Cancer Immunology, Immunotherapy

, Volume 60, Issue 5, pp 609–619

The absence of B lymphocytes reduces the number and function of T-regulatory cells and enhances the anti-tumor response in a murine tumor model


  • Tamar Tadmor
    • Haematology UnitBnai Zion Medical Center
  • Yu Zhang
    • Division of Hematology/OncologyUniversity of Miami Sylvester Comprehensive Cancer Center
  • Hyun-Mi Cho
    • Division of Hematology/OncologyUniversity of Miami Sylvester Comprehensive Cancer Center
  • Eckhard R. Podack
    • Department of Microbiology and ImmunologyUniversity of Miami Miller School of Medicine
    • Department of Microbiology and ImmunologyUniversity of Miami Miller School of Medicine
    • Division of Hematology/OncologyUniversity of Miami Sylvester Comprehensive Cancer Center
Original Article

DOI: 10.1007/s00262-011-0972-z

Cite this article as:
Tadmor, T., Zhang, Y., Cho, H. et al. Cancer Immunol Immunother (2011) 60: 609. doi:10.1007/s00262-011-0972-z


Increasing evidence suggests that B lymphocytes play a central role in inhibiting the immune response against certain tumors, but the underlying mechanisms by which B cells facilitate tumor growth are still poorly understood. In this study, we investigated how the presence or absence of B cells affects expansion and function of T-regulatory cells (‘T-regs’) in a murine breast tumor model (EMT-6). We compared tumor growth, and the number and function of T-reg cells in wild-type immune-competent mice (ICM) and B-cell-deficient mice (BCDM). Mice were either tumor-naive or implanted with EMT-6 mammary adenocarcinoma cells. Tumor growth was markedly inhibited in BCDM, compared to wild-type mice (ICM). Increased T-reg expansion as defined by CD4+/CD25+/FOXP3+ cells was evident following EMT-6 inoculation in ICM in comparison with non-tumor-bearing mice or compared to BCDM in which tumor had been implanted. The percentage and absolute number of T-regs in the spleen, tumor draining lymph nodes, and tumor bed were significantly reduced in BCDM compared to ICM. T-reg function, measured by suppression and proliferation assays, was also reduced in tumor inoculated BCDM compared to ICM. Our studies indicate that absence of B cells may play a role in augmenting the T-cell anti-tumor response, in part due to effects on T-regulatory cell expansion and function.


B-cell-deficient mice (BCDM)Immune-competent mice (ICM)T-regulatory cells (T-reg)Anti-tumor immunityB lymphocyte


The role of B lymphocytes in tumor immunity has primarily been studied in the context of their role as positive regulators of the immune response through antibody production and T-cell activation [1]. Recently, we and others have shown that B cells can negatively affect anti-tumor responses in several murine tumor models. Implantation of EL-4 T-cell lymphoma and MC38 colon carcinoma in B-cell-deficient mice results in limited tumor growth and spontaneous tumor regression as opposed to robust growth in wild-type mice [2, 3]. The absence of B lymphocytes is associated with enhanced Th1 cytokine and CTL responses [2, 3]; however, the basic mechanism for this increased cellular immune response remains unclear [24].

Other studies have documented the role of T cells termed “T-regulatory cells” (T-regs) in suppressing the immune response. T-regs in tumor-bearing animals and humans with cancer are known to inhibit T effector cell proliferation, differentiation, and activation, thereby blunting anti-tumor immune responses [5, 6].

Several studies have explored the potential interaction between T-regs and B lymphocytes in the tumor environment. B lymphocytes may facilitate the expansion and/or differentiation of T lymphocytes into T-regs capable of suppressing a Th1 response [7, 8]. They may also enhance the recruitment and expansion of T-regs via secretion of chemokines such as CCL4 [9] and may facilitate the generation of T-regs via their function as antigen presenting cells (APCs) [7]. The use of B cells as stimulators in allogeneic mixed leukocyte reaction has shown that they preferentially expand FOXP3+ CD4+ T cells and not FOXP3− CD4+ T cells [8]. B-cell-deficient individuals also tend to develop autoimmune phenomena, which may be partly mediated by a decrease in the number of T-regs [10]. We investigated how the presence or absence of B cells affects both T-reg expansion and the anti-tumor response in the murine EMT-6 mammary tumor model. Our results suggest that B cells may facilitate T-reg expansion and function after tumor implantation, which may in turn contribute to the observed decrease in anti-tumor response.

Materials and methods

Mice, cell lines, and tumor cells

Six- to eight-week-old BALB/c mice (purchased from Jackson Laboratories) and IgM −/− (μ chain knockout) B-cell-deficient mice (BCDM) on the BALB/c background were a gift from Dr Thomas Blankenstein, (Max-Delbrück-Center for Molecular Medicine, Berlin, Germany), maintained and bred in the University of Miami vivarium under standard pathogen-free conditions. All animals were cared for in accordance with the Institutional Animal Care and Use Committee guidelines.

EMT-6 mammary adenocarcinoma were obtained from ATCC and maintained in Iscove’s modification of Dulbecco’s medium containing 10% FBS, 200 lM l-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin (IMDM-10).

For tumor implantation, mice were shaved on the right flank and subcutaneously injected with 106 tumor cells. Tumor diameters were monitored twice a week using dial calipers.

Flow cytometry

The following antibodies: rat anti-mouse B220 (RA3-6B2, FITC-conjugated), rat anti-mouse CD19 (1D3, PE-conjugated), rat anti-mouse CD8α (53–6.7, APC-conjugated), anti-mouse CD127 (SB/199, PE-conjugated), anti-mouse CTLA4-PE (UC10-4F 10–11), and rat anti-mouse CD4 (RM4.5, Per-CP-conjugated) were purchased from BD Pharmingen (San Diego, CA). Rat anti-mouse CD25 (7D4, FITC-conjugated), rat anti-mouse FoxP3 (FJK-16 s, PE-conjugated), isotype control for FoxP3 IgG2a, anti-mouse GITR-PE (DTA-1), anti-mouse OX40-PE (OX-86), anti-mouse CD45RB-PE (C363.16A), and relative isotype controls were obtained from eBiosciences (San Diego, CA).

Single-cell suspensions were obtained from spleen, tumor draining lymph nodes (T-DLN), thymus, excised tumor tissue, or peripheral blood. The cells were blocked with anti-CD16/CD32 antibody (BD Pharmingen, San Diego, CA), followed by staining with the indicated antibodies at optimized concentration on ice for 15 min in the staining buffer (1 × PBS supplemented with 1% BSA and 0.09% Sodium Azide). For intracellular staining of T-reg, CD4 and CD25 were stained as above, and cells subsequently fixed and permeabilized and stained with anti-Foxp3 antibody or corresponding isotype control (eBiosciences, San Diego, CA). Stained cells were analyzed using a FACS scan flow cytometer (Becton–Dickinson, Franklin Lakes, NJ, USA), and data analyzed using Flow software (Tree Star Inc, Ashland, OR).

Calculation of absolute T-reg number

Single-cell suspensions were prepared from BCDM and ICM BALB/c spleen, and the number of splenocytes was counted and multiplied by the fractional percentage of CD4+ CD25+ FOXP3+ cells.

Purification of CD4+ CD25+ T cells and in vitro T-reg proliferation and suppression assays

Single-cell suspensions were prepared from BCDM or ICM BALB/c spleens. CD4+ CD25+/CD4+ CD25-T cells were isolated using the CD4+ CD25+ T-reg isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany).

We measured T-reg proliferation using 3H thymidine and/or BrdU incorporation assays [11, 12]. For 3H thymidine incorporation, triplicate cultures of 5 × 104 CD4+ CD25+ cells/well in 96-well plates were stimulated with plate-bound anti-CD3 (1 μg/ml)(clone 145.2C11) and soluble anti-CD28 (1 μg/ml) (clone 37.51; BD Pharmingen, San Diego, CA). CD4+ CD25+ cells were added to each well as indicated in figure legend 5, in the presence of recombinant human IL-2 at a final concentration of 10 ng/ml (Sigma, St. Louis.); 8, 32, 56, and 80 h later, the cells were pulsed for 16 h with 1 μCi of 3H thymidine. Radioactive 3H thymidine incorporation was measured by standard liquid scintillation counting, and results were expressed as counts per minute (cpm). In vitro bromodeoxyuridine (BrdU) incorporation was assayed using an APC BrdU flow kit per the manufacturer’s instructions (BD Pharmingen, San Diego, CA). Briefly, CD4+ CD25+ T-regs were isolated and cultured in triplicate using 5 × 104 CD4+ CD25 +cells/well on 24-well plates bound with 1 μg/ml anti-CD3 (clone 145.2C11) and 1 μg/ml anti-CD28 (clone 37.51; BD Bioscience, San Diego, CA). These were incubated in complete medium in the presence of recombinant human IL-2 at a final concentration of 10 ng/ml (Sigma, St. Louis.). To label cells, 10 μM of BrdU solution was added to each ml of tissue culture media for 4 h. Labeling with BrdU was performed following 48, 72, and 96 h of incubation and followed by staining of surface antigens by anti-CD4-PercP (BD Pharmingen) and intracellular staining for FOXP3+ FITC/BrdU+ APC and labeled cells analyzed by flow cytometry.

Suppression assay

Purified T-reg cells (CD4+ CD25+) were added in decreasing ratios (4:1, 2:1, and 1:1) to a constant number of T effector cells (CD4+ CD25−, 5 × 104 cells/well). A combination of 0.5 μg/ml plate-bound anti-CD3 (clone 145–2C11) and 5 μg/ml plate-bound anti-CD28 (clone 37.51; bd Bioscience, San Diego, CA) provided the polyclonal stimulus for proliferation over a 6-day culture period. Thereafter, 1 μ Ci 3H thymidine was added at day 5 for the final 6 h of culture to assess proliferation. Suppression was determined by the amount of reduction of 3H thymidine incorporation relative to control cultures without added T-regs.

CTL assays

Splenocytes from 28-day tumor-bearing BCDM or ICM were cocultured with mitomycin-C-treated EMT-6 tumor cells for 7 days at splenocyte: tumor cell ratio 8:1 in the presence of recombinant human IL-2 (50 U/ml). Effector cells obtained from 7-day cocultures were tested for CTL activity by standard chromium release assay as follows: tumor target cells were treated with mitomycin-C at 5 × 106 cells/100 μg mitomycin-C 1 ml medium for 30 min and washed three times with PBS followed by 51Cr labeling. One million mitomycin-C-treated tumor cells/100 μCi 51Cr were incubated for 1 h and washed extensively. 10451Cr-labeled tumor cells were seeded into each well, and 100 μl of effector cells were added at the indicated E: T ratio and incubated for 4 h. The plate was centrifuged; 100 μl of supernatant harvested, and 51Cr release was measured using a gamma counter (Packard Instruments, Albertville, MN). Wells containing target cells alone were used for measuring spontaneous release, whereas total incorporated radioactivity was determined by lysing the labeled target cells with 2% Triton-X 100 before harvesting. % specific lysis was calculated as = [experimental release-spontaneous release]/[maximum release-spontaneous release] × 100.

Depletion of T-reg cells in vivo

In vivo depletion of T-regs was performed as previously described [13, 14]. Briefly, mice were divided into four treatment groups: two groups (five ICM and five BCDM) underwent T-reg depletion by intraperitoneal injection of 100 μg PC61, the other two groups using PBS as control (five ICM and five BCDM). Mice were injected intraperitoneally with 100 μl of PC61 [15], a rat IgG1 anti-CD25 antibody (generated in the core facility at UM) or with PBS starting at day-3 before tumor implantation and repeated every 3 days until day +29 following tumor inoculation. Depletion of CD4+/CD25+ T-reg cells was confirmed by flow cytometry.

Statistical analysis

Statistical analysis was conducted using SPSS software. Statistical differences between groups were analyzed by ANOVA with post hoc pair wise group comparisons. Student’s t test was also used for comparisons between two groups. A value of P < 0.05 was considered statistically significant.


Increased tumor rejection of EMT-6 tumors in BALB/c mice lacking B cells is associated with enhanced anti-tumor Th1 cytokine secretion and CTL response

In previous studies, we compared the growth of MC38 and EL4 tumors in C57BL/6 immune-competent mice (ICM) and in B-cell-deficient mice (BCDM) and demonstrated reduced tumor growth and rejection of both EL4 and MC38 tumors [2]. In both tumor models, increased rejection of tumors in mice lacking B cells was associated with enhanced anti-tumor Th1 cytokine secretion and enhanced CTL response. Because C57BL/6 mice are known to manifest a tendency toward the development of Th1 responses, we decided to study responses to the EMT-6 mammary carcinoma models in BALB/c mice which preferentially mount enhanced Th2 type responses.

We compared the growth of the syngeneic EMT-6 tumor in ICM and BCDM. Mice were injected subcutaneously with 106 EMT-6 cells in the right flank, and tumor growth was monitored twice a week. B cells were assessed using flow cytometry for CD19+/B220+ cells (Fig. 1b) EMT-6 mammary tumors grew in all mice, but after days 10–14 grew progressively in ICM, while growth was significantly slower and/or tumors totally regressed in BCDM (Fig. 1a P < 0.01).
Fig. 1

Increased rejection of EMT-6 tumors in BALB/c mice lacking B cells is associated with enhanced anti-tumor Th1 cytokine secretion and CTL response. a EMT-6 mammary adenocarcinoma cells (1 × 106) were inoculated into BALB/c ICM and BCDM subcutaneously on the flank on day 0, tumor growth was monitored on indicated days. b Flow cytometric analysis of surface markers for B cells: anti-CD19+/anti-B220 + in the peripheral blood, from ICM, and BCDM (on day 0). c CTL activity in mice inoculated with EMT-6. Splenocytes from ICM (triangle) and BCDM (squares) obtained 26-day post-EMT-6 tumor inoculation were restimulated in vitro with mitomycin-C-treated EMT-6 cells for 7 days. CTL activity of effectors against EMT-6 target and non-specific B16 controls (circles) was measured in a 51Cr release assay as described in the "Methods". Data represent 4 mice/group, mean ± SD. d Interferon-γ secretion in splenocyte: tumor cell cocultures: Mice were inoculated with EMT-6 as above and splenocyte; tumor coculture was set up 26-day post-tumor inoculation as outlined in the Methods. Levels of IFN-γ in supernatants of 7-day cocultures were determined by ELISA. Data represent mean ± SEM of 5 mice/group

In order to exclude the possibility of a histocompatibility mismatch being the cause of tumor rejection, we performed skin grafts on BCDM recipients with normal tail skin grafts taken from BALB/c and C57BL/6 donors. The BALB/c skin engrafted well, whereas all the C57BL/6 grafts were rejected (data not shown).

We examined CTL activity in ICM and BCDM. Splenocytes were harvested 26-day post-tumor implantation and restimulated in vitro with mitomycin-C-treated tumor cells for 7 days. CTL activity against EMT-6 targets and non-specific B16 target cells was determined using a standard 51Cr-release assay.

Cultured splenocytes from BCDM displayed significantly greater CTL activity compared to ICM (Fig. 1c). Supernatants collected following 7 days of effector tumor cell coculture were assayed for interferon-γ levels by ELISA. Splenocytes taken from BCDM secreted significantly higher levels of interferon-γ than spleen cells taken from ICM. (Fig. 1d). These results confirmed that EMT-6 grew poorly or was completely rejected in BCDM and that the increased resistance to tumor growth was associated with an enhanced anti-tumor Th1 response.

Absolute number and percentage of T-regs (CD4+/CD25+/FOXP3+) in EMT-6 tumor-bearing mice is higher in ICM compared to BCDM

Other investigators have shown that BALB/c mice have more CD4+/CD25+ T-regs and greater suppression of their CD4+ CD25–responder T cells than C57BL/6 mice [17]. We hypothesized that differences in tumor growth might be attributed to differences in T-reg number in BCDM compared to ICM.

On day 26 post-EMT-6 tumor inoculation, spleens, tumor draining lymph nodes, thymus, tumor bed, and peripheral blood were dissected, and single-cell suspensions analyzed using flow cytometry for T-regs (CD4+/CD25+/intracellular FOXP3+).

A higher percentage of CD4+ cells was present in BCDM, in spleen (38.6% 3± in BCDM vs. 17 ± 3% in ICM), tumor draining lymph nodes (T-DLNs) (66 ± 6% in BCDM vs. 27 ± 6% in ICM), and was similar in the thymus (83 ± 9% in BCDM vs 81 ± 9% in ICM). Despite the greater overall percentage of CD4+ cells, we observed a decreased percentage of CD4+/CD25+/FOXP3+ (T-regs) in the spleen (8.7 ± 3% in BCDM vs. 13.8 ± 2% in ICM, P = 0.006), T-DLNs (9.3 ± 1% in BCDM vs. 16.3 ± 3% in ICM, P = 0.01), but not in thymus (0.6 ± 0.1% in BCDM vs. 2 ± 1.7% in ICM, P = 0.25) (Fig. 2a, c).
Fig. 2

Number of T-regulatory cells (CD4 +/CD25 +/FOXP3 +) in BCDM in comparison with ICM: Cell suspensions were prepared from spleen, T-DLNs, tumor bed, peripheral blood, and thymus 26 days after EMT-6 inoculation. a Example of flow cytometric staining for CD4+-gated cells stained with anti-CD25+/anti-FOXP3+ cells in spleen suspensions isolated from ICM and BCDM. b Absolute number of T-reg cells in the spleen of ICM and BCDM. Single-cell suspensions were prepared from BCDM and ICM BALB/c spleen, and number of splenocytes was counted and multiplied by the % CD4+CD25+FOXP3+. c Spleen, tumor draining lymph nodes, thymus, and peripheral blood were extracted 26-day post-EMT-6 tumor inoculation, and the percentage CD4+/CD25+/FOXP3+ cells in single-cell suspensions from 5 individual mice analyzed by intracellular staining. Data represent mean ± SEM of 5 mice per group. The experiment was repeated three times, and a representative experiment is shown

To compare the absolute number of T-regs in BCDM and ICM, we counted the total number of splenocytes and multiplied this by the percentage of CD4+/CD25+/FOXP3+ cells from each mouse. BCDM demonstrated a significantly lower absolute number of splenic T-regs (P < 0.001) (Fig. 2b). There were no significant differences in absolute numbers and/or percentage of infiltrating T-regs in the thymus between ICM and BCDM.

Modest changes in T-reg were accompanied by pronounced changes in tumor growth, suggesting that alteration in T-reg function, and/or in the ratio of T-reg/T effector cells may have also contributed to the observed changes in growth.

Tumor-induced expansion of T-reg cells is reduced in the absence of B cells

Recent studies suggest that T-reg numbers may be increased in tumor-bearing animals and in humans with cancer compared with healthy controls [18].

We compared levels of T-regs in naive mice and in EMT-6 tumor-bearing ICM and BCDM. On day 0, 15 of 30 mice from each group were inoculated with 106 EMT-6 into the right flank and 26 days later, single-cell suspensions of draining lymph nodes (DLNs) and spleens from naive and tumor EMT-6-bearing ICM and BCDM were analyzed for T-regs (Fig. 3). Spleens and T-DLNs from EMT-6 tumor-bearing ICM showed increased numbers of CD4+CD25+FOXP3+ cells compared to naive ICM, spleen: 15.2 ± 1% in tumor bearing vs 11.9 ± 2% in naive ICM (P < 0.05) DLNs: 13.8 ± 1% in tumor bearing vs. 8.8 ± 1% in naive ICM, respectively (P < 0.006) These differences were statistically significant and in agreement with published data on tumor-induced expansion of T-regs [18]. In contrast, little or no expansion of T-regs was observed in the spleens and/or DLNs of BCDM at different time points following tumor inoculation in comparison with naive BCDM. (Spleen: 10.1 ± 0.2% in tumor inoculated vs. 10.8 ± 1% in naive BCDM) (DLNs: 8.4 ± 1 vs. 7.0 ± 1%, respectively) (Fig. 3). These differences were not statistically significant.
Fig. 3

Tumor-Induced Expansion of Regulatory T cells in ICM and BCDM: Comparison of T-reg number in naive and tumor (EMT-6)-bearing mice 26-day post-EMT-6 inoculation, single-cell suspensions were prepared from DLNs and spleens from tumor-naive and EMT6-bearing ICM and BCDM and analyzed by intracellular staining flow cytometry for T-reg cells (anti-CD4+/anti-CD25+/anti-FOXP3+%). Data represent mean ± SEM of 5 mice per group 26 days following tumor inoculation

T-regs isolated from BCDM challenged with tumor have decreased proliferative capacity in comparison with T-regs isolated from ICM

To examine whether T-regs in ICM compared to those from BCDM also differed in functional capacity, we analyzed the in vitro proliferation of CD4+ CD25+ cells isolated from spleens of EMT-6 tumor-bearing mice. 106 EMT-6 cells were inoculated into the right flanks of ICM and BCDM, and CD4+/CD25+ T cells were isolated from spleens 26 days later. The purity of CD4+ CD25+ T-regs, as analyzed by FACS, was >95%.

To assess proliferative capacity, T-regs were incubated with plate-bound anti-CD3 (1 μg/ml) and anti-CD28 (5 μg/ml) antibodies plus recombinant human IL-2 (100 IU/ml). Proliferation was tested using flow cytometric analysis of BrdU uptake in combination with anti-CD4 surface staining and intracellular anti-FoxP3 staining, following 48, 72, and 96 h of incubation.

We also assessed proliferation by measuring 3H thymidine incorporation following 24, 48, 72, and 96 h of incubation (Fig. 4) Measurement of 3H thymidine incorporation enabled us to assess proliferation, while the BrdU assay allowed definitive identification of proliferating T-reg populations using intracellular staining for FOXP3, in combination with CD4+ staining and BrdU uptake.
Fig. 4

T-reg proliferation capacity in ICM and BCDM. a CD4+/CD25+ (T-reg 5 × 104) isolated from ICM and BCDM spleens were incubated with a combination of 1 μg/ml plate-bound anti-CD3 and 1 μg/ml plate-bound anti-CD28 plus recombinant human IL-2 at a final concentration of 10 ng/ml. Flow cytometric analysis of T-reg proliferation is shown using BrdU staining in combination with anti-CD4 surface staining and anti-FOXP3 intracellular staining following 48, 72, and 96 h of incubation. b. Representative staining of splenocytes isolated from ICM, following 96 h of incubation for BrdU uptake gated on Treg, CD8, and B (CD19)-cell markers, demonstrating that after T-reg separation, contamination by CD8 and CD19 was minimal and was not responsible to the differences in the proliferation of T-regs isolated from ICM or BCDM. The two bottom panels are derived from an ICM. c. Proliferation assessed by 3H thymidine incorporation after 24, 48, 72, and 96 h of incubation. The data are expressed as the mean of triplicates, representative of three experiments with similar results. MEAN ± SEM

Less than 1% uptake by B cells or CD8+ T cells in the purified CD4+ populations isolated from ICM and BCDM could be detected using staining for CD19+/BrdU+ and CD8+/BrdU+ cells (Fig. 4b). Both BrdU uptake and 3H thymidine incorporation were markedly decreased in T-reg populations isolated from BCDM compared to ICM-(Fig. 4a–c).

Reduced ability of T-regs derived from BCDM to suppress proliferation of T effector cells

The ability to suppress the proliferation of responding T effector cells is a hallmark of normal T-regs. A proliferation suppression assay was performed to compare the relative suppressive capacity of CD4+ CD25+ T cells isolated from ICM and BCDM.

At a T-reg to T effector ratio of 4:1, 2:1, and 1:1, T-regs from BCDM suppressed proliferation less efficiently than those derived from ICM. (P = 0.001) (Fig. 5). Differences in function of T-regs isolated from naive ICM and BCDM were not statistically significant (data not shown); this suggested a qualitative difference between T-reg from tumor-bearing ICM vs. BCDM. T-regs from EMT-6 tumor-bearing BCDM appeared to have diminished ability to suppress T effectors in vitro.
Fig. 5

In vitro assay of T-reg suppressive function in ICM and BCDM. T-regs (CD25+CD4+) were purified from EMT6-tumor-bearing mice (BCDM or BALB/c mice 30-day post-tumor inoculation) at indicated ratio (Treg: Teff) were cocultured with naive Teff (CD4+CD25-, 5 × 104 cells/well) in the presence of plate-bound anti-CD3 (0.5 ug/ml) and soluble anti-CD28 (5 ug/ml) for 6 days. cRPMI was also added to each well to reach a total volume of 200 μl. 1 μCi 3H Thymidine was added to the culture for the last 6 h, Data represent triplicate mean ± SD. The experiment has been repeated 3 times with similar results

Effects of T-reg depletion on tumor growth

In order to understand the relative effect of T-reg depletion on tumor growth, we directly depleted CD4+/CD25+ cells using an anti-CD25 (a monoclonal anti-IL2 receptor α chain) antibody, PC61.

ICM and BCDM underwent T-reg depletion by intraperitoneal injection of PC61 administered every 3 days. T-reg depletion was analyzed in the peripheral blood of the PC61-treated groups on day 0. (Results from day 0: ICM: PBS treated 14.0 ± 1%, PC61 treated: 0.3 ± 0.1%, BCDM: PBS treated: 8.8 ± 1%, PC61 treated: 0.5 ± 0.1%) (Fig. 6a). Markedly reduced EMT-6 tumor growth was seen for groups treated with PC61 compared to PBS treatment, while absence of B cells and T-reg depletion resulted in the greatest overall decrease in tumor growth (Fig. 6b). We note that the initial decrease in tumor growth following PC61 administration was followed by a gradual increase in the growth rate suggesting as previously reported that schedule of depletion might be crucial and that continued CD25+ depletion may also have affected the responding effector population as has been described related to timing of PC61 administration [14].
Fig. 6

Effects of T-reg depletion on growth of EMT-6 in ICM and BCDM: Anti-CD25 PC61 antibody (100 μg) or PBS was injected I.P. at 5-day intervals to deplete T-regs starting 3 days prior to tumor inoculation (see “Materials and methods”). a Flow cytometric analysis of surface markers for T-regs: anti-CD4+/anti-CD25+/FOXP3+ in the peripheral blood on day 0. b EMT-6 tumor growth was monitored measuring tumor diameters on indicated days. Data represent mean ± SEM of 5 mice/group. Similar results were seen in three separate experiments. A representative experiment is shown

T-reg phenotype in ICM and BCDM

To discern phenotypic changes, we performed flow cytometry analysis of CD4+/CD25+ cells purified from spleens of tumor-free and tumor-bearing ICM and BCDM.

CD4+/CD25+ T cells were stained for surface markers associated with T-regs including GITR, CTLA-4, CD44, CD45RB, CD62L, CD69, and CD103. The expression of the above markers of T-regs from ICM was indistinguishable from that of BCDM (Fig. 7).
Fig. 7

Phenotypic characterization of T-reg in ICM and BCDM. Flow cytometry analysis of CD4+/CD25+ cells purified from naive spleens of ICM and BCDM. Similar results were obtained from tumor-bearing mice (data not shown) CD4+/CD25+ isolated T cells were stained for intracellular FoxP3 and the indicated markers including CD45RB, GITR, CTLA-4, CD69, CD62L, CD103, and CD44. (Blue line = isotype, red solid = BCDM, green line = ICM)


We and others have reported the results of murine studies relating to the effects of B lymphocytes on the evolution of the anti-tumor immune response [2, 3, 19]. In earlier studies, we described the reduced growth of EL-4 lymphoma and MC38 mammary carcinoma in BCDM compared to immune-competent mice (ICM) [2]. Similar results were also reported by Qin and coworkers using the TS/A model, who also described an enhanced cytolytic response to tumor vaccination in B-cell-deficient mice, reversible after the adoptive transfer of B cells [19]. In this report, we describe reduced growth of EMT-6 and enhanced CTL response in Balb/C-derived BCDM. The mechanism for the enhanced CTL response and subsequent immune rejection of tumors in BCDM is still poorly understood.

B lymphocytes are recognized for their positive regulatory role in the immune response, functioning as antibody producing cells, antigen presenting cells (APCs), and in T-cell activation [1]. Recent reports have demonstrated that B cells can also function in a negative regulatory capacity, and a subpopulation of B cells termed B-regulatory cells (B-regs) has been identified in murine models [2023]. We observed a markedly reduced immune response in ICM compared to B-cell-deficient mice suggesting that B cells may play a similar suppressive role in the EMT-6 tumor model.

In this respect, it is known that T-regs can decrease the host immune response thereby favoring tumor growth [5, 6]. CD4+ CD25+ T cells are increased in the peripheral blood and the tumor bed of patients with different types of cancer [6, 24, 25]. In addition, anti-tumor responses have been augmented by depleting T-regs using anti-CD25 monoclonal antibodies in several mouse models [13, 14], and Daclizumab, a humanized anti-CD25 antibody, is being tested for effects on tumor growth in humans [26].

In the experiments reported here, we examined whether the presence or absence of B lymphocytes influenced the number and function of T-regs present in naive and tumor-bearing mice. As opposed to prior studies in B-cell-deficient mice using the C57BL/6 murine model [2], we studied responses in BALB/c mice because they tend to mount a reduced Th1 immune response compared to C57BL/6 mice [16] and have considerably more CD4+ CD25+ T-regs in the thymus and peripheral lymphoid organs [17]. Similar to the EL-4 and MC38 models [2], BALB/c BCDM showed increased resistance to EMT-6 tumor growth and to TSA carcinoma growth (data not shown, and reference 19), compared to ICM and an augmented immune response manifested by increased CTL activity and interferon-γ production. We now demonstrate enhanced T-reg expansion in response to tumor implantation and more potent T-reg function in wild-type mice compared to BCDM.

The reasons for reduced CTL activity in splenocyte-tumor cell cocultures is under study and may relate to functional differences in CD8+ cells or altered composition within responding splenocyte populations induced by the presence of B cells including increased T-reg expansion in vitro or changes within the responding CD8+ and CD4+ cells. Splenocytes taken from BCDM secreted significantly higher levels of interferon-γ than spleen cells taken from ICM (Fig. 1d). In contrast, we did not see major differences in levels of IL-4, IL-6, or IL-13 secreted into the supernatant of tumor—splenocyte cocultures (data not shown).

Increased rejection of primary tumors in mice lacking B cells has been reported previously [2, 3]. The postulated mechanisms involved may include decreased B-cell IL-10 production, increased T-cell infiltration in the tumors, augmented CD8+ cytotoxic T-cell responses, higher levels of NK activity, and polarization toward a Th1-enhanced cytokine response, particularly INF-γ secretion [2, 27]. Our studies suggest that decreased generation of T-reg may also contribute to increased anti-tumor activity in BCDM.

The CD4+ CD25+ FOXP3+ subpopulation is regulated by several mechanisms: T-regs are first produced in the thymus (so-called “naturally occurring T-regs”) and further expanded in the periphery, from resident precursors or through conversion from a reservoir of committed CD4+ CD25-FOXP3+ cells into so-called “de novo T-regs” [28] Interaction between T-regs and B lymphocytes has been described, but the exact relationship between them remains unclear. B-cell antigen presentation may play a role in the generation of T-regs and may participate in T-cell priming, leading to partial tolerance [7]. B-cell stimulation can also induce the generation of T cells with a regulatory phenotype in a contact-dependent manner [7]. Chen and colleagues performed mixed lymphocyte reactions (MLRs) using allogeneic CD4+ T cells isolated from BALB/c mice mixed with either B cells or DC cells isolated from C57BL/6 mice as APCs and demonstrated that overall T-cell responses were greater in MLRs primed with DCs than with B cells [8, 29]. A higher percentage of CD4+ FOXP3+ was obtained in the MLRs with B cells, and they concluded that B cells preferentially induce the generation of allogeneic Foxp3+ T cells [8]. Suto et al. first noted that in spleens of non-tumor-bearing B-cell-deficient mice the absolute number and frequency of CD4+ CD25+ T cells was significantly decreased [29]. This is similar to the results obtained in our studies, where decreased numbers of T-regs were seen in spleens and T-DLNS in BCDM compared to ICM.

Our results also suggest that tumor implantation may facilitate expansion of T-regs and are consistent with findings published by Valzasina et al. [18] showing that tumors of different origins induce an increase in T-reg numbers in tumor draining lymph nodes and spleen. In contrast to normal mice, we did not see appreciable expansion of T-regs in BCDM in response to the presence of the tumor.

In addition to effects on T-reg number, we also observed qualitative functional differences in T-regs derived from tumor-bearing mice in the presence or absence of B cells. BCDM-derived T-regs showed reduced inhibition of T-cell proliferation.

To determine whether the lack of the observed suppression was due to decreased T effector response, a series of “crossover” suppression assays was performed in parallel that involved autologous suppression assays and coincubation of T-regs from BCDM with T effector cells derived from ICM, as well as with T-regs taken from ICM mice with T effectors from BCDM. Results from these studies indicate that the origin of the T effector cells does not determine levels of suppression and support the notion that the defect resides within the T-reg population of BCDM (data not shown).

Depletion of B lymphocytes using chimeric anti-CD20 antibodies (rituximab) is used extensively in the treatment of B-cell malignancies [30] and is also a well-established therapy for some autoimmune disorders including systemic lupus erythematosus, rheumatoid arthritis, idiopathic thrombocytopenic purpura, and autoimmune hemolytic anemia [31]. Defective T-reg function has been described in the setting of autoimmune diseases such as in ITP and diabetes [32] or following administration of certain drugs such as imatinib [33]. Whether T-reg function can be altered by rituximab administration for autoimmune diseases and/or lymphoproliferative disorders is not known [34, 35]. There may also be differences due to the congenital absence of B cells vs. effects of B-cell depletion. In preliminary experiments in our laboratory, B-cell depletion with anti-murine CD20 antibody did not result in increased rejection of EMT-6 (data not shown). We are currently investigating the reason for these differences, including the potential effects of residual B cells remaining following incomplete depletion by anti-CD20 antibody. Diminished T-reg function may relate to changes in cytokine secretion or costimulatory molecule expression. Although T-reg functional capacity was altered, no significant differences were detected in the percentage of cells expressing the surface markers: GITR, CTLA-4, OX40, CD45RB CD62L, CD103, and CD127 in ICM and BCDM, respectively (Fig. 7).

The use of B-cell depletion in the treatment of non-B-cell malignancies is limited but there are several published reports [3639]. Chapoval et al. recorded that the administration of a combination of cyclophosphamide and IL-15 resulted in complete regression of lung metastasis in a model of BCDM, whereas wild-type mice showed only partial responses to this regimen [36]. Perricone et al. also demonstrated enhanced efficacy of melanoma vaccines in the absence of B lymphocytes [37]. Rituximab has also been used for patients with renal-cell carcinoma in combination with IL-2, and the combination was well tolerated but response rate did not increase, compared to other retrospective studies [38]. Barbera-Guillem described the use of rituximab in patients with advanced colorectal carcinoma with non-resectable hepatic metastases and pelvic and/or lung metastases [39]. Administration of an intraceliac arterial infusion of a rituximab-containing chemotherapy regimen resulted in regression of metastasis in 5 of 8 patients and reduction of tumor burden in 50% of the patients [39]. In all the above studies, rituximab was used only in advanced disease and in a limited number of patients and because of this definitive conclusions regarding the role and efficacy of B-cell depletion in enhancing therapeutic response cannot be drawn.

Using gene expression profiling to study changes in cytokine expression during T cell–APC interaction, Bystry et al. were able to demonstrate that B cells can also recruit activated T cells via secretion of CCL4, which also led to CD25 and CTLA4 expression, suggesting a general mechanism of T-reg recruitment by B cells via chemokine [9]. B-cell recruitment of T-regs to tumor sites may in part be responsible for the attenuated immune response in wild-type mice compared to BCDM.

Several recent studies have also postulated the existence of B-regulatory subpopulations [20, 21, 23] which also play a role in immune regulation. B cells can be divided into functionally distinct regulatory subsets capable of inhibiting immune response and inducing immune tolerance. B-regs may cause suppression of T-cell function and expansion through secretion of cytokines such as IL-10 and/or TGFβ [21, 22]. In murine models, Tedder has demonstrated a subpopulation of CD1dhiCD5+ B cells with immunosuppressive activity [21]. Whether this B-cell population exerts its immunosuppressive effects primarily through T-reg expansion is not known. In humans, a corresponding ‘B-reg’ subpopulation has yet to be defined, and targeted depletion of the inhibitory B-regulatory population may well provide additional means of augmenting the anti-tumor immune responses.

In conclusion, our data show that absence of B cells may play a useful role in augmenting the T-cell anti-tumor response, in part due to their effects on T-regulatory cell expansion and function.


This study was supported by NIH grant 5-P01-CA-109094-04 and the American Physician Fellowship (APF) for support of TT.

Copyright information

© Springer-Verlag 2011