ENHANCEMENT OF CYTOLYTIC T-CELL ACTIVITY BY INHIBITING EBAG9

Abstract

A genetically modified cytotoxic T cell includes one or more exogenous nucleic acid molecules encoding a transgenic antigen-targeting construct. Estrogen receptor-binding fragment-associated antigen 9 (EBAG9) activity is inhibited in the cells. The antigen-targeting construct can be a chimeric antigen receptor (CAR) or T cell receptor (TCR). The modified T cell can be used in the treatment of a proliferative disease, in particular for the treatment of hematologic malignancies. A pharmaceutical composition includes the modified T cell, a nucleic acid vector encoding the antigen-targeting construct and an inhibitor of EBAG9, such an RNA interference molecule. An in vitro method can increase the cytolytic activity of a cytotoxic T cell.

Claims

1. A genetically modified cytotoxic T cell comprising one or more exogenous nucleic acid molecules encoding a transgenic antigen-targeting construct, wherein in said cells estrogen receptor-binding fragment-associated antigen 9 (EBAG9) activity is inhibited.

2. The genetically modified T cell according to claim 1, wherein the inhibition of EBAG9 activity is associated with an increase in the release of cytolytic granules and/or granzyme-containing secretory lysosomes (compared to a control cytotoxic T cell).

3. The genetically modified T cell according to claim 1, wherein the transgenic antigen-targeting construct is a chimeric antigen receptor (CAR).

4. The genetically modified T cell according to claim 1, wherein the transgenic antigen-targeting construct is a T cell receptor (TCR).

5. The genetically modified T cell according to claim 1, wherein the inhibition of EBAG9 activity is obtained by knock-down of EBAG9.

6. The genetically modified T cell according to claim 5, wherein the inhibition of EBAG9 activity is obtained by genetic modification of the T cell genome with one or more exogenous nucleic acid molecules, said exogenous nucleic acid molecules comprising a vector that encodes the transgenic antigen-targeting construct and an RNA interfering sequence for knock-down of EBAG9.

7. The genetically modified T cell according to claim 1, wherein the inhibition of EBAG9 activity is obtained by genetic modification of the T cell genome by disrupting the expression and/or sequence of the EBAG9 gene.

8. A method of treating a disease in a subject, comprising administering a genetically modified T cell according to claim 1 to a subject in need thereof.

9. The method of claim 8 for the treatment of a proliferative disease, wherein the antigen targeted by the transgenic antigen-targeting construct is expressed in a target cell undergoing and/or associated with pathologic cell proliferation.

10. The method according to claim 9, wherein the proliferative disease is a hematologic malignancy and wherein the antigen targeted by the transgenic antigen-targeting construct is expressed in cancerous cells of said hematologic malignancy.

11. The method according to claim 10, wherein a. the hematologic malignancy is a CD19-expressing B-cell cancer, and wherein the transgenic antigen-targeting construct binds CD19, or b. wherein the hematologic malignancy is a BCMA-expressing B-cell cancer, and wherein the transgenic antigen-targeting construct binds BCMA, or c. wherein the hematologic malignancy is a CXCR5-expressing cancer, and wherein the transgenic antigen-targeting construct binds CXCR5.

12. The method according to claim 9 for the treatment of an autoantibody-dependent autoimmune disease, wherein the antigen targeted by the transgenic antigen-targeting construct is expressed in a target cell associated with autoantibody production.

13. A pharmaceutical composition comprising a genetically modified T cell according to claim 1, wherein the composition is suitable for the treatment of a proliferative disease, comprising additionally a pharmaceutically acceptable carrier.

14. A nucleic acid vector or combination of nucleic acid vectors comprising a sequence that encodes an antigen-targeting construct and an RNA interfering sequence for knock-down of estrogen receptor-binding fragment-associated antigen 9 (EBAG9).

15. In vitro method for increasing the cytolytic activity of a genetically modified cytotoxic T cell, said T cell comprising one or more exogenous nucleic acid molecules encoding a transgenic antigen-targeting construct, the method comprising inhibiting in said T cell the activity of estrogen receptor-binding fragment-associated antigen 9 (EBAG9), wherein inhibiting EBAG9 activity preferably comprises: a. knock-down of EBAG9 by RNA interference of EBAG9 expression, or b. genetic modification of the T cell genome by disrupting the expression and/or sequence of the EBAG9 gene.

16. The genetically modified cytotoxic T cell according to claim 1, wherein in said cells, estrogen receptor-binding fragment-associated antigen 9 (EBAG9) activity is inhibited, compared to a control cytotoxic T cell.

17. The genetically modified cytotoxic T cell according to claim 5, wherein the inhibition of EBAG9 activity is obtained by RNA interference of EBAG9 expression.

18. The genetically modified cytotoxic T cell according to claim 17, wherein the inhibition of EBAG9 activity is obtained by small interfering RNA (siRNA), short hairpin RNA (shRNA) or micro RNA (miRNA).

19. The genetically modified cytotoxic T cell according to claim 7, wherein an exogenous nucleic acid molecule encoding the transgenic antigen-targeting construct is positioned in the T cell genome within, adjacent or associated with the EBAG9 gene, thereby disrupting the expression and/or sequence of said EBAG9 gene.

20. The method according to claim 10, wherein the hematologic malignancy is non-Hodgkin lymphoma, chronic lymphocytic leukemia, acute myeloid leukemia, acute lymphoblastic leukemia or multiple myeloma.

21. The method according to claim 12, wherein the medical disorder is systemic lupus erythematosus (SLE) or rheumatoid arthritis.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0345] FIG. 1: Generating a retroviral MP71 encoding for an EBAG9-targeting miRNA and GFP.

[0346] FIG. 2: Decreased human EBAG9 expression in Jurkat cells after transduction with -retroviral vectors encoding for different EBAG9-targeting miRNAs.

[0347] FIG. 3: Experimental timeline for retroviral transduction of human primary T cells prior to functionally in vitro assays.

[0348] FIG. 4: EBAG9 downregulation facilitates the antigen-independent release of granzyme A from activated human CD8+ T cells.

[0349] FIG. 5: The MP71 vector is suitable for the simultaneous expression of an EBAG9-targeting miRNA and a CAR.

[0350] FIG. 6: BCMA and CD19 CAR expression in transduced primary human T cells.

[0351] FIG. 7: Antigen-specific cytolytic activity of BCMA CAR T cells can be increased by the downregulation of EBAG9.

[0352] FIG. 8: Increasing cytolytic activity of CAR T cells by silencing EBAG9 is a universally applicable cell biological mechanism.

[0353] FIG. 9: In vivo engineered BCMA CAR T cells with silenced EBAG9 eradicate multiple myeloma cells more efficiently.

[0354] FIG. 10: Target site validation for CRISPR-mediated EBAG9 knockout.

DETAILED DESCRIPTION OF THE FIGURES

[0355] FIG. 1: Generating a retroviral MP71 encoding for an EBAG9-targeting miRNA and GFP. EBAG9-targeting miRNAs were generated by exchanging the hairpin antisense sequence of the endogenous miRNA-155 against predicted EBAG9 target site sequences. The resulting miRNA-coding sequences were introduced into a GFP-encoding retroviral MP71 vector at an intronic position. (LTR: long terminal repeat; PRE: post-translational regulatory element; Amp R: ampicillin resistance; GFP: green fluorescence protein).

[0356] FIG. 2: Decreased human EBAG9 expression in Jurkat cells after transduction with gamma-retroviral vectors encoding for different EBAG9-targeting miRNAs. Retroviral transduction of human Jurkat cells with different GFP-encoding vectors expressing different miRNAs directed against human EBAG9. Positively transduced GFP+ cells were enriched by fluorescence-activated cell sorting (FACS) and analyzed by Western Blot. Calnexin was used as a loading control. Lysates of 1×10e6 Jurkat cells were analyzed. UT, untransduced FIG. 3: Experimental timeline for retroviral transduction of human primary T cells prior to functionally in vitro assays.

[0357] FIG. 4: EBAG9 downregulation facilitates the antigen-independent release of granzyme A from activated human CD8+ T cells. Activated human CD8+ T cells were transduced with vectors encoding an SP6 or BCMA CAR in conjunction with the EBAG9-targeting miRNA H18 or the BCMA CAR alone. Enzymatic activities in supernatants were measured on day 15 after CD8+ T cell activation. Granzyme A release was induced by re-stimulation of T cells with anti-human CD3 and anti-CD28 antibodies for 4 h. Values show the release in percentages relative to the total content. Bars represent mean values±SEM of n=3 experiments with n=4 independent donors per group. *p<0.05, **p<0.01, *p<0.001; ns, not significant. A paired t-test was performed.

[0358] FIG. 5: The MP71 vector is suitable for the simultaneous expression of an EBAG9-targeting miRNA and a CAR. Either a SP6, BCMA or CD19 CAR was introduced into a miRNA-containing MP71 vector at the indicated position by using the NotI and EcoRI restriction sites. (LTR—long terminal repeat; PRE—post-translational regulatory element; Amp R—ampicillin resistance; CAR—chimeric antigen receptor)

[0359] FIG. 6: BCMA and CD19 CAR expression in transduced primary human T cells. Primary human T cells were activated by stimulation with anti-human CD3 and anti-human CD28 antibodies for 48 h. Transduction of human T cells with vectors encoding either an BCMA (A-B) or an CD19 (C-D) CAR in conjunction with the EBAG9-targeting miRNA H18 or H17, respectively, was performed twice. Cells were cultured with IL-7/IL-15 supplementation. FACS was performed on day 6 of culture and one representative example per group is shown. Transduction rates are indicated as percentages on the gate.

[0360] FIG. 7: Antigen-specific cytolytic activity of BCMA CAR T cells can be increased by the downregulation of EBAG9. (A-B) Human CD8+ T cells were activated by stimulation with anti-human CD3 and anti-human CD28 antibodies for 48 h. Transduction of human CD8+ T cells with vectors encoding either an SP6 or BCMA CAR alone or in conjunction with the EBAG9-targeting miRNA H18 was performed twice. Cells were cultured with high IL-2 supplementation for 13 days. Prior to the in vitro cytotoxicity assay, CAR T cells were cultured with low IL-2 supplementation for 48 h. In vitro cytotoxicity assays were performed on day 15 of CAR T cell culture. Transduction rates were adjusted to 20%-30% by addition of UT. [51Cr] chromium-labeled MM (A) and B-NHL (B) cell lines were co-cultured with transduced human T cells at different effector to target ratios for 4 h. Data represent mean±SEM error bars, n=5 experiments performed in duplicates with 4-8 different donors per group. *p<0.05, **p<0.01, ***p<0.00.1; ns, not significant. A Mann-Whitney U test was employed.

[0361] FIG. 8: Increasing cytolytic activity of CAR T cells by silencing EBAG9 is a universally applicable cell biological mechanism. Human CD8+ T cells were activated by stimulation with anti-human CD3 and anti-human CD28 antibodies for 48 h. Transduction of human CD8+ T cells with vectors encoding either an SP6 or CD19 CAR alone or in conjunction with the EBAG9-targeting miRNA H17 was performed twice. Cells were cultured with high IL-2 supplementation for 13 days. Prior to the in vitro cytotoxicity assay, CAR T cells were cultured with low IL-2 supplementation for 48 h. In vitro cytotoxicity assays were performed on day 15 of CAR T cell culture. Transduction rates were adjusted to 15% by the addition of UT. [51Cr] chromium-labeled Jeko-1 cell line was co-cultured with transduced human T cells at different effector to target ratios for 4 h. Data represent mean±SEM error bars n=3 experiments performed in duplicates with 3-6 different donors per group. *p<0.05, **p<0.01, ***p<0.00.1; ns, not significant. A Mann-Whitney U test was employed.

[0362] FIG. 9: In vivo engineered BCMA CAR T cells with silenced EBAG9 eradicate multiple myeloma cells more efficiently. (A) NSG mice were engrafted with 1×10e7 MM.1S cells stably expressing GFP and a firefly luciferase. On day 6, tumor inoculation was visualized by IVIS with 150 s exposure. One day later, 1×10e6 CAR+ cells (day 10-13 of culture with II-7/IL-15 supplementation) were transferred, and treatment efficiency was observed by IVIS at 60 s exposure. (B) Mean values±SEM of bioluminescence signal intensities obtained from regions of interest covering the entire body were plotted for each group and timepoint in one experiment. (C) On days 15 and 16, animals were sacrificed and CD138+ GFP+ tumor cells in the bone marrow were quantified by flow cytometry. Mean values t SEM of n=2 experiments are plotted. *p<0.05, **p<0.01, ***p<0.00.1; ns, not significant. A Mann-Whitney U test was employed.

[0363] FIG. 10: Target site validation for CRISPR-mediated EBAG9 knockout. (A) Schematic overview of the human EBAG9 gene consisting of 7 exons. Six guide RNAs (E1-E6) were designed targeting different regions in exon 4. (B-C) Primary human T cells were activated by stimulation with human CD3/CD28 Dynabeads for 48 h prior to electroporation with various guide RNAs (E1-E6) and Cas9 protein. At day 9 after activation, genomic DNA and protein samples were taken for analysis of the gene editing efficiency (B) or EBAG9 protein expression level (C), respectively. One representative Western Blot out of three experiments is shown.

EXAMPLES

[0364] The invention is demonstrated by way of the examples disclosed below. The examples provide technical support for a more detailed description of potentially preferred, non-limiting embodiments of the invention.

Summary of the Examples

[0365] 1. miRNAs targeted at murine or human EBAG9 were cloned into a retroviral expression vector, followed by transduction of human or murine CD8+ T cells. A knockdown of EBAG9 in human and mouse CD138+ T cells of >90% was achieved, as visualized by Western blot analysis.

[0366] 2. Transduction of antigen-specific murine T cells (polyclonal) was carried out with γ-retroviruses encoding miRNAs targeted at EBAG9 and subsequently, their adoptive transfer into RAG2-KO mice was accomplished. These mice were challenged with SV40-large T-antigen (Peptide IV) pulsed splenocytes. In an in vivo killing assay, engineered T cells with an EBAG9 knockdown were more efficient in killing than unmodified or control T cells.

[0367] 3. Human T cells were transduced with γ-retrovirus encoding miRNAs targeted at EBAG9, in combination with various CARs. These cells were used in an in vitro cytotoxicity assay, where target cells expressed the cognate antigens for the CARs chosen. When CAR-positive T cell frequencies were low (<30%), we obtained enhanced cytolysis of either BCMA or CD19 expressing target cells.

[0368] 4. Human T cells were transduced with γ-retrovirus encoding miRNAs targeted at EBAG9, in combination with an anti-BCMA CAR. NSG mice were transplanted with the multiple myeloma cell line MM.1 S, and tumor onset was measured by IVIS imaging. These mice were then treated with T cells expressing a control CAR, a regular BCMA CAR, and the engineered BCMA CAR that co-expressed a miRNA targeted at human EBAG9. Tumor progression was measured 14 days alter CAR T cell administration. The number of CAR T cells was kept low to better identify the performance of few T cells with enhanced cytolytic strength. Anti-BCMA CAR T cells endowed with a miRNA that silenced EBAG9 were superior in tumor control and led to complete tumor cell eradication from bone marrow.

[0369] Retroviral Vector Design

[0370] Cloning and miRNA Sequences

[0371] For silencing of human EBAG9, different miRNAs targeting regions within the open reading frame of the human EBAG9 gene were generated. Four different target site prediction programs were used for the miRNA design reflecting the requirements of the endogenous RNAi machinery to identify suitable target sites (WlsiRNA, BlocklT, siDESIGN, OligoWalk). The miRNA secondary structure is important for recognition and processing by the RNAi machinery. Characteristic features of the RECTIFIED SHEET (RULE 91) ISA/EP miRNA structure are the rather unstructured backbone and the highly base-paired hairpin that encodes the antisense sequence.

[0372] To generate Ebag9-targeting miRNAs, the endogenous miRNA-155 was used. The 21-nucleotide containing antisense sequence within the hairpin structure was exchanged against predicted EBAG9 target site sequences (Table 1).

TABLE-US-00014 TABLE 1 Hairpin sequences directed against the human EBAG9-gene. EBAG9-targeting miRNA antisense sequence H17 (SEQ ID NO 1) 5′-AAATAACCGAAACTGGGTGAT-3′ H18 (SEQ ID NO 2) 5′-TTAAATAACCGAAACTGGGTG-3′

[0373] The resulting miRNA-coding sequences were introduced into a GFP-encoding retroviral MP71 vector at an intronic position using MluI and NsiI restriction sites. The MP71 vector is known for high transduction efficiency and stable transgene expression in primary T cells. As miRNA transcription is regulated by the polymerase II promoter, the highly active 5′ LTR of MP71 can be used to drive miRNA and transgene expression.

[0374] Knockdown Efficiency of EBAG9-Targeting miRNAs

[0375] Efficiency of miRNA-mediated EBAG9 downregulation on the protein level was tested in the human acute T cell leukemia cell line Jurkat J76. Compared to untransduced T cells (UT), an EBAG9 knockdown efficiency of around 80% could be detected for H16, H17, and H18. In contrast, H19 did not effectuate EBAG9 protein downregulation. The MP71-GFP vector without miRNA served as a control. The most efficient miRNAs H17 and H18 were selected for further analysis in primary human T cells.

Application Example

[0376] Human peripheral blood mononuclear cells (PBMCs) were isolated from healthy voluntary donors and CD8+ T cells were enriched by magnetic cell separation (negative selection). CD8+ T cells were activated by stimulation with anti-human CD3 and anti-human CD28 antibodies for 48 hours. Activated human CD8+ T cells were then subjected to two rounds of retroviral transduction and cultured with high IL-2 (100 IU/ml) and IL-15 supplementation (10 ng/ml). On day 13 after CD8+ T cell activation, cytokine supplementation was reduced to 10 IU/ml IL-2 and 1 ng/ml IL-15. Functional assays were performed 48 hours later (FIG. 3). To determine the activity of granzyme A, human CAR T cells (day 15 after activation) were restimulated for 4 hours with plate-bound anti-human CD3 and anti-human CD28 antibodies. Supernatants were analyzed for granzyme A activity by incubation with granzyme A substrate solution. Product concentration correlates with enzymatic activity. Supernatants of CAR T cells transferred to non-coated plates were used as a control for the basal secretion of granzyme A.

[0377] To prove that EBAG9 silencing increases the release of effector molecules like granzyme A from activated T cells, an in vitro release assay was performed. CD8+ T cells from healthy donors were isolated, transduced twice, and cultivated with high IL-2 supplementation for 13 days. Prior to functional in vitro assays on day 15, IL-2 supplementation was lowered for 48 h. On day 15 of culture, granzyme A release was induced by re-stimulation of T cells with anti-human CD3/CD28 antibodies for 4 h. The granzyme A amount released from BCMA CAR-transduced T cells was similar to those of UT. In contrast, T cells transduced with the H18-BCMA CAR construct released 2-fold higher amounts of granzyme A. Likewise, the H18 miRNA also endowed SP6 CAR T cells with enhanced cytolytic effector molecule secretion (FIG. 4).

[0378] Simultaneous Expression of EBAG9-Targeting miRNAs and CARs

[0379] Cloning

[0380] To generate a retroviral vector that allows for the simultaneous expression of an EBAG9-specific miRNA and a CAR, the GFP gene of the miRNA-encoding MP71-GFP vector was exchanged for a CAR cassette using the NotI and EcoRI restriction sites flanking GFP. Using this strategy, a retroviral vector encoding the EBAG9-specific miRNA H18 and the BCMA CAR and a retroviral vector encoding the EBAG9-specific miRNA H17 and the CD19 CAR were cloned (FIG. 5). As a negative control for functional assays, the SP6 CAR without any naturally occurring ligand was combined with the EBAG9-targeting miRNAs H17 and H18.

[0381] CAR Expression

[0382] Retroviral transduction of anti-human CD3/CD28-activated primary human T cells was performed twice using MP71 vectors encoding for the BCMA or the CD19 CAR alone or in conjunction with the EBAG9-specific miRNA H18 or H17, respectively. Cells were cultured with IL-7 and IL-15 supplementation. Staining of the IgG hinge region was performed on day 6 to analyze CAR expression and to determine the transduction efficiency (FIG. 6).

[0383] In Vitro Application Example

[0384] In vitro cytotoxicity assays were performed to investigate the effect of EBAG9 downregulation on the antigen-specific cytolytic capacity of CAR T cells. The in vitro cytolytic activity reports on the release of granzymes and perforin, a secretion process that is controlled by EBAG9.

[0385] CD8.sup.+ T cells from healthy donors were isolated, activated with anti-human CD3 and anti-CD28 antibodies for 48 hours, transduced twice, and cultivated with high IL-2 supplementation for 13 days. Prior to in vitro cytotoxicity assays on day 15, IL-2 supplementation was lowered for 48 h. On day 15, CAR T cells were co-cultured with target cell lines for 4 hours. Assay supernatants were counted for [.sup.51Cr]-chromium released by lysed target cells. Target cell maximum release was determined by directly counting labeled cells. Spontaneous release was measured by incubating target cells alone.

[0386] The BCMA.sup.high-expressing multiple myeloma (MM) cell line OPM-2, as well as the BCMA.sup.low-expressing B-cell non Hodgkin lymphoma (B-NHL) cell line DOHH-2, were used as target cells for BCMA CAR T cells. Prior to co-cultivation with CAR T cells on day 15 of culture, target cells were incubated with [.sup.51Cr]-chromium. After 4 h of co-cultivation with different effector to target ratios, cytolytic activity was observed in BCMA CAR-transduced CD8+ T cells, whereas no or little activity could be detected in UT or SP6 CAR T cells. Therefore, no non-specific T cell activation occurred upon EBAG9 downregulation. At the highest effector to target ratio of 80:1, target cell lysis of BCMA CAR T cells was around 30%. The combination of the BCMA CAR with EBAG9 silencing in H18-BCMA CAR T cells led to a significant increase in CAR T cell-mediated cytolytic efficiency in all cell lines tested. For example, in the MM cell line OPM-2, H18-BCMA CAR T cells had a lysis rate approximately 1.5-fold higher than the BCMA CAR only. In a different calculation, the maximal killing rate of BCMA CAR-transduced T cells (E:T 80:1) could be achieved with only one-quarter to one-eighth of EBAG9 knockdown BCMA CAR T cells. Thus, effective dose levels were substantially decreased.

[0387] To confirm the RNAi-mediated increase in CAR T cell cytotoxic activity, another miRNA sequence, H17, was used in a CD19 CAR. H17 target the same region within the open reading frame of the EBAG9 gene as H18. As for the BCMA CAR, in vitro cytotoxicity assays were performed. Transduction rates of retrovirally transduced CD8+ T cells were adjusted to around 15% using UT. The CD19.sup.high-expressing B-NHL cell line JeKo-1 was used as target cells in a chromium release assay. After 4 h of co-cultivation, almost no lysis activity could be detected for UT and the control H17-SP6 CAR T cells. In Jeko-1 cells, CD19 CAR-transduced T cells effectuated a specific target cell lysis of about 20% (E:T 80:1). Consistent with the previous results, EBAG9 silencing endowed CAR T cells with a substantial gain in killing activity. RNAi-mediated T cell engineering resulted in a cytotoxicity increase of 2-fold. Furthermore, to achieve maximal lysis of JeKo-1 cells by CD19 CAR T cells, only one-fifth to one-eighth of the H17-CD19 CAR T cells were required (FIG. 8)

In Vivo Application Example

[0388] To translate the findings of the functional in vitro assays into an in vivo model, a multiple myeloma xenograft model was established. A suitable animal model for the xenotransplantation of human multiple myeloma cell lines and primary human CAR T cells is the immunodeficient NOD scid gamma-chain deficient (NSG) mouse strain. These mice do not have mature T or B cells. In addition, NK cell differentiation is blocked. NSG mice were inoculated with the BCMA-expressing multiple myeloma cell line MM.1S. Tumor progression was monitored by bioluminescence imaging of the MM.1S cell line stably expressing a firefly luciferase-eGFP construct. Tumor cell engraftment was detected by IVIS imaging 6 days after transfer. A single dose of 1×10e6 CAR+ T cells on days 10-13 of culture with IL-7/IL-15 supplementation was injected i.v. one day later. Tumor development was followed by serial IVIS imaging until day 14 after transfer. Mice were sacrificed on days 15-16. Bone marrow was analyzed by flow cytometry for the number of remaining tumor cells and CAR T cells.

[0389] Serial IVIS imaging revealed rapid tumor growth between days 6 and 14. The highest specific luciferase signal, which correlates with tumor activity, could be localized to the bone marrow. Treatment with the non-targeting H18-SP6 CAR T cells was unable to control tumor growth. The highest tumor burden was observed in mice from this group. Hence, there was no antigen-independent T cell activation due to EBAG9 silencing and subsequently increased ability of effector molecule release. Mice treated with BCMA CAR T cells showed less tumor progression. However, clinical efficacy at this low number of effector CAR T cells was modest. In contrast, mice that received H18-BCMA CAR T cells showed almost no tumor signal (FIG. 9A-B). Accordingly, when analyzing tumor cell numbers (GFP+CD138+) in bone marrow, the prime niche for myeloma cell homing, tumor cell quantitation revealed 2-fold higher numbers in the H18-SP6 CAR control group compared to the BCMA CAR group. Notably, almost no tumor cells were present in mice treated with RNAi-mediated EBAG9 silencing in BCMA CAR T cells. Altogether, RNAi-mediated downregulation of EBAG9 led to a strongly increased antitumor efficiency even at low effector cell numbers (FIG. 9C).

[0390] CRISPR-Mediated EBAG9 Knockout

[0391] In the examples above, EBAG9 downregulation was demonstrated via transducing cells with a retroviral vector that allows for expression of an EBAG9-specific miRNA. Such vectors were also employed for the simultaneous expression of the EBAG9 miRNA together with an antigen-specific CAR construct in transduced T cells. The enhanced cytolytic capacity of CAR T cells with downregulated EBAG9 was demonstrated, both in vitro and in vivo.

[0392] To complement the examples above, additional means for downregulating EBAG9 have been assessed. The following example employs CRISPR-mediated EBAG9 knockout, demonstrating that EBAG9 can be effectively removed or reduced from treated T cells via CRISPR, thereby enabling additional means for inhibiting EBAG9 in a cytotoxic T cell.

[0393] Different guide RNAs (gRNAs) targeting the exon 4 of the human EBAG9 gene were generated (E1-E4) in order to knockout EBAG9 in primary human T cells. Human peripheral blood mononuclear cells (PBMCs) were isolated from healthy voluntary donors and CD3+ T cells were activated by stimulation with human anti-CD3/CD28 Dynabeads. Activated human CD3+ T cells were then subjected to electroporation with gRNAs and Cas9 protein and cultured under IL-2 supplementation. On day 9 after CD3+ T cell activation, genomic DNA was isolated and protein lysates were generated. Analysis of gene editing efficiency in the EBAG9 locus via TIDE analysis (tracking of Indels by decomposition) as well as western blot analysis of EBAG9 protein level revealed a 50% knockout efficiency for the gRNAs E2, E4 and E6. In contrast, gRNAs E1, E3 and E5 show just a minor decrease in the endogenous EBAG9 expression level. CD3+ T cells electroporated with a non-targeting gRNA served as control (ctl). Results are shown in FIG. 10. Thus, CRISPR-mediated gene editing can target the human EBAG9 locus and represents a viable means for EBAG9 inhibition in cytotoxic T cells expressing a transgenic antigen-targeting construct.

[0394] In vitro cytotoxicity assays to investigate the effect of CRISPR-mediated gene editing of EBAG9 on the antigen-specific cytolytic capacity of CAR T cells are ongoing and are expected to demonstrate an enhanced cytolytic activity of such EBAG9-CRISPR-edited CAR T cells via the release of granzymes and perforin, as shown for miRNA-mediated EBAG9 downregulation, as described above.