Method for determining empty, partially filled and full recombinant AAV particles

Abstract

Herein is reported a method for determining the amount or / and ratio of filled recombinant virus particles, such as recombinant adeno-associated virus particles (rAAVps), compared to empty and partially filled viral particles in a sample using transmission electron microscopy (TEM) with an EM grid without chemical staining, wherein the method comprises the step of drying the sample, preferably air-drying the sample, on the EM grid.

Description

[0001] Method for determining empty, partially filled and full recombinant AAV particles

[0002] The current invention is in the field of analytical methods. More precisely, herein is reported a method for determining the fraction of empty, partially filled and full recombinant adeno- associated viral particles using transmission electron microscopy (TEM).

[0003] Background of the Invention

[0004] Gene therapy is opening unprecedented opportunities for novel therapeutic approaches. Based on the concept of rescuing function mutations by providing a therapeutic gene with the correct sequence, to allow biological functions to be restored, it requires the use of specific delivery to ensure the proper delivery of therapeutic genes to the intended site of action. In this context, recombinant adeno-associated virus particles (rAAVps) are most widely used for therapeutic gene delivery.

[0005] The manufacturing process of therapeutic rAAVps requires the insertion of the therapeutic transgene into the recombinant AAV capsid shell (full viral particles, i.e. recombinant viral particles comprising an encapsidated nucleic acid).

[0006] However, a certain percentage of the recombinantly produced viral particles that do not contain the desired transgene might also be produced, as well as partly filled recombinant viral particles.

[0007] In view of the requirements of application of a correct dose in order to exert the intended therapeutic effect, the correct determination of at least the fraction of full rAAVps is needed.

[0008] Currently different techniques are used for determining the loading state of rAAVp preparations, such as analytical ultracentrifugation (AUC), mass photometry, determination of vector genome (vg) titer by qPCR or ddPCR and rAAVp capsid (cp) titer by ELISA or HPLC, as well as electron microscopy.

[0009] With electron microscopy (EM), a direct visualization of the individual particle of a mixture is possible. It is a calibration-free, reference-free and model-free method.

[0010] Two TEM methods are currently used: negative staining and the current gold standard CryoEM. While CryoEM has been generally accepted in the field, it is elaborate, time consuming and expensive and not suitable let alone applicable as a routine qualification method on production sites or in a clinical environment. Negative staining would, from a practical perspective, be easier, but has certain drawbacks, such as, e.g., radioactivity, as well as being is prone to misinterpretation as different serotypes react differently to the staining dye, which is especially problematic when newly generated capsid variants have to be analyzed. US 2023 / 0303983 reported a high efficiency purification of divergent AAV serotypes using AAVX affinity chromatography. In this patent application, negative staining transmission electron microscopy (TEM) has been used to analyze the percentage of empty particles in purified samples.

[0011] Sommer, J.M., et al. reported about the quantification of adeno-associated virus particles and empty capsids by optical density measurement. In more detail, Sommer et al. used as a verification a contrasting agent to preserve the sample and to selectively permeate the empty particles, thus performing negative staining electron microscopy (EM).

[0012] Pierson, E.E., et al. reported the resolving of adeno-associated viral particle diversity with charge detection mass spectrometry. Results were visualized using negative staining EM.

[0013] US 2011 / 0014472 reported the preparation of silica stabilized biological templates for the production of metal and layered nano particles. Here the biological sample was preserved and coated.

[0014] Franken, L.E., et al. reported about transmission electron microscopy as a tool for the characterization of soft materials (Adv. Sci. (Weinh.) 31 (2017) 1600476). Franken et al. concluded that in summary, drying of solvated particles is often unjustified and the misinterpretation of dried materials is a persisting problem.

[0015] CN112183433 and US2021 / 0341372 reported a method for characterization and quantification of solid and hollow virus particles based on CryoEM. Both reported a method of quantitative measurement of particle content using hydrated state imaging. In more detail, the method was for quantitative measurement of particle content using hydrated state imaging such as CryoTEM. A sample of virus-like particles (VLPs) or virus particles was used as sample which was rapidly frozen into a cryogenic liquid at a cryogenic temperature. While at the cryogenic temperature, the particle content of each VLP in the frozen sample was observed in the CryoTEM.

[0016] It has to be pointed out that both US 2021 / 0341372 and CN112183433 specifically emphasizes the need to preserve the native, hydrated state of the particle for analysis, i.e. this preservation of the hydrated state is a core part of the disclosure. It is further emphasized that drying artifacts like collapsing of particles due to varied success in preservation as well as uncontrolled variations in stain thickness are detrimental to the accuracy of the TEM analysis for E / F determination as it is the common belief in the art. This can be seen by the fact that in US 2021 / 0341372 and CN112183433 no drying is performed but only a removal of surplus liquid using an adsorbing means, followed by rapidly freezing to preserve the hydrated sample in a thin layer of solvent.

[0017] Ausman et al. (J. Pharm. Sci. 114 (2025) 1554-1562) reviewed low voltage electron microscopy as an emerging tool for AAV characterization. Summary of the Invention

[0018] Herein is reported a robust, easy and rapid method for the determination of empty, partially-filled and full rAAVps that is applicable to all natural AAV serotypes as well as artificial capsids obtained by capsid engineering.

[0019] The current invention is based, at least in part, on the unexpected finding that rAAVps can be analyzed with respect to their loading state, i.e. whether the particles are full, partially-filled or empty, by using electron microscopy, whereby the sample has been prepared by drying, such as drying in air.

[0020] This finding is indeed contrary to the common belief in the field, i.e. that biological materials suffer from drying artifacts and are unsuited to be analyzed without some form of preservation as, e.g. that done by negative staining (preservation in heavy metal salts) or CryoTEM (preservation of hydrated state by freezing). It has further to be emphasized that drying artifacts like collapsing of particles due to varied success in preservation as well as uncontrolled variations in stain thickness are detrimental to the accuracy of the TEM analysis for E / F determination.

[0021] The current invention is based, at least in part, on the unexpected finding that individual rAAVps can be recognized and characterized with respect to the loading state after being dried without staining or other forms of preserving the sample.

[0022] The current invention is based, at least in part, on the unexpected finding that the strength of electron scattering of a dried rAAVp is directly related to the loading / filling state of the particle.

[0023] One aspect of the invention is a method for determining the amount of recombinant viral particles in a sample using electron microscopy (EM) with an EM grid without chemical staining, comprising the steps of

[0024] - drying the sample on the EM grid, and

[0025] - determining and optionally classifying recombinant viral particles on the EM grid using EM and thereby determining the amount of recombinant viral particles in the sample

[0026] In certain embodiments of all aspects and embodiments, the grid is inserted into the EM and a vacuum is applied directly after the drying step. In certain embodiments of all aspects and embodiments, the drying is in air outside the EM and in vacuum inside the EM.

[0027] In certain embodiments of all aspects and embodiments, the drying is in a non-oxygen containing gas outside the EM and in vacuum inside the EM.

[0028] In certain embodiments of all aspects and embodiments, the drying is freeze-drying outside the EM and in vacuum inside the EM.

[0029] In certain embodiments of all aspects and embodiments, the drying is at room temperature.

[0030] In certain embodiments of all aspects and embodiments, the drying is at a temperature in the range between and including 20 °C to 70 °C, preferably in the range between and including 25 °C and 50 °C.

[0031] In certain embodiments of all aspects and embodiments, the EM grid comprises a support film that allows electron transmission, preferably a carbon or silicone coating.

[0032] In certain embodiments of all aspects and embodiments, the support film is dried for at least 2 days after the application to the EM grid.

[0033] In certain embodiments of all aspects and embodiments, the EM grid is a glow-discharged carbon coated EM grid.

[0034] In certain embodiments of all aspects and embodiments, the EM coating is with a glow-discharged, continuous layer of carbon, preferably without holes.

[0035] In certain embodiments of all aspects and embodiments, the carbon coating on the EM grid is as flat as possible and as thin as possible.

[0036] In certain embodiments of all aspects and embodiments, the thickness of the carbon coating on the EM grid is in the range of and including about 1 pm to about 0.1 nm.

[0037] In certain embodiments of all aspects and embodiments, the carbon coating is about 2 nm thick.

[0038] In certain embodiments of all aspects and embodiments, the method comprises prior to the drying step the step of

[0039] - applying the sample comprising the recombinant viral particles to the EM grid.

[0040] In certain embodiments of all aspects and embodiments, the sample is applied to the EM grid for about 15 to 240 seconds. In certain embodiments of all aspects and embodiments, the sample is applied to the EM grid for about 20 to 120 seconds. In certain embodiments of all aspects and embodiments, the sample is applied to the EM grid for about 25 to 65 seconds.

[0041] In certain embodiments of all aspects and embodiments, the method comprises after the applying step and before the drying step the step of

[0042] - washing the EM grid at least once with H2O.

[0043] In certain embodiments of all aspects and embodiments, the method comprises after the applying step and before the drying step the step of

[0044] - washing the EM grid three times with H2O.

[0045] In certain embodiments of all aspects and embodiments, wherein the viral particle is a recombinant adenoviral particle or a recombinant herpes viral particle or a recombinant adeno-associated viral particle (rAAVp).

[0046] In one preferred embodiment of all aspects and embodiments, the drying is a drying in air.

[0047] In certain embodiments of all aspects and embodiments, the drying is for about 30 seconds to about 90 seconds.

[0048] Thus, one aspect of the current invention is a method for determining the amount of recombinant adeno-associated virus particles (rAAVps) in a sample using electron microscopy (EM) with an EM grid without chemical staining, comprising the steps of

[0049] - drying the sample on the EM grid in air for 30 seconds to 90 seconds, and

[0050] - determining rAAVps on the EM grid using EM and thereby determining the amount of rAAVps in the sample.

[0051] In certain embodiments of all aspects and embodiments, the drying is for about 45 seconds to about 75 seconds. In certain embodiments of all aspects and embodiments, the drying is for about 60 seconds.

[0052] In certain embodiments of all aspects and embodiments, the method is for determining the amount of full rAAVps in the sample. In certain embodiments of all aspects and embodiments, the method is for determining the amount of empty rAAVps in the sample.

[0053] In certain embodiments of all aspects and embodiments, the method is for determining the amount of full and empty rAAVps in the sample.

[0054] In certain embodiments of all aspects and embodiments, the method is for determining the amount of partially filled rAAVps in the sample.

[0055] In certain embodiments of all aspects and embodiments, the method is for determining the amount of full, partially filled and empty rAAVps in the sample.

[0056] In certain embodiments of all aspects and embodiments, the sample is a rAAVp preparation or composition.

[0057] In certain embodiments of all aspects and embodiments, the method is for quality control of rAAVp preparations.

[0058] In certain embodiments of all aspects and embodiments, the method is for batch release.

[0059] In certain embodiments of all aspects and embodiments, the viral particle is a rAAVp and is of the serotype 2, 5 or 8 or an engineered variant thereof.

[0060] In certain embodiments of all aspects and embodiments, at least the last washing step or all washing steps are blotting steps.

[0061] In certain embodiments of all aspects and embodiments, the method comprises as determining step the steps of

[0062] - taking images of the EM grid with an EM,

[0063] - determining the recombinant viral particles in the images, and thereby determining the amount of recombinant viral particles.

[0064] In certain embodiments of all aspects and embodiments, the images are recorded manually or automatically. In certain embodiments of all aspects and embodiments, the images are recorded manually or automatically at precise defocus. In certain embodiments of all aspects and embodiments, depending on the scattering strength of the background and the used magnification for taking the EM image the focus has to be / is selected in order for the image quality to be sufficient and to allow to distinguish the different filling states of the rAAVps.

[0065] In certain embodiments of all aspects and embodiments, the images are recorded manually or automatically at precise defocus which takes into account the background of the grid to optimize contrast in the images to allow analysis.

[0066] In certain embodiments of all aspects and embodiments, the images are analyzed manually or automatically. In certain embodiments of all aspects and embodiments, the images are analyzed manually or automatically using existing or dedicated software or IA trained algorithms. In certain embodiments of all aspects and embodiments, the images are analyzed manually or automatically using existing or dedicated software or IA trained algorithms, by classification of particles or by sorting of particles based on their core intensity, edges, and / or image features.

[0067] In certain embodiments of all aspects and embodiments, the images are analyzed by taking the density of the core of the particle excluding focus fringes to plot these average densities into a histogram and split the data into empty and full particles based on peaks in this histogram.

[0068] In certain embodiments of all aspects and embodiments, the sample is imaged using low-dose mode in the electron microscope.

[0069] In certain embodiments of all aspects and embodiments, the sample is imaged at an electron dose that does not cause radiation damage.

[0070] In certain embodiments of all aspects and embodiments, the sample is imaged with an exposure time to the electron beam that does not cause fading of contrast.

[0071] In certain embodiments of all aspects and embodiments, the determining the viral particles is by a TOPAZ picking model that has been trained to recognize and select all viral particles. In certain embodiments of all aspects and embodiments, the determining the viral particles is by a TOPAZ picking model that has been trained to specifically recognize full, partially-filled or empty viral particles.

[0072] In certain embodiments of all aspects and embodiments, a script imports the images into RELION and performs automated selection of particles using TOPAZ.

[0073] In certain embodiments of all aspects and embodiments, a manual curation step can be / is performed to increase accuracy. In certain embodiments of all aspects and embodiments, the picked particles are extracted with a normal box size including some background and normalized.

[0074] In certain embodiments of all aspects and embodiments, the extracted viral particle images are reboxed with a smaller box-size that excludes the focus fringes and takes only the core density.

[0075] In certain embodiments of all aspects and embodiments, wherein the average gray value of each of the reboxed pixels is read out and plotted into a histogram.

[0076] In certain embodiments of all aspects and embodiments, an automated peak finding script determines the empty-full ratio based on the peaks in the histogram.

[0077] In certain embodiments of all aspects and embodiments, a script setting up a processing folder, linking the data and loading the required processing modules as well as automatically executing all steps is used, wherein a trained model (optionally an Al trained model) automatically selects the particles from the EM image (optionally a TOPAZ picker), followed by centering and extraction of the sub-images (optionally based on the RELION software), reading out the average density of the core of the particles (optionally by extracting images based on the RELION software), plotting density histogram and selecting the split (optionally manually or automatically).

[0078] In certain embodiments of all aspects and embodiments, an automated peak finding script determines the full-to-empty ratio of the rAAVps in the sample based on the peaks in a histogram, wherein from the EM images sub-images of rAAVps are identified and selected automatically by an Al-trained model using a Topaz picker, the sub-images are centered and a second level subimage of the rAAVps is extracted, the average density of the core of the rAAVps second level subimages are determined, a density histogram is plotted and the split of the density between empty and full rAAVps second level sub-images is determined.

[0079] In certain embodiments of all aspects and embodiments, the full-to-empty ratio of the rAAVps in the sample is determined based on the peaks in a histogram, wherein the histogram is generated by i) obtaining an EM image of the EM grid, ii) identifying in the EM sub-images of rAAVps, iii) selecting and extracting the sub-images from the EM image, iv) centering the sub-images and extracting a second level sub-image of the rAAVps, v) determining the average second level subimage density at the core of the rAAVps, vi) plotting a density histogram with the density at the x-axis and the number of second level sub-images at the y-axis, vii) determining the splitting density of empty and full rAAVp second level sub-images, and viii) determining the number of empty rAAVps with a density below the splitting density and the number of full rAAVps with a density above the splitting density from the histogram and thereby determining the amount of rAAVps in the sample.

[0080] In certain embodiments of all aspects and embodiments, the grids are loaded into a transmission electron microscope operating with an electron source at 120 kV.

[0081] In certain embodiments of all aspects and embodiments, images are recorded on an about 4000 by 4000 pixel charge-coupled device camera at a nominal magnification of about 30,000x yielding pictures with a pixel size corresponding to about 0.3914 nm at the specimen level.

[0082] In certain embodiments of all aspects and embodiments, the TOPAZ picking model was trained by manually picking 20 images each of samples.

[0083] In certain embodiments of all aspects and embodiments, picked particles are extracted with a normal box size (optionally of 200x200 pixels) and normalized.

[0084] In certain embodiments of all aspects and embodiments, wherein the picks are centered using a 2D classification step and then re-extracted to ensure that each particle sits exactly in the middle of the box.

[0085] In certain embodiments of all aspects and embodiments, the extracted images are reboxed (optionally with a 30x30 pixel box-size) to take only the core density.

[0086] In certain embodiments of all aspects and embodiments, the average gray value of each of the reboxed pixels is read out and plotted into a histogram.

[0087] In certain embodiments of all aspects and embodiments, the empty-full ratio is determined manually or automatically based on the peaks in this histogram.

[0088] In certain embodiments of all aspects and embodiments, the EM is a transmission EM (TEM).

[0089] In certain embodiments of all aspects and embodiments, the recombinant viral particle encapsidates a nucleic acid.

[0090] In certain embodiments of all aspects and embodiments, the recombinant viral particle comprises a protein encoding nucleic acid, an antisense oligonucleotide, a synthetic or artificial complementary region element.

[0091] Thus, the current invention comprises at least the following independent and dependent embodiments with terms in brackets being dependent embodiments, i.e. these can be present or not independently of each other even if not defined as optional: A method for determining the amount of recombinant adeno-associated virus particles (rAAVps) in a sample using electron microscopy (EM) with an EM grid without chemical staining, comprising the steps of

[0092] - drying the sample on the EM grid in air for 30 seconds to 90 seconds, and

[0093] - determining rAAVps on the EM grid using EM and thereby determining the amount of rAAVps in the sample. A method for determining the amount of recombinant, i.e. organic, viral particles in a sample using electron microscopy (EM) with an EM grid without chemical staining after application of the viral particle sample or / and without coating after application of the viral particle sample or / and without preserving the hydrated state of the viral particles by vitrification after application of the viral particle sample, comprising the steps of

[0094] - drying the sample on an EM grid (to generate denatured viral particles, i.e. to transfer the viral particles to a non-native state) or removing all solvent from the sample on the EM grid (preferably by evaporation at room temperature with optional prior blotting), and directly thereafter (i.e. without any further sample preparation step)

[0095] - determining and optionally classifying recombinant viral particles on the EM grid using EM and thereby determining the amount of recombinant viral particles in the sample. A method for determining the amount of recombinant, i.e. organic, viral particles in a sample using electron microscopy (EM) with an EM grid without chemical staining after application of the viral particle sample or / and without coating after application of the viral particle sample or / and without preserving the hydrated state of the viral particles by vitrification after application of the viral particle sample, comprising the steps of

[0096] - after application of the viral particle sample to an EM grid all solvent is removed from the EM grid by evaporation (optionally by blotting and evaporation) (preferably at room temperature), and directly thereafter (i.e. without any intervening sample preparation step)

[0097] - determining (and optionally classifying) recombinant viral particles on the EM grid using EM and thereby determining the amount of recombinant viral particles in the sample. A method for determining the amount of recombinant, i.e. organic, viral particles in a sample using electron microscopy (EM) with an EM grid without chemical staining after application of the viral particle sample or / and without coating after application of the viral particle sample or / and without preserving the hydrated state of the viral particles by vitrification after application of the viral particle sample, comprising the steps of

[0098] - washing away buffer salts, preferably with H2O and preferably by several repetitions of blotting after applying a drop of H2O and finally finishing with blotting away excess liquid,

[0099] - drying the sample on the EM grid (to generate denatured viral particles, i.e. to transfer the viral particles to a non-native state) or removing all solvent from the sample on the EM grid by evaporation (preferably at room temperature), and

[0100] - determining (and optionally classifying) recombinant viral particles on the EM grid using EM and thereby determining the amount of recombinant viral particles in the sample. A method for determining the amount of recombinant, i.e. organic, viral particles in a sample using electron microscopy (EM) with an EM grid without chemical staining after application of the viral particle sample or / and without coating after application of the viral particle sample or / and without preserving the hydrated state of the viral particles by vitrification after application of the viral particle sample, comprising the steps of

[0101] - washing the sample on the EM grid, preferably with H2O, more preferably by several (1, 2, 3, 4 or more) repetitions of blotting after applying a drop of H2O,

[0102] - drying the sample on the EM grid (to generate denatured viral particles, i.e. to transfer the viral particles to a non-native state) or removing all solvent from the sample on the EM grid by evaporation (preferably at room temperature), and directly thereafter (i.e. without any intervening sample preparation step) - determining (and optionally classifying) recombinant viral particles on the EM grid using EM and thereby determining the amount of recombinant viral particles in the sample.

[0103] 6. The method according to embodiment 5, wherein the method further comprises the step of

[0104] - cross-linking the sample, preferably by applying a cross-linking solution, wherein the cross-linking is i) before a washing step, ii) between the individual washing steps of the washing step, or iii) after the washing step, or iv) before the applying step.

[0105] 7. The method according to any one of embodiments 5 to 6, wherein the cross-linking solution comprises glutaraldehyde as crosslinking agent.

[0106] 8. The method according to any one of embodiments 5 to 7, wherein the cross-linking solution comprises glutaraldehyde at a concentration of about 2 % (v / v).

[0107] 9. The method according to any one of embodiments 5 to 8, wherein the cross-linking is by applying the cross-linking solution for about 15 to about 240 seconds.

[0108] 10. The method according to any one of embodiments 5 to 9, wherein the cross-linking is by applying the cross-linking solution for about 20 to about 120 seconds.

[0109] 11. The method according to any one of embodiments 5 to 10, wherein the cross-linking is by applying the cross-linking solution for about 60 seconds.

[0110] 12. The method according to any one of embodiments 1 to 11, wherein the grid is inserted into the EM and a vacuum is applied after the drying step.

[0111] 13. The method according to any one of embodiments 1 to 12, wherein the drying is in air outside the EM and in vacuum inside the EM.

[0112] 14. The method according to any one of embodiments 1 to 12, wherein the drying is in a nonoxygen containing gas outside the EM and in vacuum inside the EM.

[0113] 15. The method according to any one of embodiments 1 to 12, wherein the drying is freeze- drying outside the EM and in vacuum inside the EM. 16. The method according to any one of embodiments 1 to 15, wherein the drying is at room temperature.

[0114] 17. The method according to any one of embodiments 1 tol5, wherein the drying is at a temperature in the range between and including 5 °C to 70 °C, preferably in the range between and including 18 °C and 50 °C.

[0115] 18. The method according to any one of embodiments 1 to 17, wherein the EM grid comprises a support film that allows electron transmission, preferably a carbon coating.

[0116] 19. The method according to embodiment 18, wherein the support film is dried for at least 2 days after the application to the EM grid.

[0117] 20. The method according to any one of embodiments 1 to 19, wherein the EM grid is a glow- discharged carbon coated EM grid.

[0118] 21. The method according to any one of embodiments 1 to 20, wherein the EM coating is with a glow-discharged, continuous layer of carbon, preferably without holes.

[0119] 22. The method according to any one of embodiments 18 to 21, wherein the (carbon) coating on the EM grid is as flat as possible and as thin as possible

[0120] 23. The method according to any one of embodiments 18 to 22, wherein the thickness of the (carbon) coating on the EM grid is in the range of and including about 1 pm to about 0.1 nm.

[0121] 24. The method according to any one of embodiments 18 to 23, wherein the (carbon) coating is about 2 nm thick.

[0122] 25. The method according to any one of embodiments 1 to 24, wherein the method comprises prior to the drying step / as first step the step of

[0123] - applying the sample comprising the recombinant viral particles to the EM grid.

[0124] 26. The method according to embodiment 25, wherein the sample is applied to the EM grid for about 15 to 240 seconds.

[0125] 27. The method according to any one of embodiments 25 to 26, wherein the sample is applied to the EM grid for about 20 to 120 seconds.

[0126] 28. The method according to any one of embodiments 25 to 27, wherein the sample is applied to the EM grid for about 25 to 65 seconds. 29. The method according to any one of embodiments 25 to 28, wherein the method comprises after the applying step and before the drying step the step of

[0127] - washing the EM grid at least once with H2O.

[0128] 30. The method according to any one of embodiments 1 to 29, wherein the method comprises after the applying step and before the drying step the step of

[0129] - washing the EM grid three times with H2O.

[0130] 31. The method according to any one of embodiments 5 to 30, wherein the method comprises after the drying step and before the cross-linking step the step of

[0131] - washing the EM grid at least once with H2O or another dedicated solvent.

[0132] 32. The method according to any one of embodiments 5 to 31, wherein the method comprises after the drying step and before the cross-linking step the step of

[0133] - washing the EM grid three times with H2O or another dedicated solvent.

[0134] 33. The method according to any one of embodiments 1 to 32, wherein the viral particle is a recombinant adenoviral particle.

[0135] 34. The method according to any one of embodiments 1 to 32, wherein the viral particle is a recombinant herpes viral particle.

[0136] 35. The method according to any one of embodiments 1 to 32, wherein the viral particle is a recombinant adeno-associated viral particle (rAAVp).

[0137] 36. The method according to any one of embodiments 1 to 35, wherein the drying is a drying in air.

[0138] 37. The method according to any one of embodiments 2 to 36, wherein the drying is for at least 30 seconds.

[0139] 38. The method according to any one of embodiments 2 to 37, wherein the drying is for about 30 seconds to about 90 seconds.

[0140] 39. The method according to any one of embodiments 1 to 38, wherein the drying is for about 45 seconds to about 75 seconds. 40. The method according to any one of embodiments 1 to 39, wherein the drying is for about 60 seconds.

[0141] 41. The method according to any one of embodiments 1 to 40, wherein the viral particle is a composition of recombinant viral particles with different filling states.

[0142] 42. The method according to any one of embodiments 1 to 41, wherein the method is for determining the amount of full recombinant adeno-associated viral particles (rAAVps) in the sample.

[0143] 43. The method according to any one of embodiments 1 to 42, wherein the method is for determining the amount of empty rAAVps in the sample.

[0144] 44. The method according to any one of embodiments 1 to 43, wherein the method is for determining the amount of full and empty rAAVps in the sample.

[0145] 45. The method according to any one of embodiments 5 to 44, wherein the method is for determining the amount of partially filled rAAVps in the sample.

[0146] 46. The method according to any one of embodiments 5 to 45, wherein the method is for determining the amount of full, partially filled and empty rAAVps in the sample.

[0147] 47. The method according to any one of embodiments 1 to 46, wherein the sample is a rAAVp preparation or composition.

[0148] 48. The method according to any one of embodiments 1 to 47, wherein the method is for quality control of rAAVp preparations.

[0149] 49. The method according to any one of embodiments 1 to 47, wherein the method is for batch release.

[0150] 50. The method according to any one of embodiments 1 to 49, wherein the viral particle is a rAAVp and is of the serotype 2, 5 or 8 or an engineered variant thereof.

[0151] 51. The method according to any one of embodiments 1 to 50, wherein at least the last washing step or all washing steps are blotting steps.

[0152] 52. The method according to any one of embodiments 1 to 51, wherein the method comprises as determining step the steps of

[0153] - taking images of the EM grid with an EM, - determining the recombinant viral particles in the images, and thereby determining the amount of recombinant viral particles.

[0154] 53. The method according to embodiment 52, wherein the images are recorded manually or automatically.

[0155] 54. The method according to embodiment 52, wherein the images are recorded manually or automatically at precise defocus.

[0156] 55. The method according to any one of embodiments 52 to 54, wherein depending on the scattering strength of the background and the used magnification for taking the EM image the focus is selected in order for the image quality to be sufficient and to allow to distinguish the different filling states of the rAAVps.

[0157] 56. The method according to embodiment 52, wherein the images are recorded manually or automatically at precise defocus which takes into account the background of the grid to optimize contrast in the images to allow analysis.

[0158] 57. The method according to any one of embodiments 52 to 56, wherein the images are analyzed manually or automatically.

[0159] 58. The method according to any one of embodiments 52 to 56, wherein the images are analyzed manually or automatically using existing or dedicated software or artificial intelligence (Al) trained algorithms.

[0160] 59. The method according to any one of embodiments 52 to 56, wherein the images are analyzed manually or automatically using existing or dedicated software or IA trained algorithms, by classification of particles or by sorting of particles based on their core intensity, edges, and / or image features.

[0161] 60. The method according to any one of embodiments 52 to 59, wherein the images are analyzed by taking the density of the core of the particle excluding focus fringes to plot these average densities into a histogram and split the data into empty and full particles based on peaks in this histogram.

[0162] 61. The method according to any one of embodiments 52 to 60, wherein the sample is imaged using low-dose mode in the electron microscope. 62. The method according to any one of embodiments 52 to 61, wherein the sample is imaged at an electron dose that does not cause radiation damage.

[0163] 63. The method according to any one of embodiments 52 to 62, wherein the sample is imaged with an exposure time to the electron beam that does not cause fading of contrast.

[0164] 64. The method according to any one of embodiments 52 to 63, wherein the determining the viral particles is by a TOPAZ picking model that has been trained to recognize and select all viral particles.

[0165] 65. The method according to any one of embodiments 52 to 63, wherein the determining the viral particles is by a TOPAZ picking model that has been trained to specifically recognize full, partially-filled or empty viral particles.

[0166] 66. The method according to any one of embodiments 64 to 65, wherein a script imports the images into RELION and performs automated selection of particles using TOPAZ.

[0167] 67. The method according to any one of embodiments 52 to 66, wherein a manual curation step is performed to increase accuracy.

[0168] 68. The method according to any one of embodiments 52 to 67, wherein the picked particles are extracted with a normal box size including some background and normalized.

[0169] 69. The method according to any one of embodiments 52 to 68, wherein the extracted viral particle images are reboxed with a smaller box-size that excludes the focus fringes and takes only the core density.

[0170] 70. The method according to any one of embodiments 52 to 69, wherein the average gray value of each of the reboxed pixels is read out and plotted into a histogram.

[0171] 71. The method according to any one of embodiments 52 to 70, wherein an automated peak finding script determines the empty-full ratio based on the peaks in the histogram.

[0172] 72. The method according to any one of embodiments 52 to 71, wherein a script setting up a processing folder, linking the data and loading the required processing modules as well as automatically executing all steps is used, wherein a trained model (optionally an Al trained model) automatically selects the particles from the EM image (optionally a Topaz picker), followed by centering and extraction of the sub-images (optionally based on the RELION software), reading out the average density of the core of the particles (optionally by extracting images based on the RELION software), plotting density histogram and selecting the split (optionally manually or automatically). 73. The method according to any one of embodiments 52 to 72, wherein an automated peak finding script determines the full-to-empty ratio of the rAAVps in the sample based on the peaks in a histogram, wherein from the EM images sub-images of rAAVps are identified and selected automatically by an Al-trained model using a Topaz picker, the sub-images are centered and a second level sub-image of the rAAVps is extracted, the average density of the core of the rAAVps second level sub-images are determined, a density histogram is plotted and the split of the density between empty and full rAAVps second level subimages is determined.

[0173] 74. The method according to any one of embodiments 52 to 72, wherein the full-to-empty ratio of the rAAVps in the sample is determined based on the peaks in a histogram, wherein the histogram is generated by i) obtaining an EM image of the EM grid, ii) identifying in the EM sub-images of rAAVps, iii) selecting and extracting the sub-images from the EM image, iv) centering the sub-images and extracting a second level sub-image of the rAAVps, v) determining the average second level sub-image density at the core of the rAAVps, vi) plotting a density histogram with the density at the x-axis and the number of second level sub-images at the y-axis, vii) determining the splitting density of empty and full rAAVp second level sub-images, and viii) determining the number of empty rAAVps with a density below the splitting density and the number of full rAAVps with a density above the splitting density from the histogram and thereby determining the amount of rAAVps in the sample.

[0174] 75. The method according to any one of embodiments 52 to 74, wherein the grids are loaded into an electron microscope operating an electron source between 5 kV to 300 kV, preferably between 5 and 25 kV or at about 120 kV.

[0175] 76. The method according to any one of embodiments 52 to 75, wherein images are recorded on an about 4000 by 4000 pixel charge-coupled device camera at a nominal magnification of about 30,000x yielding pictures with a pixel size corresponding to about 0.3914 nm at the specimen level.

[0176] 77. The method according to any one of embodiments 52 to 76, wherein the TOPAZ picking model was trained by manually picking 20 images each of samples.

[0177] 78. The method according to any one of embodiments 52 to 77, wherein the picked particles are extracted with a normal box size (optionally of 200x200 pixels) and normalized. 79. The method according to any one of embodiments 52 to 78, wherein the picks are centered using a 2D classification step and then re-extracted to ensure that each particle sits exactly in the middle of the box.

[0178] 80. The method according to any one of embodiments 52 to 79, wherein the extracted images are reboxed (optionally with a 30x30 pixel box-size) to take only the core density.

[0179] 81. The method according to any one of embodiments 52 to 80, wherein the average gray value of each of the reboxed pixels is read out and plotted into a histogram.

[0180] 82. The method according to embodiment 81, wherein the empty-full ratio is determined manually or automatically based on the peaks in this histogram.

[0181] 83. The method according to any one of embodiments 1 to 82, wherein the EM is a transmission EM (TEM).

[0182] 84. The method according to any one of embodiments 1 to 83, wherein the recombinant viral particle encapsidates a nucleic acid.

[0183] 85. The method according to any one of embodiments 1 to 84, wherein the recombinant viral particle comprises a protein encoding nucleic acid, an antisense oligonucleotide, a synthetic or artificial complementary region element.

[0184] In addition to the various embodiments depicted and claimed, the disclosed subject matter is also directed to other embodiments having other combinations of the features disclosed or claimed herein. As such, the particular features presented herein can be combined with each other in other manners within the scope of the disclosed subject matter such that the disclosed subject matter includes any suitable combination of the features disclosed herein. The foregoing description of specific embodiments of the disclosed subject matter has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the disclosed subject matter to those embodiments disclosed. of the Figures

[0185] Figure 1 Averages of TEM images of rAAVps of serotype 2 and 8 after negative staining.

[0186] Figure 2 Scheme of different species of negatively stained images.

[0187] Figure 3 CryoEM average images of rAAVps of serotype 2 and 8.

[0188] Figure 4 Transmission EM images of rAAVps of serotype 2 and 8 after drying without staining or further conservation. Figure 5 Electron microscope imaging of rAAVp of serotype 8 with three different preparation techniques; left column: drying according to the current invention; middle column: negative staining (reference method); right column: drying followed by negative staining.

[0189] Figure 6 A) rAAVps of a serotype 2 variant prepared by drying with sample on one side of the grid; B) same rAAVps of a serotype 2 variant prepared by drying, but with sample material on both sides of the grid; C) image close to focus showing the case where the sample is on both sides of the grid and showing particles in over- and underfocus, i.e. non-focused, state within the same image.

[0190] Figure 7 Exemplary histogram obtained by automated analysis of a set of images from a dried EM grid obtained with the method according to the current invention. The arrow indicates the splitting density.

[0191] Detailed Description of the Invention

[0192] Herein is reported a robust, easy and rapid method for the determination of empty, partially-filled and full rAAVps that is applicable to all natural AAV serotypes as well as artificial capsids obtained by capsid engineering.

[0193] The current invention is based at least in part on the unexpected finding that rAAVps can be analyzed with respect to their loading state, i.e. if the particles are full, partially-filled or empty, by using electron microscopy, whereby the sample has been prepared by drying, such as drying in air.

[0194] The current invention is based at least in part on the unexpected finding that rAAVps do not completely collapse when dried, i.e. the current invention is based at least in part on the unexpected finding that individual rAAVps can be distinguished and characterized after being dried.

[0195] The current invention is based at least in part on the finding that despite the prejudices from the art, drying can be used for the preparation of rAAVp samples for EM analysis without destroying the information about the loading state of the rAAVps during the sample preparation process.

[0196] DEFINITIONS

[0197] Unless otherwise defined herein, scientific and technical terms used in connection with the present invention shall have the meanings that are commonly understood by those of ordinary skill in the art. Further, unless otherwise required by context, singular terms shall include pluralities and plural terms shall include the singular. The methods and techniques of the present disclosure are generally performed according to conventional methods well known in the art. Generally, nomenclatures used in connection with, and techniques of biochemistry, enzymology, molecular, and cellular biology, microbiology, genetics and protein and nucleic acid chemistry and hybridization described herein are those well-known and commonly used in the art.

[0198] Useful methods and techniques for carrying out the current invention are described in e.g. Ausubel, F.M. (ed.), Current Protocols in Molecular Biology, Volumes I to III (1997); Glover, N.D., and Hames, B.D., ed., DNA Cloning: A Practical Approach, Volumes I and II (1985), Oxford University Press; Freshney, R.I. (ed.), Animal Cell Culture - a practical approach, IRL Press Limited (1986); Watson, J.D., et al., Recombinant DNA, Second Edition, CHSL Press (1992); Winnacker, E.L., From Genes to Clones; N.Y., VCHPublishers (1987); Celis, J., ed., Cell Biology, Second Edition, Academic Press (1998); Freshney, R.I., Culture of Animal Cells: A Manual of Basic Technique, second edition, Alan R. Liss, Inc., N.Y. (1987); Williams and Carter: https: / / link.springer.com / book / 10.1007 / 978-l-4757-2519-3 and Frank: https: / / academic.oup.com / book / 6747. The content of which is incorporated herein by reference.

[0199] The use of recombinant DNA technology enables the generation of derivatives of a nucleic acid. Such derivatives can, for example, be modified in individual or several nucleotide positions by substitution, alteration, exchange, deletion or insertion. The modification or derivatization can, for example, be carried out by means of site directed mutagenesis. Such modifications can easily be carried out by a person skilled in the art (see e.g. Sambrook, J., et al., Molecular Cloning: A laboratory manual (1999) Cold Spring Harbor Laboratory Press, New York, USA; Hames, B.D., and Higgins, S.G., Nucleic acid hybridization - a practical approach (1985) IRL Press, Oxford, England).

[0200] It must be noted that as used herein and in the appended claims, the singular forms "a", "an", and "the" include plural reference unless the context clearly dictates otherwise. Thus, for example, reference to "a cell" includes a plurality of such cells and equivalents thereof known to those skilled in the art, and so forth. Also, the terms "a" (or "an"), "one or more" and "at least one" can be used interchangeably herein. It is also to be noted that the terms "comprising", "including", and "having" can be used interchangeably.

[0201] The term “about” denotes a range of + / - 20 % of the following numerical value. In certain embodiments, the term about denotes a range of + / - 10 % of the thereafter following numerical value. In certain embodiments, the term “about” denotes a range of + / - 5 % of the thereafter following numerical value.

[0202] The terms “comprise(s),” “include(s),” “having,” “has,” “can,” “contain(s)” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms or words that do not preclude the possibility of additional acts or structures. The term “comprising” also encompasses the term “consisting of’. The present disclosure also contemplates other embodiments “comprising”, “consisting of’ and “consisting essentially of’ the embodiments or elements presented herein, whether explicitly set forth or not.

[0203] The terms “empty recombinant AAV particle” and “empty rAAVp”, which can be used interchangeably, denote a protein shell composed of adeno-associated viral capsid polypeptides without a therein encapsidated / packaged nucleic acid. That is, an empty rAAVp is at least almost free or even free of encapsidated nucleic acid.

[0204] The terms “partially filled recombinant AAV particle” or “partially filled rAAVp”, which can be used interchangeably, denotes a non-covalent complex of a protein shell composed of adeno- associated viral capsid polypeptides and a therein encapsidated / packaged nucleic acid sequence, whereby the nucleic acid sequence is smaller / shorter than the (full) nucleic acid of interest. In principle, the encapsidated part of the nucleic acid can be a smaller, non-target nucleic acid or a fragment of the (full-length) nucleic acid of interest. Whether these shorter nucleic acids lead to a functional protein may remain open.

[0205] The terms “full recombinant AAV particle” or “full rAAVp”, which can be used interchangeably, denote a non-covalent complex formed of a protein shell composed of adeno-associated viral capsid polypeptides and a therein encapsidated / packaged (full-length) nucleic acid of interest. In certain embodiments of all aspects and embodiments, the functional protein or the functional transcript of interest has a therapeutic effect. In certain embodiments of all aspects and embodiments, the (full-length) nucleic acid of interest can be transcribed into a functional protein of interest, preferably a functional therapeutic protein of interest.

[0206] The term “full-to-empty ratio” denotes the mathematical ratio of the number of full rAAVps to the total number of rAAVps (i.e. to the sum of full, partially filled and empty rAAVps) in a sample or in a rAAVp preparation. As the number of full rAAVps can be at most the same as the total number of rAAVps, the ratio can be at most 1. Generally, the ratio is less than 1 and is expressed as a percentage. The number of full rAAVps can be determined by determining the number of nucleic acid sequences interspaced between two AAV ITRs in the sample or preparation.

[0207] The terms “transgene” and “nucleic acid of interest”, which can be used interchangeably herein, denote a nucleic acid derived from a wild-type genome of an adeno-associated virus, wherein except for the ITRs (adeno-associated virus Inverted Terminal Repeats; this term denotes the structural element as such; it encompasses wild-type ITRs as well as engineered ITRs) sequences most or all endogenous AAV nucleic acids are replaced by one or more (exogenous) nucleic acid(s) of interest. For example, such a nucleic acid of interest can be a nucleic acid transcribed into a transcript of interest or that encodes a therapeutic protein or a therapeutic nucleic acid, such as an ASO (antisense oligonucleotide), an shRNA, an siRNA etc. Typically, for a transgene one or both ITR structural elements of the wild-type AAV genome are retained either with wild-type or variant sequence. Thus, a transgene can be distinguished from a wild-type AAV genome, since all or at least a part of the viral genome has been replaced with a non-native (i.e. exogenous) nucleic acid of interest with respect to the virus. Incorporation of a non-native nucleic acid therefore defines the AAV particle as a "recombinant" AAV particle. It has to be pointed out that the serotype of the ITRs in the transgene does not need to be the same as the serotype of the adeno-associated viral capsid polypeptides forming the shell of the rAAVp comprising said transgene, nor does it need to remain wild-type.

[0208] The “transgene" is a portion of a larger nucleic acid, e.g. of a recombinant plasmid, that is ultimately packaged or encapsulated or encapsidated either directly or in form of a single strand, a double strand or in form of RNA or ASOs into a protein shell composed of adeno-associated viral capsid polypeptides to form a rAAVp. In cases where recombinant plasmids are used to construct or manufacture rAAVps, the viral particle does not include the portion of the "plasmid" that does not correspond to the transgene part of the recombinant plasmid. For example, in case of a rAAVp the recombinant vector comprises a part that is interspaced between two AAV ITRs. The non-vector portion of the recombinant plasmid is referred to as the "plasmid backbone". The plasmid backbone is important for cloning and amplification of the plasmid, a process that is needed for propagation and recombinant virus production but is not itself packaged or encapsulated or encapsidated into the rAAVp. Thus, a “transgene" refers to the nucleic acid that is packaged or encapsulated or encapsidated by a protein shell composed of adeno-associated viral capsid polypeptides, i.e. in a rAAVp.

[0209] In principle, any non- AAV nucleic acid can be packaged into a shell composed of adeno-associated viral capsid polypeptides resulting in a rAAVp, e.g. for subsequent infection (transduction) of a cell, ex vivo, in vitro or in vivo.

[0210] As used herein, the term "serotype" is used to classify different wild-type and recombinant AAV particles based on the amino acid sequence of the polypeptides forming the protein shell (capsid) of the respective AAV particles. Originally, serologic distinctiveness was determined based on the lack of cross-reactivity between antibodies to one AAV particle as compared to another AAV particle. Such cross-reactivity differences are usually due to differences in capsid polypeptide sequences and the respective antigenic determinants (e.g., due to VP1, VP2, and / or VP3 sequence differences of AAV serotypes). Despite the possibility that AAV variants including capsid variants may not be serologically distinct from a reference or wild-type AAV or other AAV serotype, they differ by at least one amino acid residue compared to the reference or wild-type or other AAV serotype.

[0211] Under the traditional definition, a serotype means that the virus of interest has been tested against serum specific for all existing and characterized serotypes for neutralizing activity and no antibodies have been found that neutralize the virus of interest. As more naturally occurring virus isolates are discovered and / or capsid mutants generated, there may or may not be serological differences with any of the currently existing serotypes. Thus, in cases where the new AAV particle has no serological difference, this new AAV particle would be a subgroup or variant of the corresponding wild-type serotype. In many cases, serology testing for neutralizing activity has yet to be performed on mutant viruses with capsid sequence modifications to determine if they are of another serotype according to the traditional definition of serotype.

[0212] THE METHOD ACCORDING TO THE INVENTION

[0213] Two EM methods are commonly used for the analysis of the loading state of rAAVps, i.e. negative staining and the current gold standard CryoEM.

[0214] In CryoEM grid-preparation is done by freezing without drying. Additionally, the demands on the microscope are high and the technological expertise needed to accurately perform the technology is high as well. Having said this, CryoEM requires a research facility equipped with N2 sensors, liquid nitrogen supply, and trained professionals (the sample has to be handled at about -190 °C all the time). This makes CryoEM a very cost and labor intensive as specialized equipment and well-trained personnel are needed to perform the method. The method is thus not applicable in a production and / or clinical environment.

[0215] Staining and freezing are commonly known methods for preserving the biologicals materials shape and texture, something that is lost when drying is used. Since the loss of shape and texture upon drying of biological materials is well known in the field, a person skilled in the art will avoid such a technology.

[0216] In contrast to CryoEM and staining-based methods the core of the current invention is the complete removal of the sample’s solvent by (physical) drying. This is in contrast to the belief in the state of the art that biological samples need to be preserved from drying before getting meaningful electron microscopy images. The current inventors have surprisingly found that in the case of viruses and the question of whether they are empty or full, no preservation of the native state of the viral particles in the sample is required, and, thus, that neither staining nor freezing is required. Amongst other things, with the method according to the current invention a comparable sensitivity as with the gold standard CryoEM can be achieved but with a much simpler and cheaper procedure. In contrast to CryoEM, the method according to the current invention can be used to quantify full- to-empty ratios. The method according to the current invention is applicable, e.g., in a hospital or production environment.

[0217] In contrast to staining, the method according to the current invention provides meaningful readouts that allow direct quantification of full-to-empty ratios.

[0218] That is, the method according to the current invention requires only a low amount of material, allows a direct per particle read on empty or full viral particles and does not require a prior assumption or models to analyze the data.

[0219] The method according to the current invention can be conducted with a low performance, ease of use, bench-top EM (at any location and around the globe). The quality of formulations (before and after shipping, freezing and thawing) can be verified with high accuracy. The method according to the current invention is a radical simplification of a high-accuracy CryoEM and a radical improvement in sensitivity of staining methods making the determination of full-to-empty ratios of rAAV preparation broadly applicable.

[0220] This is shown in detail in the following.

[0221] Whether a stain can differentiate between empty and full rAAVps at all depends on several factors, such as the capsid type, the staining technique (duration, blotting, washing, timing, etc.) as well as the staining agent. It has turned out that the negative staining method is not reliable for the analysis of rAAVps as it is questionable whether the staining agent did penetrate all empty rAAVps and more importantly whether it did penetrate full rAAVps.

[0222] This is exemplarily shown by comparing the negative staining of rAAVps of the serotypes 2 and 8 (see Figure 1) with 2 % uranyl acetate solution as staining agent. It can be seen that although the empty rAAVps of the serotype 2 are well penetrated by the staining agent and can be differentiated from full rAAVps, the same cannot be achieved for empty rAAVps of the serotype 8, which after staining cannot be differentiated from full rAAVps of the same serotype.

[0223] In Figure 2 schemes of the different species that can result from negative staining are shown. In more detail, on the left of Figure 2 an empty rAAVp that is intact but was not penetrated by the staining agent is shown. This is followed by a rAAVp that has been penetrated by the staining agent, followed by a partially collapsed rAAVp and a fully collapsed rAAVp. The middle two forms cannot always be distinguished as the amount of staining agent penetration as well as the amount of collapsing can vary. The same comparison as shown in Figure 1 has been done using CryoEM, a more complex, expensive and time-consuming method compared to negative staining. The results are shown in Figure 3. It can be seen that with CryoEM the differentiation between empty and full rAAVps of both serotypes 2 and 8 is possible.

[0224] The correct differentiation between full and empty rAAVps is required amongst other things for quality control (QC) of rAAVp preparations. QC is needed for, e.g., cell line optimization, optimization of capsid packaging properties, capsid engineering, upscaling of production methods, buffer, storage and delivery optimization as well as batch control for release.

[0225] It has now been found by the current inventors that despite the prejudices from the art, drying can be used for the preparation of rAAVp samples for EM analysis without destroying the information regarding the loading state of the rAAVps during the sample preparation process. This is shown in Figure 4. It can be seen that full and empty rAAVps can be properly distinguished after drying, in this case drying in air.

[0226] The current invention is based, at least in part, on the finding that a preparation method, which is not considered suitable for biological materials, works surprisingly for AAV particles. It has been found that the harm done by drying AAVs is contributing to the ease of distinguishing empty and full particles. On top of this, the individual particles remain recognizable and there is a different contrast due to the presence or absence of DNA. Without being bound by this theory, it is assumed that empty particles are more fragile and seem to collapse readily, a fact that increases the difference in contrast between empty and full particles.

[0227] In more detail, it can be seen in the rAAVp images of Figure 4 that after drying two distinguishable populations of rAAVps are obtained. The first population appears lighter due to less scattering and the second population appears darker due to stronger scattering. Tests with mixing different ratios of empty and full rAAVps show reproducible and consistent results.

[0228] Thus, herein is reported a robust method for the determination of empty, partially-filled and full rAAVps that is applicable to all natural AAV serotypes as well as artificial capsids obtained by capsid engineering.

[0229] Without being bound by this theory, taking the thermostability of the AAV serotypes as a measure for their overall rAAVp stability it can be expected that the different serotypes could behave differently when dried, where a more fragile empty rAAVp is more likely to collapse completely compared to a more sturdy rAAVp. The least stable rAAVp is that of AAV serotype 2 (AAV-2) which denatures completely after the temperature is increased to above 70 °C. A rAAVp of AAV serotype 5 (AAV-5) is the most stable rAAVp and is only completely denatured when the temperature is raised over 92 °C (see, e.g., Bennett, A., et al., Mol. Ther. 2017 (6) 171-182). It is not that upon drying the particles remain intact. Without being bound by this theory, it is assumed that the presence of a genome inside the capsid seems to stabilize the particle, resulting in an additional difference between full and empty capsids as full rAAVps collapse less readily compared to empty particles. The empty rAAVps can either remain intact and be lighter, or collapse and be (even more) lighter. For determining the empty-to-full ratio this is not important since the contrast is largely originating from the presence or absence of a genome inside the particle.

[0230] However, for determining intermediately filled particles it will be needed to prevent the collapse of the rAAVps during sample preparation, to prevent uneven distribution of contrast over less (uncollapsed) or more pixels (collapsed) in the image. This can be done, e.g., by fixation. That way, the diameter of the rAAVps remains intact during sample preparation. Then, when comparing, the different densities inside the capsids are represented by the same number of pixels in the images. Without being bound by this theory, it is assumed that when the size of the rAAVps remains the same, differences in intensity directly translate to differences in rAAVp filling, i.e. the size of the encapsidated nucleic acid.

[0231] The method according to the current invention is based at least in part on the finding that the difference in density between empty and full rAAVps can be used to distinguish the loading state of the rAAVp. The material inside a full capsid, i.e. the encapsidated nucleic acid, results in increased scattering of electrons compared to an empty capsid. This leads to a darker appearance in the microscope. The method according to the current invention allows amongst other things for an improved determination of the full-to-empty ratio of rAAVps in a sample, the improvement being, e.g., a reduced analysis time and improved accuracy.

[0232] It has been found that drying according to the method of the current invention also causes flattening of preferably empty capsids. Without being bound by this theory it is assumed that this flattening may play a role in the total difference in contrast that can be observed.

[0233] It has been shown that the method according to the current invention can be used to analyze samples comprising the most stable and the least stable rAAVps. This supports the general applicability of the method according to the current invention and as well shows the advantageous properties in comparison to CryoEM and negative staining methods, respectively. It further shows that the method according to the current invention is suitable for and independent of the serotype of the rAAVp, i.e. it is independent on the origin of the rAAVp, may it be natural or engineered.

[0234] The methods according to the current invention have been compared with the state of the art method of negative staining. The results are shown in Figure 5. In Figure 5 electron microscope images of samples of rAAVps of the serotype 8 prepared with three different techniques, i.e. drying according to the current invention, negative staining (reference method) and drying according to the current invention followed by negative staining, are shown (see Table 1 for details) to show the differences resulting from the individual preparation methods.

[0235] In more detail, in the top-row of Figure 5 a sample of empty rAAVps of serotype 8 (99 % empty (as determined by CryoEM)) was used. In the other two rows the same sample, a mixture of empty and full rAAVps of serotype 8, was used, whereby in the bottom row the sample was additionally fixed on the grid with glutaraldehyde after drying. It can be seen that said fixation manages to prevent flattening after drying for 60 seconds (right panel) and the difference between the empty and full particles is still visible. It is expected that full-to-empty analysis with fixed grids after drying will be working equally well and is also an aspect according to the current invention. Without being bound by this theory, it is assumed that fixing prevents the flattening of the rAAVps and thereby prevents the rAAVps from losing their shape. Additionally, it is assumed that taking an average inner density of each rAAVp after correction for the background of the image could be suitable to measure not only full and empty but also be able to identify and distinguish rAAVps comprising incomplete stretches of nucleic acid as well as overfilled rAAVps.

[0236] It has to be pointed out that when a sample is completely dried out, it, e.g., starts to physically absorb the stain. Thus, in one preferred embodiment of all aspects, the sample is dried for about 60 seconds to prevent complete drying.

[0237] Table 1: Details of the different sample preparation techniques used to generate the images of Figure 5.

[0238] AD NS AD+NS fix+AD fix+NS fix+AD+NS cryo

[0239] Sample |(|l y 30s y 30s y 30s y 30s y 30s y 30s y 30s y: yes

[0240] AD: drying

[0241] NS: negative staining fix: fixation cryo: cryo electron microscopy (CryoEM)

[0242] The current invention is also based at least in part on the finding that the sample should be on one side of the grid only to avoid misleading results. This is shown in Figure 6.

[0243] In more detail, in Figure 6 images with rAAVps of a serotype 2 variant are shown. In the left image (Figure 6A) the rAAVps are only on one side of the grid, whereas in the middle and right images (Figures 6B and 6C) the rAAVps are on both sides of the grid. It can be seen that full particles that are located on the backside of the grid appear as light as empty rAAVps on the front of the grid when going under focus. This leads to a serious overestimation of empty rAAVps.

[0244] Further, it has been found that taking an image close to focus shows some particles that are under focused whereas others are over focused. Without being bound by this theory, it is assumed that this can be used to detect whether a double layer of viral particles is present.

[0245] It has been found that, depending on the scattering strength of the background and the magnification used, a suitable focus has to be selected in order for the image quality to be sufficient and to allow distinguishing the different filling states of the rAAVps. The current invention is also based at least in part on the finding that the dried sample should be imaged using low-dose mode in the electron microscope. Prolonged exposure to the electron beam will cause fading of contrast, and thus it is important for consistent results to prevent radiation damage by limiting the electron dose to which the sample is exposed. Different full-to-empty ratios of rAAVps have been analyzed with the method according to the current invention. The results are shown in Table 2.

[0246] Table 2: Spiking data. mean %Empty: based on the combining of each measured particle separately

[0247] Consistency of measured contrast, between sample consistency, inner methodological consistency and comparison of drying (AD measurements) and cryo-TEM (cryo) can be learned from this spiking experiment.

[0248] It can be seen that the method according to the current invention determines the different full-to- empty ratios as well as percentages of full particles, respectively, correctly. In more detail, from the data of Table 2 it can be seen that results from individual datasets can be combined and normalization of images works well to correct for differences in background or imaging settings.

[0249] From the data presented in Table 3 it can be seen that the method according to the current invention and the gold standard state of the art method, i.e. CryoEM, provide for the same results. Table 3: Data comparing the method according to the current invention (drying) with cryo electron microscopy (CryoEM); given is percentage empty.

[0250] *: if more than one value is shown, different grid preparations of the same sample have been analyzed

[0251] The results obtained with the method according to the current invention are expected to be independent of microscopy grid-type, although a thin, flat and regular carbon support might be advantageous, e.g. resulting in improved contrast between full and empty rAAVps. Furthermore, the exact drying conditions, amount of sample, dilution buffer, incubation time, number of washing steps, washing buffer (as long as it is free of solutes) and method of image analysis can be varied in the method according to the current invention based on the knowledge of a person skilled in the art, resulting in alternative, equally efficient working methods.

[0252] Optionally, a fixation step, for example with glutaraldehyde, can be added to the method according to the current invention to reduce the collapsing of the rAAVps during the drying and to obtain information regarding partially-filled rAAVps. Without being bound by this theory, it is assumed that the method according to the current invention could further benefit from automatic selection of rAAVp images.

[0253] Likewise imaging with other instrumentation, camera, and controlling software as that used in the examples presented herein will be likewise suitable for performing the method according to the current invention. This could, e.g., include AFM and SEM instead of TEM as well as other scanning techniques like STEM.

[0254] Several magnifications were tested and found to be suitable for detection of empty and full rAAVps.

[0255] In order to accelerate image processing and particle counting this could be automated.

[0256] The EM images obtained with an EM grid prepared with the method according to the current invention can be analyzed using existing or dedicated software. As an example, a python script was created that uses the RELION 4.0.0 processing pipeline along with a Topaz picking algorithm. The Topaz picking model was trained to recognize and select sub-images of all viral particles, or different models can be trained to only recognize specifically full, empty or intermediate viral particles. The script was designed to import the images into RELION and to perform automated selection of sub-images of the viral particles from the EM image using Topaz. Optionally, a manual curation step can be performed to increase accuracy. The picked particles were then extracted with a normal box size (at least bigger than the particle to include some background) and normalized. The normalization step of this extraction takes into account the contribution of the background and makes sure that all particles have comparable contrast. To ensure that each particle is located exactly in the middle of the box and to allow removal of duplicate picks, the picks were centered using, e.g., a 2D classification step and then re-extracted to obtain a second level sub-image. The extracted images were reboxed with a smaller box-size that excludes the focus fringes and includes mainly the core density. The average gray value of each of the reboxed pixels was read out and plotted into a histogram. An automated peak finding script was used to determine the full-to-empty ratio based on the peak distribution in the histogram. This likewise can be done manually. An exemplary histogram with empty-full allocation is shown in Figure 7; the splitting density is indicated with an arrow.

[0257] For the CryoEM data analysis, images were imported into the EMAN2 software package and picking and sorting of viral images was done manually to assure highest accuracy and avoid software biases. In general, and in one embodiment of all aspects and embodiments according to the current invention, a script setting up a processing folder, linking the data and loading the required processing modules as well as automatically executing all steps is used. The trained model (e.g. Al trained) automatically selects the particles from the EM image (e.g., using TOPAZ picker), followed by centering and extraction of the sub-images (e.g. based on the RELION software), reading out the average density of the core of the particles, i.e. extracted images (e.g. based on the RELION software), plotting density histogram (e.g. with a script or in Excel), and selecting the split (manually or automatically).

[0258] Likewise, different picking algorithms can be used for identifying images and sub-images of full and empty rAAVps in EM images obtained with the method according to the current invention.

[0259] Likewise, any other suitable software or dedicated script or Al tool may be used.

[0260] Furthermore, image analysis can be done manually to avoid further use of software.

[0261] ***

[0262] All publications, patents, and patent applications cited herein are hereby incorporated by reference herein in their entirety for all purposes to the same extent as if each individual publication, patent, and patent application were specifically and individually indicated to be so incorporated by reference. In the event that one or more of the incorporated literature and similar materials differs from or contradicts this application, including but not limited to defined terms, term usage, described techniques, or the like, this application controls.

[0263] The following examples and figures are provided to aid the understanding of the present invention, the true scope of which is set forth in the appended claims. It is understood that modifications can be made in the procedures set forth without departing from the spirit of the invention.

[0264] That is, although the disclosed teachings have been described with reference to various applications, methods, and compositions, it will be appreciated that various changes and modifications can be made without departing from the teachings herein and the claimed invention below. The examples are provided to better illustrate the disclosed teachings and are not intended to limit the scope of the teachings presented herein. While the present teachings have been described in terms of these exemplary embodiments, the skilled artisan will readily understand that numerous variations and modifications of these exemplary embodiments are possible without undue experimentation. All such variations and modifications are within the scope of the current teachings. Examnles

[0265] Example 1

[0266] Method according to the ioveotioo - dryiog sample preparatioo

[0267] Recombinant AAV particles containing samples were diluted in D-PBS (Gibco Life Technologies) until a suitable concentration was reached. Electron microscopy grids (T600H-Cu 698 1 / inch Hex. mesh Thin Bar; EMS) were coated with an approx. 2 nm thick continuous carbon film by floating the carbon on ultrapure H2O and letting the water level drop till the carbon covered the grids. After at least 2 days of drying the grids were used. About 2-4 pL of sample was incubated on a glow- discharged carbon coated grid for 30-60 seconds, followed by three steps of washing with H2O. After the final blotting step (adding a drop of water which was then taken away by touching the grid to a filter paper to remove most of the fluid), the sample was left to dry at RT.

[0268] Example 2

[0269] Method according to the invention - drying and fixation

[0270] Recombinant AAV particles containing samples are diluted in D-PBS (Gibco Life Technologies) until a suitable concentration is reached. Electron microscopy grids (T600H-Cu 698 1 / inch Hex. mesh Thin Bar; EMS) are coated with a home-made approx. 2 nm thick carbon film by floating the carbon on ultrapure H2O and letting the water level drop till the carbon covers the grids. After at least 2 days of drying the grids are used. About 2-4 pL of the sample is incubated on a glow- discharged carbon coated grid for 30-60 seconds, followed by two steps of washing with H2O. Thereafter a fixation was done by adding a drop of 2 % (v / v) glutaraldehyde solution and incubating for 60 seconds, followed by two steps of washing with H2O. After the final blotting step, the sample was left to dry at RT.

[0271] Example 3

[0272] Comparative method - cryo-electron microscopy preparation

[0273] Samples were diluted in D-PBS (Gibco Life Technologies) to the same concentration that was used in Example 1 for the method according to the invention using drying. About 2-4 pL of the sample was incubated on a glow-discharged carbon coated Quantifoil grid for 30 seconds (Quantifoil - R2 / 1, 300, Cu + 2 nm, Germany), blotting of 3 seconds and subsequent plunging the grid into liquid ethane at -180 °C using a Leica EM GP automated plunging device (Leica Microsystems, Vienna, Austria). Example 4

[0274] Electron Microscopy

[0275] Grids were loaded into a Jeol JEM-1400 Plus transmission electron microscope operating a Lab6 electron source at 120 kV. Electron micrographs were recorded on TVIPS XF416 4000 by 4000 pixel charge-coupled device camera (Tietz Video and Image Processing System, Gauting, Germany) at a nominal magnification of 30,000x yielding pictures with a pixel size corresponding to 0.3914 nm at the specimen level. Both drying (Example 1) and CryoEM grids (Example 3) were imaged using low dose mode.

[0276] Example 5

[0277] Analysis

[0278] EM image data was analyzed using a python script that uses the RELION 4.0.0 processing pipeline (J. Zivanov, J. Oton, Z. Ke, K. Qu, D. Morado, D. Castano-Diez, A. von Kugelgen, T.A.M. Bharat, J.A.G. Briggs & S.H.W. Scheres (2022) "A Bayesian approach to single-particle electron cryotomography in RELION-4.0"; D. Kimanius, L. Dong, G. Sharov, T. Nakane & S.H.W. Scheres (2021) "New tools for automated CryoEM single-particle analysis in RELION-4.0"; S.H.W. Scheres (2012) "RELION: Implementation of a Bayesian approach to CryoEM structure determination") along with the implementation of the Topaz picking algorithm (Bepler, T., Morin, A., Rapp, M. et al. Positive-unlabeled convolutional neural networks for particle picking in cryoelectron micrographs. Nat Methods 16, 1153-1160 (2019); doi: 10.1038 / s41592-019-0575-8) and the RELION processing pipeline. The Topaz picking model was trained by manually picking 20 images each of the sample EM images as well as data from previous AAV sample EM images. The script imports the EM images into RELION and performs automated selection of particle subimages using Topaz. An optional manual curation step allows for improved picking. The picked particles were then extracted with a normal box size (200x200 pixels; pixel size of 3.914A; dependent on the magnification; in this case the box that was used was 78.3 nm) and normalized. To ensure that each particle sits exactly in the middle of the box, the picks were centered using a 2D classification step and then re-extracted without duplicates to generate second level subimages. The normalization step of this extraction took into account the contribution of the background and made sure that all particles have comparable contrast. These extracted images were reboxed with a 30x30 pixel box-size (equal to 11.7 nm corresponding to core, staying away from the edge so as not to include fringes and capsid (total diameter of AAV is 25 nm)) to take only the core density. The average gray value of each of the reboxed pixels was read out into excel and plotted into a histogram. The empty -full ratio was determined manually or automatically based on the peaks in this histogram. For the CryoEM data analysis, images were imported into the EMAN2 software package (Tang, G., et al., J. Struct. Biol. 157 (2007) 38-46) and picking and sorting of viral images was done manually to assure the highest accuracy and avoid software biases.

Claims

Patent Claims1. A method for determining the amount of recombinant adeno-associated virus particles (rAAVps) in a sample using electron microscopy (EM) with an EM grid, wherein the method is performed without chemical staining or surface coating or cryo-fixation of the sample on the EM grid, comprising the steps of- drying the sample on the EM grid in air to remove all solvent of the sample by evaporation, and without any further sample preparation steps- determining rAAVps on the EM grid using EM and thereby determining the amount of rAAVps in the sample.

2. The method according to claim 1, wherein the EM grid is a glow-discharged carbon coated EM grid with a continuous layer of carbon.

3. The method according to claim 2, wherein the carbon coating on the EM grid is about 2 nm thick.

4. The method according to any one of claims 1 to 3, wherein the method comprises prior to the drying step the step of- applying the sample to the EM grid.

5. The method according to claim 4, wherein the method comprises after the applying step and before the drying step the step of- washing the EM grid at least once with H2O.

6. The method according to any one of claims 1 to 5, wherein the method is for determining the amount of full rAAVps in the sample.

7. The method according to any one of claims 1 to 5, wherein the method is for determining the amount of empty rAAVps in the sample.

8. The method according to any one of claims 1 to 5, wherein the method is for determining the amount of full and empty rAAVps in the sample and thereby determining the full-to- empty ratio of rAAVps in the sample.

9. The method according to any one of claims 1 to 8, wherein the rAAVps are of the serotype 2, 5 or 8 or are a variant thereof.

10. The method according to any one of claims 1 to 9, wherein the full-to-empty ratio of the rAAVps in the sample is determined based on the peaks in a histogram, wherein the histogram is generated by i) obtaining an EM image of the EM grid, ii) identifying in the EM image sub-images containing an rAAVp image, iii) selecting and extracting the sub-images from the EM image, iv) centering the sub-images and extracting a second level sub-image of the rAAVps from the EM image, v) determining the average second level sub-image density or / and the scattering intensity at the core of the rAAVps in the second level sub-image, vi) plotting a density histogram with the density at the x-axis and the number of second level sub-images with said density at the y-axis, vii) determining the splitting density of empty and full rAAVp second level sub-images, and viii) determining the number of empty rAAVps with a density below the splitting density and the number of full rAAVps with a density above the splitting density from the histogram and thereby determining the amount of rAAVps in the sample.

Images