Expression of B-cell maturation antigen in multiple myeloma (MM) cell lines

The present study employed flow cytometry and immunoblot analysis to determine the expression of B-cell maturation antigen (BCMA) in multiple myeloma (MM) cell lines KMS-12-PE and NCI-H929. A CD269 (BCMA)-APC antibody and anti-human BCMA were used for flow cytometry and immunoblot analysis. Human chronic myelogenous leukemia (CML) cell line K562 was used as BCMA low or negative expression cell line. Flow cytometric analysis revealed surface expression of BCMA in MM cell lines KMS-12-PE and NCI-H929. Conversely, K562 cells did not show any detectable surface BCMA expression (0.8 ± 0.5%) (Fig. 1A, B). The MM cell lines KMS-12-PE and NCI-H929 expressed surface BCMA at 50 ± 4.2% (p < 0.0001) and 95.2 ± 3.1% (p < 0.0001), respectively, in contrast to K562 cells (Fig. 1A, B). Furthermore, the results of the immunoblot analysis were consistent with the surface expression of BCMA detected by flow cytometry. The expression of BCMA in KMS-12-PE and NCI-H929 was significantly higher than in K562 cells (p = 0.0047 and p = 0.0067, respectively) (Fig. 1C, D).

Fig. 1

Expression of B-cell maturation antigen (BCMA) in multiple myeloma (MM) cell lines. The expression of BCMA was evaluated in MM cell lines KMS-12-PE and NCI-H929, as well as in non-MM cell line K562, using flow cytometry and immunoblot analysis. A Flow cytometry histogram demonstrating surface BCMA expression compared to a matched isotype control (light gray). B The percentages of BCMA-positive cells were determined. C Immunoblot analysis demonstrated the presence of glycosylated 26 kDa BCMA, 20 kDa BCMA, and 37 kDa glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as loading control. D Densitometry analysis was performed to quantify the relative BCMA and GAPDH bands on SDS-PAGE using ImageJ software. The data were obtained from 3 independent experiments, and the results are expressed as the mean ± standard error of the mean (SEM) (N = 3). Statistical significance was determined using one-way analysis of variance (ANOVA) with Tukey's post-hoc test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001) (color figure online)

Anti-BCMA-CAR construction and expression in Lenti-X™ HEK293T

The low complete response (CR) rate of CAR T cells, which may be due to decreased efficacy of murine single-chain variable fragment (scFv) anti-BCMA chimeric antigen receptor (CAR) T cells, has motivated efforts to enhance anti-BCMA-CAR T functions. In this regard, we generated two generations of fully human scFv anti-BCMA-CARs: anti-BCMA-CAR3 (CD28/4-1BB/CD3ζ) and anti-BCMA-CAR2 (4-1BB/CD3ζ), the latter of which was generated for comparison purposes (Fig. 2A). Both CARs were produced using the EF1α promoter. The fully human scFv sequence targeting BCMA was obtained from the international publication WO 2016/014565 A2 with BCMA-10 (EG63-98LB; 139109). A c-Myc tag was inserted after the CAR sequence, followed by the CD8 hinge and CD28 or CD8 transmembrane domain (Fig. 2A).

Fig. 2

Lentiviral constructs of anti-BCMA-CARs and protein expression in Lenti-X™ HEK293T cells. A Schematic representation shows second-generation anti-BCMA-CAR (anti-BCMA-CAR2) and third-generation anti-BCMA-CAR (anti-BCMA-CAR3) lentiviral constructs containing fully human anti-BCMA scFv, c-Myc tag, hinge region, transmembrane (TM) domain, co-stimulatory domain(s), and CD3ζ. B Representative histogram demonstrates the expression of anti-BCMA-CARs on the surface of Lenti-X™ HEK293T cells analyzed by flow cytometry using by anti-cMyc-FITC antibody, and C the data were summarized as bar graphs. D Immunoblot analysis of cell lysates of Lenti-X™ HEK293T cells using anti-CD3ζ antibody shows specific bands of anti-BCMA-CAR2 and anti-BCMA-CAR3 at 62 and 66 kDa, respectively, with a loading control of 37 kDa GAPDH. E The densitometry data of anti-BCMA-CAR protein bands relative to GAPDH bands on SDS-PAGE analyzed using ImageJ software were summarized as bar graphs. The data obtained from 3 independent experiments were expressed as mean ± standard error of the mean (SEM) (N = 3). Statistical significance was determined using one-way analysis of variance (ANOVA) with Tukey’s post-hoc test (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001)

To investigate the expression of anti-BCMA-CAR proteins in Lenti-X™ HEK293T cells before T cell transduction, lentiviral constructs containing anti-BCMA-CAR2 or anti-BCMA-CAR3 were transfected into these cells. Flow cytometric analysis revealed a significant increase in surface c-Myc expression on anti-BCMA-CAR2-transfected cells (52.9 ± 1.6%) and anti-BCMA-CAR3-transfected cells (37.7 ± 1.9%) compared to un-transfected Lenti-X™ HEK293T (UTF) cells (4.0 ± 0.2%) (Fig. 2B, C). The expression of CD3ζ was assessed by immunoblotting, as shown in Fig. 2D, E. Specifically, the intracellular CD3ζ immunoblot results displayed a specific band of anti-BCMA-CAR2 and anti-BCMA-CAR3 at 62 and 66 kDa, respectively, with 37 kDa of GAPDH as a loading control (Fig. 2D). Furthermore, when analyzing the densitometry results of anti-BCMA-CAR protein bands relative to GAPDH bands on SDS-PAGE using ImageJ software, the results showed that anti-BCMA-CAR2 and anti-BCMA-CAR3 had significantly different protein expression from un-transfected Lenti-X™ HEK293T cells with p = 0.0465 and p = 0.0001, respectively. Notably, anti-BCMA-CAR3 demonstrated a significantly higher expression of anti-BCMA-CAR protein than anti-BCMA-CAR2 with p = 0.0008 (Fig. 2E). The findings of this study suggest that in the mammalian cell system, both the anti-BCMA-CAR2 and anti-BCMA-CAR3 constructs expressed CAR protein containing CD3ζ.

Generation and characterization of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells

The present study describes the generation of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells via lentiviral transduction of primary human T cells with lentiviral particles carrying genes encoding the respective CARs. Transduction efficiencies were assessed by measuring anti-BCMA-CAR surface expression on T cells, which showed that 4.5 ± 0.2% of untransduced T cells (UTD T) expressed the CAR, while 60.8 ± 9.4% and 39.2 ± 5.0% of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells, respectively, expressed the CAR (Fig. 3A, B). The immunophenotypes of anti-BCMA-CAR T cells were characterized, revealing that more than 90% of the T cell population expressed CD3 (Fig. 3C), while the populations of NK and NKT cells were low and not significantly different among the experimental groups. Interestingly, the population of B cells in PBMCs was reduced in PHA-L activated-T cells, UTD T cells, and anti-BCMA-CAR T cells (Fig. 3C). Further analysis of the CD3 + population showed that cytotoxic T cells (CD8+) were significantly more prevalent than helper T cells (CD4+) in PHA-L activated-T cells (50 ± 7.3% and 28.7 ± 5.2%), UTD T cells (67 ± 5.6% and 20.6 ± 4.3%), anti-BCMA-CAR2 T cells (61.3 ± 6.3% and 16.8 ± 3.8%), and anti-BCMA-CAR3 T cells (65.5 ± 5.1% and 21.3 ± 4.5%) (Fig. 3D). Analysis of the T cell subsets, including naive/stem cell memory, central memory, effector memory, and effector function cells, showed no significant difference between anti-BCMA-CAR T cells and UTD T cells (Fig. 3E). Finally, PD-1 and LAG-3 expression was higher in PHA-L activated-T cells compared to PBMCs, UTD T cells, anti-BCMA-CAR2, and anti-BCMA-CAR3 T cells (Fig. 3F). However, the expression of TIM-3 was increased in all groups, including PHA-L activated-T cells, UTD T cells, anti-BCMA-CAR2 T cells, and anti-BCMA-CAR3 T cells, compared to PBMCs (1.8 ± 0.7%) (Fig. 3F).

Fig. 3

Generation and characterization of anti-BCMA-CAR T cells. A Histogram plots were used to demonstrate the transduction efficiency of chimeric antigen receptor (CAR) constructs to express an anti-BCMA-CAR protein on T cells isolated from peripheral blood mononuclear cells (PBMCs) of healthy volunteer donors. The anti-c-Myc FITC antibody was employed for surface staining. B Surface expression of anti-BCMA-CAR was assessed in a sample size of N = 5. C The immune cell population was characterized based on CD3 expression (T cells), CD3-CD56+CD16+ (natural killer (NK) cells), CD3+CD56+ (natural killer T (NKT) cells), and CD3-CD19+ (B cells). D Helper (CD3+CD4+) and cytotoxic (CD3+CD8+) T-cell percentages. E Flow cytometric analysis was used to investigate the expression of T cell subsets on CD3+ lymphocytes. F Three exhaustion markers, PD-1, LAG-3, and TIM-3 were evaluated in CD3- cells using flow cytometry. Data from 5 individual healthy donors are presented as mean ± standard error of the mean (SEM) (N = 5). Unpaired Student’s t-tests were used to compare two groups, and one-way ANOVA with Tukey’s post-hoc test was employed to determine statistical significance (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001)

Anti-tumor activities of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells against multiple myeloma cells expressing BCMA

This experiment was conducted to investigate the anti-tumor activities of anti-BCMA-CAR2 and anti-BCMA-CAR3 T against BCMA-expressing multiple myeloma (MM) cell lines. K562 cells were utilized as negative target cells (BCMAneg), whereas KMS-12-PE (BCMAlow) and NCI-H929 (BCMAhigh) cells were employed as MM target cells that express BCMA. Target cells were cocultured with either untransduced (UTD) T cells or anti-BCMA-CAR T cells at various effector-to-target (E:T) ratios (1:1, 5:1, and 10:1) for a duration of 12 h. After coculture, the viability of target cells was evaluated using flow cytometry. The outcomes indicated that both anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells displayed minimal cytotoxicity against K562 (BCMAneg) cells in comparison to control UTD T cells (Supplementary Fig. 1). However, anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells exhibited specific dose-dependent killing of BCMA-expressing MM cells, KMS-12-PE, and NCI-H929 cells (Fig. 4A, B). The cytotoxicity of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells against KMS-12-PE cells was observed to be up to 43.8 ± 6.1% and 59 ± 5.4%, respectively, at an E:T ratio of 10:1 (Fig. 4A). The percentage of cytotoxicity against NCI-H929 cells by anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells was noted to be up to 56.7 ± 3.4% and 75.5 ± 3.8%, respectively, at an E:T ratio of 10:1 (Fig. 4B).

Fig. 4

The anti-tumor effect and proliferation activity of anti-BCMA-CAR T cells against multiple myeloma (MM) expressing B cell maturation antigen (BCMA). The killing activities of untransduced (UTD) T cells, anti-BCMA-CAR2 T cells, and anti-BCMA-CAR3 T cells against A KMS-12-PE (BCMAlow), and B NCI-H929 (BCMAhigh) cells were evaluated at different effector to target (E:T) ratios of 1:1, 5:1, and 10:1 over a 12-h co-culture period. The number of viable target cells was then determined using a counting bead and analyzed via flow cytometry. C The histogram and D the percentage of cell proliferation of UTD T cells, anti-BCMA-CAR2 T cells, and anti-BCMA-CAR3 T cells after activation by co-culturing with KMS-12-PE (BCMAlow) cells and NCI-H929 (BCMAhigh) at an effector to target (E:T) ratio of 5:1 for five days in the absence of exogenous cytokines were examined by carboxyfluorescein diacetate succinimidyl ester (CFSE) dilution through flow cytometry. The data were collected from 5 individual healthy donors and presented as mean values with standard error of the mean (SEM) (N = 5). One-way ANOVA with Tukey’s post-hoc test was performed to evaluate the statistical significance of the results, denoted as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Stimulation of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cell proliferation through co-culture with BCMA-expressing multiple myeloma cells

Upon co-culturing with KMS-12-PE (BCMAlow) or NCI-H929 (BCMAhigh) cells at the effector to target (E:T) ratio of 5:1, we examined the proliferation of UTD, anti-BCMA-CAR2, and anti-BCMA-CAR3 T cells by analyzing the dilution of carboxyfluorescein succinimidyl ester (CFSE) in proliferating T cells on day 5. Our results indicate that after co-culturing with KMS-12-PE (BCMAlow) cells, anti-BCMA-CAR2 T cells proliferated significantly higher than UTD T cells (58.0 ± 5.4% and 24.9 ± 4.4%, respectively; p = 0.0022). Interestingly, after co-culturing with KMS-12-PE (BCMAlow) cells, anti-BCMA-CAR3 T cells proliferated even greater than UTD T cells (78.0 ± 6.0% and 24.9 ± 4.4%, respectively; p < 0.0001) (Fig. 4C, D). We observed similar results when co-culturing with NCI-H929 (BCMAhigh) cells, where anti-BCMA-CAR2 T cells proliferated significantly higher than UTD T cells (67.9 ± 4.9% and 9.8 ± 2.3%, respectively; p < 0.0001). Additionally, anti-BCMA-CAR3 T cells proliferation was significantly higher than UTD T cells (77.7 ± 5.6% and 9.8 ± 2.3%, respectively; p < 0.0001) upon co-culturing with NCI-H929 (BCMAhigh) cells (Fig. 4C, D).

Cytokine production of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells in response to multiple myeloma cells expressing BCMA

In this study, we investigated cytokine production in response to multiple myeloma (MM) cells expressing BCMA using KMS-12-PE (BCMAlow) or NCI-H929 (BCMAhigh) cells co-cultured with UTD, anti-BCMA-CAR2, or anti-BCMA-CAR3 T cells at an effector to target (E:T) ratio of 5:1. Cytokine levels were measured by LEGENDplex™ Human CD8/NK cell panel Cytokine Bead Array (CBA) of 13 cytokines and proteins after 24 h of co-culture. Online Resource shows the cytokine concentration of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells against KMS-12-PE and NCI-H929 cells compared to UTD T cells. The data demonstrated that the levels of IL-2 and TNF-α in the culture media of anti-BCMA-CAR3 T cells co-cultured with KMS-12-PE (BCMAlow) cells were significantly increased (561.6 ± 422.3 pg/ml and 288.2 ± 117.6 pg/ml, respectively) compared to UTD T cells (16.8 ± 8.7 pg/ml and 9.8 ± 2.8 pg/ml, respectively) (Fig. 5A). However, there was no significant difference in IFN-γ levels. Additionally, the levels of granzyme A, granzyme B, and granulysin were significantly increased in the culture media of anti-BCMA-CAR3 T cells co-cultured with KMS-12-PE (BCMAlow) cells (5,995.8 ± 2,141.8 pg/ml, 27,248 ± 7,496.4 pg/ml, and 1,579.4 ± 323.8 pg/ml, respectively), compared to the UTD T cells (623.5 ± 82.6 pg/ml, 2352.7 ± 355.5 pg/ml, and 436.7 ± 112.9 pg/ml, respectively) (Fig. 5B). Furthermore, the levels of IL-4, IL-17A, and sFasL were significantly increased in the culture supernatant of anti-BCMA-CAR3 T cells against KMS-12-PE (BCMAlow) cells (7.3 ± 1.8 pg/ml, 50.7 ± 7.5 pg/ml, and 192.3 ± 50.8 pg/ml, respectively), compared to UTD T cells (0.8 ± 0.4 pg/ml, 21.7 ± 1.6 pg/ml, and 51.7 ± 10.5 pg/ml, respectively) (Fig. 5C). These findings suggest that anti-BCMA-CAR3 T cells can produce cytokines and proteins in response to BCMA-expressing MM cells, which could potentially contribute to an anti-tumor immune response.

Fig. 5

Cytokine production levels of anti-BCMA-CAR T cells against multiple myeloma (MM) cell lines expressing B-cell maturation antigen (BCMA). The levels of IL-2, TNF-α, IFN-γ, (A, D) Granzyme-A, Granzyme-B, Perforin, Granulysin (B, E), as well as IL-4, IL-6, IL-10, IL-17A, sFas, and sFasL (C, F) in the cell culture supernatants of anti-BCMA-CAR T cells, were analyzed using cytometric bead array (CBA) after 24 h of activation by culturing with KMS-12-PE (BCMAlow) cells and NCI-H929 (BCMAhigh) cells at an E:T ratio of 5:1. The data were collected from 5 individual healthy donors and presented as mean ± standard error of the mean (SEM) (N = 5). One-way ANOVA with Tukey’s post-hoc test was utilized to determine statistical significance (*p < 0.05, **p < 0.01, ***p < 0.001, and ***p < 0.0001)

The present study also examined the cytokine production of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells during co-culture with NCI-H929 (BCMAhigh) cells. The result demonstrated that the level of TNF-α was significantly elevated in the culture media of anti-BCMA-CAR3 T cells (206.4 ± 72.9 pg/ml) compared to UTD T cells (2.8 ± 0.8 pg/ml) (Fig. 5D). Furthermore, the levels of granzyme A, granzyme B, and granulysin were significantly increased in the culture media of anti-BCMA-CAR3 T cells co-cultured with NCI-H929 cells (BCMAhigh) cells (4,727.8 ± 1,679.1 pg/ml, 29,991 ± 8,511.2 pg/ml, and 14,62.4 ± 388.7 pg/ml, respectively) relative to UTD T cells (193.4 ± 37.2 pg/ml, 1,278.3 ± 334.4 pg/ml, and 270 ± 72.9 pg/ml, respectively) (Fig. 5E). Moreover, the levels of IL-4 were significantly increased in both anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells (5.9 ± 1.1 pg/ml and 5.4 ± 1.3 pg/ml, respectively) relative to UTD T cells (0.6 ± 0.3 pg/ml) (Fig. 5F). Similarly, the level of IL-17A was significantly elevated in both anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells (41.8 ± 8.1 pg/ml and 45.9 ± 9.9 pg/ml, respectively) compared to UTD T cells (9.2 ± 2.7 pg/ml) (Fig. 5F). Additionally, the level of sFasL was significantly increased in the culture media of anti-BCMA-CAR3 T cells (174.9 ± 44.7 pg/ml) compared to UTD T cells (25.9 ± 7.8 pg/ml) (Fig. 5F).

Anti-tumor efficiency of anti-BCMA-CAR2 and anti-BCMA-CAR3 T cells against multiple myeloma cells expressing BCMA in a long-term treatment

In order to evaluate anti-tumor efficiency in long-term treatment, anti-BCMA-CAR2 or anti-BCMA-CAR3 T cells were co-cultured with NCI-H929 (BCMAhigh) cells at E:T ratio of 1:2. The target MM cells were repeatedly added to the CAR T cells at days 3, 6, 9, and 12, and the residual viable NCI-H929 (BCMAhigh) cells were collected to investigate cell viability (Fig. 6A). The co-culturing with UTD T cells resulted in an increased number of tumor cells on days 6, 9, and 12, with percentages of 76.9 ± 15%, 96.3 ± 0.7%, and 96.9 ± 1.1%, respectively. However, anti-BCMA-CAR2 T cells were able to significantly reduce the number of target cells at days 3, 6, and 9 (0.9 ± 0.6%, 9.7 ± 5.5%, and 25.1 ± 13.2%, respectively), with the number gradually increasing by day 12 (36.8 ± 20.1%) (Fig. 6B). Remarkably, anti-BCMA-CAR3 T cells almost completely eliminated the target cells and were significantly different from UTD T cells on days 3, 6, 9, and 12, with the remaining target cells at 4.1 ± 2.1% (p < 0.0001) (Fig. 6B).

Fig. 6

The efficiency of anti-BCMA-CAR T cells against BCMA expressed on multiple myeloma over an extended period. A The schematic overview of the stress test experiment illustrates the co-culture of anti-BCMA-CAR T cells with NCI-H929 target cells, subjected to antigenic stimulation for three cycles at an effector (E) to target (T) ratio of 1:2. Over the 12-day period, the following parameters were assessed: B residual viable BCMA target cells, C proliferation rate, D frequency of transduced CAR+, and the expression of exhaustion markers, including E PD-1, F LAG-3, and G TIM-3. The data collected from 4 individual healthy donors are presented as the mean ± standard error of the mean (SEM) (N = 4). Two-way ANOVA was used to determine statistical significance (*p < 0.05, **p < 0.01, ***p < 0.001, and ***p < 0.0001)

This study also investigated the proliferation of CAR T cells in response to co-culturing with target MM cells expressing BCMA. The absolute effector cell counts were analyzed using a bead-based assay on days 3, 6, 9, and 12. The results demonstrated a significant increase in the proliferation rate of anti-BCMA-CAR3 T cells on day 9 (3.9 ± 0.3-fold, p = 0.0015) compared to the UTD T cells (0.5 ± 0.1-fold), while anti-BCMA-CAR2 T cells showed a higher but not statistically significant increase compared to UTD T cells (Fig. 6C). Additionally, CAR expression in CAR T cells was evaluated by flow cytometry analysis of CD3+ and Myc-tag+ cells. At day 0, CD3+ T cells expressing anti-BCMA-CAR2 and anti-BCMA-CAR3 were detected at 38.7 ± 8.6% and 28.5 ± 1.6% p < 0.001, respectively. Following co-culturing with target MM cells to day 9 and day 12, anti-BCMA-CAR3 T cells showed significantly higher populations (24.1 ± 4.6% and 22.6 ± 4.1%, respectively) compared to UTD T cells (p = 0.0483 and 0.0320), whereas anti-BCMA-CAR2 T cells did not (12.3 ± 2.9% and 13.0 ± 0.8%) (Fig. 6D).

In addition, we also investigated the expression of exhaustion markers (PD-1, LAG-3, and TIM-3) in CAR T cells following prolonged stimulation with target cells. Flow cytometry was used to analyze the expression of these markers. Our findings revealed no differences in PD-1 expression across all conditions (Fig. 6E). However, LAG-3 expression was significantly higher in anti-BCMA-CAR3 T cells on day 3 (66.1 ± 3.4%; p = 0.0205), day 6 (67.1 ± 5.2%; p = 0.0500), and day 9 (69.2 ± 6.3%; p = 0.0238) when compared to UTD T cells (25.5 ± 8.0%, 41.1 ± 6.5%, and 38.5 ± 5.3%, respectively) but did not differ significantly from anti-BCMA-CAR2 T cells (Fig. 6F). Similarly, TIM-3 expression was significantly higher in anti-BCMA-CAR2 T cells and anti-BCMA-CAR3 T cells at day 3 (90.3 ± 2.7% and 92.4 ± 1.1%, respectively) when compared to UTD T cells (55.1 ± 5.7 pg/ml). However, at day 12, the TIM-3 expression was decreased in all groups (anti-BCMA-CAR2 T cells, anti-BCMA-CAR3 T cells, and UTD T cells) (46 ± 14.2%, 61.5 ± 7.2%, and 16.4 ± 7.2%, respectively) (Fig. 6G).