T lymphocytes redirected against the kappa light chain of human immunoglobulin efficiently kill mature B lymphocyte-derived malignant cells

Abstract

There has been interest in generating T cells expressing chimeric artificial receptors (CARs) targeting CD19/CD20 antigens to treat B-cell lymphomas. If successful, however, this approach would likely impair humoral immunity because T cells may persist long-term. Most low-grade lymphoma and chronic lymphocytic leukemia (B-CLL) cells express monoclonal immunoglobulins carrying either kappa or lambda light chains. We, therefore, explored whether T lymphocytes could be genetically modified to target the tumor-associated light chain, sparing B lymphocytes expressing the reciprocal light chain, and consequently reduce impairment of humoral immunity. We found that T lymphocytes expressing the anti-kappa light chain CAR showed cytotoxic activity against Igkappa(+) tumor cell lines and B-CLL cells both in vitro and in vivo. We also found that the incorporation of the CD28 endodomain within the CAR enhanced the in vitro and in vivo expansion of transgenic T cells after tumor-associated antigen stimulation. Free Igkappa(+) did not compromise the ability of redirected T lymphocytes to eliminate Igkappa(+) tumors because these free immunoglobulins served to sustain proliferation of CAR-CD28 transgenic T cells. Thus, adoptive transfer of T lymphocytes targeting the appropriate light chain could be a useful immunotherapy approach to treat B-lymphocyte malignancies that clonally express immunoglobulin without entirely compromising humoral immunity.

Figure 1.

T lymphocytes redirected to express κ light chain-specific CAR kill Igκ⁺ tumor cell lines. (A) Immunophenotype of tumor cell lines stained with CD19, anti-κ and anti-λ light-chain–specific antibodies. (B) Cytotoxic activity of T lymphocytes obtained from healthy donors and transduced either with control-GFP vector (□) or CAR46ζ () or CAR46/28ζ (▪). Cytotoxic activity was evaluated in a standard ⁵¹Cr-release assay, and results are shown at an E/T ratio of 20:1. Data represent the mean ± SD of 6 different donors.

Figure 2.

T lymphocytes redirected to express the κ-specific CAR eliminate Igκ⁺dim tumor cells. T lymphocytes obtained from healthy donors and transduced either with control-GFP vector (A), CAR46ζ (B), or CAR46/28ζ (C) were cocultured with the CCL-120 tumor cell line (ratio, 5:1), which is Igκ^+dim. After 5 to 7 days of culture, cells were collected and stained with CD3-PerCP and CD19-PE to evaluate the growth of CD19⁺ tumor cells. No CD19⁺ cells were detectable after coculture with CAR46ζ⁺ or CAR46/28ζ⁺ T lymphocytes, whereas CD19⁺ cells were detectable when tumor cells were cocultured with control T cells.

Figure 3.

T lymphocytes from B-CLL patients expressing the κ-specific CAR kill primary κ⁺ B-CLL cells and expand when CD28 endodomain is present in the CAR. CD3⁺ T lymphocytes were isolated from patients with κ⁺ B-CLL. After activation with CD3/CD28 antibodies, T lymphocytes were transduced either with CAR46ζ or CAR46/28ζ or control GFP vector. After transduction, T lymphocytes were stained either with CD4-PE or CD8-PE antibodies and with Fc-γCy5 antibody to detect the expression of the CAR. Panel A illustrates the profile for Fc-γCy5 antibody in control T lymphocytes (top panels) and CAR46/28ζ⁺ T lymphocytes (bottom panels). Both CD4 and CD8 cells were transduced. Control, CAR46ζ⁺ and CAR46/28ζ⁺ T lymphocytes were stimulated weekly with autologous B-CLL cells (ratio, 1:1) without exogenous cytokines. Panel B illustrates that only CAR46/28ζ⁺ T lymphocytes expanded (for at least 3 weeks) after antigenic stimulation. Panel C shows that CAR46/28ζ⁺ T lymphocytes (▪) released significantly more IFN-γ (P = .01) and IL-2 (P = .02) than CAR46ζ T cells () or control T cells (□) after stimulation with autologous B-CLL cells. Data represent mean ± SD of 4 different donors. The specificity of expanded T lymphocytes was then evaluated using a CFSE-based cytotoxicity assay (E/T ratio, 20:1). Panel D illustrates that CAR46/28ζ T lymphocytes (▪) killed autologous and allogeneic κ⁺ B-CLL cells, but not allogeneic κ⁺ B-CLL cells. In contrast, no significant killing was observed for control transduced T lymphocytes expanded in the presence of exogenous IL-2 (□). Data represent mean ± SD of 3 different donors.

Figure 4.

Soluble immunoglobulins do not impair the capacity of transgenic T cells to eliminate Igκ⁺ tumor cells. The inhibitory effect of free immunoglobulins was tested using a standard 4-hour ⁵¹Cr-release assay, in which either T lymphocytes CAR46ζ⁺ () or CAR46/28ζ⁺ (▪) were incubated with Daudi cells (E/T ratio, 20:1) in the presence of serial dilutions of plasma obtained from healthy donors. The top panel illustrates the mean ± SD of residual killing by transgenic T lymphocytes in presence of the plasma for 4 different donors. The bottom panels illustrate the effects of coculturing either control T lymphocytes or CAR46ζ⁺ or CAR46/28ζ⁺ T lymphocytes with Daudi cells at a 10:1 E/T ratio, in complete media (top panels) or in 100% human plasma (bottom panels). After 5 to 7 days of culture, cells were stained with CD3-PerCP and CD19-PE to enumerate CD19⁺ tumor cells. CAR46/28ζ⁺ T lymphocytes were able to prevent the growth of tumor cells even in the presence of human plasma. This is representative of 5 independent experiments.

Figure 5.

Soluble immunoglobulins sustain the expansion of CAR46/28ζ⁺ T cells. Control T cells (□), CAR46ζ⁺ T lymphocytes (), and CAR46/28ζ⁺ T lymphocytes (▪) were incubated either with serial dilutions of plasma (A) or purified immunoglobulins (B). After 72 hours, the cells were pulsed with 1 μCi (0.037 MBq) methyl-³[H]thymidine and cultured for an additional 15 hours. Only T cells carrying the CAR46/28ζ proliferate in response to soluble immunoglobulins. Data represent mean ± SD for 3 donors. CAR46ζ⁺ (C) and CAR46/28ζ⁺ (D) T lymphocytes were also cultured for 2 weeks with the addition of plasma or soluble immunoglobulins every 3 days. Only CAR46/28ζ⁺ T cells expanded in presence of plasma or purified immunoglobulins. Data represent mean ± SD for 3 donors.

Figure 6.

T lymphocytes carrying CAR46/28ζ expand and persist in vivo after antigenic stimulation. Activated CD3⁺ lymphocytes expressing either CAR46ζ or CAR46/28ζ and double transduced to express eGFP-FLuc gene (5 × 10⁶ cells) were injected intraperitoneally into SCID mice bearing either κ⁺ Daudi cells or λ⁺ SP53 cells (5 × 10⁶ cells) labeled with RLuc gene. No exogenous cytokines were injected into the mice. Survival and expansion of transgenic T cells were monitored using an in vivo imaging system (Xenogen-IVIS Imaging System). Panel A illustrates that the persistence and intensity of the signal measured as maximum photon/sec/cm²/sr (p/s/cm²/sr) was less in mice receiving CAR46ζ⁺ T cells (top panels) than in mice given CAR46/28ζ⁺ T cells (bottom panels). Panel B illustrates the mean ± SD of 4 mice per group (P < .001). This panel also shows that CAR46/28ζ⁺ T cells did not expand/survive in mice bearing the κ^–/λ⁺ SP53 tumor cell line (broken line).

Figure 7.

CAR⁺ T lymphocytes control tumor growth in vivo. Activated CD3⁺ lymphocytes expressing CAR46ζ or CAR46/28ζ, or control T lymphocytes (5 × 10⁶ cells) were injected intraperitoneally into SCID mice bearing κ⁺ Daudi cells (0.5 × 10⁶ cells) labeled with FFLuc gene. No exogenous cytokines were injected into the mice. Day 0 is the day of adoptive transfer of T cells, 10 to 11 days after tumor implant. Tumor growth was monitored using an in vivo imaging system (Xenogen-IVIS Imaging System). Panels A illustrates that tumor growth measured as intensity of the signal (p/s/cm²/sr) was significantly greater in mice receiving control T cells (top panels) by day 25 after T-cell infusion compared to mice receiving CAR46ζ⁺ T lymphocytes (middle panels) or CAR46/28ζ⁺ T cells (bottom panels). Panel B illustrates the mean ± SD of 6 mice/group (P < .001). Panel C compares outcomes in mice bearing κ⁺ Daudi cells labeled with FFLuc gene and treated with either CAR46ζ⁺ or CAR46/28ζ⁺ T cells and receiving intraperitoneal injections of human immunoglobulins 3 times a week or no treatment. Tumor growth was monitored using an in vivo imaging system (Xenogen-IVIS Imaging System). Data are results of 5 mice/group.