Detection of variability in viruses

Abstract

Disclosed herein are methods for detecting the presence of a plurality of adeno-associated viruses (AAV) or adenoviruses (Ad) in a sample. Also described are methods for determining a plurality of serotypes of AAV or Ad viruses in a sample. Further disclosed are methods to detect the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of AAV or Ad. The use of LC-MS to detect the presence of a plurality of AAV or Ad in a sample; or to determine a plurality of serotypes of a plurality of AAV or Adin a sample; or detect the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of AAV or Ad is also disclosed.

Description

[0001] Detection of variability in viruses

[0002] Cross-References

[0003] This application claims priority to U.S. Application Number 63 / 518,147 filed on August 8, 2023.

[0004] Field of the Invention

[0005] Disclosed herein are methods for detecting the presence of a plurality of adeno-associated viruses (AAV) or adenoviruses (Ad) in a sample. Also described are methods for determining a plurality of serotypes of AAV or Ad viruses in a sample. Further disclosed are methods to detect the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of AAV or Ad. The use of LC-MS to detect the presence of a plurality of AAV or Ad in a sample; or to determine a plurality of serotypes of a plurality of AAV or Adin a sample; or detect the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of AAV or Ad is also disclosed.

[0006] Background

[0007] Adeno-associate virus and Adenovirus are virus vectors used in gene therapy. There are various serotypes of these viruses. It is an important for the quality control of gene therapy treatments that where there are multiple serotypes in a manufacturing batch or heterogeneity within a serotype, that this is detected.

[0008] Summary of the Invention

[0009] In a first aspect, there is provided a method for detecting the presence of a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample, the method comprising: a) denaturing the viral capsids in the sample; b) performing HILIC-MS (hydrophilic interaction liquid chromatography-mass spectrometry) on the sample comprising the denatured viral capsids; and c) determining the masses of one or more viral capsid proteins from a plurality of serotypes in the sample.

[0010] In a further aspect of the invention, there is provided a method for determining a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample, the method comprising: a) denaturing the viral capsids in the sample; b) performing HILIC-MS (hydrophilic interaction liquid chromatography-mass spectrometry) on the sample comprising the denatured viral capsids; c) determining the masses of one or more viral capsid proteins from a plurality of serotypes in the sample; and d) determining the two or more serotypes in the sample by comparing the determined masses of step c) with theoretical masses of viral capsid proteins for the serotypes.

[0011] In a further aspect, there is provided a method of detecting the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of adenovirus or adeno-associated virus, the method comprising: a) denaturing the viral capsids in the sample; b) performing HILIC-MS (hydrophilic interaction liquid chromatography-mass spectrometry) on the sample comprising the denatured viral capsids; and c) determining the masses of one or more viral capsid proteins in the sample; and d) comparing the determined masses of step c) with theoretical masses of viral capsid proteins for the serotype wherein multiple determined masses for the same viral capsid protein indicate the presence of heterogeneous viral capsid proteins.

[0012] In a further aspect, there is provided use of LC-MS to determine a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample.

[0013] In a further aspect, there is provided use of LC-MS to detect the presence of a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample.

[0014] In a further aspect, there is provided use of LC-MS to detect the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of adenovirus or adeno-associated virus

[0015] Further aspects of the invention can be understood from the accompanying claims and description.

[0016] Detailed description

[0017] Adenovirus and adeno-associated virus

[0018] Adenoviruses (Ads) and adeno — associated viruses (AAVs) are non-enveloped DNA viruses. The viruses are recombinant viruses. For example, recombinant viruses for use in gene therapy.

[0019] Method

[0020] The methods enclosed can be used to detect the presence of a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample, or more specifically to determine a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample. The method of detecting their presence can be carried out prior to the more specific method of determining what the viruses are.

[0021] That is, the method can comprise detecting the presence of a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample, the method comprising: a) denaturing the viral capsids in the sample; b) performing HILIC-MS (hydrophilic interaction liquid chromatography-mass spectrometry) on the sample comprising the denatured viral capsids; and c) determining the masses of one or more viral capsid proteins from a plurality of serotypes in the sample; optionally wherein the method further comprises determining the plurality of serotypes by: d) comparing the determined masses of step c) with theoretical masses of viral capsid proteins for the serotypes.

[0022] The methods also include the use of the same method of HILIC-MS to differentiate between heterogeneous capsid proteins within a serotype. That is, a method of detecting the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of adenovirus or adeno-associated virus.

[0023] The definitions below apply to all of these methods unless indicated otherwise.

[0024] Serotype

[0025] Ads and AAVs have various serotypes. A serotype refers to a variation in viral capsid proteins that is detectable by specific antibody binding. The antibody binding is capable, in certain instances, to prevent the infection of cells by biding to the virus.

[0026] There are at least 57 Ads serotypes Ad1-Ad57 which form 7 species A-G. The main target of Ad serotype specific antibodies is the hexon protein. Hexon is the largest and most abundant of the structural proteins in the Ad capsid. The 4 and 5 serotypes of Ad are the most readily used for human gene therapy. AAV serotypes include AAV1-13 and AAV hu37 which are wild-type serotypes. In addition to these wild-type serotypes, hybrid vectors (man-made variants) are also available; an example of which includes AAV DJ. Depending on their serotype, AAVs can have specific tropism for specific organs and tissues. The serotypes in the sample can comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12. Those most commonly used for gene therapy include AAV2 and AAV8.

[0027] Sample

[0028] Where the sample is for detecting the presence or determining a plurality of serotypes, the sample comprises at least two serotypes. These methods are for detecting (or identifying) two or more serotypes in a sample containing two or more serotypes. That is, it is to detect and / or identify different serotypes in a sample. The sample can comprise anyone one serotype and a distinct serotype from the first serotype. The sample can comprise anyone one of the serotypes 1 - 12 and a distinct serotype from the first serotype selected from 1- 12. The sample can combinations of serotypes 2 and 6, 2 and 8, 3 and 6, 6 and 9, 8 and 9, for instance.

[0029] In certain instances, the sample contains three serotypes. In such instances, each of the serotypes is distinct from the other. The sample can comprise anyone one of three serotypes 1 - 12, each one of the three serotypes being distinct from the others. An example of such a sample is one that has serotypes 2 and 6 and 9.

[0030] In other instances, the sample may contain more than one serotype, but only one serotype is detectable by the methods disclosed herein. In still other instances, the sample may contain more than two serotypes, but only two serotypes are detectable by the methods disclosed herein. In still further instances, the sample may contain more than three serotypes, but only three serotypes are detectable by the methods disclosed herein.

[0031] In certain instances, the sample contains only two serotypes. In other instances. The sample contains only three serotypes. In some instances, the sample only contains serotypes 2 and 6, 2 and 8, 3 and 6, 6 and 9, or 8 and 9.

[0032] The method for detecting the presence of heterogeneous viral capsid proteins uses a sample comprising at least one serotype. This method is for detecting heterogeneity within a serotype. The sample can be from a manufacturing batch with the method being used for quality control of the manufacturing batch. A sample can be the end product of a manufacturing batch, that is, the result of processing steps to purify AAV.

[0033] In other instances, the sample is a solution containing AAV, that has been subjected to previously to one or column chromatography steps. In other instances, the sample is a solution containing AAV, that has been subjected to previously to one or affinity column chromatography steps. In some instances, the sample is a solution containing AAV that has been subjected previously to one or more tangential flow filtration steps. In still other embodiments, the sample is a solution containing AAV that has been subjected previously to ultracentrifugation. In further embodiments, the sample is a solution containing AAV that was been subjected previously to at least two of the following: column chromatography, affinity chromatography, tangential flow filtration, or ultracentrifugation.

[0034] The sample can, in certain embodiments, be a solution taken at any step in a manufacturing process used to purify AAV.

[0035] To generate enough AAV for therapeutic purposes, the manufacturing process can begin from a cell culture with a volume of at least 25 litres, or 50 litres, or 100 litres, or 200 litres, or 1 ,000 litres, or 2,000 litres. More generally, these volumes can be expressed as 25-2000 litres and the volumes within this range. Accordingly, the sample is a solution containing AAV that is being purified from a starting cell culture of 25-2000 litres, such as a volume of at least 25 litres, or 50 litres, or 100 litres, or 200 litres, or 1,000 litres, or 2,000 litres.

[0036] In some instances, the sample can be a solution suspected of containing AAV, that has been previously subjected to one more of the following: column chromatography, affinity chromatography, tangential flow filtration or ultracentrifugation. The sample can be a solution suspected of containing AAV, that is derived from a starting cell culture of 25-2000 litres, such as a volume of at least 25 litres, or 50 litres, or 100 litres, or 200 litres, or 1,000 litres, or 2,000 litres.

[0037] Denaturation

[0038] The viral capsid is denatured to break apart the capsid into its component viral capsid proteins. Denaturation can be prior to loading onto the LC (liquid chromatography), for example by addition of acetic acid (or other volatile acid) to the sample. For efficiency, denaturation can be on-line mobile phase induced denaturation where the sample is loaded onto the LC. For example, the denaturation can be in the instrument where the sample is injected and put in contact with the mobile phase Mobile phase B. By the time the sample reaches the column, the viral capsid will be denatured into its substituent proteins. That is, the sample is not denatured prior to LC.

[0039] Viral capsid and viral capsid proteins

[0040] Ads have an icosahedral capsid composed of three major capsid proteins (hexon, penton base and fiber) and four minor proteins (Illa, VI, VIII and IX).

[0041] AAVs have an icosahedral, 60-mer capsid composed of three viral proteins (VPs) VP1, VP2 and VP3 in an approximate 1:1:10 ratio. The viral capsid proteins can be VP1 , VP2 and / or VP3.

[0042] Generally, AAV is a proteinaceous capsid encapsulating a single-stranded DNA molecule. Capsids are at times categorized by the amount of DNA that reside within them; filled, partially filled, empty. Whether a capsid is filled, partially filled or empty can be assessed by techniques such as cryo-electron microscopy and mass spectrometry and separated through ultracentrifugation.

[0043] By heterogeneous viral capsid proteins is meant there is more than one isoform of viral capsid protein within a serotype. For example, where the virus is adeno-associated virus and the serotype is AAV8, there can be non-modified VP3 (by non-modified is meant not modified by post-translational processes) as well as phosphorylated VP3. That is, the sample contains two isoforms of VP3. Other types of heterogeneity (other isoforms) can include glycosylated and non-glycosylated capsid proteins.

[0044] HILIC-MS

[0045] HI LIC (hydrophobic interaction liquid chromatography) can be performed on any suitable liquid chromatography (LC) system. In order to provide a lowed detection limits and good compatibility with mass spectrometry, the LC system is preferably adapted to work at follow rates of less than about 0.5 mL / min. For example, the LC system can be an ultra-high- performance liquid chromatography (UHPLC) system, optionally adapted for use at a flow rate of from about 0.05 mL / min to about 0.3 mL I min (e.g. about 0.1 mL / min).

[0046] HI LIC uses hydrophilic stationary phases in conjunction with a mobile phase containing water and a less polar solvent, e.g. acetonitrile.

[0047] For the stationary phase, any polar chromatographic surface can be used. For example: simple unbonded silica silanol or diol bonded phases; amino or anionic bonded phases; amide bonded phases; cationic bonded phases or zwitterionic bonded phases.

[0048] The stationary phase can be an amide column.

[0049] The mobile phase for use in HILIC can be provided by mixing two mobile phases such as an acidified aqueous mobile phase (Mobile phase A) and an acidified organic phase (Mobile phase B).

[0050] Mobile phase A can comprise or consist of acidified water. For example, the mobile phase A can be from 98-99.9% water and, accordingly, 2-0.1% acid. In certain instances, the mobile phase A is 99.9% water and 0.1% acid. In other instances, the mobile phase A is 99.8% water and 0.2% acid. In still other instances, the mobile phase A is 99.7% water and 0.3% acid.

[0051] Mobile phase B can comprise or consist of acidified acetonitrile (ACN). For example, the mobile phase B can be from 98-99.9% water and, accordingly, 2-0.1% ACN. In certain instances, the mobile phase B is 99.9% water and 0.1% ACN. In other instances, the mobile phase B is 99.8% water and 0.2% ACN. In still other instances, the mobile phase B is 99.7% water and 0.3% acid.

[0052] Acidification is by including an acid which can be selected from formic acid or difluoroacetic acid (DFA) or trifluoroacetic acid (TFA) or trichloroacetic acid (TCIAA). The acid can be DFA. The acid (e.g. DFA) can be in a percentage of from about 0.05% v / v to about 0.15% v / v in both Mobile phase A and Mobile phase B. For example, the acid (e.g. DFA) can be about 0.1% v / v in both Mobile phase A and Mobile phase B.

[0053] The HILIC column can be initially equilibrated with mobile phase comprising a high level of Mobile phase B (such as >80% Mobile phase B, e.g. about 85% Mobile phase B) prior to loading the sample. For example, the HILIC column can be equilibrated with mobile phase comprising about 85% mobile phase B and 15% Mobile phase A (for example, about 85% acidified ACN and about 15% acidified water).

[0054] The temperature of the column can be from 20-50°C. In some instances, the column can be about 20°, 25°, 30°, 35°, 40°, 45°, or 50°C. In some instances, the column can be about 25°C. In other instances, the column can have a temperature of 45°C. Gradient

[0055] A single HILIC run to separate one or more viral capsid proteins typically comprises applying a sample to the HILIC column (e.g. by sample injection) and then applying an elution gradient. The gradient is formed by mixing Mobile phase B and Mobile phase A, varying the relative percentage of Mobile phase B and Mobile phase A over a specified time period to apply the gradient. The elution gradient can also comprise one or more isocratic holds, i.e. specified time periods where the ratio of the two mobile phases is held constant.

[0056] During sample injection the mobile phase preferably comprises mostly acidified organic mobile phase B such as > 80% Mobile phase B, e.g. about 85% Mobile phase B and 15% Mobile phase A. That is, the sample injection phase preferably consists of about 85% acetonitrile.

[0057] The elution of the viral capsid proteins can comprise changing from about 85% Mobile phase B and about 15% Mobile phase A at the start to about 60% Mobile phase B at the end (by end is meant to the point where one or more, for example all, the viral capsid proteins are eluted).

[0058] A first portion of the elution gradient can comprise an isocratic hold of the mobile phase, with the relative percentage of Mobile phase B and Mobile phase A as per the mobile phase during sample injection. The mobile phase in the first portion can comprise > 80% Mobile phase B, e.g. the first portion can comprise about 85% Mobile phase B and about 15% Mobile phase A, e.g. the first portion consists of about 85% acidified acetonitrile (acidification can be with approximately 0.1% DFA). The first portion can be a period of from about 0.1 min to about 2 min. For example, the first portion can be a period of from about 0.25 min to about 1 min. The first portion can be a period of about 0.5 min.

[0059] A second portion of the elution gradient can comprise a linear gradient or a step gradient from where Mobile phase B is > 80% of Mobile phase B to about 64%, for example, 68-60%, for example 66-62% (all percentages in the application are v / v). Where used, the linear can be over a period of from about 0.1 min to about 2 min. For example, the linear gradient can be over a period of about 0.5 min. The linear or step gradient can be from where Mobile phase B is > 80% to Mobile phase B of about 68-60%, for example 66-62%, for example about 64%. For example, about 85% ACN to about 64% ACN (both with about 0.1% DFA). For example, the gradient can comprise a linear gradient from about 85% acidified ACN to about 64% acidified ACN over about 0.1 min to about 2 min, for example about 0.5 minutes. The second portion can also comprise an isocratic hold after the linear or step gradient. For example, an isocratic hold at about 64% Mobile phase B, for example 68-60%, 66-62%, for example 64% acidified ACN. The isocratic hold can be for 5-10 minutes, for example about 7 minutes. Again all % are v / v.

[0060] A third portion of the elution gradient can comprise a linear gradient, where the percentage of Mobile phase A is further increased in comparison to the percentage of Mobile phase A at the end of the second portion and the percentage of Mobile phase B is decreased in comparison to the end of the second portion.

[0061] The third portion can comprise a linear gradient from about 64% Mobile phase B, for example 68-60%, or 66-62% Mobile phase B (e.g. 66-64% acidified ACN), for example 64% B at the start of the third portion to about 64-56% Mobile phase B or 62-58% or about 60 % Mobile phase B at the end of the third portion. This third portion is the portion which elutes the viral capsid proteins.

[0062] The third portion can be a period of from about 15 min to about 45 min. The third portion can be a period of from about 20 min to about 40 min. The third portion can be a period of from about 15 min to about 25 min. The third portion can be a period of about 20 min.

[0063] For example, the elution gradient can comprise the following:

[0064] • A first portion comprising an isocratic hold of the mobile phase of at least 80% Mobile phase B;

[0065] • A second portion comprising a linear gradient from at least 80% Mobile phase B to about 60-68% mobile phase B;

[0066] • A third portion comprising a linear gradient from about 60-68% mobile phase B to 56- 64% mobile phase B.

[0067] The linear gradient of the third portion can be a reduction of about 0.15-0.25% v / v mobile phase B / minute. For example, a gradient of about 64% B to about 60% B over about 21 minutes. The elution of the third portion can be used to elute VP1 , VP2 and VP3 where the virus sample comprises adeno-associated serotypes.

[0068] For example, the elution gradient can comprise the following:

[0069] • A first portion comprising an isocratic hold of the mobile phase at about 85% Mobile phase B; • A second portion comprising a linear gradient from about 85% Mobile phase B to about 64-66% mobile phase B;

[0070] • A third portion comprising a linear gradient from about 64-66% mobile phase B to 58- 60% mobile phase B.

[0071] The linear gradient of the third portion can be a reduction of about 0.15-0.25% v / v mobile phase B / minute. For example, a gradient of about 64% B to about 60% B over about 21 minutes. The elution of the third portion can be used to elute VP1 , VP2 and VP3 where the virus sample comprises adeno-associated serotypes.

[0072] An exemplary elution gradient can comprise the following:

[0073] • A first portion comprising isocratic hold of the mobile phase at about 85% mobile phase B (85% acidified ACN) and about 15% mobile phase A, for about 0.5 min; and I or

[0074] • A second portion comprising a linear gradient of from about 85% Mobile phase B and about 15% Mobile phase A to about 64% Mobile phase B and about 36% Mobile phase A, over a time period of about 0.5 min followed by a hold at 64% Mobile phase B and about 36% Mobile phase A for about 7 minutes; and

[0075] • a third portion comprising a linear gradient of from about 64% Mobile phase B and about 15% Mobile phase A to about 60% Mobile phase B and about 40% Mobile phase A, over a time period of about 13 min to about 28 min (e.g. about 21 min).

[0076] The elution can be used to elute VP1, VP2 and VP3 where the virus sample comprises adeno-associated serotypes. For example, a gradient comprising a reduction of about 0.15- 0.25% Mobile phase B / minute can be used to elute VP1, VP2 and VP3, for example, a reduction of about 0.15-0.25% Mobile phase B / minute. For example, a reduction of about 0.15-0.22% Mobile phase B / minute (% refers to v / v). For example, a reduction of about 0.19% Mobile phase B / minute. The starting Mobile phase B concentration for the reduction of Mobile phase B can be about 64-66%, for example about 64%. This can be a gradient from approximately 64% v / v B to about 60% v / v Mobile phase B. This can be over about 20 minutes. The same elution profiles described above can also be used to elute Ads viral capsid proteins.

[0077] For the calculation of volume / volume, the following example can be used: By 64% v / v Mobile phase B is meant for every 100ml of mobile phase, 64ml acetonitrile and 36ml aqueous mobile phase A. Where DFA is used: 64ml acetonitrile, 35.9ml aqueous mobile phase, e.g. water, and 0.1ml DFA. Determining the masses of one or more viral capsid proteins

[0078] The eluate of the HILIC separation that comprises the viral capsid proteins is subject to mass spectrometric analysis to determine the masses of one or more viral capsid proteins (e.g. for each serotype when determining a plurality of serotypes). The mass spectrometric analysis comprises obtaining mass spectrometric (m / z) data using a mass spectrometer and subjecting the m / z data to a deconvolution process to determine the mass of each of the one or more viral capsid proteins.

[0079] The mass spectrometer can comprise an ion source that is compatible with HILIC. An example of such an ion source is an electrospray ion (ESI) source, preferably an ESI that can be adjusted in the X, Y and Z directions to provide a low detection limit.

[0080] The mass spectrometer can comprise a high resolution and high mass accuracy mass analyzer. The resolution and mass accuracy are preferably sufficient to allow accurate differentiation between exemplary AAV VPs (e.g. with mass as illustrated in Table 1 below). Examples of suitable mass analyzers include an Orbitrap mass analyzer, a Fourier-transform ion cyclotron resonance (FTICR) mass analyzer, a time-of-f light mass analyser, and the like.

[0081] The mass analyzer can have a resolution of at least about 10,000 (m / Am, 10% valley definition) over the measured mass range. For example, the mass analyzer can have a resolution of at least about 15,000 (m / Am, 10% valley definition) over the measured mass range.

[0082] The mass accuracy of the mass analyzer can be specified using an upper limit for the mass error. For example, the mass error can be less than about 10 ppm for the measured mass values.

[0083] The mass spectrometry can comprise full scan mass spectrum collection over the measured mass range. The measured mass range can be from about m / z 600 to about m / z 2,500. For example, the measured mass range can be from about m / z 850 to about m / z 2,100.

[0084] The m / z data can be subjected to the deconvolution process using any suitable software package, such as BiopharmaFinder 5.1 (Thermo Fisher Scientific, Inc.). After HI LIC but before the mass spectrometric analysis the HILIC sample can be concentrated, i.e. to increase the viral capsid concentration.

[0085] Comparing the determined masses of step c) with theoretical masses of viral capsid proteins

[0086] Theoretical masses are calculated based on the amino acid sequences of VPs. This can be done routinely by any software available to the skilled person, e.g. Biopharma Finder software. Theoretical masses for various AAV VPs are provided below in Table 1. Others could be readily calculated by the skilled person in the art. Theoretical masses for Ads viral capsid proteins could be readily calculated by the skilled person in the art, for example by reference to Benevento et al (2014), Journal of Biol. Chem. Volume 289, Issue 16, 18 April 2014, Pages 11421-11430.

[0087] Table 1 : Theoretical masses of exemplary AAV VPs

[0088] As can be seen, theoretical masses can also be calculated based on biochemical modifications which can occur during post-translational processing such as phosphorylation, oxidation or glycosylation to form different isoforms. Theoretical masses based on these further modifications can be used to determine heterogeneity of viral capsid proteins within one serotype. For example, if the virus is adeno-associated virus and there is more than one type of VP1 (e.g. glycosylated and non-glycosylated forming two isoforms of VP1).

[0089] Confirming the identity of the viral capsid proteins by matching the extraction ion chromatograms to total ion count (TIC)

[0090] In addition to confirming the identity of the AAV VPs by comparing the determined masses of the viral capsid proteins to the corresponding theoretical masses, the identity of the viral capsid proteins can be confirmed from the elution time of each given viral protein determined in the HILIC MS experiment.

[0091] As the skilled person will appreciate, the total ion count (TIC) as a function of time obtained for a HILIC-MS of a sample represents the overall chromatogram. Matching the extracted ion chromatogram for the (optionally deconvoluted) m / z data for a given viral protein to the corresponding time period on the TIC provides the elution I retention time of the said given viral protein. This retention time will be reproducible for the given viral protein under the HILIC conditions used in the HILIC-MS experiment. Accordingly, where the retention time is known for given viral proteins, retention time (and extracted ion chromatogram) can provide an additional confirmation of the identity of the viral proteins.

[0092] Use

[0093] The above definitions and features apply to the use of LC-MS to detect the presence of a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample; or to determine a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample; or to detect the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of adenovirus or adeno-associated virus. The LC-MS can be HILIC-MS. The viruses in the sample can be adeno-associated viruses. That is, use of HILIC-MS to detect the presence or to determine of a plurality of serotypes of adeno-associated viruses in a sample.

[0094] Throughout the specification, unless the context demands otherwise, the terms ‘comprise’ or ‘include’, or variations such as ‘comprises’ or ‘comprising’, ‘includes’ or ‘including’ will be understood to imply the method or kit includes a stated integer or group of integers, but not the exclusion of any other integer or group of integers.

[0095] Each document, reference, patent application or patent cited in this text is expressly incorporated herein in their entirety by reference, which means it should be read and considered by the reader as part of this text. That the document, reference, patent application or patent cited in the text is not repeated in this text is merely for reasons of conciseness. Reference to cited material or information contained in the text should not be understood as a concession that the material or information was part of the common general knowledge or was known in any country.

[0096] Description of the Figures

[0097] Figure 1 shows a general workflow used for AAV capsid protein identification by LC-MS.

[0098] Figure 2 shows FLR chromatograms of empty AAV2 (A), empty AAV8 (B), and a mixture of AAV2:AAV8 at a particle ratio of 4:10 (C) obtained from the injection of 5 pL samples.

[0099] Figure 3 shows FLR chromatograms obtained from the injection of 5 pL of AAV2:AAV8 binary mixtures at particle ratios of 1.6: 1.0 (A) and 1.2: 1.0 (B). Note that the mixture includes full AAV2 (QtC) and empty AAV2

[0100] Figure 4 shows AAV2 / AAV8 mixture LC-MS data processed in BiopharmaFinder 5.1.

[0101] Figure 5 shows AAV2 VP3 ID (BiopharmaFinder 5.1) in the 1.6: 1.0 AAV2 / AAV8 mixture with AAV2 VP3 identified across 3 replicates.

[0102] Figure 6 shows AAV2 VP2 ID (BiopharmaFinder 5.1) in the 1.6: 1.0 AAV2 / AAV8 mixture with AAV2 VP2 identified across 3 replicates.

[0103] Figure 7 shows AAV2 VP1 ID (BiopharmaFinder 5.1) in the 1.6: 1.0 AAV2 / AAV8 mixture with AAV2 VP1 identified across all 3 replicates.

[0104] Figure 8 shows AAV8 VP3 ID (BiopharmaFinder 5.1) in the 1.6: 1.0 AAV2 / AAV8 mixture with AAV8 VP3 identified across 3 replicates.

[0105] Figure 9 shows AAV2 / AAV8 mixture LC-MS data processed in BiopharmaFinder 5.1.

[0106] Figure 10 shows LC-FLR analysis of an AAV2:AAV8 mixture (0.9: 1.0). FLR chromatograms obtained from the injection of 5 pL of AAV2:AAV8 binary mixture at particle ratios of 0.9: 1.0. The mixture includes full AAV2 (QtC) and empty AAV8.

[0107] Figure 11 shows LC-FLR analysis of an AAV6:AAV9 mixture (0.1:0.9). FLR chromatograms obtained from the 6 pL injection of the AAV6:AAV9 binary mixture at particle ratios of 0.1:0.9. The mixture includes full AAV9 (QtC) and empty AAV6 (Virovek). Figure 12 shows LC-FLR analysis of an AAV8:AAV9 mixture (0.25:0.75). FLR chromatograms obtained from the 6 pL injection of the AAV8:AAV9 binary mixture at particle ratios of 0.25:0.75. The mixture includes full AAV9 (QtC) and empty AAV8 (Virovek).

[0108] Figure 13 shows LC-MS analysis of an AAV2:AAV6 mixture (0.94:0.06). Total Ion Count (TIC) chromatograms obtained from the 6 pL injection of the AAV2:AAV6 binary mixture at particle ratios of 0.94:0.06. The mixture includes full AAV2 (QtC) and empty AAV6 (Virovek).

[0109] Figure 14 shows LC-MS analysis of an AAV2:AAV9 mixture (0.94:0.06). Total Ion Count (TIC) chromatograms obtained from the 6 pL injection of the AAV2:AAV9 binary mixture at particle ratios of 0.94:0.06. The mixture includes full AAV2 (QtC) and empty AAV9 (Virovek).

[0110] Figure 15 shows LC-MS analysis of an AAV6:AAV8 mixture (0.94:0.06). Total Ion Count (TIC) chromatograms obtained from the 6 pL injection of the AAV6:AAV8 binary mixture at particle ratios of 0.94:0.06. The mixture includes empty AAV6 (Virovek) and full AAV8 (Virovek).

[0111] Figure 16 shows LC-MS analysis of an AAV2:AAV6:AAV9 ternary mixture (0.86:0.07:0.07). Total Ion Count (TIC) chromatograms obtained from the 6 pL injection of the AAV2:AAV6:AAV9 ternary mixture at particle ratios of 0.86:0.07:0.07. The mixture includes full AAV2 (QtC):empty AAV6 (Virovek) and full AAV9 (Virovek).

[0112] Figure 17 shows LC-MS analysis of an Adenovirus serotype 5 Clarified Bulk Harvest sample (in-house). Total Ion Count (TIC) chromatogram obtained from the 7 pL injection of the Adenovirus 5 sample. The inset chromatogram depicts the region of interest for the elution of Ad5 viral proteins.

[0113] Figure 18 shows LC-MS analysis of an Adenovirus serotype 5 Clarified Bulk Harvest sample (in-house). Total Ion Count (TIC) chromatogram obtained from the 7 pL injection of the Adenovirus 5 sample and Extracted Ion Chromatogram (XIC) of Hexon-ll Ad5 structural protein. Raw mass spectra and deconvoluted mass spectra for Hexon-ll protein also shown.

[0114] Figure 19 shows LC-MS analysis of an Adenovirus serotype 5 Clarified Bulk Harvest sample (in-house). Total Ion Count (TIC) chromatogram obtained from the 7 pL injection of the Adenovirus 5 sample and Extracted Ion Chromatogram (XIC) of Penton-Ill Ad5 structural protein. Raw mass spectra and deconvoluted mass spectra for Penton-Ill protein also shown.

[0115] Figure 20 shows LC-MS analysis of an Adenovirus serotype 5 Clarified Bulk Harvest sample (in-house). Total Ion Count (TIC) chromatogram obtained from the 7 pL injection of the Adenovirus 5 sample and Extracted Ion Chromatogram (XIC) of Fiber-IV Ad5 serotype specific protein. Raw mass spectra and deconvoluted mass spectra for Fiber-IV protein also shown.

[0116] Examples Aspects of the present invention will now be illustrated by way of example only and with reference to the following experimentation.

[0117] An outline of the general method can be seen in Figure 1.

[0118] Materials

[0119] 1. Equipment / Software

[0120] • Thermo Vanquish LIHPLC system with Binary pump H (VH-P10-A), Sampler Module (VH-A10-A), Column Compartment (VH-C10-A), and Fluorescence Detector (VF-D50- A).

[0121] • Thermo Orbitrap MS Exploris 240

[0122] • Waters Acquity LIPLC Glycoprotein BEH Amide Column, 300A, 1.7pm, 2.1 x 150 mm

[0123] • Chromeleon 7.3.1 and Biopharma Finder 5.1

[0124] 2. Materials and Reagents

[0125] • LC-MS grade lonHance difluoroacetic acid

[0126] • Optima™ LC-MS Grade acetonitrile (referred to as B below)

[0127] • Optima™ LC-MS Grade water

[0128] 3. Samples (particle titer determined by SEC-MALS)

[0129] • AAV2 Material Empty (8.77E+12 capsids (cp) / mL)

[0130] • AAV2 Quick-to-Clinic Pilot run 1 (6.80E+12 cp / mL)

[0131] • AAV2 Quick-to-Clinic Pilot run 2 (6.47 E+12 cp / mL)

[0132] • AAV6 Material Empty (2.49E+13 cp / mL)

[0133] • AAV8 Material Empty (2.10E+13 cp / mL)

[0134] • AAV8-CMV-GFP Material (1.57E+13 cp / mL)

[0135] • AAV9-CMV-GFP Material (3.16E+13 cp / mL)

[0136] • AAV9 Quick-to-Clinic Confirmation run 5 (1.86 E+13 cp / mL)

[0137] Example 1 : Liquid Chromatography-Mass Spectrometry can distinguish between serotypes in a mixed serotype sample

[0138] To determine whether AAV serotypes can be distinguished in a sample having more than one serotype, the methodology following was used. Five microliters of AA2:AAV8 mixtures at particle ratios of 1.6: 1.0 and 1.2: 1.0 were separated using a Thermo Vanquish UHPLC system with Binary pump H (VH-P10-A), Sampler Module (VH-A10-A), Column Compartment (VH-C10-A). The liquid chromatography (LC) settings were set as follows: Column temperature 25°; autosampler temperature 5°C; fluorescence detection (excitation wavelength 280 nm, emission wavelength 348 nm, and sensitivity 3). Mobile phase A = acidified water. Mobile phase B = acidified 100% acetonitrile. Acidification is due to adding DFA 0.1% v / v to Mobile phase A and Mobile phase B. The gradient was as follows a flow rate of 0.1 mL / min: Isocratic hold at 85% B (acetonitrile) over 0.5 min followed by a linear decrease to 64 % B over 0.5 mins followed by an isocratic hold at 64% B for over 7 minutes. An elution gradient was then performed for 21.3 mins by decreasing mobile phase B from 64% to 60%. A column wash was further conducted by an isocratic hold of mobile phase B to 5% over 6.67 mins. Column was then re-equilibrated by increasing mobile phase B to the initial conditions (85% B) over 14.5 min. Separated AAV viral proteins were detected in both fluoresce and MS domains.

[0139] Mass Spectrometry (MS) settings were set as follows: The ion source parameters included the H-ESI ion source with a static spray voltage of 2700 V, sheath gas 20 arb, aux gas 5 arb, and sweep gas 0 arb. Ion transfer tube temp and vaporizer temp were selected at 320°C and 150°C, respectively.

[0140] Full scan MS analysis was conducted in the mass range of m / z 850 - 2100 with an Orbitrap resolving power of 15,000 at m / z 200, RF lens of 125%, normalized AGC target 50%, maximum injection time 200 ms, 10 microscans, and in-source fragmentation voltage of 35 V.

[0141] Data analysis

[0142] MS data analysis was performed in BiopharmaFinder 5.1. Briefly, theoretical masses and potential Post Translational Modifications (PTMs) of AAV VPs are defined in BiopharmaFinder 5.1 based on VPs amino acid sequences. A multiconsensus analysis is performed on the raw data files collected. A mass deconvolution of the experimental raw MS spectra is performed resulting in a list of identified proteoforms. A further comparison of VP theoretical and experimental masses permits a reliable fingerprint of AAV serotypes via accurate identification of VPs (mass error < 10 ppm).

[0143] All VPs from AAV serotypes 2 and 8 elute in the range 64% B to 60% B over 21 minutes.

[0144] Theoretical masses can be derived from the table below for various AAV serotypes. Theoretical masses are calculated based on the amino acid sequences of VPs. This is done by importing sequences in Biopharma Finder software. The user can also predefine the number and type of potential post translational modifications to consider:

[0145] Table 1.

[0146] Results:

[0147] The results are shown in Figures 2A-C. A clear separation of AAV2 and AAV8 VPs in the LC domain using the optimized method is shown in Figures 2A and 2B, respectively. The intact mass analysis of the raw MS data confirmed the identity of VP1 , VP2, and VP3 in both serotypes (see Figure 2A and 2B inset tables). We further pushed the capabilities of our LC-MS method by analyzing a mixture of empties AAV2:AAV8 at a particle ratio of 0.4:1 . The FLR chromatogram depicted in Figure 2C shows that the described method can accurately discriminate between both serotypes.

[0148] Example 2: LC-MS can distinguish different serotypes across a variety of ratios in the original sample

[0149] The capabilities of the method for fingerprinting full AAV2 in AAV2:AAV8 mixtures at different particle ratios, were also evaluated.

[0150] Results:

[0151] The results are shown in Figures 3-10.

[0152] The FLR chromatograms depicted in Figures 3A and 3B evidenced the effectiveness of the proposed method for separating and detecting full AAV2 in mixtures of serotypes 2 and 8 at particle ratios of 1.6: 1.0 and 1.2: 1.0, respectively. The analysis of MS raw data confirmed the identity of the AAV2 VPs in both mixtures. However, AAV8 VP3 was the only VP identified in the analyzed mixtures, suggesting that VP1 and VP2 abundances were near or below the limit of identification of the method. The results presented here demonstrate the discrimination of AAV in mixtures of two serotypes using intact mass LC-MS. Slight modifications of the current workflow could enable the identification of AAVs in mixtures of 3 or more serotypes.

[0153] Separation of the viral proteins across three replicates of an AAV2 and AAV8 mixture (1.6: 1.0) is shown in the TIC chromatograms of Figure 4. The gradient conditions allowed for the early elution of AAV8 VP3s at 13-15 mins while AAV2 VP3 main proteoform elutes later at 19 minutes. Despite that AAV8 VP1 and VP2 could not be identified in the MS due to low ion signals, the elution of these two species were confirmed in the FLR chromatograms (see Figure 3). The deconvoluted mass spectra retrieved from Biopharma Finder 5.1 show the identification of all AAV2 VPs and VP3 from AVV8 serotype across the three analyzed replicates (Figure 4). A detailed inspection of Figures 5 - 8 evidenced that VPs are accurately identified with assignment errors below 10 ppm, clear matching of extracted ion chromatograms (XIC) with TICs, and over 30 charge states matched for all VPs.

[0154] Similar results were obtained for the AAV2:AAV8 mixture (1.2: 1.0). That is, AAV8 VP3 and AAV2 VP1 , VP2, and VP3 can be separated and identified through their intact mass with assignment errors below 10 ppm. However, AAV8 VP1 and VP2 were not identified in the MS domain (see Figure 9). A third AAV2:AAV8 mixture (0.9: 1.0) analyzed using the same optimized LC-MS method revealed distinct results. The unexpected drop in signal for AAV8 VP3 observed in the FLR chromatogram shown in Figure 10 suggests a potential issue during sample preparation.

[0155] The analysis of an AAV6:AAV9 binary mixture (0.1 :0.9) evidenced the suitability of the described LC-MS method for fingerprinting a targeted AAV9 serotype even in the presence of AAV6 at 10% residual levels (Figure 11). These results effectively demonstrate the versatility of the current LC-MS method, highlighting its applicability not only for drug product characterization, but also as a reliable residual assay.

[0156] Results obtained for AAV9:AAV8, AAV2:AAV6, AAV2:AAV9, and AAV6:AAV8 binary mixtures using both a modified elution gradient (mobile phase B from 65% to 61% along 21.3 mins) and higher column temperature, 45 °C, are depicted in Figures 12-15. Moreover, a ternary mixture of AAV2, AAV6, and AAV9 (0.86:0.07:0.07) analyzed using the modified LC conditions confirmed the suitability of the method for the identification of a targeted AAV serotype even in a complex mixture with two other AAV serotypes (Figure 16).

[0157] Overall, it can be seen that the method provides a sensitive way of detecting and differentiating between not only serotypes but their ratios in samples.

[0158] Example 3: HILIC based LC-MS method can identify serotype-specific Fiber protein of Human Adenovirus 5 in a Clarified Bulk Harvest sample

[0159] LC settings were set as follows: Column temperature 25°; autosampler temperature 5°C. Mobile phase A = water and DFA 0.1% v / v. Mobile phase B = 100% acetonitrile and DFA 0.1% v / v. Single detection via mass spectrometry was selected.

[0160] Several ion signals in the retention time range 10-35 mins are observed in the Total Ion Count (TIC) chromatogram of Figure 17. Further deconvolution of raw MS spectra allowed for the identification of the Ad5 three main structural proteins, Hexon, Penton, and Fiber, with mass errors below 10 ppm (Figures 18-20). Since Fiber proteins from different Adenovirus variants have distinct amino acid sequences, their molecular composition can be reliably used for serotype identification. Consequently, the accurate assignment of Ad5 Fiber protein highlighted in Figure 20 demonstrates fingerprinting Human Adenovirus serotypes using the described LC-MS method.

Claims

CLAIMS1. A method for detecting the presence of a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample, the method comprising: a) denaturing the viral capsids in the sample; b) performing HILIC-MS (hydrophilic interaction liquid chromatography-mass spectrometry) on the sample comprising the denatured viral capsids; and c) determining the masses of one or more viral capsid proteins from a plurality of serotypes in the sample.

2. A method for determining a plurality of serotypes of adenoviruses or adeno- associated viruses in a sample, the method comprising: a) denaturing the viral capsids in the sample; b) performing HILIC-MS (hydrophilic interaction liquid chromatography-mass spectrometry) on the sample comprising the denatured viral capsids; and c) determining the masses of one or more viral capsid proteins from a plurality of serotypes in the sample; and d) determining the two or more serotypes in the sample by comparing the determined masses of step c) with theoretical masses of viral capsid proteins for the serotypes.

3. The method of claim 2, wherein the viruses are adeno-associated viruses and step d) compares: a) the masses of two or more of: VP1 , VP2 and VP3, for two or more serotypes; or b) the masses of VP1, VP2 and VP3 for two or more serotypes.

4. A method of detecting the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of adenovirus or adeno-associated virus, the method comprising: a) denaturing the viral capsids in the sample; b) performing HILIC-MS (hydrophilic interaction liquid chromatography-mass spectrometry) on the sample comprising the denatured viral capsids; and c) determining the masses of one or more viral capsid proteins in the sample; and d) comparing the determined masses of step c) with theoretical masses of viral capsid proteins for the serotype wherein multiple determined masses for the same viral capsid protein indicate the presence of heterogeneous viral capsid proteins.

5. The method of claim 1 or claim 4, wherein the virus is adeno-associated virus and the viral capsid proteins are VP1, VP2 and / or VP3.

6. The method of any of the preceding claims, wherein the identity of the viral capsid proteins is additionally confirmed by matching the extraction ion chromatograms to total ion count (TIC).

7. The method of any of the preceding claims wherein denaturation of the viral capsids is by on-line mobile phase induced denaturation.

8. The method of any of the preceding claims wherein the HILIC is performed using an amide column.

9. The method of any of the preceding claims, wherein the temperature of the LC column is 20-30°C, preferably 25°C.

10. The method of any of the previous claims wherein the HILIC uses: a) an aqueous mobile phase (mobile phase A); and b) a mobile phase B comprising acetonitrile.

11. The method of claim 10 wherein mobile phase A and mobile phase B comprise difluoroacetic acid (DFA).

12. The method of any of claims 10-11, wherein a gradient comprising the reduction of about 0.15-0.25% v / v mobile phase B / minute is used to elute the one or more viral capsid proteins from the HILIC column.

13. The method of claim 12, wherein the viruses in the sample are adeno-associated viruses and the gradient comprising the reduction of about 0.15-0.25% v / v mobile phase B / minute is used to elute VP1, VP2 and VP3 from the HILIC column.

14. The method of any of claims 12-13 wherein the concentration of mobile phase B at the start of the gradient is about 68-60% v / v.

15. The method of claim 14, wherein a gradient of about 68-60% v / v mobile phase B to about 64-56% v / v mobile phase B is used.

16. The method of claim 15, wherein the gradient is over about 20 minutes.

17. The method of any of claims 12-16, wherein prior to the gradient the column is equilibrated with about 80-85% v / v mobile phase B.

18. The method of claim 17, wherein following sample injection there is an isocratic hold of mobile phase B at about 80-85% v / v mobile phase B.

19. The method of any of the preceding claims, wherein the MS is performed on a mass spectrometer having a resolution of at least 15,000 (m / Am, 10% valley definition) over the measured mass range.

20. The method of any of the preceding claims, wherein the MS is configured to have a mass error of less than 10 ppm for the measured mass values.

21. Use of LC-MS to detect the presence of a plurality of serotypes of adenoviruses or adeno-associated viruses in a sample.

22. Use of LC-MS to determine a plurality of serotypes of adenoviruses or adeno- associated viruses in a sample.

23. Use of LC-MS to detect the presence of heterogeneous viral capsid proteins in a sample comprising a serotype of adenovirus or adeno-associated virus.

Images