WO2024163383A2 - Treating spinal muscular atrophy (sma) by modulating mir34 and use of mir34 as a predictive biomarker of sma - Google Patents

Abstract

The present invention relates to a MiR34 as a biomarker for spinal muscular atrophy (SMA) and therapeutic/prognosis applications. In particular, the present invention provides a method for treating SMA via modulating MiR34. The present invention also provides a method for prognosis of SMA based on a baseline level of MiR34 in an SMA patient before treatment. The present invention further provides a method for predicting an SMA patient's response to treatment against SMA based on the change of a MiR34 level in the patient before and after the treatment or during the period of the treatment.

Description

TITLE OF THE INVENTION

TREATING SPINAL MUSCULAR ATROPHY (SMA) BY MODULATING MIR34 AND USE OF MIR34 AS A PREDICTIVE BIOMARKER OF SMA

REFERENCE TO ELECTRONIC SEQUENCE LISTING

The application contains a Sequence Listing which has been submitted electronically in .XML format and is hereby incorporated by reference in its entirety. Said .XML copy, created on January 29, 2024, is named “20240129-Sequence-Listing” and is 20.6 KB in size. The sequence listing contained in this .XML file is part of the specification and is hereby incorporated by reference in its entirety.

RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. provisional application number 63/441,950, filed January 30, 2023 under 35 U.S.C. §119, the entire content of which is incorporated herein by reference.

TECHNOLOGY FIELD

[0002] The present invention relates to a MiR34 as a biomarker for spinal muscular atrophy (SMA) and therapeutic/prognosis applications. In particular, the present invention provides a method for treating SMA via modulating MiR34. The present invention also provides a method for prognosis of SMA based on a baseline level of MiR34 in an SMA patient before treatment. The present invention further provides a method for predicting an SMA patient’s response to treatment against SMA based on the change of a MiR34 level in the patient before and after the treatment or during the period of the treatment.

BACKGROUND OF THE INVENTION

[0003] Spinal muscular atrophy (SMA) is an autosomal recessive neurodegenerative disease manifesting as degeneration and dysfunction of spinal motor neurons (MNs). SMA is caused by functional defects in the Survival motor neuron 1 (SMN1) gene. SMN1 is a conserved and essential gene in metazoans, with loss of SMN1 function usually resulting in embryonic lethality. ¹ Humans carry a unique hypomorphic paralogue, SMN2. which is 99% identical to SMN1 but harbors a C-to-T nucleotide variant in exon 7 that leads to exon exclusion. ² As a result, only ~ 10% of SMN2 transcripts are complete and translated into functional SMN protein. The limited amount of SMN protein generated from SMN2 transcript allows humans to survive upon loss of SMN1, with SMN2 copy number being a key genetic determinant of SMA disease severity. ^3-5

[0004] Given that SMA is a well-studied monogenetic disease and the existence of near-identical

1

SUBSTITUTE SHEET (RULE 26) SMN2 provides additional targets for intervention, several approved therapies aimed at restoring SMN levels are now being applied to tackle this once incurable disease.⁵ For instance, nusinersen is a splicing-corrective anti-sense oligonucleotide (ASO) that promotes inclusion of SMN2 exon 7, ^7-10 onasemnogene abeparvovec is an SMN gene therapy delivered by self-complementary adeno- associated virus (scAAV), ⁿ’ ¹² and risdiplam is a small molecule that functions as a splicing modifier to promote SMN2 exon 7 inclusion. ^{13, 14} This emerging array of therapeutic options with distinct modes of action is enabling personalized treatment protocols for optimal prognoses. However, several genetic modifiers are known to regulate SMN transcription, splicing, and mRNA stability, ¹⁵‘ ¹⁸ which might explain why SMN2 copy number does not always precisely predict disease severity, despite being the sole source of SMN protein in SMA patients. ¹⁹ Thus, these genetic modifiers could also interact with currently available medications and result in distinct patient responses, ^{20, 21} so potential biomarkers that can reliably predict and monitor treatment responses are sorely needed to facilitate choice of therapeutic plans and enable timely adjustment during the treatment course. ^{22, 23}

SUMMARY OF THE INVENTION

[0005] In the present invention, it is first demonstrated that MiR34 in the spinal cords (motor neurons, MNs) is critical in motor function, and restoration of motor function in SMA animals can be achieved by MiR34 treatment. It is also found that a higher baseline level of MiR34 in SMA patients before treatment is indicative of favorable outcome of treatment, and a trend of decreasing MiR34 during treatment is correlated with improvement of motor function.

[0006] Therefore, in one aspect, the present invention provides a method for treating SMA, the method comprising administrating to a SMA patient an effective amount of a MiR34 molecule. [0007] In some embodiments, the MiR34 includes MiR34a, MiR34b, MiR34c or any combination thereof.

[0008] In some embodiments, administering the effective amount of the MiR34 molecule results in restoring motor function of the subject.

[0009] In some embodiments, the MiR34 molecule is a single-strand RNA molecule or a duplex RNA molecule.

[00010] In some embodiments, the MiR34 molecule is encoded by an expression vector.

[00011] In some embodiments, the MiR34 molecule is administered intravenously, intramuscularly, intranasally or intrathecally.

[00012] In some embodiments, the patient is an infant, a child, an adolescent or an adult. [00013] In some embodiments, the patient has been diagnosed with type 0 SMA, type 1 SMA, type 2 SMA, ty pe 3 SMA or type 4 SMA.

0

SUBSTITUTE SHEET (RULE 26) [00014] Also provided is use of a MiR34 molecule as described herein for manufacturing a medicament for treating SMA. Furter provided is a pharmaceutical composition for use in treating SMA, which comprises a MiR34 molecule as described herein and a pharmaceutically acceptable carrier.

[00015] In another aspect, the present invention provides a method for prognosis of spinal muscular atrophy (SMA) in an SMA patient, the method comprising

(i) providing a biological sample obtained from the patient before treatment; and

(h) detecting a biomarker in the biological sample to obtain a baseline detection level, comparing the baseline detection level with a baseline reference level to obtain a comparison result, and determining the prognosis of the patient based on the comparison result, wherein the biomarker includes a MiR34 and a higher baseline detection level as compared to the baseline reference level is indicative of a positive prognosis..

[00016] In some embodiments, the MiR34 includes MiR34a, MiR34b, MiR34c or any combination thereof; and/or the biomarker further includes pNfH.

[00017] In some embodiments, the positive prognosis includes improvement of motor function. [00018] In some embodiments, the patient is an infant, a child, an adolescent or an adult.

[00019] In some embodiments, the patient has been diagnosed with type 0 SMA, type 1 SMA, type 2 SMA, type 3 SMA or type 4 SMA.

[00020] In some embodiments, the treatment comprises administration of an SMA drug. [00021] In some embodiments, the SMA drug comprises nusinersen, zolgensma or risdiplam.

[00022] In a further aspect, the present invention provides a method of predicting an SMA patient’s response to a treatment, comprising

(i) providing a first biological sample from the patient at a first time point;

(li) providing a second biological sample from the patient at a second time point, which is later than the first time point;

(lii) detecting a biomarker in the first biological sample and the second biological sample to obtain a first detection level and a second detection level, respectively;

(iv) comparing the second detection level with the first detection level to obtain a comparison result, and determining the patient’s response to the therapy regimen based on the comparison result, wherein the biomarker includes a MiR34, the first time point is before the treatment and the second time point is after the treatment, or the first time point and the second time point are after the treatment, and

3

SUBSTITUTE SHEET (RULE 26) a decrease in the second detection level as compared to the first detection level indicates that the patient has positively responded to the therapy regimen.

[00023] In some embodiments, the MiR34 includes MiR34a, MiR34b, MiR34c or any combination thereof; and/or the biomarker further includes pNfH.

[00024] In some embodiments, the patient positively responded to the therapy regimen exhibits improvement of motor function.

[00025] In some embodiments, the patient is an infant, a child, an adolescent or an adult.

[00026] In some embodiments, the patient has been diagnosed with type 0 SMA, type 1 SMA, type 2 SMA, type 3 SMA or type 4 SMA.

[00027] In some embodiments, the treatment comprises administration of an SMA drug.

[00028] In some embodiments, the SMA drug comprises nusinersen, zolgensma or risdiplam.

[00029] In some embodiments, the second time point is at 64 days, 183 days, 482 days or more later than the first time point.

[00030] In some embodiments, the methods described herein can further comprise conducting a proper method for treating SMA, based on the results of prognosis or prediction of response to the treatment.

[00031] Also provided is a kit or composition for performing the method as described herein, comprising a reagent that specifically recognizes the biomarker as described herein, and instructions for using the kit to detect the presence or amount of the biomarker as described herein.

[00032] Further provided is use of a reagent that specifically recognizes the biomarker as described herein for prognosis of SMA, or for predicting an SMA patient’s response to a treatment, or for manufacturing a kit or a composition for prognosis of SMA or for predicting an SMA patient’s response to a treatment.

[00033] The details of one or more embodiments of the invention are set forth in the description below. Other features or advantages of the present invention will be apparent from the following detailed description of several embodiments, and also from the appending claims

BRIEF DESCRIPTION OF THE DRAWINGS

[00034] The foregoing summary, as well as the following detailed description of the invention, will be better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, there are shown in the drawings embodiments which are presently preferred. It should be understood, however, that the invention is not limited to the precise arrangements and instrumentalities shown.

[00035] In the drawings:

4

SUBSTITUTE SHEET (RULE 26) [00036] Figs. 1A to IE show identification of spinal miRNAs. (Fig. 1A) Summary of known possible causes of selective motor neuron degeneration in SMA. PN: proprioceptive sensory neurons; IN: interneuron; MN: motor neuron; NMJ: neuromuscular junction. Adapted from the previous study of Tisdale et al.⁴⁴ (Fig. IB) Schematic illustration of the differentiation process from Hb9::GFP ESCs to spina] MNs and INs. RA: retinoic acid; SAG: Smoothened agonist; ESC: embryonic stem cell; MN: motor neuron; IN: interneuron. (Fig. 1C and Fig. ID) Heatmaps (TPM) presenting miRNA abundances. miRN As enriched in MNs and INs were selected for cross-referencing against postnatal expression patterns and those enriched in the spinal cord.⁴⁹ (Fig. IE) In situ hybridizations of MiR17a, MiR27a, and MiR34a for brachial spinal cord sections taken from postnatal day 5 (P5) wild type mouse pups. Scale bar is 50 pm in (Fig. IE).

[00037] Figs. 2A to 2E show dysregulated spinal miRNAs in SMA iPSC-derived MNs. (Fig. 2A) Schematic illustration of the differentiation process from two type I SMAiPSCs to spinal MNs. iPSC: induced pluripotent stem cells; EB: embryoid bodies; MN: motor neurons. (Fig. 2B and Fig. 2C) MN survival in long-term MN cultures was assessed by quantifying the ISLl/SMI32^on iPSC-derived MNs and determining the ratio to the day 4 population by immunostaining, which revealed a significant decline in SMA MNs after the 7^th -8^th week (quantified in Fig. 2C, statistical analysis of each cell line compared to the Ctrl- 1 iPSC line). (Fig. 2D and Fig. 2E) Expression of spinal miRNAs in SMA MNs relative to Ctrl- 1 MN cultures at week 6 (6W, before MN degeneration, Fig. 2D) and week 8 (8W, after MN degeneration, Fig. 2E). We observed a notable decline in MiR34a levels prior to MN degeneration. Data are shown as mean ± SD of fold-change (FC) relative to the Ctrl- 1 cell line. T. test by simple main effect analysis for Fig. 2C (FC: fold-change; n.s.: not significant; error bars represent SD, n = 3~4 independent experiments; * p-value<0.05, by Student’s t-test). Two tailed t test for (Fig. 2D and Fig. 2E); * denotes P< 0.05.

[00038] Figs. 3A to 3D show dysregulated spinal miRNAs in SMNA7 mice. (Fig. 3A) Immunostainings of spinal cord from Pl, P5, and P10 SMNA7 mice reveals reduced ChAT⁺ MN signal at postnatal day 10 (P10). (Fig. 3B) Quantification of ChAT⁺ MN number in SMNA7 mouse spinal cord. (Fig. 3C) Expression of spinal miRNAs in P10 spinal cord of SMNA7 mice. Data are shown as fold-change (FC) relative to Ctrl mice. (Fig. 3D) In situ hybridization of MiR34a in Pl, P5 and P10 ventral-half spinal cords. Data are presented as mean ± SD; * denotes P< 0.05; Scale bars represent 100 pm in (Fig. 3 A) and 50 pm in (Fig. 3D).

[00039] Figs. 4A to 4H show that Mir34/449 TKO mice display axon swelling and compromised NMJ end-plates. (Fig. 4A) Immunostainings for NMJs in the intercostal muscles of Mir 34/449 TKO and SMNA7 mice using anti-NF and SV2 antibody (green) and a-BTX (red). Abnormal axonal

5

SUBSTITUTE SHEET (RULE 26) swellings are indicated by arrowheads. (Fig. 4B) Quantification (as percentage) of swollen axonal terminals from (Fig. 4A). (Fig. 4C) Representative images of a-BTX-mediated labeling of AChR (red) in intercostal muscles from wild type (WT) and Mir34/449 TKO mice at P7. SMNA7 and control (SMN^+/‘) mice were analyzed in parallel to compare phenotypes. (Fig. 4D) Quantification of NMJ endplate area from (Fig. 4C). (Fig. 4E and Fig. 4G) Immunostaining of laminin (green) in the intercostal (Fig. 4E) and gastrocnemius (GA) muscle (Fig. 4G) of P7 Mir 34/449 TKO mice and their Mir 34^' ; 449" littermate control. Nuclei were labeled with DAPI (white). (Fig. 4F and Fig. 4H) Quantification of the myofiber size from intercostal (Fig. 4E) and GA (Fig. 4G) muscle immunostainings. Myofibers in the Mir 34/449 TKO muscle are smaller than for the Ctrl group. Scale bars represent 10 pm in (Fig. 4A) and (Fig. 4C) and 20 pm in (Fig. 4E) and (Fig. 4G). Data are presented as mean ± SD; one-way ANOVA with Tukey’s multiple comparisons test was performed in (Fig. 4B) and (Fig. 4D); Student t test was applied in (Fig. 4F) and (Fig. 4H). * denotes P< 0.05; "NS” denotes ‘‘not significant”.

[00040] Figs. 5A to 5C show that the MiR34 family regulates synapse formation pathways. (Fig. 5A) Experimental strategy to perform RNA-seq on thoracic and lumbar segments from Mir 34/449 TKO and WT spinal cords at P14. (Fig. 5B and Fig. 5C) Strategy to identify potential MiR34/449 targets in Alzr34/449-depleted spinal cord. The Venn diagram shows overlap of genes upregulated in the thoracic (4021 genes; Fig. 5B) or lumbar spinal cord (2265 genes; Fig. 5C), and the predicted MiR34/449 targets according to TargetScan (666 genes). Differentially expressed genes between Mir 34/449 TKO versus WT were filtered out according to the criteria of p < 0.01 and log2 fold-change > 0.5. Intersecting genes from thoracic segment (n=140) as well as lumbar segment (n=84) are predicted to be in vivo targets dysregulated in the Mir 34/449 TKO spinal cord. Intersecting genes from left panel were subjected to gene ontology analysis in the right panel. Potential pathways related to the synaptogenesis and neuromuscular phenotypes are highlighted in red and green, respectively.

[00041] Figs. 6A to 6H show that neonatal delivery of MiR34a partially rescues the disease phenotype of SMN A 7 mice. (Fig. 6A) Schematic illustration of the MiR34a overexpression experiments using the SMNA7 mouse model. (Fig. 6B) Induction of MiR34a expression by scAAV9 vector in the spinal cord of SMN" mice was verified via qPCR. (Fig. 6C) Immunostaining for NMJs in the intercostal muscles was performed using anti -NF and SV2 antibody (green) and a-BTX (red). Arrowheads indicate swollen axonal terminals. Areas enclosed by the red rectangles have been enlarged to help visualize the size of motor end-plates. Quantitative analyses are presented in bar graphs for (Fig. 6D) the swollen axonal terminals and (Fig. 6E) the size of motor end-plates. Righting reflex latency (Fig. 6F) and success rate (Fig. 6G) were quantified as indexes of motor function for mice at P7. (Fig. 6H) Pearson’s correlation analysis was performed to evaluate the correlation

6

SUBSTITUTE SHEET (RULE 26) coefficient (r) between righting reflex latency and end-plate area. Data are presented as mean ± SD; * denotes P< 0.05; Scale bar: 50 pm.

[00042] Fig. 7 A to 71 show the trend for MiR34 reduction in type I SMA patients following nusinersen treatment. (Fig. 7A) Schematic illustration of the treatment time-course for nusinersen (an ASO). (Fig. 7B) Levels of phosphorylated neurofilament heavy chain (pNfH) in the CSF of type I SMA patients before (Day 0) and after two months (Day 64) of nusinersen treatment. (Figs. 7C-7E) Expression of MiR34a (Fig. 7C), MiR34b (Fig. 7D), and MiR34c (Fig. 7E) in the CSF of type I SMA patients after two months of nusinersen treatment. (Figs. 7F-7H) Correlation between changing HINE- 2 scores over the nusinersen treatment time-course and the change in MiR34b levels after two months (Day 64) of nusinersen treatment (Days 64-Day 0, Fig. 7F) (Days 183-Day 0, Fig. 7G) (Days 482-Day 0, Fig. 7H). Coefficients (r) lower than -0.5 indicate a strong inverse correlation. (Fig. 71) Correlation between changing HINE-2 scores after 482 days of the nusinersen treatment and the basal expression level in CSF of the MiR34 family and pNfH before nusinersen treatment (day 0). Coefficients (r) higher than 0.5 indicate a strong correlation.

[00043] Figs. 8A to 8H show the expression profiles of the candidate miRNAs. Expression profiles for the indicated miRNAs in mouse spinal cord from available data sources ^{42, 89} (miR-17-5p expression in spinal cord, Fig. 8A) (miR-23a-3p expression in spinal cord, Fig. 8B) (miR-24-3p expression in spinal cord, Fig. 8C) (miR-27a-3p expression in spinal cord, Fig. 8D) (miR-34a-5p expression in spinal cord, Fig. 8E) (miR-125b-5p expression in spinal cord, Fig. 8F) (miR-181a-5p expression in spinal cord, Fig. 8G) (miR-218-5p expression in spinal cord, Fig. 8H).

[00044] Figs. 9A to 9D show the establishment and characterization of induced pluripotent stem cells (iPSCs) from a type I SMA patient. (Fig. 9A) Schematic illustration of the SMAiPSC derivation process. Peripheral blood mononuclear cells (PBMCs) were reprogrammed into iPSCs through transient expression of OCT4, SOX2, KLF4, LIN28, and SHP53 via electroporation. Colonies with an ESC-like morphology were then picked at day 16. (Fig. 9B) The iPSCs formed ESC-like colonies under both feeder-dependent and feeder-free conditions. (Fig. 9C) MN markers HB9, LHX3 and FOXP1 are expressed in healthy controls and SMA iPSC derived MNs. (Fig. 9D) Western blotting reveals compromised SMN protein levels in SMA iPSCs relative to healthy control iPSCs, indicating that patient-derived iPSCs could recapitulate the molecular pathology of SMA.

[00045] Figs. 10A and 10B show the representative images illustrating the severe disease phenotype of mutant mice. (Fig. 10A) SMNA7 mice at Pl, P5 and P10, and (Fig. 10B) Mir 34/449 TKO mice at early postnatal stages demonstrating their smaller body size compared to the littermate Ctrl group.

[00046] Figs. HA and 11B show the GO analysis of downregulated genes in Mir34/449 TKO

7

SUBSTITUTE SHEET (RULE 26) mice. Downregulated genes from the thoracic spinal cord (n=2532; Fig. 11A) and lumbar spinal cord (n=1140; Fig. 1 IB) derived as in Fig. 5A were subjected to GO analysis. Differentially expressed genes between Mir 34/449 TKO and WT mice were filtered according to the criteria of p < 0.01 and log2 fold-change < -0.5. Potential pathways related to muscle development and synaptic transmission phenotypes are highlighted in green and orange, respectively.

[00047] Figs. 12A to 121 show the correlation between HINE2 scores and altered miRNA expression in CSF. Correlation between changing HINE-2 scores over the nusinersen treatment timecourse and the change in pNfH (Days 64-Day 0, Fig. 12A) (Days 183-Day 0, Fig. 12B) (Days 482- Day 0, Fig. 12C), MiR34a (Days 64-Day 0, Fig. 12D) (Days 183-Day 0, Fig. 12E) (Days 482-Day 0, Fig. 12F), and MiR34c (G-I) (Days 64-Day 0, Fig. 12G) (Days 183-Day 0, Fig. 12H) (Days 482-Day 0, Fig. 121) levels in CSF after two months (Day 64) of nusinersen treatment. Coefficients (r) lower than -0.5 indicate a strong inverse correlation.

[00048] Fig. 13 show the correlation between the righting reflex latency and spinal MiR34a level of SMNA7 mice. All SMNA7 mice were analyzed, including those treated with vehicle, MiR34a, or with unsuccessful intravenous MiR34a deliveries (n=20).

DETAILED DESCRIPTION OF THE INVENTION

[00049] The following description is merely intended to illustrate various embodiments of the invention. As such, specific embodiments or modifications discussed herein are not to be construed as limitations to the scope of the invention. It will be apparent to one skilled in the art that various changes or equivalents may be made without departing from the scope of the invention.

[00050] As used herein, the singular forms “a”, “an”, and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to “a component” includes a plurality of such components and equivalents thereof known to those skilled in the art.

[00051] The term “comprise” or “comprising” is generally used in the sense of include/including which means permitting the presence of one or more features, ingredients or components. The term “comprise” or “comprising” encompasses the term “consists” or “consisting of.”

[00052] As used herein, the term “about” or “approximately” refers to a degree of acceptable deviation that will be understood by persons of ordinary skill in the art, which may vary to some extent depending on the context in which it is used. In general, “about” or “approximately” may mean a numeric value having a range of ± 10% around the cited value.

[00053] As used herein, the term “nucleic acid fragment,” “nucleic acid” and “polynucleotide,” used interchangeably herein, refer to a polymer composed of nucleotide units, including naturally occurring nucleic acids, such as deoxyribonucleic acid (“DNA”) and ribonucleic acid (“RNA”) as

8

SUBSTITUTE SHEET (RULE 26) well as nucleic acid analogs including those which have non-naturally occurring nucleotides. Thus, these terms include, but are not limited to, single-, double-, or multi-stranded DNA or RNA, genomic DNA, cDNA, mRNA, DNA-RNA hybrids, or a polymer comprising purine and pyrimidine bases or other natural, chemically or biochemically modified, non-natural, or derivatized nucleotide bases. It will be understood that when a nucleic acid fragment is represented by a DNA sequence (i.e., A, T, G, C), this also includes an RNA sequence (i.e., A, U, G, C) in which “U” replaces “T.” [00054] As used herein, the term “primer” as used herein refers to a specific oligonucleotide sequence which is complementary to a target nucleotide sequence and used to hybridize to the target nucleotide sequence. A primer serves as an initiation point for nucleotide polymerization catalyzed by either DNA polymerase, RNA polymerase or reverse transcriptase. For example, primers for CIGALTs and galectin-4, as used herein, respectively, are those which are capable to hybridize to the nucleotide sequence of the individual target genes to initiate nucleotide polymerization and produce the nucleotide products as expected based on the design of the sequences of the primers.

[00055] As used herein, the term “probe” as used herein refers to a defined nucleic acid segment (or nucleotide analog segment, e.g., polynucleotide as defined herein) which can be used to identify a specific polynucleotide sequence present in samples during hybridization, said nucleic acid segment comprising a nucleotide sequence complementary of the specific polynucleotide sequence to be identified. Typically, a probe can produce a detectable signal since it is labeled in some way, for example, by incorporation of a reporter molecule such as a fluorophore or radionuclide or an enzyme. For example, probes for MiR34, as used herein, respectively, are those which are capable to specifically hybridize to the corresponding nucleotide sequence of the individual target gene and produce detectable signals caused by such hybridization.

[00056] As used herein, the term “hybridization” as used herein shall include any process by which a strand of nucleic acid joins with a complementary strand through base pairing. Relevant technologies are well known in the art and described in, for example, Sambrook et al., Molecular Cloning: A Laboratory Manual, 2^nd ed., Cold Spring Harbor Laboratory Press (1989), and Frederick M.A. et al., Current Protocols in Molecular Biology, John Wiley & Sons, Inc. (2001). Typically, stringent conditions are selected to be about 5 to 30°C lower than the thermal melting point (T_m ) for the specified sequence at a defined ionic strength and pH. More typically, stringent conditions are selected to be about 5 to 15°C lower than the T _m for the specified sequence at a defined ionic strength and pH. For example, stringent hybridization conditions will be those in which the salt concentration is less than about 1.0 M sodium (or other salts) ion, typically about 0.01 to about 1 M sodium ion concentration at about pH 7.0 to about pH 8.3 and the temperature is at least about 25°C for short probes (e.g., 10 to 50 nucleotides) and at least about 55°C for long probes (e.g., greater than

9

SUBSTITUTE SHEET (RULE 26) 50 nucleotides). An exemplary non-stringent or low stringency condition for a long probe (e.g., greater than 50 nucleotides) would comprise a buffer of 20 mM Tris, pH 8.5, 50 mM KC1, and 2 mM MgCh , and a reaction temperature of 25°C.

[00057] As used herein, the term “encode” as used herein refers to the inherent property of specific sequences of nucleotides in a polynucleotide (e.g., a gene, a cDNA, or an mRNA) to serve as templates for synthesis of a gene product having either a defined sequence of nucleotides (i.e., rRNA, tRNA and mRNA) or a defined sequence of amino acids and the biological properties resulting therefrom.

[00058] As used herein, the term “expression” as used herein refers to the realization of genetic information encoded in a gene to produce a gene product such as an unspliced RNA, an mRNA, a splice variant mRNA, a polypeptide or protein, a post-translationaly modified polypeptide, a splice variant polypeptide and so on.

[00059] As used herein, the term “expression level” refers to the amount of a gene product expressed by a particular gene in cells which can be determined by any suitable method known in the art.

[00060] As used herein, a biological marker (or called biomarker or marker) is a characteristic that is objectively measured and evaluated as an indicator of normal or abnormal biologic processes/conditions, diseases, pathogenic processes, or responses to treatment or therapeutic interventions. Markers can include presence or absence of characteristics or patterns or collections of the characteristics which are indicative of particular biological processes/conditions. A marker is normally used for diagnostic and prognostic purposes. However, it may be used for therapeutic, monitoring, drug screening and other purposes described herein, including evaluation the effectiveness of a drug for treating a disease.

[00061] As used herein, a biological sample to be analyzed by any of the methods described herein can be of any type of samples obtained from a subject to be diagnosed. In some embodiments, a biological sample can be a body fluid sample such as a blood sample, a cerebrospinal fluid (CSF) or a urine sample. In some embodiments, a biological sample is a cerebrospinal fluid. In other embodiments, a blood sample can be whole blood or a faction thereof e.g. serum or plasma, heparinized or EDTA treated to avoid blood clotting.

[00062] MicroRNAs (miRNAs) are short non-coding RNAs that regulate either mRNA decay or translation efficiency. Unlike coding genes, miRNAs are usually expressed in specific tissues²⁴ and often fine-tune the hubs of gene regulatory networks to confer robustness on functional gene modules. ²⁵ As a result, disease-linked miRNAs are often dysregulated in a tissue or cell typespecific fashion. For instance, a wide spectrum of miRNAs is dysregulated in MN diseases, such as

10

SUBSTITUTE SHEET (RULE 26) amyotrophic lateral sclerosis (ALS) ^26-28 and SMA.^29-31 In MN diseases, miRNAs are not only affected at the molecular pathology level, but also actively regulate key facets of disease, including cell death, neurite outgrowth, and excitotoxicity. ²⁷- ²³- -¹²-³³ Although often specifically expressed in certain cell types, miRNAs are not limited to their host cells, since they also participate in cell-cell communication. Some miRNAs are actively sorted and packaged for exocytosis in various cell types, ^34-36 and miRNA-containing extracellular vesicles can be received by other cell types where they regulate target genes in the recipient cell. ³⁷ Given their cell type-specificity, direct involvement in pathogenesis, and extracellular presence, miRNAs are considered not only critical to understanding the molecular basis of diseases, but also represent promising candidate biomarkers to reflect disease progression and prognosis. ²⁹

[00063] As used herein, “miR-34” refers to one or more of miR-34a, miR-34b and miR34c.

[00064] As used herein, the terms “subject,” “individual” and “patient,” used interchangeably herein, refer to a mammalian subject for whom diagnosis, prognosis, treatment, or therapy is needed, particularly humans. Other subjects may include cattle, dogs, cats, guinea pigs, rabbits, rats, mice, horses, and so on.

[00065] As used herein, the term “diagnosis” as used herein generally includes determination as to whether a subject is likely affected by a given disease, disorder or dysfunction. The skilled artisan often makes a diagnosis on the basis of one or more diagnostic indicators, i.e., a marker, the presence, absence, or amount of which is indicative of the presence or absence of the disease, disorder or dysfunction. It will be understood in the art that diagnosis does not mean determining the presence or absence of a particular disease with 100% accuracy, but rather an increased likelihood of the presence of certain disease in a subject.

[00066] As used herein, the term “prognosis” as used herein generally refers to a prediction of the probable course and outcome of a clinical condition or disease. Prognosis of a patient is usually made by evaluating factors or symptoms of a disease that are indicative of a favorable or unfavorable course or outcome of the disease. It is understood that the term “prognosis” does not necessarily refer to the ability to predict the course or outcome of a condition with 100% accuracy. Instead, the skilled artisan will understand that the term “prognosis” refers to an increased probability that a certain course or outcome will occur; that is, that a course or outcome is more likely to occur in a patient exhibiting a given condition, when compared to those individuals not exhibiting the condition. It would be understandable that a positive prognosis typically refers to a beneficial clinical outcome or outlook, such as improvement or enhancement in motor function or a longer survival rate whereas a negative prognosis typically refers to a negative clinical outcome or outlook, such as declining or losing motor function or a reduced survival rate.

11

SUBSTITUTE SHEET (RULE 26) [00067] As used herein, the term “treatment” refers to the application or administration of one or more active agents to a subject afflicted with a disorder, a symptom or condition of the disorder, or a progression of the disorder, with the purpose to cure, heal, alleviate, relieve, alter, remedy, ameliorate, improve, or affect the disorder, the symptom or condition of the disorder, the disabilities induced by the disorder, or the progression or predisposition of the disorder.

[00068] As used herein, the term “effective amount” refers to the amount of an active ingredient to confer a desired biological effect in a treated subject or cell. The effective amount may change depending on various reasons, such as administration route and frequency, body weight and species of the individual receiving said pharmaceutical, and purpose of administration. Persons skilled in the art may determine the dosage in each case based on the disclosure herein, established methods, and their own experience.

[00069] As used herein, the term “normal individual” may refer to an individual who is healthy and does not suffer from the disease (e.g., SMA), and may refer to a single normal individual or a group of normal individuals.

[00070] As used herein, an “aberrant level” can refer to a level that is increased or decreased compared with a reference level. For example, an aberrant level can be higher or lower than a reference level by more than 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or 100%. In some embodiments, the expression level of a biomarker as described herein in a subject to be tested is compared to a reference level based on historical values. For example, the reference level can be set based on an average or median expression level of such biomarker in corresponding biological samples obtained from a cohort of subjects. For instance, the cohort of subjects can be a group of SMA patients enrolled in a clinical trial. In particular embodiments, the cohort of subjects can be a group of SMA patients in early stage of disease onset without severe progression in phenotype or function. In some embodiments, a reference level can refer to the level measured in normal individuals.

[00071] As used herein, the term “low expression” and “high expression” for a biomarker as used herein are relative terms that refer to the level of the biomarker found in a sample. In some embodiments, low and high expression can then be assigned to each sample based on whether the expression of such biomarker in a sample is above (high) or below (low) the average or median expression level.

[00072] As used herein, the term “response” refers to the clinical response to a drug/treatment of the patients suffering from a disease which is treatable with said drug/treatment. A positive response may describe that patents receiving a drug/treatment achieve the purpose to cure, heal, alleviate, relieve, alter, remedy, ameliorate, improve, or affect the disease and benefit from the

12

SUBSTITUTE SHEET (RULE 26) drug/treatment to obtain improved clinical responses. A positive response may achieve partial or complete treatment of the disease. A non-response as used herein may describe that the predicted response of patients to the treatment/ drug is negative, or absent.

[00073] SMAis an autosomal recessive genetic disorder involving a mutation or deletion in the Survival Motor Neuron 1 (SMN1) gene. Specifically, SMA is caused by decrease in the level of functional SMN protein which is required for maintaining normal, motor neuron function.

Although SMA patients have defects in SMN1 gene, the paralogous gene, SMN2, produces low levels of functional SMN protein that may compensate for the defect of SMN1 and reduce the SMA disease severity. SMA can be classified as type 0, 1, 2, 3 or 4 depending on age of onset. Type 0 SMA is the most severe version of SMA and is diagnosed prenatally with decreased fetal movement in utero. At birth, the infant is very weak and typically requires respiratory and feeding support. Type 1 SMA, also known as Werdnig-Hoffmann disease, the most common form, is usually diagnosed during an infant’s first 6 months. Infants with type 1 SMA never learn to sit independently. Type 2 SMA is usually diagnosed after 6 months of age, but before 2 years of age. Patients with type 2 SMA are able to sit unassisted, but cannot walk without aid. Type 3 SMA, also known as Kugelberg- Welander disease, emerges in children 18 months old or older (before 3 years of age, or in the teenage years). Children with type 3 SMA are initially able to walk, but have increasingly limited mobility as they grow; many need to use a wheelchair later in life. Type 4 SMA is adult onset, mild in phenotype, and very rare.

[00074] According to the present invention, MiR34 in the spinal cords play a critical role in maintaining motor function, and introducing MiR34 improves the motor ability of SMA animals. Therefore, the present invention provides a method for treating SMA by administering MiR34 to a SMA patient.

[00075] In some embodiments, the MiR34 includes MiR34a, MiR34b, MiR34c or any combination thereof.

[00076] In some embodiments, the MiR34 molecule is a single-strand RNA molecule or a duplex RNA molecule.

[00077] In some embodiments, the MiR34 molecule is encoded by an expression vector. Such an expression vector can be constructed by inserting a nucleotide sequence encoding a microRNA into a suitable vector in which the microRNA sequences are in operable linkage with a proper promoter. [00078] Examples of a proper vector include a viral vector such as retroviral, adenoviral, adeno- associated viral (AAV), and lentiviral vector. A recombinant AAV (rAAV) is typically composed of a transgene and its regulatory sequences (e g. a promoter), and 5' and 3' AAV inverted terminal repeats (ITRs). Examples of promoters include but not limited to a retroviral Rous sarcoma virus

13

SUBSTITUTE SHEET (RULE 26) (RSV) LTR promoter (optionally with a RSV enhancer), a cytomegalovirus (CMV) promoter (optionally with a CMV enhancer), a SV40 promoter, a dihydrofolate reductase promoter, a [Lactin promoter, a phosphoglycerol kinase (PGK) promoter and a EFla promoter.

[00079] A mi croRNA molecule as described herein is used as an active ingredient may be formulated in a proper carrier into a composition for purpose of delivery'. As used herein, “pharmaceutically acceptable” means that the carrier is compatible with an active ingredient in the composition, and preferably can stabilize said active ingredient and is safe to the receiving individual. Said carrier may be a diluent, vehicle, excipient, or matrix to the active ingredient. Typically, a composition comprising MiR34 as described herein as an active ingredient can be in a form of a solution such as an aqueous solution e.g. a saline solution or it can be provided in powder form. Appropriate excipients also include lactose, sucrose, dextrose, sorbose, mannose, starch, Arabic gum, calcium phosphate, alginates, tragacanth gum, gelatin, calcium silicate, microcrystalline cellulose, polyvinyl pyrrolidone, cellulose, sterilized water, syrup, and methylcellulose. The composition may further contain pharmaceutically acceptable auxiliary substances as required to approximate physiological conditions, for example, pH adjusting and buffering agents, such as sodium acetate, sodium chloride, potassium chloride, calcium chloride, sodium lactate and the like. The formulations are easily administered in a variety of dosage forms such as injectable solutions, drug-release capsules, and the like. Delivery vehicles such as liposomes, nanocapsules, microparticles, microspheres, lipid particles, vesicles, and the like, may be used for the introduction of the compositions of the present invention into suitable host cells. The composition of the present invention may be delivered via any physiologically acceptable route, such as oral, parenteral (such as intramuscular, intravenous, subcutaneous, and intraperitoneal), transdermal, intrathecal, suppository, and intranasal methods. In some embodiments, the composition of the present invention is administered intravenously, intramuscularly, intranasally or intrathecally.

[00080] According to the present invention, a baseline level of MiR34 in a SMA patient before treatment can be used as a predictor for the outcome of treatment. Specifically, the higher baseline level of MiR34, the better outcome of treatment. Therefore, the present invention provides a method for prognosis of SMA by measuring a MiR34 baseline level before treatment and comparing it with a baseline reference level where a higher MiR34 baseline level as compared to the baseline reference level indicates a positive prognosis.

[00081] In addition, the change of MiR34 level of SMA patients during treatment is correlated with SMA patients’ response to the treatment. Specifically, a trend of decreasing MiR34 during treatment is correlated with improvement of motor function. Therefore, the present invention provides a method for predicting an SMA patient’s response to a treatment, comprising detecting

14

SUBSTITUTE SHEET (RULE 26) MiR34 levels at a first time point and a later second time point, comparing the MiR34 levels at the first time point and the second time point, and determining the patient’s response based on the comparison result, wherein a decreased MiR34 level from the first time point to the second time point indicates a positive response of the treatment.

[00082] In some embodiments, the first time point is before the treatment and the second time point is after the treatment.

[00083] In some embodiments, the first time point and the second time point are after the treatment. [00084] To perform the methods described herein, a biological sample can be obtained from a subject in need and a biomarker in the sample can be detected or measured via any methods known in the art. Assays based on the use of primers or probes that specifically recognize the nucleotide sequence of the gene as the biomarker may be used for the measurement, which include but are not limited to reverse transferase-polymerase chain reaction (RT-PCR) and in situ hybridization (ISH), the procedures of which are well known in the art. Primers or probes can readily be designed and synthesized by one of skill in the art based on the nucleic acid region of interest. It will be appreciated that suitable primers or probes to be used in the invention can be designed using any suitable method in view of the nucleotide sequences of the gene of interest as disclosed in the art.

[00085] In some embodiments, the amount of a biomarker in the sample derived from the candidate individual in need can be compared to a standard value to determine whether the candidate individual has a positive prognosis of SMA or a positive response to the treatment as applied. The standard value may represent the average or median amount of a biomarker as described herein in a population of SMA patients. Typically, such population of SMA patients are chosen to be matched to the candidate individual in, for example, age and/or ethnic background. Preferably, such population of SMA patients and the candidate individual are of the same species.

[00086] Also provided is a kit for performing the method of the invention. Specifically, the kit comprises a reagent (e.g., a primer, a probe, or a labeling reagent) that can specifically detect the marker(s) as described herein. The kit can further instructions for using the kit to detect the presence or amount of the marker(s) in a biological sample for prognosis and/or monitoring patient’s response to SMA treatment. The components including the detection reagents as described herein can be packaged together in the form of a kit. For example, the detection reagents can be packaged in separate containers, e.g., a nucleic acid (a primer or a probe), a control reagent (positive and/or negative), and/or a detectable label, and the instructions (e.g., written, tape, VCR, CD-ROM, etc.) for performing the assay can also be included in the kit. The assay format of the kit can be a Northern hybridization or a chip, for example. Further provided is use of such reagent for performing a method for prognosis and/or monitoring patient’s response to SMA treatment. The reagent may be

15

SUBSTITUTE SHEET (RULE 26) mixed with a earner e.g. a pharmaceutically acceptable carrier to form a composition for the detection or diagnosis purpose. Examples of such carrier include injectable saline, injectable distilled water, an injectable buffer solution and the like.

[00087] The present invention is further illustrated by the following examples, which are provided for the purpose of demonstration rather than limitation. Those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention.

[00088] Examples

[00089] Spinal muscular atrophy (SMA) is a neurodegenerative disease characterized by the selective loss of spinal motor neurons (MNs) and concomitant muscle weakness. Mutation of SMNI is known to cause SMA, and restoring SMN protein levels via antisense oligonucleotide treatment is effective for ameliorating symptoms. However, this approach is hindered by exorbitant costs, invasive procedures, and poor treatment responses of some patients. Here, we seek to circumvent these hurdles by identifying reliable biomarkers that could predict treatment efficacy. Indeed, a previous assessment of miRNAs known to be dysregulated in SMA revealed their value as accessible biomarkers. ³⁰ Although rigorous studies have been conducted to identify biomarkers of SMA in patients and/or animal models that reflect disease severity and treatment responses, ^{22, 23, 38} they have only had limited success and their reliability remains controversial. Thus, we decided to systematically identify miRNAs that reflect MN physiology and pathology to seek potential bona fide biomarkers that could predict SMA treatment efficacy since: (1) expanding our search beyond canonical candidates (e.g., circulating proteins) represents a promising strategy to accelerate the discovery of applicable biomarkers; ³¹’ ³⁹ and (2) spinal cord tissue is usually overlooked in biomarker searches due to the invasiveness of obtaining such tissue. We uncovered that MiR34 exhibits consistent downregulation during SMA progression in both human and rodent contexts. Importantly, Mir34 family -knockout mice display axon swelling and reduced neuromuscular junction (NMJ) end-plates, recapitulating SMA pathology. Introducing MiR34a via scAAV9 improved the motor ability of SMNA7 mice, possibly by restoring NMJ end-plate size. Finally, we observed a consistent decreasing trend in MiR34 family expression in the cerebrospinal fluid (CSF) of type I SMA patients during the loading phase of nusinersen treatment. Baseline CSF MiR34 levels before nusinersen injection proved predictive of patient motor skills one year later. Thus, we propose that MiR34 may serve as a novel biomarker of SMA since it contributes directly to pathology and can help evaluate the therapeutic effects of nusinersen.

16

SUBSTITUTE SHEET (RULE 26) [00090] 1. Material and Methods

[00091] 1.1 Patients

[00092] We enrolled a consecutive cohort of seven type I SMA infants (three males and four females) meeting clinical definitions and genetically confirmed as harboring a defective SMN1 gene, all of whom underwent nusinersen therapy at Kaohsiung Medical University Hospital. Nusinersen administration was scheduled as a standard protocol according to the manufacturer’s instructions. Baseline demographics and clinical motor-function scales of these patients, including HINE-2 scores, were collected each time the patients visited the hospital. HINE-2 is a quantifiable scale to assess motor-function development according to achievement of motor milestones, i.e., voluntary grasp, kicking, head control, rolling, sitting, crawling, standing, and walking. Total HINE-2 scores range from 0 to 26, with higher scores indicating better motor function. The patients were considered as displaying a treatment response if they showed improvements in at least one milestone or if there were more milestones displaying improvements than declining performance. Targeted CSF molecular biomarkers, i.e., pNfH and MiR34, were compared with HINE-2 motor performance, as determined before beginning nusinersen treatment (day 0) and after treatment on days 64, 183, and 482. CSF was then extracted from the patients, the samples were stored frozen at -80 °C, and diluted prior to use to the minimum required concentration using assay dilution buffer (available as part of the respective assay kit, see below). This study was approved by the Institutional Review Board of Kaohsiung Medical Universify Hospital and Academia Sinica. Informed consent was obtained from the guardians of all of the enrolled type I SMA patients.

[00093] 1.2 pNfH measurement

[00094] Levels of phosphorylated neurofilament heavy chain (pNfH) were measured using a commercially-available enzyme-linked immunosorbent assay (ELISA) kit according to the manufacturer’s instructions (Euroimmun, Lubeck, Germany).

[00095] 1.3 Mouse breeding and maintenance

[00096] Heterozygous breeder pairs of SMNA7 SMA model mice (Smn : SMN2 : Smn/17 ) ^NB£g-Grm7^Ts^fSMN2^^S9Ahmb Smnl^tmlMsd Tg(SMN2* delta?) 4299Ahmb/J, Stock #005025) were imported from Jackson Laboratory and the Mir34/449 TKO mice were generated as previously reported in the C57BL/6J background. ⁴³ The mutant mouse lines were maintained and bred according to the following intercross mating: Smrr' SMN2 innA7 and Mir 34c Mir34bc Mir 44⁽ All age-matched littermates from such matings served as a control (Ctrl) group for all experiments, unless otherwise specified. The primers used for genotyping are listed in Table 1.

17

SUBSTITUTE SHEET (RULE 26) [00097] Table 1 : List of primers for mouse genotyping

[00098] We employed the smallest sample size that would still detect significant differences, in accordance with 3Rs principles. No animals were involved in prior unrelated experimental procedures. The influence of mouse sex was not evaluated in this study. All of the live animals were maintained in a specific-pathogen-free (SPF) animal facility, approved and overseen by IACUC Academia Sinica.

[00099] 1.4 Tissue collection

[000100] Mice were sacrificed under deep anesthesia using 20 mg/mL Avertin (2,2,2- Tribromoethanol, Sigma), with dosage based on mouse body weight. Then, cardiac perfusion of cold PBS was performed before collecting tissues and placing them in Trizol (Thermo Scientific) for RNA extraction. A subsequent perfusion was performed with freshly prepared 4% paraformaldehyde (PFA) in PBS, followed by whole spinal cord dissection, for immunostaining or in situ hybridization. Spinal cords were sucrose-cryoprotected and embedded in FSC 22 frozen section media (Leica), before being cut into 20-25 pm cryostat sections as previously described. ²⁸ [000101] 1.5 Mouse ESC culture and MN differentiation

[000102] ESCs were cultured and differentiated into spinal MNs as previously described. ⁴⁰ Cells were trypsinized and harvested for fluorescence-activated cell sorting (FACS) at day seven to obtain GFP^on and GFP^off neurons for small RNA-seq when required. All cell lines used in this study are

18

SUBSTITUTE SHEET (RULE 26) subjected to regular mycoplasma tests.

[000103] 1.6 Human ESC/iPSC culture and differentiation

[000104] Peripheral blood was collected after obtaining consent from parents of two type I SMA patients and from healthy volunteers (approved IRB: AS-IRB02-105064), and then reprogrammed as described previously. ⁸¹ ESCs and iPSCs were cultured in Essential 8 medium (Al 517001, Thermo Fisher Scientific) on a Vitronectin-coated dish (A14700, Thermo Fisher Scientific).

[000105] Direct differentiation into spinal motor neurons (MNs) was performed as described previously. ⁴⁸ In brief, ESCs/iPSCs were dissociated into single cells with Accutase (Al 110501, Thermo Fisher Scientific) and resuspended in a differentiation medium to form embryoid bodies (EB) [Advanced DMEM (12634010, Thermo Fisher Scientific): Neurobasal medium (21103-049, Thermo Fisher Scientific) 1: 1, supplemented with 1% N2 supplement (17512048, Thermo Fisher Scientific), 2% B27 supplement (17504044, Thermo Fisher Scientific), 2 mM L-Glutamine (25030081, Thermo Fisher Scientific), 50 U/mL penicillin-streptomycin (15070063, Thermo Fisher Scientific)].

[000106] Early neutralization was first induced by dual SMAD inhibition, with subsequent caudalization and ventralization induced by retinoic acid (R2625-100MG, Sigma-Aldrich) and smoothened agonist (566650-5, Millipore). Gamma-secretase inhibitor (565784-19, Millipore) was added to enhance the transition from neural progenitor to post-mitotic MNs. EBs were dissociated on day 16 of differentiation, with further seeding on poly-L-lysine (P4707, Sigma- Aldrichj/Laminin (23017-015, Thermo Fisher Scientific)-coated plates. Dissociated MNs were then cultured for 2 weeks in differentiation medium containing Y-27632 (688000, Calbiochem), BDNF (450-02, Peprotech), GDNF (450-51, Peprotech), and L-ascorbic acid.

[000107] 1.7 microRNA in situ hybridization and miRNAscope assay

[000108] For all experimental procedures, we used diethylpyrocarbonate (DEPC)-treated water or PBS for washing steps or reagent preparation. Spinal cord cryosections were initially treated with 10 pg/mL proteinase K (Invitrogen), followed by acetylation in acetic anhydride/triethanolamine, and then fixed again with 4% PFA. Next, sections were pre-hybridized in hybridization solution [50% formamide, 5 X SSC, 0.5 mg/rnL yeast tRNA (Ambion), 5 X Denhardt’s solution (Fisher), 0.5 mg/ml salmon sperm DNA (Thermo Fisher Scientific), 0.02% Roche blocking reagent] at room temperature for 2~4 h, followed by hybridization with each miRNA probe overnight at 55 °C. After posthybridization washes in 2 X SSC and then 0.2 X SSC at 55 °C, the in situ hybridization signals were detected using the NBT/BCIP (Roche) system according to the manufacturer’s instructions. After termination of color development, slides were subjected to immunostaining as described below. Slides were mounted in Aqua-Poly /Mount and analyzed with a Zeiss LSM 780 confocal microscope. The 5’

19

SUBSTITUTE SHEET (RULE 26) FITC-labeled LNA miR-17-3p probe (ACUGCAGUGAGGGCACUUGUAG) (SEQ ID NO: 17), miR-27a-3p probe (UUCACAGUGGCUAAGUUCCGC) (SEQ ID NO: 18), and miR-34a-5p probe (UGGCAGUGUCUUAGCUGGUUGU) (SEQ ID NO: 19) were purchased from Exiqon.

[000109] To detect miR-34c-5p, spinal cord samples were dissected and fixed with 4% PFA, as described previously. ^{28, 41} The cryostat sections were processed using miRNAscope™ technolog}' (Advanced Cell Diagnostics, ACD) according to the manufacturer’s instructions. The probe to detect mmu-miR-34c-5p was customized from 896831 -SI (ACD).

[000110] 1.8 RNA isolation from CSF

[000111] Total RNA was isolated from CSF using a mtRNeasy Serum/Plasma Advanced Kit (Cat No.: 217204, Qiagen) by following the manufacturer’s protocol and then eluted in 30 pl of nuclease- free water.

[000112] 1.9 RNA isolation and qRT-PCR

[000113] Total RNA was isolated using a Quick-RNA MiniPrep kit (Zymo Research). For miRNA expression analyses, 100-200 ng of total RNA from each sample was reverse-transcribed using a miRNA-specific primer from TaqMan MicroRNA Assays (Life Technology). The following assays were used: miR-17-5p (Assay ID: 002308); miR-27a-3p (Assay ID: 000408); miR-34a-5p (Assay ID: 000426); miR-218-5p (Assay ID: 000521); miR-34b-5p (Assay ID: 002617); and miR-34c-5p (Assay ID: 000428). Ubiquitous miR-16 (Assay ID: 000391) and a small nucleolar RNA (RNU48, Assay ID: 001006) were used as endogenous internal controls for mouse and human samples, respectively. The cel-miR-39 miRNA (Assay ID: 000200) was used as an exogenous spike-in normalization control for CSF samples. Each quantitative real-time PCR was performed in duplicate per sample, with at least three different experimental samples.

[000114] 1.10 RNA-seq analysis

[000115] For the small RNA-seq analysis, the raw reads were preprocessed with the adapter trimmer EARRINGS, ⁸² and aligned with the ad hoc NTA aligner Tailor (reference: mmlO). ^8j We configured Tailor by setting the minimum sequence alignment length to 18 nucleotides and the maximum hit occurrence to at most 10 positions. After the alignment, we annotated each hit with information retrieved from several genomic databases, as follows: we used GenCode (version 27) ⁸⁴ to annotate gene names and types; miRbase (version 22) ⁸⁵ for pri-miRNA and pre-miRNA classification; and Repeat-Masker (RMSK) ⁸⁶ to mask repeat regions in the genomes. Each read could exhibit multiple annotations. To resolve any conflicts, we defined a hierarchy of annotations, whereby higher-order annotations overwrote lower-ranked annotations. The annotation hierarchy was as follows: miRNA > RMSK > others.

[000116] We collected and analyzed datasets sourced from different labs and derived from different

20

SUBSTITUTE SHEET (RULE 26) experimental conditions and designs, necessitating data normalization to facilitate comparisons. We applied two normalization approaches, i.e., a correction according to reads per million (RPM) and quantile normalization. The RPM correction accounts for the bias in sequencing depth, whereas the quantile normalization balances the scope of data distributions across different samples. Before applying the quantile normalization, we removed reads with an RPM value less than 5, since they were likely to be insignificant or to represent sequencing error. Moreover, in the early stage of analyzing some samples, reads aligned to miR-lOa apparently comprised up to 90% of all reads, strongly biasing the RPM calculation and quantile normalization. To eliminate such artefacts, when comparing samples, we masked miRNAs whose expression level (in RPM) ranked above the top 10% and below the bottom 10% during normalization, thereby preventing outliers from affecting the robustness of our downstream analyses.

[000117] Bulk RNA-seq was performed on an Illumina HiSeq 2500 sequencing system as described previously. ⁴¹

[000118] 1.11 Immunostaining

[000119] Immunostaining was performed as described previously. ^{28, 41} The commercially-available primary antibodies used in this study are: guinea pig polyclonal anti -Isl 1/2 (a gift from Thomas Jessell Laboratory, 1: 1000), guinea pig anti-Hb9 (a gift from Dr. H. Wichterle, 1: 1000), mouse monoclonal anti-NANOG (MABD24, Millipore, 1:500), rabbit polyclonal anti-OCT3/4 (SC-5279, Santa Cruz, 1 : 1000), rabbit monoclonal anti-ChAT (ZRB1012, Millipore, 1 : 100), goat polyclonal anti-ChAT (AB144P, Millipore, 1 :100), mouse monoclonal anti-NeuN (MAB377, Millipore, 1 :500), and rabbit polyclonal anti-Laminin (L9393, Millipore, 1:100). Alexa488-, Cy3- and Cy 5 -conjugated secondary antibodies were obtained from either Invitrogen or Jackson Immunoresearch and used at 1: 1000 dilutions. Images were acquired by using a Zeiss LSM710 or LSM780 confocal microscope.

[000120] Whole-mount staining of the muscle NMJs was performed as described previously. ²⁸ In brief, freshly dissected muscles were fixed in 4% PFA for 2 h and then washed overnight with PBS. Next, the muscles were permeabilized and blocked in 5% bovine serum albumin (BSA) in 1% Triton X-100/PBS at 4 °C overnight, followed by incubation with antibodies against neurofilament NF-M (2H3, Developmental Studies Hybridoma Bank, 1: 100), synaptic vesicle protein 2 (SV2, Developmental Studies Hybridoma Bank, 1 : 100), and a-bungarotoxin (BTX, B 13422, Alexa fluor 555 conjugate, Thermo Fisher Scientific, 1:500) in blocking buffer at 4 °C for 2 days. After several PBS washes, the muscle tissues were incubated overnight at 4 °C with secondary antibodies. Muscles were teased apart and flattened before mounting on slides before being imaged.

[000121] 1.12 Predicted miR-34/449 target genes and gene ontology analysis

[000122] We used the online database TargetScan (Release 7.1 or 7.2,

21

SUBSTITUTE SHEET (RULE 26) htp://www.targetscan.org/mmu 72/) to identify putative MiR34/449-regulated genes. To refine selected candidates, we conducted gene ontology analysis in the Database for Annotation, Visualization, and Integrated Discovery (DAVID).⁸⁷ Functional clustering annotated from DAVID, expressed as gene enrichment, is presented in.⁸⁸

[000123] 1.13 scAAV9-MiR34a preparation and injection

[000124] A 700-bp mouse DNA fragment encompassing Mir 34a was subcloned from pENTR/D- TOPO-Afr/Wo plasmid⁵⁰ into scAAV9-CMV plasmid. This scAAV9-MiR34a construct was first transfected into mouse MiR34a-KO ESCs to verify expression and then it was packaged by the AAV Core Facility in Academia Sinica. For neonatal intravenous administration, scAAV9 solution was freshly prepared in sterile PBS containing 0.3% Fast Green FCF (Sigma- Aldrich). At P0, mouse pups were cold-anesthetized on ice for 1 min, before injecting 10¹⁰ viral genomes of scAAV9 in 30 pL solution into the temporal vein with a 30G needle (BD Ultra-Fine If) under microscopy. Injections were confirmed successful if the pup immediately turned blue. Injected pups were allowed to recover on a heat pad before being placed back in their home cage.

[000125] 1.14 Statistical analysis

[000126] Results are expressed as mean ± SD (standard deviation). Each experiment was repeated at least three times to confirm the findings, unless otherwise specified. Statistical analysis was performed in GraphPad Prism 6.0 (GraphPad Software). Multiple groups were analyzed by one-way or two-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test, as indicated in the figure legends. Two groups were analyzed by Student's t-test. P-values of less than 0.05 were considered significant. Spearman's correlation coefficient was used to assess associations between variables. Baseline CSF miRNA and pNfH levels were used to fit individual linear models predicting HINE-2 score improvements on Day 482 to prevent overfiting, as subject numbers were limited. Individual models were assessed by F-test. Raw p-values from the F-tests were corrected by means of the Benjamini-Hochberg method (with a false discovery rate threshold set at 5%).

[000127] 2. Results

[000128] 2.1 Identification of spinal miRNAs as candidate biomarkers

[000129] Although SMA primarily manifests postnatally, many studies have illustrated that increased apoptosis and gene dysregulation occur during gestation.^{40, 41} Recent studies have further demonstrated that although SMA pathology mainly impacts motor neurons (MNs), effects on multiple interneurons (INs) in the spinal cord may also cause aberrant circuitry that could prompt MN degeneration (Fig. 1 A).⁴²'⁴⁴ Early molecular dysregulation induces postnatal neuronal loss and motor function deterioration, representing a critical window for disease intervention (Fig. 1 A).^{45 46} To reflect

22

SUBSTITUTE SHEET (RULE 26) the molecular pathogenesis of SMA during this early period, we hypothesized that good and reliable biomarkers for predicting or impeding SMA disease progression might fit two criteria: (1) show a consistent decrease throughout SMA-linked MN degeneration, and (2) participate in SMA pathology directly. To identify miRNAs that are expressed from embryonic to adult spinal cords, we began by identifying miRNAs in MNs and INs from an embryonic stem cell (ESC) line expressing an MN transgenic reporter (Hb9::GFP).⁴⁰ Previously, we have shown that Hb9::GFP^0N cells are largely enriched in Islll/Isl2 and Chat, whereas Hb9::GFP^0FF cells (Shh^low condition) are primarily defined by Lhxl/5, Bm3a, and Gadl/2 expression.⁴¹ The Hb9::GFP^0N embryonic MNs and Hb9::GFP^0FF INs were then isolated by means of fluorescence-activated cell sorting (FACS) before undergoing small RNA-sequencing (Fig. IB and Fig. 1C) ⁴¹ (see details in Materials and methods;). Next, we cross- referenced the identified embryonic spinal miRs against the postnatal spinal neuronal miRNA datasets from a public neural miRNA database, ⁴² which revealed 187 candidates enriched in both (Fig. ID). After checking the literature for candidates reported to manifest miRNA functions in the spinal cord, 28, 29, 43-45 _{we se}]_ect_ec[ MiR17, 23, 24, 27, 34, 125 and 181 for further assessments (Fig. ID and Figs. 8A to 8H)iIn this study, we use 'Mir 34' and ‘MiR34’ to indicate the mouse gene and the mature form of miRNA, respectively. Additionally, we selected one MN-enriched expressed miRNA (i.e., MiR218), and two previously identified SMA-related biomarkers (i.e., MiR133 and MiR206) to test in parallel their possible roles in SMA.^{46, 47} Finally, to validate expression of our candidate spinal miRNAs in vivo, we performed in situ hybridization and verified that MiR17a, 27a, and 34a are highly expressed in the ventral horn of mouse spinal cord sections at postnatal day 5 (P5) (Fig. IE).

[000130] 2.2 Early dysregulation of MiR34 among SMA patient iPSC-differentiated MNs

[000131] To examine if our candidate spinal miRNAs reflect human pathology, we modeled type I SMA by reprogramming patient-sourced peripheral blood into induced pluripotent stem cells (iPSCs) (Fig. 9A). The two independent patient-derived iPSC lines exhibited compromised SMN protein levels when compared to the two well-characterized control iPSCs used in our previous study (Fig. 9B). To assess the pattern of dysregulation for candidate miRNAs, we differentiated healthy control iPSCs and type I SMA iPSCs into human spinal MNs (Fig. 2A). ^{28, 48} After two weeks, ~ 80% of the cells expressed ISL1 and SMI32, two generic human MN markers (Fig. 2B). Subsequently, the embyroid body (EB)-derived colonies were dissociated for long-term MN culture at day 11. ²⁸ We defined the “critical window” based on the appearance of significant MN loss. MN survival was assessed by quantifying the ratio of MNs (ISL 1/2^ON and SMI32^ON). ⁴⁹ Consistent with a previous study,⁵⁰ we observed that numbers of both type I SMA MN lines were significantly reduced relative to the two healthy controls at 6 weeks from the start of directed differentiation when normalized against total neurons (Fig. 2C).

23

SUBSTITUTE SHEET (RULE 26) [000132] To determine if our candidate miRNAs are dysregulated during and after that critical window, we examined their expression levels at week 6 (within the critical window) and week 8 (after the critical window) in long-term spinal neuron cultures. Only MiR34a exhibited significant downregulation within the critical window (Fig. 2D), but MiR24, MiR27a, MiR34a, MiR125b, MiR133a/b were all significantly downregulated in SMA-MN cultures after the critical window (MiR 206 was undetectable in the MN culture and was omitted from further analysis) (Fig. 2E). These results indicate that MiR34a is a spinal miRNA exhibiting early dysregulation in our human iPSC model of SMA.

[000133] 2.3 Verification of spinal miRNA dysregulation in SMN A 7 mice during the critical window for intervention

[000134] To identify actionable biomarkers that could be informative for the early critical window and thus guide SMA therapeutic plans, we sought to identify spinal miRNA candidates that displayed dysregulation prior to MNs degenerating in vivo. To do so, we utilized homozygous mutant SMN A 7 mice that exhibit smaller body size and motor deficiency relative to heterozygotic mice (Control) in the first two weeks after birth (Fig. 10A). ⁵¹ Given that we observed a significant decrease in the MN population at P10 based on ChAT⁺ expression (Fig. 3 A and Fig. 3B), we defined P10 as the end of the critical window in SMN A 7 mice. Next, we evaluated expression of the spinal miRNAs — including MiR17, MiR27, MiR34a, MiR125b, MiR181a, and MiR218 — by qPCR analysis at P10. Notably, MiR34a presented robust and consistent downregulation in the whole spinal cord (Fig. 3C), which was further corroborated by in situ hybridization (Fig. 3D). Considering that (1) previous studies have highlighted roles for MiR34 in premotor INs and regulation of programmed cell death, ^{43, 52, 53} and (2) Mir34 displayed the most consistent dysregulated trend among mouse and human SMA models, we focused on MiR34 in this study to examine its clinical value as a biomarker and its involvement in SMA pathogenesis.

[000135] 2.4 The MiR34 family regulates motor-end plate function

[000136] We reasoned that a reliable disease biomarker is typically more easily interpretable and specific if it plays a direct role in disease pathogenesis, so we investigated if the Mir34 family (MiR34a/bc) plays a role in regulating MN function. However, the MiR449 cluster is functionally redundant to the Mir34 family and bears the same seed sequence, which makes it challenging to assess their respective roles by means of genetic manipulations. To circumvent this problem, we established a Mir 34/449 triple knockout (TKO) mouse model in which all six alleles are completely deleted to prevent compensatory effects (Mir34a^J~ Mir34bc ⁱ' Mir 449" or Mir 34/449 TKO), thereby providing us with a unique opportunity to examine the mechanisms in MNs that are regulated by the MiR34/449

24

SUBSTITUTE SHEET (RULE 26) families. ⁴³

[000137] Mir 34/449 TKO mice exhibited two waves of postnatal lethality, i.e., in the first week and fourth week after birth. Relative to littermate control (Ctrl) groups at P7, the Mir 34/449 TKO mice displayed retarded growth (Fig. 10B). Our previous study indicated that until P20, numbers of MNs are comparable between Ctrl w Mir 34/449 TKO mice. ⁴³ Though loss of MNs is a hallmark of SMA, several lines of evidence support that the functional deficits could be as important as cell death. For instance, life expectancy is prolonged when SMA mouse models are treated with Rho-associated kinase to disrupt P53 function, yet MN loss is not rescued. ^{54, 55} In contrast, whereas Bax knockout rescues MN number and survival, growth retardation remains and survival is only partially rescued.⁵⁶ These observations prompted us to speculate that SMN deficiency could perturb multiple parallel pathways, highlighting the importance of investigating functional aspects other than MN loss alone. Thus, we further tested if the morphology of neuromuscular junctions (NMJ) is disrupted in our Mir 34/449 TKO mice.

[000138] First, we assessed the presynaptic components (MNs) in the NMJ motor end-plate via a- Bungarotoxin (a-BTX) labeling combined with immunostainings at P7 for neurofilaments (NF) and synaptic vesicle protein 2 (SV2) in intercostal muscles — one of the muscle types most affected by SMA both in patients and in the SMNA7 mouse model ⁵⁷ ’ ⁵⁸ — from wild type ( [V),Mir34a/449 double knockout (Mir 34a": Mir449^J' or Mir34a/449 DKO), Mir 34/449 TKO (Mir34a^J/ Mir34bc '/ Mir 449' '). and SMNA7 mice. Interestingly, both Mir 34/449 TKO and Mir34a/449 DKO littermate mice displayed significant pre-synaptic NF aggregation (Fig. 4A, quantification in Fig. 4B). Next, we assessed the post-synaptic components by examining the size of acetylcholine receptor (AChR) clusters in the intercostal muscles. Similar to SMNA7 mice, NMJ end-plates were smaller inMir 34/449 TKO mice compared to both DKO and WT groups (Fig. 4C, quantification in Fig. 4D). We also examined myofiber growth retardation, another phenotype reported for SMNA7 mice, ⁵⁷ and detected a reduction in myofiber diameter in both the intercostal and gastrocnemius muscles (GA) of Mir 34/449 TKO mice at P7 (Fig. 4E and Fig. 4G, quantification in Fig. 4F and Fig. 4H, respectively).

[000139] 2.5 The MiR34 family regulates synapse formation pathways

[000140] To identify genes regulated by MiR34/449 in the spinal cord, we conducted RNA-seq on thoracic and lumbar segments of spinal cord tissues at P14 from Mir 34/449 TKO and age-matched WT mice serving as the Control group (Fig. 5A), as these two segments are more affected in SMA. In the thoracic segments, we identified 4021 upregulated and 2532 downregulated genes upon MiR34/449 depletion (Fig. 5B and Fig. 11 A). In contrast, 2265 upregulated and 1140 downregulated genes were uncovered in the lumbar segments (Fig. 5C and Fig. 1 IB). Next, we aimed to uncover

25

SUBSTITUTE SHEET (RULE 26) target genes directly associated with MiR34/449, so we cross-referenced the upregulated genes against predicted MiR34/449 targets in the thoracic (Fig. 5B) and lumbar (Fig. 5C) regions. In both segments, gene ontology (GO) analysis revealed several convergent terms, such as synapse-related and axon guidance terms (Fig. 5B and Fig. 5C), consistent with the major dysregulated GO terms i den ti Pied in a series of studies on SMAmice. ^59-61 Strikingly, we also noticed neuromuscular junction related genes were aberrantly upregulated in the lumbar regions, which is in concordance with the above described phenotype for Mir 34/449 TKO mice (Figs. 4Ato 4H and Fig. 5C). Thus, the MiR34 family may target synapse formation and several core pathways that are disrupted in SMA at early postnatal stages.

[000141] 2.6 Neonatal delivery of MiR34a partially rescues the disease phenotype of SMN A 7 mice

[000142] As MiR34 was consistently reduced in our SMA models and could regulate NMJs, we sought to establish if reintroducing MiR34 could ameliorate SMA-linked phenotypes. To address this question, we administered a single dose of self-complementary adeno-associated vector serotype 9 (scAAV9) expressing MiR34a intravenously into SMNA7 mice right after they were bom (postnatal dayO, P0), before subjecting them to histological and behavioral assessments at P7 (Fig. 6A). At P7, we verified by qPCR that levels of MiR34a were elevated in the spinal cords of the MiR34a-treated SMA mice (Fig. 6B). Although the control vehicle deliver}' group still displayed SMA phenotypes, including aggregated NF in the pre-synaptic terminals of motor axons and smaller a-BTX-labeled endplate area in the intercostal muscles (Fig. 6C, Fig. 6D, and Fig. 6E), the AAV-MiR34a treatment cohort displayed significantly enlarged NMJ end-plates (Fig. 6C and Fig. 6E). Importantly, the righting reflex test also revealed that MiR34a treatment partially rescued the impaired motor function of SMA mice (Fig. 6F and Fig. 6G). Notably, we uncovered a strong negative correlation between the righting reflex index and end-plate area of SMA mice (n=l 3, Fig. 6H), as well as between their righting reflex latency and spinal MiR34a levels (n=20, Fig. 13), indicating that restoration of motor function by MiR34a treatment was achieved by maintaining NMJ integrity. Together, these findings substantiate evidence for a mechanistic role of MiR34 in SMA pathology, warranting further investigation regarding its predictive power in clinical settings.

[000143] 2.7 MiR34 in the cerebrospinal fluid predicts the responsiveness of type I SMA patients to nusinersen treatment

[000144] Given existing programs for carrier and prenatal screening of SMN deficiency, a useful SMA biomarker would reliably predict disease progression and patient responsiveness to treatments. Therefore, we evaluated longitudinal changes in MiR34 family expression in CSF samples obtained from type I SMA patients undergoing ASO (nusinersen) treatment. We enrolled a total of seven type I

26

SUBSTITUTE SHEET (RULE 26) SMA patients receiving nusinersen therapy at Kaohsiung Medical University in this study. SMN2 copy number was determined for each patient, and longitudinal assessments of clinical features and motor function were also recorded (Table 2).

[000145] Table 2: Clinical features and motor function evaluations of SMA type 1 patients before and after nusinersen treatment.

[000146] The mean age of SMA disease onset was 2.9 months, and the mean age at administration of the first nusinersen dose was 8.6 months. Loading doses of nusinersen were given to patients on days 0, 14, 28, and 64, followed by scheduled maintenance doses every 4 months thereafter (Fig. 7 A). We profiled MiR34a, MiR34b, and MiR34c expression levels in the CSF at baseline (before the first nusinersen dosage) and on the day of the last loading dose (Day 64) (Figs. 7C-7E). Levels of axonal phosphorylated neurofilament heavy chain (pNfH) proteins in blood and CSF have been proposed to reflect active axonal loss and may act as a sensitive tool for detecting early MN degeneration. ⁶² Moreover, the rapid decline in pNfH levels observed in SMA patients upon SMN-restorative treatment has been suggested as reflecting early rescue of axonal pathology. ^63-65 Therefore, we used pNfH as a known biomarker to examine the predictive power of our baseline measurements in terms of motor function. Similar to pNfH (Fig. 7B), we found that MiR34a, MiR34b, and MiR34c all exhibited a decline in patient expression levels upon nusinersen treatment (except for patient SMA-2; Figs. 7C-7E). Since comparing baseline (day 0) and day 64 measurements is confounded by the varied individual treatment responses among patients, we assessed the degree of correlation between changes in pNfH and M1R34 family expression levels in the CSF and improved motor function. To do so, we performed Spearman’s correlation analysis on the motor-function scale of the Hammersmith Infant Neurological Examination (HINE-2) against pNfH and MiR34 expression in our type I SMA patients at Day 64, Day 183 and Day 482 of nusinersen treatment (Figs. 7F-7H, Fig. 12). Although we detected no significant correlations for the loading phase of nusinersen therapy, we did find that relative changes

27

SUBSTITUTE SHEET (RULE 26) in both pNfH and the MiR34 family, and particularly for MiR34b levels, were negatively correlated with improved HINE-2 scores at the later treatment phases of Day 183 and Day 482 (Figs. 7F-7H).

[000147] Finally, that correlation between treatment responses and CSF levels of pNfH and the MiR34 family prompted us to examine if their baseline measurements are predictive. Therefore, we fitted individual linear regression models against HINE-2 changes on Day 482 as the dependent variable and pNfH/MiR34 expression at baseline as the independent variables, and performed an F- test to assess the predictive power of each model. After multiple comparison corrections, we found that baseline MiR34b in CSF was significantly and positively correlated with HINE-2 score on Day 482 (Fig. 71), whereas pNfH failed to reach significance. This outcome is concordant with a recent report indicating that plasma pNfH is only predictive of motor function after treatment for more than 3 years. ⁶⁶ Taken together, our findings indicate that the MiR34 family is an early and constant biomarker that is downregulated in human and rodent models of SMA, and it might account for synaptogenesis pathway regulation that maintains motor end-plate integrity. Loss of MiR34 function leads to the impairment of NMJs and compromises muscle fiber size, whereas delivery of MiR34a can ameliorate SMA symptoms, which might explain why disruption of SMN function leads to SMA pathology. Importantly, MiR34 family members are present in the CSF and could serve as an early predictor of patient responses to nusinersen treatment.

[000148] 3. DISSCUSSION

[000149] Although miRNAs have been well documented over the past two decades as exerting versatile roles in regulating neural development, their roles in neurodegenerative diseases are just beginning to emerge. Significantly, deletion of the miRNA biogenesis enzyme Dicer from MNs either using Olig2-Cre or ChAT-Cre leads to MN degeneration, ²⁶‘ ^{28> 67} mimicking certain hallmarks of SMA and ALS and eliciting the hypothesis that miRNAs might be critical regulators of MN diseases. Mechanistically, SMN engages with miRNA-RBP (RNA-binding protein) complexes and likely regulates miRNA biogenesis and metabolism. SMN deficiency may alter miRNAs or miRNPs (ribonucleoproteins harboring miRNAs), ⁶⁸ which might result in MN death. Previously, we proposed mechanisms by which SMN-associated RBPs might be involved in miRNA biogenesis, including by facilitating recruitment of Drosha to specific miRNAs, by binding to components of the Drosha and Dicer complexes, or by acting as regulators of the RISC complex. ²⁹ Consequently, disruption of the SMN-RBP complex due to SMN deficiency in the SMA pathological background might cause dysregulated processing of miRNAs and pre-mRNA splicing. We reasoned that these miRNAs are more likely to be affected by SMA, a disease known to manifest prenatally. Among spinal miRNAs, we found that the MiR34 family is downregulated in early stages of SMA in both rodent and human MNs. Using Mir 34/449 TKO mice, we circumvented the functional redundancy among MiR34 family

28

SUBSTITUTE SHEET (RULE 26) members and reveal that the M1R34 family is required to maintain motor end-plate integrity. We hypothesized that MiR34 downregulation is a result of SMN deficiency. Consistently, we observed several major dysregulated pathways vs\MiR34 TKO and SMA model mice, as well as in human iPSC models. Finally, our longitudinal study of CSF profile and motor development of type I SMA patients undergoing nusinersen treatment reveals that baseline levels of MiR-34b in CSF is a significant predictor of treatment outcome after one and half years of therapy and that changes in CSF-circulating MiR34 are correlated with improved motor function. Thus, we propose that MiR-34b and other MiR- 34 family members represent promising biomarkers for predicting patient responsiveness to SMA treatment and can reflect treatment progress in real-time because they maintain the NMJ motor endplates that are negatively affected early in SMA pathogenesis. ⁶⁹

[000150] What makes the MiR34 family reflect SMA pathology? Previously, we have shown that MiR34 maintains postnatal IN numbers to modulate sensory perception and motor output. ⁴³ Several studies have demonstrated that disruption of the sensory-motor circuitry contributes to MN degeneration early in SMA onset. ⁷⁰ In particular, altered sensory-motor systems preceding MN death have been reported for various SMA mouse models, all of which are characterized by hyperexcitability, increased input resistance, and decreased synaptic efficacy. ^71-74 Since both disrupted circuitry and motor end-plate dysfunction manifest early in SMA, the early changes in MiR34 family expression levels might signify a complex interplay among the multiple components of the sensory-motor circuitry that could ultimately contribute to MN dysfunction and death during SMA pathogenesis.

[000151] In this study, we used entire thoracic and lumbar segments to perform RNAseq on the MIR34/449 TKO mice, largely due to the difficulty of acquiring the TKO mice. Since it has been reported previously that some specific segments (i.e., T9 and LI) manifest more prominent MN degeneration than other segments and that a differential medial-to-lateral and cranial-caudal MN vulnerability exists in SMA,^{72, 75} our approach might have overlooked some dysregulated MiR34/449- mediated genes involved in SMA. Identifying the full spectrum of dysregulated genes mediated by MiR34 and establishing their functions and possible links to SMA according to specific vulnerable spinal segments via a single-cell RNAseq approach would be illuminating. MiR34 TKO alters the expression of genes involved in axon guidance and synaptogenesis, so we cannot rule out the possibility that the MiR34 family could contribute to SMA pathology via sensory-motor circuit dysfunction and indirectly cause motor end-plate dysfunction. Nevertheless, our neuropathological findings fromAfir 34/449 TKO mice indicate that synaptic defects might manifest at multiple levels of the spinal sensory-motor and peripheral neuromuscular circuitry, which was has also been observed early in an SMA mouse model. ⁷¹ Thus, the multifaceted and versatile roles of the MiR34 family in spinal neurons might enhance its representation in the CSF, rendering detection less noisy.

29

SUBSTITUTE SHEET (RULE 26) [000152] Why is it that expression levels of the MiR34 family are downregulated after efficient nusinersen treatment in patients, even though they play positive roles in maintaining NMJ formation? Although it remains largely unclear how SMA-associated miRNAs end up in body fluid, several models may explain how miRNAs in CSF can act as biomarkers. First, if the primary source of the miRNAs is leakage from necrotic MNs, then levels of MN-miRNAs in the CSF will gradually increase in SMA patients as the MNs progressively degenerate with SMA onset. Our results reveal a trend of decreasing MiR34 family levels that are highly correlated with motor improvements, which supports this model. Moreover, a previous study reported a positive role for miRNAs in MN function but reduced levels in the CSF.⁷⁶ Alternatively, miRNAs are known to undergo active and selective exocytosis. In that scenario, miRNAs undergo selective packaging prior to exiting cells, ³⁶ so SMA could prompt MiR34 family members to be actively packaged and removed from MNs. Both models are consistent with the intracellular dow nregulation of the MiR34 family in untreated MNs and their decreased extracellular levels when MNs are rescued with appropriate treatments. Further investigation is required to understand if SMN protein interacts with the molecular machinery that packs and secretes miRNAs, thereby enabling targeted discovery and interpretation of other miRNA biomarkers.

[000153] Over the past decade, rapid advancements in developing SMA treatments represent one of the most exciting research directions in the field of rare monogenetic disorders and neurodegenerative diseases. Several therapeutic approaches for SMA are feasible, such as by increasing SMN production, which can be achieved by modifying SMN2 mR A splicing or by replacing the defective SMN1 gene. Other approaches, such as increasing SMN transcript levels, stabilizing SMN protein, and neuroprotective or cell therapies are currently undergoing development. ^{77, 78} Here, we have shown that MiR34b in the CSF exhibits equivalent or a better ability to predict treatment responses to an ASO-based treatment when compared with pNfH. Our preliminary multivariate linear model that simultaneously considers pNfH and the MiR34 family presented improved prediction accuracy compared to individual pNfH/MiR34b models. Although we cannot rule out a contribution of multivariate model over-fitting given our current sample size, improved model accuracy by enhancing sample size could indicate that pNfH and MiR34b are not redundant and that multiple weak predictors could form a stronger ensemble model. A strong predictive model could help determine if dose adjustment or a combinatorial approach is needed for SMApatients that are predicted to respond poorly to treatment. In this study, we used mouse ESC-derived MNs and INs as paradigms to search for potential miRNAs that are involved in SMA disease progression and that might serve as novel mechanistic biomarkers for predicting disease treatment outcomes. It remains to be tested if other miRNAs that are expressed in the adult spinal cord or some specific miRNAs that only exist in

30

SUBSTITUTE SHEET (RULE 26) humans may also act as good candidates. It will be tantalizing to explore if the MiR34 family, together with other identified miRNA biomarkers such as MiR-133/181/206,

as well as other yet-to-be- discovered candidates, could be pooled as an “SMA disease treatment prediction biomarker panel” for clinical application in the real world.

[000154] In conclusion, the MiR34 family is a mechanistic biomarker of SMA and exerts multifaceted functions in regulating both MNs and sensory-motor circuits in the spinal cord. MiR34b levels at baseline and during nusinersen therapy are predictive and are correlated with motor function after treatment. Therefore, we propose that the MiR34 family represents a set of promising biomarkers to assess responses to SMN-restorative therapies.

Sequence Information Mouse

MiR34a uggcagugucuuagcugguugu (SEQ ID NO: 20)

MiR34b aggcaguguaauuagcugauugu (SEQ ID NO: 21)

MiR34c aggcaguguaguuagcugauugc (SEQ ID NO: 22)

> Human

MiR34a uggcagugucuuagcugguugu (SEQ ID NO: 20)

MiR34b uaggcagugucauuagcugauug (SEQ ID NO: 23)

MiR34c aggcaguguaguuagcugauugc (SEQ ID NO: 22)

References

1. Schmid, A, and DiDonato, CJ (2007). Animal models of spinal muscular atrophy. J Child Neurol 22: 1004-1012.

2. Rochette, C, Gilbert, N, and Simard, L (2001). SMN gene duplication and the emergence of the SMN2 gene occurred in distinct hominids: SMN2 is unique to Homo sapiens. Human Genetics 108: 255-266.

3. Lunn, MR, and Wang, CH (2008). Spinal muscular atrophy. Lancet 371: 2120-2133.

4. Monani, UR (2005). Spinal muscular atrophy: a deficiency in a ubiquitous protein; a motor neuron-specific disease. Neuron 48: 885-896.

5. Burghes, AH, and Beattie, CE (2009). Spinal muscular atrophy: why do low levels of survival motor neuron protein make motor neurons sick? Nat Rev Neurosci 10: 597-609.

6. Chen, T, and Wang, C (2020). Emerging Innovative Therapies of Spinal Muscular Atrophy:

31

SUBSTITUTE SHEET (RULE 26) Current Knowledge and Perspectives. In: Atta-ur-Rahman and Amtul, Z (eds). Frontiers in Clinical Drug Research-CNS and Neurological Disorders , vol. 8. Bentham Science Publishers: Sharjah, United Arab Emirates, pp 1-34.

7. Finkel, RS, Mercuri, E, Darras, BT, Connolly, AM, Kuntz, NL, Kirschner, J, et al. (2017). Nusinersen versus Sham Control in Infantile-Onset Spinal Muscular Atrophy. The New England journal of medicine 377: 1723-1732.

8. Mercuri, E, Darras, BT, Chiriboga, CA, Day, JW, Campbell, C, Connolly, AM, et al. (2018). Nusinersen versus Sham Control in Later-Onset Spinal Muscular Atrophy. The New England journal of medicine 378: 625-635.

9. Hua, Y, Sahashi, K, Hung, G, Rigo, F, Passini, MA, Bennett, CF, et al. (2010). Antisense correction of SMN2 splicing in the CNS rescues necrosis in a type III SMA mouse model. Genes Dev 24: 1634-1644.

10. Hua, Y, Vickers, TA, Baker, BF, Bennett, CF, and Krainer, AR (2007). Enhancement of SMN2 exon 7 inclusion by antisense oligonucleotides targeting the exon. PLoSBiol 5: e73-e73.

11. Foust, KD, Wang, X, McGovern, VL, Braun, L, Bevan, AK, Haidet, AM, et al. (2010), Rescue of the spinal muscular atrophy phenotype in a mouse model by early postnatal delivery of SMN. Nature Biotechnology 28: 271-274.

12. Mendell, JR, Al-Zaidy, S, Shell, R, Arnold, WD, Rodino-Klapac, LR, Prior, TW, et al. (2017). Single-Dose Gene-Replacement Therapy for Spinal Muscular Atrophy. New England Journal of Medicine 31'1: 1713-1722.

13. Ratni, H, Ebeling, M, Baird, J, Bendels, S, Bylund, J, Chen, KS, et al. (2018). Discovery of Risdiplam, a Selective Survival of Motor Neuron-2 (SMN2) Gene Splicing Modifier for the Treatment of Spinal Muscular Atrophy (SMA). Journal of Medicinal Chemistry 61: 6501-6517.

14. Baranello, G, Darras, BT, Day, JW, Deconinck, N, Klein, A, Masson, R et al. (2021). Risdiplam in Type 1 Spinal Muscular Atrophy. New England Journal of Medicine 384: 915- 923.

15. Farooq, F, Molina, FA, Hadwen, J, MacKenzie, D, Witherspoon, L, Osmond, M, et al. (2011). Prolactin increases SMN expression and survival in a mouse model of severe spinal muscular atrophy via the STAT5 pathway. The Journal of clinical investigation 121: 3042-3050.

16. Farooq, F, Balabanian, S, Liu, X, Holcik, M, and MacKenzie, A (2009). p38 Mitogen-activated protein kinase stabilizes SMN mRNA through RNA binding protein HuR. Human molecular genetics 18: 4035-4045.

17. Hofmann, Y, Lorson, CL, Stamm, S, Androphy, EJ, and Wirth, B (2000). Htra2-pl stimulates an exonic splicing enhancer and can restore full-length SMN expression to survival motor neuron 2 (SMN2). Proceedings of the National Academy of Sciences 97: 9618-9623.

18. d'Ydewalle, C, Ramos, DM, Pyles, NJ, Ng, SY, Gorz, M, Pilato, CM, et al. (2017). The Antisense Transcript SMN-AS1 Regulates SMN Expression and Is a Novel Therapeutic Target for Spinal Muscular Atrophy. Neuron 93: 66-79.

19. Wirth, B, Karakaya, M, Kye, MJ, and Mendoza-Ferreira, N (2020). Twenty -Five Years of Spinal Muscular Atrophy Research: From Phenotype to Genotype to Therapy, and What Comes Next. Annu Rev Genomics Hum Genet 21: 231-261.

20. Ramdas, S, and Servais, L (2020). New treatments in spinal muscular atrophy: an overview of currently available data. Expert opinion on pharmacotherapy 21: 307-315,

21. Baranello, G, Gomi, K, Daigl, M, Kotzeva, A, Evans, R, Hawkins, N, etal. (2021). Prognostic Factors and Treatment-Effect Modifiers in Spinal Muscular Atrophy. Clin Pharmacol Ther.

22. Kariyawasam, DST, D'Silva, A, Lin, C, Ryan, MM, and Farrar, MA (2019). Biomarkers and the Development of a Personalized Medicine Approach in Spinal Muscular Atrophy. Frontiers in neurology 10: 898.

23. Smeriglio, P, Langard, P, Querin, G, and Biferi, MG (2020). The Identification of Novel Biomarkers Is Required to Improve Adult SMA Patient Stratification, Diagnosis and Treatment.

32

SUBSTITUTE SHEET (RULE 26) J Pers Med 10.

24. Ludwig, N, Leidinger, P, Becker, K, Backes, C, Fehlmann, T, Pallasch, C, et al. (2016). Distribution of miRNA expression across human tissues. Nucleic Acids Research 44: 3865- 3877.

25. Pelaez, N, and Carthew, RW (2012). Biological robustness and the role of microRNAs: a network perspective. Curr Top Dev Biol 99: 237-255.

26. Haramati, S, Chapnik, E, Sztainberg, Y, Eilam, R, Zwang, R, Gershoni, N. etal. (2010). miRNA malfunction causes spinal motor neuron disease. Proc Natl Acad Sci USA 107: 13111-13116.

27. Reichenstein, I, Eitan, C, Diaz-Garcia, S, Haim, G, Magen, I, Siany, A, U al. (2019). Human genetics and neuropathology suggest a link between miR-218 and amyotrophic lateral sclerosis pathophysiology'. Sci Tr ansi Med 11.

28. Tung, YT, Peng, KC, Chen, YC, Yen, YP, Chang, M, Thams, S, et al. (2019). Mir-17 approximately 92 Confers Motor Neuron Subtype Differential Resistance to ALS-Associated Degeneration. Cell Stem Cell 25: 193-209 el97.

29. Chen, TH, and Chen, JA (2019). Multifaceted roles of microRNAs: From motor neuron generation in embryos to degeneration in spinal muscular atrophy. eLife 8.

30. Catapano, F, Zaharieva, I, Scoto, M, Marrosu, E, Morgan, J, Muntoni, F, et al. (2016). Altered Levels of MicroRNA-9, -206, and -132 in Spinal Muscular Atrophy and Their Response to Antisense Oligonucleotide Therapy. Molecular therapy Nucleic acids 5: e331.

31. Abiusi, E, Infante, P, Cagnoli, C, Lospinoso Severini, L, Pane, M, Coratti, G, t al- (2021). SMA-miRs (miR-181a-5p, -324-5p, and -451a) are overexpressed in spinal muscular atrophy skeletal muscle and serum samples. eLife 10: e68054.

32. Wertz, MH, Winden, K, Neveu, P, Ng, S-Y, Ercan, E, and Sahin, M (2016). Cell-type-specific miR-431 dysregulation in a motor neuron model of spinal muscular atrophy. Human molecular genetics 25: 2168-2181.

33. Verma, P, Augustine, GJ, Ammar, MR, Tashiro, A, and Cohen, SM (2015). A neuroprotective role for microRNA miR-1000 mediated by limiting glutamate excitotoxicity. NatNeurosci 18: 379-385.

34. McKenzie, AJ, Hoshino, D, Hong, NH, Cha, DJ, Franklin, JL, Coffey, RJ, etal. (2016). KRAS- MEK signaling controls Ago2 sorting into exosomes. Cell reports 15: 978-987.

35. Mittelbrunn, M, Gutierrez- Vazquez, C, Villarroya-Beltri, C, Gonzalez, S, Sanchez-Cabo, F, Gonzalez, MA, et al. (2011). Unidirectional transfer of microRNA-loaded exosomes from T cells to antigen-presenting cells. Nature communications 2: 1-10.

36. Shurtleff, MJ, Temoche-Diaz, MM, Karfilis, KV, Ri, S, and Schekman, R (2016). Y-box protein

1 is required to sort microRNAs into exosomes in cells and in a cell-free reaction. eLife 5: el9276.

37. Tsai, Y-W, Sung, H-H, Li, J-C, Yeh, C-Y, Chen, P-Y, Cheng, Y-J, et al. (2019). Glia-derived exosomal miR-274 targets Sprouty in trachea and synaptic boutons to modulate growth and responses to hypoxia. Proceedings of the National Academy of Sciences 116: 24651-24661.

38. Navarrete-Opazo, A, Garrison, S, and Waite, M (2020). Molecular biomarkers for spinal muscular atrophy: A systematic review. Neurology: Clinical Practice: 10. 1212/CP J.0000000000000872.

39. Saffari, A, Carmet, C, Blaschek, A, Hahn, A, Hoffmann, GF, Johannsen, J, et al. (2021). 1H- NMR-based metabolic profiling identifies non-invasive diagnostic and predictive urinary fingerprints in 5q spinal muscular atrophy. Orphanet Journal of Rare Diseases 16: 441.

40. Wichterle, H, Lieberam, I, Porter, JA, and Jessell, TM (2002). Directed differentiation of embryonic stem cells into motor neurons. Cell 110: 385-397.

41. Yen, YP, Hsieh, WF, Tsai, YY, Lu, YL, Liau, ES, Hsu, HC, et al. (2018). Dlkl-Dio3 locus- derived IncRNAs perpetuate postmitotic motor neuron cell fate and subtype identity. eLife 1.

42. Pomper, N, Liu, Y, Hoye, ML, Dougherty, JD, and Miller, TM (2020). CNS mi croRNA profiles:

33

SUBSTITUTE SHEET (RULE 26) a database for cell type enriched microRNA expression across the mouse central nervous system. Sci Rep 10: 4921.

43. Chang, SH, Su, YC, Chang, M, and Chen, JA (2021). MicroRNAs mediate precise control of spinal interneuron populations to exert delicate sensory -to-motor outputs. Elife 10.

44. Li, CJ, Liau, ES, Lee, YH, Huang, YZ, Liu, Z, Willems, A, et al. (2021). MicroRNA governs bistable cell differentiation and lineage segregation via a noncanonical feedback. Mol Syst Biol 17: e9945.

45. Li, CJ, Hong, T, Tung, YT, Yen, YP, Hsu, HC, Lu, YL, et al. (2017). MicroRNA filters Hox temporal transcription noise to confer boundary formation in the spinal cord. Nature communications 8: 14685.

46. Amin, ND, Senturk, G, Costaguta, G, Driscoll, S, O'Leary, B, Bonanomi, D, et al. (2021). A hidden threshold in motor neuron gene networks revealed by modulation of miR-218 dose. Neuron 109: 3252-3267 e3256.

47. Thiebes, KP, Nam, H, Cambronne, XA, Shen, R, Glasgow, SM, Cho, HH, et al. (2015). miR- 218 is essential to establish motor neuron fate as a downstream effector of Isll-Lhx3. Nature communications 6: 7718.

48. Maury, Y, Come, J, Piskorowski, RA, Salah-Mohellibi, N, Cheval eyre, V, Peschanski, M, et al. (2015). Combinatorial analysis of developmental cues efficiently converts human pluripotent stem cells into multiple neuronal subtypes. Nat Biotechnol 33: 89-96.

49. Mullen, RJ, Buck, CR, and Smith, AM (1992). NeuN, a neuronal specific nuclear protein in vertebrates. Development 116: 201-211.

50. Fuller, HR, Mandefro, B, Shirran, SL, Gross, AR, Kaus, AS, Botting, CH, etal. (2015). Spinal Muscular Atrophy Patient iPSC-Derived Motor Neurons Have Reduced Expression of Proteins Important in Neuronal Development. Front Cell Neurosci 9: 506.

51. Le, TT, Pham, LT, Butchbach, ME, Zhang, HL, Monani, UR, Coovert, DD, et al. (2005). SMNDelta7, the major product of the centromeric survival motor neuron (SMN2) gene, extends survival in mice with spinal muscular atrophy and associates with full-length SMN. Hum Mol Genet 14: 845-857.

52. Hermeking, H (2010). The miR-34 family in cancer and apoptosis. Cell Death & Differentiation 17: 193-199.

53. Simon, CM, Dai, Y, Van Alstyne, M, Koutsioumpa, C, Pagiazitis, JG, Chalif, JI, et al. (2017). Converging Mechanisms of p53 Activation Drive Motor Neuron Degeneration in Spinal Muscular Atrophy. Cell reports 21: 3767-3780.

54. Bowerman, M, Murray, LM, Boyer, JG, Anderson, CL, and Kothary, R (2012). Fasudil improves survival and promotes skeletal muscle development in a mouse model of spinal muscular atrophy. BMC Medicine 10: 24.

55. Courtney, NL, Mole, AJ, Thomson, AK, and Murray, LM (2019). Reduced P53 levels ameliorate neuromuscular junction loss without affecting motor neuron pathology in a mouse model of spinal muscular atrophy. Cell Death Dis 10: 515.

56. Tsai, MS, Chiu, YT, Wang, SH, Hsieh-Li, HM, Lian, WC, and Li, H (2006). Abolishing Bax- Dependent Apoptosis Shows Beneficial Effects on Spinal Muscular Atrophy Model Mice. Molecular Therapy 13: 1149-1155.

57. Kong, L, Wang, X, Choe, DW, Polley, M, Burnett, BG, Bosch-Marce, M, et al. (2009). Impaired synaptic vesicle release and immaturity of neuromuscular junctions in spinal muscular atrophy mice. The Journal of neuroscience : the official journal of the Society for Neuroscience 29: 842-851.

58. Kariya, S, Park, GH, Maeno-Hikichi, Y, Leykekhman, O, Lutz, C, Arkovitz, MS, et al. (2008). Reduced SMN protein impairs maturation of the neuromuscular junctions in mouse models of spinal muscular atrophy. Hum Mol Genet 17: 2552-2569.

59. Zhang, Z, Pinto, AM, Wan, L, Wang, W, Berg, MG, Oliva, I, et al. (2013). Dysregulation of

34

SUBSTITUTE SHEET (RULE 26) synaptogenesis genes antecedes motor neuron pathology in spinal muscular atrophy. Proceedings of the National Academy of Sciences of the United States of America 110: 19348- 19353.

60. Jangi, M, Fleet, C, Cullen, P, Gupta, SV, Mekhoubad, S, Chiao, E, et al. (2017). SMN deficiency in severe models of spinal muscular atrophy causes widespread intron retention and DNA damage. Proc Natl Acad Sci USA 114: E2347-E2356.

61. Baumer, D, Lee, S, Nicholson, G, Davies, JL, Parkinson, NJ, Murray, LM, et al. (2009). Alternative splicing events are a late feature of pathology in a mouse model of spinal muscular atrophy. PLoS Genet 5: el000773.

62. Zucchi, E, Bonetto, V, Soraru, G, Martinelli, I, Parchi, P, Liguori, R, et al. (2020). Neurofilaments in motor neuron disorders: towards promising diagnostic and prognostic biomarkers. Mol Neurodegener 15: 58.

63. De Vivo, DC, Bertini, E, Swoboda, KJ, Hwu, WL, Crawford, TO, Finkel, RS, et al. (2019). Nusinersen initiated in infants during the presymptomatic stage of spinal muscular atrophy: Interim efficacy and safety results from the Phase 2 NURTURE study. Neuromuscular disorders : NMD 29: 842-856.

64. Darras, BT, Crawford, TO, Finkel, RS, Mercuri, E, De Vivo, DC, Oskoui, M, et al. (2019). Neurofilament as a potential biomarker for spinal muscular atrophy. Annals of clinical and translational neurology 6: 932-944.

65. Winter, B, Guenther, R, Ludolph, AC, Hermann, A, Otto, M, and Wurster, CD (2019). Neurofilaments and tau in CSF in an infant with SMAtype 1 treated with nusinersen. J Neurol Neurosurg Psychiatry 90: 1068-1069.

66. Farrar, MA, Muntoni, F, Sumner, CJ, Crawford, TO, Finkel, RS, Mercuri, E, et al. (2021). Plasma Phosphorylated Neurofilament Heavy Chain (pNF-H) Level is Associated with Future Motor Function in Nusinersen-treated Individuals with Later-onset Spinal Muscular Atrophy (SMA) (2248). Neurology 96: 2248.

67. Crawford, TO, Paushkin, SV, Kobayashi, DT, Forrest, SJ, Joyce, CL, Finkel, RS, et al. (2012). Evaluation of SMN protein, transcript, and copy number in the biomarkers for spinal muscular atrophy (BforSMA) clinical study. PloS one 7: e33572.

68. Nelson, PT, Hatzigeorgiou, AG, and Mourelatos, Z (2004). miRNP:mRNA association in polyribosomes in a human neuronal cell line. RNA 10: 387-394.

69. Kong, L, Valdivia, DO, Simon, CM, Hassinan, CW, Delestree, N, Ramos, DM, et al. (2021). Impaired prenatal motor axon development necessitates early therapeutic intervention in severe SMA. Sci Transl Med 13.

70. Shorrock, HK, Gillingwater, TH, and Groen, EJN (2019). Molecular Mechanisms Underlying Sensory-Motor Circuit Dysfunction in SMA. Front Mol Neurosci 12: 59.

71. Ling, KK, Lin, MY, Zingg, B, Feng, Z, and Ko, CP (2010). Synaptic defects in the spinal and neuromuscular circuitry in a mouse model of spinal muscular atrophy. PLoS One 5: el5457.

72. Mentis, GZ, Blivis, D, Liu, W, Drobac, E, Crowder, ME, Kong, L, et al. (2011). Early functional impairment of sensory-motor connectivity in a mouse model of spinal muscular atrophy. Neuron 69: 453-467.

73. Shorrock, HK, van der Hoom, D, Boyd, PJ, Llavero Hurtado, M, Lamont, DJ, Wirth, B, et al. (2018). UBAl/GARS-dependent pathways drive sensory-motor connectivity defects in spinal muscular atrophy. Brain 141: 2878-2894.

74. Fletcher, EV, Simon, CM, Pagiazitis, JG, Chalif, JI, Vukojicic, A, Drobac, E, et al. (2017). Reduced sensory synaptic excitation impairs motor neuron function via Kv2.1 in spinal muscular atrophy. Nat Neurosci 20: 905-916.

75. Buettner, JM, Sime Longang, JK, Gerstner, F, Apel, KS, Blanco-Redondo, B, Sowoidnich, L, et al. (2021). Central synaptopathy is the most conserved feature of motor circuit pathology across spinal muscular atrophy mouse models. iScience 24: 103376.

35

SUBSTITUTE SHEET (RULE 26) Hoye, ML, Regan, MR, Jensen, LA, Lake, AM, Reddy, LV, Vidensky, S, et al. (2018). Motor neuron-derived microRNAs cause astrocyte dysfunction in amyotrophic lateral sclerosis. Brain 141 2561-2575. Mercuri, E, Pera, MC, Scoto, M, Finkel, R, and Muntoni, F (2020). Spinal muscular atrophy - insights and challenges in the treatment era. Nat Rev Neurol. Messina, S, and Sframeli, M (2020). New Treatments in Spinal Muscular Atrophy: Positive Results and New Challenges. J Clin Med 9 Abiusi, E, Infante, P, Cagnoli, C, Lospinoso Severini, L, Pane, M, Coratti, G, et al. (2021). SMA-miRs (miR-181a-5p, -324-5p, and -451a) are overexpressed in spinal muscular atrophy skeletal muscle and serum samples. eLife 10. Spicer, C, Lu, CH, Catapano, F, Scoto, M, Zaharieva, I, Malaspina, A etal. (2021). The altered expression of neurofilament in mouse models and patients with spinal muscular atrophy. Annals of clinical and translational neurology. Okita, K, Matsumura, Y, Sato, Y, Okada, A, Morizane, A, Okamoto, S, et al. (2011). A more efficient method to generate integration-free human iPS cells. Nature methods 8: 409-412. Wang, TH, Huang, CC, and Hung, JH (2021). EARRINGS: an efficient and accurate adapter trimmer entails no a priori adapter sequences. Bioinformatics 37: 1846-1852. Chou, MT, Han, BW, Hsiao, CP, Zamore, PD, Weng, Z, and Hung, JH (2015). Tailor: a computational framework for detecting non-templated tailing of small silencing RNAs. Nucleic Acids Res 43: el09. Frankish, A, Diekhans, M, Ferreira, AM, Johnson, R, Jungreis, I, Loveland, J, et al. (2019). GENCODE reference annotation for the human and mouse genomes. Nucleic Acids Res 47: D766-D773. Kozomara, A, Birgaoanu, M, and Griffiths-Jones, S (2019). miRBase: from microRNA sequences to function. Nucleic Acids Res 47: D155-D162. Shirak, A, Grabherr, M, Di Palma, F, Lindblad-Toh, K, Hulata, G, Ron, M, et al. (2010). Identification of repetitive elements in the genome of Oreochromis niloticus: tilapia repeat masker. Mar Biotechnol (NY) 12: 121-125. Huang da, W, Sherman, BT, and Lempicki, RA (2009). Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. NatProtoc 4: 44-57. Tung, YT, Lu, YL, Peng, KC, Yen, YP, Chang, M, Li, J, et al. (2015). Mir-17 approximately 92 Governs Motor Neuron Subtype Survival by Mediating Nuclear PTEN. Cell reports 11: 1305- 1318. Hoye, ML, Koval, ED, Wegener, AJ, Hyman, TS, Yang, C, O'Brien, DR, et al. (2017). MicroRNA Profiling Reveals Marker of Motor Neuron Disease in ALS Models. JNeurosci 37: 5574-5586.

SUBSTITUTE SHEET (RULE 26)

Claims

CLAIMS What is claimed is:

1. A method for treating spinal muscular atrophy (SMA) in an SMA patient, the method comprising administrating to the patient an effective amount of a MiR34 molecule.

2. The method of claim 1, wherein the MiR34 includes MiR34a, MiR34b, MiR34c or any combination thereof.

3. The method of claim 1 or 2, wherein administering the effective amount of the MiR34 molecule results in restoring motor function of the patient.

4. The method of any of claims 1-3, wherein the MiR34 molecule is a single-strand RNA molecule or a duplex RNA molecule, or the MiR34 molecule is encoded by an expression vector.

5. The method of any of claims 1-4, wherein the MiR34 molecule is administered intravenously, intramuscularly, intranasally or intrathecally.

6. The method of any of claims 1-5, wherein the patient is an infant, a child, an adolescent or an adult.

7. The method of any of claims 1-6, wherein the patient has been diagnosed with type 0 SMA, type 1 SMA, type 2 SMA, type 3 SMA or type 4 SMA.

8. Use of a MiR34 molecule as described in any of preceding claims for manufacturing a medicament for treating SMA

9. A pharmaceutical composition for use in treating SMA, which comprises an effective amount of a MiR34 molecule as described in any of preceding claims and a pharmaceutically acceptable carrier.

10. A method for prognosis of spinal muscular atrophy (SMA) in an SMA patient, the method comprising

(i) providing a biological sample obtained from the patient before treatment; and

(li) detecting a biomarker in the biological sample to obtain a baseline detection level, comparing the baseline detection level with a baseline reference level to obtain a comparison result, and determining the prognosis of the patient based on the comparison result, wherein the biomarker

37

SUBSTITUTE SHEET (RULE 26) includes a MiR34 and a higher baseline detection level as compared to the baseline reference level is indicative of a positive prognosis.

11. The method of claim 10, wherein the MiR34 includes MiR34a, MiR34b, MiR34c or any combination thereof; and/or the biomarker further includes pNfH.

12. The method of claim 10 or 11, wherein the positive prognosis includes improvement of motor function.

13. The method of any of claims 10-12, wherein the patient is an infant, a child, an adolescent or an adult.

14. The method of any of claims 10-13, wherein the patient has been diagnosed with type 0 SMA, type 1 SMA, type 2 SMA, type 3 SMA or type 4 SMA.

15. The method of any of claims 10-14, wherein the treatment comprises administration of an SMA drug.

16. The method of claim 19, wherein the SMA drug comprises nusinersen, zolgensma or risdiplam.

17. A method of predicting an SMA patient's response to a treatment, comprising

(i) providing a first biological sample from the patient at a first time point;

(li) providing a second biological sample from the patient at a second time point, which is later than the first time point;

(lii) detecting a biomarker in the first biological sample and the second biological sample to obtain a first detection level and a second detection level, respectively;

(iv) comparing the second detection level with the first detection level to obtain a comparison result, and determining the patient’s response to the therapy regimen based on the comparison result, wherein the biomarker includes a MiR34, the first time point is before the treatment and the second time point is after the treatment, or the first time point and the second time point are after the treatment, and a decrease in the second detection level as compared to the first detection level indicates that the patient has positively responded to the therapy regimen.

18. The method of claim 17, wherein

38

SUBSTITUTE SHEET (RULE 26) the MiR34 includes MiR34a, MiR34b, MiR34c or any combination thereof; and/or the biomarker further includes pNfH.

19. The method of claim 17 or 18, wherein the patient positively responded to the treatment exhibits improvement of motor function.

20. The method of any of claims 17-19, wherein the patient is an infant, a child, an adolescent or an adult.

21. The method of any of claims 17-20, wherein the patient has been diagnosed with type 0 SMA, type 1 SMA, type 2 SMA, type 3 SMA or type 4 SMA.

22. The method of any of claims 17-21, wherein the treatment comprises administration of an SMA drug.

23. The method of claim 22, wherein the SMA drug comprises nusinersen, zolgensma or risdiplam.

24. The method of any of claims 17-23, wherein the second time point is at 64 days, 183 days, 482 days or more later than the first time point.

25. The method of any of claims 10-24, further comprising conducting a method for treating SMA.

26. A kit or composition for use in performing a method of any of claims 10-24, which comprises a reagent that specifically recognizes the biomarker, and instructions for using the kit to detect the presence or amount of the biomarker.

27. The kit or composition for use of claim 26, wherein the reagent is linked to a detectable label.

28. Use of a reagent that specifically recognizes the biomarker as described in any of the preceding claims for performing a method for prognosis of SMA as defined in any of claims 10-16 or a method for predicting an SMA patient’s response to a treatment as defined in any of claims 17-24, or in the manufacture of a kit or a composition for performing a method for prognosis of SMA as defined in any of claims 10- 16 or a method for predicting an SMA patient’s response to a treatment as defined in any of claims 17-24.

39

SUBSTITUTE SHEET (RULE 26)