • Figure 31 provides a workflow schematic diagram of an iterative form of umExM Lighter- colored captions highlight the key differences from iExM [Nishino, M. et al.
  • Fig.31 panel (i) a specimen is chemically fixed with 4% paraformaldehyde (PFA) + 0.5% calcium chloride (CaCl2) at 4 0C for 24 hours. The brain is sliced on a vibratome at 100 ⁇ m thickness at 4 0C.
  • the specimen is treated with pGk13a at 4 0C overnight (unless otherwise noted, overnight means >16 hours throughout the figure).
  • the specimen is treated with acrylic acid N-hydroxysuccinimide ester (AX) at 4 0C overnight.
  • Fig.31 panel (iv) the specimen is embedded in an expandable hydrogel (made with cleavable crosslinker DATD [Boutin, J. A. Cell. Signal.9, 15–35 (1997)] at 4 0C for at least 24 hours.
  • the specimen-embedded hydrogel is mechanically softened with enzymatic cleavage of proteins (i.e., specific cleavage with proteinase-k) at room temperature ( ⁇ 24 0C), overnight.
  • the specimen-embedded hydrogel is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes).
  • the specimen-embedded hydrogel is re-embedded into a non-expandable hydrogel at 50 0C, for >4 hours.
  • the sample is treated with AX as done in Fig.31 panel (iii).
  • the sample was gelled again but with a monomer solution that contains the non-cleavable crosslinker (made with BIS) at room temperature ( ⁇ 24 0C), 22 #17358430v1 overnight.
  • the specimen-embedded hydrogel is incubated in a gel cleaving solution (contains sodium metaperiodate) at room temperature ( ⁇ 24 0C) for 1 hour.
  • This step cleaves the initial gel and re-embedding gel that was formed in Fig.31 panel (iv) and panel (vi).
  • Fig.31 panel (ix) the pGk13a, that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight.
  • the specimen-embedded hydrogel is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes).
  • DETAILED DESCRIPTION The invention, in part, provides methods and probes for visualizing membrane ultrastructure through dense labeling of membranous structures and visualization using ordinary microscopy with expansion microscopy (ExM).
  • umExM methods of the invention result in ⁇ 60nm resolution through the low-distortion ( ⁇ 3%) physical 4x expansion of a thick (100 ⁇ m) mice brain tissue. It has now been shown that in thick brain tissue sections, methods of the invention preserve ultrastructural context. Studies were performed to quantitatively examine the anatomical properties of ultrastructure features 23 #17358430v1 (i.e., diameter of unmyelinated and myelinated axons, and cilia) that were previously measured via EM. Methods and compositions of the invention can be used for light microscope imaging of ultrastructure feature (i.e., membrane vesicle around the cilia), features previously only visualized using electron microscopy (EM).
  • ultrastructure feature i.e., membrane vesicle around the cilia
  • Methods and probes of the invention provide a high signal-to-background (S/B) and uniform labeling of membrane and permit simultaneous visualization of biomolecules of interest in the ultrastructural context.
  • umExM methods and probes can achieve higher resolution (for example, ⁇ 30nm) when combined with existing fluorescence fluctuation imaging methods (i.e., SRRF).
  • umExM probes and methods of the invention have broad utility in visualization of ultrastructural context along with the key biomolecules and are useful to describe and characterize ultrastructural features that were previously only possible to visualize with EM.
  • methods and compositions of the invention can be used to visualize and reconstruct dense neuronal circuitries with biomolecule of interest to better understand the brain and other tissues.
  • Embodiments of methods and probes of the invention can be used to visualize the ultrastructure of membranes.
  • Methods of the invention comprise labeling membranous structures and visualizing the labeled structures with standard fluorescence microscopy (FM) techniques.
  • Methods and probes of the invention may be used to obtain images of resolution similar to low-resolution EM images.
  • Use of embodiments of methods and compositions of the invention permit visualization of anatomical properties and characteristics of ultrastructure features previously identified with EM including features that with prior methods were only visible using EM methods.
  • methods and compositions of the invention may be used to visualize features that could not be fully characterized using FM methods.
  • Embodiments of methods and probes of the invention also permit visualization of ultrastructural features across a large volume of tissue with a high signal-to-background (S/B) ratio.
  • S/B signal-to-background
  • the utility of umExM methods and probes of the invention has now been demonstrated for segmentation and tracing of neuronal processes, such as axons, in tissues such as, but not limited to, mouse brain tissue.
  • Additional aspects of the invention comprises combining umExM methods and probes with optical fluctuation imaging such as, but not limited to SACD.
  • umExM methods of the invention may also include iterative expansion methods, which can be used to increase resolution.
  • umExM methods of the invention comprising iterative expansion yielded ⁇ 35 nm resolution imaging, 24 #17358430v1 indicating use of such methods and probes to obtain electron microscopy resolution visualization of membranes on ordinary light microscopes.
  • the invention comprises rationally and systematically designed novel membrane-labeling probes that may be used in conjunction with novel ExM protocols. Probes and methods of the invention can be used to achieve dense labeling of membranes, including plasma membranes, mitochondrial membranes, nuclear membranes, ciliary membranes, myelin sheaths, and extracellular vesicle membranes in tissues.
  • ultrastructural membrane expansion microscopy (umExM), using the word ultrastructure in 27 #17358430v1 the same sense as an earlier protocol in the expansion microscopy community, called ultrastructure expansion microscopy [Gambarotto, D. et al. Nat. Methods 16, 71–74 (2019)].
  • umExM preserves ultrastructure and enables the visualization of membranous structures in tissue slices.
  • umExM methods of the invention have been used with 100 ⁇ m-thick slices of fixed mouse brain at a resolution of ⁇ 60 nm with excellent uniformity and continuity of membrane labeling as well as a high signal-to-background ratio (40-80 fold higher than background).
  • umExM methods and probes of the invention can be used to co-visualize membranous structures along with proteins and RNAs.
  • the dense membrane labeling of umExM methods and probes of the invention can be used for the segmentation of neuronal compartments (e.g., cell bodies, dendrites, and axons), and tracing of neuronal processes (e.g., axons).
  • umExM methods of the invention achieve dense labeling of membranes and high- integrity expansion and enable imaging of membranous structures using a standard confocal microscope. In some embodiments, methods of the invention achieve ⁇ 60nm lateral resolution and enable co-visualization of membranous structures in a wide range of brain regions along with proteins and RNAs. umExM methods of the invention enable segmentation of cell bodies, dendrites, and axons (>200nm in diameter) and can be used to trace axons. Probes The invention, in part, provides probes that can be used with ExM chemistry. In some embodiments, a probe of the invention is used for nanoscale resolution imaging and permits continuous tracing of membranous structures with ExM chemistry.
  • a probe of the invention is an unnatural synthetic amphiphilic membrane probe comprising characteristics that improve imaging capability of traditional ExM methods.
  • a probe of the invention may comprise one or more of the following characteristics: (1) lipophilicity, which in some embodiments is similar to art-known traditional fluorescent lipophilic dyes such as DiI and/or other dyes that preferentially localize to and diffuse within membranes permitting membrane labeling; (2) a chemical handle for chemoselective conjugation of a fluorophore following the formation of an expandable hydrogel network, so the initial label remains as small as possible to facilitate diffusion and to avoid potential fluorophore degradation during the free-radical polymerization, and (3) a polymer-anchorable handle for binding the probe to an ExM-gel matrix, to allow expansion.
  • a probe of the invention comprises a chain of lysines.
  • Lysines contain primary amines that serve as sites for binding to a polymer-anchorable handle.
  • a non-limiting example of a polymer-anchorable handle is acryloyl–X (AcX), which has been used to attach proteins via their amines to an in situ synthesized ExM hydrogel.
  • lysine chain of a probe of the invention comprises D-lysines. The presence of D-lysines minimized degradation during the mechanical softening step of ExM, which in some instances includes a proteinase K softening step.
  • a probe of the invention comprises one or more D-cystines.
  • a probe of the invention is an acrylic acid conjugated probe.
  • a non-limiting example of an acrylic acid conjugated probe that may be used in certain embodiments of the invention is: palmitoyl-glycine-acrylic-lysine-lysine-lysine-lysine-lysine- lysines-lysine-lysine-lysine-lysine-lysine-lysine-azide (palmitoyl-glycine-acrylic- KKKKKKKKKKK-azide (SEQ ID NO: 1), which is also referred to herein as probe pGk13A.
  • one or more lysines included in an acrylic acid conjugated probe of the invention is a D-lysine.
  • a probe of the invention comprises a lipid tail on the amine terminus of the chain.
  • one or more glycine linkers are positioned between the lipid tail and the amine terminus of the lysine chain.
  • a glycine linker comprises 1, 2, or 3 glycines. Inclusion of a glycine linker assists in membrane labeling and provides mechanical flexibility.
  • a probe of the invention also comprises a chemical “handle” attached to the carboxy terminus of the terminal lysine.
  • palmitoyl and biotin are used for lipid tails and chemical handles respectively and to attach five (5) lysines in the backbone.
  • a probe prepared as described in the above non-limiting example comprised a glycine and penta-lysine D-peptidic backbone, with a palmitoyl lipid group on the amine-terminus and a biotin on the carboxy-terminus.
  • the probe is referred to herein as “pGk5b” and comprises the structure: palmitoyl-glycine-lysine-lysine-lysine-lysine-lysine- biotin (palmitoyl-glycine-KKKKK-biotin (SEQ ID NO: 2).
  • Another probe of the invention is “pGk5a” which comprises the same structure as pGk5b except the biotin is replaced with an azide.
  • pGk5a probe has the structure: palmitoyl-glycine-KKKKK-biotin (SEQ ID NO: 3).
  • Another probe of the invention is “pGk13b” which has the same structure as pGk13b except the azide is replaced with biotin.
  • the pGk13b probe has the structure: (palmitoyl-glycine- acrylic-KKKKKKKKKKKKK-biotin (SEQ ID NO: 4).
  • Non-limiting examples of other probes that may be used in methods of the invention are (palmitoyl-glycine-acrylic- 29 #17358430v1 KKKKKKKKKKKK-azide (SEQ ID NO: 5), (palmitoyl-glycine-acrylic- KKKKKKKKKK-azide (SEQ ID NO: 6), (palmitoyl-glycine-acrylic-KKKKKKKKKK- azide (SEQ ID NO: 7), (palmitoyl-glycine-acrylic-KKKKKKKKK-azide (SEQ ID NO: 8); (palmitoyl-glycine-acrylic-KKKKKKK-azide (SEQ ID NO: 9); (palmitoyl-glycine-acrylic- KKKKKK-azide (SEQ ID NO
  • a probe of the invention comprises a sequence set forth as one of SEQ ID NOs: 1-16, but instead of an azide or biotin, the probe includes a different linkable group for lipid staining. ExM and umExM Expansion microscopy (ExM) may be an ideal approach to this challenge.
  • ExM physically magnifies biological specimens by covalently anchoring biomolecules or labels to a swellable polymer network (typically sodium polyacrylate) synthesized in situ throughout the specimen, followed by chemical softening of the tissue, and the addition of water to swell the hydrogel.
  • a swellable polymer network typically sodium polyacrylate
  • the anchored biomolecules or labels are pulled apart from each other, isotropically, in the most popular form to a physical magnification of ⁇ 4x in linear dimension.
  • Established ExM methods generally comprise four steps: Anchoring, Gelation, Softening/Denaturation, and Expansion.
  • Anchoring involves attaching polymerization monomers, a non-limiting example of which is acrylamide, to fixed biological samples through covalent modification.
  • Gelation is a process of forming an acrylamide-acrylate hydrogel within the biological sample through polymerization.
  • Digestion/Denaturation comprises breaking down (digestion of) biomolecules using methods including but not limited to contact with enzymes, heat, denaturing agents, reducing agents, or a combination thereof.
  • Expansion comprises soaking the hydrogel in an aqueous solution (a non-limiting example of which is water) to physically expand (also referred to herein as “to magnify”) the 30 #17358430v1 biological sample that is embedded in the hydrogel.
  • Methods of the invention include modifications of standard ExM methods and may also comprise use of a probe of the invention in the modified ExM methods.
  • umExM method of the invention may include modified ExM methods and a probe of the invention as described herein.
  • Some embodiments of umExM methods of the invention comprise a modified expansion microscopy methods and include use of an amphiphilic membrane probe (a non- limiting example of which is pGk13a).
  • Some embodiments of a umExM method of the invention comprise an ExM method in which the softening step (Fig.1) is modified such that instead of softening using proteinase-k, the softening step is done using lysC/trypsin.
  • the softening step is modified such that instead of softening using proteinase-k, the softening step is done using lysC/trypsin.
  • methods and probes of the invention can be used to visualize cell boundaries (Fig.2).
  • Methods and probes of the invention are compatible with conventional antibody staining (Fig.3) if the proteins are cleaved in the softening step [Fig.1B panel (v)] using lysC/trypsin, instead of proteinase-k.
  • biomolecule may be used in reference to a protein molecule, a lipid molecule, a glycoprotein molecule, a polynucleotide molecule, or a carbohydrate molecule.
  • a method of the invention is performed on a single type of biomolecule and in certain embodiments a method of the invention is performed on a plurality of a single type of biomolecule.
  • a single type of protein may be assessed using an embodiment of a method of the invention, and the method may be used on a plurality of the single type of protein molecule.
  • a method of the invention is performed on two or more different biomolecules.
  • a biological sample may include a plurality of different protein molecules, and each may be assessed using a method of the invention.
  • a biological sample may include one or more polynucleotide molecules and one or more protein molecules, each of which may 31 #17358430v1 be assessed using a method of the invention.
  • the term plurality means more than one, which may be 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, or more.
  • biospecimen and “biological sample” are used interchangeably herein.
  • the term “clinical sample” is used herein to mean a sample obtained from a subject.
  • methods include embedding one or a plurality of a cell/tissue and/or membrane in a polymer, for example but not limited to an acrylamide polymer, followed by digestion, such as but not limited to proteolytic digestion, and swelling of the polymer comprising the embedded biomolecule(s).
  • a method and/or a probe of the invention may be used for to assess ultrastructure of a membrane in the imbedded sample and/or to assess one or more biomolecules of interest.
  • an embodiment of a method of the invention can be used to assess a protein in a biological sample, to detect and assess genomic DNA in a sample, to detect and assess a carbohydrate molecule(s) in a biological sample, etc.
  • Methods of the invention can be used to detect and identify one or more alternations in proteins, lipids, glycoproteins, polynucleotides, etc.
  • an embodiment of a method of the invention may be used to identify and assess a polynucleotide (DNA or RNA) sequence such as, but not limited to: a genomic DNA sequence from a subject; a wild-type (control) genomic DNA sequence; a wild-type RNA sequence; a genetically modified RNA sequence; a genetically engineered genomic DNA sequence, a genomic DNA sequence or RNA sequence known to be or suspected of being associated with a disease or condition.
  • DNA or RNA polynucleotide
  • Methods of the invention can be used to identify biomolecule components (e.g., amino acid sequences, nucleic acid sequences, etc.) and membrane ultrastructures as well as differences in one or more biomolecules obtained from different sources.
  • methods of the invention may be used to compare structure and/or sequence/components of a normal (e.g., control) biomolecule to structure and/or sequence/components of a biomolecule obtained from a biological sample.
  • a biological sample examined using a method and or probe of the invention is obtained from a subject who has, or is suspected of having, a disease or condition. Differences between the determined biomolecule and the control biomolecule may assist in identifying a biomolecule variation or abnormality associated with the subject’s disease or condition.
  • Methods of the invention may be used to provide structure and component information beyond that obtainable from assessment of spatial localization of biomolecules when examined in unexpanded conformations or beyond information obtainable using traditional light microscopy and similar to information obtainable using electron microscopy.
  • 32 #17358430v1 Polynucleotides/Proteins/Lipids/Carbohydrates/Glycoproteins
  • the term “nucleotide” as used herein includes a phosphoric ester of nucleoside—the basic structural unit of nucleic acids (DNA or RNA).
  • polynucleotide” and “nucleic acid” refer to a polymer comprising multiple nucleotide monomers and may be used interchangeably herein.
  • a polynucleotide may be either single stranded, or double stranded with each strand having a 5' end and a 3' end.
  • a nucleotide in a polynucleotide may be a natural nucleotide (deoxyribonucleotides A, T, C, or G for DNA, and ribonucleotides A, U, C, G for RNA)
  • protein and “polypeptide” refer to nitrogenous organic compounds comprising chains of amino acids and the terms may be used interchangeably herein.
  • lipid refers to organic compounds comprising fatty acids or their derivatives.
  • carbohydrate refers to a molecule that includes carbon (C), hydrogen (H) and oxygen (O) atoms.
  • saccharide is a group that includes sugars, starch, and cellulose.
  • Monosaccharides and disaccharides are relatively low molecular weight carbohydrates. Larger saccharides include polysaccharides and oligosaccharides.
  • glycoprotein as used herein refers to any of a class of proteins that have carbohydrate groups attached to a polypeptide chain.
  • a glycoprotein comprises oligosaccharide chains (glycans) that are covalently attached to amino acid sidechains.
  • a biomolecule assessed using a method of the invention is a “modified biomolecule,” which, as used herein, refers to a biomolecule that comprises one or more non-natural or derivatized components.
  • the term “component” means a portion of the biomolecule.
  • a protein biomolecule may be an amino acid
  • a polynucleotide biomolecule may be a nucleic acid
  • a lipid biomolecule may be a fatty acid
  • (4) a carbohydrate molecule may be a saccharide or polysaccharide
  • a glycoprotein molecule may be a carbohydrate molecule, an amino acid, or a protein molecule.
  • a component of a biomolecule is chemically or biochemically modified.
  • one or more modified components are incorporated into a biomolecule.
  • Modified biomolecules may confer desirable properties absent or lacking in the natural biomolecule and biomolecules comprising one or more modified components may be used in the compositions and methods of the invention.
  • a “modified 33 #17358430v1 biomolecule” refers to a biomolecule comprising at least one modified component.
  • a modified biomolecule may comprise one, two, three, four, five, or more modified components.
  • a polynucleotide may be DNA (including but not limited to cDNA or genomic DNA), RNA, or hybrid polymers (e.g., DNA/ RNA).
  • polynucleotide and “nucleic acid” do not refer to any particular length of polymer.
  • Polynucleotides used in embodiments of methods of the invention may be at least 1, 2, 3, 4, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 500, 1000, 2000, or 5000 kb or more in length.
  • protein does not refer to any particular length of the molecule.
  • a protein used in embodiments of methods of the invention may be at least 1, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 500, or more amino acids in length.
  • sequence refers to a contiguous series of nucleotides or amino acids, respectively.
  • structure refers to overall sequence organization of the polynucleotide, including “structural variations” such as insertions, deletions, repeats, and rearrangements.
  • a polynucleotide or other biomolecule may be chemically or biochemically synthesized, or may be isolated from a subject, cell, tissue, or other source or sample that comprises, or is believed to comprise, the biomolecule.
  • structure refers to overall s component organization of the biomolecule.
  • a membrane assessed using a method and/or probe of the invention may be a cultured membrane or may be obtained from a subject.
  • Embedding Embodiments of methods of the invention may include embedding the biomolecule in a polymer material for example using certain methods of standard ExM practice.
  • a means for embedding the biomolecule in the polymer material comprises incubating the biomolecule in a polymer monomer and polymerizing the monomer.
  • the polymer material is a swellable polymer material.
  • a non-limiting example of a swellable polymer material comprises an acrylamide-co-acrylate copolymer.
  • the term “swellable polymer material” generally refers to a material that expands when contacted with a liquid, such as water or other solvent [Wassie A., et al., Nat. Methods 16, 33-41 (2019); and US Patent 10,059,990 in relation to swellable and non-swellable materials, each publication is incorporated by reference herein in its entirety.] 34 #17358430v1
  • the swellable material may uniformly expand in three dimensions. Additionally, or alternatively, the material is transparent such that, upon expansion, light can pass through the sample.
  • the swellable polymer material is a swellable polymer or hydrogel.
  • the swellable polymer is formed in situ from precursors thereof: for example, one or more polymerizable materials, monomers or oligomers may be used, such as monomers selected from the group consisting of water-soluble groups containing a polymerizable ethylenically unsaturated group.
  • Monomers or oligomers may comprise one or more substituted or unsubstituted methacrylates, acrylates, acrylamides, methacrylamides, vinylalcohols, vinylamines, allylamines, allylalcohols, including divinylic crosslinkers thereof (e.g., N, N-alkylene bisacrylamides).
  • Precursors may also comprise polymerization initiators and crosslinkers.
  • a swellable polymer is an acrylamide-co-acrylate copolymer, polyacrylate, or polyacrylamide, or co-polymers or cross-linked co-polymers thereof.
  • the swellable polymer may be formed in situ by chemically cross-linking water-soluble oligomers or polymers.
  • the invention envisions adding precursors, such as water-soluble precursors, of the swellable polymer to the sample and rendering the precursors swellable in situ.
  • Certain embodiments of the invention include embedding a biomolecule in a non- swellable polymer material capable of conversion to a swellable polymer material.
  • the method may also include polymerizing the non-swellable polymer and then converting the polymerized non-swellable polymer material into a swellable polymer material.
  • non-swellable polymer material comprises a polymer material capable of conversion to a swellable polymer material, including a non-swellable hydrogel comprising one or more of an acrylamide and polyacrylate [Ueda H., et al., Nat. Rev. Neurosci.21, 61-79 (2020)].
  • the polymer is not a polyacrylade polymer.
  • the non- swellable polymer material is converted into a swellable polymer material before the physical expanding step of the method.
  • a non-swellable polymer can comprise various materials.
  • a non-swellable polymer material may include a non-swellable hydrogel.
  • a non-swellable hydrogel may include one or more of an acrylamide and polyacrylate.
  • a non-swellable polymer used in an embodiment of a method of the invention may be a polymer that can be chemically converted into a swellable polymer.
  • such a non-swellable polymer may be acrylamide; acrylamide can later be converted into an acrylamide-co-acrylate copolymer after treatment 35 #17358430v1 with a strong base such as sodium hydroxide, which can then swell after dialysis with water.
  • a umExM method of the invention comprise an ExM method in which the gelation temperature (also referred to herein as polymerization temperature) is modified such that instead of a standard gelation at room temperature or above, the umExM method includes gelation of the expansion sample at approximately 40C overnight.
  • Homogenization Embodiments of methods of the invention include a homogenization step. As used herein and in the expansion microscopy arts, the term “homogenization” refers to a process that frees up, also referred to as “releases” intra-sample connections before expansion. For example, a homogenization step may be used to release connections within the biomolecule.
  • a term for a biomolecule following a homogenization step in a method of the invention is “homogenized biomolecule material,” which indicates the biomolecule has been homogenized and is in a condition in which the biomolecule is capable of expansion.
  • a means of homogenization of a biomolecule includes one or more of an enzyme- based digestion and a heat-based denaturation.
  • a means of homogenizing an embedded biomolecule comprises contacting the polymer material in which the biomolecule is embedded with one or more of (i) a strong detergent or surfactant (e.g., sodium dodecyl sulfate); (ii) one or more enzymes; and (iii) denaturing heat.
  • a strong detergent or surfactant e.g., sodium dodecyl sulfate
  • enzymes e.g., sodium dodecyl sulfate
  • enzymes e.g., sodium dodecyl sulfate
  • denaturing heat e.g., denaturing heat.
  • enzymes that may be used in a homogenization step of a method of the invention are proteinase K (proK), an endoproteinase, non-limiting examples of which are: LysC and trypsin.
  • a umExM method of the invention comprise an ExM method in which the softening step (also referred to herein as a homogenization step) is modified such that instead of softening using proteinase K, the homogenization is done by contacting the gel with lysC and trypsin.
  • Polymer and Biomolecule Physical Expansion The polymer within which homogenized biological sample is embedded is isotropically expanded.
  • a solvent or liquid is added to the polymer 36 #17358430v1 containing the homogenized biomolecule material and the solvent or liquid is absorbed by the swellable material and causes swelling.
  • a means of physically expanding the sample material includes expanding the polymer material in which the biological sample material is embedded, wherein the expansion of the polymer material expands the homogenized biological sample material (and biomolecules in that biological sample) isotropically in at least a linear manner within the polymer material.
  • the physically expanded polymer and biomolecule materials are not re-embedded in a polymer prior to being detected. In certain embodiments of the invention, the physically expanded polymer and biomolecule materials are re-embedded in a polymer prior to being detected.
  • a method of the invention includes re-embedding the physically expanded polymer and biomolecule materials into a swellable polymer. Labelling A biomolecule or biomolecule material embedded in a polymer may be “labelled” or “tagged” with a probe of the invention.
  • a probe used in a method of the invention does not include a 37 #17358430v1 detectable label, but probe is prepared such that it comprises a component that permits inclusion of a detectable label on the probe at a subsequent time.
  • pGk5b or pGk13a may be used as a probe in a method of the invention without including a detectable label such as a fluorophore, but the probe may comprise azide, which may be used to attach a detectable label, a non-limiting example of which is a fluorophore, to the probe at a subsequent time.
  • a probe of the invention comprises a chemical handle, a non-limiting example of which is an azide, for chemoselective conjugation of a detectable label following the formation of an expandable hydrogel network, permitting the initial probe molecule to be small in size to facilitate diffusion and to avoid potential degradation of a detectable label, a non-limiting example of which is a fluorophore, during the free-radical polymerization of the hydrogel network.
  • a chemical handle a non-limiting example of which is an azide
  • a detectable label a non-limiting example of which is a fluorophore
  • a method of the invention includes incubating a tissue sample with a pGk13a probe for at least 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 36, or 48 hours, including all times within the range.
  • a sample tissue is incubated in a concentration of pGk13a probe that is at least 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, mg/ml, including all concentrations within that range.
  • the incubation is performed at about 4 o C.
  • a probe of the invention comprises a detectable label either initially or one subsequently attached to the probe.
  • detectable label means a label or tag chemically bound to the probe and when the probe binds a biomolecule in a biological sample or to a component thereof, the biomolecule or component is labelled. The label may be detected using microscopy or one or more other means of detection.
  • a detectable label may be selective for a specific target (e.g., a biomarker or class of molecule), as may be accomplished with an antibody or other target specific binder, or the detectable label on a probe of the invention may be an affinity label, including one or more of biotin, digoxigenin, and a hapten.
  • a specific target e.g., a biomarker or class of molecule
  • the detectable label on a probe of the invention may be an affinity label, including one or more of biotin, digoxigenin, and a hapten.
  • a detectable label of a probe of the invention comprises a visible component, as is typical of a dye or fluorescent molecule, a luminescent label, a radiolabel, an enzymatic label, a contrast agent, a heavy metal, or a heavy element such as bromine or iodine, or metals such as gold, osmium, rhenium, etc.; however, any signaling means used by the label is also contemplated.
  • enzymatic methods for detectable labeling are used, including contacting the biomolecule and/or biomolecule material with one or more enzymes, under suitable conditions for activity of the one or more enzymes to result in detectable labeling of biomolecule and/or biomolecule material, respectively.
  • Detecting structure and sequence methods of the invention allow detection of spatial structures and components of expanded biomolecule material and ultrastructural components of membranes using microscopic visualization methods.
  • the signal from individual molecules may be spatially punctate due to the homogenization step. However, these puncta will be spatially proximal, allowing the overall length of the biomolecule to be inferred based on the dimension in which spatial proximity is highest.
  • the structure of the biomolecule may be inferred over distances up to the entire length of the biomolecule.
  • detecting means using one or both of an imaging method and a sequencing method to identify the spatial position and components of biomolecules.
  • Imaging methods include but are not limited to light microscopy, epi-fluorescence microscopy, confocal microscopy, spinning disk microscopy, multi-photon microscopy, light-sheet microscopy, total internal reflection (TIRF) microscopy, light-field microscopy, imaging mass spectrometry, imaging Raman spectroscopy, super-resolution microscopy, or transmission electron microscopy.
  • enzymatic detection methods may be used to detect spatial structures and components of expanded biomolecule material and ultrastructural components of membranes.
  • Enzymatic means that can be used in conjunction with methods and probes of the invention include but are not limited to random primer extension, terminal transferase tailing, padlock probe rolling circle amplification [Larsson, C., et al. Nat. Methods 1(3): 227-232 (2004)], in situ PCR [Hodson, R., et al. Appl. Environ. Microbiol.4074-4082 (1995)], horseradish peroxidase tyramide signal amplification [Schonhuber, W., et al. Appl. Environ.
  • spatial position refers to the location of a biomolecule, biomolecule material, and/or biomolecule component relative to the location of another biomolecule, biomolecule material, and/or biomolecule component, respectively.
  • Certain embodiments of methods of the invention are useful to determine relative positions of one or more 39 #17358430v1 components or biomolecule materials generated from a single biomolecule. Spatial positions and relative spatial positions of ultrastructural components of a membrane can also be determined using embodiments of methods and probes of the invention. Thus, embodiments of methods of the invention can be used to disarticulate a biomolecule and/or ultrastructural membrane features into components in a controlled manner and then to identify the components and/or features and their relative positions in the expanded conformation. Segmenting cell compartments with umExM Methods and probes of the invention can be used for segmentation of cell compartments.
  • methods and probes of the invention are used for segmentation of neuronal compartments (i.e., cell bodies, dendrites, axons).
  • segmentation using umExM methods and probes is used to help analyze signaling proteins within distinct neuronal compartments.). It was determined that umExM images could capture and support the segmentation of neuronal compartments, and umExM methods of the invention enable capture and segmentation of cell compartments in fixed tissue, such as but not limited to neuron compartments, with a standard confocal microscope. Tracing Axons with umExM Methods and probes of the invention can be used to trace cell processes, such as axons.
  • methods of the invention comprise manual axon tracing using umExM images.
  • umExM methods of the invention may be used to trace neuronal processes, such as myelinated and unmyelinated axons.
  • methods and probes of the invention can be used to trace neurons in tissues such as, but not limited to cortex, hippocampus, and corpus callosum.
  • Antibody staining with umExM umExM methods and probes of the invention may, in some embodiments, be used with antibody staining of endogenous protein epitope. Methods of the invention may include pre-expansion antibody staining and/or post-expansion antibody staining.
  • Post-expansion antibody staining in conjunction with a method of the invention may be used to reveal previously unknown proteins and even cellular structures, in part because the antibodies are applied to expanded samples, where densely packed proteins are decrowded, making more room for antibody staining. Selection of antibodies may be based on routine parameters, such as, but not limited to the type of tissue and the goal of visualization. 40 #17358430v1 Visualization of proteins and RNAs with umExM Methods and probes of the invention may be used with RNA visualization methods.
  • umExM methods comprise methods of ExM visualization of RNA (ExFISH).
  • a umExM method of the invention comprises an RNA anchoring step, for example, using an art-known RNA anchor (for example but not intended to be limiting, LabelX) [Chen, F. et al. Nat. Methods 13, 679–684 (2016)].
  • RNA anchor for example but not intended to be limiting, LabelX
  • probe pGk13a is applied to label a membrane in a sample
  • LabelX anchoring solution is applied followed by AX anchoring solution, and then gelling, all at 4 0C.
  • the tissue is softened with proteinase K softening solution
  • pGk13a is fluorescently labeled
  • RNAs are labeled with a standard FISH hybridization chain reaction (HCR) protocol.
  • umExM methods and probes of the invention may be used to simultaneously visualize membranous structures along with proteins and RNAs, with a standard confocal microscope.
  • Biotin-DHPE, BODIPY-lipid, mCling staining and proExM umExM methods and probes of the invention can be used in conjunction with staining methods such as, but not limited to: BODIPY-lipid staining methods and mCling staining methods.
  • the biotin-DHPE, BODIPY, or mCling staining methods may be processed using an art-known proExM method.
  • Higher resolution imaging with umExM umExM methods of the invention may include by imaging samples with super- resolution imaging methods, or by increasing the expansion of the sample.
  • umExM methods of the invention comprise a super-resolution imaging method.
  • a non-limiting example of a super-resolution-imaging method is a super-resolution imaging based on autocorrelation with two-step deconvolution (i.e., SACD) in which a SACD algorithm is used to resolve fluctuations.
  • Some embodiments of umExM methods of the invention comprise an iterative form of umExM, in which the sample is be expanded 2, 3, 4, 5, or more times. Analysis Certain embodiments of methods of the invention may be used to analyze structure, spatial organization, and sequence of one or more ultrastructural membrane components and/or biomolecules of interest that may be known or may be suspected of being associated 41 #17358430v1 with a disease or condition.
  • Some embodiments of methods of the invention can be used to identify a biomolecule or ultrastructural membrane component associated with a disease or condition.
  • diseases and conditions that can be assessed using embodiments of the invention are: disease and conditions such as but not limited to: foodborne illness, food poisoning associated conditions; bacterial infection; viral infections; parasitic infections; poisoning; contamination and/or poisoning with one or more toxins and heavy metals; sickle cell anemia; hemophilia; cystic fibrosis; Tay Sachs disease; Huntington’s disease; fragile X syndrome; chromosomal disorders such as but not limited to: Down syndrome and Turner syndrome; polygenic disorders such as but not limited to Alzheimer’s disease, heart disease, cancers, and diabetes, etc.
  • tissue sample means a tissue obtained from a source, such as a subject or cultured source.
  • a tissue sample comprises a slice of a tissue.
  • a non-limiting example of a tissue slice is a slice of a brain.
  • the tissue is a fixed tissue.
  • Methods and probes of the invention may be used on tissue slices 50 to 100 microns thick, and in some embodiments, may be used on thinner or thicker samples.
  • Methods used with thicker samples may, compared with tissue slices that are 50 to 100 microns think, include longer probe incubation times, a higher concentration of probe, or both, as slice thickness increases.
  • a method of the invention that includes a pGk13a probe and a tissue sample thicker than 100 microns includes at least a 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 36, 48-hours-long pGk13a incubation time, including all times within the range.
  • a method of the invention that includes a pGk13a probe and a tissue sample thicker than 100 microns, the tissue is incubated in a concentration of the pGk13a probe at a concentration of at least 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, mg/ml, including all concentrations within that range.
  • Subjects and cells A tissue sample used in a method of the invention may be obtained from a subject or from a cultured source.
  • subject may refer to human or non-human animals, including mammals and non-mammals, vertebrates and invertebrates, and may also be any multicellular organism or single-celled organism such as a eukaryotic (including plants and 42 #17358430v1 algae) or prokaryotic organism, archaeon, microorganisms (e.g., bacteria, archaea, fungi, protists, viruses), and aquatic plankton.
  • a subject may be considered a normal subject or may be a subject known to have or suspected of having a disease or condition.
  • an organism is a genetically modified organism.
  • a subject is a plant.
  • a cell sample comprises a plurality of cells.
  • the term “plurality” means more than one. In some instances, a plurality of cells is at least 1, 10, 100, 1,000, 10,000, 100,000, 500,000, 1,000,000, 5,000,000, or more cells.
  • a plurality of cells from which biomolecules are obtained for use in methods of the invention may be a population of cells.
  • a plurality of cells may include cells that are of the same cell type.
  • a cell from which one or more biomolecules are obtained for use in methods of the invention is a healthy normal cell, which is not known to have a disease, disorder, or abnormal condition.
  • a plurality of cells from which biomolecules are isolated for use in methods of the invention includes cells having a known or suspected disease or condition or other abnormality, for example, a cell obtained from a subject diagnosed as having a disorder, disease, or condition, including, but not limited to a degenerative cell, a neurological disease-bearing cell, a cell model of a disease or condition, an injured cell, etc.
  • a cell is an abnormal cell obtained from cell culture, a cell line known to include a disorder, disease, or condition.
  • Non-limiting examples of diseases or conditions include disorders, such as sickle cell anemia, hemophilia, cystic fibrosis, Tay Sachs disease, Huntington’s disease, and fragile X syndrome; chromosomal disorders, such as Down syndrome and Turner syndrome; Alzheimer’s disease, heart disease, diabetes; and cancers.
  • a plurality of cells is a mixed population of cells, meaning all cells are not of the same cell type.
  • Cells may be obtained from any organ or tissue of interest, including but not limited to skin, lung, cartilage, brain, CNS, PNS, breast, blood, blood vessel (e.g., artery or vein), fat, pancreas, liver, muscle, gastrointestinal tract, heart, bladder, kidney, urethra, and prostate gland.
  • a cell from which one or more biomolecules are isolated for use in methods of the invention is a control cell.
  • cells from which one or more biomolecules are isolated for use in methods of the invention may be genetically modified or not genetically modified.
  • 43 #17358430v1 A cell or tissue for use in methods of the invention may be obtained from a biological sample obtained directly from a subject.
  • biological samples are samples of blood, saliva, lymph, cerebrospinal fluid, vitreous humor, aqueous humor, mucous, tissue, surgical specimen, biopsy specimen, tissue explant, organ culture, biological fluid or any other tissue or cell preparation, or fraction or derivative thereof or isolated therefrom, etc.
  • one or more biomolecules may be obtained from primary cells, cell lines, freshly isolated cells or tissues, frozen cells or tissues, paraffin embedded cells or tissues, fixed cells or tissues, and/or laser dissected cells or tissues.
  • a sample for use in a method of the invention is a fixed sample.
  • methods and probes of the invention may be used to achieve ⁇ 60nm lateral resolution and to enable co-visualization of membranous structures in a wide range of brain regions along with proteins and RNAs.
  • umExM methods and probes of the invention enable segmentation of cell bodies, dendrites, and axons (>200nm in diameter) and enable tracing of axons.
  • Studies set forth herein also demonstrate that ⁇ 35nm resolution imaging of membrane structures is possible by combining umExM with super-resolution imaging (e.g., SACD) or through an iterative form of umExM.
  • results of studies set forth herein demonstrate use of 4x umExM methods set forth herein to visualize mitochondria (Fig.4A, B, & G; Fig.5A) and ER (Fig.5B; Fig.6A and Fig.7 panel (iv), and use of the method to reveal some features of cytoplasmic vesicles (i.e., synaptic vesicles; Fig.3C-F).
  • the iterative form of umExM which provides higher resolution (i.e., ⁇ 35nm resolution, Fig.8G) compared to 4x umExM, is shown herein to reveal mitochondria cristae (Fig.9), with an appearance similar to that shown with isoSTED [Schmidt, R. et al.
  • Methods used with thicker samples may include longer pGk13a incubation times, or a higher concentration of pGk13a, or both, as slice thickness increases.
  • Methods Membrane probe synthesis Membrane probes were commercially synthesized (Anaspec). They were purified to >95% purity. They were aliquoted in 1 mg quantities into tubes, lyophilized to powder, and stored at -20 0C until stock solutions were prepared. Stock solutions were stored at -20 0C until use. Brain tissue preparation for umExM All procedures involving animals were in accordance with the US National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Massachusetts Institute of Technology Committee on Animal Care.
  • Wild type mice both male and female, used without regard to sex, C57BL/6 or Thy1-YFP, 6-8 weeks old, from either Taconic or JAX mice were first terminally anesthetized with isoflurane. Then, ice-cold 1x phosphate-buffered saline (PBS, Corning, catalog no.21031CM) was transcardially perfused until the blood cleared (approximately 25 ml). For all umExM experiments, the mice were then transcardially perfused with 4% PFA + 0.5% CaCl2 fixative solution (see Table 2 “fixative solution”).
  • PBS ice-cold 1x phosphate-buffered saline
  • Fixative solution (prepared fresh and used immediately): *Chilled on ice before use **Kept on ice during perfusion pGk13a stock solution (prepared at RT and immediately stored at -20 0C): 45 #17358430v1 pGk13a membrane labeling stock solution (prepared fresh and used immediately at 4 0C) * Chilled on ice before use AX buffer solution (prepared fresh and stored at 4 0C) * Chilled on ice before use AX stock solution (prepared fresh and used immediately at 4 0C, and stored at -20 0C*) * Aliquot 20ul into a PCR tube, and store at -20 0C in a sealed container (e.g., 50mL tube) with drying agents (e.g., Drierite) ** Anhydrous DMSO (Thermo Fisher, cat.
  • a sealed container e.g., 50mL tube
  • drying agents e.g., Drierite
  • umExM monomer solution (9.4ml, aliquoted to 10 tubes of 940 ⁇ l and stored at -20 0C): 46 #17358430v1 *All concentrations are in g/100 ml except 10xPBS.
  • umExM gelling solution (1ml, prepared and gelled at 4 0C): *All concentrations are in g/100 ml except umExM Monomer Solution. All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015).
  • umExM Digestion buffer* 100ml, prepared and applied at RT, and stored at 4 0C: dilution in Digestion buffer. All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015).
  • Second monomer solution (34.5ml, prepared fresh): 47 #17358430v1 *All concentrations are in g/100 ml.
  • Second gelling solution 50ml, prepared fresh: *All concentrations are in g/100 ml except cleavable second monomer solution.
  • 250 ⁇ l of TEMED solution (0.1g/ml in water) to 34.5mL of second monomer solution, vortex for ⁇ 10 seconds, add 250 ⁇ l of APS solution (0.1g/ml in water), vortex for ⁇ 10 seconds.
  • Third monomer solution (9.4ml, aliquoted to 10 tubes of 940 ⁇ l and stored at -20 0C): *All concentrations are in g/100 ml except 10xPBS.
  • Third gelling solution (prepared in fresh and applied at RT): *All concentrations are in g/100 ml. 48 #17358430v1 ** To make the third gelling solution, add 20 ⁇ l of 4-hydroxy-TEMPO solution (0.005g/ml in water) and 20 ⁇ l of TEMED solution (0.1g/ml in water) to 940 ⁇ l of monomer solution, vortex for 2-3 seconds, add 20 ⁇ l of APS solution (0.1g/ml in water), vortex for 2-3 seconds.
  • the fixative was kept on ice during perfusion. After the perfusion step, brains were dissected out, stored in fixative on a shaker ( ⁇ 10-20 rpm) at 4 0C for 24 hours for further fixation, and sliced on a vibratome (Leica VT1000S) at 100 ⁇ m thickness. For the slicing, the tray was filled with ice-cold PBS, and the tray was surrounded by ice. The slices were then transferred to a 50-ml tube filled with 40 ml of ice-cold quenching solution (100mM Glycine in PBS) on the shaker ( ⁇ 10-20 rpm) at 4 0C, overnight (> 8hrs).
  • the slices were washed 3-4 times with ice-cold PBS on the shaker ( ⁇ 10-20 rpm) at 4 0C, for 1-2 hours each and stored in PBS at 4 0C.
  • umExM for brain tissue slices 1.
  • the fixed tissue slices (as described in the Brain tissue preparation for umExM section) were incubated in membrane labeling solution (Table 2, “pGk13a stock solution”) on the shaker ( ⁇ 10-20 rpm) at 4 0C, overnight (unless otherwise noted, overnight means >16 hours). 2.
  • the fixed tissue slices were then incubated in AX stock solution (Table 2, “AX stock solution”) on the shaker ( ⁇ 10-20 rpm) at 4 0C, overnight.
  • the tissue was then washed 2-3 times in PBS on the shaker ( ⁇ 10-20 rpm) at 4 0C, 1 hour each. 3.
  • the fixed tissue slices were then incubated in gelling solution (Table 2, “umExM gelling solution”) 30 minutes on the shaker ( ⁇ 10-20 rpm) at 4 0C for pre-gelation incubation.
  • gelation chamber was constructed similarly as previously described [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)].
  • two spacers VWR, catalog no.48368-085) were placed on a microscope slide (VWR, catalog no.48300-026).
  • the two spacers were separated from each other enough so the brain tissue slice could be placed in between them.
  • the brain tissue slice was placed between the spacers and sliced with a razor blade (VWR, catalog no.55411-050) into two equally sized 49 #17358430v1 half-coronal sections.
  • the lid (VWR, catalog no.87001-918) was then placed on top of the spacers as well as the brain tissue slices.
  • the empty space between the half-coronal sections and spacers was fully filled with the gelling solution.
  • the chamber was transferred to a plastic jar with a lid (Fisher Scientific, catalog no. R685025) at 4 0C to initiate free-radical polymerization for >24 hours.
  • the gelation chamber containing the sample (tissue- embedded hydrogel) was taken out. 4.
  • the sample was trimmed with a razor blade (VWR, catalog no.55411-050) to have two gelled half-coronal sections.
  • Each gel (each half-coronal section) was then transferred from the chamber to a 12-well plate (Fisher Scientific, catalog no. FB012928) that contained proteinase K digestion solution (Table 2, “umExM Digestion buffer”) in the well (2 ml of digestion solution per well per half-coronal section).
  • the gel was then digested at room temperature (RT, 24 0C) on the shaker (50 rpm), overnight.
  • the brain tissue slice was prepared as described in step 1 in the umExM for brain tissue slices section. Then, 1ml of permeabilization solution (0.005%-0.01% of saponin (Sigma Aldrich, catalog no.84510) or triton (Sigma, catalog no. X100), 1% Bovine Serum Albumin (BSA, Sigma Aldrich, catalog no. A3294) in PBS) was applied and held at 4 0C, overnight. Then 10 ⁇ l of primary antibody, rabbit anti-SV2A (Abcam, catalog no.
  • step 6 in the umExM for brain tissue slices section was performed.
  • steps 1-3 in the umExM for brain tissue slices section were performed.
  • step 4 in the umExM for brain tissue slices section was performed, but with 2 ml of Trypsin+Lys-C softening solution (Table 2, “umExM Trypsin+Lys-C softening solution”) instead of proteinase K digestion solution, for each half coronal slice sample.
  • each half coronal slice sample (tissue-embedded hydrogel) was incubated in PBS containing primary antibodies, rabbi anti-PSD95 (Thermo Fisher, catalog no. MA1-046), at a concentration of 10 ⁇ g/ml at 4 0C for 2-3 days.
  • the samples were washed 3-4 times in PBS at RT, 30 minutes each.
  • step 5 in the umExM for brain tissue slices section was performed.
  • each half coronal slice sample was incubated in PBS containing primary antibodies, goat anti-rabbit ATTO 647N conjugated secondary antibody (Rockland Immunochemicals, catalog no.50-194-3924), at a concentration of 5-10 ⁇ g/ml at 4 0C for 2-3 days.
  • step 6 in the umExM for brain tissue slices section was performed.
  • Antibody staining of fluorescent proteins for umExM The expanded samples, after either proteinase K digestion (steps 1-4 in the mExM for brain tissue slices section) or Trypsin+Lys-C softening treatment (post-expansion antibody staining protocol in Immunohistochemistry-compatible umExM section), were incubated in PBS containing ATTO 647N fluorophore-conjugated nanobody against the green fluorescent protein (GFP, ChromoTek, catalog no.
  • GFP green fluorescent protein
  • sample tissue-embedded hydrogel
  • hybridization buffer (10% formamide, 2 ⁇ SSC) at RT for 0.5-1 h
  • ACTB probe was applied (Molecular instruments) at 8nM concentration, overnight at 37 0C (buffered in HCR.v3.0 Wash Buffer [Choi, H. M. T. et al. Development 145, (2018)]).
  • the gel was then washed with HCR v3.0 Wash Buffer for 2-3 times at 37 0C followed by another washing with second washing buffer (5x SSC buffer + 0.1% Tween 20) 30 minutes for 4 times at 37 0C, followed by treating the sample with fluorescently (Alexa 647) labeled HCR hairpin amplifiers (1:100) at RT, overnight. Then the samples were washed with 5 ⁇ SSCT, 20 minutes for 4 times at RT. The samples were expanded ( ⁇ 3x; similar to the expansion factor of ExFISH9 that used LabelX for anchoring RNAs) with 0.05 ⁇ SSCT, 10-20 minutes each time, 3 times.
  • Confocal imaging, deconvolution, and visualization Confocal images in the Figures were obtained on an Andor spinning disk (CSU-W1 Yokogawa) confocal system on a Nikon Eclipse Ti-E inverted microscope body with a Zyla 5.5 camera or a Hamamatsu qCMOS camera.
  • a 4x 0.2 NA, 10x 0.45 NA, 40x 1.15 NA, or 60x 1.27 NA lens was used for all imaging.
  • Large-scan imaging was performed with the confocal microscope and then stitched with a shading correction function via the default setting in Nikon element software version 4.0. All confocal images in the main figures were deconvoluted with the Sparse-deconvolution [Zhao, W. et al.
  • umExM+SACD images 40 frames of umExM images were captured and divided into two sets of 20 frames by separating odd and even images, and SACD (see umExM with optical fluctuation imaging section below) was performed to generate two SACD images (each derived from 20 frames). Subsequently, these two images underwent FRC resolution analysis using the same Fiji plugin. The best FRC value obtained across the blocks in each image pair was used to quantify the resolution of umExM, umExM+SACD and the iterative form of umExM. Analysis of the biotin (pGk13b) vs.
  • ExM was performed as described in the umExM for brain tissue slices section. The samples were then imaged in a random region in the hippocampus with the confocal microscope with 10x, 0.45NA objective. The mean pGk13a from each condition was then measured and an unpaired t-test function in RStudio 2021.09.2+382 with R version 4.1.2 was performed. Signal-to-background analysis The umExM samples were prepared as described in the umExM for brain tissue slices section.
  • the laser excitation power (mW) was measured. To do so, a Nikon W1 spinning disk equipped with a four-line laser system was used. Because the 561nm laser line was used for pGk13a signals, and the exposure time for this one laser line was reported, the laser excitation power (mW) of this line was measured.
  • pGk5b stock solution prepared at RT and immediately stored at -20 0C: 55 #17358430v1 pGk5b membrane labeling stock solution (prepared fresh and used immediately at 4 0C) * Chilled on ice before use * Aliquot 20ul into a PCR tube, and store at -20 0C in a sealed container (e.g., 50mL tube) with drying agents (e.g., Drierite) ** Anhydrous DMSO (Thermo Fisher, cat. no. D12345) Monomer solution aka StockX (9.4ml, aliquoted to 10 tubes of 940 ⁇ l and stored at -20 0C): *All concentrations are in g/100 ml except 10xPBS.
  • Digestion buffer* (100ml, prepared and stored at RT, applied at 37 0C): 56 #17358430v1
  • Digestion buffer (100ml, prepared and stored at RT, applied at 37 0C): 56 #17358430v1
  • NEB dilute Proteinase-K
  • All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015).
  • Fixation Reversal buffer (10ml, prepared at RT and used immediately): *All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015).
  • Cell Signaling Technology **Santa Cruz Biotechnology The sample was then gelled at 37 0C, using the gelation chamber described in step 3 of the umExM for brain tissue slices section.
  • the initial gel was treated with a cleaving solution (50mM sodium metaperiodate in 0.1M sodium acetate buffer, pH 5.0) for one hour on the shaker ( ⁇ 100-150 rpm), at RT. Then the sample was washed 4 times in 100mM glycine PBS on the shaker ( ⁇ 50-100 rpm) at RT, 30 minutes each, and then the sample was washed 3-4 times with PBS on the shaker ( ⁇ 50-100 rpm) at RT, 15 minutes each. The membrane probe was then fluorescently labelled, and the sample expanded as described in steps 5-6 of the umExM for brain tissue slices section.
  • a cleaving solution 50mM sodium metaperiodate in 0.1M sodium acetate buffer, pH 5.0
  • 57 #17358430v1 Accuracy (Rand score) of segmentation and tracing of pGk13a signals umExM was performed with fixed brain slices from Thy1-YFP mice and boosted YFP signals with anti-GFP (as described in the Antibody staining of fluorescent proteins for umExM section). Volumes of a random region in somatosensory cortex L6 and hippocampus dentate gyrus were imaged, with two labels (anti-GFP antibody and pGk13a for membranes). Segmentation To identify neuronal compartments, a maximum-intensity z-projected (max-z projected) image was generated from the anti-GFP channel of the volume.
  • ROIs were employed to crop the pGk13a channel and anti-GFP channel of the volume.
  • a single z-plane was randomly selected from the cropped volume, compartments were manually segmented based on pGk13a signals, and then the same compartments were segmented based on anti-GFP signals, all with ITK-SNAP software [Yushkevich, P. A. et al. Neuroimage 31, 1116–1128 (2006)].
  • the pGk13a-guided segmentation was then quantitively compared to the anti-GFP-guided segmentation using the Rand score [Unnikrishnan, R. et al., IEEE Trans. Pattern Anal. Mach. Intell.29, 929–944 (2007); Arbeláez, P.
  • umExM with Optical fluctuation imaging (umExM with SACD)
  • the samples were prepared as described in umExM for brain tissue slices.
  • the samples were imaged with Andor spinning disk (CSU-W1 Tokogawa) confocal system with a 60x, 1.27NA objective with either a Zyla 5.5 camera or a Hamamatsu qCMOS, with an optional ⁇ 1.5 magnification.20 frames of images (exposure time, 50ms; laser power 90%) were used, which took ⁇ 1 second in total.
  • the chest was cut open, and a 21-gauge butterfly needle was inserted into the left ventricle. A small incision was made in the right atrium to facilitate outflow.
  • the mouse was perfused transcardially at a flow rate of 10 mL/min using a Masterflex Peristaltic pump. Fresh aCSF was flown for 2-3 minutes to clear out the blood. This was followed by perfusion 59 #17358430v1 with 15% mannitol in aCSF solution for 1 minute, and then a 6% mannitol aCSF solution for 5 minutes.
  • the mouse was perfused with an ice-cold fixative containing 5% mannitol, 4% paraformaldehyde, 2mM CaCl 2 , 4mM MgCl 2 , and 150mM sodium cacodylate buffer (pH 7.4) for 5 minutes.
  • the brain was carefully removed from the skull and placed in a vial containing the same fixative solution. It was then fixed for at least 24 hours with gentle agitation at 4 0C.100 ⁇ m sections were cut using a Leica VT1000 S vibrating blade microtome and collected in the cold fixative solution.
  • umExM samples were prepared as described in the umExM for brain tissue slices section, except for fluorescently labeling the membrane probe (step 5).
  • the expanded samples were incubated in a cleavable re-embedding solution (Table 2, “Second gelling solution”) for 1 hour on a shaker ( ⁇ 50rpm) at RT for pre- gelation incubation.
  • the sample was gelled at 50 0C, for >4 hours, with the same gelation chambers used in step 3 of the umExM for brain tissue slices section.
  • the re- embedded samples were washed 3-4 times in PBS at RT, 30 minutes each.
  • the re-embedded samples were then treated with AX solution and washed in PBS as described in step 2 of the umExM for brain tissue slices section.
  • the samples were trimmed into smaller samples with razor blades and then gelled again with a non-cleavable gelling solution (Table 2, “Third gelling solution”) 30 minutes on the shaker ( ⁇ 10-20 rpm) at 4 0C for pre-gelation incubation.
  • the sample was gelled at 37 0C, overnight, with the same gelation chambers used above.
  • the samples were treated with the cleaving solution (50mM sodium metaperiodate in 0.1M sodium acetate buffer, pH 5.0) for one hour, at RT.
  • the samples were washed 4 times in 100mM glycine PBS on the shaker ( ⁇ 50-100 rpm) at RT, 30 minutes each, and then the sample was washed 3-4 times with PBS on the shaker ( ⁇ 50-100 rpm) at RT, 15 minutes each.
  • the membrane probe was then fluorescently labeled, and the sample expanded as described in steps 5-6 of the umExM for brain tissue slices section.
  • tissue slices of 100 ⁇ m thickness were treated with lipid labeling solution as used for mExM, except azide was used instead of biotin as the linkable group of the lipid stain (pGk5a for short):first the tissue was incubated in 1ml of pGk5 (0.1 ⁇ g/ml in ice-cold PBS) at 4 0C overnight (>16hrs) to let the 60 #17358430v1 labels diffuse and intercalate thoroughly throughout. Subsequently, the sample was washed 2x using PBS at 4 0C for 1 hour each to remove any excess lipid label.
  • EM electron microscopy
  • the azide-DBCO chemistry served to link the lipid label with a gold nanoparticle. Thereafter the sample was washed 3x with 0.15M sodium cacodylate buffer at room temperature for 30 minutes each to remove unbound nanoparticles. The samples were sent to the Harvard Medical School Electron Microscopy Core to be stained, then embedded and sliced using a standard EM preparation protocol [Eltoum, I. et al., J. Histotechnol.24, 173– 190 (2001)]. In summary, the tissue was stained with 1% uranyl acetate (UA) for 1 hour at room temperature, embedded in resin, and sliced in ultrathin sections (40nm thickness).
  • U uranyl acetate
  • a common UA staining protocol (1% UA for 1 hour at RT) would be used to enhance the pGk5a signals on top of signals from gold nanoparticles as UA can react to amino groups of pGk5a.
  • the protocol was adjusted by replacing pGk5a staining with common osmium staining (1% OsO4 for 1 hour at RT). Samples were imaged on a JEOL 1200EX transmission electron microscope using 80keV transmitted voltage. The images were captured with an AMT 2k CCD camera.
  • HEK293 or HeLa or U2OS cells were plated in the well ( ⁇ 40k cells/ml in cell culture medium (described in next paragraph) per well.)
  • the plate was then moved to a humidified cell culture incubator (set at 37°C, 20% oxygen, and 5% CO 2 ) for at least 6 hours for cells to adhere.
  • the cells were fixed with 4% paraformaldehyde (PFA) and 61 #17358430v1 0.1% glutaraldehyde in Dulbecco's 1x phosphate buffered saline (PBS) at room temperature (RT) for 15 minutes. Fixed cells were washed 4 times with PBS for 10 minutes each at 4 0C, and kept in PBS at 4 0C.
  • PFA paraformaldehyde
  • PBS Dulbecco's 1x phosphate buffered saline
  • the U2OS cell culture medium used was Dulbecco’s modified Eagle’s medium (DMEM, Corning, catalog no.10013CV) supplemented with 10% heat-inactivated fetal bovine serum (FBS, Thermo Fisher, catalog no. A3840001), 1% penicillin-streptomycin (Thermo Fisher, catalog no.15140122) and 1% sodium pyruvate (Thermo Fisher, catalog no. 11360070).
  • FBS heat-inactivated fetal bovine serum
  • Thermo Fisher, catalog no.15140122 1% penicillin-streptomycin
  • sodium pyruvate Thermo Fisher, catalog no. 11360070.
  • Transduction of cells via BacMam virus The adherent cells were prepared as described in the Cell Preparation section herein. The cells were transduced by directly adding 12 ⁇ l of BacMam reagent (either CellLightTM Mitochondria-GFP, catalog no.
  • the fixed cells (as described in the cell preparation methods herein) were incubated in the pGk5b solution (Table 4, “pGk5b membrane labeling stock solution”) at 4 0C overnight. 2. The cells were then incubated in the AcX solution (Table 4, “AcX stock solution”) overnight at 4 0C. Then the cells were washed with ice-cold PBS 2 times, 30min each at 4 0C. 3. The cells were then incubated in the gelling solution (Table 4, “Gelling solution”) 30min at 4 0C for pre-gelation. During this step, the gelation chamber was constructed as described previously 3 .
  • VWR spacers
  • catalog no.48368-085 two spacers were placed on a microscope slide (VWR, catalog no.48300-026). The spacers were separated from each other enough so an adherent cell-containing cover glass could be placed in between them. The adherent cell-containing cover glass was then placed between the spacers on the slide. The lid (VWR, catalog no.87001-918) was placed on top of the spacers, covering the cell-containing cover glass. Then the empty space between the cells and spacers was fully filled with the gelling solution. Next, the chamber was transferred to a 37 0C incubator to initiate free-radical polymerization. After 2 hours, the gelation chamber containing cells was taken out 4.
  • the gel was trimmed with a razor blade (VWR, cat. no.55411-050) and transferred from the chamber to a 6-well plate (Thermo Fisher, catalog no.140675) that contained proteinase K digestion buffer (Table 4, “Digestion buffer”) in the well (3mL of digestion buffer per well). The gel was then digested at 37C on a shaker overnight (> 16 hours). After digestion, the gel was washed 4 times in PBS at RT, 30 minutes each. 5.
  • the digested gels were labeled with 0.3mg/ml of streptavidin labeled with Atto 565 (Atto 565-Streptavidin; Sigma Aldrich, catalog no.56304-1MG-F) buffered in PBS overnight at RT, and then washed 4 times in PBS at room temperature (RT), 30 minutes each. 6. The gels were placed 4 times in excess water at RT for expansion, 30 minutes each.
  • the fixed tissue slices (as described in the Brain tissue preparation for mExM section herein) were incubated in a lipid labeling solution (Table 4, “Lipid labeling stock solution”) at 4 0C overnight (>16 hours) to let the labels diffuse and intercalate thoroughly throughout the tissue slices. 2.
  • the tissue slices were then incubated in an AcX stock solution (Table 4, “AcX stock solution”) overnight (>16 hours) at 4 0C.
  • the tissue was then washed 2 times in PBS at 4 0C, 1 hour each. 3.
  • the tissue slices were then incubated in gelling solution (Table 4, “Gelling solution”) 30min at 4 0C for pre-gelation incubation. During this step, the gelation chamber was constructed as previously described 7 .
  • VWR spacers
  • VWR catalog no. 48368-085
  • the two spacers were separated from each other enough so the brain tissue slice could be placed in between them.
  • the brain tissue slice was placed between the spacers.
  • the lid (VWR, catalog no.87001-918) was placed on top of the spacers as well as the brain tissue slice.
  • the empty space between the brain tissue slice and spacers were fully filled with the gelling solution.
  • the chamber was transferred to a 37 0C incubator to initiate free-radical polymerization. After 2 hours, the gelation chamber containing the tissue was taken out. 4.
  • the gel was trimmed with a razor blade (VWR, cat.
  • the digested gels were labeled with 0.3mg/ml of streptavidin labeled with Atto 565 (Atto 565-Streptavidin; Sigma Aldrich, catalog no.56304-1MG-F) buffered in PBS overnight at RT, and then washed 4 times in PBS at room temperature (RT), 30 minutes each. 6. The gels were placed 4 times in excess water at RT for expansion, 30 minutes each.
  • Immunohistochemistry-compatible mExM The aforementioned mExM steps were carried out the same, except for the digestion step (i.e., step 4 in the mExM for cells and mExM for brain tissue slices sections).
  • the sample was heated in fixation reversal (FR; Table 64 #17358430v1 1, “Fixation Reversal buffer”) buffer for 30 minutes at 100 0C and then held for 2 hours at 80 0C.
  • the FR buffer consisted of 0.5% PEG20000, 100mM DTT, 4% SDS, in 100mM Tris pH8. After this, the FR-digested sample was washed in 1x PBS 4 times at RT for 1 hour before proceeding to the immunohistochemistry steps.
  • the expanded gels were first blocked with MAXblock Blocking Medium (Active Motif, catalog no.15252) for 4-6 hours at room temperature and incubated in MAXbind Staining Medium (Active Motif, catalog no.15251) containing primary antibodies at a concentration of 10 ⁇ g/ml overnight at 4 0C. Then, the sample was washed with MAXwash Washing Medium (Active Motif, catalog no.15254) at RT 4 times, 30 minutes each and subsequently incubated in secondary antibodies buffered in MAXbind Staining Medium at a concentration of 10 ⁇ g/ml for 10-12 hours at 4 0C.
  • MAXblock Blocking Medium Active Motif, catalog no.15252
  • MAXbind Staining Medium Active Motif, catalog no.15251
  • the secondary antibodies were washed, again, with MAXwash Washing Medium at RT 4 times, 30 minutes each time.
  • anti-TOM20 Cell Signaling Technology, catalog no.42406S, rabbit; Santa Cruz Biotechnology, catalog no. sc-17764, mouse
  • anti-NUP98 Cell Signaling Technology, catalog no.2597S, rabbit
  • anti-myelin basic protein MBP; Cell Signaling Technology, catalog no.78896S, rabbit
  • Abcam catalog no. ab40390, rabbit
  • AVES catalog no. AB_2313550, chicken
  • the pGk5b probes that were conjugated to the gel were then labeled with 0.3mg/ml of streptavidin labeled with Atto 565 (Atto 565-Streptavidin; Sigma Aldrich, catalog no.56304- 1MG-F) buffered in PBS overnight at RT, and then washed 4 times in PBS at RT, 30 minutes each. Finally, the gel was placed 4 times in excess water at RT for expansion, 30 minutes each.
  • the lipid labels that were 65 #17358430v1 conjugated to the gel were then labeled with 0.3mg/ml of streptavidin labeled with Atto 565 (Atto 565-Streptavidin; Sigma Aldrich, catalog no.56304-1MG-F) buffered in PBS overnight at RT, and then washed 4 times in PBS at RT, 30 minutes each. Finally, the gel was placed 4 times in excess water at RT for expansion, 30 minutes each. Expansion factor and degree of isotropy analysis For umExM, expansion factor and degree of isotropy analysis were carried out with cells.
  • mExM protocol for cells is described mExM for cells method section herein.
  • umExM protocol for cells is similar as mExM for cells but using the pGk13a membrane labeling stock solution for step 1, AX stock solution for step 2, umExM gelling solution for step 3, umExM Digestion buffer for step 4, and pGk13a was fluorescently labeled with 1ml of Cy3 conjugated DBCO (Cy3 DBCO Click chemistry tools, catalog no. A140-1) buffered in PBS at a concentration of 0.03mg/ml on the shaker (50 rpm) at RT, overnight (>16 hours). The samples were expanded with water. The expansion factor was evaluated as previously described [Chen, F.
  • BacMam virus was used to express GFP proteins in the matrix of mitochondria in HEK293 and U2OS cells.
  • the cell was imaged with SIM before expansion and re-imaged the same region after expansion with a confocal microscope.
  • the SIM image and the confocal image were non-rigidly registered, then the root-mean-square (RMS) length measurement error was calculated as a function of measurement length for SIM vs. expanded-cell images.
  • RMS root-mean-square
  • expansion factor was measured with slices of fixed mouse brain. The gel size was measured before and after expansion (i.e., after 2 nd round of expansion) and divided the measured gel size to obtain expansion factor.
  • Colocalization analysis for mExM images A colocalization analysis for mExM was performed by adopting recommended colocalization methods for light microscopy studies [Bolte, S. & Cordeli Guatemala, F. P. J. 66 #17358430v1 Microsc.224, 213–232 (2006); Lunde, A. & Glover, J. C. Scientific Reports 202010:110, 1– 26 (2020)].
  • the foreground and background fluorescence of GFP channels of mExM images was first segmented using the Otsu image processing algorithm [Otsu, N. IEEE Trans. Syst. Man Cybern. SMC-9, 62–66 (1979)] as previously done for segmenting signals [Sarkar, D. et al.
  • a binary signal mask was then created based on the foreground signals and the signal mask was used to segment the pGk5b signals.
  • the fraction of expressed GFP and antibody signals that had pGk5b signals was evaluated by counting the pixels containing pGk5b signals that were above 1x standard deviation below the mean of the pGk5b signal intensity in the image. The analysis was performed with RStudio 2021.09.2+382 with R version 4.1.2. ExM with fGk5b and fGk5b+pGk5b.
  • step 1 in the umExM for brain tissue slices methods section herein was performed.
  • 0.1% GMA in 100 mM sodium bicarbonate was used on a shaker ( ⁇ 10-20 rpm) at room temperature, overnight.
  • step 3-6 the umExM method in the brain tissue slices method section herein were performed.
  • Biotin-DHPE, BODIPY-lipid, mCling staining and proExM Biotin-DHPE was adopted from ref [Wang, U.-T. T.
  • biotin-DHPE solution Biotium, catalog no.60022; 0.1 mg/ml in 50% ethanol
  • Premixed ABC solution was prepared by using equal amounts of avidin and biotin from the VECTASTAIN ABC-HRP Kit via mixing them in 67 #17358430v1 TBS at a 1:50 dilution and incubated for 30 minutes.
  • biotin-DHPE-tissue was extensively washed at RT with PBS to remove unbound biotin-DHPE.
  • the premixed ABC solution was then added to the biotin-DHPE-labeled tissue for 1 hour, followed by extensive washing at RT with PBS. After washing, Alexa Fluor-555-conjugated tyramide solution (Invitrogen, catalog no. B40955) in TRIS (1:100 dilution) containing hydrogen peroxide (0.03%; note that ref [Wang, U.-T. T. et al. Sci. Rep.13, 21922 (2023)] used amplification dilution buffer which is currently discontinued), was applied to the tissue for 20 minutes at RT, and followed by extensive washing at RT with PBS.
  • BODIPY-lipid The BODIPY-FL-C12 (BODIPY) staining protocol was adopted from ref [Cui, Y. et al. PLoS One 18, e0291506 (2023)].
  • BODIPY BODIPY-FL-C12 staining protocol
  • 4% fixed mouse brain tissue was incubated in 10 ⁇ g/mL BODIPY (Invitrogen, catalog no. D3822) in PBS at RT, overnight. BODIPY stained tissue was then extensively washed with PBS.
  • mCling The mCling staining protocol was adopted from ref [Damstra, H. G. J. et al. Elife 11, (2022)].
  • the membrane labeling probe should have a chemical handle that allows for selective conjugation of fluorophores subsequent to the formation of the ExM polymer. This design ensured that the membrane labeling probe remained small in size, facilitating its diffusion and preventing potential degradation of the fluorophore during free-radical polymerization of the ExM gel [Tillberg, P. W. et al. Nature Biotechnology 2016 34:934, 987–992 (2016)].
  • the membrane labeling probe should have a polymer- anchorable handle to incorporate into the ExM gel network for physical expansion.
  • These three features collectively are included in the design of membrane labeling probes that achieve both dense membrane coverage and compatibility with ExM chemistry, allowing for nanoscale imaging of membranous structures with a standard confocal microscope.
  • the probe design proceeded in two phases – a preliminary phase and a final phase.
  • the preliminary phase was used to explore certain aspects of chemical space, and to validate certain aspects of dense membrane staining in ExM.
  • the final phase was then used to refine the properties of the stain for optimal performance, and to perform an even more detailed validation of the density of the membrane staining possible.
  • the membrane labeling probe was designed to contain a chain of lysines with primary amines for binding to a polymer-anchorable handle, such as acryloyl–X (AcX) [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)], previously used to anchor protein amines to the ExM hydrogel [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)].
  • a lipid tail was included on the amine terminus of the lysine chain, with a glycine in between, to provide mechanical flexibility [Yan, B. X. & Sun Qing, Y. J. Biol.
  • D-lysines were used, rather than the biologically typical L-lysines, to minimize degradation during the chemical softening step of ExM, which in its most popular form involves a proteinase K softening step [Tillberg, P. W. et al. Nature Biotechnology 201634:9 34, 987–992 (2016)].
  • a chemical handle was attached to the carboxy terminus of the lysine chain, for selective conjugation to fluorophore(s) after expansion.
  • palmitoyl and biotin were selected as the lipid tail and chemical handle, respectively, and five D-lysines were included in the backbone.
  • This design resulted in a glycine and penta-D-lysine peptidic backbone, with a palmitoyl group on the amine-terminus and a biotin 69 #17358430v1 on the carboxy-terminus.
  • This preliminary probe was named: pGk5b (palmitoyl-glycine-(D- lysine)5-biotin).
  • Electron microscopy (EM) was used to validate the preliminary probe design (with biotin replaced with azide (and denoted pGk5a) so that gold nanoparticles could be added via click chemistry for EM imaging) and results indicated that membranes were labeled (see Fig. 17).
  • pGk5a an azide version of pGk5b (referred to herein as pGk5a) was used to conjugate gold nanoparticle-DBCO for EM imaging, instead of using pGk5b with gold nanoparticle-streptavidin.
  • This approach was selected because applying streptavidin to the tissue sample typically requires detergent to remove membranes, which could impact downstream processing.
  • Mouse brain tissue sections were incubated with 100 ⁇ M of the membrane probe pGk5a (pGk5b, with an azide replacing the biotin). The specimens were post-labeled with gold nanoparticles modified with a dibenzocyclooctyne (DBCO) handle, for EM visualization (see further detail elsewhere herein).
  • DBCO dibenzocyclooctyne
  • EM sample processing involving PFA+glutaraldehyde fixation followed by osmium staining, has been optimized over decades [Chen, F. et al., Science 347, 543–548 (2015); Chang, J. B. et al. Nature Methods 201714:614, 593–599 (2017); Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)], and the clear visualization of organelles is protocol-dependent.
  • matrix-targeted GFP is indistinguishable from mitochondrial membrane in the context of the current experiment.
  • mExM was then performed on, and imaged in, the very same cells with a confocal microscope. Comparing pre-expansion SIM images of mitochondrial matrix-targeted GFP to post-expansion images of either GFP (with anti-GFP labeling for boosting GFP signals), or pGk5b, indicated the same low distortion (a few percent, over ⁇ 10 ⁇ m) as was found for previous ExM protocols (see further detail elsewhere herein; Fig.4F and Fig.4I). By comparing the distance between two landmarks in pre- vs.
  • mExM could accurately visualize mitochondria and ER in cells.
  • mExM was compatible with slices of fixed mouse brain and provided more details compared to the unexpanded state (Fig.7) and was compatible with antibody staining.
  • a commonly used ExM softening protocol [Chen, F. et al. Nat. Methods 13, 679–684 (2016)] (i.e., SDS solution at a high temperature) was used that could reveal previously unseen structures by preserving protein epitopes through the expansion process [Chen, F. et al. Nat. Methods 13, 679–684 (2016)].
  • This protocol builds upon post- expansion protein-retention ExM (proExM) protocols [Klimas, A. et al.
  • mExM was applied to fixed brain tissue from mice and antibody staining was performed against organelle-specific membrane-localized proteins including TOM20 for mitochondria, NUP98 for the nuclear pore complex, and MBP for myelin, and results indicated in all cases that >98% of the pixels exhibiting these reference indicators also exhibited pGk5b signals (Fig.18).
  • mExM was performed with antibody staining (see further detail elsewhere herein), using antibodies against organelle-specific membrane-localized proteins, TOM20 for mitochondria (Fig.18A), and Nup98 for the nuclear pore complex (Fig.18B).
  • myelin was labeled using an antibody against myelin basic protein (MBP) (Fig.18C).
  • MBP myelin basic protein
  • lipids were compared, while keeping the rest of the probe design constant, and it was observed that the palmitoylated probe achieved a denser membrane labeling compared to the farnesylated one (Fig.19A-B). Furthermore, using a mixture of palmitoylated and farnesylated probe did not achieve denser membrane labeling (Fig.19C). Omitting the glycine linker caused a loss of detail (Fig.6). Finally, the total number of lysines in the backbone of the membrane labeling probe was varied.
  • lysines i.e., 3, 7, 9, 11, 13, and 19 lysines
  • these probes were applied to slices of fixed mice brain and ExM was performed.
  • the probe containing 13 or more lysines appeared to show the boundary of neuronal processes the best (Fig.20).
  • a probe with 13 lysines (pGk13b) was selected to minimize probe size, to facilitate its diffusion throughout brain tissue.
  • pGk13b was applied to slices of fixed Thy1-YFP mouse brain, which expresses cytosolic yellow fluorescent protein (YFP) under the Thy1 promoter in subsets of neurons [Feng, G. et al. Neuron 28, 41–51 (2000)].
  • ExM was performed and pGk13b was fluorescently labeled with Cy3-conjugated streptavidin, treating the sample with anti-GFP (many fluorescent proteins survive proteinase K softening [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)]; anti-GFP binds YFP) to boost YFP signals.
  • pGk13b + Cy3-streptavidin (for which each streptavidin bears more than one Cy3, according to the vendor) was compared with pGk13a + Cy3-DBCO (exhibiting one Cy3 per DBCO) in the context of ExM imaging of the hippocampus in fixed mouse brain slices. It was determined that the mean signal of pGk13a was >2x higher than that of pGk13b (Fig.13). Probe pGk13a (palmitoyl-glycine-(D-lysine) 13 -azide, Fig.1A), was used for the rest of the studies.
  • brain slices were prepared from mice that were fixed with 4% PFA and 0.5% CaCl 2 at 4 0C, and standard ExM (37 0C gelation) or modified ExM procedure (4 0C gelation) were performed, and hippocampal regions were imaged with a confocal microscope, with results indicating the mean signal of pGk13a from the modified ExM procedure (4 0C gelation) was ⁇ 50% higher than from the standard ExM procedure (37 0C gelation) (Fig.22).
  • the protocol was finalized as follows: mouse brain fixed in 4% PFA and 0.5% CaCl2, the brain sectioned, excess aldehydes quenched with a commonly used 100 mM glycine 1x phosphate-buffered saline (PBS) solution, pGk13a applied at 150 ⁇ M, applied a previously established biomolecule anchoring solution (acrylic acid N-hydroxy succinimide ester (AX, a reagent that is smaller, more cost-effective, yet functionally analogous to AcX [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)] in the context of ExM) in MES buffer, pH 6.0) [Shen, F. Y. et al.
  • PBS glycine 1x phosphate-buffered saline
  • AX acrylic acid N-hydroxy succinimide ester
  • umExM The isotropy of umExM was validated by quantitatively comparing pre-expansion structured illumination microscopy (SIM) images to post-expansion confocal images of the 74 #17358430v1 same sample and calculating the distortion across the images as was done for membrane expansion microscopy (mExM) above herein (see further detail elsewhere herein; Fig.4C-I).
  • SIM pre-expansion structured illumination microscopy
  • mExM membrane expansion microscopy
  • umExM was measured via Fourier Ring Correlation (FRC) resolution analysis [Banterle, N. et al., J. Struct. Biol.183, 363–367 (2013); Culley, S. et al. Nat. Methods 15, 263 (2018)], a gold-standard method which uses Fourier transformation of images to measure resolution, on pGk13a signals from expanded samples.
  • FRC Fourier Ring Correlation
  • umExM was applied to fixed brain slices from Thy1-YFP mice and the YFP signals were boosted with anti-GFP treatment.
  • the hippocampal dentate gyrus (Fig. 2H-I), third ventricle (Fig.2J), and somatosensory cortex layer (L) 6 (Fig.2K-L) were then imaged. Axons were identified by examining pGk13a signal flanking anti-GFP signals (Fig. 2M).
  • the diameter of axons is known to be diverse across brain regions [Sun, D.-E. et al. Nat. Methods 18, 107–113 (2021)]. However, results of studies described herein aligned with measured axon diameters from EM images of the same brain regions (i.e., cortex and dentate gyrus) [Sun, D.-E. et al. Nat. Methods 18, 107–113 (2021); Götz, R. et al. Nat. Commun.11, 6173 (2020); Wen, G. et al. ACS Nano 14, 7860–7867 (2020); Wang, U.-T. T. et al. Sci. Rep. 13, 21922 (2023)].
  • Motile cilia were also identified in the third ventricle by their fingerlike morphologies (Fig.2Q); their diameter (Fig.2R) was comparable to previous measurements made using EM [Reynolds, M. J. et al. Sci. Rep.8, 7977 (2016)].
  • a volume of the third ventricle was imaged and visualized through 3D volume rendering (Fig.2S), and membrane vesicles (known as extracellular vesicles; arrows in Fig.2S and Fig.24) were found around cilia, similar to what was previously seen with EM [Wood, C. R. & Rosenbaum, J. L.Trends Cell Biol.25, 276–285 (2015); Kesimer, M.
  • S/B average signal-to-background ratio
  • pGk13a signal divided by the background background was calculated as the average of images of empty gel regions
  • umExM was applied to a 100 ⁇ m thick fixed coronal slice of mouse brain (Fig.26A) and large-scan imaging of the expanded sample was performed with a low magnification objective (4x, 0.2NA; Fig.26B) at 30 milliseconds (ms) laser exposure time (see Table 3 for details).
  • the mean S/B ratio of a single z-plane was then measured at different depths of the volume (Fig.26D) and results indicated a consistently high mean S/B ratio (>40 fold higher than background, Fig.26D) throughout the slice.
  • ciliary membrane was manually traced and the number of gaps along them were counted, with a gap defined as a region with intensity smaller than two standard deviations below the mean pGk13a signal along the ciliary membrane, that was longer than 60nm (the effective resolution of umExM using a 60x, 1.27NA water objective; Fig.2G). Results indicated that >97% of the ciliary membrane was continuous by this metric at a gap measurement length of 60nm (Fig.2V). Studies were then performed to compare umExM to prior commercially available membrane probes used for lipid or membrane imaging, namely BODIPY FL C12, mCling and Biotin-DHPE.
  • ExM was performed, and the same S/B ratio analysis and continuity analysis as described elsewhere herein were performed. Results 77 #17358430v1 indicated the S/B ratio for umExM images was ⁇ 30-39 times higher than that of Biotin- DHPE, BODIPY FL C12, and mCling images (Fig.28A-C). A random part of the CA1 region of the mouse hippocampus was imaged with a 4x objective at 50ms laser exposure time for samples that were stained with biotin-DHPE, BODIPY FL C12, or mCling (Fig. 28A-C). umExM (pGk13a) was also performed for comparison (Fig.28D). Qualitatively, umExM generated the highest contrast image, compared to the others.
  • the signal-to- background (S/B, where the background was determined as the average across images of empty gel regions) was measured for the images obtained from each sample.
  • S/B The signal-to- background
  • Results indicated that the S/B for umExM images was many times higher than those of Biotin-DHPE, BODIPY FL C12, and mCling sample images.
  • Results also indicated that the signals of existing membrane probes were not dense enough to trace ciliary membranes, thus the aforementioned continuity analysis was not possible to perform (Fig.28F-G). Cilia in the 3rd ventricle were imaged to perform continuity analysis, as was done in Fig.2V. However, except for the sample stained with mCling, no signals were observed.
  • post-expansion staining can reveal previously unknown proteins and even cellular structures [Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)], as antibodies are applied to expanded samples, where densely packed proteins are decrowded, making more room for antibody staining.
  • a small amount of detergent i.e., 0.005%-0.01% of saponin or triton-x
  • PLoS One 18, e0291506 (2023)] that enabled antibody staining after expansion was adapted.
  • a softening solution that contained site-specific proteases including trypsin and LysC was used, and then immunostaining was performed after sample expansion.
  • umExM was performed with post-expansion antibody staining, using an antibody against PSD95 (Fig.3G-L).
  • These signals were adjacent to pGk13a signals (Fig.3L, arrows), consistent with the known role of PSD95 as a postsynaptic density protein.
  • RNA anchoring step was added to the umExM protocol using a previously established RNA anchor (i.e., LabelX) [Chen, F. et al. Nat. Methods 13, 679–684 (2016)], so the protocol became as follows: pGk13a was applied to label the membrane, LabelX anchoring solution was applied followed by AX anchoring solution, and then gelling, all at 4 0C. The tissue was then softened with proteinase K softening solution, pGk13a was fluorescently labeled, and RNAs were labeled with a standard FISH hybridization chain reaction (HCR) protocol.
  • HCR FISH hybridization chain reaction
  • umExM enables simultaneous visualization of membranous structures along with proteins and RNAs, with a standard confocal microscope. Segmenting neuron compartments with umExM Studies were performed to investigate whether umExM could support the segmentation of neuronal compartments (i.e., cell bodies, dendrites, axons) to help with the analysis of signaling proteins within distinct neuronal compartments.
  • neuronal compartments i.e., cell bodies, dendrites, axons
  • umExM provides ⁇ 60nm lateral resolution (Fig.2G)
  • Fig.2G lateral resolution
  • umExM images could capture neuronal 79 #17358430v1 processes that are larger than roughly >120nm (resolution of umExM multiplied by two).
  • umExM was applied to fixed brain slices from Thy1-YFP mice, anti-GFP staining was performed to boost YFP signals, and volumes of random regions of somatosensory cortex L6 and hippocampal dentate gyrus were imaged.
  • Fig.29A Cell bodies were randomly selected using anti-GFP signals, they were manually segmented based on pGk13a signals, and then the same cell body was segmented based on anti-GFP signals (Fig.29A) through the commonly used EM image segmentation software, ITK-SNAP [Yushkevich, P. A. et al. Neuroimage 31, 1116–1128 (2006)]. This procedure was repeated for dendrites (Fig. 29B), myelinated axons (Fig.29C), and unmyelinated axons (Fig.29D, see Methods for details).
  • the Rand score was utilized, which is a recommended and commonly used metric for assessing EM-based imaging segmentations [Unnikrishnan, R. et al., IEEE Trans. Pattern Anal. Mach. Intell.29, 929–944 (2007); Arbeláez, P. et al., IEEE Trans. Pattern Anal. Mach. Intell.33, 898–916 (2011)], with a Rand score of 0 meaning no similarity between the pGk13a signal-guided versus anti-GFP signal-guided segmentations, and a Rand score of 1 meaning segmentations from the two signals are identical.
  • umExM images could capture and support the segmentation of neuronal compartments
  • thin processes such as tiny axons (as they can be ⁇ 50nm in diameter [Helmstaedter, M. Nature Methods 201310:610, 501–507 (2013)]) and spine necks (known to be ⁇ 40-50nm in diameter [Helmstaedter, M. Nature Methods 201310:610, 501–507 (2013)]) cannot be yet resolved, as umExM provides ⁇ 60nm resolution (Fig.2G).
  • umExM still enables capture and segmentation of neuron compartments that are larger, in fixed tissue, with a standard confocal microscope.
  • the tracing was then repeated using the anti-GFP signals (Fig.30A, column “GFP”).
  • the tracing results based on the pGk13a and anti-GFP signals were visually indistinguishable (Fig.30B).
  • the corpus callosum a brain region containing densely packed myelinated axons.
  • manual tracing of neuronal processes was challenging in this region as only a subset of the processes were visually distinguishable (Fig.16A-B), perhaps due to light scattering; this optical phenomenon was not observed in the somatosensory cortex and hippocampus.
  • Previous studies reported that a subset of native lipids, which causes scattering, may still remain even after tissue clearing [Murray, E. et al. Cell 163, 1500 (2015)] and expansion processes [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)].
  • umExM was performed on a fixed brain slice until the softening step was completed, and then biomolecule anchoring (AX) solution was applied again (so that pGk13 probes in the initial gel could be transferred from the initial gel to a subsequently formed ExM gel; the newly applied AX would react to unreacted amines in pGk13a), an expandable 81 #17358430v1 gel was cast that was prepared with non-cleavable crosslinker N,N - methylenebis(acrylamide) (BIS) in the initial gel, the initial gel (which was made with cleavable crosslinker N,N'-Diallyl-L-tartardiamide (DATD)) was chemically cleaved, pGk13a was fluorescently labeled via click chemistry, and the sample was expended with water.
  • AX biomolecule anchoring
  • umExM of fixed mouse brain slices was performed and a confocal microscope used to image 20 frames of a hippocampal region at an imaging rate of 50ms/frame (Fig.8A-B). Then the SACD algorithm [Zhao, W. et al. Nat. Photonics 1–8 (2023)] was used to resolve the fluctuations (Fig.8C). The resolution of the resulting image was measured the same way as for umExM. umExM+SACD provided a final effective resolution of ⁇ 33nm (Fig.8D). Next studies were performed to create an iterative form of umExM, adapted from the previously established iterative form of ExM (iExM) [Chang, J. B. et al.
  • umExM was performed on fixed brain slices but without fluorescently labeling pGk13a.
  • the expanded 82 #17358430v1 sample was then embedded into a re-embedding gel (uncharged gel) prepared with a cleavable crosslinker (DATD) to preserve the expanded state during subsequent steps [Chang, J. B. et al.
  • Example 3 Studies were performed to quantitatively examine the ultrastructure preservation of the umExM by comparing it to known ultrastructure features. To achieve this, umExM was applied to the adult Thy1-YFP mouse brain and it was treated with anti-GFP labeling (i.e., atto647n fluorophore attached GFP Nanobody) for boosting GFP signals, then the cortex L6, third ventricle, and hippocampus dentate gyrus were imaged.
  • anti-GFP labeling i.e., atto647n fluorophore attached GFP Nanobody
  • Axons were identified by examining the pGk13a signal bearing GFP signals.
  • the finding agrees with previously reported values via EM images [Claiborne, B. J.
  • the cilia in the third ventricle was also identified by its fingerlike morphologies and its location in the expanded mouse brain.
  • umExM provides dense labeling of membranes in a wide range of thick mouse brain tissue section that enables visualization of the ultrastructural details. To demonstrate this, umExM protocol was applied to the 100 ⁇ m thick brain tissue sections from adult C57BL/6 mice and SDCM was used to image a wide range of brain regions including cortex layers, corpus callosum, third ventricle, choroid plexus, and hippocampus.
  • the labeling was dense enough that it only required 100ms laser exposure time to take a single image slice (13.65 ⁇ m x 13.65 ⁇ m) with SDCM equipped with a standard sCMOS camera (i.e., Zyla 5.5) and 60x 1.27NA water immersion objective. Due to the dense labeling of membranes, the surface topologies, such as cilia, of the epithelial cells in the third ventricle of the mouse brain could be visualized through 3D volume rendering. The membrane vesicles (known as extracellular vesicles) were also found. There are currently no clear molecule markers (e.g., an antibody) with which to characterize the vesicles using FM [Sun, S. et al.
  • umExM enabled visualization of ultrastructure features, similar to low-resolution EM images, to visualize ultrastructure context of neural circuits.
  • Example 4 the uniformity of labeling throughout the 3D volume of umExM-processed thick tissue (100 ⁇ m thick, equivalent to half of a coronal slice) was evaluated, utilizing two approaches: (1) the variations in overall labeling were investigated, as quantified by the signal-to-background (S/B) ratio of the entire XY plane, at different depths of the expanded tissue volume; (2) the changes in signal-to-noise (S/B) ratio along different line profiles within each XY plane were analyzed.
  • S/B signal-to-background
  • umExM was applied to the thick (100 ⁇ m) adult C57BL/6 mouse brain tissue section and it was imaged with low and high magnification objectives (4x objective and 60x NA objective) with varying laser exposure 84 #17358430v1 time (30ms, 50ms, and 100ms), and the signal-to-background (S/B) ratio as described above was measured.
  • the background was measured by imaging a region where the tissue was not embedded.
  • the mean pGk13a signals across the z-axis were obtained (i.e., from the bottom and top of the sample).
  • the S/B ratio was measured by dividing pGk13a signals by background signals. Consistently high S/B ratio (minimum of ⁇ 40 folds higher than the background, was observed across the entire z-axis.
  • the pGk13a signal across the brain regions was analyzed by measuring S/B ratio.
  • the pGk13a signal pattern was similar to what was previously observed with large-scale EM imaging [Mikula, S. & Denk, W. Nature Methods 201512:612, 541–546 (2015)].
  • a consistently high S/B minimum of ⁇ 20 folds higher than the background, was observed in the cortex regions and a higher S/B (i.e., ⁇ 2x) in the corpus callosum region compared to cortex regions.
  • umExM enables multiscale imaging and visualization of ultrastructure features across the wide brain regions.
  • Example 4 umExM, as described above herein, is compatible with mouse brain tissue, and provides ultrastructural details similar to low-resolution EM images. However, with proteinase K treatment, proteins would be unavailable for antibody staining except for fluorescent proteins [Honig, M. G. & Hume, R. I. Trends Neurosci 12, 333–341 (1989)]. Certain embodiments of methods of the invention can be used to image antibody-stained proteins and lipids in the same specimen, to localize proteins amidst specific ultrastructural 85 #17358430v1 features, and to help with the analysis of signaling proteins amidst cellular organization.
  • Certain embodiments of methods of the invention can be used for post-expansion staining, retaining proteins through the umExM procedure, as post-expansion staining can reveal previously unknown structures [Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)].
  • previously established antibody labeling strategies for ExM [Tillberg, P. W. et al. Nat Biotechnol 34, 987–992 (2016); Cui, Y. et al. bioRxiv 2022.06.19.496699 (2022) doi:10.1101/2022.06.19.496699] were adopted, namely pre-expansion staining [Scicchitano, M. S.
  • a small amount of detergent i.e., 0.005%-0.01% of saponin or triton-x
  • a small amount of detergent i.e., 0.005%-0.01% of saponin or triton-x
  • saponin or triton-x was used to permeabilize the membrane in thick mice brain tissue, which was stained with primary antibody against sv2a, a pre-synapse marker and synaptic vesicle marker, at 4 o C, umExM was performed, and then the sample was incubated with a secondary antibody.
  • hippocampus CA4 region although there was not sufficient resolution to resolve individual synaptic vesicles, the pool of synaptic vesicles in the mossy fiber bouton was observed, similar to EM observation [Rollenhagen, A. et al.
  • milder enzyme i.e., Trypsin/Lys-C instead of proteinase-k
  • umExM was performed on thick mice brain tissue but the tissue was softened with Trypsin/Lys-C instead of using proteinase-k, then stained with primary antibody against PSD-95, a post-synapse marker and post-synaptic density marker, followed by secondary antibody incubation.
  • umExM methods of the invention were determined to be compatible with conventional antibody staining, allowing simultaneous visualization of the biomolecule of interest in the ultrastructural context through dense labeling of membranous structures, with ordinary FM.
  • Example 5 To achieve an ⁇ 30nm resolution with a standard confocal microscope, the technology was first checked with another super-resolution microscope (e.g., Airyscan, STED), to ensure there were enough continuous signals from the pGk13a probe at a higher resolution. Airyscan was used to image a mouse brain tissue section that underwent umExM protocol, and it was observed that the method indeed provides continuous signals. The method was then combined with fluctuation imaging methods (i.e., SRRF).
  • SRRF fluctuation imaging methods

Landscapes

  • Chemical & Material Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Physics & Mathematics (AREA)
  • Analytical Chemistry (AREA)
  • General Physics & Mathematics (AREA)
  • Pathology (AREA)
  • Organic Chemistry (AREA)
  • Molecular Biology (AREA)
  • Medicinal Chemistry (AREA)
  • Engineering & Computer Science (AREA)
  • Urology & Nephrology (AREA)
  • Biophysics (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Biomedical Technology (AREA)
  • Hematology (AREA)
  • Genetics & Genomics (AREA)
  • Biotechnology (AREA)
  • Cell Biology (AREA)
  • Microbiology (AREA)
  • Food Science & Technology (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

The invention relates, in part, to methods and compounds for ultrastructure membrane expansion microscopy (umExM), which permits high-resolution visualization of membrane ultrastructure using light microscopy.

Description

ULTRASTRUCTURAL MEMBRANE EXPANSION MICROSCOPY RELATED APPLICATIONS This application claims benefit under 35 U.S.C. § 119(e) of U.S. Provisional application serial number 63/520,702 filed August 21, 2023, the disclosure of which is incorporated by reference herein in its entirety. GOVERNMENT SUPPORT This invention was made with government support under 1R01EB024261, 1R01AG070831, 1R01MH123403, 1R01MH124606, U01MH114819 awarded by the National Institutes of Health; and 1122374 awarded by the National Science Foundation. The government has certain rights in the invention. FIELD OF THE INVENTION The invention relates, in part, methods and compounds for ultrastructure membrane expansion microscopy (umExM), which permits high-resolution visualization of membrane ultrastructure using light microscopy. BACKGROUND OF THE INVENTION Neuroscientists have long studied complex neural circuits by examining their molecular features and the ultrastructure context of the neural circuits. Fluorescence microscopy (FM), including super-resolution microscopy, allowed neuroscientists to identify specific biomolecules, while electron microscopy (EM) allowed neuroscientists to reveal ultrastructural context through dense labeling of membranous structures. Both FM and EM greatly advanced fundamental understandings of neurobiology. However, there are clear limitations on each of imaging modalities: FM lacks the ability to provide the biomolecule localization in the ultrastructural context, and EM cannot identify specific biomolecules in the ultrastructural context. Although there has been a long effort to visualize biomolecules of interest on the ultrastructural context with EM to comprehensively study and understand neural circuitries, such as correlative EM (CLEM) technology and gold-nanoparticle labeling, it remains a major technological challenge [Chen, F. et al., Science 347, 543–548 (2015); Chang, J. B. et al. Nature Methods 201714:614, 593–599 (2017)]. Despite many technological challenges, it would be ideal if such imaging could be achieved with ordinary FM (i.e., confocal 1 #17358430v1 microscope), allowing anyone to identify and localize biomolecules of interest in the ultrastructural context of neural circuits, across the large volume, in the thick brain tissue. Expansion microscopy (ExM) [Chen, F. et al., Science 347, 543–548 (2015)] physically magnifies preserved biological specimens by covalently anchoring biomolecules and/or their labels to a swellable polymer network (such as sodium polyacrylate) synthesized in situ throughout a specimen, followed by chemical softening of the sample, and the addition of water to swell the hydrogel. As the hydrogel swells, the anchored biomolecules and/or their labels are pulled apart from each other isotropically, typically to a physical magnification of ~4-10x in each linear dimension. With an iterative form of ExM, [Chang, J. B. et al. Nature Methods 201714:614, 593–599 (2017); Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)] the expanded sample can be expanded a second time, resulting in an overall physical magnification of beyond 10x in each linear dimension. The net result of the expansion is that biomolecules and/or their labels that are initially localized within the diffraction limit of a traditional optical microscope can now be separated in space to distances far enough to resolve them. Expansion microscopy protocols are increasingly prevalent in biology for visualizing proteins [Yu, C. C. et al. Elife 9, 1–78 (2020); Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016); Damstra, H. G. J. et al. Elife 11, (2022); M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022); Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)], nucleic acids [Chen, F. et al. Nat. Methods 13, 679–684 (2016); Alon, S. et al. Science 371, (2021); Cui, Y. et al. PLoS One 18, e0291506 (2023)], and membrane or lipids [Damstra, H. G. J. et al. Elife 11, (2022); M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022); Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023). ExM; Cui, Y. et al. PLoS One 18, e0291506 (2023); White, B. M., et al. Chemrxiv (2022) doi:10.26434/chemrxiv-2022-n4cq4; Sun, D.-E. et al. Nat. Methods 18, 107–113 (2021); Götz, R. et al. Nat. Commun.11, 6173 (2020); Wen, G. et al. ACS Nano 14, 7860–7867 (2020); Wang, U.-T. T. et al. Sci. Rep.13, 21922 (2023)] also enables the visualization of anatomical features of specimens through dense labeling of total protein [Yu, C. C. et al. Elife 9, 1–78 (2020); M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022); Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)] via N- hydroxyl succinimide (NHS) ester staining. While several ExM methods have been reported for membrane or lipid labeling and visualization [Damstra, H. G. J. et al. Elife 11, (2022); M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022); Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023). ExM; Cui, Y. et al. PLoS One 18, e0291506 (2023); White, B. M., et al. Chemrxiv (2022) doi:10.26434/chemrxiv-2022-n4cq4; Sun, D.-E. et al. Nat. 2 #17358430v1 Methods 18, 107–113 (2021); Götz, R. et al. Nat. Commun.11, 6173 (2020); Wen, G. et al. ACS Nano 14, 7860–7867 (2020); Wang, U.-T. T. et al. Sci. Rep.13, 21922 (2023)], achieving dense labeling in fixed tissues has remained challenging. SUMMARY OF ELEMENTS OF INVENTION According to an aspect of the invention, a probe molecule is provided, the probe molecule including: (a) a chain including 2, 3, 4, 5, 6, 7, 8,9, 10, 11, 12, 13, 14, 15 or more lysines and/or cystines; (b) a chemical handle; and (c) a polymer-anchorable handle. In some embodiments, the probe molecule includes (a) a chain including 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more lysines; (b) a chemical handle; and (c) a polymer-anchorable handle. In certain embodiments, one or more of the lysines are D-lysines. In certain embodiments, one or more of the cystines are D-cystines. In some embodiments, the chain includes acrylic acid. In some embodiments, the chemical handle includes azide. In certain embodiments, the chemical handle includes biotin. In certain embodiments, the polymer-anchorable handle is attached to a primary amine of the chain. In certain embodiments, the polymer-anchorable handle includes acryloyl–X (AcX). In some embodiments, the chain includes a lipid tail on the amine terminus of the chain. In some embodiments, the lipid tail includes palmitoyl. In some embodiments, the probe molecule includes: palmitoyl-glycine-acrylic-lysine-lysine- lysine-lysine-lysine-lysines-lysine-lysine-lysine-lysine-lysine-lysine-lysine-azide (SEQ ID NO: 1). In some embodiments, a glycine linker molecule is positioned between the lipid tail and the amine terminus of the lysine chain. In certain embodiments, the lysine chain includes 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, or more lysines. In certain embodiments, the lysine chain includes five lysines. In certain embodiments, the probe molecule includes a glycine and penta-lysine D-peptidic chain, with a palmitoyl lipid group on the amine- terminus and a biotin on the carboxy-terminus, wherein the probe molecule includes: palmitoyl-glycine-lysine-lysine-lysine-lysine-lysine-biotin (SEQ ID NO: 2). In some embodiments, the probe molecule includes pGk5b, pGk5a, pGk13a, or pGk13b. In some embodiments, the probe molecule is one of SEQ ID NOs: 1-16. According to another aspect of the invention, a method for preparing a biological sample for detecting ultrastructural features of a membrane in the biological sample is provided: the method including: (a) contacting the biological sample with a probe of any embodiment the aforementioned probes.; (b) embedding the contacted biological sample within a polymer material; (c) homogenizing the embedded biological sample; and (d) physically expanding the homogenized embedded sample. In certain embodiments, 3 #17358430v1 embedding the biological sample in the polymer material includes incubating the biological sample in a polymer monomer and polymerizing the monomer. In certain embodiments, the polymer material includes a swellable polymer material. In some embodiments, the swellable polymer material includes an acrylamide-co-acrylate copolymer. In certain embodiments, the method also includes polymerizing the polymer material. In some embodiments, the polymerization includes polymerization at approximately 4⁰C for a time period of at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more hours. In some embodiments, the polymer material includes a non-swellable polymer material capable of conversion to a swellable polymer material, and the method further includes converting the polymerized non-swellable polymer material into a swellable polymer material. In certain embodiments, if the polymerized polymer is a swellable polymer, a means for the physical expansion of the biological sample includes contacting the homogenized embedded biological sample with a solvent or liquid that swells the swellable polymer. In some embodiments, the liquid includes water. In some embodiments, homogenizing the embedded biological sample includes contacting the polymer material in which the biological sample is embedded with one or more enzymes. In certain embodiments, the contacting enzyme is an endoproteinase, optionally wherein the endoproteinase is LysC or Trypsin. In certain embodiments, the homogenization includes contacting the embedded biological sample with one or both of LysC and Trypsin. In certain embodiments, the physically expanding of the biological sample includes expanding the polymer material in which the biological sample is embedded, wherein the expansion of the polymer material expands the homogenized biological sample isotropically in at least a linear manner within the polymer material. In some embodiments, the polymer material includes a hydrogel and a means of expanding the hydrogel includes contacting the hydrogel with an aqueous solution, optionally wherein the aqueous solution includes water. In some embodiments, the method also includes attaching a detectable label to the probe molecule at one or both of before and after the physical expansion step. In some embodiments, the method also includes re-embedding the physically expanded polymer and homogenized biological sample in either a swellable or a non-swellable polymer. In some embodiments, the method also includes expanding the embedded sample 1, 2, 3, 4, 5, or more times. In certain embodiments, the method also includes detecting one or more of a spatial position, a structure, a component of, and an identity of the expanded biological sample. In certain embodiments, the method also includes imaging the biological sample after the physical expansion. In some embodiments, the imaging includes optical microscopy. In some embodiments, the microscopy is light microscopy, fluorescence microscopy or electron 4 #17358430v1 microscopy. In some embodiments, the imagining includes optical fluctuation imaging. In certain embodiments, the method also includes detecting one or more proteins and/or polynucleotide molecules in the embedded, expanded sample. In some embodiments, the polynucleotide molecule comprises an RNA molecule. In certain embodiments, the method also includes antibody staining the expanded sample. In some embodiments, biological sample includes a cell or tissue. In some embodiments, the cell or tissue is a fixed cell or tissue. In some embodiments, the cell or tissue is obtained from a subject. In certain embodiments, the subject is a mammal, optionally is a human or a rodent. In certain embodiments, the cell or tissue is a cultured cell or cultured tissue. In some embodiments, the biological sample includes a cell membrane. In some embodiments, the tissue is a tissue slice. In certain embodiments, the tissue includes CNS tissue. In certain embodiments, the tissue is brain tissue. In some embodiments, the tissue slice is between about 50 and 100 microns thick. In some embodiments, the tissue slice is less than 50 microns thick or is more than 100 microns thick. In certain embodiments, the method also includes segmenting cell compartments of the expanded sample, optionally wherein the compartments are cell bodies, dendrites and/or axons in the expanded sample. In certain embodiments, the method also includes tracing cell processes in the expanded sample, optionally wherein the cell processes are myelinated axons and/or unmyelinated axons in the expanded sample. According to yet another aspect of the invention, a method is provided, the method including use of a probe of any embodiment of an aforementioned aspect of the invention in an imaging method. In some embodiments, the imaging method includes imaging an ExM- processed biological sample. In certain embodiments, the imaging method comprises light microscopy. In some embodiments, the imaging method comprises confocal microscopy. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1A-B provides schematic diagrams illustrating ultrastructural membrane expansion microscopy (umExM) concept and workflow. umExM is a modified form of expansion microscopy with a custom-designed amphiphilic membrane labeling probe (termed pGk13a). Fig.1A shows chemical structure of pGk13a. The probe does not contain any fluorophore but has an azide to bind a fluorophore later. Fig.1B is a schematic diagram of umExM workflow. Lighter text highlights key differences from ExM [Chen, F. et al., Science 347, 543–548 (2015)] and proExM [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)], whereas darker fine text highlight the same steps as ExM and proExM. In Fig.1B panel (i) a specimen is perfused and chemically fixed with 4% paraformaldehyde (PFA) + 5 #17358430v1 0.5% calcium chloride (CaCl2) at 4 ⁰C for 24 hours. The brain is sliced on a vibratome to 100 μm thickness at 0-4 ⁰C. In Fig.1B panel (ii) the specimen is treated with pGk13a (structure is depicted in (Fig.1A) at 4 ⁰C overnight (unless otherwise noted, overnight means >16 hours). In Fig.1b panel (iii) the specimen is treated with acrylic acid N-hydroxysuccinimide ester (AX) at 4 ⁰C overnight. In Fig.1 B panel (iv) the specimen is embedded in an expandable hydrogel (made with N,N'-Diallyl-L-tartardiamide (DATD) crosslinker [Yu, C. C. et al. Elife 9, 1–78 (2020)]) at 4 ⁰C for at least 24 hours. In Fig.1B panel (v) the sample (specimen- embedded hydrogel) is chemically softened with enzymatic cleavage of proteins (i.e., non- specific cleavage with proteinase K) at room temperature (~24 ⁰C) overnight. The probe is not digested during proteinase K treatment because it is composed of D-amino acids. In Fig. 1B panel (vi) the sample is treated with 1x phosphate-buffered saline (PBS) to partially expand it. The probe, pGk13a, that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight. In fig.1B panel (vii) the sample is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes). Figure 2A-V provides photomicrographic images and graphs illustrating embodiments of resolution and distortion of umExM. Fig.2A provides representative (n=3 cells from one culture) single z-plane structured illumination microscopy (SIM) image of a pre-expanded HEK293 cell expressing mitochondrial matrix-targeted green fluorescent protein (GFP, shown). Fig.2B provides a single z-plane confocal image of the same HEK293 cell as in Fig. 2A, after undergoing the umExM protocol, showing expression of mitochondrial matrix- targeted GFP in the same field of view as shown in Fig.2A. GFP shown. Fig.2C provides a single z-plane confocal image of the same umExM-expanded fixed HEK293 cell as in Fig. 2A, showing pGk13a staining of the membrane in the same field of view as shown in Fig. 2A. pGk13a staining shown. Fig.2D is a graph showing root-mean-square (RMS) length measurement error as a function of measurement length, comparing pre-expansion SIM images of fixed HEK293 cells expressing mitochondrial matrix-targeted GFP and post- expansion confocal images of the same cells after umExM processing, showing mitochondrial matrix-targeted GFP (blue line, mean; shaded area, standard deviation; n=3 cells from one culture). Fig.2E, as in Fig.2D, but with post-expansion images showing pGk13a staining of the membrane. Fig.2F is a boxplot showing measured expansion factor as described (n=8 pairs of landmark points; from 3 fixed brain slices from two mice; median, middle line; 1st quartile, lower box boundary; 3rd quartile, upper box boundary; error bars 6 #17358430v1 are the 95% confidence interval; black points, individual data points; used throughout this manuscript unless otherwise noted). Fig.2G is a boxplot showing resolution of post- expansion confocal images (60x, 1.27NA objective) of umExM-processed mouse brain tissue slices showing pGk13a staining of the membrane (n=5 fixed brain slices from two mice). Fig. 2H shows representative (n=5 fixed brain slices from two mice) single z-plane confocal image of expanded Thy1-YFP mouse brain tissue (hippocampus, dentate gyrus) showing pGk13a staining of the membrane. pGk13a staining of the membrane visualized in inverted color throughout this figure (dark signals on light background) except for Fig.2S. Fig.2I shows magnified view of black boxed region in Fig.2H. Fig.2J, as in Fig.2H, but imaging of the third ventricle. Fig.2K as in Fig.2H but imaging of mouse somatosensory cortex layer 6 (L6). Fig.2L shows magnified view of black boxed region in Fig.2K. Fig.2M shows representative (n=2 fixed brain slices from two mice) single z-plane confocal image of expanded Thy1-YFP mouse brain tissue (hippocampus dentate gyrus), that underwent umExM protocol and anti-GFP labeling (here labeling YFP), showing YFP and pGk13a staining of the membrane (inverted gray). Fig.2N shows diameter of unmyelinated axons (n=17 axons from three fixed brain slices from two mice). Fig.2O, as in Fig.2M, but imaging of somatosensory cortex L6 that was used for measuring the diameter of myelinated axons. Fig.2P shows diameter of myelinated axons (n=21 axons from two fixed brain slices from two mice). Fig.2Q as in Fig.2M but imaging of the third ventricle that was used for measuring the diameter of cilia. Fig.2R shows diameter of cilia (n=19 cilia from two fixed brain slices from two mice). Fig.2S left panel shows representative (n=4 slices of fixed brains from three mice) volume rendering of epithelial cells in the third ventricle from mouse brain tissue, showing pGk13a staining of the membrane. pGk13a staining of the membrane visualized in gray color. Fig.2S right panel shows magnified view of boxed region in left panel. Arrows indicate putative extracellular vesicles. Serial image sections that were used for the 3D rendering are in Fig.23. Fig.2T shows single z-plane confocal image of expanded mouse brain tissue (third ventricle) processed by umExM, showing pGk5b staining (gray), focusing on the plasma membrane of cilia (i.e., ciliary membrane). Fig.2U shows transverse profile of cilia in the dotted boxed region in Fig.2U after averaging down the long axis of the box and then normalizing to the peak of pGk13a signal. Fig.2V is a boxplot showing the percent continuity of the membrane label (n=5 separate cilia from two fixed brain slices from one mouse), where a gap was defined as a region larger than the resolution of the images (~60 nm, from Fig.2G), over which the pGk13a signal was two standard deviations below the mean of the intensity of pGk13a along the ciliary membrane. Scale bars are provided in 7 #17358430v1 biological units throughout all figures (i.e., physical size divided by expansion factor): Fig. 2A, B, & D 5μm; Fig.2H 5μm; Fig.2I 2μm; Fig.2J 5μm; Fig.2K 5μm; Fig.2L 5μm; fig. 2M 0.25 μm; Fig.2O & Q 1μm; Fig.2S left panel (x); 13.57μm (y); and 7.5μm (z); Fig.2S right panel 3.76μm (x); 3.76μm (y); 1.5μm (z; Fig.2T 2μm. Figure 3A-O shows photomicrographic images of umExM with antibody staining and RNA fluorescence in situ hybridization (FISH). Fig.3A provides representative (n=5 slices of fixed brain from two mice) single z-plane confocal image of expanded mouse brain tissue (hippocampus, CA3) after umExM processing with a pre-expansion antibody staining protocol (Fig.10), showing immunostaining with an antibody against the synaptic vesicle protein SV2A. Fig.3B provides magnified view of the box in Fig.3A. Fig.3C shows single z-plane confocal image of the specimen of Fig.3A, showing pGk13a staining of the same field of view as in Fig.3A. pGk13a staining of the membrane visualized in inverted gray color throughout this figure. Fig.3D shows magnified view of the box in Fig.3C. Fig.3E shows an overlay of Fig.3A and Fig.3C. Fig.3F shows a magnified view of the box in Fig. 3E. Fig.3G shows representative (n=5 slices of fixed brain from two mice) single z-plane confocal image of expanded mouse brain tissue (hippocampus, CA1) after umExM processing with a post-expansion antibody staining protocol (Fig.11), showing immunostaining against the post-synaptic density protein PSD-95. Fig.3H shows magnified view of the box in Fig.3G. Fig.3I shows single z-plane confocal image of the specimen of Fig.3G, showing pGk13a staining of the same field of view as in Fig.3G. Fig.3J shows magnified view of the box in Fig.3I. Fig.3K shows overlay of Fig.3G and Fig.3I. Fig.3L shows magnified view of the box in Fig.3C. The examples of PSD95 signals that were aligned with pGk13a signals were pinpointed with arrows. Fig.3M shows representative (n=3 slices of fixed brain from one mouse) single z-plane confocal image of expanded mouse brain tissue (hippocampus, CA1) after umExM processing with a FISH protocol (Fig.12), showing HCR-FISH targeting ACTB. Fig.3N shows single z-plane confocal image of the specimen of Fig.3J, showing pGk13a staining of the same field of view as in Fig.3J. Fig.3O shows an overlay of Fig.3M and Fig.3N. Scale bars: (Fig.3A, B, C, G, H, & J) 5 μm; Fig. 3D, E, F, J, K, & L 1μm; Fig.3M, N, and O 20 μm. Figure 4A-L provides photomicrographic images and graphs showing results of certain embodiments of methods of the invention. Fig.4A shows representative (n=12 cells from two cultures) single z-plane confocal image of expanded HEK293 cell that underwent the 8 #17358430v1 mExM protocol, showing pGk5b staining of the membrane. Image visualized in inverted gray color (dark signals on light background). Fig.4B shows representative (n=12 cells from two cultures) single z-plane confocal image of expanded HeLa cell that underwent the mExM protocol, showing pGk5b staining of the membrane. Image visualized in inverted gray color. Fig.4C shows representative (n=3 cells from one culture) single z-plane structured illumination microscopy (SIM) image of a pre-expanded U2OS cell expressing mitochondrial matrix-targeted green fluorescent protein (GFP). Fig.4D shows single z-plane confocal image of the same U2OS cell as in Fig.4C, after undergoing the mExM protocol, showing expression of mitochondrial matrix-targeted GFP in the same field of view as shown in Fig. 4C. Fig.4E shows non-rigidly registered and overlaid pre-expansion SIM image of the U2OS cell expressing mitochondrial matrix-targeted GFP in Fig.4C, and post-expansion confocal image of the same fixed U2OS cell after mExM processing, showing the mitochondrial matrix-targeted GFP channel in Fig.4D. fig.4F shows root mean square (RMS) length measurement error as a function of measurement length, comparing pre-expansion SIM images of fixed U2OS cell expressing mitochondrial matrix-targeted GFP and post-expansion confocal images of the same cells after mExM processing, showing mitochondrial matrix- targeted GFP (blue line, mean; shaded area, standard deviation; n=3 cells from one culture). Fig.4G shows single z-plane confocal image of the same mExM-expanded fixed U2OS cell as in Fig.4D, showing pGk5b staining of the membrane in the same field of view as shown in Fig.4C. Fig.4H shows non-rigidly registered and overlaid pre-expansion SIM image of the U2OS cell expressing mitochondrial matrix-targeted GFP in Fig.4C and post-expansion confocal image of the same U2OS cell in Fig.4E after mExM processing, showing pGk5b staining. Fig.4I shows root mean square (RMS) length measurement error as a function of measurement length, comparing pre-expansion SIM images of fixed U2OS cells expressing mitochondrial matrix-targeted GFP, and post-expansion confocal images of the same cell after mExM processing, showing pGk5b staining (line, mean; shaded area, standard deviation; n=3 cells from one culture). Fig.4J-K shows results of expansion factor analysis on HEK293 cells, which underwent mExM, after expressing mitochondrial matrix-targeted GFP. Two landmark points in pre-expansion images were randomly selected and the corresponding landmarks were found in expanded sample images. The distance the distance between the points was then calculated, in both pre- and post-expansion images, and the ratio was calculated to obtain the expansion factor. Fig.4J shows representative (out of 10 cells from two cultures) single z-plane confocal image of pre-expanded HEK293 cell. Fig.4K, as in Fig.4J, but post-expansion, for the same field of view shown in Fig.4J. Fig.4L provides a 9 #17358430v1 boxplot showing measured expansion factor (n =10 cells from two cultures; black points, individual measured expansion factor, median, middle line; 1st quartile, lower box boundary; 3rd quartile, upper box boundary; error bars are the 95% confidence interval). Scale bars: Fig. 4A-B 5 μm; fig.4C, D, E, G, H, & J 2 μm in biological units; Fig.4K 10μm in post- expansion units. Figure 5A-C provides photomicrographic images and a graph. Fig.5A panel (i) shows representative (n=4 cells from one culture) single z-plane confocal image of expanded HEK293 cell expressing mitochondrial matrix-targeted GFP, after mExM processing. Fig.5A panel (ii) shows single z-plane confocal image of the same expanded HEK293 cell as in Fig. 5 panel (i), showing pGk5b staining in the same field of view as in Fig.5 panel (i). Fig.5 panel (iii) provides an overlay of Fig.5 panel (i) and panel (ii). Fig.5B, as in Fig.5A, but for a HEK293 cell expressing ER membrane-targeted GFP (n=3 cells from one culture). Fig.5C shows fraction of the pixels containing mitochondrial matrix-targeted GFP signal (left, n=3 cells from one culture; black points, individual measured fractions of expressed organelle targeted GFP that contained pGkb5 signal; median, middle line; 1st quartile, lower box boundary; 3rd quartile, upper box boundary; error bars are 95% confidence interval, used throughout unless otherwise noted) or ER membrane-targeted GFP signal (right, n=3 cells from one culture) that also exhibited pGk5b signal, in mExM-processed HEK293 cells. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig.5A- B 5 μm. Figure 6 provides photomicrographic images showing effect of the glycine linker attached to the palmitoyl group on the efficacy of membrane labeling in fixed brain tissue. Image visualized in inverted gray color. Two versions of the palmitoylated 5 lysine biotin membrane probe were tested. Fig.6 Left panel shows results with a probe that contained a glycine linker attached to the palmitoyl group enabling flexibility of the lipid relative to the peptide carrier. Fig.6 Right panel shows results with a probe that did not contain a glycine but in which the lipid was directly attached to the lysine backbone. In the case of the glycine linker, the level of detail achieved in labeling membranes was superior to that achieved without the glycine linker. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): 5 μm. 10 #17358430v1 Figure 7 provides photomicrographic images of results. Fig.7 panel (i) shows representative (n=5 fixed brain slices from two mice) single z-plane confocal image of pre-expanded mouse brain tissue (cortex) showing pGk5b staining. Images visualized in inverted gray color throughout this figure (dark signals on light background). Fig.7 panel (ii) shows representative (n=5 fixed brain slices from two mice) single z-plane confocal image of expanded mouse brain tissue after mExM processing, showing pGk5b staining. Fig.7 panel (iii) as in Fig.7 panel (i) but focused on one cell body. Fig.7 panel (iv), as in Fig.7 panel (ii) but focused on one cell body. Note that the pre-expansion images of mouse brain tissue in Fig.7 contain native lipids, which were not removed before imaging. Accordingly, the pre- expansion sample exhibits substantial light scattering and a mismatch in refractive index, which significantly impacts image contrast20. Meanwhile, both light scattering as well as mismatch in refractive index are ameliorated by ExM7. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig.7 panel (i) 10 μm, panel (ii) 10 μm, panel (iii) 5 μm, panel (iv) 5 μm. Figure 8A-G provides photomicrographic images and graphs of higher resolution umExM. Fig.8A shows representative (n=5 fixed brain slices from 2 mice) single z-plane confocal image of post-expansion mouse brain tissue (Somatosensory cortex, L4) that underwent the umExM protocol. Images were taken at 50ms/frame for 20 frames with a confocal microscope with 1.5x optical zoom. pGk13a staining of the membrane visualized in inverted gray color throughout this figure. Fig.8B shows average z-projection of images in Fig.8A. Fig.8C shows fluctuations in the acquired frames (as in Fig.8A) were resolved with the ‘super-resolution imaging based on autocorrelation with a two-step deconvolution’ (SACD) algorithm [Zhao, W. et al. Nat. Photonics 1–8 (2023)]. Fig.8D provides a boxplot showing resolution of post-expansion confocal images (60x, 1.27NA objective) of umExM + SACD- processed mice brain tissue slices showing pGk13a staining of the membrane (n=5 fixed brain slices from two mice). Fig.8E shows representative (n=6 fixed brain slices from one mouse) single z-plane confocal image of post-expansion mouse brain tissue (Somatosensory cortex, L4) after the iterative form of umExM processing (Fig.31), showing pGk13a staining of the membrane. Fig.8F shows magnified view of box in Fig.8E. Fig.8G, as in Fig.8D but for the iterative form of umExM (n=6 fixed brain slices from one mouse). Scale bars: Fig. 8B-C 10μm; Fig.8E-F 1.5μm. 11 #17358430v1 Figure 9 shows photomicrographic images of five examples of single z-plane confocal images of post-expansion mouse brain tissue (somatosensory cortex, L4) after the iterative form of umExM processing, showing pGk13a staining of the mitochondrial cristae. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): 0.2 μm. Figure 10 provides a workflow schematic diagram of umExM with pre-expansion antibody staining. Lighter captions highlight the key differences from ExM [M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022)] and proExM [Klimas, A. et al. Nature Biotechnology 202341:6 41, 858–869 (2023)], whereas darker captions highlight steps similar to those of earlier protocols. In Fig.10 panel (i) a specimen is chemically fixed with 4% PFA + 0.5% CaCl2 at 4 ⁰C for 24 hours. The brain is sliced on a vibratome at 100 μm thickness at 4 ⁰C. In Fig.10 panel (ii) the specimen is treated with either 0.005-0.01% detergent (i.e., saponin or triton) at 4 ⁰C overnight (unless otherwise noted, overnight means >16 hours throughout the figure). Then specimen is incubated with a primary antibody. In Fig.10 panel (iii) the specimen is treated with pGk13a at 4 ⁰C overnight. In Fig.10 panel (iv) the specimen is treated with AX at 4 ⁰C overnight. In Fig.10 panel (v) the specimen is embedded in an expandable hydrogel (made with DATD crosslinker [Boutin, J. A. Cell. Signal.9, 15–35 (1997)]) at 4 ⁰C overnight. In Fig.10 panel (vi) the sample (specimen-embedded hydrogel) is chemically softened with enzymatic cleavage of proteins (i.e., non-specific cleavage with proteinase K) at room temperature (~24 ⁰C), overnight. In Fig.10 panel (vii) the sample is then treated with PBS to partially expand it. The pGk13a, that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight. In Fig.10 panel (viii) the sample is then incubated with a secondary antibody at 4 ⁰C for 2-3 days. In Fig.10 panel (ix) the sample is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes). Figure 11 provides a workflow schematic diagram of umExM with post-expansion antibody staining. Light-colored captions highlight the key differences from ExM [M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022)] and proExM [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)], whereas dark-colored captions highlight steps similar to those of ExM and proExM. In Fig.11 panel (i) a specimen is chemically fixed with 4% PFA + 0.5% CaCl2 at 4 ⁰C for 24 hours. The brain is sliced on a vibratome at 100μm thickness at 4 ⁰C. Fig.11 panel (ii) the specimen is treated with pGk13a at 4 ⁰C overnight (unless otherwise noted, overnight means >16 hours throughout the figure). In Fig.11 panel (iii) the specimen 12 #17358430v1 is treated with AX at 4 ⁰C overnight. In Fig.11 panel (iv) the specimen is embedded in an expandable hydrogel (made with DATD crosslinker [Boutin, J. A. Cell. Signal.9, 15–35 (1997)]) at 4 ⁰C overnight. In Fig.11 panel (v) the sample (specimen-embedded hydrogel) is mechanically softened with enzymatic cleavage of proteins (i.e., specific cleavage with Trypsin and LysC) at room temperature (~24 ⁰C), overnight. In Fig.11 panel (vi) the sample is then treated with PBS to partially expand it. The pGk13a, that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight. In Fig.11 panel (vii) the sample is then incubated with a primary antibody at ~4 ⁰C, for 48-72 hours. In Fig.11 panel (vii) the sample is then incubated with a secondary antibody at ~4 ⁰C, for 48-48 hours. In Fig.11 panel (ix) the sample is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes). Figure 12 provides a workflow schematic diagram of umExM with FISH staining. Light- colored captions highlight the key differences from ExM [M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022)], proExM [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)] and ExFISH [Sarrazin, S. et al., Cold Spring Harb. Perspect. Biol.3, 1–33 (2011)], whereas dark-colored captions highlight steps similar to those of ExM, proExM and ExFISH. In Fig.12 panel (i) a specimen is chemically fixed with 4% paraformaldehyde (PFA) + 0.5% calcium chloride (CaCl2) at 4 ⁰C for 24 hours. The brain is sliced on a vibratome at 100 μm thickness at 4 ⁰C. In Fig.12 panel (ii) the specimen is treated with the pGk13a at 4 ⁰C overnight (unless otherwise noted, overnight means >16 hours throughout the figure). In Fig.12 panel (iii) the specimen is treated with acrylic acid N-hydroxysuccinimide ester (AX) at 4 ⁰C overnight. In Fig.12 panel (iv) the specimen is embedded in an expandable hydrogel (made with DATD crosslinker [Wen, G. et al. ACS Nano 14, 7860–7867 (2020)]) at 4 ⁰C for at least 24 hours. In Fig.12 panel (v) the specimen is mechanically softened with enzymatic cleavage of proteins (i.e., specific cleavage with Proteinase k) at room temperature (~24 ⁰C), overnight. In Fig.12 panel (vi) the specimen-embedded hydrogel is treated with PBS to partially expand it. The pGk13a, that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight. In Fig. 12 (vii) the specimen-embedded hydrogel is incubated with HCR-FISH probe at 37 ⁰C, overnight. In Fig.12 panel (vii) the specimen-embedded hydrogel is incubated with fluorescently labeled HCR-hairpin amplifiers at ~24 ⁰C, overnight. In Fig.12 panel (viii) the specimen-embedded hydrogel is expanded with 0.05x SSCT at room temperature for 1.5 hours (exchanging water every 30 minutes). 13 #17358430v1 Figure 13A-C provides photomicrographic images and a graph illustrating the performance of a biotin handle (pGk13b) vs. azide handle (pGk13a) with fixed mouse brain tissues in the context of ExM. Mouse brain tissues were fixed with ice-cold 4% PFA. pGk13b or pGk13a was applied overnight, and the standard ExM protocol [Yu, C. C. et al. Elife 9, 1–78 (2020)] was then followed. In short, the tissues were processed with AcX, and the ExM gel was formed. After the tissue softening with proteinase K, the probe was fluorescently labeled with fluorescent streptavidin (i.e., cy3-streptavidin, >1 fluorophore per streptavidin), expanded, and imaged (results show in Fig.13A) or with fluorescent DBCO (i.e., cy3-DBCO, 1 fluorophore per DBCO), expanded and imaged (results shown in Fig.13B). Fig.13A shows representative (n = 6 fixed brain slices from two mice) single z-plane confocal image of expanded mouse brain tissue (hippocampus), showing the pGk13b staining of the membrane. The image is visualized in inverted gray color. Fig.13B is as in Fig.13A but with pGk13a, showing the pGk13a staining of the membrane. Images in Fig.13A-B were visualized with the same brightness and contrast with ImageJ software to highlight the difference between the two images. Fig.13C shows the signal intensity of the pGk13b (left; n=6 fixed brain slices from two mice) and the pGk13a (right; n=6 fixed brain slices from two mice). Black points, individual measured average intensity of each image; median, middle line; 1st quartile, lower box boundary; 3rd quartile, upper box boundary; error bars are the 95% confidence interval; p-value, unpaired two-sided t-test between signals from the pGk13b (left) and the pGk13a handle probe (right). Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig.13A-B 20 μm. Figure 14 provides a provides a workflow schematic diagram of umExM with double gelation. Lighter-colored captions highlight the key differences from ExM [M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022)] and proExM [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)], whereas darker-colored captions highlight steps similar to those in ExM and proExM. In Fig.14 panel (i) a specimen is chemically fixed with 4% paraformaldehyde (PFA) + 0.5% calcium chloride (CaCl2) at 4 ⁰C for 24 hours. The brain is sliced on a vibratome at 100 μm thickness at 4 ⁰C. In Fig.14 panel (ii) the specimen is treated with pGk13a at 4 ⁰C overnight (unless otherwise noted, overnight means >16 hours throughout the figure). In Fir.14 panel (iii) the specimen is treated with acrylic acid N- hydroxysuccinimide ester (AX) at 4 ⁰C overnight. In Fig.14 panel (iv) the specimen is embedded in an expandable hydrogel (made with cleavable crosslinker DATD) at 4 ⁰C for at 14 #17358430v1 least 24 hours. In Fig.14 panel (v) the specimen is mechanically softened with enzymatic cleavage of proteins (i.e., specific cleavage with proteinase-k) at room temperature (~24 ⁰C), overnight. In Fig.14 panel (vi) the specimen-embedded hydrogel is treated with PBS to partially expand it. Next, the sample is treated with AX as was done in Fig.14 panel (iii). Subsequently, the sample was gelled again but with a monomer solution that contains the non-cleavable crosslinker N,N -Methylenebis(acrylamide) (BIS) at room temperature (~24 ⁰C), overnight. In Fig.14 panel (vii) the specimen-embedded hydrogel is incubated in a gel cleaving solution (containing sodium metaperiodate) at room temperature (~24 ⁰C) for 1 hour. This step cleaves the initial gel that was formed in Fig.14 panel (iv). In Fig.14 panel (viii) the pGk13a, that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight. The specimen-embedded hydrogel is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes). The modified protocol requires additional time and resources (i.e., necessitating two gelations) compared to the unmodified protocol, which is suitable for all these regions except the corpus callosum. Hence, this modified protocol is suggested specifically for the corpus callosum region. Figure 15 provides a photomicrographic image showing a representative (n=2 fixed brain slices from one mouse) single z-plane confocal image of expanded mouse brain tissue that underwent umExM protocol with the GMA anchor instead of AX anchor, showing pGk13a staining of the membrane in the hippocampus dentate gyrus. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): 10 μm. Figure 16A-D provides photomicrographic images. Fig.16A shows representative (n=5 brain tissue sections from two mice) single z-plane confocal image of expanded mouse brain tissue after umExM protocol processing, showing pGk13a staining of the membrane in the corpus callosum region. The image is visualized in inverted gray color. Fig.16B shows magnified view of boxed region in Fig.16A. Only a subset of axons can be identified in the images. Fig.16C shows representative (n=5 brain tissue sections from 2 mice) single z-plane confocal image of expanded mouse brain tissue after modified umExM protocol processing, showing pGk13a staining of the membrane in the corpus callosum region. The image is visualized in inverted gray color. Fig.16D shows magnified view of boxed region in Fig. 16C. The modification of the protocol drastically improved the visualization of axons in the 15 #17358430v1 corpus callosum. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig.16A & C 10 μm; Fig.16B & D 2 μm. Figure 17A-B provides electron microscopy images of membrane label (pGk5a)-stained mouse brain slices (hippocampus region). In brief, 100 μM of palmitoylated glycine pentalysine peptide, equipped with an azide group (instead of biotin; termed pGk5a), without osmium counterstain (in Fig.17A), or no membrane probe but with osmium tetroxide counterstain in Fig.17B), was applied to 100-μm thick tissue slices. Mouse brain tissue was preserved in 4% PFA and 0.1% glutaraldehyde at 4 ⁰C, and labeled with pGk5a for >16 hours at 4 ⁰C. The tissue was post-fixed in 2% PFA and 2% glutaraldehyde and labeled with 1.8nm undecagold gold nanoparticles, conjugated to dibenzocyclooctyne to attach to the azide handle on pGk5a. The tissue was counter-labeled with uranyl acetate, embedded in resin, sliced, and imaged on a TEM scope. Because the common practice of uranyl acetate (UA) staining without osmium does not clearly visualize the membrane, a common UA staining protocol (1% UA for 1 hour at room temperature) was used to enhance pGk5a signals on top of signals from gold nanoparticles. As a control, the tissue underwent the same protocol as described above, but without membrane probe incubation and with osmium. When labeling membranes with the membrane probe in the absence of the osmium counterstain, the stain appeared to label the membranes as with osmium, but with slightly lower contrast. Scale bars: 1 μm. Figure 18A-D provides images and a graph of results of certain studies described herein. Fig. 18A panel (i) shows representative (n=5 cortex or hippocampus regions from the same mouse brain) single z-plane confocal image of expanded mouse brain tissue after mExM processing with immunostaining, showing immunolabeling against the mitochondrial membrane protein Tom20, using antibodies from two separate vendors (see Table 4 for details). Fig.18 panel (ii) shows single z-plane confocal image of the specimen of Fig.18 panel (i), showing pGk5b staining of the same field of view as in Fig.18 panel (i). Fig.18 panel (iii) shows an overlay of Fig 18 panel (i) and panel (ii). Fig.18 panel (iv) shows magnified views of white boxed regions in Fig.18 panel (iii) showing TOM20 signal. Fig.18 panel (vi) as in Fig.18 panel (iv) but showing pGk5b signal. Fig.18 panel (vii) shows overlay of Fig.18 panel (iv) and panel (v). In Fig.18B, as in Fig.18A, but for the nuclear pore protein NUP98. Fig.18C as in Fig.18A, but for myelin basic protein (MBP), using antibodies from three separate vendors (see Table 4 for details). Fig.18D shows fraction of the pixels containing each membrane 16 #17358430v1 protein signal, that also contained pGk5b signal, in mExM-processed mouse brain tissue, for each of the antibodies used above (see Example 1, methods; for each box, n=5 cortex or hippocampus regions from the same mouse brain; black points, individual measured fraction of expressed membrane-localized proteins that contained pGk5b signal; median, middle line; 1st quartile, lower box boundary; 3rd quartile, upper box boundary; error bars are the 95% confidence interval). Scale bars: Fig.18 panels (i) – (iii) 10μm, Fig.18 panel (iv) –(vi) 1μm. Figure 19A-D provides photomicrographic images of results from tests of two versions of the membrane labeling probe with ExM: Fig.19A shows results using palmitoyl-glycine-(D- lysine)5-biotin (pGk5b), and Fig.19B shows result using farnesyl-glycine-(D-lysine)5-biotin (i.e., replacing palmitoyl in pGk5b with farnesyl, fGk5b), as well as (Fig.19C-D) a mixture of pGk5b+fGk5b at varying concentrations .Fig.19A shows representative (n=5 fixed brain slices from two mice) single z-plane confocal image of expanded mouse brain tissue (hippocampus) after mExM processing, showing pGk5b (i.e., the palmitoylated form of the membrane labeling probe; used at 10mM) staining of the membrane. The image is visualized in inverted gray color. Fig.19B as in Fig.19A, but with a farnesylated form of the membrane labeling probe (used at 10mM, n=5 fixed brain slices from two mice). Fig.19C as in Fig. 19A, but with a mixture of 5mM pGk5b + 5mM fGk5b (n=2 fixed brain slices from two mice). Fig.19D as in Fig.19A, but with a mixture of 10mM pGk5b + 10mM fGk5b (n=2 fixed brain slices from two mice). Images Fig.19A-D are visualized with the same brightness and contrast with ImageJ software. Scale bars are provided in biological units: Fig.19A-D 5 μm. Figure 20A-G provides photomicrographic images of results of testing varying numbers of lysines (i.e., 3, 5, 7, 9, 11, 13, and 19 lysines) in the backbone of the membrane labeling probe while holding other moieties (palmitoyl tail, glycine, and biotin) constant. Fig.20A shows representative (n=3 fixed brain slices from two mice) single z-plane confocal image of expanded mouse brain tissue (hippocampus) after mExM processing, but with a membrane labeling probe containing 3 lysines, showing the probe staining of the membrane. The image is visualized in inverted gray color. Fig.20B-G as in Fig.20A but with membrane labeling probes containing 5, 7, 9, 11, 13, and 19 lysines, respectively. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig.20A-G 2 μm. 17 #17358430v1 Figure 21A-F provides photomicrographic images of certain experimental results. Fig.21A shows representative (n=5 fixed brain slices from two mice) single z-plane confocal image of expanded Thy1-YFP mouse brain tissue (hippocampus) after mExM processing using pGk13b and stained with an anti-GFP antibody to boost the YFP signal, showing the pGk13b staining of the membrane. The image is visualized in inverted gray color. Fig.21B shows single z-plane confocal image of the specimen of Fig.21A showing anti-GFP staining of the same field of view as in Fig.21A. Fig.21C is an overlay of Fig.21A and Fig.21B. Fig.21D shows magnified views of boxed region in Fig.21A. Fig.21E shows magnified views of boxed region in Fig.21B. Fig.21F shows magnified views of boxed region in Fig.21C. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig. 21A-C 5μm, Fig.21D-F 0.5μm. Figure 22A-C provides photomicrographic images and a graph of results of pGk13a staining of the membrane of mouse brain tissue fixed with 4%PFA and 0.5% CaCl2 at 4 ⁰C and proceeded with the standard ExM protocol (37 ⁰C gelation) [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)] vs. modified ExM protocol (i.e., 4 ⁰C gelation). Mouse brain tissues were fixed with ice-cold 4% PFA and 0.5% CaCl2 fixatives. The pGk13a probe was applied overnight at 4 ⁰C. The standard ExM protocol or modified ExM protocol (i.e., 4 ⁰C gelation instead of 37 ⁰C gelation) was then performed. In short, the tissues were processed with AcX, and ExM gel was formed at 37 ⁰C (see Fig.22A) or 4 ⁰C (see Fig.22B). After the tissue softening with proteinase K, the pGk13a was fluorescently labeled with fluorescent DBCO (i.e., cy3-DBCO, 1 fluorophore per DBCO), expanded, and imaged. Fig.22A shows representative (n=6 fixed brain slices from two mice) single z-plane confocal image of expanded mouse brain (hippocampus) tissue showing pGk13a staining of the membrane after the standard ExM protocol. The image is visualized in inverted gray color. Fig.22B as in Fig.22A but with 4 ⁰C gelation. Images Fig.22A-B were visualized with the same brightness and contrast with ImageJ software to highlight the difference between the two images. Fig.22C is graph showing the intensity of 37 ⁰C gelation (left; n=6 fixed brain slices from 2 mice) and 4 ⁰C gelation (right; n=6 fixed brain slices from two mice). Black points, individual measured average intensity of each image; median, middle line; 1st quartile, lower box boundary; 3rd quartile, upper box boundary; error bars are the 95% confidence interval; p-value, unpaired two-sided t-test between signals from the 37 ⁰C gelation (left bar), and 4 ⁰C gelation (right bar). Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig.22A-B 20 μm. 18 #17358430v1 Figure 23A-B provides photomicrographic images with Fig.23A showing representative (n=2 fixed brain slices from two mice) single z-plane confocal image of expanded mouse brain tissue (CA1), that underwent the umExM protocol, but without pGk13a staining of the membrane, showing DBCO-Cy3 staining of the membrane (inverted gray). Fig.23B an in Fig.23A but with pGk13a staining of the membrane. Images Fig.23A-B were taken under identical optical conditions and visualized with the same brightness and contrast with ImageJ software. Scale bars are provided in biological units: Fig.23A-B 40μm. Figure 24 shows images of fifteen serial sections from the 3D volume rendering in Fig.2S, right. The arrows indicate membrane vesicles. Scale bar in biological units (i.e., physical size divided by expansion factor): 1 μm. Figure 25 shows a representative (n=5 fixed brain slices from two mice) single z-plane confocal image of expanded mouse brain tissue after umExM processing, showing the pGk13a staining of the membrane in the choroid plexus region. The image is visualized in inverted gray color. Example microvilli in the choroid plexus are pointed at with an arrow. Scale bar in biological units (i.e., physical size divided by expansion factor): 10 μm. Figure 26A-H provides photomicrographic images and graphs of results of certain studies. Fig.26A shows photograph of a fixed 100μm thick adult mouse coronal slice that underwent the umExM protocol. Fig.26B shows single z-plane confocal image of green boxed region in Fig.26A. Images were taken with a 4x objective at 30ms laser exposure time, and they were stitched with shading correction function via default setting from Nikon Element software version 4.30. pGk13a staining of the membrane visualized in inverted gray color throughout this figure (dark signals on light background). No image processing (e.g., denoising or deconvolution) other than stitching was performed for images presented throughout this figure. Fig.26C is a volume rendering of the box (i) in Fig.26B. Images were taken with a 4x objective at 50ms laser exposure time with a z step size of 0.375μm (in biological unit). Unless otherwise noted, clipping planes that are red colored indicate the portion that has been clipped out to expose the inside of the volume for 3D images presented throughout this figure. Fig.26D shows profile of mean pGk13a signal intensity of XY planes taken along the depth of the volume in Fig.26C. Fig.26E is a volume rendering of the white box (ii) in Fig. 26B. Images were taken with a 60x objective at 100ms laser exposure time with a z step size 19 #17358430v1 of 0.125μm. Fig.26F shows magnified view of green boxed region in Fig.26E. Fig.26G shows profile of mean pGk13a signal intensity of XY planes taken along the depth of the volume in Fig.26E. Fig.26H shows cross-sectional images of dentate gyrus region in 100- μm thick mouse coronal slices that underwent the umExM protocol, showing pGk13a staining of the membrane. Images were taken with a 60x objective at 100ms laser exposure time with a z step size of 0.075μm. Lines indicate the cross-sectional views in y-z and x-z images. Scale bars are provided in biological units (i.e., physical size divided by expansion factor) Fig.26B 500μm; Fig.26C 340μm (x); 340μm (y); and 100μm (z); Fig.26E 62μm (x); 62μm (y); and 20μm (z); Fig.26H 5μm (x-y); 1μm (y-z); 1μm (x-z). Figure 27A-B provides photomicrographic images with Fig.27A showing representative (n=5 fixed brain slices from two mice) single z-plane confocal image of expanded mouse brain tissue after umExM processing, showing pGk13a staining of the membrane in hippocampal CA2. The image is visualized in inverted gray color. Fig.27B shows magnified view of boxed region in Fig.27A. The image is visualized in inverted gray color. Scale bars are provided in biological units (i.e., physical size divided by expansion factor): Fig.27A 10 μm; Fig.27B 2 μm. Figure 28A-F provides photomicrographic images and graphs with Fig.28A showing a representative (n=2 fixed brain slices from two mice) single z-plane confocal image of expanded CA1 region of mouse brain tissue, prepared using the biotin-DHPE staining protocol established in ref [Stephan, T., et al., Scientific Reports 20199:19, 1–6 (2019)] and followed by proExM [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)], showing biotin-DHPE staining of the membrane. Inset: magnified view of the yellow boxed region. Fig.28B as in Fig.28A but with the BODIPY staining protocol established in ref [Schmidt, R. et al. Nano Lett.9, 2508–2510 (2009)] and followed by proExM [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)], showing BODIPY staining of the membrane. Fig.28C as in Fig.28A but with the mCling staining protocol established in ref [Linder, M. E. & Deschenes, R. J. Nat. Rev. Mol. Cell Biol.8, 74–84 (2007)] and followed by proExM, showing mCling staining of the membrane. Fig.28D as in Fig.28A but with the umExM protocol, showing pGk13a staining of the membrane. Fig.28E is a boxplot showing measured S/B ratio for each of the probes used in Fig.28A-D (n=2 fixed brain slices from two mice; black points, individual measured S/B ratio, median, middle line; 1st quartile, lower box boundary; 3rd quartile, upper box boundary; error bars are the 95% confidence 20 #17358430v1 interval). Fig.28F as in Fig.28C, but with imaging of the 3rd ventricle. Image visualized in inverted gray color (dark signals on light background). Fig.28G shows magnified view of boxed region in Fig.28F. Scale bars are provided in biological units: Fig.28A-D 100 μm; Fig.28F 10μm; Fig.28G 2μm. Figure 29A-E provides photomicrographic images and a graph illustrating segmentation ability of umExM. Fig.29A, panel (i) shows single z-plane confocal image of expanded Thy1-YFP mouse brain tissue after umExM processing, showing pGk13a staining of the membrane. Fig.29A, panel (ii) shows single z-plane image showing manual segmentation of the cell body in Fig.29A, panel (i). Fig.29A, panel (iii) shows overlay of Fig.29A, panel (i) and panel (ii). Fig.29A, panel (iv) shows single z-plane confocal image of the specimen of Fig.29A, panel (i) showing GFP signal of the same field of view as in Fig.29A, panel (i). Fig.29A, panel (v shows single z-plane image showing manual segmentation of the cell body from Fig.29A, panel (iv). Fig.29A, panel (vi) shows an overlay of Fig.29A, panel (iv) and panel (v). Fig.29B, as in Fig.29B, but for segmenting dendrites. Fig.29C, left panel shows single z-plane confocal image of expanded Thy1-YFP mouse brain tissue showing pGk13a staining of the membrane. Fig.29C panel (i) shows magnified view of the box on the left. Fig.29C panel (ii) shows single z-plane image showing manual segmentation of the myelinated axon in Fig.29C panel (i). Fig.29C panel (iii) shows overlay of Fig.29C panel (i) and panel (ii). Fig.29C panel (iv shows single z-plane confocal image of the specimen of Fig.29C panel (i), showing GFP signal of the same field of view as in Fig.29C panel (i). Fig. 29C panel (v) shows single z-plane image showing manual segmentation of the myelinated axon in Fig.29C panel (iv). Fig.29C panel (vi) shows overlay of Fig.29C panel (iv) and panel (v). Fig.29D as in Fig.29C, but for segmenting unmyelinated axons. Fig.29E shows Rand score of pGk13a signal-guided segmentation of cell body, dendrites, myelinated axon and unmyelinated axons, using anti-GFP signal-guided segmentation as a “ground truth.” (n=3 cell bodies and n=3 dendrites from two fixed brain slices from two mice, and n=5 myelinated axons and n=5 unmyelinated axons from two fixed brain slices from two mice). Scale bars: Fig.29A panels (i-vi) 5 μm; Fig.29B panels (i-vi 5 μm; Fig.29C left panel 2 μm; Fig.29C panels (i-vi) 0.5 μm; Fig.29D left panel 2 μm; Fig.29D panels (i-vi) 0.5 μm. Figure 30A-I shows results demonstrating traceability of umExM. Fig.30A (pGk13a column) shows serial confocal images of expanded Thy1-YFP mouse brain tissue after umExM processing, showing pGk13a staining of the membrane. (GFP column) anti-GFP 21 #17358430v1 signal of the same sample in the same field of view. Fig.30B, left shows pGk13a signal- guided manually traced and reconstructed myelinated axon from (Fig.30A, pGk13a column). Fig.30B, right, as in Fig.30B left, but with anti-GFP signals. Fig.30C shows Rand score (n=3 myelinated axons from two fixed brain slices from two mice) of pGk13a signal-guided manual tracing of myelinated axons, using anti-GFP signal-guided tracing as a “ground truth.” Fig.30D, as in Fig.30A, but with an unmyelinated axon. Fig.30E, as in Fig.30B but for Fig.30D. Fig.30F, as in Fig.30C but for unmyelinated axons (n=3 unmyelinated axons from two fixed brain slices from two mice). Fig, 30G shows representative (n=4 fixed brain slices from two mice) single z-plane confocal image of expanded mouse brain tissue (corpus callosum) after umExM with double gelation processing (see Fig.14), showing pGk13a staining of the membrane. The seeding points for manual segmentation are labeled. Fig.30H shows magnified view of the box in Fig.30G. Fig.30I shows 3D rendering of 20 manually traced and reconstructed myelinated axons in the corpus callosum. Planes were visualized from raw umExM images that were used for tracing. Scale bars: Fig.30A 0.5μm; Fig.30D 0.2μm; Fig.30G 18μm; Fig.30I 39.25μm (x); 39.25μm (y); and 20μm (z). Figure 31 provides a workflow schematic diagram of an iterative form of umExM Lighter- colored captions highlight the key differences from iExM [Nishino, M. et al. PLoS One 15, e0236373 (2020)], whereas darker colored captions highlight steps similar to those in iExM. In Fig.31 panel (i) a specimen is chemically fixed with 4% paraformaldehyde (PFA) + 0.5% calcium chloride (CaCl2) at 4 ⁰C for 24 hours. The brain is sliced on a vibratome at 100 μm thickness at 4 ⁰C. In Fig.31 panel (ii) the specimen is treated with pGk13a at 4 ⁰C overnight (unless otherwise noted, overnight means >16 hours throughout the figure). In Fig.31 panel (iii) the specimen is treated with acrylic acid N-hydroxysuccinimide ester (AX) at 4 ⁰C overnight. In Fig.31 panel (iv) the specimen is embedded in an expandable hydrogel (made with cleavable crosslinker DATD [Boutin, J. A. Cell. Signal.9, 15–35 (1997)] at 4 ⁰C for at least 24 hours. In Fig.31 panel (v) the specimen-embedded hydrogel is mechanically softened with enzymatic cleavage of proteins (i.e., specific cleavage with proteinase-k) at room temperature (~24 ⁰C), overnight. In Fig.31 panel (vi) the specimen-embedded hydrogel is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes). Then, the specimen-embedded hydrogel is re-embedded into a non-expandable hydrogel at 50 ⁰C, for >4 hours. In Fig.0 panel (vii) the sample is treated with AX as done in Fig.31 panel (iii). Subsequently, the sample was gelled again but with a monomer solution that contains the non-cleavable crosslinker (made with BIS) at room temperature (~24 ⁰C), 22 #17358430v1 overnight. In Fig.31 panel (viii) the specimen-embedded hydrogel is incubated in a gel cleaving solution (contains sodium metaperiodate) at room temperature (~24 ⁰C) for 1 hour. This step cleaves the initial gel and re-embedding gel that was formed in Fig.31 panel (iv) and panel (vi). In Fig.31 panel (ix) the pGk13a, that is anchored to the gel matrix, is fluorescently labeled via click-chemistry (i.e., DBCO-fluorophore) at room temperature, overnight. The specimen-embedded hydrogel is expanded with water at room temperature for 1.5 hours (exchanging water every 30 minutes). DETAILED DESCRIPTION The invention, in part, provides methods and probes for visualizing membrane ultrastructure through dense labeling of membranous structures and visualization using ordinary microscopy with expansion microscopy (ExM). The invention, in part, includes an imaging method, which is referred to herein as “ultrastructure membrane expansion microscopy” (umExM), Fig.1). The invention also includes novel membrane probes, a non- limiting example of which is referred to herein as “pGk13a” (Fig.2A). In some embodiments of methods of the invention, a probe of the invention is used in modified ExM protocols, for example a umExM method (Fig.2) to visualize membrane ultrastructure. Methods and probes of the invention provide uniform and continuous membrane labeling, yielding a high signal- to-background ratio, which preserves ultrastructure alongside visualization of associated proteins, in fixed tissues. Membrane labeling methods and probes of the invention enable imaging membrane components with nanoscale registration relevant to membrane landmarks, and also facilitate the segmentation and tracing of membranous structures, such as axons and dendrites, using conventional microscopy, such as but not limited to confocal microscopy. Methods and compositions of the invention can also be used to label and visualize one or more biomolecules of interest that are present in membrane ultrastructure. Methods and probes of the invention can be used in fields such as but not limited to neuroscience and biology, for the investigation of ultrastructure, cellular compartments, and molecular content, in intact tissues, with nanoscale precision. Certain umExM methods of the invention result in high resolution low-distortion expansion. For example, though not intended to be limiting, in some embodiments, umExM methods of the invention result in ~60nm resolution through the low-distortion (~3%) physical 4x expansion of a thick (100μm) mice brain tissue. It has now been shown that in thick brain tissue sections, methods of the invention preserve ultrastructural context. Studies were performed to quantitatively examine the anatomical properties of ultrastructure features 23 #17358430v1 (i.e., diameter of unmyelinated and myelinated axons, and cilia) that were previously measured via EM. Methods and compositions of the invention can be used for light microscope imaging of ultrastructure feature (i.e., membrane vesicle around the cilia), features previously only visualized using electron microscopy (EM). Methods and probes of the invention provide a high signal-to-background (S/B) and uniform labeling of membrane and permit simultaneous visualization of biomolecules of interest in the ultrastructural context. umExM methods and probes can achieve higher resolution (for example, ~30nm) when combined with existing fluorescence fluctuation imaging methods (i.e., SRRF). umExM probes and methods of the invention have broad utility in visualization of ultrastructural context along with the key biomolecules and are useful to describe and characterize ultrastructural features that were previously only possible to visualize with EM. In addition, methods and compositions of the invention can be used to visualize and reconstruct dense neuronal circuitries with biomolecule of interest to better understand the brain and other tissues. Embodiments of methods and probes of the invention can be used to visualize the ultrastructure of membranes. Methods of the invention comprise labeling membranous structures and visualizing the labeled structures with standard fluorescence microscopy (FM) techniques. Methods and probes of the invention may be used to obtain images of resolution similar to low-resolution EM images. Use of embodiments of methods and compositions of the invention permit visualization of anatomical properties and characteristics of ultrastructure features previously identified with EM including features that with prior methods were only visible using EM methods. Thus, methods and compositions of the invention may be used to visualize features that could not be fully characterized using FM methods. Embodiments of methods and probes of the invention also permit visualization of ultrastructural features across a large volume of tissue with a high signal-to-background (S/B) ratio. The utility of umExM methods and probes of the invention has now been demonstrated for segmentation and tracing of neuronal processes, such as axons, in tissues such as, but not limited to, mouse brain tissue. Additional aspects of the invention comprises combining umExM methods and probes with optical fluctuation imaging such as, but not limited to SACD. umExM methods of the invention, may also include iterative expansion methods, which can be used to increase resolution. In some embodiments, umExM methods of the invention comprising iterative expansion yielded ~35 nm resolution imaging, 24 #17358430v1 indicating use of such methods and probes to obtain electron microscopy resolution visualization of membranes on ordinary light microscopes. As described herein, the invention comprises rationally and systematically designed novel membrane-labeling probes that may be used in conjunction with novel ExM protocols. Probes and methods of the invention can be used to achieve dense labeling of membranes, including plasma membranes, mitochondrial membranes, nuclear membranes, ciliary membranes, myelin sheaths, and extracellular vesicle membranes in tissues. Methods of the invention have been demonstrated to be more effective for imaging than prior methods, and a comparison of methods of the invention with some previous methods are shown in Table 1. Table 1. A comparison of protocols. The comparison table excludes TREx [Linder, M. E. & Deschenes, R. J. Nat. Rev. Mol. Cell Biol.8, 74–84 (2007)], LExM [Feng, G. et al. Neuron 28, 41–51 (2000)], TRITON ExM [Hendrickson, W. A. et al. Proc. Natl. Acad. Sci. U. S. A. 86, 2190 (1989)], uniExM [Schmidt, R. et al. Nano Lett.9, 2508–2510 (2009)], sphingolipid ExM [McKay, B. E. et al., Methods Mol. Biol.418, 111–128 (2008)], and TT-ExM [Stephan, T., et al., Scientific Reports 20199:19, 1–6 (2019)] as they did not show membrane or lipid labeling in tissue.
Figure imgf000026_0001
25 #17358430v1
Figure imgf000027_0001
26 #17358430v1
Figure imgf000028_0001
“N/A” denotes not available * Measured with 100% laser and 100ms exposure time and 60x, 1.27NA water immersion lens umExM The protocol comprising methods of the invention is referred to herein as ultrastructural membrane expansion microscopy (umExM), using the word ultrastructure in 27 #17358430v1 the same sense as an earlier protocol in the expansion microscopy community, called ultrastructure expansion microscopy [Gambarotto, D. et al. Nat. Methods 16, 71–74 (2019)]. umExM preserves ultrastructure and enables the visualization of membranous structures in tissue slices. As a non-limiting example, umExM methods of the invention have been used with 100µm-thick slices of fixed mouse brain at a resolution of ~60 nm with excellent uniformity and continuity of membrane labeling as well as a high signal-to-background ratio (40-80 fold higher than background). umExM methods and probes of the invention can be used to co-visualize membranous structures along with proteins and RNAs. The dense membrane labeling of umExM methods and probes of the invention can be used for the segmentation of neuronal compartments (e.g., cell bodies, dendrites, and axons), and tracing of neuronal processes (e.g., axons). umExM methods of the invention achieve dense labeling of membranes and high- integrity expansion and enable imaging of membranous structures using a standard confocal microscope. In some embodiments, methods of the invention achieve ~60nm lateral resolution and enable co-visualization of membranous structures in a wide range of brain regions along with proteins and RNAs. umExM methods of the invention enable segmentation of cell bodies, dendrites, and axons (>200nm in diameter) and can be used to trace axons. Probes The invention, in part, provides probes that can be used with ExM chemistry. In some embodiments, a probe of the invention is used for nanoscale resolution imaging and permits continuous tracing of membranous structures with ExM chemistry. In some embodiments, a probe of the invention, is an unnatural synthetic amphiphilic membrane probe comprising characteristics that improve imaging capability of traditional ExM methods. A probe of the invention may comprise one or more of the following characteristics: (1) lipophilicity, which in some embodiments is similar to art-known traditional fluorescent lipophilic dyes such as DiI and/or other dyes that preferentially localize to and diffuse within membranes permitting membrane labeling; (2) a chemical handle for chemoselective conjugation of a fluorophore following the formation of an expandable hydrogel network, so the initial label remains as small as possible to facilitate diffusion and to avoid potential fluorophore degradation during the free-radical polymerization, and (3) a polymer-anchorable handle for binding the probe to an ExM-gel matrix, to allow expansion. 28 #17358430v1 In some embodiments a probe of the invention comprises a chain of lysines. Lysines contain primary amines that serve as sites for binding to a polymer-anchorable handle. A non-limiting example of a polymer-anchorable handle is acryloyl–X (AcX), which has been used to attach proteins via their amines to an in situ synthesized ExM hydrogel. In some embodiments, lysine chain of a probe of the invention comprises D-lysines. The presence of D-lysines minimized degradation during the mechanical softening step of ExM, which in some instances includes a proteinase K softening step. In some embodiments, a probe of the invention comprises one or more D-cystines. In some embodiments, a probe of the invention is an acrylic acid conjugated probe. A non-limiting example of an acrylic acid conjugated probe that may be used in certain embodiments of the invention is: palmitoyl-glycine-acrylic-lysine-lysine-lysine-lysine-lysine- lysines-lysine-lysine-lysine-lysine-lysine-lysine-lysine-azide (palmitoyl-glycine-acrylic- KKKKKKKKKKKKK-azide (SEQ ID NO: 1), which is also referred to herein as probe pGk13A. In some embodiments, one or more lysines included in an acrylic acid conjugated probe of the invention is a D-lysine. Some embodiments of a probe of the invention comprises a lipid tail on the amine terminus of the chain. In some embodiments, one or more glycine linkers are positioned between the lipid tail and the amine terminus of the lysine chain. In some embodiments, a glycine linker comprises 1, 2, or 3 glycines. Inclusion of a glycine linker assists in membrane labeling and provides mechanical flexibility. A probe of the invention also comprises a chemical “handle” attached to the carboxy terminus of the terminal lysine. As a non-limiting example, in some embodiments, palmitoyl and biotin are used for lipid tails and chemical handles respectively and to attach five (5) lysines in the backbone. A probe prepared as described in the above non-limiting example, comprised a glycine and penta-lysine D-peptidic backbone, with a palmitoyl lipid group on the amine-terminus and a biotin on the carboxy-terminus. The probe is referred to herein as “pGk5b” and comprises the structure: palmitoyl-glycine-lysine-lysine-lysine-lysine-lysine- biotin (palmitoyl-glycine-KKKKK-biotin (SEQ ID NO: 2). Another probe of the invention is “pGk5a” which comprises the same structure as pGk5b except the biotin is replaced with an azide. pGk5a probe has the structure: palmitoyl-glycine-KKKKK-biotin (SEQ ID NO: 3). Another probe of the invention is “pGk13b” which has the same structure as pGk13b except the azide is replaced with biotin. The pGk13b probe has the structure: (palmitoyl-glycine- acrylic-KKKKKKKKKKKKK-biotin (SEQ ID NO: 4). Non-limiting examples of other probes that may be used in methods of the invention are (palmitoyl-glycine-acrylic- 29 #17358430v1 KKKKKKKKKKKKKK-azide (SEQ ID NO: 5), (palmitoyl-glycine-acrylic- KKKKKKKKKKKK-azide (SEQ ID NO: 6), (palmitoyl-glycine-acrylic-KKKKKKKKKK- azide (SEQ ID NO: 7), (palmitoyl-glycine-acrylic-KKKKKKKKK-azide (SEQ ID NO: 8); (palmitoyl-glycine-acrylic-KKKKKKK-azide (SEQ ID NO: 9); (palmitoyl-glycine-acrylic- KKKKKK-azide (SEQ ID NO: 10), (palmitoyl-glycine-acrylic-KKKKKKKKKKKKKK- biotin (SEQ ID NO: 11), (palmitoyl-glycine-acrylic-KKKKKKKKKKKK-biotin (SEQ ID NO: 12), (palmitoyl-glycine-acrylic-KKKKKKKKKK-biotin (SEQ ID NO: 13), (palmitoyl- glycine-acrylic-KKKKKKKKK-biotin (SEQ ID NO: 14); (palmitoyl-glycine-acrylic- KKKKKKK-biotin (SEQ ID NO: 15); (palmitoyl-glycine-acrylic-KKKKKK-biotin (SEQ ID NO: 16). In some embodiments, instead of a single glycine as in SEQ ID NOs: 1-16, there are 2 glycines at the position of the single glycine, or there are three glycines at the position of the single glycine. In some embodiments, a probe of the invention comprises a sequence set forth as one of SEQ ID NOs: 1-16, but instead of an azide or biotin, the probe includes a different linkable group for lipid staining. ExM and umExM Expansion microscopy (ExM) may be an ideal approach to this challenge. ExM physically magnifies biological specimens by covalently anchoring biomolecules or labels to a swellable polymer network (typically sodium polyacrylate) synthesized in situ throughout the specimen, followed by chemical softening of the tissue, and the addition of water to swell the hydrogel. At the hydrogel swells, the anchored biomolecules or labels are pulled apart from each other, isotropically, in the most popular form to a physical magnification of ~4x in linear dimension. The net result is that biomolecules or labels that are initially localized within the diffraction limit of a traditional optical microscope can now be separated in space to distances far enough that they can be resolved on ordinary FM. Established ExM methods generally comprise four steps: Anchoring, Gelation, Softening/Denaturation, and Expansion. Anchoring involves attaching polymerization monomers, a non-limiting example of which is acrylamide, to fixed biological samples through covalent modification. Gelation is a process of forming an acrylamide-acrylate hydrogel within the biological sample through polymerization. Digestion/Denaturation comprises breaking down (digestion of) biomolecules using methods including but not limited to contact with enzymes, heat, denaturing agents, reducing agents, or a combination thereof. Expansion comprises soaking the hydrogel in an aqueous solution (a non-limiting example of which is water) to physically expand (also referred to herein as “to magnify”) the 30 #17358430v1 biological sample that is embedded in the hydrogel. Methods of the invention include modifications of standard ExM methods and may also comprise use of a probe of the invention in the modified ExM methods. umExM method of the invention may include modified ExM methods and a probe of the invention as described herein. Some embodiments of umExM methods of the invention comprise a modified expansion microscopy methods and include use of an amphiphilic membrane probe (a non- limiting example of which is pGk13a). A pGk13a membrane probe of the invention comprises a glycine and 13-lysine D-peptidic backbone, with a palmitoyl lipid group on the amine-terminus and an azide on the carboxy-terminus (termed pGk13a). Some embodiments of a umExM method of the invention comprise an ExM method in which the gelation (also referred to herein as polymerization) temperature is modified such that instead of a standard gelation at room temperature or above, the umExM method includes gelation of the expansion sample at approximately 4⁰C overnight. Some embodiments of a umExM method of the invention comprise an ExM method in which the softening step (Fig.1) is modified such that instead of softening using proteinase-k, the softening step is done using lysC/trypsin. As a non-limiting example, unlike prior ExM methods, such as mExM methods (see for example US publication no. US2010130882), methods and probes of the invention can be used to visualize cell boundaries (Fig.2). Methods and probes of the invention are compatible with conventional antibody staining (Fig.3) if the proteins are cleaved in the softening step [Fig.1B panel (v)] using lysC/trypsin, instead of proteinase-k. The invention, in part, provides methods for obtaining a structure and identity of biomolecules. As used herein the term, “biomolecule” may be used in reference to a protein molecule, a lipid molecule, a glycoprotein molecule, a polynucleotide molecule, or a carbohydrate molecule. In some embodiments, a method of the invention is performed on a single type of biomolecule and in certain embodiments a method of the invention is performed on a plurality of a single type of biomolecule. For example, though not intended to be limiting, a single type of protein may be assessed using an embodiment of a method of the invention, and the method may be used on a plurality of the single type of protein molecule. In some embodiments, a method of the invention is performed on two or more different biomolecules. For example, though not intended to be limiting, a biological sample may include a plurality of different protein molecules, and each may be assessed using a method of the invention. As another non-limiting example, a biological sample may include one or more polynucleotide molecules and one or more protein molecules, each of which may 31 #17358430v1 be assessed using a method of the invention. As used herein, the term plurality means more than one, which may be 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, or more. The terms “biospecimen” and “biological sample” are used interchangeably herein. The term “clinical sample” is used herein to mean a sample obtained from a subject. In certain embodiments, methods include embedding one or a plurality of a cell/tissue and/or membrane in a polymer, for example but not limited to an acrylamide polymer, followed by digestion, such as but not limited to proteolytic digestion, and swelling of the polymer comprising the embedded biomolecule(s). In certain embodiments, a method and/or a probe of the invention may be used for to assess ultrastructure of a membrane in the imbedded sample and/or to assess one or more biomolecules of interest. For example, though not intended to be limiting, an embodiment of a method of the invention can be used to assess a protein in a biological sample, to detect and assess genomic DNA in a sample, to detect and assess a carbohydrate molecule(s) in a biological sample, etc. Methods of the invention can be used to detect and identify one or more alternations in proteins, lipids, glycoproteins, polynucleotides, etc. In a non-limiting example, an embodiment of a method of the invention may be used to identify and assess a polynucleotide (DNA or RNA) sequence such as, but not limited to: a genomic DNA sequence from a subject; a wild-type (control) genomic DNA sequence; a wild-type RNA sequence; a genetically modified RNA sequence; a genetically engineered genomic DNA sequence, a genomic DNA sequence or RNA sequence known to be or suspected of being associated with a disease or condition. Methods of the invention can be used to identify biomolecule components (e.g., amino acid sequences, nucleic acid sequences, etc.) and membrane ultrastructures as well as differences in one or more biomolecules obtained from different sources. As a non-limiting example, methods of the invention may be used to compare structure and/or sequence/components of a normal (e.g., control) biomolecule to structure and/or sequence/components of a biomolecule obtained from a biological sample. In some embodiments a biological sample examined using a method and or probe of the invention is obtained from a subject who has, or is suspected of having, a disease or condition. Differences between the determined biomolecule and the control biomolecule may assist in identifying a biomolecule variation or abnormality associated with the subject’s disease or condition. Methods of the invention may be used to provide structure and component information beyond that obtainable from assessment of spatial localization of biomolecules when examined in unexpanded conformations or beyond information obtainable using traditional light microscopy and similar to information obtainable using electron microscopy. 32 #17358430v1 Polynucleotides/Proteins/Lipids/Carbohydrates/Glycoproteins The term “nucleotide” as used herein includes a phosphoric ester of nucleoside—the basic structural unit of nucleic acids (DNA or RNA). The terms “polynucleotide”" and “nucleic acid” refer to a polymer comprising multiple nucleotide monomers and may be used interchangeably herein. A polynucleotide may be either single stranded, or double stranded with each strand having a 5' end and a 3' end. A nucleotide in a polynucleotide may be a natural nucleotide (deoxyribonucleotides A, T, C, or G for DNA, and ribonucleotides A, U, C, G for RNA) The term “protein” and “polypeptide” refer to nitrogenous organic compounds comprising chains of amino acids and the terms may be used interchangeably herein. The term “lipid” as used herein refers to organic compounds comprising fatty acids or their derivatives. The term “carbohydrate” refers to a molecule that includes carbon (C), hydrogen (H) and oxygen (O) atoms. Another term for carbohydrate is saccharide, which is a group that includes sugars, starch, and cellulose. Monosaccharides and disaccharides are relatively low molecular weight carbohydrates. Larger saccharides include polysaccharides and oligosaccharides. The term “glycoprotein” as used herein refers to any of a class of proteins that have carbohydrate groups attached to a polypeptide chain. A glycoprotein comprises oligosaccharide chains (glycans) that are covalently attached to amino acid sidechains. In some embodiments, a biomolecule assessed using a method of the invention is a “modified biomolecule,” which, as used herein, refers to a biomolecule that comprises one or more non-natural or derivatized components. As used herein the term “component” means a portion of the biomolecule. As non-limiting examples the term “component” used in reference to (1) a protein biomolecule may be an amino acid, (2) a polynucleotide biomolecule may be a nucleic acid; (3) a lipid biomolecule may be a fatty acid; (4) a carbohydrate molecule may be a saccharide or polysaccharide; and (5) a glycoprotein molecule may be a carbohydrate molecule, an amino acid, or a protein molecule. In some embodiments, a component of a biomolecule is chemically or biochemically modified. In some embodiments of the invention, one or more modified components are incorporated into a biomolecule. Modified biomolecules may confer desirable properties absent or lacking in the natural biomolecule and biomolecules comprising one or more modified components may be used in the compositions and methods of the invention. As used herein, a “modified 33 #17358430v1 biomolecule” refers to a biomolecule comprising at least one modified component. In some embodiments, a modified biomolecule may comprise one, two, three, four, five, or more modified components. A polynucleotide may be DNA (including but not limited to cDNA or genomic DNA), RNA, or hybrid polymers (e.g., DNA/ RNA). The terms “polynucleotide” and “nucleic acid” do not refer to any particular length of polymer. Polynucleotides used in embodiments of methods of the invention may be at least 1, 2, 3, 4, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 500, 1000, 2000, or 5000 kb or more in length. The term “protein” does not refer to any particular length of the molecule. A protein used in embodiments of methods of the invention may be at least 1, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 500, or more amino acids in length. The term “sequence,” used herein in reference to a polynucleotide or protein, refers to a contiguous series of nucleotides or amino acids, respectively. The term “structure” as used herein in reference to a polynucleotide refers to overall sequence organization of the polynucleotide, including “structural variations” such as insertions, deletions, repeats, and rearrangements. A polynucleotide or other biomolecule may be chemically or biochemically synthesized, or may be isolated from a subject, cell, tissue, or other source or sample that comprises, or is believed to comprise, the biomolecule. The term “structure” as used herein in reference to a protein, lipid, or glycoprotein refers to overall s component organization of the biomolecule. A membrane assessed using a method and/or probe of the invention may be a cultured membrane or may be obtained from a subject. Embedding Embodiments of methods of the invention may include embedding the biomolecule in a polymer material for example using certain methods of standard ExM practice. In some instances, a means for embedding the biomolecule in the polymer material comprises incubating the biomolecule in a polymer monomer and polymerizing the monomer. In some embodiments, the polymer material is a swellable polymer material. A non-limiting example of a swellable polymer material comprises an acrylamide-co-acrylate copolymer. As used herein, the term “swellable polymer material” generally refers to a material that expands when contacted with a liquid, such as water or other solvent [Wassie A., et al., Nat. Methods 16, 33-41 (2019); and US Patent 10,059,990 in relation to swellable and non-swellable materials, each publication is incorporated by reference herein in its entirety.] 34 #17358430v1 The swellable material may uniformly expand in three dimensions. Additionally, or alternatively, the material is transparent such that, upon expansion, light can pass through the sample. In some embodiments, the swellable polymer material is a swellable polymer or hydrogel. In one embodiment, the swellable polymer is formed in situ from precursors thereof: for example, one or more polymerizable materials, monomers or oligomers may be used, such as monomers selected from the group consisting of water-soluble groups containing a polymerizable ethylenically unsaturated group. Monomers or oligomers may comprise one or more substituted or unsubstituted methacrylates, acrylates, acrylamides, methacrylamides, vinylalcohols, vinylamines, allylamines, allylalcohols, including divinylic crosslinkers thereof (e.g., N, N-alkylene bisacrylamides). Precursors may also comprise polymerization initiators and crosslinkers. In some embodiments, a swellable polymer is an acrylamide-co-acrylate copolymer, polyacrylate, or polyacrylamide, or co-polymers or cross-linked co-polymers thereof. Alternatively, or additionally, the swellable polymer may be formed in situ by chemically cross-linking water-soluble oligomers or polymers. Thus, the invention envisions adding precursors, such as water-soluble precursors, of the swellable polymer to the sample and rendering the precursors swellable in situ. Certain embodiments of the invention include embedding a biomolecule in a non- swellable polymer material capable of conversion to a swellable polymer material. In this instance, the method may also include polymerizing the non-swellable polymer and then converting the polymerized non-swellable polymer material into a swellable polymer material. As used herein, the term “non-swellable polymer material” comprises a polymer material capable of conversion to a swellable polymer material, including a non-swellable hydrogel comprising one or more of an acrylamide and polyacrylate [Ueda H., et al., Nat. Rev. Neurosci.21, 61-79 (2020)]. In some embodiments of the invention, the polymer is not a polyacrylade polymer. In some embodiments of methods of the invention, the non- swellable polymer material is converted into a swellable polymer material before the physical expanding step of the method. A non-swellable polymer can comprise various materials. As a non-limiting example, a non-swellable polymer material may include a non-swellable hydrogel. As another non-limiting example, a non-swellable hydrogel may include one or more of an acrylamide and polyacrylate. A non-swellable polymer used in an embodiment of a method of the invention may be a polymer that can be chemically converted into a swellable polymer. For example, such a non-swellable polymer may be acrylamide; acrylamide can later be converted into an acrylamide-co-acrylate copolymer after treatment 35 #17358430v1 with a strong base such as sodium hydroxide, which can then swell after dialysis with water. Other polymers such as polyacrylate may also be used in certain embodiments of the invention. Some embodiments of a umExM method of the invention comprise an ExM method in which the gelation temperature (also referred to herein as polymerization temperature) is modified such that instead of a standard gelation at room temperature or above, the umExM method includes gelation of the expansion sample at approximately 4⁰C overnight. Homogenization Embodiments of methods of the invention include a homogenization step. As used herein and in the expansion microscopy arts, the term “homogenization” refers to a process that frees up, also referred to as “releases” intra-sample connections before expansion. For example, a homogenization step may be used to release connections within the biomolecule. The release of connections loosens the biomolecule in place and renders the biomolecule capable of expanding in the expansion step of the method. A term for a biomolecule following a homogenization step in a method of the invention is “homogenized biomolecule material,” which indicates the biomolecule has been homogenized and is in a condition in which the biomolecule is capable of expansion. In some embodiments of methods of the invention, a means of homogenization of a biomolecule includes one or more of an enzyme- based digestion and a heat-based denaturation. In some embodiments of methods of the invention, a means of homogenizing an embedded biomolecule comprises contacting the polymer material in which the biomolecule is embedded with one or more of (i) a strong detergent or surfactant (e.g., sodium dodecyl sulfate); (ii) one or more enzymes; and (iii) denaturing heat. Non-limiting examples of enzymes that may be used in a homogenization step of a method of the invention are proteinase K (proK), an endoproteinase, non-limiting examples of which are: LysC and trypsin. Some embodiments of a umExM method of the invention comprise an ExM method in which the softening step (also referred to herein as a homogenization step) is modified such that instead of softening using proteinase K, the homogenization is done by contacting the gel with lysC and trypsin. Polymer and Biomolecule Physical Expansion The polymer within which homogenized biological sample is embedded is isotropically expanded. In some embodiments, a solvent or liquid is added to the polymer 36 #17358430v1 containing the homogenized biomolecule material and the solvent or liquid is absorbed by the swellable material and causes swelling. For example, if the mechanism of expansion is the polyelectrolyte effect, the polymer may be dialyzed against water or an aqueous solution to expand. In one embodiment, the addition of water allows the embedded sample to expand at least 3, 4, 5, or more times its original size in three dimensions. Thus, the sample may be increased 100-fold or more in volume. In some embodiments of methods of the invention, a means of physically expanding the sample material includes expanding the polymer material in which the biological sample material is embedded, wherein the expansion of the polymer material expands the homogenized biological sample material (and biomolecules in that biological sample) isotropically in at least a linear manner within the polymer material. In certain embodiments, the polymer material comprises a hydrogel and a means of expanding the hydrogel includes contacting the hydrogel with an aqueous solution, optionally water. Re-embedding Certain embodiments of the invention an expanded swellable polymer comprising homogenized biological sample (e.g., biomolecule) material may be re-embedded in a non- swellable or in a swellable polymer prior to detection of the biomolecule material. A re- embedded swellable polymer may be partially or completely degraded chemically, provided the biomolecule material in the polymer either remains anchored or is transferred to the non- swellable polymer. In some embodiments of the invention, non-charged polymer chemistries may be used to avoid charge passivation. In certain embodiments of the invention, the physically expanded polymer and biomolecule materials are not re-embedded in a polymer prior to being detected. In certain embodiments of the invention, the physically expanded polymer and biomolecule materials are re-embedded in a polymer prior to being detected. Although not intended to be limiting, in some embodiments in which an extended imaging time is desirable, a method of the invention includes re-embedding the physically expanded polymer and biomolecule materials into a swellable polymer. Labelling A biomolecule or biomolecule material embedded in a polymer may be “labelled” or “tagged” with a probe of the invention. As described elsewhere herein, “pGk5b” and “pGk13a” are non-limiting examples of a probe that can be used in a method of the invention. In some embodiments, a probe used in a method of the invention does not include a 37 #17358430v1 detectable label, but probe is prepared such that it comprises a component that permits inclusion of a detectable label on the probe at a subsequent time. For example, though not intended to be limiting, pGk5b or pGk13a may be used as a probe in a method of the invention without including a detectable label such as a fluorophore, but the probe may comprise azide, which may be used to attach a detectable label, a non-limiting example of which is a fluorophore, to the probe at a subsequent time. In some embodiments, a probe of the invention comprises a chemical handle, a non-limiting example of which is an azide, for chemoselective conjugation of a detectable label following the formation of an expandable hydrogel network, permitting the initial probe molecule to be small in size to facilitate diffusion and to avoid potential degradation of a detectable label, a non-limiting example of which is a fluorophore, during the free-radical polymerization of the hydrogel network. Methods and probes of the invention may be used on tissue slices 50 to 100 microns thick, and in some embodiments, may be used on thinner or thicker samples. In some embodiments, a method of the invention includes incubating a tissue sample with a pGk13a probe for at least 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 36, or 48 hours, including all times within the range. In another non-limiting example, a sample tissue is incubated in a concentration of pGk13a probe that is at least 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, mg/ml, including all concentrations within that range. In some embodiments, of methods of the invention, the incubation is performed at about 4oC. In some embodiments, the incubation is performed at about 2 oC, 3 oC, 4 oC, 5 oC, 6 oC, 7 oC, 8 oC, 9 oC, 10 oC. In some embodiments, a probe of the invention comprises a detectable label either initially or one subsequently attached to the probe. As used herein, the term “detectable label” means a label or tag chemically bound to the probe and when the probe binds a biomolecule in a biological sample or to a component thereof, the biomolecule or component is labelled. The label may be detected using microscopy or one or more other means of detection. A detectable label may be selective for a specific target (e.g., a biomarker or class of molecule), as may be accomplished with an antibody or other target specific binder, or the detectable label on a probe of the invention may be an affinity label, including one or more of biotin, digoxigenin, and a hapten. In some embodiments, a detectable label of a probe of the invention comprises a visible component, as is typical of a dye or fluorescent molecule, a luminescent label, a radiolabel, an enzymatic label, a contrast agent, a heavy metal, or a heavy element such as bromine or iodine, or metals such as gold, osmium, rhenium, etc.; however, any signaling means used by the label is also contemplated. 38 #17358430v1 In other embodiments, enzymatic methods for detectable labeling are used, including contacting the biomolecule and/or biomolecule material with one or more enzymes, under suitable conditions for activity of the one or more enzymes to result in detectable labeling of biomolecule and/or biomolecule material, respectively. Detecting structure and sequence In certain embodiments, methods of the invention allow detection of spatial structures and components of expanded biomolecule material and ultrastructural components of membranes using microscopic visualization methods. The signal from individual molecules may be spatially punctate due to the homogenization step. However, these puncta will be spatially proximal, allowing the overall length of the biomolecule to be inferred based on the dimension in which spatial proximity is highest. Thus, the structure of the biomolecule may be inferred over distances up to the entire length of the biomolecule. As used herein, “detecting” means using one or both of an imaging method and a sequencing method to identify the spatial position and components of biomolecules. Imaging methods include but are not limited to light microscopy, epi-fluorescence microscopy, confocal microscopy, spinning disk microscopy, multi-photon microscopy, light-sheet microscopy, total internal reflection (TIRF) microscopy, light-field microscopy, imaging mass spectrometry, imaging Raman spectroscopy, super-resolution microscopy, or transmission electron microscopy. In some embodiments of methods of the invention, enzymatic detection methods may be used to detect spatial structures and components of expanded biomolecule material and ultrastructural components of membranes. Enzymatic means that can be used in conjunction with methods and probes of the invention include but are not limited to random primer extension, terminal transferase tailing, padlock probe rolling circle amplification [Larsson, C., et al. Nat. Methods 1(3): 227-232 (2004)], in situ PCR [Hodson, R., et al. Appl. Environ. Microbiol.4074-4082 (1995)], horseradish peroxidase tyramide signal amplification [Schonhuber, W., et al. Appl. Environ. Microbiol.3268-3273 (1997)], luciferase-catalyzed pyrophosphate chemiluminescence [Nyren, P., et al. Anal. Biochem.208:171-175 (1993)], or other PCR-based or DNA sequencing methods [Ståhl, P. L., et al. Science 353.6294: 78-82 (2016); Rodriques, S. G., et al., Science 363.6434: 1463-1467 (2019)]. As used herein, “spatial position” refers to the location of a biomolecule, biomolecule material, and/or biomolecule component relative to the location of another biomolecule, biomolecule material, and/or biomolecule component, respectively. Certain embodiments of methods of the invention are useful to determine relative positions of one or more 39 #17358430v1 components or biomolecule materials generated from a single biomolecule. Spatial positions and relative spatial positions of ultrastructural components of a membrane can also be determined using embodiments of methods and probes of the invention. Thus, embodiments of methods of the invention can be used to disarticulate a biomolecule and/or ultrastructural membrane features into components in a controlled manner and then to identify the components and/or features and their relative positions in the expanded conformation. Segmenting cell compartments with umExM Methods and probes of the invention can be used for segmentation of cell compartments. In some embodiments, methods and probes of the invention are used for segmentation of neuronal compartments (i.e., cell bodies, dendrites, axons). As a non- limiting example, segmentation using umExM methods and probes is used to help analyze signaling proteins within distinct neuronal compartments.). It was determined that umExM images could capture and support the segmentation of neuronal compartments, and umExM methods of the invention enable capture and segmentation of cell compartments in fixed tissue, such as but not limited to neuron compartments, with a standard confocal microscope. Tracing Axons with umExM Methods and probes of the invention can be used to trace cell processes, such as axons. In some embodiments, methods of the invention comprise manual axon tracing using umExM images. umExM methods of the invention may be used to trace neuronal processes, such as myelinated and unmyelinated axons. In some embodiments, methods and probes of the invention can be used to trace neurons in tissues such as, but not limited to cortex, hippocampus, and corpus callosum. Antibody staining with umExM umExM methods and probes of the invention may, in some embodiments, be used with antibody staining of endogenous protein epitope. Methods of the invention may include pre-expansion antibody staining and/or post-expansion antibody staining. Post-expansion antibody staining in conjunction with a method of the invention may be used to reveal previously unknown proteins and even cellular structures, in part because the antibodies are applied to expanded samples, where densely packed proteins are decrowded, making more room for antibody staining. Selection of antibodies may be based on routine parameters, such as, but not limited to the type of tissue and the goal of visualization. 40 #17358430v1 Visualization of proteins and RNAs with umExM Methods and probes of the invention may be used with RNA visualization methods. In some embodiments, umExM methods comprise methods of ExM visualization of RNA (ExFISH). In some embodiments, a umExM method of the invention comprises an RNA anchoring step, for example, using an art-known RNA anchor (for example but not intended to be limiting, LabelX) [Chen, F. et al. Nat. Methods 13, 679–684 (2016)]. In a non-limiting example, probe pGk13a is applied to label a membrane in a sample, LabelX anchoring solution is applied followed by AX anchoring solution, and then gelling, all at 4 ⁰C. The tissue is softened with proteinase K softening solution, pGk13a is fluorescently labeled, and RNAs are labeled with a standard FISH hybridization chain reaction (HCR) protocol. umExM methods and probes of the invention may be used to simultaneously visualize membranous structures along with proteins and RNAs, with a standard confocal microscope. Biotin-DHPE, BODIPY-lipid, mCling staining and proExM umExM methods and probes of the invention can be used in conjunction with staining methods such as, but not limited to: BODIPY-lipid staining methods and mCling staining methods. In some embodiments, the biotin-DHPE, BODIPY, or mCling staining methods may be processed using an art-known proExM method. Higher resolution imaging with umExM umExM methods of the invention may include by imaging samples with super- resolution imaging methods, or by increasing the expansion of the sample. Some embodiments of umExM methods of the invention comprise a super-resolution imaging method. A non-limiting example of a super-resolution-imaging method is a super-resolution imaging based on autocorrelation with two-step deconvolution (i.e., SACD) in which a SACD algorithm is used to resolve fluctuations. Some embodiments of umExM methods of the invention comprise an iterative form of umExM, in which the sample is be expanded 2, 3, 4, 5, or more times. Analysis Certain embodiments of methods of the invention may be used to analyze structure, spatial organization, and sequence of one or more ultrastructural membrane components and/or biomolecules of interest that may be known or may be suspected of being associated 41 #17358430v1 with a disease or condition. Some embodiments of methods of the invention can be used to identify a biomolecule or ultrastructural membrane component associated with a disease or condition. Non-limiting examples of diseases and conditions that can be assessed using embodiments of the invention are: disease and conditions such as but not limited to: foodborne illness, food poisoning associated conditions; bacterial infection; viral infections; parasitic infections; poisoning; contamination and/or poisoning with one or more toxins and heavy metals; sickle cell anemia; hemophilia; cystic fibrosis; Tay Sachs disease; Huntington’s disease; fragile X syndrome; chromosomal disorders such as but not limited to: Down syndrome and Turner syndrome; polygenic disorders such as but not limited to Alzheimer’s disease, heart disease, cancers, and diabetes, etc. Methods of the invention can also be used in forensic examination. Tissues Methods and probes of the invention may be used on a tissue sample. As used herein the term “tissue sample” means a tissue obtained from a source, such as a subject or cultured source. In some embodiments, a tissue sample comprises a slice of a tissue. A non-limiting example of a tissue slice is a slice of a brain. In some embodiments, the tissue is a fixed tissue. Methods and probes of the invention may be used on tissue slices 50 to 100 microns thick, and in some embodiments, may be used on thinner or thicker samples. Methods used with thicker samples may, compared with tissue slices that are 50 to 100 microns think, include longer probe incubation times, a higher concentration of probe, or both, as slice thickness increases. In a non-limiting example, a method of the invention that includes a pGk13a probe and a tissue sample thicker than 100 microns, includes at least a 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 36, 48-hours-long pGk13a incubation time, including all times within the range. In another non-limiting example, a method of the invention that includes a pGk13a probe and a tissue sample thicker than 100 microns, the tissue is incubated in a concentration of the pGk13a probe at a concentration of at least 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, mg/ml, including all concentrations within that range. Subjects and cells A tissue sample used in a method of the invention may be obtained from a subject or from a cultured source. The term “subject” may refer to human or non-human animals, including mammals and non-mammals, vertebrates and invertebrates, and may also be any multicellular organism or single-celled organism such as a eukaryotic (including plants and 42 #17358430v1 algae) or prokaryotic organism, archaeon, microorganisms (e.g., bacteria, archaea, fungi, protists, viruses), and aquatic plankton. A subject may be considered a normal subject or may be a subject known to have or suspected of having a disease or condition. In some embodiments, an organism is a genetically modified organism. In some embodiments, a subject is a plant. As used herein the term “genetically modified” is used interchangeably with the term “genetically engineered.” Cells, tissues, or other sources or samples may include a single cell, a variety of cells, or organelles. It will be understood that a cell sample comprises a plurality of cells. As used herein, the term “plurality” means more than one. In some instances, a plurality of cells is at least 1, 10, 100, 1,000, 10,000, 100,000, 500,000, 1,000,000, 5,000,000, or more cells. A plurality of cells from which biomolecules are obtained for use in methods of the invention may be a population of cells. A plurality of cells may include cells that are of the same cell type. In some embodiments, a cell from which one or more biomolecules are obtained for use in methods of the invention is a healthy normal cell, which is not known to have a disease, disorder, or abnormal condition. In some embodiments, a plurality of cells from which biomolecules are isolated for use in methods of the invention includes cells having a known or suspected disease or condition or other abnormality, for example, a cell obtained from a subject diagnosed as having a disorder, disease, or condition, including, but not limited to a degenerative cell, a neurological disease-bearing cell, a cell model of a disease or condition, an injured cell, etc. In some embodiments, a cell is an abnormal cell obtained from cell culture, a cell line known to include a disorder, disease, or condition. Non-limiting examples of diseases or conditions include disorders, such as sickle cell anemia, hemophilia, cystic fibrosis, Tay Sachs disease, Huntington’s disease, and fragile X syndrome; chromosomal disorders, such as Down syndrome and Turner syndrome; Alzheimer’s disease, heart disease, diabetes; and cancers. In some embodiments of the invention, a plurality of cells is a mixed population of cells, meaning all cells are not of the same cell type. Cells may be obtained from any organ or tissue of interest, including but not limited to skin, lung, cartilage, brain, CNS, PNS, breast, blood, blood vessel (e.g., artery or vein), fat, pancreas, liver, muscle, gastrointestinal tract, heart, bladder, kidney, urethra, and prostate gland. In some embodiments, a cell from which one or more biomolecules are isolated for use in methods of the invention is a control cell. In various embodiments, cells from which one or more biomolecules are isolated for use in methods of the invention may be genetically modified or not genetically modified. 43 #17358430v1 A cell or tissue for use in methods of the invention may be obtained from a biological sample obtained directly from a subject. Non-limiting examples of biological samples are samples of blood, saliva, lymph, cerebrospinal fluid, vitreous humor, aqueous humor, mucous, tissue, surgical specimen, biopsy specimen, tissue explant, organ culture, biological fluid or any other tissue or cell preparation, or fraction or derivative thereof or isolated therefrom, etc. In some embodiments of the invention, one or more biomolecules may be obtained from primary cells, cell lines, freshly isolated cells or tissues, frozen cells or tissues, paraffin embedded cells or tissues, fixed cells or tissues, and/or laser dissected cells or tissues. In some embodiments, a sample for use in a method of the invention is a fixed sample. EXAMPLES Example 1 Studies have been performed that demonstrate use of umExM methods and probes of the invention to achieve dense labeling of membranes, and high-integrity expansion, and to enable imaging of membranous structures using a standard confocal microscope. In some embodiments, methods and probes of the invention may be used to achieve ~60nm lateral resolution and to enable co-visualization of membranous structures in a wide range of brain regions along with proteins and RNAs. umExM methods and probes of the invention enable segmentation of cell bodies, dendrites, and axons (>200nm in diameter) and enable tracing of axons. Studies set forth herein also demonstrate that ~35nm resolution imaging of membrane structures is possible by combining umExM with super-resolution imaging (e.g., SACD) or through an iterative form of umExM. Results of studies set forth herein demonstrate use of 4x umExM methods set forth herein to visualize mitochondria (Fig.4A, B, & G; Fig.5A) and ER (Fig.5B; Fig.6A and Fig.7 panel (iv), and use of the method to reveal some features of cytoplasmic vesicles (i.e., synaptic vesicles; Fig.3C-F). The iterative form of umExM, which provides higher resolution (i.e., ~35nm resolution, Fig.8G) compared to 4x umExM, is shown herein to reveal mitochondria cristae (Fig.9), with an appearance similar to that shown with isoSTED [Schmidt, R. et al. Nano Lett.9, 2508–2510 (2009)]. The cost of pGk13a falls within the price range of commercially available membrane labeling probes used in other ExM technologies (DiD for MAGNIFY [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)] and PacSph for panExM-t [M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022)]; Table 1). In studies set forth herein, methods and 44 #17358430v1 probes of the invention have been used on tissue slices 50 to 100 microns thick. Methods of the invention may also be used on thinner and thicker samples. Methods used with thicker samples may include longer pGk13a incubation times, or a higher concentration of pGk13a, or both, as slice thickness increases. Methods Membrane probe synthesis Membrane probes were commercially synthesized (Anaspec). They were purified to >95% purity. They were aliquoted in 1 mg quantities into tubes, lyophilized to powder, and stored at -20 ⁰C until stock solutions were prepared. Stock solutions were stored at -20 ⁰C until use. Brain tissue preparation for umExM All procedures involving animals were in accordance with the US National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Massachusetts Institute of Technology Committee on Animal Care. Wild type (both male and female, used without regard to sex, C57BL/6 or Thy1-YFP, 6-8 weeks old, from either Taconic or JAX) mice were first terminally anesthetized with isoflurane. Then, ice-cold 1x phosphate-buffered saline (PBS, Corning, catalog no.21031CM) was transcardially perfused until the blood cleared (approximately 25 ml). For all umExM experiments, the mice were then transcardially perfused with 4% PFA + 0.5% CaCl2 fixative solution (see Table 2 “fixative solution”). Fixative solution (prepared fresh and used immediately):
Figure imgf000046_0001
*Chilled on ice before use **Kept on ice during perfusion pGk13a stock solution (prepared at RT and immediately stored at -20 ⁰C): 45 #17358430v1
Figure imgf000047_0001
pGk13a membrane labeling stock solution (prepared fresh and used immediately at 4 ⁰C)
Figure imgf000047_0002
* Chilled on ice before use AX buffer solution (prepared fresh and stored at 4 ⁰C)
Figure imgf000047_0003
* Chilled on ice before use AX stock solution (prepared fresh and used immediately at 4 ⁰C, and stored at -20 ⁰C*)
Figure imgf000047_0004
* Aliquot 20ul into a PCR tube, and store at -20 ⁰C in a sealed container (e.g., 50mL tube) with drying agents (e.g., Drierite) ** Anhydrous DMSO (Thermo Fisher, cat. no. D12345) umExM monomer solution (9.4ml, aliquoted to 10 tubes of 940μl and stored at -20 ⁰C):
Figure imgf000047_0005
46 #17358430v1
Figure imgf000048_0001
*All concentrations are in g/100 ml except 10xPBS. umExM gelling solution (1ml, prepared and gelled at 4 ⁰C):
Figure imgf000048_0002
*All concentrations are in g/100 ml except umExM Monomer Solution. All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015). **To make umExM gelling solution, add 20μl of 4-hydroxy-TEMPO solution (0.005g/ml in water) and 20μl of TEMED solution (0.1g/ml in water) to 940μl of umExM monomer Solution, vortex for 2-3 seconds, add 20μl of APS solution (0.1g/ml in water), vortex for 2-3 seconds, and add 4μl of 1M HCl, and vortex for 2-3 second. umExM Digestion buffer* (100ml, prepared and applied at RT, and stored at 4 ⁰C):
Figure imgf000048_0003
dilution in Digestion buffer. All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015). Trypsin+Lys-C softening solution (2ml, prepared fresh and applied at RT):
Figure imgf000048_0004
Second monomer solution (34.5ml, prepared fresh): 47 #17358430v1
Figure imgf000049_0001
*All concentrations are in g/100 ml. Second gelling solution (50ml, prepared fresh):
Figure imgf000049_0002
*All concentrations are in g/100 ml except cleavable second monomer solution. To make the second gelling solution, 250μl of TEMED solution (0.1g/ml in water) to 34.5mL of second monomer solution, vortex for ~10 seconds, add 250μl of APS solution (0.1g/ml in water), vortex for ~10 seconds. Third monomer solution (9.4ml, aliquoted to 10 tubes of 940μl and stored at -20 ⁰C):
Figure imgf000049_0003
*All concentrations are in g/100 ml except 10xPBS. Third gelling solution (prepared in fresh and applied at RT):
Figure imgf000049_0004
*All concentrations are in g/100 ml. 48 #17358430v1 ** To make the third gelling solution, add 20μl of 4-hydroxy-TEMPO solution (0.005g/ml in water) and 20μl of TEMED solution (0.1g/ml in water) to 940μl of monomer solution, vortex for 2-3 seconds, add 20μl of APS solution (0.1g/ml in water), vortex for 2-3 seconds. List of antibodies used for umExM:
Figure imgf000050_0001
The fixative was kept on ice during perfusion. After the perfusion step, brains were dissected out, stored in fixative on a shaker (~10-20 rpm) at 4 ⁰C for 24 hours for further fixation, and sliced on a vibratome (Leica VT1000S) at 100 μm thickness. For the slicing, the tray was filled with ice-cold PBS, and the tray was surrounded by ice. The slices were then transferred to a 50-ml tube filled with 40 ml of ice-cold quenching solution (100mM Glycine in PBS) on the shaker (~10-20 rpm) at 4 ⁰C, overnight (> 8hrs). The slices were washed 3-4 times with ice-cold PBS on the shaker (~10-20 rpm) at 4 ⁰C, for 1-2 hours each and stored in PBS at 4 ⁰C. umExM for brain tissue slices 1. The fixed tissue slices (as described in the Brain tissue preparation for umExM section) were incubated in membrane labeling solution (Table 2, “pGk13a stock solution”) on the shaker (~10-20 rpm) at 4 ⁰C, overnight (unless otherwise noted, overnight means >16 hours). 2. The fixed tissue slices were then incubated in AX stock solution (Table 2, “AX stock solution”) on the shaker (~10-20 rpm) at 4 ⁰C, overnight. The tissue was then washed 2-3 times in PBS on the shaker (~10-20 rpm) at 4 ⁰C, 1 hour each. 3. The fixed tissue slices were then incubated in gelling solution (Table 2, “umExM gelling solution”) 30 minutes on the shaker (~10-20 rpm) at 4 ⁰C for pre-gelation incubation. During this step, the gelation chamber was constructed similarly as previously described [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)]. In summary, two spacers (VWR, catalog no.48368-085) were placed on a microscope slide (VWR, catalog no.48300-026). The two spacers were separated from each other enough so the brain tissue slice could be placed in between them. The brain tissue slice was placed between the spacers and sliced with a razor blade (VWR, catalog no.55411-050) into two equally sized 49 #17358430v1 half-coronal sections. The lid (VWR, catalog no.87001-918) was then placed on top of the spacers as well as the brain tissue slices. The empty space between the half-coronal sections and spacers was fully filled with the gelling solution. The chamber was transferred to a plastic jar with a lid (Fisher Scientific, catalog no. R685025) at 4 ⁰C to initiate free-radical polymerization for >24 hours. Then, the gelation chamber containing the sample (tissue- embedded hydrogel) was taken out. 4. The sample was trimmed with a razor blade (VWR, catalog no.55411-050) to have two gelled half-coronal sections. Each gel (each half-coronal section) was then transferred from the chamber to a 12-well plate (Fisher Scientific, catalog no. FB012928) that contained proteinase K digestion solution (Table 2, “umExM Digestion buffer”) in the well (2 ml of digestion solution per well per half-coronal section). The gel was then digested at room temperature (RT, 24 ⁰C) on the shaker (50 rpm), overnight. After digestion, the gels were washed 3-4 times in PBS on the shaker (50 rpm) at RT, 30 minutes each. 5. Each sample was labeled with 0.5ml of Cy3 conjugated DBCO (Cy3 DBCO Click chemistry tools, catalog no. A140-1) buffered in PBS at a concentration of 0.1mg/ml on the shaker (50 rpm) at RT, overnight. Then, the samples were washed 2-3 times in PBS on the shaker (50 rpm) at room temperature (RT), 30 minutes each. The samples were then transferred to 4 ⁰C, overnight. 6. The samples were placed 2-3 times in excess water on the shaker (50 rpm) at RT for expansion, 30 minutes each. Immunohistochemistry-compatible umExM For pre-expansion antibody staining (Fig.10), the brain tissue slice was prepared as described in step 1 in the umExM for brain tissue slices section. Then, 1ml of permeabilization solution (0.005%-0.01% of saponin (Sigma Aldrich, catalog no.84510) or triton (Sigma, catalog no. X100), 1% Bovine Serum Albumin (BSA, Sigma Aldrich, catalog no. A3294) in PBS) was applied and held at 4 ⁰C, overnight. Then 10μl of primary antibody, rabbit anti-SV2A (Abcam, catalog no. ab32942), was added to the permeabilization solution, and then held for 24 hours at 4 ⁰C on the shaker (50 rpm). Then, the tissues were washed 3-4 times in PBS at 4 ⁰C, 1 hour each. Next, steps 2-5 in the umExM for brain tissue slices sections were performed. Subsequently, each half coronal slice sample was incubated it in 50 #17358430v1 PBS containing primary antibodies, goat anti-rabbit ATTO 647N conjugated secondary antibody (Rockland Immunochemicals, catalog no.50-194-3924), at a concentration of 5-10 μg/mL at 4 ⁰C for 2-3 days. The samples (tissue-embedded hydrogel) were washed 3-4 times in PBS at RT, 30 minutes each. Finally, step 6 in the umExM for brain tissue slices section was performed. For post-expansion staining (Fig.11), steps 1-3 in the umExM for brain tissue slices section were performed. Then, step 4 in the umExM for brain tissue slices section was performed, but with 2 ml of Trypsin+Lys-C softening solution (Table 2, “umExM Trypsin+Lys-C softening solution”) instead of proteinase K digestion solution, for each half coronal slice sample. Then each half coronal slice sample (tissue-embedded hydrogel) was incubated in PBS containing primary antibodies, rabbi anti-PSD95 (Thermo Fisher, catalog no. MA1-046), at a concentration of 10 μg/ml at 4 ⁰C for 2-3 days. The samples were washed 3-4 times in PBS at RT, 30 minutes each. Then, step 5 in the umExM for brain tissue slices section was performed. Subsequently, each half coronal slice sample was incubated in PBS containing primary antibodies, goat anti-rabbit ATTO 647N conjugated secondary antibody (Rockland Immunochemicals, catalog no.50-194-3924), at a concentration of 5-10 μg/ml at 4 ⁰C for 2-3 days. Then, the samples were washed 3-4 times in PBS at RT, 30 minutes each. Finally, step 6 in the umExM for brain tissue slices section was performed. Antibody staining of fluorescent proteins for umExM The expanded samples, after either proteinase K digestion (steps 1-4 in the mExM for brain tissue slices section) or Trypsin+Lys-C softening treatment (post-expansion antibody staining protocol in Immunohistochemistry-compatible umExM section), were incubated in PBS containing ATTO 647N fluorophore-conjugated nanobody against the green fluorescent protein (GFP, ChromoTek, catalog no. gba647n) or ATTO 488 fluorophore-conjugated nanobody against the green fluorescent protein (GFP, ChromoTek, catalog no. gba488) at a concentration of 10 μg/ml for overnight at 4 ⁰C. The samples were washed 3-4 times in PBS at RT, 30 minutes each. Then steps 5 and 6 in the umExM for brain tissue slices section were performed. umExM with RNA For umExM with RNA (Fig.12), brain tissue slices were prepared as described in step 1 in the umExM for brain tissue slices section. The sample was then incubated in 1mL of LabelX solution9 (10 μL of AcX (ThermoFisher, catalog no. A20770), 10 mg/ml in DMSO, 51 #17358430v1 was reacted with 100 μL of Label-IT Amine Modifying Reagent (Mirus Bio, catalog no. MIR3900), overnight at RT with shaking). Then step 2 in the umExM for brain tissue slices section was performed, but with the 0.05 mg/ml AX in MES buffer (Table 2, “AX buffer solution”) for 24 hours at 4 ⁰C. Then steps 3-6 in the umExM for brain tissue slices section were performed. Next, the standard FISH hybridization chain reaction (HCR) protocol was performed, similar to earlier ExM protocols that visualized RNAs [Chen, F. et al. Nat. Methods 13, 679–684 (2016); Alon, S. et al. Science 371, (2021); Cui, Y. et al. PLoS One 18, e0291506 (2023)]. In particular, the sample (tissue-embedded hydrogel) was incubated in hybridization buffer (10% formamide, 2× SSC) at RT for 0.5-1 h, and ACTB probe was applied (Molecular instruments) at 8nM concentration, overnight at 37 ⁰C (buffered in HCR.v3.0 Wash Buffer [Choi, H. M. T. et al. Development 145, (2018)]). The gel was then washed with HCR v3.0 Wash Buffer for 2-3 times at 37 ⁰C followed by another washing with second washing buffer (5x SSC buffer + 0.1% Tween 20) 30 minutes for 4 times at 37 ⁰C, followed by treating the sample with fluorescently (Alexa 647) labeled HCR hairpin amplifiers (1:100) at RT, overnight. Then the samples were washed with 5× SSCT, 20 minutes for 4 times at RT. The samples were expanded (~3x; similar to the expansion factor of ExFISH9 that used LabelX for anchoring RNAs) with 0.05× SSCT, 10-20 minutes each time, 3 times. Confocal imaging, deconvolution, and visualization Confocal images in the Figures were obtained on an Andor spinning disk (CSU-W1 Yokogawa) confocal system on a Nikon Eclipse Ti-E inverted microscope body with a Zyla 5.5 camera or a Hamamatsu qCMOS camera. A 4x 0.2 NA, 10x 0.45 NA, 40x 1.15 NA, or 60x 1.27 NA lens was used for all imaging. Large-scan imaging was performed with the confocal microscope and then stitched with a shading correction function via the default setting in Nikon element software version 4.0. All confocal images in the main figures were deconvoluted with the Sparse-deconvolution [Zhao, W. et al. Nature Biotechnology 2021 40:440, 606–617 (2021)] software (version 1.0.3) using the software provided in GitHub (github.com/WeisongZhao/Sparse-SIM). Gaussian filter function (sigma=2) in ImageJ (version 1.53q) was applied to all antibody signals (anti-GFP, anti-SV2A and anti-PSD95). The 3D volume renderings of confocal images were generated using the volume viewer or 3D viewer function in ImageJ (version 1.53q). All images were visualized with an auto-scaling function in ImageJ (version 1.53q) except for Figs.13, 22 and 23, for which the same 52 #17358430v1 brightness and contrast with ImageJ software was used to highlight the difference between experimental outcomes. Resolution analysis For the resolution analysis, blockwise Fourier Ring Correlation (FRC) resolution analysis [Culley, S. et al. Nat. Methods 15, 263 (2018)] was adopted to measure the resolution of umExM as well as umExM+SACD and the iterative form of umExM. The pixel size of the umExM, umExM+SACD and iterative umExM images were normalized by the expansion factor, so the resolution would be described in biologically relevant terms. For umExM and the iterative form of umExM images, the same region of umExM samples was imaged twice for independent noise realization. Then the NanoJ-SQUIRREL Fiji plugin [Culley, S. et al. Nat. Methods 15, 263 (2018)] was used to perform FRC resolution analysis. In the case of umExM+SACD images, 40 frames of umExM images were captured and divided into two sets of 20 frames by separating odd and even images, and SACD (see umExM with optical fluctuation imaging section below) was performed to generate two SACD images (each derived from 20 frames). Subsequently, these two images underwent FRC resolution analysis using the same Fiji plugin. The best FRC value obtained across the blocks in each image pair was used to quantify the resolution of umExM, umExM+SACD and the iterative form of umExM. Analysis of the biotin (pGk13b) vs. azide (pGk13a) version of the membrane probe The brain tissue sections were prepared as described in the Brain tissue preparation for umExM section but with 4% PFA (Electron Microscopy Sciences, catalog no.15710) solution instead of 4% PFA + 0.5% CaCl2 (Table 2, “fixative solution”). To compare the biotin version (pGk13b) of the probe with azide version (pGk13a) of the probe, ExM was performed as described in the umExM for brain tissue slices section, but with either pGk13b or pGk13a in step 1 and a typical ExM gelation temperature in step 3 (pre-gelation 4 ⁰C and gelation at 37 ⁰C). For fluorescently labeling the pGk13b and pGk13a, an excessive amount of Cy3-conjugated streptavidin or DBCO was used for a long time (~2 days at RT) to fluorescently label the membrane probes, as much as possible. In particular, 1ml of PBS containing Cy3 conjugated streptavidin (Invitrogen, catalog no. SA1010) was used at a concentration of 0.1mg/ml for 2 days at RT. For fluorescently labeling pGk13a, 1 ml of PBS containing Cy3 conjugated DBCO (Click chemistry tools, catalog no. A140-1) at a concentration of 0.1mg/ml for 2 days at RT was used. Both samples were washed 3-4 times 53 #17358430v1 in PBS at RT, 30 minutes each, and expanded with water. A random region in the hippocampus was then imaged with the confocal microscope with 10x, 0.45NA objective. The mean pGk13a and pGk13b signals were then measured. An unpaired two-sided t-test function in RStudio 2021.09.2+382 with R version 4.1.2 was then performed. Analysis of 37 ⁰C vs.4 ⁰C ExM protocols For control experiments, ExM was performed as described in the umExM for brain tissue slices section but with typical ExM gelation temperature (i.e., gelled at 37 ⁰C for 2 hours in step 3). For 4 ⁰C gelation, ExM was performed as described in the umExM for brain tissue slices section. The samples were then imaged in a random region in the hippocampus with the confocal microscope with 10x, 0.45NA objective. The mean pGk13a from each condition was then measured and an unpaired t-test function in RStudio 2021.09.2+382 with R version 4.1.2 was performed. Signal-to-background analysis The umExM samples were prepared as described in the umExM for brain tissue slices section. To obtain the mean pGk13a signal, a volume covering the depth from z=0 μm to z=100 μm with a z-step size of 0.375 μm (in biological units) was imaged, using a 4x 0.2 NA lens and Zyla 5.5 camera with a 50ms laser exposure time (see below and Table 3 for details) for each z-plane. The laser excitation power (mW) was measured. To do so, a Nikon W1 spinning disk equipped with a four-line laser system was used. Because the 561nm laser line was used for pGk13a signals, and the exposure time for this one laser line was reported, the laser excitation power (mW) of this line was measured. This measurement was performed using a power meter to directly measure the excitation light output from the 4x, 10x, 40x and 60x objective lenses that were used for imaging pGk13a signals throughout See Table 3. Table 3: Laser excitation power (mW) measurements
Figure imgf000055_0001
54 #17358430v1
Figure imgf000056_0001
To obtain the mean background, random empty regions in the gel were imaged with the same imaging conditions (i.e., 4x lens, Zyla 5.5 camera, 50ms laser exposure time) and averaged. Then the mean signal-to-background (S/B) was measured by dividing the mean pGk13a signal captured in the XY plane by the mean background (i.e., mean pGk13a signal/mean background). Subsequently the mean signal-to-background (S/B) ratio was calculated for a single z-plane at various depths within the volume. This was repeated with a 60x, 1.27 NA lens for a volume covering the depth from z=0 μm to z=10 μm, with a z step size = 0.125 μm. Continuity of labeled membrane analysis The umExM samples were prepared as described in the umExM for brain tissue slices section. The ciliary membrane (n=5 separate cilia from two fixed brain slices from one mouse) was randomly traced. The starting point of tracing was chosen randomly. Based on the traced ciliary membrane, the number of gaps were counted with gaps defined as a region with intensity smaller than a 2x standard deviation below the mean along the pGk13a labeled ciliary membrane, which was longer than 60nm (in biological units, the effective resolution of the 60x 1.27NA objective that was used for imaging; Fig.2g). umExM with double gelation (for corpus callosum) For umExM with double gelation (for corpus callosum) (Fig.14), samples were prepared as described in the umExM for brain tissue slices section, except for fluorescently labeling the membrane probe and expansion (step 5-6). Then, the sample was incubated in a non-cleavable gelling solution (Table 4, “Monomer solution”) for 30 minutes on the shaker (~10-20 rpm) at 4 ⁰C for pre-gelation incubation. Table 4. Details of solutions used in certain embodiments of the invention. pGk5b stock solution (prepared at RT and immediately stored at -20 ⁰C):
Figure imgf000056_0002
55 #17358430v1
Figure imgf000057_0001
pGk5b membrane labeling stock solution (prepared fresh and used immediately at 4 ⁰C)
Figure imgf000057_0002
* Chilled on ice before use
Figure imgf000057_0003
* Aliquot 20ul into a PCR tube, and store at -20 ⁰C in a sealed container (e.g., 50mL tube) with drying agents (e.g., Drierite) ** Anhydrous DMSO (Thermo Fisher, cat. no. D12345) Monomer solution aka StockX (9.4ml, aliquoted to 10 tubes of 940μl and stored at -20 ⁰C):
Figure imgf000057_0004
*All concentrations are in g/100 ml except 10xPBS. All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015). Gelling solution (1ml, prepared at 4 ⁰C, gelled at 37 ⁰C):
Figure imgf000057_0005
*All concentrations are in g/100 ml except Monomer Solution. All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015). To make the gelling solution, add 20μl of 4-hydroxy-TEMPO solution (0.005g/ml in water) and 20μl of TEMED solution (0.1g/ml in water) to 940μl of monomer solution, vortex for 2-3 seconds, add 20μl of APS solution (0.1g/ml in water), vortex for 2-3 seconds. Digestion buffer* (100ml, prepared and stored at RT, applied at 37 ⁰C): 56 #17358430v1
Figure imgf000058_0001
To formulate the Digestion solution, dilute Proteinase-K (NEB, cat. no. P8107S) at 1:100 dilution in Digestion buffer. All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015). Fixation Reversal buffer (10ml, prepared at RT and used immediately):
Figure imgf000058_0002
*All stock solutions are formulated in water (Thermo Fisher, cat. no.10977015).
Figure imgf000058_0003
* Cell Signaling Technology **Santa Cruz Biotechnology The sample was then gelled at 37 ⁰C, using the gelation chamber described in step 3 of the umExM for brain tissue slices section. After the gelation, the initial gel was treated with a cleaving solution (50mM sodium metaperiodate in 0.1M sodium acetate buffer, pH 5.0) for one hour on the shaker (~100-150 rpm), at RT. Then the sample was washed 4 times in 100mM glycine PBS on the shaker (~50-100 rpm) at RT, 30 minutes each, and then the sample was washed 3-4 times with PBS on the shaker (~50-100 rpm) at RT, 15 minutes each. The membrane probe was then fluorescently labelled, and the sample expanded as described in steps 5-6 of the umExM for brain tissue slices section. 57 #17358430v1 Accuracy (Rand score) of segmentation and tracing of pGk13a signals umExM was performed with fixed brain slices from Thy1-YFP mice and boosted YFP signals with anti-GFP (as described in the Antibody staining of fluorescent proteins for umExM section). Volumes of a random region in somatosensory cortex L6 and hippocampus dentate gyrus were imaged, with two labels (anti-GFP antibody and pGk13a for membranes). Segmentation To identify neuronal compartments, a maximum-intensity z-projected (max-z projected) image was generated from the anti-GFP channel of the volume. Using this max-z projected image, cell bodies, dendrites, and axons were pinpointed. However, anti-GFP signal alone cannot differentiate between myelinated and unmyelinated axons. Thus, the pGk13a signal was used to assist in identifying myelinated axons, as myelinated axons exhibited strong pGk13a signals compared to unmyelinated axons (Fig.2M for unmyelinated axon; Fig.2O for myelinated axon). Subsequently, several regions of interest (ROIs) were randomly created, each containing a portion of identified neuronal compartments. These ROIs were employed to crop the pGk13a channel and anti-GFP channel of the volume. A single z-plane was randomly selected from the cropped volume, compartments were manually segmented based on pGk13a signals, and then the same compartments were segmented based on anti-GFP signals, all with ITK-SNAP software [Yushkevich, P. A. et al. Neuroimage 31, 1116–1128 (2006)]. The pGk13a-guided segmentation was then quantitively compared to the anti-GFP-guided segmentation using the Rand score [Unnikrishnan, R. et al., IEEE Trans. Pattern Anal. Mach. Intell.29, 929–944 (2007); Arbeláez, P. et al., IEEE Trans. Pattern Anal. Mach. Intell.33, 898–916 (2011)]. This experiment and analysis were repeated (n=3 cell bodies and n=3 dendrites from two fixed brain slices from two mice, and n=5 myelinated axons and n=5 unmyelinated axons from two fixed brain slices from two mice). Tracing Myelinated axons were identified by inspecting anti-GFP signals as well as pGk13a signals in the same way as described above herein. Among the identified myelinated axons, some were randomly selected and traced from z=0 to z=10.5 μm based on the pGk13a signals and the same myelinated axons were also traced based on anti-GFP signals, all with ITK- SNAP software [Yushkevich, P. A. et al. Neuroimage 31, 1116–1128 (2006)]. Specifically, 58 #17358430v1 myelinated axons were traced by annotating the centroid of the myelinated axons with brush size=8 in ITK-SNAP software. The pGk13a-guided tracing was then quantitatively compared to the anti-GFP-guided tracing using the Rand score. This experiment and analysis was repeated (n=3 myelinated axons from two fixed brain slices from two mice). Next, the unmyelinated axons were identified by inspecting anti-GFP signals, as was done for the segmentation study above herein. Then, one was randomly selected and traced from z=0 to z=5 μm based on the pGk13a and anti-GFP signals and the Rand score [Unnikrishnan, R. et al., IEEE Trans. Pattern Anal. Mach. Intell.29, 929–944 (2007); Arbeláez, P. et al., IEEE Trans. Pattern Anal. Mach. Intell.33, 898–916 (2011)] was calculated as was done for myelinated axons. This experiment and analysis was repeated (n=3 myelinated axons from two fixed brain slices from two mice). For tracing myelinated axons in the corpus callosum, umExM with double gelation protocol (Fig.15) was applied to mouse brain tissue section. A random volume (39.25 by 39.25 by 20 μm) was assigned to the corpus callosum. Then webKnossos [Boergens, K. M. et al. Nature Methods 201714:714, 691–694 (2017)] was used to trace n=20 myelinated axons that spanned the entire dataset. umExM with Optical fluctuation imaging (umExM with SACD) The samples were prepared as described in umExM for brain tissue slices. The samples were imaged with Andor spinning disk (CSU-W1 Tokogawa) confocal system with a 60x, 1.27NA objective with either a Zyla 5.5 camera or a Hamamatsu qCMOS, with an optional ×1.5 magnification.20 frames of images (exposure time, 50ms; laser power 90%) were used, which took ~1 second in total. Then the SACD ImageJ plugin as provided in the Github (//github.com/WeisongZhao/SACDj) was used. The plugin with the default hyper- parameters [Zhao, W. et al. Nat. Photonics 1–8 (2023)] (i.e., 1st=10, fourier=2, 2nd=10, order=2, scale=2) was used. Finally, CLAHE was applied for visualization purposes. ECS preservation protocol ECS perfusion was adapted from the published protocol [Lu, X. et al. Cell reports methods 3, (2023)]. The mouse was terminally anesthetized with isoflurane and placed on a dissection tray. The chest was cut open, and a 21-gauge butterfly needle was inserted into the left ventricle. A small incision was made in the right atrium to facilitate outflow. The mouse was perfused transcardially at a flow rate of 10 mL/min using a Masterflex Peristaltic pump. Fresh aCSF was flown for 2-3 minutes to clear out the blood. This was followed by perfusion 59 #17358430v1 with 15% mannitol in aCSF solution for 1 minute, and then a 6% mannitol aCSF solution for 5 minutes. Finally, the mouse was perfused with an ice-cold fixative containing 5% mannitol, 4% paraformaldehyde, 2mM CaCl2, 4mM MgCl2, and 150mM sodium cacodylate buffer (pH 7.4) for 5 minutes. After perfusion, the brain was carefully removed from the skull and placed in a vial containing the same fixative solution. It was then fixed for at least 24 hours with gentle agitation at 4 ⁰C.100µm sections were cut using a Leica VT1000 S vibrating blade microtome and collected in the cold fixative solution. Iterative form of umExM For the iterative form of umExM (Fig.16), umExM samples were prepared as described in the umExM for brain tissue slices section, except for fluorescently labeling the membrane probe (step 5). The expanded samples were incubated in a cleavable re-embedding solution (Table 2, “Second gelling solution”) for 1 hour on a shaker (~50rpm) at RT for pre- gelation incubation. Next, the sample was gelled at 50 ⁰C, for >4 hours, with the same gelation chambers used in step 3 of the umExM for brain tissue slices section. The re- embedded samples were washed 3-4 times in PBS at RT, 30 minutes each. The re-embedded samples were then treated with AX solution and washed in PBS as described in step 2 of the umExM for brain tissue slices section. The samples were trimmed into smaller samples with razor blades and then gelled again with a non-cleavable gelling solution (Table 2, “Third gelling solution”) 30 minutes on the shaker (~10-20 rpm) at 4 ⁰C for pre-gelation incubation. Next, the sample was gelled at 37 ⁰C, overnight, with the same gelation chambers used above. The samples were treated with the cleaving solution (50mM sodium metaperiodate in 0.1M sodium acetate buffer, pH 5.0) for one hour, at RT. Then the samples were washed 4 times in 100mM glycine PBS on the shaker (~50-100 rpm) at RT, 30 minutes each, and then the sample was washed 3-4 times with PBS on the shaker (~50-100 rpm) at RT, 15 minutes each. The membrane probe was then fluorescently labeled, and the sample expanded as described in steps 5-6 of the umExM for brain tissue slices section. Electron microscope imaging, visualization, and analysis To validate pGk5b labeling by electron microscopy (EM), tissue slices of 100 μm thickness were treated with lipid labeling solution as used for mExM, except azide was used instead of biotin as the linkable group of the lipid stain (pGk5a for short):first the tissue was incubated in 1ml of pGk5 (0.1μg/ml in ice-cold PBS) at 4 ⁰C overnight (>16hrs) to let the 60 #17358430v1 labels diffuse and intercalate thoroughly throughout. Subsequently, the sample was washed 2x using PBS at 4 ⁰C for 1 hour each to remove any excess lipid label. Then the sample was placed in 2% PFA, 2% glutaraldehyde in PBS at 4 ⁰C for 6 hours for post-fixation of the sample for EM staining. This fix also served for further EM processing, to preserve the state of ultrastructure [Skepper, J. N., J. Microsc.199, 1–36 (2000); Eltoum, I. et al., J. Histotechnol.24, 173–190 (2001)]. These steps were performed at 4 ⁰C to promote the stability of the lipids and the lipid label in the sample. The sample was moved to 1.8nm undecagold-DBCO conjugate solution (2.5mg/1mL Nanopartz, part no. CK11) at 4 ⁰C for 12 hours. The azide-DBCO chemistry served to link the lipid label with a gold nanoparticle. Thereafter the sample was washed 3x with 0.15M sodium cacodylate buffer at room temperature for 30 minutes each to remove unbound nanoparticles. The samples were sent to the Harvard Medical School Electron Microscopy Core to be stained, then embedded and sliced using a standard EM preparation protocol [Eltoum, I. et al., J. Histotechnol.24, 173– 190 (2001)]. In summary, the tissue was stained with 1% uranyl acetate (UA) for 1 hour at room temperature, embedded in resin, and sliced in ultrathin sections (40nm thickness). As discussed in Fig.17, it was decided a common UA staining protocol (1% UA for 1 hour at RT) would be used to enhance the pGk5a signals on top of signals from gold nanoparticles as UA can react to amino groups of pGk5a. As for control experiments, the protocol was adjusted by replacing pGk5a staining with common osmium staining (1% OsO4 for 1 hour at RT). Samples were imaged on a JEOL 1200EX transmission electron microscope using 80keV transmitted voltage. The images were captured with an AMT 2k CCD camera. Acquired images were processed with a Gaussian filter (Radius = 1) and Enhance contrast (Saturated pixels = 10.5%) function in ImageJ (version 1.53q). pGk5a treated sample without OsO4 clearly showed membranes (e.g., mitochondrial membrane and vesicle membranes; Fig. 1A) similar to the control experiment (i.e., that is, only with OsO4; Fig.17B) but with slightly lower contrast. Cell preparation A 13-mm-diameter coverslip (Thermo Fisher, catalog no.174950) was inserted into one well of a 24-glass well plate (Cellvis 24 WELL GLASS BTTM PLATE 20/CS, catalog no. NC0397150). Then, either HEK293 or HeLa or U2OS cells were plated in the well (~40k cells/ml in cell culture medium (described in next paragraph) per well.) The plate was then moved to a humidified cell culture incubator (set at 37°C, 20% oxygen, and 5% CO2) for at least 6 hours for cells to adhere. The cells were fixed with 4% paraformaldehyde (PFA) and 61 #17358430v1 0.1% glutaraldehyde in Dulbecco's 1x phosphate buffered saline (PBS) at room temperature (RT) for 15 minutes. Fixed cells were washed 4 times with PBS for 10 minutes each at 4 ⁰C, and kept in PBS at 4 ⁰C. The HEK293 cell culture medium used was: Dulbecco’s modified Eagle’s medium (DMEM, Corning, catalog no.10013CV) supplemented with 10% heat-inactivated fetal bovine serum (FBS, Thermo Fisher, catalog no. A3840001), 2 mM GlutaMax (Thermo Fisher Scientific, cat. No.3505006), and 1% penicillin-streptomycin (Thermo Fisher, catalog no. 15140122). The HeLa cell culture medium used was DMEM supplemented with 10% heat- inactivated fetal bovine serum (FBS) and 1% penicillin-streptomycin (Thermo Fisher, catalog no.15140122). The U2OS cell culture medium used was Dulbecco’s modified Eagle’s medium (DMEM, Corning, catalog no.10013CV) supplemented with 10% heat-inactivated fetal bovine serum (FBS, Thermo Fisher, catalog no. A3840001), 1% penicillin-streptomycin (Thermo Fisher, catalog no.15140122) and 1% sodium pyruvate (Thermo Fisher, catalog no. 11360070). Transduction of cells via BacMam virus The adherent cells were prepared as described in the Cell Preparation section herein. The cells were transduced by directly adding 12μl of BacMam reagent (either CellLight™ Mitochondria-GFP, catalog no. C10508 or CellLight™ ER-GFP, catalog no. C10590) to the cell medium. The cells were then placed in the culture incubator overnight (>16hrs). The cells were then fixed and washed as described herein in the cell preparation methods. Brain tissue preparation for mExM All procedures involving animals were in accordance with the US National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Massachusetts Institute of Technology Committee on Animal Care. Wild type (both male and female, C57BL/6 or Thy1-YFP, 6-8 weeks old, from either Taconic or JAX) mice were first terminally anesthetized with isoflurane. Then, ice-cold PBS was transcardially perfused until the blood cleared (approximately 25ml). For all mExM experiments, the mice were then transcardially perfused with 4% paraformaldehyde (PFA) and 0.1% glutaraldehyde in ice- cold PBS. The fixative was kept on ice during perfusion. After the perfusion step, brains were dissected out, stored in fixative at 4 ⁰C for 12 hours for further fixation, and sliced on a vibratome (Leica VT1000S) at 100 μm thickness. For the slicing, the tray was filled with ice- 62 #17358430v1 cold PBS, and the tray was surrounded by ice. The slices were kept in PBS at 4 ⁰C overnight for washing and storing. mExM for cells 1. The fixed cells (as described in the cell preparation methods herein) were incubated in the pGk5b solution (Table 4, “pGk5b membrane labeling stock solution”) at 4 ⁰C overnight. 2. The cells were then incubated in the AcX solution (Table 4, “AcX stock solution”) overnight at 4 ⁰C. Then the cells were washed with ice-cold PBS 2 times, 30min each at 4 ⁰C. 3. The cells were then incubated in the gelling solution (Table 4, “Gelling solution”) 30min at 4 ⁰C for pre-gelation. During this step, the gelation chamber was constructed as described previously3. In summary, two spacers (VWR, catalog no.48368-085) were placed on a microscope slide (VWR, catalog no.48300-026). The spacers were separated from each other enough so an adherent cell-containing cover glass could be placed in between them. The adherent cell-containing cover glass was then placed between the spacers on the slide. The lid (VWR, catalog no.87001-918) was placed on top of the spacers, covering the cell-containing cover glass. Then the empty space between the cells and spacers was fully filled with the gelling solution. Next, the chamber was transferred to a 37 ⁰C incubator to initiate free-radical polymerization. After 2 hours, the gelation chamber containing cells was taken out 4. The gel was trimmed with a razor blade (VWR, cat. no.55411-050) and transferred from the chamber to a 6-well plate (Thermo Fisher, catalog no.140675) that contained proteinase K digestion buffer (Table 4, “Digestion buffer”) in the well (3mL of digestion buffer per well). The gel was then digested at 37C on a shaker overnight (> 16 hours). After digestion, the gel was washed 4 times in PBS at RT, 30 minutes each. 5. The digested gels were labeled with 0.3mg/ml of streptavidin labeled with Atto 565 (Atto 565-Streptavidin; Sigma Aldrich, catalog no.56304-1MG-F) buffered in PBS overnight at RT, and then washed 4 times in PBS at room temperature (RT), 30 minutes each. 6. The gels were placed 4 times in excess water at RT for expansion, 30 minutes each. mExM for brain tissue slices 63 #17358430v1 1. The fixed tissue slices (as described in the Brain tissue preparation for mExM section herein) were incubated in a lipid labeling solution (Table 4, “Lipid labeling stock solution”) at 4 ⁰C overnight (>16 hours) to let the labels diffuse and intercalate thoroughly throughout the tissue slices. 2. The tissue slices were then incubated in an AcX stock solution (Table 4, “AcX stock solution”) overnight (>16 hours) at 4 ⁰C. The tissue was then washed 2 times in PBS at 4 ⁰C, 1 hour each. 3. The tissue slices were then incubated in gelling solution (Table 4, “Gelling solution”) 30min at 4 ⁰C for pre-gelation incubation. During this step, the gelation chamber was constructed as previously described7. In summary, two spacers (VWR, catalog no. 48368-085) were placed on a microscope slide (VWR, catalog no.48300-026). The two spacers were separated from each other enough so the brain tissue slice could be placed in between them. The brain tissue slice was placed between the spacers. The lid (VWR, catalog no.87001-918) was placed on top of the spacers as well as the brain tissue slice. The empty space between the brain tissue slice and spacers were fully filled with the gelling solution. The chamber was transferred to a 37 ⁰C incubator to initiate free-radical polymerization. After 2 hours, the gelation chamber containing the tissue was taken out. 4. The gel was trimmed with a razor blade (VWR, cat. no.55411-050) and transferred from the chamber to a 6-well plate (Thermo Fisher, catalog no.140675) that contained proteinase K digestion buffer (Table 4, “Digestion buffer”) in the well (4mL of digestion buffer per well). The gel was then digested at 37C on a shaker overnight (>16 hours). After digestion, the gel was washed 4 times in PBS at RT, 30 minutes each. 5. The digested gels were labeled with 0.3mg/ml of streptavidin labeled with Atto 565 (Atto 565-Streptavidin; Sigma Aldrich, catalog no.56304-1MG-F) buffered in PBS overnight at RT, and then washed 4 times in PBS at room temperature (RT), 30 minutes each. 6. The gels were placed 4 times in excess water at RT for expansion, 30 minutes each. Immunohistochemistry-compatible mExM The aforementioned mExM steps were carried out the same, except for the digestion step (i.e., step 4 in the mExM for cells and mExM for brain tissue slices sections). Instead of using the proteinase K digestion buffer, the sample was heated in fixation reversal (FR; Table 64 #17358430v1 1, “Fixation Reversal buffer”) buffer for 30 minutes at 100 ⁰C and then held for 2 hours at 80 ⁰C. The FR buffer consisted of 0.5% PEG20000, 100mM DTT, 4% SDS, in 100mM Tris pH8. After this, the FR-digested sample was washed in 1x PBS 4 times at RT for 1 hour before proceeding to the immunohistochemistry steps. The expanded gels were first blocked with MAXblock Blocking Medium (Active Motif, catalog no.15252) for 4-6 hours at room temperature and incubated in MAXbind Staining Medium (Active Motif, catalog no.15251) containing primary antibodies at a concentration of 10 μg/ml overnight at 4 ⁰C. Then, the sample was washed with MAXwash Washing Medium (Active Motif, catalog no.15254) at RT 4 times, 30 minutes each and subsequently incubated in secondary antibodies buffered in MAXbind Staining Medium at a concentration of 10 μg/ml for 10-12 hours at 4 ⁰C. Finally, the secondary antibodies were washed, again, with MAXwash Washing Medium at RT 4 times, 30 minutes each time. For primary antibodies, anti-TOM20 (Cell Signaling Technology, catalog no.42406S, rabbit; Santa Cruz Biotechnology, catalog no. sc-17764, mouse), anti-NUP98 (Cell Signaling Technology, catalog no.2597S, rabbit), anti-myelin basic protein (MBP; Cell Signaling Technology, catalog no.78896S, rabbit; Abcam, catalog no. ab40390, rabbit; AVES, catalog no. AB_2313550, chicken), were used. For secondary antibodies, anti-chicken Alexa Fluor Plus 488 (Thermo Fisher, catalog no. A32931), anti- rabbit Alexa Fluor Plus 488 (Thermo Fisher, catalog no. A32731), and anti-mouse Alexa Fluor Plus 647 (Thermo Fisher, catalog no. A32728) were used. After antibody staining, the pGk5b probes that were conjugated to the gel were then labeled with 0.3mg/ml of streptavidin labeled with Atto 565 (Atto 565-Streptavidin; Sigma Aldrich, catalog no.56304- 1MG-F) buffered in PBS overnight at RT, and then washed 4 times in PBS at RT, 30 minutes each. Finally, the gel was placed 4 times in excess water at RT for expansion, 30 minutes each. Antibody staining of fluorescent proteins for mExM The expanded samples, after either proteinase-k digestion or high-temperature softening, were incubated with MAXblock Blocking Medium (Active Motif, catalog no. 15252) for 4-6 hours at room temperature and incubated in MAXbind Staining Medium (Active Motif, catalog no.15251) containing fluorophore-conjugated primary antibody against the green fluorescent protein (GFP) at a concentration of 10 μg/mL overnight (> 16 hours) at 4 ⁰C. Anti-GFP (Thermo, catalog no. A-21311) was used for the primary antibody. Next, the sample was washed with MAXwash Washing Medium (Active Motif, catalog no. 15254) at RT 4 times, 30 minutes each. After antibody staining, the lipid labels that were 65 #17358430v1 conjugated to the gel were then labeled with 0.3mg/ml of streptavidin labeled with Atto 565 (Atto 565-Streptavidin; Sigma Aldrich, catalog no.56304-1MG-F) buffered in PBS overnight at RT, and then washed 4 times in PBS at RT, 30 minutes each. Finally, the gel was placed 4 times in excess water at RT for expansion, 30 minutes each. Expansion factor and degree of isotropy analysis For umExM, expansion factor and degree of isotropy analysis were carried out with cells. mExM protocol for cells is described mExM for cells method section herein. umExM protocol for cells is similar as mExM for cells but using the pGk13a membrane labeling stock solution for step 1, AX stock solution for step 2, umExM gelling solution for step 3, umExM Digestion buffer for step 4, and pGk13a was fluorescently labeled with 1ml of Cy3 conjugated DBCO (Cy3 DBCO Click chemistry tools, catalog no. A140-1) buffered in PBS at a concentration of 0.03mg/ml on the shaker (50 rpm) at RT, overnight (>16 hours). The samples were expanded with water. The expansion factor was evaluated as previously described [Chen, F. et al., Science 347, 543–548 (2015); Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)]. In particular, HEK293 and U2OS cells transfected with BacMam viruses expressing GFP proteins targeted to the matrix of mitochondria were used. Two landmarks were randomly chosen in pre-expansion images and the corresponding landmarks found in expanded-cell images, and the ratio was calculated. Distortion was evaluated as previously described [Chen, F. et al., Science 347, 543– 548 (2015); Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)]. In summary, BacMam virus was used to express GFP proteins in the matrix of mitochondria in HEK293 and U2OS cells. The cell was imaged with SIM before expansion and re-imaged the same region after expansion with a confocal microscope. The SIM image and the confocal image were non-rigidly registered, then the root-mean-square (RMS) length measurement error was calculated as a function of measurement length for SIM vs. expanded-cell images. For the iterative form of umExM, expansion factor was measured with slices of fixed mouse brain. The gel size was measured before and after expansion (i.e., after 2nd round of expansion) and divided the measured gel size to obtain expansion factor. Colocalization analysis for mExM images A colocalization analysis for mExM was performed by adopting recommended colocalization methods for light microscopy studies [Bolte, S. & Cordelières, F. P. J. 66 #17358430v1 Microsc.224, 213–232 (2006); Lunde, A. & Glover, J. C. Scientific Reports 202010:110, 1– 26 (2020)]. The foreground and background fluorescence of GFP channels of mExM images was first segmented using the Otsu image processing algorithm [Otsu, N. IEEE Trans. Syst. Man Cybern. SMC-9, 62–66 (1979)] as previously done for segmenting signals [Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022); Valdes, P. A. et al. Sci. Transl. Med.16, eabo0049 (2024)]. A binary signal mask was then created based on the foreground signals and the signal mask was used to segment the pGk5b signals. The fraction of expressed GFP and antibody signals that had pGk5b signals was evaluated by counting the pixels containing pGk5b signals that were above 1x standard deviation below the mean of the pGk5b signal intensity in the image. The analysis was performed with RStudio 2021.09.2+382 with R version 4.1.2. ExM with fGk5b and fGk5b+pGk5b. The mExM protocol described above herein was followed. For ExM with fGk5b, 10mM fGk5b, was used, which was the same concentration as the membrane probe used in the mExM protocol (10mM pGk5b), and the rest of the procedure was identical to that of mExM. Similarly, for ExM with fGk5b + pGk5b, 5mM fGk5b + 5mM pGk5b or 10mM fGk5b + 10mM pGk5b was used, and for ExM with pGk5b, 10mM pGk5b was used, while the rest of the downstream protocol remained the same as mExM. umExM with GMA anchoring The GMA anchoring protocol was adapted from the published protocol [Cui, Y. et al. PLoS One 18, e0291506 (2023)]. Briefly, step 1 in the umExM for brain tissue slices methods section herein was performed. Then 0.1% GMA in 100 mM sodium bicarbonate was used on a shaker (~10-20 rpm) at room temperature, overnight. Then 3-6 the umExM method in the brain tissue slices method section herein were performed. Biotin-DHPE, BODIPY-lipid, mCling staining and proExM Biotin-DHPE The Biotin-DHPE followed by tyramide signal amplification protocol was adopted from ref [Wang, U.-T. T. et al. Sci. Rep.13, 21922 (2023)]. In summary, 4% PFA fixed mouse brain tissue was incubated in biotin-DHPE solution (Biotium, catalog no.60022; 0.1 mg/ml in 50% ethanol) overnight at RT. Premixed ABC solution was prepared by using equal amounts of avidin and biotin from the VECTASTAIN ABC-HRP Kit via mixing them in 67 #17358430v1 TBS at a 1:50 dilution and incubated for 30 minutes. During this time, biotin-DHPE-tissue was extensively washed at RT with PBS to remove unbound biotin-DHPE. The premixed ABC solution was then added to the biotin-DHPE-labeled tissue for 1 hour, followed by extensive washing at RT with PBS. After washing, Alexa Fluor-555-conjugated tyramide solution (Invitrogen, catalog no. B40955) in TRIS (1:100 dilution) containing hydrogen peroxide (0.03%; note that ref [Wang, U.-T. T. et al. Sci. Rep.13, 21922 (2023)] used amplification dilution buffer which is currently discontinued), was applied to the tissue for 20 minutes at RT, and followed by extensive washing at RT with PBS. BODIPY-lipid The BODIPY-FL-C12 (BODIPY) staining protocol was adopted from ref [Cui, Y. et al. PLoS One 18, e0291506 (2023)]. In summary, 4% fixed mouse brain tissue was incubated in 10 μg/mL BODIPY (Invitrogen, catalog no. D3822) in PBS at RT, overnight. BODIPY stained tissue was then extensively washed with PBS. mCling The mCling staining protocol was adopted from ref [Damstra, H. G. J. et al. Elife 11, (2022)]. In summary, 4% fixed mouse brain tissue was incubated in 5 μM BODIPY (Synaptic Systems, catalog no.710006AT1) in PBS at RT, overnight. mCling stained tissue was then extensively washed with PBS. proExM After treating tissue with either biotin-DHPE, BODIPY, or mCling staining protocols, they were processed using the standard proExM [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)] protocol. In brief, tissue was incubated in 0.1 mg/ml of AcX (Thermo Fisher, catalog no. A20770), then gelled with the gelling solution (see Table 4), digested with the digestion solution (see Table 4), and finally expanded with water. Results Design of ultrastructure membrane expansion microscopy chemistry To develop a membrane labeling probe that labels membranes densely enough to support nanoscale resolution imaging and allow continuous tracing of membranous structures, with ExM chemistry, an unnatural synthetic amphiphilic membrane labeling probe was designed with the following features. First, the membrane labeling probe should exhibit 68 #17358430v1 lipophilicity, similar to traditional fluorescent lipophilic dyes like DiI, to enable its preferential localization and diffusion within membranes [Honig, M. G. & Hume, R. I. Trends Neurosci.12, 333–341 (1989)]. The lipophilic hydrocarbon side chains of DiI, for example, are inserted into the hydrophobic regions of membranes [Bruce, L. L. J. Neurosci. Methods 73, 107–112 (1997)]. Second, the membrane labeling probe should have a chemical handle that allows for selective conjugation of fluorophores subsequent to the formation of the ExM polymer. This design ensured that the membrane labeling probe remained small in size, facilitating its diffusion and preventing potential degradation of the fluorophore during free-radical polymerization of the ExM gel [Tillberg, P. W. et al. Nature Biotechnology 2016 34:934, 987–992 (2016)]. Third, the membrane labeling probe should have a polymer- anchorable handle to incorporate into the ExM gel network for physical expansion. These three features collectively are included in the design of membrane labeling probes that achieve both dense membrane coverage and compatibility with ExM chemistry, allowing for nanoscale imaging of membranous structures with a standard confocal microscope. The probe design proceeded in two phases – a preliminary phase and a final phase. The preliminary phase was used to explore certain aspects of chemical space, and to validate certain aspects of dense membrane staining in ExM. The final phase was then used to refine the properties of the stain for optimal performance, and to perform an even more detailed validation of the density of the membrane staining possible. During the preliminary phase, the membrane labeling probe was designed to contain a chain of lysines with primary amines for binding to a polymer-anchorable handle, such as acryloyl–X (AcX) [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)], previously used to anchor protein amines to the ExM hydrogel [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)]. To achieve membrane labeling, a lipid tail was included on the amine terminus of the lysine chain, with a glycine in between, to provide mechanical flexibility [Yan, B. X. & Sun Qing, Y. J. Biol. Chem.272, 3190–3194 (1997)]. D-lysines were used, rather than the biologically typical L-lysines, to minimize degradation during the chemical softening step of ExM, which in its most popular form involves a proteinase K softening step [Tillberg, P. W. et al. Nature Biotechnology 201634:9 34, 987–992 (2016)]. Finally, a chemical handle was attached to the carboxy terminus of the lysine chain, for selective conjugation to fluorophore(s) after expansion. In the preliminary design, palmitoyl and biotin were selected as the lipid tail and chemical handle, respectively, and five D-lysines were included in the backbone. This design resulted in a glycine and penta-D-lysine peptidic backbone, with a palmitoyl group on the amine-terminus and a biotin 69 #17358430v1 on the carboxy-terminus. This preliminary probe was named: pGk5b (palmitoyl-glycine-(D- lysine)5-biotin). Electron microscopy (EM) was used to validate the preliminary probe design (with biotin replaced with azide (and denoted pGk5a) so that gold nanoparticles could be added via click chemistry for EM imaging) and results indicated that membranes were labeled (see Fig. 17). In this study, an azide version of pGk5b (referred to herein as pGk5a) was used to conjugate gold nanoparticle-DBCO for EM imaging, instead of using pGk5b with gold nanoparticle-streptavidin. This approach was selected because applying streptavidin to the tissue sample typically requires detergent to remove membranes, which could impact downstream processing. Mouse brain tissue sections were incubated with 100 μM of the membrane probe pGk5a (pGk5b, with an azide replacing the biotin). The specimens were post-labeled with gold nanoparticles modified with a dibenzocyclooctyne (DBCO) handle, for EM visualization (see further detail elsewhere herein). The resulting specimens were then imaged with EM. Similar details of intact membranes, organelles, and synapses (Fig.17A), were seen when compared to classical OsO4 membrane visualization (Fig.17B), in EM. Note that EM sample processing, involving PFA+glutaraldehyde fixation followed by osmium staining, has been optimized over decades [Chen, F. et al., Science 347, 543–548 (2015); Chang, J. B. et al. Nature Methods 201714:614, 593–599 (2017); Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)], and the clear visualization of organelles is protocol-dependent. Early EM protocols often yielded images where mitochondrial cristae and synaptic vesicles were challenging to identify (see Fig 6, 9, &10 in [Yu, C. C. et al. Elife 9, 1–78 (2020)]). Similarly, further optimization of EM processing and membrane probe treatment could lead, in principle, to clear visualization of, for example, vesicles and mitochondrial cristae through the lipid stain and EM imaging. pGk5b was applied to a standard cell line (HEK 293), ExM was performed, and samples were imaged with a confocal microscope (unless otherwise noted, a spinning disk confocal microscope was used throughout), and labeling of membranes (Fig.4A-B) was observed. This preliminary method, using pGk5b, was called membrane expansion microscopy (mExM). The isotropy of mExM and expansion factor was evaluated as commonly done with ExM technologies [Chen, F. et al., Science 347, 543–548 (2015); Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)] and the observed distortion and expansion factor were comparable to previous ExM protocols (Fig.4C-L). 70 #17358430v1 The isotropy of mExM expansion was evaluated by quantitatively comparing pre- expansion structured illumination microscopy (SIM) images to post-expansion confocal images, of the same sample, and calculating the distortion across the images. Fixed U2OS cells expressing mitochondrial matrix-targeted GFP were imaged with SIM (see further detail elsewhere herein; Fig.4C, without anti-GFP labeling; given that resolving the mitochondrial matrix vs. membrane requires 30 nm resolution [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016); Damstra, H. G. J. et al. Elife 11, (2022)], and a classical ExM (that expands ~4x) offers ~60-70 nm resolution [M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022); Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)], matrix-targeted GFP is indistinguishable from mitochondrial membrane in the context of the current experiment. mExM was then performed on, and imaged in, the very same cells with a confocal microscope. Comparing pre-expansion SIM images of mitochondrial matrix-targeted GFP to post-expansion images of either GFP (with anti-GFP labeling for boosting GFP signals), or pGk5b, indicated the same low distortion (a few percent, over ~10 μm) as was found for previous ExM protocols (see further detail elsewhere herein; Fig.4F and Fig.4I). By comparing the distance between two landmarks in pre- vs. post-expansion images (Fig.4J and Fig.4K) of the same sample, the expansion factor could be calculated; an expansion factor (~4.4, Fig.4L) was obtained, which was similar to what was previously reported [M’Saad, O. et al. bioRxiv 2022.04.04.486901 (2022); Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)]. Almost all of the pixels exhibiting reference indicators (e.g., mitochondrial matrix- targeted GFP, which is indistinguishable from mitochondrial membrane after ~4x expansion; it requires ~30 nm resolution to distinguish them [Stephan, T., et al., Scientific Reports 2019 9:19, 1–6 (2019); Schmidt, R. et al. Nano Lett.9, 2508–2510 (2009)], and ER membrane- targeted GFP), also exhibited pGk5b labeling (Fig.5). Mitochondrial matrix-targeted GFP or endoplasmic reticulum (ER) membrane- targeted GFP was expressed in HEK293 cells via BacMam virus. mExM was performed on the cells Fig.5A-B; see further detail elsewhere herein), and then imaged the expanded cells, in order to quantify the fraction of mitochondrial matrix- and ER membrane-targeted fluorescent protein signal that also exhibited pGk5b signal (see further detail elsewhere herein; in summary, a pixel was considered pGk5b-positive if it was brighter than one standard deviation below the pGk5b mean that was measured across the whole images). As a result, it was observed that >99% of the mitochondrial matrix-targeted and ER membrane- 71 #17358430v1 targeted GFP signals also exhibited pGk5b signals (n=3 separate cells from 1 culture; Fig. 5D). Thus, mExM could accurately visualize mitochondria and ER in cells. mExM was compatible with slices of fixed mouse brain and provided more details compared to the unexpanded state (Fig.7) and was compatible with antibody staining. To enable post- expansion antibody staining, a commonly used ExM softening protocol [Chen, F. et al. Nat. Methods 13, 679–684 (2016)] (i.e., SDS solution at a high temperature) was used that could reveal previously unseen structures by preserving protein epitopes through the expansion process [Chen, F. et al. Nat. Methods 13, 679–684 (2016)]. This protocol builds upon post- expansion protein-retention ExM (proExM) protocols [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)], as well as tissue proteomics protocols for formalin-fixed paraffin-embedded (FFPE) tissues [Alon, S. et al. Science 371, (2021); Cui, Y. et al. PLoS One 18, e0291506 (2023)]. In brief, the sample were heated for half an hour at 100 ⁰C and for 2 hours at 80 ⁰C, in a “fixation reversal” (FR) buffer12 containing 0.5% PEG20000, 100mM DTT, 4% SDS, in 100mM Tris pH8 (see Table 4, for details). mExM was applied to fixed brain tissue from mice and antibody staining was performed against organelle-specific membrane-localized proteins including TOM20 for mitochondria, NUP98 for the nuclear pore complex, and MBP for myelin, and results indicated in all cases that >98% of the pixels exhibiting these reference indicators also exhibited pGk5b signals (Fig.18). Using the softening solution (see above herein, and Table 4), mExM was performed with antibody staining (see further detail elsewhere herein), using antibodies against organelle-specific membrane-localized proteins, TOM20 for mitochondria (Fig.18A), and Nup98 for the nuclear pore complex (Fig.18B). IN addition, myelin was labeled using an antibody against myelin basic protein (MBP) (Fig.18C). The fraction of signals from antibodies against membrane-localized proteins that also contained pGk5b signals was quantified as was done for membrane-targeted GFP in cultured cells (see further detail elsewhere herein). Because different antibodies may not react to the same sites on the same target protein [Chen, F. et al. Nat. Methods 13, 679–684 (2016)], multiple antibodies against the same protein for TOM20 (two separate vendors) and MBP (three separate vendors) were used to further validate the method technology. These results indicated it was possible to make a label capable of supporting low- distortion, high-fidelity (as reflected by organelle reference marker colocalization) membrane staining for ExM tissue processing, but some issues remained – for example, the plasma membrane, key to tracing the boundary of neuronal processes, remained hard to see. 72 #17358430v1 Having finished the preliminary phase of the project, the next step was to optimize mExM further. Membrane probes with saturated (palmitoyl [Linder, M. E. & Deschenes, R. J. Nat. Rev. Mol. Cell Biol.8, 74–84 (2007)]) and unsaturated (farnesyl [Boutin, J. A. Cell. Signal.9, 15–35 (1997)]) lipids were compared, while keeping the rest of the probe design constant, and it was observed that the palmitoylated probe achieved a denser membrane labeling compared to the farnesylated one (Fig.19A-B). Furthermore, using a mixture of palmitoylated and farnesylated probe did not achieve denser membrane labeling (Fig.19C). Omitting the glycine linker caused a loss of detail (Fig.6). Finally, the total number of lysines in the backbone of the membrane labeling probe was varied. To explore this, a series of probes were prepared in which the number of lysines (i.e., 3, 7, 9, 11, 13, and 19 lysines) in the backbone of the probe were varied while holding other moieties known to be useful (i.e., palmitoyl tail, glycine, and biotin) constant. These probes were applied to slices of fixed mice brain and ExM was performed. The probe containing 13 or more lysines appeared to show the boundary of neuronal processes the best (Fig.20). A probe with 13 lysines (pGk13b) was selected to minimize probe size, to facilitate its diffusion throughout brain tissue. To confirm whether the probe labeled the boundary of neuronal processes, pGk13b was applied to slices of fixed Thy1-YFP mouse brain, which expresses cytosolic yellow fluorescent protein (YFP) under the Thy1 promoter in subsets of neurons [Feng, G. et al. Neuron 28, 41–51 (2000)]. ExM was performed and pGk13b was fluorescently labeled with Cy3-conjugated streptavidin, treating the sample with anti-GFP (many fluorescent proteins survive proteinase K softening [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)]; anti-GFP binds YFP) to boost YFP signals. It was observed that YFP-filled processes were flanked by pGk13b staining Fig.21), confirming the successful visualization of neuronal boundaries (i.e., plasma membranes) with a standard confocal microscope. Use of azide as a chemical handle was explored (resulting in a reagent referred to as pGk13a) instead of biotin (the aforementioned pGk13b) to increase membrane signals in the context of ExM. pGk13b + Cy3-streptavidin (for which each streptavidin bears more than one Cy3, according to the vendor) was compared with pGk13a + Cy3-DBCO (exhibiting one Cy3 per DBCO) in the context of ExM imaging of the hippocampus in fixed mouse brain slices. It was determined that the mean signal of pGk13a was >2x higher than that of pGk13b (Fig.13). Probe pGk13a (palmitoyl-glycine-(D-lysine)13-azide, Fig.1A), was used for the rest of the studies. 73 #17358430v1 It was reasoned that preserving membrane integrity in the sample was critical for achieving dense labeling of membranes via pGk13a. However, achieving this was not trivial, in part because many lipids [Harayama, T. & Riezman, H. Nature Reviews Molecular Cell Biology 201819:519, 281–296 (2018)] are not fixed through standard paraformaldehyde (PFA) chemical fixation [Bullock, G. R. J. Microsc.133, 1–15 (1984)]. To better preserve membrane integrity, a small amount (0.5%) of calcium chloride was added to 4% PFA fixative. A consistent temperature of 4 ⁰C (a cold temperature at which lipids are more ordered, thus reducing the possibility of them diffusing out of the sample [Mabrey, S. & Sturtevant, J. M. Proceedings of the National Academy of Sciences 73, 3862–3866 (1976)]) was maintained throughout tissue processing until the completion of ExM gel formation, to mitigate potential lipid loss, as higher temperatures can exacerbate this process [Seifert, U. Adv. Phys.46, 13–137 (1997)]. To assess whether this was helpful, brain slices were prepared from mice that were fixed with 4% PFA and 0.5% CaCl2 at 4 ⁰C, and standard ExM (37 ⁰C gelation) or modified ExM procedure (4 ⁰C gelation) were performed, and hippocampal regions were imaged with a confocal microscope, with results indicating the mean signal of pGk13a from the modified ExM procedure (4 ⁰C gelation) was ~50% higher than from the standard ExM procedure (37 ⁰C gelation) (Fig.22). The protocol was finalized as follows: mouse brain fixed in 4% PFA and 0.5% CaCl2, the brain sectioned, excess aldehydes quenched with a commonly used 100 mM glycine 1x phosphate-buffered saline (PBS) solution, pGk13a applied at 150 μM, applied a previously established biomolecule anchoring solution (acrylic acid N-hydroxy succinimide ester (AX, a reagent that is smaller, more cost-effective, yet functionally analogous to AcX [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)] in the context of ExM) in MES buffer, pH 6.0) [Shen, F. Y. et al. Nature Communications 202011:111, 1–12 (2020)] to the pGk13a labeled tissue, and finally cast the expandable hydrogel in the tissue — all at 4 ⁰C. The sample was then softened with proteinase K softening solution [Chen, F. et al., Science 347, 543–548 (2015); Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)], the pGk13a was fluorescently labeled via click-chemistry, and the sample expanded with water. This protocol, which used probe pGk13a, (Fig.1A) and optimized ExM protocol (Fig.1B), was called umExM. Validation of umExM The isotropy of umExM was validated by quantitatively comparing pre-expansion structured illumination microscopy (SIM) images to post-expansion confocal images of the 74 #17358430v1 same sample and calculating the distortion across the images as was done for membrane expansion microscopy (mExM) above herein (see further detail elsewhere herein; Fig.4C-I). In summary, fixed cells expressing mitochondria matrix-targeted GFP with SIM were imaged, umExM was performed, and the same cells imaged with a confocal microscope. Comparing pre-expansion SIM images of expressed GFP (Fig.2A) to post-expansion images of anti-GFP (Fig.2B), or of pGk13a (Fig.2C), the same low distortion (Fig.2D-E) as was found in previous ExM protocols [Chen, F. et al., Science 347, 543–548 (2015); Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)] was observed. By comparing the distance between two landmarks in pre- vs. post-expansion images of the same sample, the expansion factor could be calculated; and it was possible to obtain an expansion factor (~4x) similar to what was previously reported [Chen, F. et al., Science 347, 543–548 (2015); Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)] (Fig.2F). Finally, studies were performed to determine how DBCO-Cy3 itself might contribute to membrane labeling, as DBCO is lipophilic. It was observed that umExM without pGk13a staining did not reveal overt staining, when compared to umExM images acquired with pGk13a (Fig.23). Having established the isotropy of umExM expansion, studies were then performed to examine whether this was sufficient to resolve known ultrastructural features previously reported using EM or super-resolution microscopy. To explore this, the effective resolution of umExM was measured via Fourier Ring Correlation (FRC) resolution analysis [Banterle, N. et al., J. Struct. Biol.183, 363–367 (2013); Culley, S. et al. Nat. Methods 15, 263 (2018)], a gold-standard method which uses Fourier transformation of images to measure resolution, on pGk13a signals from expanded samples. umExM was applied to fixed brain slices from mice and imaged the hippocampus was imaged with a resolution of ~60 nm (Fig.2G) obtained with a 60x, 1.27NA water objective, similar to the previously reported effective resolution of ExM protocols with similar expansion factor [Chen, F. et al., Science 347, 543–548 (2015); Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)]. To explore ultrastructural features, umExM was applied to fixed brain slices from Thy1-YFP mice and the YFP signals were boosted with anti-GFP treatment. The hippocampal dentate gyrus (Fig. 2H-I), third ventricle (Fig.2J), and somatosensory cortex layer (L) 6 (Fig.2K-L) were then imaged. Axons were identified by examining pGk13a signal flanking anti-GFP signals (Fig. 2M). The diameter of unmyelinated axons (i.e., in the dentate gyrus; Fig.2N) and myelinated axons (i.e., in the somatosensory cortex; Fig.2O-P), were quantified and axon diameters were determined to be comparable to those obtained from the same brain regions imaged with EM [Claiborne, B. J. et al., J. Comp. Neurol.246, 435–458 (1986); Innocenti, G. M. et al., Cereb. 75 #17358430v1 Cortex 24, 2178–2188 (2014); Wang, S. S. H. et al. Journal of Neuroscience 28, 4047–4056 (2008)]. The diameter of axons is known to be diverse across brain regions [Sun, D.-E. et al. Nat. Methods 18, 107–113 (2021)]. However, results of studies described herein aligned with measured axon diameters from EM images of the same brain regions (i.e., cortex and dentate gyrus) [Sun, D.-E. et al. Nat. Methods 18, 107–113 (2021); Götz, R. et al. Nat. Commun.11, 6173 (2020); Wen, G. et al. ACS Nano 14, 7860–7867 (2020); Wang, U.-T. T. et al. Sci. Rep. 13, 21922 (2023)]. Motile cilia were also identified in the third ventricle by their fingerlike morphologies (Fig.2Q); their diameter (Fig.2R) was comparable to previous measurements made using EM [Reynolds, M. J. et al. Sci. Rep.8, 7977 (2018)]. A volume of the third ventricle was imaged and visualized through 3D volume rendering (Fig.2S), and membrane vesicles (known as extracellular vesicles; arrows in Fig.2S and Fig.24) were found around cilia, similar to what was previously seen with EM [Wood, C. R. & Rosenbaum, J. L.Trends Cell Biol.25, 276–285 (2015); Kesimer, M. et al. FASEB J.23, 1858–1868 (2009)]. The choroid plexus (Fig.25) was also imaged and microvilli were observed, also showing a similar topology to what was previously seen with EM (Fig.2 from ref [Johanson, C. et al. Toxicol. Pathol.39, 186–212 (2011)]). Uniform and continuous labeling of membranes by umExM The uniformity of labeling throughout 3D volumes of umExM-processed slices of mouse brain were evaluated, using a confocal microscope. The studies included investigation of the variation in overall labeling, as quantified by the average signal-to-background ratio (S/B; pGk13a signal divided by the background; background was calculated as the average of images of empty gel regions) of each XY plane, at different depths in the expanded tissue volume. umExM was applied to a 100 μm thick fixed coronal slice of mouse brain (Fig.26A) and large-scan imaging of the expanded sample was performed with a low magnification objective (4x, 0.2NA; Fig.26B) at 30 milliseconds (ms) laser exposure time (see Table 3 for details). A volume [i.e., entire depth, from z=0 μm to z=100 μm with z step size=0.375 μm, in biological units (that is, divided by the expansion factor) throughout] of a random part of the CA1 region with the same objective at 50ms laser exposure time for each z-plane was imaged (Fig.26C). The mean S/B ratio of a single z-plane was then measured at different depths of the volume (Fig.26D) and results indicated a consistently high mean S/B ratio (>40 fold higher than background, Fig.26D) throughout the slice. A volume (from z=0 μm to z=10 μm, with z step size = 0.125 μm) of the dentate gyrus region of the hippocampus was then 76 #17358430v1 imaged with a high magnification objective (60x, 1.27NA water immersion lens) with 100ms laser exposure time (Fig.26E). Nanoscale features, such as neuronal processes, were observed with zooming into the raw dataset (Fig.26F). The same analysis (as in Fig.26D) was performed, and results indicated consistently high mean S/B (>80 fold higher than background, Fig.26G). Neuronal processes were clearly delineated, when zoomed into a cross-sectional image of the volume (Fig.26H). The experiments were repeated (n=3 fixed brain coronal slices from two mice) and similar results observed. A 60x 1.27NA water immersion objective was used to image the somatosensory cortex (L6) and hippocampus [Cornu ammonis (CA) 2, Fig.27; dentate gyrus].100 μm thick coronal slices of fixed brain from mice were used and the expanded samples were imaged using the same imaging conditions (i.e., 60x lens, 100ms laser exposure time) throughout the study unless specified otherwise. The studies next quantified the continuity of labeled membranes. Specifically, studies focused on individual membranes that could be visualized with umExM in the expanded brain samples, such as the ciliary membrane (Fig.2T-V), which could easily be identified since they are not in close apposition to a second membrane. Distinct peaks of pGk13a signals corresponding to ciliary membranes (Fig.2T) were observed. To quantify the continuity of pGk13a labeled membranes, ciliary membrane was manually traced and the number of gaps along them were counted, with a gap defined as a region with intensity smaller than two standard deviations below the mean pGk13a signal along the ciliary membrane, that was longer than 60nm (the effective resolution of umExM using a 60x, 1.27NA water objective; Fig.2G). Results indicated that >97% of the ciliary membrane was continuous by this metric at a gap measurement length of 60nm (Fig.2V). Studies were then performed to compare umExM to prior commercially available membrane probes used for lipid or membrane imaging, namely BODIPY FL C12, mCling and Biotin-DHPE. These probes were applied to tissue. Note that none of these lipid stains were reported for tissue application in the ExM context, except mCling, which, in a study using it in tissue, did not provide much in the way of experimental detail [Gambarotto, D. et al. Nat. Methods 16, 71–74 (2019)]. Given the lack of tissue protocol available for these probes, staining protocols were utilized that had been established for cultured cells. These protocols were applied to standard 4% PFA-fixed tissue, and the most commonly used form of ExM, proExM [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)] were performed (see further detail elsewhere herein). ExM was performed, and the same S/B ratio analysis and continuity analysis as described elsewhere herein were performed. Results 77 #17358430v1 indicated the S/B ratio for umExM images was ~30-39 times higher than that of Biotin- DHPE, BODIPY FL C12, and mCling images (Fig.28A-C). A random part of the CA1 region of the mouse hippocampus was imaged with a 4x objective at 50ms laser exposure time for samples that were stained with biotin-DHPE, BODIPY FL C12, or mCling (Fig. 28A-C). umExM (pGk13a) was also performed for comparison (Fig.28D). Qualitatively, umExM generated the highest contrast image, compared to the others. The signal-to- background (S/B, where the background was determined as the average across images of empty gel regions) was measured for the images obtained from each sample. Results indicated that the S/B for umExM images was many times higher than those of Biotin-DHPE, BODIPY FL C12, and mCling sample images. Results also indicated that the signals of existing membrane probes were not dense enough to trace ciliary membranes, thus the aforementioned continuity analysis was not possible to perform (Fig.28F-G). Cilia in the 3rd ventricle were imaged to perform continuity analysis, as was done in Fig.2V. However, except for the sample stained with mCling, no signals were observed. Furthermore, although mCling was able to visualize cilia to some extent, the signals were not dense enough for membranes to be traced, so continuity analysis was not possible (Fig.28F-G). Visualization of proteins and RNAs with umExM To explore the compatibility of umExM with antibody staining of endogenous protein epitopes, previously established antibody labeling strategies for ExM were adopted, namely pre-expansion antibody staining [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016)] and post-expansion antibody staining [Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)]. Notably, post-expansion staining can reveal previously unknown proteins and even cellular structures [Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)], as antibodies are applied to expanded samples, where densely packed proteins are decrowded, making more room for antibody staining. For umExM with pre-expansion antibody staining, a small amount of detergent (i.e., 0.005%-0.01% of saponin or triton-x) was used to permeabilize membranes in slices of fixed mouse brain tissue, slices were incubated with primary antibody at 4 ⁰C, umExM performed, and then the expanded sample was incubated with a secondary antibody. Using this protocol (Fig.10), umExM was performed with pre-expansion antibody staining against SV2A, a synaptic vesicle marker (Fig.3A-F). Results showed regions of SV2A presence (Fig.3A-B) in hippocampal area CA3. These signals exhibited pGk13a signals (Fig.3C-D), consistent with these signals being from synaptic vesicles (Fig.3E-F). 78 #17358430v1 For umExM with post-expansion antibody staining, a previous softening method [Tillberg, P. W. et al. Nature Biotechnology 201634:934, 987–992 (2016); Cui, Y. et al. PLoS One 18, e0291506 (2023)] that enabled antibody staining after expansion was adapted. In particular, a softening solution that contained site-specific proteases including trypsin and LysC was used, and then immunostaining was performed after sample expansion. Using this protocol (Fig.11), umExM was performed with post-expansion antibody staining, using an antibody against PSD95 (Fig.3G-L). Results indicated a PSD95 expression pattern (Fig.3G- L) similar to previous post-expansion antibody labeling of PSD953. These signals were adjacent to pGk13a signals (Fig.3L, arrows), consistent with the known role of PSD95 as a postsynaptic density protein. To explore whether umExM is compatible with RNA visualization, studies were performed that combined umExM and ExM visualization of RNA (ExFISH). In particular, an RNA anchoring step was added to the umExM protocol using a previously established RNA anchor (i.e., LabelX) [Chen, F. et al. Nat. Methods 13, 679–684 (2016)], so the protocol became as follows: pGk13a was applied to label the membrane, LabelX anchoring solution was applied followed by AX anchoring solution, and then gelling, all at 4 ⁰C. The tissue was then softened with proteinase K softening solution, pGk13a was fluorescently labeled, and RNAs were labeled with a standard FISH hybridization chain reaction (HCR) protocol. Note that studies were performed to investigate the use of glycidyl methacrylate (GMA) [Cui, Y. et al. PLoS One 18, e0291506 (2023)], a previously established reagent for anchoring proteins and RNAs. However, results indicated suboptimal membrane visualization after expansion (Fig.15), suggesting the need for separate optimization in this regard. Therefore, LabelX was selected for use as the RNA anchor for umExM. This protocol (Fig.12) was used to target ACTB mRNA in fixed brain slices (Fig.3M-O; used 40x lens). Results indicated similar gene expression (ACTB) patterns (Fig.3M) as with the earlier ExFISH protocol [Chen, F. et al. Nat. Methods 13, 679–684 (2016); Alon, S. et al. Science 371, (2021); Cui, Y. et al. PLoS One 18, e0291506 (2023)]. Thus, umExM enables simultaneous visualization of membranous structures along with proteins and RNAs, with a standard confocal microscope. Segmenting neuron compartments with umExM Studies were performed to investigate whether umExM could support the segmentation of neuronal compartments (i.e., cell bodies, dendrites, axons) to help with the analysis of signaling proteins within distinct neuronal compartments. As umExM provides ~60nm lateral resolution (Fig.2G), it reasoned that umExM images could capture neuronal 79 #17358430v1 processes that are larger than roughly >120nm (resolution of umExM multiplied by two). To explore this, umExM was applied to fixed brain slices from Thy1-YFP mice, anti-GFP staining was performed to boost YFP signals, and volumes of random regions of somatosensory cortex L6 and hippocampal dentate gyrus were imaged. Cell bodies were randomly selected using anti-GFP signals, they were manually segmented based on pGk13a signals, and then the same cell body was segmented based on anti-GFP signals (Fig.29A) through the commonly used EM image segmentation software, ITK-SNAP [Yushkevich, P. A. et al. Neuroimage 31, 1116–1128 (2006)]. This procedure was repeated for dendrites (Fig. 29B), myelinated axons (Fig.29C), and unmyelinated axons (Fig.29D, see Methods for details). In summary, dendrites and unmyelinated axons were randomly selected using anti- GFP signals, and for myelinated axons, both anti-GFP and pGk13a signals were employed. This combination was necessary because anti-GFP signals alone could not precisely identify myelinated axons, whereas pGk13a signals were effective in pinpointing them (i.e., strong and thick pGk13a signals due to myelin sheaths; Fig.2O). Qualitatively, the morphologies of pGk13a signal-guided segmentations were very similar to anti-GFP signal-guided segmentations (Fig.29A-D). To quantitatively evaluate the accuracy of pGk13a signal- guided segmentation, the Rand score was utilized, which is a recommended and commonly used metric for assessing EM-based imaging segmentations [Unnikrishnan, R. et al., IEEE Trans. Pattern Anal. Mach. Intell.29, 929–944 (2007); Arbeláez, P. et al., IEEE Trans. Pattern Anal. Mach. Intell.33, 898–916 (2011)], with a Rand score of 0 meaning no similarity between the pGk13a signal-guided versus anti-GFP signal-guided segmentations, and a Rand score of 1 meaning segmentations from the two signals are identical. It was observed that pGk13a signal-guided segmentation achieved Rand scores of 0.988 ± 0.015 (n=3 cell bodies from two fixed brain slices from two mice), 0.940 ± 0.004 (n=3 dendrites from two fixed brain slices from two mice), 0.946 ± 0.013 (n=5 myelinated axon from two fixed brain slices from two mice), and 0.890 ± 0.053 (n=5 unmyelinated axon from two fixed brain slices from two mice) for cell bodies, dendrites, myelinated axons, and unmyelinated axons, respectively (Fig.29E). Although it was determined that umExM images could capture and support the segmentation of neuronal compartments, thin processes such as tiny axons (as they can be ~50nm in diameter [Helmstaedter, M. Nature Methods 201310:610, 501–507 (2013)]) and spine necks (known to be ~40-50nm in diameter [Helmstaedter, M. Nature Methods 201310:610, 501–507 (2013)]) cannot be yet resolved, as umExM provides ~60nm resolution (Fig.2G). However, umExM still enables capture and segmentation of neuron compartments that are larger, in fixed tissue, with a standard confocal microscope. 80 #17358430v1 Tracing axons with umExM Studies were then performed to explore manual axon tracing supported by umExM images. To explore this, umExM samples were prepared, volumes of expanded samples were imaged, and myelinated and unmyelinated axons were randomly selected as described above herein. First, pGk13a signals of myelinated axons were traced across the entire image stack (from z=0 to z=10.5μm; Fig.30A, column “pGk13a”) by annotating the centroids of axons in the stacks using the same segmentation software as above (see Methods for details; in summary, brush size=8 was used to manually annotate through the stacks). The tracing was then repeated using the anti-GFP signals (Fig.30A, column "GFP"). The tracing results based on the pGk13a and anti-GFP signals were visually indistinguishable (Fig.30B). The Rand score, the same evaluation metric used above herein, was calculated and results indicated 0.995 ± 0.004 (n=3 myelinated axons from two fixed brain slices from two mice) when anti- GFP-guided tracing was used as the ‘ground truth’. This procedure was repeated for unmyelinated axons in the dentate gyrus (Fig.2M), and results indicated 0.993 ± 0.006 (from z=0 to z=5.0 μm; Fig.30B, n=3 myelinated axons from two fixed brain slices from two mice). However, due to the axial resolution of umExM, which is ~125nm (axial resolution of a confocal microscope divided by expansion factor; ~500nm/4) in principle, tracing unmyelinated axons with pGk13a signals alone posed a limitation beyond z=~5 μm (on average, n=3). Next, the corpus callosum, a brain region containing densely packed myelinated axons, was imaged. However, it was determined that manual tracing of neuronal processes was challenging in this region as only a subset of the processes were visually distinguishable (Fig.16A-B), perhaps due to light scattering; this optical phenomenon was not observed in the somatosensory cortex and hippocampus. Previous studies reported that a subset of native lipids, which causes scattering, may still remain even after tissue clearing [Murray, E. et al. Cell 163, 1500 (2015)] and expansion processes [Klimas, A. et al. Nature Biotechnology 202341:641, 858–869 (2023)]. It was determined that transferring pGk13a and biomolecules to an ExM gel matrix formed post-expansion, and then chemically cleaving the initial ExM gel, exhibited improved visualization of axons in this brain region (Fig.16C-D). In more detail, umExM was performed on a fixed brain slice until the softening step was completed, and then biomolecule anchoring (AX) solution was applied again (so that pGk13 probes in the initial gel could be transferred from the initial gel to a subsequently formed ExM gel; the newly applied AX would react to unreacted amines in pGk13a), an expandable 81 #17358430v1 gel was cast that was prepared with non-cleavable crosslinker N,N - methylenebis(acrylamide) (BIS) in the initial gel, the initial gel (which was made with cleavable crosslinker N,N'-Diallyl-L-tartardiamide (DATD)) was chemically cleaved, pGk13a was fluorescently labeled via click chemistry, and the sample was expended with water. Thus protocol (Fig.14), was used to image corpus callosum covering a volume of 39.25 by 39.25 by 20 μm, at 50ms laser exposure time for each single z section. When zoomed into the dataset, it was possible to clearly identify neuronal processes in the corpus callosum, similar to what was observed in other brain regions such as cortex and hippocampus. This dataset was used to manually trace 20 axons in the bundle of myelinated axons (Fig.30G-I) that spanned the entire dataset without any challenges. Higher resolution imaging with umExM ExM can support higher resolution imaging, by imaging ExM-processed samples with other super-resolution imaging methods [Klimas, A. et al. Nature Biotechnology 202341:6 41, 858–869 (2023); Lee, H. et al., Scientific Reports 202111:111, 1–7 (2021)], or by expanding beyond 4 times, e.g. through iterative forms of ExM [Chang, J. B. et al. Nature Methods 201714:614, 593–599 (2017); Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022); Yu, C. C. et al. Elife 9, 1–78 (2020)]. Studies were performed to assess these possibilities. First, umExM was combined with an existing super-resolution imaging method. Inspired by recent progress in optical fluctuation imaging with ordinary confocal microscopes [Zhao, W. et al. Nat. Photonics 1–8 (2023); Gustafsson, N. et al. Nature Communications 20167:17, 1–9 (2016); Dertinger, T. et al., Proc. Natl. Acad. Sci. U. S. A.106, 22287–22292 (2009)], studies were performed using “super-resolution imaging based on autocorrelation with two-step deconvolution” (i.e., SACD) [Zhao, W. et al. Nat. Photonics 1–8 (2023)], as this method requires fewer frames to resolve fluctuations compared to other methods [Zhao, W. et al. Nat. Photonics 1–8 (2023)]. umExM of fixed mouse brain slices was performed and a confocal microscope used to image 20 frames of a hippocampal region at an imaging rate of 50ms/frame (Fig.8A-B). Then the SACD algorithm [Zhao, W. et al. Nat. Photonics 1–8 (2023)] was used to resolve the fluctuations (Fig.8C). The resolution of the resulting image was measured the same way as for umExM. umExM+SACD provided a final effective resolution of ~33nm (Fig.8D). Next studies were performed to create an iterative form of umExM, adapted from the previously established iterative form of ExM (iExM) [Chang, J. B. et al. Nature Methods 201714:614, 593–599 (2017)]. umExM was performed on fixed brain slices but without fluorescently labeling pGk13a. The expanded 82 #17358430v1 sample was then embedded into a re-embedding gel (uncharged gel) prepared with a cleavable crosslinker (DATD) to preserve the expanded state during subsequent steps [Chang, J. B. et al. Nature Methods 201714:614, 593–599 (2017)], the specimen was treated with biomolecule anchoring (AX) solution again so that the pGk13a probes could be transferred from the initial gel to a subsequent gel, a new expandable gel prepared with a non-cleavable crosslinker (BIS) was cast, and the initial (composed of cleavable crosslinker DATD, as noted above) and re-embedding gels were chemically cleaved, as in the previously established iterative form of ExM [Chang, J. B. et al. Nature Methods 201714:614, 593–599 (2017); Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)]. Finally, pGk13a was fluorescently labeled via click chemistry and the sample expanded. Inspired by recent advancements in extracellular space preservation (ECS) fixation [Lu, X. et al. Cell reports methods 3, (2023)], this protocol (Fig.31) was applied to ECS-preserved fixed brain slices (Fig.8E-F) and achieved ~12x expansion. The resolution of the resulting image was measured as described here, and results indicated the iterative form of umExM achieved a final effective resolution of ~35nm (Fig.8G). With this protocol, mitochondrial cristae were observed (Fig.9), showing similar appearance as seen with earlier super- resolution imaging methods (e.g., Fig 2b from ref [Schmidt, R. et al. Nano Lett.9, 2508– 2510 (2009)]). Example 3 Studies were performed to quantitatively examine the ultrastructure preservation of the umExM by comparing it to known ultrastructure features. To achieve this, umExM was applied to the adult Thy1-YFP mouse brain and it was treated with anti-GFP labeling (i.e., atto647n fluorophore attached GFP Nanobody) for boosting GFP signals, then the cortex L6, third ventricle, and hippocampus dentate gyrus were imaged. Axons were identified by examining the pGk13a signal bearing GFP signals. The diameter of unmyelinated axons (i.e., parallel axons in the dentate gyrus) and myelinated axons (i.e., myelinated axons in the cortex) were quantified, and 0.187 ± 0.03 μm (n=17 axons from 2 brain tissue sections from 2 mice; and 0.553 ± 0.17 μm (n=21 axons from 2 brain tissue sections from 2 mice) respectively, were observed. The finding agrees with previously reported values via EM images [Claiborne, B. J. et al., Journal of Comparative Neurology 246, 435–458 (1986); Wang, S. S. H. et al. Journal of Neuroscience 28, 4047–4056 (2008)] The cilia in the third ventricle was also identified by its fingerlike morphologies and its location in the expanded mouse brain. The diameter of the cilia was measured and 0.246 ± 0.03μm (n=19 cilia from 2 83 #17358430v1 brain tissue sections from 2 mice) was observed, which also agrees with the previously reported value via EM [Sun, S. et al. Proc Natl Acad Sci U S A 116, 9370–9379 (2019)] and stimulated emission depletion microscopy (STED) super-resolution microscopy [Yoon, J. et al. Biophys J 116, 319–329 (2019)]. umExM provides dense labeling of membranes in a wide range of thick mouse brain tissue section that enables visualization of the ultrastructural details. To demonstrate this, umExM protocol was applied to the 100μm thick brain tissue sections from adult C57BL/6 mice and SDCM was used to image a wide range of brain regions including cortex layers, corpus callosum, third ventricle, choroid plexus, and hippocampus. The labeling was dense enough that it only required 100ms laser exposure time to take a single image slice (13.65μm x 13.65μm) with SDCM equipped with a standard sCMOS camera (i.e., Zyla 5.5) and 60x 1.27NA water immersion objective. Due to the dense labeling of membranes, the surface topologies, such as cilia, of the epithelial cells in the third ventricle of the mouse brain could be visualized through 3D volume rendering. The membrane vesicles (known as extracellular vesicles) were also found. There are currently no clear molecule markers (e.g., an antibody) with which to characterize the vesicles using FM [Sun, S. et al. Proc Natl Acad Sci U S A 116, 9370–9379 (2019)], though EM has previously been used to visualize membrane vesicles, (see Fig.5 from ref [Sun, S. et al. Proc Natl Acad Sci U S A 116, 9370–9379 (2019)]) in epithelial cells. Results of the studies described herein demonstrate that methods of the invention were successfully used to visualize membrane vesicles without requiring use of any specific molecule marker for the visualization. As expected, no microtubules in the cilia were observed, as no protein staining (e.g., NHS total protein staining) was performed. Thus, umExM enabled visualization of ultrastructure features, similar to low-resolution EM images, to visualize ultrastructure context of neural circuits. Example 4 Next, the uniformity of labeling throughout the 3D volume of umExM-processed thick tissue (100 µm thick, equivalent to half of a coronal slice) was evaluated, utilizing two approaches: (1) the variations in overall labeling were investigated, as quantified by the signal-to-background (S/B) ratio of the entire XY plane, at different depths of the expanded tissue volume; (2) the changes in signal-to-noise (S/B) ratio along different line profiles within each XY plane were analyzed. To demonstrate this, umExM was applied to the thick (100μm) adult C57BL/6 mouse brain tissue section and it was imaged with low and high magnification objectives (4x objective and 60x NA objective) with varying laser exposure 84 #17358430v1 time (30ms, 50ms, and 100ms), and the signal-to-background (S/B) ratio as described above was measured. First, a large-scale imaging of umExM processed sample with a low magnification objective (4x) with 30ms laser exposure time; obtained images were stitched with shading correction function via default setting from Nikon element software) was performed. Then a large volume (i.e., entire depth, from z=0μm to z=100μm, of a subset of CA1 region) of umExM processed tissue section was imaged with the same low magnification objective but with 50ms laser exposure time, and the pGk13a signal was analyzed. In particular, the background was measured by imaging a region where the tissue was not embedded. Then the mean pGk13a signals across the z-axis were obtained (i.e., from the bottom and top of the sample). The S/B ratio was measured by dividing pGk13a signals by background signals. Consistently high S/B ratio (minimum of ~40 folds higher than the background, was observed across the entire z-axis. Next, the pGk13a signal across the brain regions was analyzed by measuring S/B ratio. In the cortex regions, the pGk13a signal pattern was similar to what was previously observed with large-scale EM imaging [Mikula, S. & Denk, W. Nature Methods 201512:612, 541–546 (2015)]. In particular, a consistently high S/B (minimum of ~20 folds higher than the background, was observed in the cortex regions and a higher S/B (i.e.,~2x) in the corpus callosum region compared to cortex regions. In the hippocampus, consistently high S/B in the CA1 region (minimum of ~20 folds higher than the background, and dentate gyrus region (minimum of ~20 folds higher than the background, was observed, similar to what was observed in the cortex region. Finally, a volume of the dentate gyrus region of the hippocampus was imaged with 60x, 1.27NA water objective with 100ms laser exposure time. Nanoscale features, such as neuronal processes, were observed as images zoomed into the raw dataset. Similar to the analysis result above, consistently high S/B was observed (minimum of ~80 folds higher than the background) across the z-axis from 0μm to 10μm. Thus, umExM enables multiscale imaging and visualization of ultrastructure features across the wide brain regions. Example 4 umExM, as described above herein, is compatible with mouse brain tissue, and provides ultrastructural details similar to low-resolution EM images. However, with proteinase K treatment, proteins would be unavailable for antibody staining except for fluorescent proteins [Honig, M. G. & Hume, R. I. Trends Neurosci 12, 333–341 (1989)]. Certain embodiments of methods of the invention can be used to image antibody-stained proteins and lipids in the same specimen, to localize proteins amidst specific ultrastructural 85 #17358430v1 features, and to help with the analysis of signaling proteins amidst cellular organization. Certain embodiments of methods of the invention can be used for post-expansion staining, retaining proteins through the umExM procedure, as post-expansion staining can reveal previously unknown structures [Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)]. To explore this, previously established antibody labeling strategies for ExM [Tillberg, P. W. et al. Nat Biotechnol 34, 987–992 (2016); Cui, Y. et al. bioRxiv 2022.06.19.496699 (2022) doi:10.1101/2022.06.19.496699] were adopted, namely pre-expansion staining [Scicchitano, M. S. et al., Journal of Histochemistry and Cytochemistry 57, 849–860 (2009)] and post-expansion staining [[Tillberg, P. W. et al. Nat Biotechnol 34, 987–992 (2016); Sarkar, D. et al. Nature Biomedical Engineering 20226:96, 1057–1073 (2022)]. For pre- expansion staining, a small amount of detergent (i.e., 0.005%-0.01% of saponin or triton-x) was used to permeabilize the membrane in thick mice brain tissue, which was stained with primary antibody against sv2a, a pre-synapse marker and synaptic vesicle marker, at 4oC, umExM was performed, and then the sample was incubated with a secondary antibody. In hippocampus CA4 region, although there was not sufficient resolution to resolve individual synaptic vesicles, the pool of synaptic vesicles in the mossy fiber bouton was observed, similar to EM observation [Rollenhagen, A. et al. Journal of Neuroscience 27, 10434–10444 (2007)], and a strong and dense SV2A expression on the pool of synaptic vesicles was found. Although SV2A was expressed in the ultrastructural context, the membranous structures were not well preserved, perhaps due to detergent treatment that were done prior to performing umExM, similar to what was observed with EM imaging that were treated with detergents prior to the osmium staining [Fang, T. et al. Nature Methods 201815:1215, 1029–1032 (2018)]. Next, umExM was performed by adopting an established softening method [Tillberg, P. W. et al. Nat Biotechnol 34, 987–992 (2016); Cui, Y. et al. bioRxiv 2022.06.19.496699 (2022) doi:10.1101/2022.06.19.496699] that used milder enzyme (i.e., Trypsin/Lys-C instead of proteinase-k) for post-expansion antibody labeling. In particular, umExM was performed on thick mice brain tissue but the tissue was softened with Trypsin/Lys-C instead of using proteinase-k, then stained with primary antibody against PSD-95, a post-synapse marker and post-synaptic density marker, followed by secondary antibody incubation. In hippocampus dentate gyrus, although there was not enough resolution to distinguish the clear boundary of synapses to post-synaptic proteins, which is known to be ~20-30nm apart [Chen, X. et al. Proc Natl Acad Sci U S A 105, 4453–4458 (2008)], the pGk13a signals that are aligned with 86 #17358430v1 PSD-95 expression were observed. Furthermore, compared to pre-expansion antibody staining, membranous structures were well preserved, perhaps due to omitting permeabilization step with detergents prior to pGk13a labeling. Studies were performed to examine umExM under various conditions, as a non-limiting example, to examine various antibody incubation times. umExM methods of the invention were determined to be compatible with conventional antibody staining, allowing simultaneous visualization of the biomolecule of interest in the ultrastructural context through dense labeling of membranous structures, with ordinary FM. Example 5 To achieve an ~30nm resolution with a standard confocal microscope, the technology was first checked with another super-resolution microscope (e.g., Airyscan, STED), to ensure there were enough continuous signals from the pGk13a probe at a higher resolution. Airyscan was used to image a mouse brain tissue section that underwent umExM protocol, and it was observed that the method indeed provides continuous signals. The method was then combined with fluctuation imaging methods (i.e., SRRF). Equivalents Although several embodiments of the present invention have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and/or structures for performing the functions and/or obtaining the results and/or one or more of the advantages described herein, and each of such variations and/or modifications is deemed to be within the scope of the present invention. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and/or configurations will depend upon the specific application or applications for which the teaching of the present invention is/are used. Those skilled in the art will recognize or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto; the invention may be practiced otherwise than as specifically described and claimed. The present invention is directed to each individual feature, system, article, material, and/or method described herein. 87 #17358430v1 In addition, any combination of two or more such features, systems, articles, materials, and/or methods, if such features, systems, articles, materials, and/or methods are not mutually inconsistent, is included within the scope of the present invention. All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and/or ordinary meanings of the defined terms. The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.” The phrase “and/or,” as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Other elements may optionally be present other than the elements specifically identified by the “and/or” clause, whether related or unrelated to those elements specifically identified, unless clearly indicated to the contrary. All references, patents and patent applications and publications that are cited or referred to in this application are incorporated by reference in their entirety herein. What is claimed is: 88 #17358430v1

Claims

CLAIMS 1. A probe molecule comprising: (a) a chain comprising 2, 3, 4, 5, 6, 7, 8 or more lysines and/or cystines; (b) a chemical handle; and (c) a polymer-anchorable handle.
2. The probe molecule of claim 1, wherein one or more of the lysines are D-lysines.
3. The probe molecule of claim 1, wherein one or more of the cystines are D-cystines.
4. The probe molecule of claim 1, wherein the chain comprises acrylic acid.
5. The probe molecule of claim 1, wherein the chemical handle comprises azide.
6. The probe molecule of claim 1, wherein the chemical handle comprises biotin.
7. The probe molecule of claim 1, wherein the polymer-anchorable handle is attached to a primary amine of the chain.
8. The probe molecule of claim 1, wherein the polymer-anchorable handle comprises acryloyl–X (AcX).
9. The probe molecule of claim 1, wherein the chain comprises a lipid tail on the amine terminus of the chain.
10. The probe molecule of claim 9, wherein the lipid tail comprises palmitoyl.
11. The probe molecule of claim 1, comprising palmitoyl-glycine-acrylic-lysine-lysine- lysine-lysine-lysine-lysines-lysine-lysine-lysine-lysine-lysine-lysine-lysine-azide.
12. The probe molecule of claim 9, wherein a glycine linker molecule is positioned between the lipid tail and the amine terminus of the lysine chain. 89 #17358430v1
13. The probe molecule of claim 1, wherein the lysine chain comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or more lysines.
14. The probe molecule of claim 1, wherein the lysine chain comprises five lysines.
15. The probe molecule of claim 1, comprising a glycine and penta-lysine D-peptidic chain, with a palmitoyl lipid group on the amine-terminus and a biotin on the carboxy- terminus, wherein the probe molecule comprises: palmitoyl-glycine-lysine-lysine-lysine- lysine-lysine-biotin.
16. The probe molecule of claim 1, wherein the probe molecule comprises pGk5b, pGk5a, pGk13a, or pGk13b.
17. The probe molecule of claim 1, wherein the probe molecule is one of SEQ ID NOs: 1- 16.
18. A method for preparing a biological sample for detecting ultrastructural features of a membrane in the biological sample: the method comprising: (a) contacting the biological sample with a probe of any one of claims 1-17; (b) embedding the contacted biological sample within a polymer material; (c) homogenizing the embedded biological sample; and (d) physically expanding the homogenized embedded sample.
19. The method of claim 18, wherein embedding the biological sample in the polymer material comprises incubating the biological sample in a polymer monomer and polymerizing the monomer.
20. The method of claim 18, wherein the polymer material comprises a swellable polymer material.
21. The method of claim 20, wherein the swellable polymer material comprises an acrylamide-co-acrylate copolymer.
22. The method of claim 18, further comprising polymerizing the polymer material. 90 #17358430v1
23. The method of claim 22, wherein the polymerization comprises polymerization at approximately 4⁰C for a time period of at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more hours.
24. The method of claim 18, wherein the polymer material comprises a non-swellable polymer material capable of conversion to a swellable polymer material, and the method further comprises converting the polymerized non-swellable polymer material into a swellable polymer material.
25. The method of claim 18, wherein if the polymerized polymer is a swellable polymer, a means for the physical expansion of the biological sample comprises contacting the homogenized embedded biological sample with a solvent or liquid that swells the swellable polymer.
26. The method of claim 25, wherein the liquid comprises water.
27. The method of claim 18, wherein homogenizing the embedded biological sample comprises contacting the polymer material in which the biological sample is embedded with one or more enzymes.
28. The method of claim 27, wherein the contacting enzyme is an endoproteinase, optionally wherein the endoproteinase is LysC or Trypsin.
29. The method of claim 28, wherein the homogenization comprises contacting the embedded biological sample with LysC and Trypsin.
30. The method of claim 18, wherein the physically expanding of the biological sample comprises expanding the polymer material in which the biological sample is embedded, wherein the expansion of the polymer material expands the homogenized biological sample isotropically in at least a linear manner within the polymer material. 91 #17358430v1
31. The method of claim 18, wherein the polymer material comprises a hydrogel and a means of expanding the hydrogel comprises contacting the hydrogel with an aqueous solution, optionally water.
32. The method of claim 18, further comprising attaching a detectable label to the probe molecule at one or both of before and after the physical expansion step.
33. The method of claim 18, further comprising re-embedding the physically expanded polymer and homogenized biological sample in either a swellable or a non-swellable polymer.
34. The method of claim 18, further comprising expanding the embedded sample 1, 2, 3, 4, 5, or more times.
35. The method of claim 18, further comprising detecting one or more of a spatial position, a structure, a component of, and an identity of the expanded biological sample.
36. The method of claim 18, further comprising imaging the biological sample after the physical expansion.
37. The method of claim 36, wherein the imaging comprises optical microscopy.
38. The method of claim 37, wherein the microscopy is light microscopy, fluorescence microscopy or electron microscopy.
39. The method of claim 36, wherein the imagining comprises optical fluctuation imaging.
40. The method of claim 18, further comprising detecting one or more proteins and/or RNA in the embedded, expanded sample.
41. The method of claim 18, further comprising antibody staining the expanded sample.
42. The method of claim 18, wherein the biological sample comprises a cell or tissue. 92 #17358430v1
43. The method of claim 42, wherein the cell or tissue is a fixed cell or tissue.
44. The method of claim 42, wherein the cell or tissue is obtained from a subject, optionally wherein the subject is a mammal, and optionally wherein the subject is a human or a rodent.
45. The method of claim 42, wherein the cell or tissue is a cultured cell or cultured tissue.
46. The method of claim 18, wherein the biological sample comprises a cell membrane.
47. The method of claim 47, wherein the tissue is a tissue slice.
48. The method of claim 47, wherein the tissue slice is between about 50 and 100 microns thick.
49. The method of claim 47, wherein the tissue slice is less than 50 microns thick or is more than 100 microns thick.
50. The method of claim 18, further comprising segmenting cell compartments of the expanded sample, optionally wherein the compartments are cell bodies, dendrites and/or axons in the expanded sample.
51. A method of imaging a sample is provided, the method including using a probe of any one of claims 1-17 in an imaging method.
52. The method of claim 51, wherein the imaging method includes imaging an ExM- processed biological sample.
53. The method of claim 51, wherein the imaging method includes light microscopy, and optionally confocal microscopy. 93 #17358430v1
PCT/US2024/043153 2023-08-21 2024-08-21 Ultrastructural membrane expansion microscopy Pending WO2025042943A2 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US202363520702P 2023-08-21 2023-08-21
US63/520,702 2023-08-21

Publications (2)

Publication Number Publication Date
WO2025042943A2 true WO2025042943A2 (en) 2025-02-27
WO2025042943A3 WO2025042943A3 (en) 2025-04-17

Family

ID=94732744

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2024/043153 Pending WO2025042943A2 (en) 2023-08-21 2024-08-21 Ultrastructural membrane expansion microscopy

Country Status (1)

Country Link
WO (1) WO2025042943A2 (en)

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12265004B2 (en) * 2019-11-05 2025-04-01 Massachusetts Institute Of Technology Membrane probes for expansion microscopy
WO2021113505A1 (en) * 2019-12-05 2021-06-10 Massachusetts Institute Of Technology Method for preparing a specimen for expansion microscopy

Also Published As

Publication number Publication date
WO2025042943A3 (en) 2025-04-17

Similar Documents

Publication Publication Date Title
Wassie et al. Expansion microscopy: principles and uses in biological research
Klimas et al. Magnify is a universal molecular anchoring strategy for expansion microscopy
Chang et al. Iterative expansion microscopy
Wen et al. Current progress in expansion microscopy: chemical strategies and applications
JP6456969B2 (en) Expansion microscopy
US11408890B2 (en) Iterative expansion microscopy
Karagiannis et al. Expansion microscopy of lipid membranes
US12265004B2 (en) Membrane probes for expansion microscopy
JP2019528052A (en) Highly multiplexed fluorescence imaging
WO2017027368A1 (en) Protein retention expansion microscopy
US20230332207A1 (en) Epoxides for multimodal detection of biomolecules
US11802822B2 (en) Multiplexed expansion (MultiExM) pathology
US20190256633A1 (en) Swellable and Structurally Homogenous Hydrogels and Methods of Use Thereof
CN120700116A (en) Whole tissue sequencing at clinical and industrial scale
Lee et al. Fluorescent labeling of abundant reactive entities (FLARE) for cleared-tissue and super-resolution microscopy
Shin et al. Dense, continuous membrane labeling and expansion microscopy visualization of ultrastructure in tissues
Wang et al. Universal, label-free, single-molecule visualization of DNA origami nanodevices across biological samples using origamiFISH
Cho et al. Expansion microscopy
JP2024521142A (en) Next-generation volumetric in situ sequencing
Wen et al. Trifunctional linkers enable improved visualization of actin by expansion microscopy
Bianchini et al. Expansion microscopy at the nanoscale: the nuclear pore complex as a fiducial landmark
WO2025042943A2 (en) Ultrastructural membrane expansion microscopy
Michalska et al. Uncovering brain tissue architecture across scales with super-resolution light microscopy
US11333588B1 (en) Matrix-assisted methods and compositions to prepare biological samples for super-resolution imaging
Cui et al. A multifunctional anchor for multimodal expansion microscopy

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 24857212

Country of ref document: EP

Kind code of ref document: A2