Modified AAV Capsid Proteins for Treatment of Arthritic Disease

Abstract

The invention relates to recombinant adeno-associated virus (rAAV) virions for gene therapy, wherein the rAAV virions comprise a novel capsid protein. In particular, the invention relates to the use of such virions in gene therapy for the treatment of an arthritic disease, such as for example rheumatoid arthritis, or symptoms thereof, preferably by intraarticular administration.

Claims

1.-18. (canceled)

19. A method of treating or preventing an arthritic disease or of treating or preventing symptoms associated with an arthritic disease, the method comprising administering to a subject in need thereof a recombinant adeno-associated virus (rAAV) virion comprising a modified capsid protein and a promoter operably linked to a nucleotide sequence encoding a gene product of interest, wherein: (a) the gene product of interest is an IL-6 inhibitor; and (b) the modified capsid protein comprises in the C-terminal part of the protein an amino acid sequence Z, residues of which are exposed on the surface of the capsid protein, wherein the amino acid sequence Z: a. comprises or consists of a sequence of amino acid residues of the formula I: TABLE-US-00011 y-G-Q-x-G-(x).sub.3-R-(x).sub.3-y-A-Q-A-A wherein x represents a single amino acid residue and wherein y represents 0, 1 or 2 amino acid residues; and b. is present at a location corresponding to a position 100-200 amino acid residues from the C terminus of a wild-type AAV capsid protein.

20. The method according to claim 19, wherein the amino acid sequence Z is present at a location corresponding to a position 120-180 amino acid residues from the C terminus of a wild-type AAV capsid protein.

21. The method according to claim 20, wherein the amino acid sequence Z is present at a location corresponding to a position 130-170 amino acid residues from the C terminus of a wild-type AAV capsid protein.

22. The method according to claim 21, wherein the amino acid sequence Z is present at a location corresponding to a position 140-160 amino acid residues from the C terminus of a wild-type AAV capsid protein.

23. The method according to claim 19, wherein the IL-6 inhibitor is selected from the group consisting of an IL-6 receptor antagonist, a humanized anti-IL-6 monoclonal antibody, a chimeric anti-TL-6 monoclonal antibody, a humanized rabbit anti-IL-6 monoclonal antibody, and a soluble IL-6 receptor.

24. The method according to claim 23, wherein the IL-6 receptor antagonist is tocilizumab or sarilumab.

25. The method according to claim 23, wherein the humanized anti-IL-6 monoclonal antibody is olokizumab or sirukumab.

26. The method according to claim 23, wherein the chimeric anti-IL-6 monoclonal antibody is siltucimab.

27. The method according to claim 23, wherein the humanized rabbit anti-IL-6 monoclonal antibody is clazakizumab.

28. The method according to claim 19, wherein the promoter is a constitutive promoter or an inducible promoter.

29. The method according to claim 28, wherein the promoter is an NFκB responsive promoter.

30. The method according to claim 29, wherein the NFκB responsive promoter is an NFκB responsive CMV promoter.

31. The method according to claim 30, wherein the NFκB responsive CMV promoter is an NFκB responsive minimal CMV promoter.

32. The method according to claim 19, wherein the sequence Z comprises or consists of a sequence of amino acid residues of the formula II: TABLE-US-00012 EEEIxxxxPVATExxGxxxxNxQy-Z-(x).sub.nLPGMVWQxRDVYLQGPIW AKIPHTDG a. wherein Z, x and y are as defined in claim 2; and b. wherein n is 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 or 15.

33. The method according to claim 19, wherein the capsid protein comprises an amino acid sequence selected from the group consisting of: i) an amino acid sequence having at least 70% sequence identity with an amino acid sequence having SEQ ID NO: 1 and wherein amino acids at positions 588-602 of SEQ ID NO: 1 have at least 80% sequence identity with SEQ ID NO: 11, ii) an amino acid sequence having at least 70% sequence identity with an amino acid sequence having SEQ ID NO: 2 and wherein amino acids at positions 585-602 of SEQ ID NO: 2 have at least 80% sequence identity with SEQ ID NO: 10, iii) an amino acid sequence having at least 70% sequence identity with an amino acid sequence having SEQ ID NO: 3 and wherein amino acids at positions 587-601 of SEQ ID NO: 3 have at least 80% sequence identity with SEQ ID NO: 9, iv) an amino acid sequence having at least 70% sequence identity with an amino acid sequence having SEQ ID NO: 4 and wherein amino acids at positions 586-600 of SEQ ID NO: 4 have at least 80% sequence identity with SEQ ID NO: 8, v) an amino acid sequence having at least 70% sequence identity with an amino acid sequence having SEQ ID NO: 5 and wherein amino acids at positions 588-602 of SEQ ID NO: 5 have at least 80% sequence identity with SEQ ID NO: 9, vi) an amino acid sequence having at least 70% sequence identity with an amino acid sequence having SEQ ID NO: 6 and wherein amino acids at positions 588-602 of SEQ ID NO: 6 have at least 80% sequence identity with SEQ ID NO: 8, and vii) an amino acid sequence having at least 70% sequence identity with an amino acid sequence having SEQ ID NO: 7 and wherein amino acids at positions 587-604 of SEQ ID NO: 7 have at least 80% sequence identity with SEQ ID NO: 12, wherein the modified capsid protein provides for an at least two-fold increase in expression in comparison to an unmodified capsid protein with an amino acid sequence selected from the group consisting of SEQ ID NO: 13-19, when tested under the same conditions.

34. The method according to claim 33, wherein the modified capsid protein provides for an at least two-fold increase in expression in human FLS cells, in comparison to an unmodified capsid protein with an amino acid sequence selected from the group consisting of SEQ ID NO: 13-19, when tested under the same conditions.

35. The method according to claim 33, wherein the unmodified capsid protein has the amino acid sequence SEQ ID NO: 19 or has the same serotype as the modified capsid protein.

36. The method according to claim 33, wherein the capsid protein comprises or consists of an amino acid sequence selected from the group consisting of SEQ ID NO:1-7.

37. The method according to claim 36, wherein the capsid protein comprises or consists of SEQ ID NO: 4 or SEQ ID NO: 6.

38. The method of claim 19, wherein the rAAV virion is provided in a pharmaceutical composition, wherein the pharmaceutical composition comprises the rAAV virion and a pharmaceutically acceptable carrier.

39. The method of claim 38, the method further comprising administering an immunosuppressant to the subject.

40. The method of claim 19, wherein the arthritic disease is selected from the group consisting of rheumatoid arthritis (RA), juvenile rheumatoid arthritis, osteoarthritis (OA), gout, pseudogout, spondyloarthritis (SpA), psoriatic arthritis, ankylosing spondylitis, septic arthritis, arthritis, juvenile idiopathic arthritis, blunt trauma, joint replacement and Still's disease.

41. The method of claim 19, wherein the rAAV virion is administered systemically or locally.

42. The method of claim 41, wherein the local administration is intraarticular administration.

43. The method of claim 39, wherein at least one of the pharmaceutical composition and the immunosuppressant is administered locally.

44. The method of claim 43, wherein the local administration is intraarticular administration.

Description

DESCRIPTION OF THE FIGURES

[0476] The present invention will be discussed in more detail below, with reference to the attached drawings:

[0477] FIG. 1: Screening of capsid serotypes on HEK293T and FLS cells. Crude lysate containing 7 mutant capsid serotypes (plus AAV5) expressing yellow fluorescent protein (YFP) were used to transduce HEK293T cells or 3 different FLS cell lines (each from a different RA patient). After 72 hours (HEK293T) or 6 days (FLS), cells were assayed for percentage of cells expressing YFP by FLOW cytometry. Panel A shows % of HEK 293T cells expressing YFP; Panel B shows the % of YFP-expressing cells in 3 different FLS cell lines; Panel C shows the mean fluorescent intensity (MFI) in HEK293T cells; Panel D shows MFI in 3 different FLS cell lines (all cells); Panel E shows MFI in 3 different FLS cell lines (only positive population). The sample legend is depicted in Table 2.

[0478] FIG. 2: Capsid mutants show increased luciferase expression vs wt-AAV5 in FLS cells. Purified AAV (4 mutant serotypes or AAV5) expressing YFP-Luc fusion protein were used to transduce three different FLS cell lines from different RA patients: BB5498 (FLS 1), BB5540 (FLS 2) and BB7144 (FLS 3) using two MOIs (20000 or 100000 rAAV particles per cell). After 4 days, cells were lysed and luciferase expression was measured. Data is presented as absolute luciferase expression levels (RLU; white bars) or fold increase over AAV5 (black bars). Panel A shows FLS 1 at MOI 20K; Panel B shows FLS 1 at MOI 100K; Panel C shows FLS 2 at MOI 20K; Panel D shows FLS 2 at MOI 100K; Panel E shows FLS 3 at MOI 20K; and Panel F shows FLS 3 at MOI 100K. Open bars show luciferase (RLU) and filled bars show “fold increase” over AAV5. In a different experiment, three additional FLS cell lines from RA patients were transduced with AAV (7 mutant serotypes or AAV5) expressing luciferase: BB4308 (FLS 4), BX 1592 (FLS 5), BB4426 (FLS 6) using 2 MOIs (10K or 100K rAAV particles per cell). Panel G shows FLS 4 at MOI 10K; Panel H shows FLS 4 at MOI 100K; Panel I shows FLS 5 at MOI 10K; Panel J shows FLS 5 at MOI 100K; Panel K shows FLS 6 at MOI 10K; and Panel L shows FLS 3 at MOI 100K. Transduction efficacy of the 7 mutant serotypes or AAV5 (MOI 100K) was also evaluated in HEK293T cells (Panel M). Open bars show luciferase expression (RLU) and filled bars show “fold increase” over AAV5.

[0479] FIG. 3A: Capsid mutants exhibit increased gene expression in vivo. Two capsid mutants (AAV9-A2 and AAV7-A6) were compared with wtAAV5 using the air pouch synovium model. Luciferase-expressing vector was administered on day 0 following air pouch formation and luciferase expression was measured by live animal imaging (IVIS) on day 3 following transduction. Data shown is the luminescence (photon/second/square centimeter m2/steradian) in air pouch in mean+SEM.

[0480] FIG. 3B: In a second experiment, 5 selected capsid mutants (AAV1-P4, AAV7-A6, AAV9-A2, AAVrh10-A2, AAVrh10-A6) and wtAAV5 were injected into the knee joints of mice. A luciferase expressing vector was injected on day 0 and expression was measured by live imaging (IVIS) at indicated time points after administration. Data shown is the luminescence (photon/second/square centimeter m2/steradian)(left panel) in mean+SEM. **P<0.05, ***P<0.01, ****P<0.00001 vs. wtAAV5 at day 14. FIG. 3C: Fold increase vs. wtAAV5.

[0481] FIG. 4: CLUSTAL format alignment by MAFFT FFT-NS-I (v7.215). Below the alignment is a key denoting a conserved residue (*); and a non-conservative mutation ( ).

[0482] FIG. 5: CLUSTAL multiple sequence alignment by MUSCLE (3.8). Below the alignment is a key denoting a conserved residue (*); a conservative mutation (:); a semi-conservative mutation (.); and a non-conservative mutation ( ).

[0483] FIG. 6: CLUSTAL format alignment of inserts P4, A2, A6, P2 and QR-P2 (SEQ ID NO's: 8-12) by MAFFT FFT-NS-I (v7.215). Below the alignment is a key denoting a conserved residue (*); and a non-conservative mutation ( ).

[0484] FIG. 7: CLUSTAL multiple sequence alignment of inserts P4, A2, A6, P2 and QR-P2 (SEQ ID NO's: 8-12) by MUSCLE (3.8). Below the alignment is a key denoting a conserved residue (*); and a non-conservative mutation ( ).

[0485] FIG. 8: hTNFR2 expression levels in supernatant of EF1α-hTNFR-IgG1 AAV capsid mutants transduced HEK293T/17 cells.

SEQUENCE LISTING

[0486] Table 1 provides an explanation of the sequence references in correlation with the SEQ ID No's.

TABLE-US-00008 TABLE 1 Explanation of sequence references SEQ ID NO: serotype Modified capsid/insert/wild-type 1 AAV1 Modified capsid 2 AAV2 Modified capsid 3 AAV7 Modified capsid 4 AAV9 Modified capsid 5 AAVrh10 Modified capsid 6 AAVrh10 Modified capsid 7 AAV DJ-QR Modified capsid 8 Insert A2 Insert 9 Insert A6 Insert 10 Insert P2 Insert 11 Insert P4 Insert 12 Insert QR-P2 Insert 13 AAV1 Wild-type capsid 14 AAV2 Wild-type capsid 15 AAV7 Wild-type capsid 16 AAV9 Wild-type capsid 17 AAVrh10 Wild-type capsid 18 AAV DJ-QR Synthetic capsid 19 AAV5 Wild-type capsid

EXAMPLES

Example 1

Initial Screening of Capsid Library

1.1. Materials and Methods

[0487] 96-well plates spotted (and subsequently dried) with crude lysate containing AAV from 91 different AAV capsid serotypes were obtained from Dirk Grimm and Kathleen Börner at the University of Heidelberg. Each vector encoded a YFP transgene driven by a CMV promoter. As FLS are the primary target cells in the joint, an AAV capsid mutant library was screened for serotypes that show increased expression in human FLS isolated from joints of rheumatoid arthritis patients (RA-FLS) (as described in van de Sande M G et al., (2011) Ann Rheum Dis 70: 423-427). RA-FLS were plated (2500/well, 37° C./5% CO.sub.2) directly onto the spotted plates (DMEM-GlutaMAX-I (Gibco, ref. 31966-021), 10% FBS (heat inactivated (HI) Bovine Serum Gold, Gibco, ref A15-151), 10 mM HEPES (Gibco, ref. 15630-056), 50 μg/ml gentamycin (Gibco, ref. 15710-049), 100U/ml penicillin/100 μg/ml streptomycin (Sigma-Aldrich, ref.P0781) and all wells were visualized for YFP expression by fluorescence microscopy after 6 days.

1.2. Results

[0488] Transduction Efficacy of Capsid Mutants Vs. WT-AAV5 in FLS from RA Patients.

[0489] In screening of the 91 capsid mutants, while the overall expression levels were low, the present inventors identified 7 different serotypes that showed higher expression than wtAAV5: AAV9-A2, AAV7-A6, AAV1-P4, AAVDJ-QR-P2, AAVrh10-A6, AAVrh10-A2 and AAV2-P2 (amino acid sequences SEQ ID NO: 1-7; wtAAV5 SEQ ID NO: 19).

[0490] Crude lysates of all 7 vectors were used in an in vitro transduction assay in 3 different patient FLS cell lines and in HEK293T cells (example 2).

TABLE-US-00009 TABLE 2 Sample legend for FIG. 1 Capsid Position Sample serotype Insert/modified sequence insert in VP1 SEQ ID NO:  5 5 none none — 19 61 AAV1 GQSGNDVRSANAQAA P4 588-602  1 33 AAV9 GQRGNYSRGVDAQAA A2 586-600  4 34 AAVrh10 GQRGNYSRGVDAQAA A2 588-602  6 50 AAV2 QGQSGCDCRGDCFCA(QAA) P2 585-602  2 88 AAV-DJ-QR QGQRGCDCRGDCFCA(QAA) QR-P2 587-604  7 43 AAV7 GQRGNEARVREAQAA A6 587-601  3 46 AAVrh10 GQRGNEARVREAQAA A6 588-602  5

Example 2

Expression of Crude Lysates of 7 Selected Mutants

2.1. Materials and Methods

AAV Production

[0491] Details on the production of the crude AAV lysates can be found in Grosse et al. (J. Virol, 2017, doi: 10.1128/JVI.01198-17).

[0492] Aliquots of crude lysate for each of the selected 7 capsid mutants (plus wtAAV5 as a control) were used to transduce cells (HEK293T or 3 different FLS cell lines isolated from RA patients) and YFP expression was measured by flow cytometry 3 (HEK293T)-5 days (FLS) following transduction. In detail, HEK293T were seeded in a 96-well plate (Greiner Bio-One, ref. 655180) at 45000 cells per well. RA-FLS were seeded in a 96-well plate at 2500 cells per well. After overnight incubation, the cell supernatants were replaced with 40 μl DMEM-glutaMAX-I (Gibco 31966-021) containing 0.001% pluronic F68 solution (Sigma P5556). The virus lysates were added in duplo, 10 μl per well. After 4 hours, doxorubicin (final concentration 0.4 μM) (Sigma D1515) (Doxorubicin improves AAV transduction efficiency) in DMEM-glutaMAX-I containing FBS (heat inactivated (HI) Bovine Serum Gold, Gibco, ref A15-151), final concentration 1%) was added to the wells (50 μl per well). The day after, the medium of FLS was removed and DMEM-glutaMAX-I (10% FBS (heat inactivated (HI) Bovine Serum Gold, Gibco, ref A15-151), 10 mM HEPES (Gibco, ref. 15630), 50 μg/ml gentamycin (Gibco ref. 15710-049), 100 U/ml penicillin/100 μg/ml streptomycin (Sigma-Aldrich, ref. P0781)) was added (200 μl per well). The medium of HEK293T cells was not changed. Three (HEK293T cells) or 6 days (FLS) after transduction cells were trypsinized using 0.5% Trypsin/EDTA (Gibco ref. 15400-054) in PBS (Gibco, ref. 10010) and analyzed for YFP expression by FLOW cytometry (FACSCanto II, BD Biosciences). Both percentage of expressing cells and mean fluorescence intensity (MFI) for all cells was determined.

2.2. Results

[0493] Crude lysates of all 7 vectors were used in an in vitro transduction assay in 3 different patient FLS cell lines and in HEK293T cells. Cells were assayed for the percentage of cells expressing YFP by fluorescence microscopy (data not shown) or FLOW cytometry (FIG. 1 panels A-E). While there was some variability between cell types, all mutant capsids gave higher expression in both, FLS and HEK293T cells than AAV5-WT (FIG. 1). Table 2 provides the sample legend for FIG. 1. Based on these results, four capsid mutants were selected for further investigation (see example 3).

Example 3

In Vitro Testing of Capsid Variants in HEK293T and FLS

3.1 Materials and Methods

[0494] 3.1.1 Four of the mutant capsid proteins, AAV9-A2, AAV7-A6, AAV1-P4, and AAVDJ-QR-P2, were further investigated. Purified vector (lodixanol gradient) expressing a YFP-Luciferase fusion protein (to allow for visualization (YFP) as well as quantification by luciferase assay) was generated. Three different primary FLS lines isolated from rheumatoid arthritis patients (as described in van de Sande M G et al., (2011) Ann Rheum Dis 70: 423-427) were transduced with each serotype at 2 vector doses (MOI 20,000 or 100,000) and after 4 days, cells were harvested and gene expression was quantified by luciferase assay (Promega Luciferase assay Kit).

[0495] In detail, RA-FLS were plated at 2500 cell/well in a 96-well plate (Greiner Bio-One, ref. 655207) in medium (DMEM-GlutaMAX (Gibco ref. 31966-021), 10% FBS (heat inactivated (HI) Bovine Serum Gold, ref A15-151), 10 mM HEPES (Gibco ref. 15630-056), 50 μg/ml gentamycin (Gibco, ref 15710-049), 100u/ml penicillin/100 μg/ml streptomycin (Sigma-Aldrich Merck ref. P0781). After 48h, medium was removed and virus (in DMEM-Glutamax containing 0.001% Pluronic-68 (Sigma, ref. p5556)) was added at an MOI of 20,000 or 100,000. After 4h, medium containing Doxorubicin (Sigma, ref. D1515, final concentration 0.4 μM) and FBS (final concentration 1%) was added. 24h later, medium was replaced with DMEM-GlutaMAX, (10% FBS, 10 mM HEPES, 50 μg/ml gentamycin, 100u/ml penicillin, 100 μg/ml streptomycin). Four days post-transduction, cells were washed 1× with 100 μl PBS (Gibco, ref. 10010) and luciferase activity was determined using the ONE Glo™ luciferase assay system (Promega, ref. E6110): 100 μl Lysis buffer was added and cells were placed on a shaker for 10′, 900 rpm at RT. Subsequently, 20 μl lysate was transferred to a white 96-well plate, 80 μl substrate (was added for 3′ (dark) and luciferase activity was determined on a luminometer (1 sec/well, synergy HT, Biotek).

[0496] 3.1.2. In a similar experiment, three additional FLS cell lines isolated from rheumatoid arthritis patients were transduced with AAV5 and 7 capsid mutants from a different AAV preparation than described in 3.1.1 (AAV9-A2, AAV1-P4, AAV7-A6, AAVDJ-QR-P2, AAVrh10-A6, AAVrh10-A2, AAV2-P2) containing a luciferase gene (MOI 10,000 and 100,000). The number of empty particles differed between the AAV preparations. To exclude a possible effect on transduction efficacy, empty capsid correction was done by adding AAV5 empty particles to equalize the percentage of empty particles per preparation.

[0497] 3.1.3. The 7 capsid mutants from the same preparation as described in 3.1.2 were also tested in HEK293T cells. In detail, HEK293T were seeded in a 96-well plate (Greiner Bio-One, ref. 655180) at 50000 cells per well. After overnight incubation, the cell supernatants were replaced with DMEM-glutaMAX-1 (Gibco 31966-021) containing 0.001% pluronic F68 solution (Sigma P5556). The different vectors were added in duplo, at an MOI of 100,000. In this protocol, empty capsid correction was done as described for 3.1.2. After 4 hours, doxorubicin (final concentration 0.4 μM) (Sigma D1515) in DMEM-glutaMAX-I-containing FBS (heat inactivated (HI) Bovine Serum Gold, Gibco, ref A15-151), final concentration 1%, was added to the wells. Three days after transduction, cells were harvested and gene expression was quantified by luciferase assay (Promega Luciferase assay Kit) on a luminometer (BMG Labtech Fluostar Omega).

3.2. Results

[0498] 3.2.1 In vitro transductions of three different FLS cell lines were performed using recombinant AAV comprising one of the 4 mutant capsids (as well as AAV5 as control, made in the identical manner) following the protocol described in 3.1.1. All 4 serotypes showed increased expression levels when compared with AAV5, ranging from 2-fold to 35-fold increases, depending on the serotype and cell line used (FIG. 2A-F).

[0499] 3.2.2 In another series of experiments, in vitro transduction efficacy of 7 mutant capsids (as well as AAV5 control, made in the identical manner) was assessed in 3 FLS cell lines. All 7 serotypes showed increased luciferase expression levels when compared with AAV5, ranging from 6-fold to 55-fold increases depending on the serotype and cell line used (FIG. 2G-L)

[0500] 3.2.3 A similar experiment was performed in HEK293T cells. Transduction with all 7 serotypes resulted in enhanced luciferase expression compared with wtAAV5, ranging from 2-fold to 12-fold increases (FIG. 2M).

Example 4

In Vivo Study in the Air Pouch Synovium Model

4.1. Materials and Methods

Animals

[0501] Female Balb/c mice (8-10 weeks old and weighing 20-25 g; (Harlan, Boxmeer, the Netherlands)) were housed in individual ventilated cages at the animal facility of the Academic Medical Center, Amsterdam. Food and water were available ad libitum. All animal experiments were performed according to the guidelines of the Animal Research Ethics Committee of the University of Amsterdam.

Air Pouch Synovium (APS) Model

[0502] Two serotypes, AAV9-A2 and AAV7-A6, were compared against wtAAV5. The air pouch synovium model was adapted from Edwards et al (1981; J Pathol 134: 147-156). At day 0, 3 ml of air was injected subcutaneously into the dorsal skin of 7-9 week-old female Balb/cOlaHsd mice (Harlan) (day 0). Immediately following the formation of the air pouch, 1 ml of air was removed and 1 ml of AAV (2e10 vector genomes/mouse in PBS (Gibco, ref. 10010 containing 0.001% pluronic F68 (Sigma, ref. p5556) was added directly into the air pouch. Three days following transduction, gene expression was measured by in vivo animal imaging.

Imaging of Luciferase Expression

[0503] Luciferase expression was measured at day 3. It was initially planned to continue monitoring expression for up to 3 months following vector administration, however, a parvovirus infection of the animal facility resulted in the premature termination of all ongoing experiments. D-luciferin potassium-salt substrate (Caliper Life Sciences, Hopkinton, Mass., USA) was injected intraperitoneally (150 mg/kg of body weight, in a volume of approximately 200 μl). Photon counts were acquired 10 min after substrate administration for 5 min using a cooled charge-coupled device (CCD) camera system (Photon Imager, Biospace Lab, Paris, France) and image processing and signal intensity quantification and analysis were performed using M3 Vision (Biospace Lab). The number of photons emitted per second per square centimeter per steradian was calculated as a measure of luciferase activity.

General Animal Conditions and Ethics Statement

[0504] Air pouch formation, vector administration and in vivo imaging were performed under isoflurane anaesthesia (3% isoflurane and oxygen). At the end of the experiments, animals were sacrificed by cardiac puncture under isoflurane anaesthesia, followed by cervical dislocation. The studies were reviewed and approved by the animal care and use committee of the University of Amsterdam and carried out in strict accordance with the recommendations in the Dutch Law on Animal Welfare (Dutch: “Wet op Dierproeven”). Animals were maintained under pathogen-free conditions in the animal facility of the University of Amsterdam.

4.2. Results

[0505] Based on these promising results, a preliminary in vivo study was performed using the air pouch synovium (APS) model, where two serotypes, AAV9-A2 and AAV7-A6, were compared against wtAAV5. Due to an unfortunate infection in the animal facility that necessitated the premature termination of this study, we were only able to obtain data from a single time point, day 3 post vector administration. At this time point it was clear that the capsid mutants were giving rise to increased gene expression when compared with AAV5, with AAV7-A6 showing -6-fold increased expression and AAV9-A2˜22-fold (FIG. 3A).

Example 5

In Vivo Study: Intra-Articular Injections in Healthy Animals

5.1. Material and Methods

Animals

[0506] Male DBA1/J mice (12 weeks old, Envigo) were housed in individual ventilated cages at the animal facility of the Academic Medical Center, Amsterdam. Food and water were available ad libitum. All animal experiments were performed after approval of the Central Commission Animal Experiments (CCD) and the Animal Research Ethics Committee of the University of Amsterdam, the Netherlands.

Expression Study

[0507] Five rAAV comprising capsid mutants, i.e., AAV9-A2, AAV1-P4, AAV7-A6, AAVrh10-A6 and AAVrh10-A2, were compared against wtAAV5. As capsid load may affect expression (Aalbers C J et al., Hum Gene Ther 2017; 28 (2):168-178), rAAV preparations were corrected for the capsid load by adding wtAAV5 empty particles. Healthy mice (n=9 per group) received intra-articular injections of AAV vector carrying the luciferase gene in both knees (7.5×10.sup.9 viral genomes per knee). Gene expression was determined by in vivo imaging at several time points after vector administration.

Imaging of Luciferase Expression

[0508] Luciferase expression was determined at indicated time points (FIG. 3B). At each time point, D-luciferin potassium-salt substrate (Caliper Life Sciences, Hopkinton, Mass., USA) was injected intraperitoneally (150 mg/kg of body weight, in a volume of approximately 200 μl). Photon counts were acquired 15 min after substrate administration for 5 min using a cooled charge-coupled device (CCD) camera system (Photon Imager, Biospace Lab, Paris, France). Image processing and signal intensity quantification and analysis were performed using M3 Vision (Biospace Lab). The number of photons emitted per second per square centimeter per steradian was calculated as a measure of luciferase activity.

General Animal Conditions and Ethics Statement

[0509] Vector administration and in vivo imaging were performed under isoflurane anaesthesia (4% isoflurane and oxygen). The studies were carried out in strict accordance with the recommendations in the Dutch Law on Animal Welfare (Dutch: “Wet op Dierproeven”). Animals were maintained under pathogen-free conditions in the animal facility of the University of Amsterdam.

5.2. Results

[0510] At the first time point, day 3, AAV-mediated expression in the knee is detected in all groups and increases in time (FIG. 3B). All capsid mutants except AAV1-P4 show increased expression vs. wtAAV5 with AAV9-A2 showing the highest expression (−5 fold increased vs. wtAAV5 at day 14) (FIG. 3C). Expression levels at day 14 from high to low: AAV9-A2>AAVrh10-A2>AAVrh10-A6>AAV7-A6>wtAAV5>AAV1-P4. On day 7, AAVrh10-A2, AAV9-A2 and AAVrh10-A6 show significantly increased expression vs. wtAAV5 (**P<0.05, ***P<0.01, ****P<0.00001 vs. wtAAV5 at day 14. (FIG. 3B).

Example 6

Determination of Neutralizing Antibody Titers Against Capsid Mutants in Human Sera

6.1 Material and Methods

[0511] HEK293T cells were plated in DMEM containing 9% FBS, 0.9% penicillin/streptomycin in 96-well clear-bottomed plates. Cells were allowed to rest for 24 hours (at 37° C., 5% CO.sub.2) before transduction. Human serum samples (obtained from the French blood institute) where diluted as follows: neat undiluted serum—1:4-1:16-1:64-1: 256-1:1,024 (neat serum means 1 volume of virus for 1 volume of serum). A pooled mouse plasma sample (from 10 DBA/1 mice, taken 42 days after intra-articular injection of an AAV5-vector) was serially diluted in FBS as follows: 1:10-1:50-1:250-1:6,250-1:31,250. A solution of human Intravenous Immunoglobulin (IVig, Sanquin, lot 15D30H4560A) was serially semi-log diluted from 1:10 down to 1:10,000. Samples and controls were incubated together with the appropriate capsid mutant or wtAAV5 vector for 30′±5 min. at 35-38° C. at an MOI of 2,500 (as determined previously). After 48±2 hours, luciferase reagent was added and luminescence emission was measured with the VictorX microplate reader. Transduction inhibition titers were determined as the highest dilution of serum still associated to a detectable neutralizing activity, i.e. a neutralizing activity >50%.

6.2 Results

[0512] As presented in Table 3, 70%-85% of the samples did not contain neutralizing antibodies against wtAAV5 or the 7 capsid mutants. Most of the samples shared reactivity against the seven capsid mutants, thus a serum sample having reactivity against the wild type AAV5 capsid also reacted against other capsids. In terms of the level of response, they were also comparable between capsid mutants. The number of samples that did not react (ND=not detected) is indicated for each capsid mutant. These data are only given as information as it is very difficult to compare titers with different vectors. Regarding the pooled mouse serum sample from intra-articular injected joints, it only reacted against the WT AAV5 capsid that was used to immunize the animals, whereas no response was observed against mutant capsids (Table 3). All capsid mutants and WT AAV5 were neutralized by IVIg (titers >100) (data not shown).

TABLE-US-00010 TABLE 3 For each serum sample as well as the pooled mouse plasma, the inhibitory titer is reported and corresponds to the highest dilution still associated to a detectable neutralizing activity AAV5 AAV9A2 AAV-DJ-QR-P2 AAVrh10-A2 AAV1-P4 AAV2-P2 AAV7-A6 AAVrh10-A6 sample 1 sample 2 4 ND ND ND ND ND ND ND sample 3 ND ND ND ND ND ND ND ND sample 4 ND ND ND ND ND ND ND ND sample 5 ND ND ND ND ND ND ND ND sample 6 ND ND ND ND ND ND ND ND sample 7 sample 8 sample 9 ND ND ND ND ND ND ND ND sample 10 4 4 4 sample 11 ND ND ND ND ND ND ND ND sample 12 ND ND ND ND ND 1 ND ND sample 13 ND ND ND ND ND ND ND ND sample 14 ND 1 ND ND 4 1 ND 1 sample 15 ND ND ND ND ND ND ND ND sample 16 1 1 4 4 4 1 1 sample 17 ND ND ND ND ND ND ND ND sample 18 ND ND ND ND ND ND ND ND sample 19 ND ND ND ND 4 ND ND ND sample 20 4 4 ND % negative samples 85  80  75  85  80  70  75  75  mouse plasma ND ND ND ND ND ND ND Titers > 8 are considered as seropositive. Positive signals are highlighted in bold/italic. ND = Not Detectable

Example 7

[0513] In Vitro Expression of AAV Capsid Mutants Containing the EF1α Promoter and hTNFR2-IgG1 Transgene in HEK293T/17Cells.

Material and Methods

[0514] Three AAV vectors (AAV5 wt, AAV9A2, AAVrh10A2) (antibody based affinity and Cesium chloride gradient purified) expressing human TNFR2-IgG1 (etanercept) were generated and expression levels were assessed in HEK293T/17 cells. AAV5 wt is wild type AAV5, AAV9A2 comprises the modified capsid protein AAV9 A2 and AAVrh10A2 comprises the modified capsid protein AAVrh10 A2.

[0515] In detail, HEK293T/17 cells were seeded in a 48-well plate (Costar, ref #3548.) at 105,000 cells per well. After overnight incubation, the cell supernatants were replaced with DMEM glutaMAX-I (Gibco 31966-021) containing 0.001% pluronic F68 solution (Sigma P5556). The different vectors were added in duplicate, at an multiplicity of infection (MOI) of 100,000. After 4 hours, doxorubicin (final concentration 0.4 μM) (Sigma D1515) or doxorubicin (final concentration 0.4 μM), (Sigma D1515) in combination with LPS (final concentration 1 μg/ml) (Sigma, ref #L2880-10 mg) (LPS induces inflammation of the cells) in DMEM-glutaMAX-I-containing FBS (heat inactivated (HI)) Bovine Serum Gold, (Gibco, ref 10270-106), (final concentration 1%), was added to the wells. Three days after transduction the supernatant was harvested and hTNFR2-IgG1 levels in the supernatant were quantified by a soluble hTNFR2 DuoSet Elisa (R&D systems ref #DY726).

Results

[0516] Transduction with AAV9A2 and AAVrh10A2 resulted in enhanced hTNFR2 expression levels compared to wtAAV5, ranging from 9- to 13-fold increases in non-stimulated and LPS-stimulated cells. (FIG. 8).

CONCLUSION

[0517] Virions comprising the modified capsid protein AAV9A2 or AAVrh10A2 result in higher gene expression levels in HEK293/17 cells as compared to AAV5 wt virions.

[0518] The present invention has been described above with reference to a number of exemplary embodiments as shown in the drawings. Modifications and alternative implementations of some parts or elements are possible, and are included in the scope of protection as defined in the appended claims