Immunotherapy of Skeletal Myopathies Using ANTI-FAP CAR-T Cells

Abstract

The invention relates to an immune cell engineered to express a chimeric antigen receptor (CAR) which specifically binds Fibroblast Activation Protein (FAP) for use in the treatment of skeletal muscle fibrosis in muscular dystrophies.

Claims

1. A method for treating skeletal muscle fibrosis in muscular dystrophies comprising administering to a patient in need thereof a therapeutically effective amount of an immune cell engineered to express a chimeric antigen receptor (CAR) which specifically binds Fibroblast Activation Protein (FAP).

2. The method according to claim 1, wherein the CAR comprises: (i) an extracellular domain comprising at least one antigen-binding domain that specifically binds FAP, (ii) a transmembrane domain, and (iii) an intracellular domain comprising an intracellular signaling domain capable of activating an immune cell, and optionally comprising one or more co-stimulatory signaling domains.

3. The method according to claim 1, wherein the antigen-binding domain is a single-chain variable fragment (scFv) of a monoclonal antibody that specifically binds FAP.

4. The method according to claim 3, wherein the single-chain variable fragment (scFv) that binds FAP comprises a heavy chain variable domain comprising the amino acid sequence SEQ ID NO: 1 and a light chain variable domain comprising the amino acid sequence SEQ ID NO: 2.

5-15. (canceled)

16. The method according to claim 3 wherein the single-chain variable fragment (scFv) that binds FAP comprises the amino acid sequence SEQ ID NO: 8.

17. The method according to claim 2, wherein the extracellular domain further comprises a hinge domain from IgG4 heavy chain.

18. The method according to claim 2, wherein the transmembrane domain is from CD28.

19. The method according to claim 2, wherein the intracellular signaling domain is a CD3 zeta signaling domain.

20. The method according to claim 2, wherein the intracellular domain further comprises one or more co-stimulatory signaling domains from CD28 or 4-1BB.

21. The method according to claim 2, wherein the CAR comprises from its N- to C-terminus: a signal peptide from mouse Ig-kappa light chain, a scFv fragment from an anti-FAP monoclonal antibody, a modified hinge domain from human IgG4 heavy chain, a transmembrane domain from human CD28, a first co-stimulatory domain from human CD28, a second co-stimulatory domain from human 4-1BB and an intracellular signaling domain from human CD3 zeta chain.

22. The method according to claim 2 wherein the CAR comprises the amino acid sequence SEQ ID NO:16.

23. The method according to claim 1, wherein said immune cell is a T cell or NK cell.

24. The method according to claim 1 wherein said immune cell is a cytolytic T cell.

25. The method according to claim 1, wherein said immune cell is modified with an expression vector comprising a nucleic acid construct encoding the CAR chosen from a lipid nanoparticle packaging an RNA molecule and a lentiviral vector.

26. The method according to claim 25 wherein said expression vector is a self-inactivating or VSVG pseudotyped lentiviral vector.

27. The method according to claim 1, which reduces the expression level of at least one biomarker of fibrosis.

28. The method according to claim 1, wherein said immune cell is administered in combination with a vector for gene therapy of muscular dystrophies.

29. The method according to claim 1, wherein the muscular dystrophies are chosen from Dystrophynopathies, Limb-girdle muscular dystrophies and Congenital muscle dystrophies.

30. The method according to claim 1, wherein the muscular dystrophies is Duchenne muscular dystrophy.

31. The method of claim 28, wherein said immune cell is administered in combination with a reduced dose of vector for gene therapy compared to the use of the vector for gene therapy without the immune cell.

Description

FIGURE LEGENDS

[0148] FIG. 1: Schematic representation of the third-generation lentiviral vectors (A) pCCL-EF1a-scFvFAP-CD28-4.1BB-CD3-T2A-CD19-WPRE and (B) pCCL-EF1a-scFvFAP-CD28-4.1BB-CD3-WPRE where scFv against mFAP is fused to T-cell signaling domains CD28, 4.1BB and CD3. The selection marker CD19 is also indicated (A only).

[0149] FIG. 2: Phenotypic multi-color flow cytometric analysis of T lymphocytes (LT) non-transduced (left panels) or transduced (right panels) after 7 days of activation including 5 days of transduction (PT). Cells have been immuno-labelled prior analysis with a mouse anti-CD3, -CD4 and -CD8, and a human anti-CD19 and analysis is performed on live-gated cells.

[0150] FIG. 3: Murine FAP expression in 3T3 and 3T3-FAP cells. (A) mRNA FAP expression measured by ddPCR in 3T3, 3T3-FAP and a mixed (50:50) cells population of 3T3 and 3T3-FAP cells. Absence of FAP expression is confirmed in 3T3 and is specifically expressed in the stable cell line 3T3-FAP. (B) Murine FAP expression revealed by immunocytochemistry using a murine anti-FAP antibody.

[0151] FIG. 4: Luciferase assay after 24 h coculture between 3T3-Luc2 or 3T3-FAP-Luc2 and T-cells or FAP-CAR-T following ratio 1:7.5 and 1:15 (1 3T3 for 7.5 or 15 T). These results correspond to two experiments done in triplicates (one dot corresponds to one value). p values were calculated with Student's T-test

[0152] FIG. 5: Degranulation assay after 6 h coculture between 3T3-Luc2 or 3T3-FAP-Luc2 and T-cells or FAP-CAR-T following the ratio 1:15.

[0153] FIG. 6: Kinetic of FAP (top) and Col3 (bottom) expression relative to P0 in the Tibialis (TA), Gastrocnemius (GA) and Extensor Digitorum Longus (EDL) from DBA2 and DBA2-MDX mice aged from 2 to 4 month-old. Statistical analysis: unpaired Student's T-test.

[0154] FIG. 7: Schematic representation of the FAP-CAR-T treatment and details for each study. Two to three month-old DBA2-MDX mice receive 2 consecutive doses of FAP-CAR-T cells (low or high) and skeletal muscles were collected 2 weeks after the second infusion for molecular and histological analyses. The number (Nb) of T-cells injected represents the total number of cells in the preparation of CAR-T cells. The Input is calculated from the vector copy number (VCN) in the preparation of CAR-T cells and which represents the active principle. These Input numbers vary depending on the separate preparations of CAR-T cells which were made and are reported in the table for each injection.

[0155] FIG. 8: mRNA expression of FAP (top) and Col3 (bottom) in the Tibialis (TA), Extensor Digitorum Longus (EDL) and heart from DBA2 (WT) and DBA2-MDX mice controls (white bars) or treated with LT (darker grey bars) or FAP-CAR-T cells at r high doses (110.sup.6 cells light grey bars) (n=6 mice per group). Statistical analysis: Mann and Whitney unpaired t test.

[0156] FIG. 9: Sirius Red histological sections of TA, EDL and Heart from DBA2 mice, untreated DBA2-MDX controls and treated DBA2-MDX mice. The dark intensity staining represents collagen fibers (Obf: 10) (A). Collagen deposit were quantified on muscle section (B).

[0157] FIG. 10: Schematic representation of the combined FAP-CAR-T treatment and AAV-microdystrophin gene therapy in MDX mice model of DMD.

[0158] FIG. 11: Increase of microdystrophin mRNA expression in TA and EDL after FAP-CAR-T treatment. MDX mice were treated as shown on FIG. 10. MDX mice treated only with a 10-fold higher dose of AAV-MD1 (410.sup.13 vg/kg) are used as positive control. Microdystrophin mRNA levels were determined by digital droplet PCR using primers from Table 4 and normalized to levels of murine Mpz (myelin protein zero or PO) mRNA.

[0159] FIG. 12: Increase of AAV viral genome copies in TA and EDL after FAP-CAR-T treatment. MDX mice were treated as shown on FIG. 10. MDX mice treated only with a 10-fold higher dose of AAV-MD1 (410.sup.13 vg/kg) are used as positive control. Viral genome copies was determined by digital droplet PCR using primers from Table 4.

[0160] FIG. 13: Increase of microdystrophin positive fibers in TA and EDL after FAP-CAR-T treatment. MDX mice were treated as shown on FIG. 10. MDX mice treated only with a 10-fold higher dose of AAV-MD1 (410.sup.13 vg/kg) are used as positive control. Microdystrophin positive fibers were quantified after immunostaining on muscle sections.

[0161] FIG. 14: Increase of microdystrophin protein expression in TA and EDL after FAP-CAR-T treatment. MDX mice were treated as shown on FIG. 10. MDX mice treated only with a 10-fold higher dose of AAV-MD1 (410.sup.13 vg/kg) are used as positive control. Microdystrophin protein expression was quantified with simple western blot.

[0162] FIG. 15: Decrease of collagen deposit area in TA and EDL after FAP-CAR-T treatment. MDX mice were treated as shown on FIG. 10. MDX mice treated only with a 10-fold higher dose of AAV-MD1 (410.sup.13 vg/kg) are used as positive control. Collagen deposit were quantified on muscle section.

EXAMPLES

Materials and Methods

1. Plasmid Construction

1.1 Construction of pCCL-EF1a-scFvFAP-CD28-4.1BB-CD35-T2A-4CD19-WPRE

[0163] From the construct, pCCL-EF1a-scFvCD123-CD28-4.1BB-CD3-T2A-CD19-WPRE previously described in Bole-Richard et al., Leukemia, 2020, 34, 3228-3241, the fragment scFvCD123 was removed using the restriction enzymes PspXI and BstEII generating an acceptor backbone for the scFv FAP fragment. The fragment scFv FAP was synthetized from the sequence of an anti-murine FAP monoclonal antibody described as SEQ ID NO: 3 in the international application WO 2014/055442 and listed below as SEQ ID NO: 18, adding at each extremity the restriction enzyme sequences PspXI and BstEII. The scFvFAP fragment was digested with the restriction enzymes PspXI and BstEII, ligated into the PspXI and BstEII-digested acceptor pCCL backbone and the product was used to transform competent XL10 cells. Colonies were screened by digestion and the correct clones were confirmed by sequencing. pCCL-EF1a-scFvCD123-CD28-4.1BB-CD32-T2A-CD19-WPRE corresponds to the nucleotide sequence SEQ ID NO: 19.

1.2 Construction of pCCL-EF1a-scFvFAP-CD28-4.1BB-CD3-WPRE

[0164] From the pCCL-EF1a-scFvFAP-CD28-4.1BB-CD32-T2A-CD19-WPRE plasmid (a) a PCR reaction was done in order to generate the fragment EcoR1-scFv-FAP-T2A-Sal1-EcoR1, (b) the fragment scfvFAP-CD28-4.1BB-CD32-T2A-CD19 was removed using the restriction enzyme EcoRI and (c) the PCR fragment generated in (a) that does not contain CD19 anymore was ligated to the EcoR1-digested pCCL backbone obtained in (b) to generate the pCCL-EF1a-scfvFAP-CD28-4.1BB-CD32-WPRE construct (SEQ ID NO: 20). Such product was used to transform competent XL10 cells. Colonies were screened by digestion and correct clones were confirmed by sequencing.

1.3 Construction of pCCL-SFFV-mFAP-WPRE

[0165] From the construct pCCL-SFFV-Luc2-WPRE available in the laboratory, the Luciferase 2 fragment was removed using the restriction enzymes BamHI and SalI. A plasmid containing the sequence coding for the mouse FAP protein was purchased from R&D Systems (RDC2905) and the mFAP cDNA fragment was extracted using the restriction enzymes BamHI and SalI prior ligation into the PCCL backbone in order to obtain pCCL-SFFV-mFAP-WPRE. The new product was used to transform competent XL10 cells. Colonies were screened by digestion and correct clones were confirmed by sequencing.

2. Lentiviral Production and Generation of Stably-Transduced Target and Control Cell Lines

[0166] Lentiviral vectors were produced by transient transfection of HEK293T cells using calcium phosphate and 4 plasmids including a transfer plasmid and three accessory plasmids (HIV-1 gagpol, HIV-Rev, and VSV-G). The following transfer plasmids were used: pCCL-EF1a-scFvFAP-CD28-4.1BB-CD32-T2A-CD19-WPRE, pCCL-EF1a-scFvFAP-CD28-4.1BB-CD32-WPRE, pCCL-SFFV-mFAP-WPRE and pCCL-SFFV-Luc2-WPRE. The harvested particles were concentrated about 500 fold by ultracentrifugation (50,000g, 2 h, 12 C.), suspended in phosphate buffered saline and cryopreserved at 80 C. The infectious titer of the vector was determined as infectious genome (IG)/mL titers on HCT116 cells using the ddPCR provirus primers normalized to albumin (see primers and probes sequences below):

TABLE-US-00001 TABLE1 Oligonucleotidesequences Oligo Sequence5.fwdarw.3 Alb.fw GCTGTCATCTCTTGTGGGCTGT(SEQIDNO:21) Alb.rv ACTCATGGGAGCTGCTGGTTC(SEQIDNO:22) Alb.pr CGCACGGCAAGAGGCGAGG(SEQIDNO:23) PRO.fw CACTCCCAACGAAGACAAGA(SEQIDNO:24) PRO.rv TCTGGTTTCCCTTTCGCTTT(SEQIDNO:25) PRO.pr TCTCTAGCAGTGGCGCCCGAACAGG(SEQID NO:26)

[0167] The following stable cell lines expressing FAP, Luc2 or both transgenes: 3T3-FAP, 3T3-Luc2, 3T3-FAP-Luc2 were generated by lentiviral transduction in 1 ml complete DMEM media (1% glutamine, 1% Pen/Strep, 10% FBS) of the NIH-3T3 fibroblastic cell line using the vectors described above (110.sup.5 cells, 210.sup.6 IG/mL). Expression of the transgenes were confirmed by immunochemistry or bioluminescence assays.

3. FAP-CAR-T Cells Preparation

[0168] DBA2 primary murine splenic T cells were isolated using the Pan T cell Negative Selection kit as suggested by the manufacturer (Miltenyi Biotec), and transduced with the indicated LV. The cells (110.sup.6 cells/well) were incubated in 1 mL of complete medium in 24-well plates (complete medium=RPMI-1640+1% glutamine+1% Pen/Strep+10% FBS+50 uM Beta-mercaptoethanol+50 g/mL IL-2+CD3/CD28 activation beads at a ratio of 1:1 (Gibco) prepared freshly). After 48 hours, cells (110.sup.6 cells/well in 1 mL complete medium) were mixed with FAP-CAR-T lentivirus (210.sup.7 IG/mL) and lentiboost (0.5 mg/mL) in a 24-well plate. After overnight incubation, cells were expanded in complete media for another 4 days. T-cells transduction efficacy was either determined by flow cytometry using an anti-human CD19 when using the LV pCCL-EF1a-scFvFAP-CD28-4.1BB-CD3-T2A-CD19-WPRE (see below) or by ddPCR to determine the vector copy number per cell in the case of the LV pCCL-EF1a-scFvFAP-CD28-4.1BB-CD3-WPRE (using provirus ddPCR primers referenced to murine titin).

TABLE-US-00002 TABLE2 Oligonucleotidesequences mTitin-F 5-AAAACGAGCAGTGACGTGAGC-3 (SEQIDNO:27) mTitin-R 5-TTCAGTCATGCTGCTAGCGC-3 (SEQIDNO:28) mTitin-P 5-TGCACGGAAGCGTCTCGTCTCAGTC-3 (SEQIDNO:29)

4. FAP-CAR-T Cells Phenotyping

[0169] After 7 days of culture, LT and FAP-CAR-T cells were incubated for 1 h with the following conjugated antibodies prior to FACS analysis: anti mouse-CD3, anti-mouse CD8 and anti-mouse CD4 and with the anti-human CD19, when using the LV pCCL-EF1a-scFvFAP-CD28-4.1BB-CD3-T2A-CD19-WPRE. After washing cells and adding the viability marker 7AAD, cells were analyzed by flow cytometry.

5. Luciferase Assay for FAP-CAR-T Cell Recent Cytotoxic Activity

[0170] 5000 3T3-Luc2 control cells and 3T3-FAP-Luc2 target cells were plated in a 96 well plate (ViewPlate, Perkin Elmer, 6005181) in 100 L of complete DMEM media (1% glutamine, 1% Pen/Strep, 10% FBS). The following day, FAP-CAR-T cells or control non-specific T lymphocytes (TL) were added at 2 different concentrations in 100 L of medium containing 37,500 or 75,000 cells, to cuo-culture cells and targets for 24h. At the end of the co-culture, 100 L of supernatant medium were removed from each well and mixed with 100 L of luciferin solution (Bright-Glo Luciferase assay, Promega) to measure bioluminescence levels extemporaneously using a luminometer (560 nm).

6. Degranulation Assay to Measure FAP-Specific Cytotoxicity

[0171] Functional cytotoxic activity of FAP-CAR T cells was measured by the detection of CD107 lysosomal-associated membrane protein (LAMP) on the surface of recently degranulated cells. 5000 control 3T3 cells and 3T3-FAP target cells were plated in a 96 well plate in 100 l of complete DMEM medium. The following day, 100 L1 containing 75,000 FAP-CAR-T cells or control non-specific T lymphocytes (TL) were added to the corresponding wells together with brefeldin-A (1/2000) and 20 L/mL of anti-CD107a. After 6 h incubation at 37 C., cells were washed with PBS-1 and stained with anti-CD8 antibody. After adding the viability marker 7AAD, cells were analyzed by flow cytometry.

7. Mice

[0172] For this entire study, only males were used. The DBA2 controls were purchased directly from Charles River Laboratories (DBA/2J, Ref: 625) while the DBA2/MDX are bred and obtained from the Centre d'Exploration et de Recherche Fonctionnelle Exprimentale (CERFE; Evry, France). Originally, the DBA2-MDX have been purchased from Jackson Laboratories (D2.B10-Dmd.sup.mdx/J, Strain #013141) and they have been at CERFE for about 8-10 generations.

8. Expression of Fibrosis Marker Genes

[0173] The levels of murine collagen 3a1 and murine FAP gene mRNAs was measured by ddPCR in different muscle tissues and normalized to levels of murine Mpz (myelin protein zero or PO) mRNA. Tibialis anterior (TA), gastrocnemius (GA), extensor digitorum longus (EDL) and heart, were isolated from euthanized DBA2 and DBA2-MDX and directly conserved in RNAlater. RNA was extracted from these tissues using the RNeasy fibrous tissue kit (Qiagen) following manufacturer's recommendations. RNA was then reverse transcribed using the Verso cDNA Synthesis kit (ThermoFisher Scientific). For the ddPCR, 1 of ddPCR Supermix for Probes no dUTP (Biorad), 16 ng of complementary DNA and the following primers sets were used: ddPCR Gene Expression Assay: Fap, Mouse (Biorad, 10031252), ddPCR Gene Expression Assay: Col3a1, Mouse (Biorad, 10031252) and ddPCR Gene Expression Assay: Mpz, Mouse (Biorad, 10031255).

TABLE-US-00003 TABLE3 Ampliconsequences Gene Coordinates Sequenceoftheamplicon Mpz mm10|chr1: ATCAAAAGGGCTGGGGGAGTCTCGCAAGGATAAG 171160004- AAATAGCGGTTAGCGGGCCGGGGGGGGGTCGGG 171160125:+ GGTCTGCGACGGAGTCCTCCAAAGGCTCTCAGGT GGTGGTCATCGAGATGGAGC(SEQIDNO:30) Fap mm10|chr2: GAGCTCTCTTTGTGTTTCCTTCAGGTTTGTAAACTC 62556167- TTGAGGGACGTAAGACAATGCATATCACCACTAA 62573849:+ AGCAAGCGCAGCCAGGGTGGTAACTCCAAAGACA GTTTTCAGCCATGTCTTC(SEQIDNO:31) Col3a1 mm10|chr1: GTGGAACCTGGTTTCTTCTCACCCTTCTTCATCCCA 45311767- CTCTTATTTTGGCACAGCAGTCCAACGTAGATGAA 45321568:+ TTGGGATGCAGCCACCTTGGTCAGTCCTATGAGTC TAGAGATGTCTGGAAGCCAGAACCATGTCAAA (SEQIDNO:32)

[0174] After running the PCR [95 C. for 10 min+40(94 C. for 30 sec, 60 C. for 1 min)+98 C. for 10 min], droplets are generated using the droplet generator QX200 (Biorad) and reactions are read and analyzed using the QuantaLife software (Biorad).

9. Mice and In Vivo Studies

[0175] Two to three month old anesthetized DBA2-MDX mice were injected intravenously into retro-orbital vein with either 510.sup.5 (Low dose) or 110.sup.6 (high dose) of control LT or specific FAP-CAR-T cells resuspended in 100 l PBS-1. Mice were reinjected one week later with the same dose and same cells. After two weeks, age-matched control mice (DBA2) and treated mice were euthanized and skeletal muscles collected for molecular and histological analyses.

[0176] For subsequent gene therapy, two to three month old DBA2-MDX mice were injected with two consecutive doses (110.sup.6) of FAP-CAR-T cells, and after two weeks of treatment the mice were injected with one dose (5.Math.10.sup.12 vg/kg) of rAAV9-microdystrophin (MD1) vector. For concomitant gene therapy, DBA2-MDX mice were injected with a first dose (110.sup.6) of FAP-CAR-T cells, and after two weeks of treatment the mice were injected concomitantly with a second dose (110.sup.6) of FAP-CAR-T cells and one dose (5.Math.10.sup.12 vg/kg) of rAAV9-microdystrophin (MD1) vector. Control DBA2-MDX mice were injected with two consecutive doses (110.sup.6) of LT, and two weeks later with one dose (5.Math.10.sup.12 vg/kg) of rAAV9-microdystrophin (MD1) vector (Bourg et al., Int. J. Mol. Sci., 2022, 23, 2016; doi: 10.3390/ijms23042016). After 6 weeks, blood was collected, and force tests (escape test an dTA in situ) were performed on control and treated mice (DBA2-MDX). After five weeks, control and treated mice (DBA2-MDX) were euthanized and heart and skeletal muscles (TA, ED) were collected for molecular and histological analyses.

10. Histological Analysis

[0177] TA, GA and EDL collected from study mice were immediately snap frozen in liquid nitrogen. Cryopreserved muscles were then sectioned (8 m) using a cryostat and stained with the Sirius Red dye using standard procedures. Images were then acquired using the Axioscan microscope (10 Obj, Zeiss).

Quantification of Microdystrophin Positive Fibers

[0178] Fiber cytoplasmic region enclosed within the membrane staining (Laminin) are segmented by morphological segmentation after contrast enhancement and artefact filtering (FiJi software 2.0.0-rc/1.52 p, Morpholib plugin v 1.4.1). Fibers are enlarged according to the magnification to capture the membrane region. Nuclei are detected from the DAPI channel using the local maxima detection. The fluorescence intensity in each object (fibers, fibers membrane and nuclei) is measured for each channel together with fibers shape and size.

[0179] Nuclei are associated to their parent fiber using the R software and all fluorescence and shape datas aggregated together. Non-fiber objects are filtered-out based on shape, size and fluorescence criteria. Positive fibers for any channel (Laminin, microdystrophin and nuclei) are detected based on the fluorescence distribution of negative control slices or slices from known negative condition and the percentage of microdystrophin fibers were determined

Collagen Quantification on Muscle Section

[0180] Collagen deposits on muscle section are quantified using an open-source software for bioimage analysis: Qupath Software. For that, we created two Pixel classifiers to train the software on 3 muscle sections to identify representative region (collagen region) by surrounding them. The first one is for delimiting the tissue to be analysed and the second one is for identifying the region of collagen deposits. We quantified the surface area occupied by the collagen in relation with the total muscle section surface area.

11. Molecular Analysis

[0181] Microdystrophin mRNA expression levels and AAV vector copy number in muscle of DBA2-MDX mice receiving combined therapy were analysed by digital droplet PCR on cDNA or gDNA using the oligonucleotide primers and probes listed below.

TABLE-US-00004 TABLE4 Oligonucleotidesequences Targetgene Name Sequence ITRG AAV22mers.F CTCCATCACTAGGGGTTCCTTG (SEQIDNO:33) ITRG AAV18mers.R GTAGATAAGTAGCATGGC (SEQIDNO:34) ITRG AAV_MGB.P TAGTTAATGATTAACCC (SEQIDNO:35) Microdystrophin dys.R GGTTGTGCTGGTCCAGGGCGT (SEQIDNO:36) Microdystrophin dys.F CCAACAAAGTGCCCTACTACATC (SEQIDNO:37) Microdystrophin dys.P CCGAGCTGTACCAGAGCCTGGCC (SEQIDNO:38) MurineMpz MH181P0.F CTCCAAGCAGATGCAGCAGA (PO) (SEQIDNO:39) MurineMpz M267PO.R ATAGCCTTGCGCATCATGGT (PO) (SEQIDNO:40) MurineMpz M225PO.P CCGTGGTGCTGATGGGCAAGAA (PO) (SEQIDNO:41) MurineTitin mTitin-F AAAACGAGCAGTGACGTGAGC (SEQIDNO:27) MurineTitin mTitin-R TTCAGTCATGCTGCTAGCGC (SEQIDNO:28) MurineTitin mTitin-P TGCACGGAAGCGTCTCGTCTCAGTC (SEQIDNO:29)
Quantification of Microdystrophin Expression with Simple Western Blot

[0182] Muscle proteins were extracted in RIPA buffer supplemented with Protease Inhibitor Cocktail EDTA-free (Roche) and Benzonase by homogenization. Total proteins were then quantified by BCA method, thanks to the Pierce BCA protein assay kit (Invitrogen) accordingly to manufacturer's instructions. Protein detection has been performed by capillary western blot, thanks to the JESS protein simple (Bio-techne), accordingly to manufacturer's directions. Micro-dystrophin detection has been performed by the antibody DysB (NCL-DYSB, Leica, 1:20) and its expression has been quantified by compass software.

12. Statistical Analyses

[0183] Differences between two groups were compared using Student's T test. Differences between multiple groups were compared with one-way analysis of variance (ANOVA). The statistics were calculated using the GraphPad Prism 7 software.

Results

1. Construction of Lentiviral Gene Transfer Plasmids to Produce mFAP-Directed CAR T Cells

[0184] Advanced-generation HIV-1-derived lentiviral gene transfer plasmids were generated to express third-generation chimeric antigen receptor (CAR) constructs directed against murine fibroblast activation protein (FAP), containing the human CD28, 4.1BB and CD3zeta chain modules for co-stimulation, survival and signal transduction, eventually combined to a truncated human CD19 marker (FIG. 1A) or not (FIG. 1B). These constructs were obtained by replacing the single-chain variable Fragment (ScFv) immunoglobulin fusion protein sequence of another CAR previously described in Ble-Richard et al., Leukemia, 2020, 34, 3228-3241, by a scFv specific for mFAP which was synthesized from the sequence published in the International application WO 2014/055442. Both lentiviral plasmids shown in FIG. 1 have been used to produce lentiviral vector for this study. The vectors were pseudotyped with VSVG and titered according to standard techniques present in the laboratory (Corre et al., Gene Therapy, 2022, 29, 536-543).

2. Production and Characterization of Murine FAP-CAR-T Cells

[0185] T lymphocytes (LT) isolated from spleens of 2 to 3 month-old DBA2 mice, were cultured, activated by CD28/CD3 beads and Interleukin-2 for 48h prior to lentiviral transduction. Five days post-transduction, the cell surface phenotype of the cells was characterized by flow cytometry (FIG. 2). Results show that after 7 days of activation, cells are mainly CD8+ and it can be extrapolated that 30% of cells are effectively transduced (CD19+).

3. FAP-CAR T Cell Generation and Functional Characterization In Vitro

3.1 Generation of Cells Expressing Mouse FAP

[0186] In order to test the specificity and efficacy of FAP-CAR-T in vitro, the murine fibroblastic cell line NIH-3T3 cells known to lack the mouse FAP, was used as negative control and to generate specific target cells. A lentiviral construct encoding the murine FAP was produced and used to transduce NIH-3T3 cells to generate the stable cell line 3T3-FAP which expresses high levels of the mFAP mRNA (FIG. 3-A) and of FAP protein (FIG. 3-B).

3.2 Generation of Cells for the Cytotoxicity Assay and Validation of FAP-CAR-T Cell Specific Cytotoxic Activity

[0187] In order to measure FAP-CAR-T cell specific killing of FAP-bearing cells in vitro, two new stable cell lines were generated using lentiviral vectors coding for the bioluminescent marker Luciferin 2: 3T3-Luc2 and 3T3-FAP-Luc2 (FIG. 4). Results show that cell mortality (indicated by a decreased bioluminescence relative to controls) only occurs in a statistically significant manner when FAP-CAR-T cells are cocultured with cells expressing the FAP target.

3.3 Degranulation Assay

[0188] The functional properties of FAP-CAR-T cells were also tested in vitro using CD107a degranulation assay after coculture between 3T3 or 3T3-FAP and T-cells or FAP-CAR-T (FIG. 5). Results show that the FAP-CAR-T cells only degranulate when they are in presence of the murine FAP target.

4. Skeletal Muscle Fibrosis in the DBA2-MDX Model of Duchenne Muscular Dystrophy

[0189] Biomarkers of fibrosis such as FAP expression as well as another marker of fibrosis, Collagen III, were quantified in skeletal muscles of interest (TA=tibialis anterior; GA=gastrocnemius; EDL: extensor digitorum longus) in the DBA2-MDX mouse model for Duchenne muscular dystrophy, over time and in the steady-state, using ddPCR (FIG. 6). Results show that independently of the age, these two makers of fibrosis are always overexpressed in skeletal muscles from DBA2-MDX compared to their age-match DBA2 controls. Significant differences are identified in the TA at 3 and 4 months, and in the GA at 2 or 3 months, depending on the biomarker. Similar trends are also seen in the EDL although it did not reach statistical significance. Lack of significance for some groups might be due to low mice number per group (n=3).

5. Effect of FAP-CAR-T Cells In Vivo on DBA2-MDX Associated Fibrosis

[0190] To test the efficacy of the FAP-CAR-T treatment on DBA2-MDX-associated fibrosis in vivo, two to three month-old DBA2-MDX mice were injected with two consecutive doses of FAP-CAR-T cells prepared and characterized as previously described (see schematic representation of the protocol in FIG. 7). In order to assess the effects of CAR-T cells on fibrosis, mRNA from skeletal muscles of control or treated mice were extracted and after reverse transcription, expression of FAP and Collagen III were determined by ddPCR (FIG. 8). Results show that FAP-CAR-T cells injected systemically are able to decrease biomarkers of fibrosis in skeletal muscles in DBA2-MDX mice. The reduction of biomarkers by FAP-CAR T cells is more pronounced in the TA and EDL where these two biomarkers are expressed at higher levels compared to the GA. Results show also that injection of FAP-CAR-T cells has no significant impact on the decrease biomarkers of fibrosis in heart in DBA2-MDX mice because FAP is not expressed (FIG. 8).

[0191] To confirm effect on fibrosis, histological sections of skeletal muscles from control and treated mice were stained with Sirius red in order to visualize the amount of collagen fibers present in the tissue. Results show a decrease in collagen in skeletal muscle from DBA-MDX treated with the FAP-CAR-T as well as an improvement in the structural organization of muscle fibers (FIGS. 9A and 9B).

6. Effect of FAP-CAR-T Cells Mediated Fibrosis Reduction on Gene Therapy in DMD Mouse Model

[0192] To evaluate the effect of the FAP-CAR-T mediated fibrosis reduction on gene therapy for Duchenne Muscular Dystrophy, two combinations of treatment are being tested: subsequent and concomitant injections of FAP-CAR-T and rAAV vector expressing microdystrophin. For subsequent gene therapy, two to three-month-old DBA2-MDX mice were injected with two consecutive doses of FAP-CAR-T cells, and after two weeks of treatment the mice were injected with one dose of rAAV9-microdystrophin (MD1) vector. For concomitant gene therapy, two to three-month-old DBA2-MDX mice were injected with one dose of FAP-CAR-T cells, and after two weeks of treatment the mice were injected concomitantly with a second dose of FAP-CAR-T cells and one dose of rAAV9-microdystrophin (MD1) vector (FIG. 10). Different tests were performed to assess the expression of microdystrophin mRNA and protein, to assess the quantity of AAV vectors copies in muscle and to evaluate the muscle integrity from control or treated mice.

[0193] Results of subsequent gene therapy show that FAP-CAR-T cells mediatedfibrosis reduction systemically leads to highermicrodystrophin expression levels in muscle fibers in skeletal muscles in DBA2-MDX mice (FIGS. 11 and 14). There is an increase of AAV vectors copies in muscle and an increase of microdystrophin positive fibers (FIGS. 12 and 13). These effects seem to be more pronounced in the TA and the EDL where the the fibrosis is reduced. To confirm effect on muscle integrity, histological sections of skeletal muscles from control and treated mice were stained with Sirius red and hemotoxylin eosine in order to visualize the amount of collagen fibers present in the tissue. Results show a decrease in collagen in skeletal muscle from DBA-MDX treated with the FAP-CAR-T (FIG. 15) as well as an improvement in the structural organization of muscle fibers. Functional analyses of the skeletal muscle are being performed to confirm the efficacy of the combined therapy approach: FAP-CAR-T cells and AAV-MD1, to improve gene therapy for Duchenne muscular dystrophy.

[0194] For concomitant injections of FAP-CAR-T cells and AAV-MD1, it is expected that levels of dystrophin expression and reduction of collagen deposits are less important than in the case of subsequent injections of FAP-CAR-T cells and AAV-MD1.