WO2025097023A1

WO2025097023A1 - Electrokinetic assessment of critical quality attributes of nanoparticles

Info

Publication number: WO2025097023A1
Application number: PCT/US2024/054218
Authority: WO
Inventors: Mikhail Skliar; Yunshan WANG
Original assignee: University of Utah Research Foundation Inc
Current assignee: University of Utah Research Foundation Inc
Priority date: 2023-11-03
Filing date: 2024-11-01
Publication date: 2025-05-08
Anticipated expiration: 2026-05-03

Abstract

Methods of analyzing samples containing a plurality of particles using electrokinetic microfluidic devices.

Description

ELECTROKINETIC ASSESSMENT OF CRITICAL QUALITY ATTRIBUTES OF NANOPARTICLES

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Application No. 63/547,259, filed November 3, 2023. which is incorporated into this application in its entirety.

BACKGROUND

[0002] Biotherapeutic nanoparticles (bioNPs) are an important class of biologies. Examples include formulated nanoparticle (NP) therapeutics, such as lipid nanoparticles (LNPs) and viral vectors containing biologically active cargo. Cell-secreted and engineered therapeutic extracellular vesicles (EVs) and exosomes (the smallest EVs) are rapidly advancing to treat acute and intractable diseases and chronic conditions with precision and minimal side effects.

[0003] COVID-19 vaccine development highlighted biomanufacturing as the critical bottleneck for separating successful bioNP products from the competition. In April 2021, there were 237 vaccine candidates. By October, only six vaccines, of which four were bioNP products (two mRNA LNPs and two viral-vectored), managed to overcome manufacturing challenges and were eventually produced in billions of doses worldwide. The central problem that complicated at-scale and flexible production of bioNPs with consistent efficacy is the lack of feedback on product quality during biomanufacturing to control product variability.

[0004] Without real time and dynamic feedback, the biomanufacturing of bioNPs remains open-loop and recipe-based and must rely on laboratory' analyses of physical, compositional, and potency characteristics of bioNPs used to infrequently validate the expected product properties. Such time-consuming and costly laboratory methods cannot be used to automatically compensate unavoidable variability during bioNP production. Laboratory testing is also ill-suited for quality assurance within the supply chain, which may require worldwide cold-chain distribution and storage.

[0005] Extracellular vesicles (EVs) are a class of bioNPs that are nano- to micro- sized particles which inherit molecular content from secreting cells. EVs have shown promise in treating, for example, neurological, cardiac, and renal diseases. Therapeutic EVs are produced in clinically meaningful quantities inside bioreactors. However, biomanufacturing EVs with consistent therapeutic properties remains a challenge. Grow th conditions vary inside bioreactors, strongly influence the EVs’ secretion attributes and composition. A rapid screening method to frequently assess produced EVs for compliance with critical quality attributes (CQA) and detect and compensate for variations in growth conditions is needed to ensure the therapeutic consistency of EVs produced at scale. Such a screening is also useful in controlling the reprogramming of producer cells by transient exposure to varying growth conditions and biologically active substances. Screening techniques which can assess other types of nanoparticles, such as peptides, polymer nanoparticles, and metallic nanoparticles, without a required step of labeling such nanoparticles, are also desired. This disclosure addresses these needs in the art.

SUMMARY

[0006] Described is a method of analyzing samples comprising a plurality of particles, comprising: a) introducing the sample into a system comprising a microfluidic electrokinetic device which comprises: (i) at least one flow channel comprising a fluid inlet and a fluid outlet, wherein the flow channel extends from the fluid inlet to the fluid outlet along a first axis, wherein the flow channel comprises a first wall and an opposing second wall, wherein the first and second walls have opposing bodies that converge to create a constriction within the flow channel, the opposing bodies comprising an electrical insulator; (ii) a power supply and a plurality⁷ of electrodes in electrical communication with the pow er supply, wherein the power supply is configured to apply a voltage between at least two electrodes of the plurality of electrodes to provide a non-uniform electric field across the flow' channel; (iii) a light source having an output beam path configured to irradiate the plurality of particles in the flow channel; and (iv) an optical device comprising a photon detector configured to detect light scattered or emitted by the plurality of particles; b) applying a voltage to the electrodes via the power supply to provide the non-uniform electric field, which exerts an electrokinetic (EK) force on the plurality of particles within the flow channel; c) irradiating the plurality of particles w ith the output beam path, wherein the output beam path is irradiated along a second axis; d) detecting the light scattered or emitted by the plurality⁷ of particles using the optical device, wherein detection generates a profile for the sample comprising the plurality of particles; and e) determining the deviation of the profile for the sample comprising the plurality of particles from a standard profile, wherein the standard profile is characteristic of one or more of a particle size distribution, a mean particle size, a zeta potential distribution, a mean zeta potential, a Clausius-Mossotti factor distribution, a dielectric permittivity, or a mean particle Clausius-Mossotti factor.

[0007] Also disclosed are microfluidic electrokinetic devices useful for analyzing samples comprising a plurality of particles using the methods disclosed. In some aspects, the electrokinetic devices comprise a flow channel comprising multiple constrictions that serially narrow along the first axis. In other aspects, the electrokinetic devices comprises multiple flow channels arranged in a parallel configuration. In such aspects, each flow channel comprises one constriction. The electric fields and electrokinetic forces created inside the flow channel can be created by two or more electrodes externally excited by a constant or alternating electrical potentials.

BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The foregoing summary, as well as the following description of the disclosure, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the disclosure, the drawings illustrate some, but not all, alternative embodiments. This disclosure is not limited to the precise arrangements and instrumentalities shown. The following figures, which are incorporated into and constitute part of the specification, assist in explaining the principles of the disclosure.

[0009] FIG. 1 is a schematic showing an exemplary aspect of a system comprising a electrokinetic device.

[0010] FIG. 2 is a schematic of an exemplary⁷ aspect of an electrokinetic device comprising multiple flow channels.

[0011] FIG. 3 is a schematic showing the gating forces exerted on particles and the formation of the retention patterns as calculated using 2D Multiphysics simulations.

[0012] FIG. 4 are representative results from 2D Multiphysics simulations showing the effects of particle zeta potential ranging from -10, -15, and -20 mV (FIG. 4A), particle sizes changing from 100, 150. to 200 nm (FIG. 4B), and applied voltage set to 200, 500. and 800 V (FIG. 4C) on the trapping patterns. FIG. 4D shows the retention zones for particles with 100, 150, and 200 nm diameters at three gates in the channel with three sequential gates. The gap size decreased from 2 to 1.2 to 0.6 pm from left to right. The sample flows from the left to the right.

[0013] FIG. 5 is a schematic conceptually illustrating deviations in the sample profile from a standard profile shown in the top row of the image.

[0014] FIG. 6 are schematics show ing the electrokinetic chip and holder. FIG. 6A show s the design of the EK chip. FIG. 6B show s the design of the chip holder. [0015] FIG. 7 shows representative results from COMSOL mathematical modeling. FIG. 7A is a schematic showing the PDMS microchannel with triangular dielectric features modifying the electric field imposed by the electrodes shown inserted into the inlet and outlet of a microfluidic channel. FIG. 7B show s the opening region located between the insulating features (posts). FIG. 7C is a plot of the net particle velocity in the gated region imposed by electrokinetic forces. The zone of negative particle velocities indicates that particles without the trapping zones may move within it without escaping.

[0016] FIG. 8 shows representative simulation data for the DEP force and the EO and EP forces as a function of voltage. FIG. 8 A is the cut-point for the analysis of EK forces. FIG.

8B is a plot showing the DEP and EO+EP forces as a function of the applied voltage for a distance between the insulating posts of 2 pm and a particle zeta potential of -15 mV. All EK forces balance when the applied voltage is -680 V.

[0017] FIG. 9 shows representative simulation data for the trapping analysis in one opening region and variation of the DC applied voltage for a particle diameter of 100 nm, a 2 pm distance between insulating posts, and -15 mV of particle zeta potential. Trapping profiles are shown for DC applied voltages of 100 V (FIG. 9A), 200 V (FIG. 9B), 500 V (FIG. 9C), and 800 V (FIG. 9D).

[0018] FIG. 10 shows representative simulation data for the trapping analysis in one opening region and variation of the particle zeta potential for a particle diameter of 100 nm, a 2 pm distance between insulating posts, and 500 V of applied voltage. Trapping profiles are shown for particle zeta potentials of -5 mV (FIG. 10 A), -10 mV (FIG. 10B), -15 mV (FIG. 10C), and -20 mV (FIG. 10D).

[0019] FIG. 11 shows representative simulation data for the trapping analysis in one opening region and variation of Re{K} for a particle diameter of 100 nm, a 2 pm distance between insulating posts, and 500 V of applied voltage. Trapping profiles are shown for Re{K} of -0.5 (FIG. 11 A), -0.25 (FIG. 1 IB), 0 (FIG. 11C), 0.25 (FIG. HD), 0.5 (FIG. HE), and 1 (FIG. 1 IF).

[0020] FIG. 12 shows representative simulation data for the trapping analysis in one opening region and variation of particle diameter for a particle zeta potential of -15 mV, a 2 pm distance between insulating posts, and 500 V of applied voltage. Trapping profiles are shown for particle diameters of 30 nm (FIG. 12A), 50 nm (FIG. 12B), 100 nm (FIG. 12C), 150 nm (FIG. 12D), and 200 nm (FIG. 12E).

[0021] FIG. 13 shows representative simulation data for trapping analysis in a multiinsulating post flow channel with 3 gates of different widths and 3 different particle sizes. The trapping profiles are shown for the first gate (FIG. 13A), second gate (FIG. 13B), and third gate (FIG. 13C).

[0022] FIG. 14 shows representative simulation data for three gates in series in a microfluidic device. FIG. 14A shows the trapping field at the three gates (gi = 2 pm, g2 = 1.5 pm, g3 = 0.6 pm). FIG. 14B shows that selective trapping in a microfluidic platform with consecutive gates of widths 2 pm (particle diameter > 140 nm), 1.5 pm (particle diameter > 70 nm), and 0.6 pm (particle diameter > 15 nm) can take place at 75V.

DETAILED DESCRIPTION

[0023] Provided are methods of analyzing samples comprising a plurality of particles, comprising: a) introducing the sample into a system comprising a microfluidic electrokinetic device which comprises: (i) at least one flow channel comprising a fluid inlet and a fluid outlet, wherein the flow channel extends from the fluid inlet to the fluid outlet along a first axis, wherein the flow channel comprises a first wall and an opposing second wall, wherein the first and second walls have opposing bodies that converge to create a constriction within the flow channel, the opposing bodies comprising an electrical insulator; (ii) a power supply and a plurality of electrodes in electrical communication with the power supply, wherein the power supply is configured to apply a voltage between at least two electrodes of the plurality of electrodes to provide a non-uniform electric field across the flow channel; (iii) a light source having an output beam path configured to irradiate the plurality of particles in the flow channel; and (iv) an optical device comprising a photon detector configured to detect light scattered or emitted by the plurality of particles; b) applying a voltage to the electrodes via the power supply to provide the non-uniform electric field, which exerts an electrokinetic (EK) force on the plurality of particles within the flow channel; c) irradiating the plurality of particles with the output beam path, wherein the output beam path is irradiated along a second axis; d) detecting the light scattered or emitted by the plurality of particles using the optical device, wherein detection generates a profile for the sample comprising the plurality of particles; and e) determining the deviation of the profile for the sample comprising the plurality⁷ of particles from a standard profile, wherein the standard profile is characteristic of one or more of a particle size distribution, a mean particle size, a zeta potential distribution, a mean zeta potential, a Clausius-Mossotti factor distribution, a dielectric permittivity, or a mean particle Clausius-Mossotti factor.

[0024] Accordingly, provided are methods for analyzing a sample comprising a plurality of particles comprising introducing the sample into a system comprising a microfluidic electrokinetic device in combination with detection of the light scattered or emitted by the plurality of particles. The light scattered from the one or more particles can be detected using, for example. Nanoparticle Tracking Analysis (NT A). NTA offers the advantage of label-free nanoparticle visualization and analysis in real time, allowing for the simultaneous characterization and separation of samples with mixed and unknown particles.

[0025] A “trapping zone” is a collection of points in the flow channel where at least a portion of the plurality of particles are stationary at the boundary due to a balance (i.e.. net zero) of electrokinetic forces consisting of dielectrophoretic, electroosmotic and electrophoretic forces. The boundary' of the “trapping zone” can also be described as the location where a particle’s velocity' along the field line is zero. Inside the trapping zone, particles may move until reaching the zone's boundary’ where the motion seizes. In some aspects, the sample profile generated by the methods described is characterized by trapping zones at the opposing bodies that converge to create a constriction within the flow channel. In such aspects, the trapping zones “trap” subpopulations in the plurality' of particles.

[0026] In some aspects, the sample profile generated by the methods described is characterized by changes in the flow profile of subpopulations in the plurality of particles due to the applied non-uniform electric field.

I. Methods of Analyzing a Sample

A. Insulator-Based Dielectrophoretic Devices

[0027] The methods disclosed relate to insulator-based electrokinetic microfluidic devices. Dielectrophoresis (DEP) is an electrodynamic transport mechanism with a nonlinear dependence on electric field. A non-uniform electric field produces an electrodynamic force on a dielectric particle producing a force toward the either region of higher (positive DEP) or lower (negative DEP) electric field gradient. In addition to the DEP force, other electrokinetic forces, including electroosmotic (caused by the entraining motion of the fluid under the influence of the elective field) and electrophoretic (caused by the electric field force on the charged surface of a particle) can contribute to the net electrokinetic force on a particle. The net electrokinetic force results in the particle motion that can occur in either direct (DC), alternating (AC) electric fields, or a combination thereof. Insulator-based dielectrophoresis (iDEP) is an alternative to conventional electrode-based dielectrophoresis (eDEP) systems. In iDEP, insulating structures are used to generate nonuniform electric fields. iDEP method differs from traditional DEP separation in that a voltage, created by either DC, AC, or a combination thereof, is applied to electrodes located in remote inlet and outlet reservoirs and the field nonuniformities are generated by arrays of insulating posts located within the channel.

[0028] iDEP offers several advantages compared with traditional DEP. The use of remote electrodes avoids many of the problems associated with embedded electrodes, such as electrochemical reactions and bubble generation at the electrode surfaces. Additionally, the use of DC voltages in eDEP creates many issues, which are not encountered in iDEP. The use of a DC field can be advantageous because it can be used to drive both electrophoretic and dielectrophoretic transports, allowing greater control over particle movement. The combination of iDEP and eDEP, obtained when the DC and AC voltages are superimposed, has the advantage of revealing frequency-dependent dielectric permittivity of particles.

[0029] As illustrated in FIG. 1, the electrokinetic microfluidic device 100 can comprise at least one flow channel 105 comprising a fluid inlet 107 and a fluid outlet 109, wherein the flow channel 105 extends from the fluid inlet 107 to the fluid outlet 109 along a first axis 118, wherein the flow channel 105 comprises a first wall 110 and a second wall 115, wherein the first 110 and second walls 115 have opposing bodies 117 that converge to create a constriction 120 within the flow channel 105, the opposing bodies comprising an electrical insulator 119. The electrokinetic microfluidic device 100 can also comprise a power supply 125, a plurality of electrodes 130 in electrical communication with the power supply 125, a light source 135 having an output beam path 140 configured to irradiate the plurality of particles 102 along a second axis 145, and an optical device 150 comprising a photon detector 153 configured to detect light scattered or emitted 155 by the plurality of particles 102. The aspect shown in FIG. 1 comprises one flow⁷ channel 105 with multiple constrictions 120. The multiple constrictions in FIG. 1 can serially decrease in size along the flow of the channel (left to right). In such a configuration, a sample comprising a plurality of particles having, for example, multiple mean particle sizes, can be analyzed. The largest particles are trapped at the widest constriction, and the smallest particles are trapped at the narrowest constriction. [0030] In some aspects, and as shown in FIG. 2. the electrokinetic microfluidic device 220 can comprise multiple flow channels 230. In this aspect, each flow channel 230 comprises one constriction 240 formed by the first wall 245 and the second wall 250 having opposing bodies 252 that converge to create the constriction 240. In this aspect, the electrokinetic microfluidic device 220 can comprise a single inlet or multiple inlets (not shown) through which the sample comprising the plurality of particles can be introduced into the multiple flow channels. The electrokinetic microfluidic device can also comprise a single outlet or multiple outlets (not shown). The electrokinetic microfluidic device 220 can also comprise a power supply 260 and a plurality of electrodes 255 in electrical communication with the power supply 260. The constrictions in FIG. 2 can have different sizes. EK forces are smaller when the constriction is larger. Thus, the smallest subpopulation of particles are trapped at the largest constriction. As the constriction narrows (for example, from left to right in FIG. 2), a larger subpopulation of particles are trapped.

[0031] In some aspects, the constriction is between 0.5 pm to 10 pm. The constriction can be, for example, between 0.5 pm to 9 pm, 0.5 pm to 8 pm, 0.5 pm to 7 pm, 0.5 pm to 6 pm, 0.5 pm to 5 pm, 0.5 pm to 4 pm, 0.5 pm to 3 pm, 0.5 pm to 2 pm, 0.5 pm to 1 pm, 1 pm to 10 pm, 1 pm to 8 pm, 1 pm to 6 pm, 1 pm to 5 pm, 1 pm to 4 pm, 1 pm to 3 pm, 1 pm to 2 pm, 2 pm to 8 pm, 2 pm to 6 pm. 2 pm to 5 pm, 2 pm to 3 pm, 3 pm to 10 pm, 3 pm to 8 pm, 3 pm to 6 pm, 3 pm to 5 pm, 5 pm to 10 pm, 5 pm to 8 pm, 5 pm to 6 pm, 6 pm to 10 pm, 6 pm to 8 pm, or 8 pm to 10 pm. The constriction can be, for example, 0.5 pm, 0.6 pm, 0.7 pm, 0.8 pm, 0.9 pm, 1 pm, 1.5 pm, 2 pm, 2.5 pm, 3 pm, 3.5 pm, 4 pm, 4.5 pm, 5 pm, 5.5 pm, 6 pm, 6.5 pm, 7 pm, 7.5 pm, 8 pm, 8.5 pm, 9 pm, 9.5 pm, or 10 pm.

[0032] In some aspects, the flow channel in the electrokinetic microfluidic device is fabricated at the millimeter to nanometer scale.

[0033] In some aspects, the flow channel has a length along the first axis from 1 mm to 20 mm. The flow’ channel can have a length along the first axis of, for example, 1 mm to 15 mm, 1 mm to 10 mm, 1 mm to 5 mm, 1 mm to 3 mm. 3 mm to 20 mm, 3 mm to 15 mm, 3 mm to 10 mm, 3 mm to 5 mm, 5 mm to 20 mm, 5 mm to 15 mm, 5 mm to 10 mm, 10 mm to 20 mm, 10 mm to 15 mm, or 15 mm to 20 mm. The flow channel can have a length along the first axis of, for example, 1 mm, 3 mm, 5 mm, 7 mm, 10 mm, 13 mm, 15 mm, 18 mm, or 20 mm. In some aspects, increasing the number of constrictions in the flow channel increases the length of the flow channel.

[0034] The opposing bodies that converge to create a constriction within the flow channel comprise an electrical insulator. In some aspects, the electrical insulator can be composed of a polymer, glass, silicon, or combinations thereof. In further aspects, the electrical insulator is a polymer. In yet further aspects, the electrical insulator is polymethylsiloxane (PDMS).

[0035] In some aspects, the opposing bodies comprising an electrical insulator that converge to create a constriction within the How channel are configured to selectively separate at least a portion of the plurality of particles in the sample and allow passage of a second portion of the plurality of particles. In further aspects, the number of constrictions in the flow channel is determined by the number of separations observed in a standard sample having one or more of particle size distribution, mean particle size, zeta potential distribution, mean zeta potential, Clausius-Mossotti factor distribution, or mean particle Clausius-Mossotti factor. Larger number of constrictions create a more detailed nanoparticle retention pattern for a heterogeneous samples comprising pluralities of particles. For example, when heterogeneity is in particle sizes only (all other properties are the same), the number and the width of constrictions and the applied external electrical excitation can be selected such that subpopulations with the same particle diameter are trapped at distinct constrictions, with sizes of trapped particles decreasing with the direction of the flow, and smaller particles trapped at the narrower constrictions.

[0036] In some aspects, the flow channel comprises multiple constrictions formed by the first and second walls converging within the flow channel. In a further aspect, the flow channel comprises one constriction formed by the first and second walls converging within the flow channel.

[0037] In some aspects, the flow channel comprises multiple constrictions that serially narrow along the first axis. In a further aspect, the constrictions serially narrow by from 0. 1 pm to 5 pm. The constrictions can serially narrow by. for example, from 0. 1 pm to 4 pm, 0. 1 pm to 3 pm, 0. 1 pm to 2 pm, 0. 1 pm to 1 pm, 0. 1 pm to 0.5 pm, 0.5 pm to 5 pm, 0.5 pm to 4 pm, 0.5 pm to 3 pm, 0.5 pm to 2 pm, 0.5 pm to 1 pm, 1 pm to 5 pm, 1 pm to 4 pm, 1 pm to 3 pm, 1 pm to 2 pm, 2 pm to 5 pm, 2 pm to 4 pm, 2 pm to 3 pm, 3 pm to 5 pm, 3 pm to 4 pm, or 4 pm to 5 pm.

[0038] In some aspects, the microfluidic electrokinetic device comprises multiple flow channels arranged in a parallel configuration. Such aspects are particularly suitable for interference-free light scattering measurement of the particles’ position and motion. In a parallel configuration, different voltages can be applied across each flow channel. Without wishing to be bound by theory, such a configuration can generate a sample profile wherein, when a sample comprising a plurality of particles with particle subpopulations A, B, and C, the first constriction in a first flow channel can “trap” subpopulation A. The second gate in a second flow channel can “trap” subpopulations A and B. Finally, the third gate in a second flow channel can “trap” subpopulations A, B, and C. The subpopulations can be in a state of motion between trapped and untrapped in the parallel configuration. The amount of trapped vs. untrapped particles generally increases as the width of the constriction decreases.

[0039] In some aspects, the applied voltage is from 5 V to 1000 V. The applied voltage can be, for example, from 5 V to 900 V, 5 V to 500 V, 5 V to 100 V, 5 V to 50 V, 5 V to 25 V, 5 V to 10 V, 25 V to 1000 V. 25 V to 500 V. 25 V to 100 V. 50 V to 1000 V, 50 V to 500 V, 50 V to 100 V, 100 V to 900 V, 100 V to 800 V, 100 V to 700 V, 100 V to 600 V. 100 V to 500 V. 100 V to 400 V, 100 V to 300 V, 100 V to 200 V. 200 V to 1000 V, 200 V to 900 V, 200 V to 800 V, 200 V to 700 V, 200 V to 600 V, 200 V to 500 V, 200 V to 400 V, 200 V to 300 V, 300 V to 1000 V, 300 V to 900 V, 300 V to 800 V, 300 V to 700 V, 300 V to 600 V, 300 V to 500 V, 300 V to 400V, 400 V to 1000 V, 400 V to 900 V, 400 V to 800 V, 400 V to 700 V. 400 V to 600 V, 400 V to 500 V, 500 V to 1000 V, 500 V to 900 V, 500 V to 800 V, 500 V to 700. 500 V to 600 V, 600 V to 1000 V. 600 V to 900 V, 600 V to 800 V, or 800 V to 1000 V. The applied voltage can be, for example, 5 V, 10 V, 25 V, 50 V, 75 V, 100 V, 150 V, 200 V, 250 V, 300 V, 350 V, 400 V, 450 V, 500 V, 550 V, 600 V, 650 V, 700 V, 750 V, 800 V, 850 V. 900 V, 950 V, or 1000 V.

[0040] In some aspects, the applied voltage is applied via the power supply using direct current, alternating current, or a combination thereof. In a further aspect, the applied voltage is applied via the power supply using direct current.

[0041] The light source having an output beam path configured to irradiate the plurality of particles can be a coherent light source or an incoherent light source. In some aspects, the light source is a coherent light source. In a further aspect, the light source is a laser having a wavelength of from 300 nm to 800 nm. The wavelength of the laser can be, for example, from 300 nm to 700 nm, 300 nm to 600 nm, 300 nm to 500 nm, 300 nm to 400 nm, 400 nm to 800 nm, 400 nm to 700 nm, 400 nm to 600 nm, 400 nm to 500 nm, 500 nm to 800 nm, 500 nm to 700 nm, 500 nm to 600 nm, 600 nm to 800 nm. 600 nm to 700 nm. or 700 nm to 800 nm. The wavelength of the laser can be, for example, 300 nm, 350 nm, 400 nm, 450 nm, 500 nm, 550 nm, 600 nm, 650 nm, 700 nm, 750 nm, or 800 nm. In a yet further aspect, the light source is a laser having a wavelength of from 300 nm to 500 nm. An exemplary' laser suitable for the methods described is a NanoSight LM 10 405 nm laser device manufactured by Malvern Panalytical. In some aspects, the light source is an incoherent light source, such as a white light source.

[0042] In some aspects, the light source comprises a scanning beam, a stationary' beam, or a combination thereof.

[0043] In some aspects, the output beam path irradiates the plurality of particles at multiple locations in the flow channel. In a further aspect, two or more output beam paths irradiate the plurality' of particles at the same location or multiple locations.

[0044] The optical device can be any optical device known in the art which can be configured to detect light scattered or emitted from nano or microscale particles. In some aspects, the optical device is an optical microscope fitted with one or more scanning objectives that direct light to the photon detector.

[0045] The photon detector is configured to detect light scattered or emitted by the plurality of particles or a single particle in the plurality of particles. In some aspects, the photon detector includes a camera, such as a charge-coupled device (CCD) detector or a complementary' metal-oxide-semiconductor (CMOS) detector. In a further aspect, the camera detects light scattered by the plurality of particles or a single particle in the plurality of particles. In aspects wherein the camera detects light scattered, the method described is a label-free particle analysis, i.e., the plurality of particles or a single particle in the plurality of particles are not labeled using dyes, fluorophores, radioisotopes, etc.

[0046] In some aspects, the photon detector is configured to detect light scattered by the plurality of particles at multiple locations within the flow channel.

[0047] In some aspects, the photon detector is configured to detect light emitted by the plurality of particles. In a further aspect, the photon detector is configured to detect fluorescence emitted by the plurality' of particles. The plurality' of particles can be labeled with a fluorophore or a dye. The fluorophore or dye can be any fluorophore or dye which is known in the art as suitable for labeling nanoparticles. Non-limiting examples of fluorophores or dyes include fluorescein, Alexa Fluor dyes, BODIPY dyes, green fluorescent protein, (GFP), rhodamine B, Nile Red, Cy3, Cy5, Cy 7, coumarin dyes, phycobilliproteins, phycocyanin, phycoery thrin, and phycoerythrocyanin. The photon detector may include fluorescence filters that are configured to filter out the excitation wavelength of light produced by the light source so that fluorescence emitted from the plurality of particles can be detected.

[0048] The light scattered by particles could be used to observe their individual motion under the effect of electrokinetic forces. Such motion can then be used to infer the size of a particle using principles of nanoparticle tracking analysis that correlates the Brownian fluctuation in particle’s motion to its hydrodynamic diameter. The motion of a particle of thus determined size can then be used to infer their other properties, such as zeta potential distribution, mean zeta potential, dielectric permittivity, Clausius-Mossotti factor distribution, or mean particle Clausius-Mossotti factor.

B. Samples

[0049] The sample comprising a plurality of particles can comprise any particles which are suitable for analysis in a microfluidic device and are can be dispersed in a fluid within the flow channel. Without wishing to be bound by theory, parameters such as the applied voltage, flow channel dimensions, number of flow channels, and number of constrictions can be adjusted to be suitable for analysis for a specific sample of interest.

[0050] In some aspects, the plurality of particles are selected from extracellular vesicles, peptides, proteins, DNA, RNA, micro-organisms, amino acids, nucleotides, nucleic acid molecules, glycoproteins, carbohydrates, lipids, lectins, cells, viruses, viral particles, bacterial, organelles, spores, protozoa, yeasts, molds, fungi, pollens, diatmons, toxins, biotoxins, immunoglobulins, antibodies, supramolecular assemblies, quantum dots, metallic nanoparticles, polymeric nanoparticles, dendrimers, carbon-based nanomaterials, liposomes, semiconductor nanoparticles, or a combination thereof. In a further aspect, the plurality of particles are extracellular vesicles.

[0051] The plurality of particles can include particles having a variety of shapes, including generally spherical or irregular shapes, flakes, needle-like particles, chips, fibers, equiaxed particles, etc.

[0052] In some aspects, the sample comprising a plurality of particles is homogeneous. In a further aspect, the sample comprising a plurality of particles is heterogeneous and comprises subpopulations of particles.

[0053] In some aspects, the plurality' of particles have a mean diameter of from 15 to 200 nm. The plurality of particles can have a mean diameter of, for example, from 15 nm to 150 nm, 15 nm to 125 nm, 15 nm to 100 nm. 15 nm to 75 nm, 15 nm to 50 nm, 15 nm to 30 nm, 15 nm to 20 nm, 20 nm to 200 nm, 20 nm to 150 nm, 20 nm to 125 nm, 20 nm to 100 nm, 20 nm to 75 nm, 20 nm to 50 nm, 20 nm to 30 nm, 30 nm to 200 nm, 30 nm to 150 nm, 30 nm to 125 nm, 30 nm to 100 nm, 30 nm to 75 nm, 30 nm to 50 nm, 50 nm to 200 nm, 50 nm to 150 nm, 50 nm to 125 nm, 50 nm to 100 nm. 50 nm to 75 nm, 75 nm to 200 nm, 75 nm to 150 nm, 75 nm to 125 nm, 75 nm to 100 nm. 100 nm to 200 nm, 100 nm to 150 nm, or 150 nm to 200 nm. The plurality of particles can have a mean particle diameter of, for example, 15 nm, 20 nm, 25 nm, 30 nm, 35 nm, 40 nm, 45 nm, 50 nm, 55 nm, 60 nm, 65 nm, 70 nm, 75 nm, 80 nm, 85 nm. 90 nm, 100 nm, 110 nm. 120 nm, 130 nm, 140 nm, 150 nm, 160 nm, 170 nm, 180 nm, 190 nm, or 200 nm.

[0054] In some aspects, the plurality of particles have a mean zeta potential of from -2 mV to -50 mV. The plurality' of particles can have a mean zeta potential, for example, of from -2 mV to -45 mV, -2 mV to -40 mV, -2 mV to -30 mV, -2 mV to -20 mV, -2 mV to -10 mV, - 2 mV to -5 mV, -5 mV to -50 mV. -5 mV to -40 mV, -5 mV to -30 mV, -5 mV to -20 mV, -5 mV to -10 mV. -10 mV to -50 mV, -10 mV to -40 mV, -10 mV to -30 mV. -10 mV to -20 mV, -20 mV to -30 mV, -20 mV to -50 mV, -20 mV to -40 mV, -20 mV to -30 mV, -30 mV to -50 mV, -30 mV to -40 mV. or -40 mV to -50 mV. The plurality of particles can have a mean zeta potential, for example, of -2 mV, -5 mV, -10 mV, -15 mV, -20 mV, -25 mV, -30 mV, -40 mV, or -50 mV.

[0055] In some aspects, the plurality' of particles have a mean zeta potential of from -2 mV to +50 mV. The plurality of particles can have a mean zeta potential, for example, of from -2 mV to +45 mV, -2 mV to +40 mV, -2 mV to +30 mV, -2 mV to +20 mV, -2 mV to + 10 mV, -2 mV to +5 mV, +5 mV to +50 mV, +5 mV to +40 mV, +5 mV to +30 mV, +5 mV to +20 mV, +5 mV to +10 mV, +10 mV to +50 mV, +10 mV to +40 mV, +10 mV to +30 mV, +10 mV to +20 mV, +20 mV to +30 mV, +20 mV to +50 mV, +20 mV to +40 mV, +20 mV to +30 mV, +30 mV to +50 mV, +30 mV to +40 mV. or +40 mV to +50 mV. The plurality of particles can have a mean zeta potential, for example, of -2 mV, +5 mV, +10 mV, +15 mV, +20 mV, +25 mV, +30 mV, +40 mV, or +50 mV.

[0056] In some aspects, the plurality of particles have a mean particle Clausius-Mossotti factor of from -1 to 1. The Clausius-Mossottii factor of a particle characterizes the particle’s polarizability. This factor arises when dealing with the polarization of a particle embedded in a medium whose dielectric properties differ from that of the particle. The plurality of particles can have, for example, a mean particle Clausius-Mossotti factor of from -1 to 0.7, -1 to 0.5, -1 to 0.3, -1 to 0, -1 to -0.7, -1 to -0.5, -1 to -0.3, -0.7 to 1, -0.7 to 0.7, -0.7 to 0.5, -0.7 to -0.3, -0.7 to 0, -0.7 to -0.3, -0.7 to -0.5, -0.5 to 1, -0.5 to 0.5, -0.5 to 0, 0 to 1, 0 to 0.5, or 0.5 to 1. The plurality of particles can have, for example, a mean particle Clausius-Mossottii factor of -1, -0.7, -0.5, -0.3, 0, 0.3, 0.7, or 1.

[0057] The sample can comprise any fluid suitable for suspension of the plurality' of particles. Non-limiting examples of fluids which are suitable for the samples include water, cell-growth medium, saliva and other biological fluids, organic solvents, and emulsions. [0058] In some aspects, the sample comprises an electrolyte.

C. Sample Profiles

[0059] As discussed above, detecting the light scattered or emitted by the plurality of particles using the optical device generates a profile for the sample comprising the plurality' of particles. The profile is characterized by the extent of how the particles in the plurality of particles are driven toward or away from the one or more constrictions in the flow channel due to the applied non-uniform electric field. [0060] In some aspects, the sample profile generated by the methods described is characterized by trapping zones at the opposing bodies that converge to create a constriction within the flow channel. In such aspects, the trapping zones "trap’’ one or more subpopulations or particles in the plurality of particles.

[0061] In some aspects, the sample profile generated by the methods described is characterized by separating the plurality of particles into one or more subpopulations of particles at the opposing bodies that converge to create a constriction within the flow channel. The separation profile of the one or more subpopulations of particles may be controlled using the applied voltage. For example, the separation profile may be stationary' using direct current, where one or more subpopulations of particles are separated and specific fractions are captured at trapping zones. Additionally or alternatively, the separation profile may be transitory using a voltage sweep, a time-dependent change, or the motion of the particles as they transit through zones of electrokinetic forces.

[0062] Critical Quality Attributes (QCAs) are measurable chemical, biological, physical, material, microbiological, or their combination, properties that should be within an appropriate limit, range, or distribution to ensure desired product quality. Examples of QCAs which can be evaluated using the methods described are mean particle sizes, mean particle size distributions, mean particle zeta potentials, mean particle zeta potential distributions, and mean particle Clausius-Mossotti factors.

[0063] The deviation of the sample profile from a standard profile is determined. The standard profile can be obtained using the methods described from a standard sample having a known or target one or more mean particle size, mean particle size distribution, mean particle zeta potential, mean particle zeta potential distribution, or mean particle Clausius- Mossotti factor, distribution of dielectric permittivity and its mean value, or the combined effect of multiple properties on the zones of particle trapping and particle motion in the electrokinetic force field. In some aspects, the standard profile is a Critical Quality Attribute. [0064] The deviation of the sample profile from the standard profile can be determined based on changes in trapping zones, sizes of particle subpopulations trapped in trapping zones, intensity’ of scattered light, which is affected by the concentration and sizes of the trapped particles, a trajectories of particles traversing or approaching the constrictions. The deviation can be determined, for example, based on relative scattering intensity in trapping zones, the traj ectories of the particles, including the loci of their position as a function of time, or velocity of the motion within particular constrictions. The deviation of the sample profile from a standard profile is determined qualitatively or quantitatively. In some aspects. the deviation of the sample profile from a standard profile is determined qualitatively. For example, the number of particles in different trapping zones may be different from what is observed with the standard samples, as reflected by different intensities of the light scattered by particles trapped at one or more trapping zones inside the channel.

[0065] In some aspects, a sample profile is determined to not meet a QCA standard if there is no particle trapping in a trapping zone when the standard profile exhibits particle trapping in a corresponding trapping zone.

[0066] In some aspects, the deviation of the sample profile from a standard profile is determined quantitatively.

[0067] In some aspects, the sample is determined to meet a QCA standard if a relative intensity of scattered light in a subpopulation of particles within a trapping zone is within 1%, 2%, 5%, 10%, 15%, 20%, or 25% of a relative intensity of scattered light in a subpopulation of particles within a corresponding trapping zone in the standard profile. In some aspects, the sample is determined to meet a QCA standard if a mean velocity of a subpopulation of particles at a constriction is within 1%, 2%, 5%. 10%. 15%. 20%. or 25% of a mean velocity in a subpopulation of particles at a corresponding constriction in the standard profile.

[0068] In some aspects, the sample is determined to meet a QCA standard if a mean travel time of a subpopulation of particles from a first constriction to a second constriction is within 1%, 2%, 5%, 10%, 15%, 20%, or 25% of a mean travel time of a subpopulation of particles from a corresponding first constriction to a corresponding second constriction in the standard profile.

[0069] In some aspects, the sample is determined to meet a QCA standard if the sample profile deviates no more than 1%, 5%, 10%, 15%, or 20% from a standard profile.

EXAMPLES

[0070] The following examples further illustrate this disclosure. The scope of the disclosure and claims is not limited by the scope of the following examples.

[0071] Features and advantages of this disclosure are apparent from the detailed specification, and the claims cover all such features and advantages. Numerous variations will occur to those skilled in the art, and any variations equivalent to those described in this disclosure fall within the scope of this disclosure. Those skilled in the art will appreciate that the conception upon which this disclosure is based may be used as a basis for designing other compositions and methods for carry ing out the several purposes of this disclosure. As a result, the claims should not be considered as limited by the description or examples. Example 1: Computer-aided Design of DEP Device by Multiphysics Computer Simulations [0072] A dielectric particle in microfluidic channel experiences electrokinetic (FEK) forces containing dielectrophoretic (FDEP), electroosmotic (FEO), and electrophoretic (FEP) forces in electric field nonuniformly perturbed by dielectric insulators (gray triangles) forming a gate. As shown in FIG. 3, FEO + FEP acts toward the negative electrode, while FDEP has the opposite direction away from the highest field gradient (negative DEP force). In the region where FDEP exceeds that of FEO + FEP, defined by the condition that FEO =FEP, particles are entrained in a retention zone. Particles trapped in the retention zone have a combination of properties (e.g., size and zeta potential) leading to the balance of forces in the electric field profile at the gate. The curves in FIG. 3 defining retention boundaries w ere calculated using 2D Multiphysics simulations. Though the effects of particles on the electric field, particleparticle interactions, and particle-wall interactions were ignored in simulations, these results explain the formation of the retention pattern as a function of particle properties and are consistent with experiments.

[0073] FIG. 4 shows changes in DEP trapping for particles with different zeta potentials, sizes, and applied voltages, suspended in liquid with conductivity similar to the cell growth medium. In a single gate, particles with smaller zeta potential (-10 or -15 mV), may be trapped at both insulators of the gates (FIG. 4A) without forming an ark connecting the insulators. Particles with -20 mV zeta potential have a larger continuous trapping pattern bridging two insulators. If multiple trapping zones are possible, thermal fluctuations may assist particles' escape to a new downstream zone (direction of the fluid flow in FIG. 4D is from the left to the right). Transient evolution of trapping w as found to occur as the particle sizes change while the applied voltage is constant. In this case, smaller particles are only- trapped close to the insulators forming the constriction, where the electrical filed gradients are the largest in the absolute value, while larger particles (FIG. 4B) are retained in the ark connecting the constriction (FIG. 4C). The results show a slight change in size (<50nm) or zeta potential (< 5mV) produces significant changes in trapping patterns, demonstrating the high sensitivity of the proposed approach to variations in samples.

[0074] FIG. 4D shows retention patterns inside a three-gate EK microfluidic device when the distance between the dielectric teeth decreases from 2 to 1.2 and 0.6 pm in the direction of the flow. Only large particles were retained at the gate with the smallest FDEP (left gate), while the right gate with the highest FDEP retained all particles. Therefore, the sample introduced in the left port of the device will be fractionated by size and charge. The distinct retention pattern observed in FIG. 4D depends on the properties on the particles in the sample.

[0075] The fractionation pattern grows in specificity with the number of gates for heterogeneous samples. Multiple gates with decreasing spacing and increasing FDEP will fractionate heterogeneous nanoparticle samples with finer discrimination, create more complex retention patterns (fingerprints), and allow us to distinguish samples with finer variations in size and charge distributions.

[0076] FIG. 5 conceptually illustrates deviations in EK fingerprints from therapeutic EVs' nominal (expected) fingerprint, signaling an "off-spec" product and the need to adjust biomanufacturing conditions, terminate the batch, or discard the obtained product. Vertical dashed lines indicate the nominal mean values of distributions particles' size and zeta potential. The second row gives electrokinetic fingerprinting for a sample, Off spec 1.

[0077] Sample Off spec 1 contains EVs with smaller than expected sizes and zeta potential. Such deviations shrink the trapping zones and break the continuous retention pattern at zones 1 and 2 into disconnected regions. EV sizes and zeta potentials in the Off spec 2 sample exceed the nominal, producing different fingerprints from the On spec sample that shows a larger number of particles are captured at all DEP gates. Off spec 3 pattern occurs when some EVs have sizes larger than present in the nominal sample and the subpopulation of particles with the zeta potential smaller than expected is also present. In all patterns, the concentration of particles retained at different gates is revealed by the scattering intensity.

[0078] The simulations illustrate mapping sample properties to the retention pattern of the iDEP device. They revealed the design variables we could adjust to increase the specificity (if not uniqueness) in mapping heterogeneous EV samples. The analysis may be extended by including additional variables, such as particles' concentration, applied voltage, the geometry of g-iDEP gates and their number, and the evolution of the pattern due to thermal fluctuations. The effects of the superimposed flow pattern created when the sample is flowed through the EK device may include cases of constant, time-varying, and oscillating pressure drops driving the flow. Combining electrokinetically and pressure drop-driven flows will modify the liquid flow distribution and, thus, the EV retention pattern. Insulator dielectrophoresis created by applying DC voltage may be supplemented with the electrode dielectrophoresis created by one, two, or more electrodes excited by applying frequencydependent AC volatge. By sweeping the frequency of AC excitation, the influence of particles' complex dielectric permittivity spectra on the pattern of particle retention and motion in the EK force field is revealed.

[0079] The effect of particle non-sphericity. elasticity, particle-particle, particle-wall interactions, Coulombic heating of the fluid, and deviations in gate geometries, may need to be considered during the computer-aided design of EK devices and comparing samples with the standard.

Example 2: Fabrication of EK Device

[0080] The initial design comprises the EK chip and its holder. The EK chip (design in FIG. 6A) was 3D nano-printed with high spatial resolution using 3D nano-printing by Nanoscribe 3D Photonic Professional GT2 printer, which has 400 nm lateral resolution, sufficient for the desired gate geometry. The vertical resolution is an adequate 1 pm. A non- cytotoxic photocurable resin with a very low autofluorescence and high transparency was used in printing.

[0081] The small size of the EK chip (~1 mm in the longest dimension) complicates its handling, predictable positioning in the laser beam, and reading the retention pattern created by particles' light scattering. A holder was designed (FIG. 6B) for positioning consistency, which accepts the chip in a predetermined location and orientation. The holder is 3D printed at a lower resolution than the chip. It contains microfluidic channels to direct the sample into the EK chip and interfaces with the electrodes applying an electrical field to impose the EK separation. The holder can also be made from Polydimethylsiloxane (PDMS) using soft lithography.

Example 3: Integration of EK Device with NTA-like Readout of CQA Fingerprint [0082] The retention patterns in the prototype EK device are interrogated by light scattering. The initial results of this approach were obtained by adapting an existing nanoparticle tracking analysis (NT A) instrument (Malvern NanoSight, Model LM10) to the imaging of the retention pattern. The holder and the EK chip on the LM10 stage were illuminated by a cylindrical laser beam (2 = 405 nm). A large diameter of the beam (~50 pm) relative to the width of illuminated EK gates leads to scattering artifacts, which overwhelm the scattering by nanoparticles retained at the gates.

[0083] EK retention patterns may be interrogated by adjusting the width of the laser beam to conform to the gaps between dielectric structures and electrodes shaping electrical gradients and electrokinetic forces.

[0084] The described EK+NTA configuration may be used in the fluorescent NTA mode to block the scattering interferences and reveal compositional target molecule-fluorophore dependence of the retention patterns and particle motion. Only fluorescent light emitted bylabelled test particles will be visualized in this arrangement, allowing for the testing and the optimization of EK designs without the custom optical platform with optimized laser widths and positioning needed for the label-free readout. Synthetic (latex) nanoparticles ranging from 30 to 200 nm diameters or fluorescently labeled biotherapeutic nanoparticles may be used for this purpose. The particle size distribution (d_p) and zeta potential ty_p) may be characterized by NTA and DLS (Malvern Zetasizer). The experimental testing of the DEP+NTA prototype may start with monodispersed fluorescent beads with independently validated properties. After successfully validating distinct capture patterns with monodispersed samples, mixtures of particles of different d_p

may be used to confirm unique capture patterns and their shift with changes in the particle properties, as illustrated in simulations (FIG. 4 and FIG. 5). The pattern's sensitivity⁷ to changes in d_p

may be determined in spike-in experiments to establish limits on samples distinguishable from the nominal, as illustrated in FIG. 5. In label-free implementation, the described fluorescent NTA visualization of the retention pattern must be repeated using label-free scattered light readout. Collectively, the experiments with synthetic or biological particles inform the design of the EK device and its operating conditions for robust differentiation of nanoparticle populations with high sensitivity, shown to be feasible in simulations (FIG. 5). The effect of particle concentration on the brightness of the retention pattern may be used to differentiate concentrations of the nominal sample with samples compared for compliance.

[0085] The correlation between the retention patterns and properties of nanoparticles maybe established experimentally. For example, the properties of biological nanoparticles (e.g., non-therapeutic EVs isolated from serum), including their size distribution, surface charges, and molecular composition, may be first characterized by conventional methods such as NTA, electron microscopy, BCA protein assay, mass spectroscopy, and RNA-seq analysis. Additional characterization examples are given below in the context of reno-therapeutic EVs. The CQA signature of same samples (particle retention pattern at EK gates) may then be obtained to establish the corresponding retention patters for different designs of the EK device and operating conditions, such as applied electrical potentials.

[0086] The described workflow may be used to validate the design principles established with synthetic or appropriately selected biological nanoparticles and requirements on the EK device to differentiate signatures of slightly different samples, such as EVs isolated from the same biofluid (e.g., serum) sample by different techniques. The described workflow applied to serum EVs would start in a florescent NTA mode after fluorescently labeling EVs (e.g., using labeled antibodies for CD63 and CD81 tetraspanins).

[0087] The EK design can also be modified to have a longer channel to allow time for bubble dissolution. Alternatively, bubble traps will be incorporated into the design. The sample can be fractionated into the size and/or density fraction to simplify the problem and demonstrate the fingerprinting for highly heterogenous samples after fractionating them into subpopulations with reduced heterogeneity.

Example 4: Testing EK Signatures on Therapeutic EVs

[0088] The proposed electrokinetic CQA may be used to assess therapeutic EV samples. The first type of EVs secreted by human amniotic fluid (AF) derived stem cells (AFSC) are preventative and therapeutic against CKD (chronic kidney disease) and AKI (acute kidney injury) from various etiologies. Therapeutic EVs may be isolated from the growth medium of bone marrow hMSCs (human mesenchymal stem cells) grown inside a bioreactor, such as a hollow fiber bioreactor (Quantum™, Terumo BCT). MSC-EVs are the most studied and widely used class, known to be broadly therapeutic owing to their immunomodulatory effects. The second type of EVs may be selected to have no therapeutic benefits (such as EVs isolated from a serum of a diabetic patient).

[0089] The goal in using these two EV types is to find how similar their CQAs are compared to non- therapeutic serum EVs.

Characterization of therapeutic EVs

[0090] Reno-therapeutic EVs is a class of biomanufactured EVs that may be secreted by, for example, clonal human amniotic fluid stem cells. They can be characterized by laboratory techniques to quantify⁷ their size distribution, concentration, surface charges, vesicle-to- protein content by the NTA and BCA protein assay, expression of EV-canonical biomarkers (e.g.. CD63, CD81, ALIX. FLOT1, ICAM1, EpCam, ANXA5, and TSG101), stem-cell specific, and reno-therapeutic biomarkers, and microRNA cargo specific to renal diseases, and involved in pathways related to mTOR, TGFfy VEGF, Hippo, focal adhesions, ERB signaling, MAPK, Smad binding, and cell junctions. The morphological structure of vesicles may be determined by electron microscopy. Negative controls (cytoplasmic and extracellular) may be used to assess sample contamination. A similarly detailed characterization of MSC- EVs could be performed.

Sensitivity of EK + NTA CQA Sisnature to Non-therapeutic Content

[0091] A small fraction of non-therapeutic (serum) EVs may be sequentially added to therapeutic samples until changes in the retention pattern of the electrokinetic CQA assay reveal their presence. This titration may be used to quantify the CQA assay's sensitivity to "off-spec" biologies by a given design of an EK device. Several devices may be tested to elucidate the design- and operating-conditions dependence of off-spec product detection. The impact of impurities co-isolated with EVs from the growth medium (e.g., solution proteins) on the EK signature may be determined for different samples (e.g., as received vs. dialyzed EV samples) and quantified for its impact on the ability to distinguish between EV samples with different non-therapeutic EVs titers.

[0092] The acquired experimental data may be used to develop a "mixture model" describing the impact of mixing samples with know n EK signatures at different ratios on the mixture's fingerprint. Simulations predict linear mixing. However, limits of the EK pattern imaging, dictated by the optics, digital camera, and strong particle-size dependence of scattering intensity, will likely lead to a nonlinear pattern mixing and stronger sensitivity to changes in larger-size nanoparticles.

[0093] Results of carefully controlled titration experiments, first, with latex beads, may be used to produce EK patterns in response to known sample changes and statistically quantify (e.g., with weak and strong confidence: p<0. 1 and <0.01) the detectible changes in EV size distribution, their surface changes, variations in concentration and particle permittivity, and simultaneous changes in multiple properties. The obtained detection thresholding to determine when a simple is different from the standard will likely be nonlinear to the listed properties impacting the EK fingerprinting.

Example 5: Exosome Trapping Based on Dielectrophoresis

[0094] Manipulation of the nanoparticle or biomolecules is possible because of the influence of the electrokinetic (EK) typically comprising of the dielectrophoretic (DEP), electroosmotic (EO) and electrophoretic (EP) forces. The electroosmosis phenomenon is quite common in microfluidics as surfaces of microfluidic channels are usually charged. The surface charge induces an accumulation of counter-ions and a depletion of co-ions inside the channel. As a consequence, when the voltage is applied across the microchannel, the unbalance between positive and negative charges inside the channel results in a net force on the solvent so strong as to move the fluid, which then creates entraining force on the suspended particles. Electrophoresis is widely utilized in applications to create a driving force to translocate charged molecules and particles. EO can be used as a driving force to move analytes through membrane nanopores and nanochannels, for example, during DNA sequencing inside a nanopore, and to characterize proteins, peptides, and nanoparticles. A voltage applied by electrodes inside a microfluidic channel creates an electrical field. E. Charged particles are driven toward or away from electrodes according to the sign of their charge. Dieelectric structures modify this field in both straight and the direction of forces it creates. Dielectrophoretic force is exerted on polanzable particle inside nonuniform electrical field. Polarizable particles carrying an intrinsic dipole orient themselves along the field lines and move in the direction of increasing |E|. The propensity to polarization may be characterized by complex dialectic permittivity, expressing the dielectric loss and storage as a complex number that changes with frequency. The electrodes creating the electric filed may be placed in inlet and outlet reservoirs and the changes in the imposed electrical field may be imposed by insulating posts, or boundary features, such as triangular constructions (or gates). Addition electrodes may also be used to modify the EK force field.

COMSOL Mathematical Modelins

[0095] A two dimensional model of a microchannel containing triangular insulating features creating flow constrictions and modifying the EK force field was considered in the configuration shown in FIG. 7A. The distribution of the electric potential was solved using the following Laplace equation in AC/DC module of COMSOL Multiphysics 5.5 (COMSOL Inc., MA, USA):

V²cp = 0

[0096] The surface of the microchannel with insulating constrictions are defined as boundaries and Neumann boundary conditions (n • J = 0) were applied, where n and J are the normal vector from the surface and the electrical current densify, respectively. The DC voltage, applied at the inlet (V_m = variable) and outlet ( _out = 0) of the microchannel, create the electric field (external excitation). The triangular insulating posts have 60 degrees of interior angle and distance between the triangular insulating pillars is denoted by AH (FIG. 7B). This width is adjusted and optimized according to the size of the particle to be trapped. The insulating constrictions create nonuniform electric field shown in FIG. 7B. High electric field gradients occur at the sharp tip of triangular insulating posts. Particles are considered to be trapped when the DEP force counteracts the particles from moving downstream with the fluid under the effect of EO and EP forces. In certain instances (the case of nonlinear electrophoresis), determined by the properties of the fluid, electrical excitation and particle properties, the balance of all electrokinetic forces may occur when the DEP force is relatively small and the particle trapping is primarily determined by the balance EO and EP forces.

[0097] FIG. 7C shows the net particle velocity as an example of trapping that occurs in negative DEP (-0.5 of Re{K}). The particle net velocity along the center cut-line of the insulating posts is shown in FIG. 7B. A negative sign on net particle velocity indicates that the particles move back to the channel inlet. This shows that the particles are trapped in the left side of the insulating posts between the positions where the net velocity under the combined influence of DEP, EO, and EP forces is zero.

Effect of Applied Volta e

[0098] Trapping occurs in the closed regions defined by locations where all EK forces are balanced yielding net particle velocity equal to zero. EK forces with an applied voltage of lOOEto 800 V at the point in trapping region shown in FIG. 8 A were calculated, and the results are shown in FIG. 8B. EO+EP forces increases linearly with the applied voltage and DEP force changes nonlinearly. FIG. 9 presents the modeling predictions for the effect of applied voltage. The streamline in FIG. 9 represents the particle velocity and the grey region is trapping region. Below 100E there was no trapping because the DEP force is insufficient to balance other EK forces (FIG. 9A). After that, when the applied voltage is increased, trapping appears and particles are fully trapped at 800 V (FIG. 9D). As a result, the applied voltage increases, the electric field increases, so both the DEP force and the EP+EO forces increase. In low applied voltage, the DEP force is insufficient for a balance and no trapping occurs. However, since the DEP force increases more rapidly (nonlinearly) it can balance the other EK forces when a sufficient high DC voltage is applied, leading to nanoparticle trapping.

Effect of Particle Zeta Potential

[0099] The zeta potential of the particle affects the EP velocity. The EP velocity and the electroosmotic (EO) velocity may be in the same or opposite directions depending on the particle change and the presence of nonlinear electrophoresis, occurring when EP force nonlinearly depends on the strength of the electric field due to the distortion of an ionic cloud surrounding a particle with the charges surface. Considering the balance between all EK forces, as the particle zeta potential increases, we established the conditions for trapping exosomes with a zeta potential in the range between -5/wEto ~20mV and the negatively changed walls of the microfluidic channel. FIG. 10 shows under these conditions that, as the zeta potential increases, the trapping region is increased, and at -20 mV, exosomes are trapped in a large area connecting triangular insulators.

[00100] Previously, it was calculated that exosomes have a (Re{K}) of -0.5. Since changes in the Calusisus-Mossotti (CM) factor may lead to negative or positive DEP force (aways or towards the largest gradients in the electric field, respectively), its impact on nanoparticle trapping was analyzed. FIG. 11 shows changes in the trapping region for different values of the CM factor, obtained assuming the suspending medium has the conductivity o_m = 15mS/cm. The streamline in the FIG. 11 represents the particle velocity and grey region is trapping region. The Re{K} affects the DEP force and changes the size and position of the trapping region. The trapping region expanded when the magnitude of Re{K} is increased under considered conditions. To understand changes in the location of the trapping zone, note the electric field in FIG. 7. In negative DEP (negative Re K}). the DEP force is repelled from the regions of the high gradient in the electric field and in the region to the left side of the insulating post (upstream), is the location where the DEP force balances the other EK forces and the trapping occurs. Conversely, in positive DEP (positive Re{K} the DEP force is towards peak gradients in the electric field and the trapping region occurs downstream (to the right) of insulating constriction.

Effect of Particle Size

[00101] The EK forces depend on the particle size. To retain a particle inside the trapping the trapping zone, all forces must balance. When the DEP force counteracts other EK forces, it requires larger applied voltage to trap smaller particles. FIG. 12 shows changes in the trapping region and streamlines for particles of different sizes. The results were obtained for the negative DEP (Re{K}=-0.5) and a particle zeta potential equal to -15 mV. The results were obtained for sizes typical for exosomes and other extracellular vesicles and are shown for the particle diameters between 30 nm and 200 nm. The results indicate enlargement of trapping regions as particle size increased. They further indicate the fractionation of exosomes by sizes at the gates with different widths.

Particle Separation Based on Particle Size

[00102] We analyzed the particle separation inside an EK device with three gates formed by triangular isolating features modifying the EK forces inside a microfluidic channel (FIG. 13). The gate widths, AH (the distance between triangular features), were 0.6, 1.2, and 2 pm. The sample containing particle 100 nm, 150 nm, and 200nm in diameters were considered. The largest 200 nm nanoparticles were only trapped at the first gate with AH=2 pm. All smaller particles are not retained at this gate. The second gate (AH=1.2 pm) traps 150 nm nanoparticles and any 200 nm particles that escaped the trapping at the first gate (such escape could be caused by thermal fluctuations, for example). The smallest 100 nm nanoparticles are not trapped at gates one and tw o and can only be retained at the third gate with the smallest width (AH-0.6 pm), which would also retain all larger particles that escapped the rapping at the preceding gates (FIG. 13). Therefore, the EK device fractionates the particles based on their sizes and creates a distinct trapping pattern that reflects the distribution of sizes in the sample. Any heterogeneity’ in the nanoparticle population other than their sizes, including zeta potential and dielectric permittivity, will also influence the trapping patterns and the motion of particles in a microfluidic channel.

Example 6: The Impact of Non-linear Electrophoresis

[00103] COMSOL simulation experiments were conducted to assess the effect of nonlinear electrophoresis.

[00104] When the applied field exceeds ~V_Tfa (where a is the particle size and V_T is the thermal voltage V_T = k_BT /e), differences in electrophoretic mobility' between particles and ions cannot be perfectly compensated by the (diffusive) replenishment of ions from the bulk solution into the ion cloud of the particles, resulting in deformed ion clouds, which induce a concentration polarization on the diffuse layer that sharply deviates from electroneutrality.

[00105] This deformation changes the electrophoretic mobility, adding a correction term to the electrophoretic velocity' due to Non-Linear Electrophoresis (NLEP) effects. NLEP is driven by surface conductance resulting from the deformation of the ionic cloud surrounding particles. Consequently, NLEP depends on the zeta potential of the particles, their surface area, and the ionic strength of the solution.

[00106] For Ea/V_T « 1, linear retardation occurs as the deformed ion cloud, skewing opposite to the EP velocity, and pulling the particle in that direction.

[00107] For Ea!\I_T ~O(1). an E³ speed-up occurs as the ion cloud separates from the particle, leading to a larger effective surface charge density.

[00108] For Ea/V_T » 1, an E^{3 2} speed-up occurs as ions lagging behind the particle are found even separated from the particle.

[00109] Analytically, this means the electrophoretic velocity should be corrected by the addition of a nonlinear term

[00110] Here, Du — is the Dukhin number (ratio of surface conductivity to bulk

conductivity, with ._D the Debye length) and

[00111] As before, trapping would take place when the net velocity of the particles due to EP. EO, and DEP forces is zero,

^UNET ^{= U}EP + ^UDEP + ^UEO ⁼ 0-

[00112] Note that for the trapping to occur, the streamlines associated with the net velocity of particles must converge to a specific region (trapping zone). This condition can be assessed by the connectedness of u_NET ■ E = 0. Note that in all cases, including linear electrophoresis, some of the boundaries of the trapping region may be formed by impermeable walls of the microfluidic channel.

[00113] Analy zing the trapping voltage for various particle charges, sizes, and bulk ionic strength, it is observed that NLEP tends to dominate over DEP at lower ionic strengths, smaller particle sizes, and higher zeta potential, ^_p.

[00114] Lower trapping voltages are sufficient for lower ionic strengths of the solution and higher zeta potentials of the particles; these are the conditions when NLEP dominates DEP. As shown in FIG. 14A and FIG. 14B, which shows the trapping regions at three different gates of a microfluidic device (FIG. 14A), selective trapping can occur at 75 V at consecutive widths of the gates, decreasing from 2 pm to 1.5 pm and 0.6 pm.

Claims

CLAIMS What is claimed is:

1. A method of analyzing a sample comprising a plurality of particles, comprising: a) introducing the sample into a system comprising a microfluidic electrokinetic device which comprises:

(i) at least one flow channel comprising a fluid inlet and a fluid outlet, wherein the flow channel extends from the fluid inlet to the fluid outlet along a first axis, wherein the flow channel comprises a first wall and an opposing second wall, wherein the first and second walls have opposing bodies that converge to create a constriction within the flow channel, the opposing bodies comprising an electrical insulator;

(ii) a power supply and a plurality of electrodes in electrical communication with the power supply, wherein the power supply is configured to apply a voltage between at least two electrodes of the plurality of electrodes to provide a non-uniform electric field across the flow channel;

(iii) a light source having an output beam path configured to irradiate the plurality of particles in the flow channel; and

(iv) an optical device comprising a photon detector configured to detect light scattered or emitted by the plurality of particles; b) applying a voltage to the electrodes via the power supply to provide the non- uniform electric field, which exerts an electrokinetic (EK) force on the plurality’ of particles within the flow channel; c) irradiating the plurality of particles with the output beam path, wherein the output beam path is irradiated along a second axis; d) detecting the light scattered or emitted by the plurality of particles using the optical device, wherein detection generates a profile for the sample comprising the plurality⁷ of particles; and e) determining the deviation of the profile for the sample comprising the plurality of particles from a standard profile, wherein the standard profile is characteristic of one or more of a particle size distribution, a mean particle size, a zeta potential distribution, a mean zeta potential, a Clausius-Mossotti factor distribution, a dielectric permittivity, or a mean particle Clausius-Mossotti factor.

2. The method of claim 1, wherein the flow channel has a length along the first axis from 1 mm to 20 mm.

3. The method of claim 1, wherein the applied voltage is from 5 V to 1000 V.

4. The method of claim 1, wherein the constriction is between 0.5 pm to 10 pm.

5. The method of claim 4, wherein the flow channel comprises multiple constrictions that serially narrow along the first axis.

6. The method of claim 5, wherein the constrictions serially narrow by 0. 1 pm to 5 pm.

7. The method of claim 1, wherein the microfluidic electrokinetic device comprises multiple flow channels arranged in a parallel configuration.

8. The method of claim 1, wherein the applied voltage is applied via the power supplyusing direct current, alternating current, or a combination thereof.

9. The method of claim 1, wherein the light source is a laser having a wavelength of from 300 to 500 nm.

10. The method of claim 1, wherein the photon detector includes a camera which is a charge-coupled device (CCD) detector or a complementary metal-oxide-semiconductor (CMOS) detector.

11. The method of claim 8, wherein a different voltage is applied across each flow channel.

12. The method of claim 1, wherein the flow channel comprises multiple constrictions formed by the first and second walls converging within the flow channel.

13. The method of claim 10, wherein the camera detects light scattered by the plurality of particles

14. The claim of claim 10, wherein the camera comprises at least one fluorescent filter and detects light emitted from the plurality’ of particles.

15. The method of claim 7, wherein each flow channel comprises one constriction formed by the first and second walls converging within the flow' channel.