TGF-β family cytokines play multiple roles in immune responses. TGF-β1-null mice suffer from multi-organ infiltration that leads to their premature death. T cells play a central role in the TGF-β1 phenotype, as deficiency of TGF-β1 only in T cells reproduces the lethal phenotype. Although it is known that TGF-β1 controls B cells isotype switch and homeostasis, the source responsible for this control has not been characterized. Because of the major role that T cells play in regulating B cell responses, we addressed the T cell dependency of the TGF-β1 control of B cells. The analysis of T cell-deficient, TGF-β1 knockout mice and the production of chimeras in which B but not T cells lacked TGF-β1 allowed us to show that B cells are controlled in part by cell autonomous production of TGF-β1.

Transforming growth factor-β (TGF-β) family cytokines are implicated in multiple biological processes including embryonic development, cell cycle control, and cell survival (1). The importance of TGF-β1 in controlling immune cells was demonstrated by the deleterious effects of its deletion in mice, resulting in multiple-organ infiltration and premature death (2). This phenotype is largely T cell mediated. Indeed, neither infiltration nor premature death occurs in MHC-deficient or T cell-depleted mice (3, 4, 5). More recently, the production of transgenic mice with a T lymphocyte-specific expression of a dominant negative form of the TGF-βR and, more conclusively, the specific ablation of TGF-βR or TGF-β1 in T cells have demonstrated that this cytokine controls T cells by different mechanisms (6, 7). Natural regulatory T cells (Tregs) require TGF-β1 signaling for their maintenance in the periphery and use TGF-β1 production to control naive T cell activation. In addition, the activation of naive T cells is limited by their own production of TGF-β1. These mechanisms are essential for prevention of wasting systemic inflammatory disease leading to premature death.

TGF also plays a central role in B cell biology. In vitro, TGF-β induces the B cell switch to IgA by controlling the transcription of sterile Igα transcripts and inhibits B cell proliferation and survival. The importance of TGF-β in controlling B cell homeostasis and responses has been firmly established by specifically deleting TGF-βR in B cells (8). This ablation renders B cells unresponsive to TGF-β1, 2, and 3, and results in uncontrolled B cell responses including complete IgA deficiency and increased IgG production. It was proposed that this increase could be due to an arrested class switch recombination at γ1 during sequential class switch from μ to α. In agreement with a role for TGF-β1 but not TGF-β2 or 3 in controlling IgA and IgG switch, TGF-β1-deficient mice show a similar IgG1 increase and IgA deficit (9). TGF-βR deletion in B cells also documented a critical role for TGF in regulating B cell homeostasis as evidenced by the increase in B1 and splenic B cells in these mice. Consistent with the observation made in T cells, the predominant effect of TGF-β on B cell homeostasis occurred via induction of cell death rather than inhibition of cell proliferation. Given the critical role that T cells play in controlling immune responses through TGF-β1 production and in helping B cell responses, it was tempting to speculate that T cell-derived TGF-β1 would regulate B cells. To define the cellular origin of TGF-β1 implicated in B cell phenotypes and in particular the role of T cell-derived TGF-β1, we developed TGF-β1-deficient mice without T cells by crossing them with CD3ε-deficient mice (10). Those mice and chimeras allowed us to define a cell autonomous control of TGF-β1 in B cells.

The manuscript does not contain human studies.

Experiments using mice were handled according to the rules of the french government décret n° 87–848 from October 10th 1987, Paris. Accreditation number for the institute is B13.055.10. Accreditation number for the investigator is 13.70.

TGF-β-deficient mice were obtained from the Mouse Models of Human Cancers Consortium Repository as RAG/TGF knockout (KO)3 and crossed with CD3ε-deficient mice kindly provided by Bernard and Marie Malissen (Centre d’Immunologie de Marseille Luminy, Marseille, France). Mice were analyzed between 4 and 8 wk of age. For bone marrow (BM) chimeras, μMT and CD3ε KO mice were sacrificed between 4 and 8 wk, and tibias and femurs were flushed in PBS. A total of 4 × 106 cells were used to reconstitute sublethally irradiated (3 gray) RAG 2 KO mice. When two different BMs served for reconstitution, 2 × 106 cells of each were injected. Mice were either analyzed 2 mo after reconstitution or immunized.

Immunizations were with either 100 μg of OVA (Sigma-Aldrich) or Bovine type II collagen (Chondrex) in CFA either s.c. (Collagen) or i.p. (OVA) for T-dependent Ags. Mice were re-immunized similarly after 4 wk. T cell-deficient chimeras were immunized either i.p. using 100 μg of TNP-Ficoll or orally by gavage with 200 μg of FITC-Dextran and 10 μg of Cholera Toxin (Sigma-Aldrich). For oral immunization, mice received 3 gavages, one every week, and bled 1 wk later. Blood was collected via retro-orbital bleeding.

Serum and fecal pellets were tested for Ig levels by ELISA using a goat anti-mouse isotyping kit (Southern Biotechnology Associates). Anti-DNA reactivity was determined by ELISA. Plates were blocked with methylated BSA (100 μg/ml) (Sigma-Aldrich) before coating with sonicated salmon sperm DNA (50 μg/ml) (Invitrogen). Sera were tested at 1/100 and Ig revealed using a mixture of anti IgG2a, IgG2b, IgG3, and IgG1 HRP-coupled Abs (Southern Biotechnology Associates).

For cytometric analysis, cells from peritoneal cavity (PC) were stained with FITC-coupled anti-CD11b, PE-coupled anti-B220, allophycocyanin-coupled anti-IgM, biotin-coupled anti-CD5, PerCP CY5.5-coupled streptavidin, PE Cy7-coupled anti-CD45.1, and allophycocyanin Cy7-coupled anti-CD45.2 mAb (BD Pharmingen). For lymph node (LN), mesenteric LN (MLN), and blood B cells, cells were stained with PE-coupled anti-B220, FITC-coupled anti-IgG1, biotin-coupled anti-IgA, PerCP CY5.5-coupled streptavidin, PE Cy7-coupled anti-CD45.1, and allophycocyanin Cy7-coupled anti-CD45.2 mAbs. For Treg analysis, spleen (Sp), LN, and MLN cells were stained using anti-CD4, anti-CD8, anti-CD69, and anti-CD25 mAbs, and the percentage of Treg cells (CD4+CD8CD69CD25+) among CD4-positive cells was plotted.

For BrdU staining, mice were given 1 mg/ml BrdU in the drinking water over a period of 15 days. PC cells were stained for the expression of surface B220, CD11c, CD45.1, and CD45.2. Cells were then fixed, permeabilized using cytofix/cytoperm kit (BD Pharmingen), and stained intracellularly with an anti-BrdU mAb (BD Biosciences). Signals were acquired on a Canto II cytometer and analyzed using FlowJo software.

The MFBF11 TGF reporter cells were kindly provided by Ina Tesseur (Stanford University School of Medicine, Stanford, CA). They consist of TGF-β1-deficient embryonic fibroblasts expressing a reporter plasmid containing SMAD-binding elements driving the expression of secreted alkaline phosphatase (SEAP). Cells were treated as described (11). Sera were used at 1/100 dilution, either untreated or acidified and in the presence or absence of the anti-TGF-β1 Ab 1D11 (BD/Pharmingen) at 10 μg/ml.

In CD3ε-deficient mice (CD3 KO), low levels of IgG1 and IgA are present in the serum. As shown in Fig. 1,A, the mean serum IgG1 concentration is 0.003 mg/ml in TGF wild-type (WT)/CD3 KO compared with 0.56 mg/ml in the presence of T cells (Fig. 1 C). Upon introduction of the TGF-β1 deficiency on the CD3ε-null background, we noted a marked decrease in serum IgA (0.006 mg/ml in TGF KO vs 0.016 mg/ml in TGF WT) and a parallel increase in serum IgG1 (0.01 mg/ml TGF KO vs 0.003 mg/ml in TGF WT). Gut IgA has been shown to be largely produced via T-independent mechanisms (11). Consistently, fecal IgA level was close to normal in CD3ε-deficient mice but the absence of TGF-β1 induced a profound depletion of this isotype. Therefore, TGF-β1 produced by non-T cells controls T cell-independent switch to IgA and IgG1.

FIGURE 1.

Serum and fecal pellets IgG1 and IgA in mice with TGF-β1-deficient B cells. Sera and fecal pellets from CD3KO (A), T-def chim (B), and T-suff chim (C), with either TGF-β1 WT (•) or KO B cells (Δ) were analyzed. The concentrations of IgG1 and IgA in the serum are in mg/ml (Left axis) and the concentration of IgA in the fecal pellet in μg/mg of protein (Right axis). Two-tailed Student’s t test comparing WT and KO are included (p values).

FIGURE 1.

Serum and fecal pellets IgG1 and IgA in mice with TGF-β1-deficient B cells. Sera and fecal pellets from CD3KO (A), T-def chim (B), and T-suff chim (C), with either TGF-β1 WT (•) or KO B cells (Δ) were analyzed. The concentrations of IgG1 and IgA in the serum are in mg/ml (Left axis) and the concentration of IgA in the fecal pellet in μg/mg of protein (Right axis). Two-tailed Student’s t test comparing WT and KO are included (p values).

Close modal

To identify whether B lymphocytes or other tissues were the source of TGF-β1 responsible for IgA and IgG1 production, chimeras were produced by reconstituting sublethally irradiated RAG KO mice with TGF-β1/CD3 KO BM (T-def chim (Fig. 2)). Preliminary reconstitution of Ly5.2 RAG mice with Ly5.1 BM showed that in such mice 90% of myeloid cells are RAG-derived (data not shown). Therefore in T-def chimeras, B cells are all deficient (or all sufficient) for TGF-β1 while the majority of the other hematopoietic cells come from the RAG background. We first evaluated the level of serum TGF-β in TGF-β1 null mice and chimeras using reporter cells that allows for the detection of bioactive TGF-β1, β2 and β3 (12). In control mice, TGF activity was totally inhibited using an anti TGF-β1 Ab and therefore totally TGF-β1-derived (Fig. 3). Consistently, TGF-β1 null mice were negative for serum TGF activity showing no compensation by TGF-β2 or β3. In contrast, chimeras lacking B cell-derived TGF-β1, with or without T cells, had normal level of seric TGF-β. Therefore neither B nor T cells represent a major source of serum TGF-β1.

FIGURE 2.

Design of chimeric mice. Mice and chimeras that are used in this study are indicated. CD3ε−/−TGF-β1−/− double KO mice were obtained by crossing CD3ε−/− mice to TGF-β1−/− mice (CD3 KO mice model). T-deficient chimeras (T-def chim) were obtained using sublethally irradiated RAG KO mice reconstituted with CD3ε−/− or CD3ε−/−TGF-β1−/− BM. T-sufficient chimeras (T-suff chim) were generated by reconstituting sublethally irradiated RAG KO mice with a mixture of CD3ε KO and μMT BMs or a mixture of CD3ε−/−TGF-β1−/− and μMT BMs. Mixed BM chimeras (mixed BM chim) were generated by sublethally irradiating RAG KO mice and reconstituting them either with an equal mixture of CD3ε KO plus Β6 Ly5.1 BMs or an equal mixture of CD3ε−/−TGF-β1−/− plus Β6 Ly5.1 BMs.

FIGURE 2.

Design of chimeric mice. Mice and chimeras that are used in this study are indicated. CD3ε−/−TGF-β1−/− double KO mice were obtained by crossing CD3ε−/− mice to TGF-β1−/− mice (CD3 KO mice model). T-deficient chimeras (T-def chim) were obtained using sublethally irradiated RAG KO mice reconstituted with CD3ε−/− or CD3ε−/−TGF-β1−/− BM. T-sufficient chimeras (T-suff chim) were generated by reconstituting sublethally irradiated RAG KO mice with a mixture of CD3ε KO and μMT BMs or a mixture of CD3ε−/−TGF-β1−/− and μMT BMs. Mixed BM chimeras (mixed BM chim) were generated by sublethally irradiating RAG KO mice and reconstituting them either with an equal mixture of CD3ε KO plus Β6 Ly5.1 BMs or an equal mixture of CD3ε−/−TGF-β1−/− plus Β6 Ly5.1 BMs.

Close modal
FIGURE 3.

Serum TGF activity in mice with TGF-β1-deficient B cells. Sera used were from CD3KO, T-def chim, and T-suff chim, with either TGF-β1 WT (•) or KO cells (○). Serum were either untreated (−) or acidified (+), then neutralized and added to MFBF11 reporter cells. Blocking anti-TGF-β1 mAb was added when indicated (+). After 24 h, SEAP activity was quantified in the supernatant.

FIGURE 3.

Serum TGF activity in mice with TGF-β1-deficient B cells. Sera used were from CD3KO, T-def chim, and T-suff chim, with either TGF-β1 WT (•) or KO cells (○). Serum were either untreated (−) or acidified (+), then neutralized and added to MFBF11 reporter cells. Blocking anti-TGF-β1 mAb was added when indicated (+). After 24 h, SEAP activity was quantified in the supernatant.

Close modal

We quantified IgA and IgG production in serum and fecal pellets in these T-deficient chimeric mice. Remarkably, chimeras with TGF-β1−/− B cells showed a significant decrease and increase of seric IgA and IgG1, respectively (Fig. 1,B) (IgG1 = 0.023 mg/ml with TGF-β1−/− B cells, 0.006 mg/ml with WT; IgA = 0.012 mg/ml with TGF-β1−/− B cells, 0.024 mg/ml with WT). Therefore, in the absence of T cells, B cell-derived TGF-β1 controls systemic IgA and IgG1 production. Fecal pellet IgA was also reduced in the absence of B cell-derived TGF-β1, but the decrease was less significant than that observed in CD3/TGF-β1 double KO mice. Therefore, these results suggest that TGF-β1 produced by both B cells and other gut non-T cells regulate IgA secretion. To further confirm that B cell-derived TGF-β1 controls T-independent responses, T-def chimeras were immunized with the T-independent Ag TNP-Ficoll. Chimeras with TGF-β1−/− B cells showed increased production of IgM, IgG3, and IgG1 (Fig. 4 A). To analyze the production of IgA, chimeras were immunized orally with a T-independent Ag in the presence of cholera toxin. Such immunization resulted in significant IgA responses in serum and feces that were dramatically blunted in the absence of B cell-derived TGF-β1. Therefore, the intrinsic TGF-β1 production by B cells controls the extent of systemic T-independent responses and promotes IgA production in the gut.

FIGURE 4.

Ab responses in mice with TGF-β1-deficient B cells. A, T-def chim with either TGF-β1 WT (•) or KO (Δ) B cells were immunized i.p. with TNP-Ficoll and anti-TNP Abs were titrated by ELISA (Left panel) or were immunized orally with FITC-Dextran and IgA anti-FITC titers in the serum and feces were calculated (Right panel). B, T-suff chim with either TGF-β1 WT (•) or KO (Δ) B cells were immunized with collagen or OVA in CFA. Anti-OVA or anti-collagen titers were measured by ELISA. The values represent the mean ± SD of four to six mice per group. (*) Two-tailed Student’s t test comparing WT and KO responses gave values of p < 0.02.

FIGURE 4.

Ab responses in mice with TGF-β1-deficient B cells. A, T-def chim with either TGF-β1 WT (•) or KO (Δ) B cells were immunized i.p. with TNP-Ficoll and anti-TNP Abs were titrated by ELISA (Left panel) or were immunized orally with FITC-Dextran and IgA anti-FITC titers in the serum and feces were calculated (Right panel). B, T-suff chim with either TGF-β1 WT (•) or KO (Δ) B cells were immunized with collagen or OVA in CFA. Anti-OVA or anti-collagen titers were measured by ELISA. The values represent the mean ± SD of four to six mice per group. (*) Two-tailed Student’s t test comparing WT and KO responses gave values of p < 0.02.

Close modal

Because T cells play a major role in controlling the B cell isotype switch, we next addressed whether the lack of B cell-derived TGF-β1 would be compensated by T cell-derived TGF-β1. We reconstituted RAG KO mice with a mixture of TGF-β1/CD3 KO and μMT BM (T-suff chim (Fig. 2)). In these mice, B but not T lymphocytes fail to produce TGF-β1. In control chimeras, serum IgA and IgG1 production was restored to normal levels; in contrast, the absence of B cell-derived TGF-β1 again provoked an increase in IgG1 and decrease in IgA production (Fig. 1,C) (serum IgA: 0.05 mg/ml with TGF-β1−/− B cells, 0.1 mg/ml with WT; IgG1: 1.2 mg/ml with TGF-β1−/− B cells, 0.56 mg/ml with WT). Importantly, the absence of B cell-derived TGF-β1 resulted in a significant decrease in fecal IgA production, confirming that B cell-produced TGF-β1 indeed participates in gut IgA production and that T cell-produced TGF-β1 does not compensate for this loss. Those phenotypes were confirmed by analyzing membrane Ig expression on splenic and MLN B cells that showed increased and decreased proportions of IgG1- and IgA-positive cells, respectively (Fig. 5). Therefore, these results show that even in the presence of normal T cells, B cell-derived TGF-β1 plays a major role in controlling IgA and IgG1 production.

FIGURE 5.

B cell-produced TGF-β1 regulates the number of IgG1- and IgA-positive cell. T-suff chim with either TGF-β1 WT (•) or KO (Δ) B cells were analyzed for the presence of Sp and MLN IgA- and IgG1-positive B cells. The percentage of IgA- or IgG1-positive cells among B220 cells is plotted. Two-tailed Student’s t test comparing WT and KO values are included (p values).

FIGURE 5.

B cell-produced TGF-β1 regulates the number of IgG1- and IgA-positive cell. T-suff chim with either TGF-β1 WT (•) or KO (Δ) B cells were analyzed for the presence of Sp and MLN IgA- and IgG1-positive B cells. The percentage of IgA- or IgG1-positive cells among B220 cells is plotted. Two-tailed Student’s t test comparing WT and KO values are included (p values).

Close modal

TGF-βR-deficient mice demonstrated the role of TGF-β1 in controlling adaptive immune responses. Upon immunization with T-dependent Ags, these mice mounted significantly increased IgM and IgG3 responses and blunted IgA production. To address the contribution of B cell-derived TGF-β1 in this process, we immunized T-suff chim with OVA or bovine collagen in CFA. We found that in the absence of B cell-derived TGF-β1, IgM and IgG3 responses were increased but again IgA production was drastically depressed (Fig. 4 B). The production of other isotypes did not show significant differences. Therefore, and surprisingly considering the importance of T cell-derived cytokines in controlling B cell activation and Ig switch, TGF-β1 regulation of B cell responses showed a clear B cell autonomous effect.

TGF-βR-deficient mice demonstrated a role for TGF-β in controlling B1 cell homeostasis. These cells are dependent upon chronic recognition of self or microbial Ags and are responsible for the T-independent production of natural Abs (13). In CD3/TGF-β1 double-deficient mice, we found an increased proportion of B1 cells. This showed that TGF-β1 controls B1 cell homeostasis in a T-independent manner. Furthermore, the analysis of T-suff chim in which TGF-β1-deficient B cells were introduced in a normal environment showed that B cell-derived TGF-β1 mediated this effect (Fig. 6 and Table I).

FIGURE 6.

Cell autonomous regulation of B1 cells by TGF-β1. CD3KO, T-suff chim, and mixed BM chimera with either TGF-β1 WT (•) or KO B cells (Δ) were analyzed. PC cells were analyzed by flow cytometry using B220 and CD11b staining plus CD45.1 and CD45.2 for Ly5.1/5.2 chimeras. Data represents the B1 to B2 ratio or the Ly5.2 to Ly5.1 ratio of B220-positive cells. Two-tailed Student’s t test comparing WT and KO values gave p values <0.01.

FIGURE 6.

Cell autonomous regulation of B1 cells by TGF-β1. CD3KO, T-suff chim, and mixed BM chimera with either TGF-β1 WT (•) or KO B cells (Δ) were analyzed. PC cells were analyzed by flow cytometry using B220 and CD11b staining plus CD45.1 and CD45.2 for Ly5.1/5.2 chimeras. Data represents the B1 to B2 ratio or the Ly5.2 to Ly5.1 ratio of B220-positive cells. Two-tailed Student’s t test comparing WT and KO values gave p values <0.01.

Close modal
Table I.

Absolute numbers of B cells in CD3 KO, T-sufficient chimeras, and mixed BM chimerasa

CD3 KOT-suff chimMixed BM chimMixed BM chim
TGF-β1 WTTGF-β1 KOTGF-β1 WTTGF-β1 KOCD45.1 WTCD45.2 WTCD45.1 WTCD45.2 KO
PC B1 0.92 (±0.4) 3.32 (±1.5) 0.55 (±0.2) 2.4 (±0.5) 1.19 (±0.6) 1.4 (±0.74) 0.8 (±0.35) 3.2 (±1.8) 
PC B2 0.42 (±0.2) 0.40 (±0.2) 2.2 (±0.4) 1.95 (±0.5) 1.45 (±0.7) 1.7 (±0.63) 1.2 (±0.88) 1.4 (±0.61) 
Sp B 30.4 (±8.6) 40.4 (±19.5) 41.4 (±9.2) 63.4 (±22.5) 19.5 (±6.8) 21.3 (±13.2) 21 (±6.5) 49.6 (±14.2) 
LN B 6.9 (±3.1) 7.3 (±4.8) 5.7 (±2.1) 8.2 (±4.2) 2.6 (±1.5) 2.8 (±1.5) 2.7 (±1.1) 3.6 (±1.2) 
MLN B 5.6 (±3) 6.1 (±2.7) 6.6 (±2.4) 6.3 (±3.1) 2.2 (±0.7) 2.6 (±0.8) 1.9 (±0.9) 2.9 (±1.2) 
CD3 KOT-suff chimMixed BM chimMixed BM chim
TGF-β1 WTTGF-β1 KOTGF-β1 WTTGF-β1 KOCD45.1 WTCD45.2 WTCD45.1 WTCD45.2 KO
PC B1 0.92 (±0.4) 3.32 (±1.5) 0.55 (±0.2) 2.4 (±0.5) 1.19 (±0.6) 1.4 (±0.74) 0.8 (±0.35) 3.2 (±1.8) 
PC B2 0.42 (±0.2) 0.40 (±0.2) 2.2 (±0.4) 1.95 (±0.5) 1.45 (±0.7) 1.7 (±0.63) 1.2 (±0.88) 1.4 (±0.61) 
Sp B 30.4 (±8.6) 40.4 (±19.5) 41.4 (±9.2) 63.4 (±22.5) 19.5 (±6.8) 21.3 (±13.2) 21 (±6.5) 49.6 (±14.2) 
LN B 6.9 (±3.1) 7.3 (±4.8) 5.7 (±2.1) 8.2 (±4.2) 2.6 (±1.5) 2.8 (±1.5) 2.7 (±1.1) 3.6 (±1.2) 
MLN B 5.6 (±3) 6.1 (±2.7) 6.6 (±2.4) 6.3 (±3.1) 2.2 (±0.7) 2.6 (±0.8) 1.9 (±0.9) 2.9 (±1.2) 
a

The absolute numbers of peritoneal cavity B1 and B2 cells and of Sp, LN, and MLN B220-positive cells are given in million cells. Each value represents the average of at least six mice. Two-tailed Student’s t test comparing WT and KO B cells from Sp of T-suff chim and mixed BM chim, as well as B1 cells from all types of mice gave values of p < 0.01.

To further define whether B1 cell intrinsic production of TGF-β1 was responsible for their homeostatic control, we prepared mixed BM chimeras in which Ly5.2 TGF-β1-deficient (or control cells) compete with similar proportions of control Ly5.1 B cells. These mice were analyzed for the presence of B cells in the PC. In mixed chimeras, KO B1 cells had a clear homeostatic advantage over control B1 cells as noted by the high Ly5.2/Ly5.1 ratio (Fig. 6 and Table I). Therefore, TGF-β1 controls B1 cells in a cell autonomous manner. By contrast, TGF-β1-deficient B2 cells had no competitive advantage over control B2 cells. Ly5.1/Ly5.2 mixed chimeras showed equal numbers of WT and TGF-β1 KO B cells among, recirculating, LN or MLN B cells (Fig. 7). Therefore, no cell autonomous role for TGF-β1 could be defined for the homeostatic control of these populations. By contrast, disequilibrium was evident among splenic B cells. Although the WT Ly5.2/WT Ly5.1 ratio was close to 1, the KO Ly5.2/WT Ly5.1 ratio increased in B cells suggesting that, as in the case of B1 cells, TGF-β1 controls B cell splenocytes in a cell autonomous manner (Fig. 7). A similar increase in splenic TGF-β1 KO B cells was also evident in CD3KO and in T-suff chim (Table I). Despite this, no major consistent differences were found between marginal zone, follicular, and immature B cell populations that would be compatible with a role for TGF-β1 in controlling all these cells or their precursors (data not shown). The accumulation of B1 cells in the absence of TGF-β1 might be the consequence of different mechanisms. As B1 and B2 cell precursors originate from different pools, TGF-β1 could control B1 cell precursor frequency. Also, B1 cell maintenance is mostly due to self-renewal and TGF-β1 could limit B1 cell homeostatic proliferation. To address these hypotheses, we injected BrdU for 15 days in mixed BM chimeras and followed B cell BrdU incorporation that sums input from precursors and homeostatic proliferation. In agreement with what was reported in TGF-βR-deficient mice, the percentage of BrdU-positive cells was identical between WT and KO B cells in those chimeras, suggesting that the accumulation of B1 cell might originate from a different cause (Fig. 8). Because B1 cells are the main precursors of intestinal IgA (11), the concomitant raise in B1 cell and fall of IgA might be linked.

FIGURE 7.

Cell autonomous regulation of Splenic B cells by TGF-β1. Mixed BM chimera with either TGF-β1 WT (•) or KO B cells (Δ) were tested. BM, recirculating, splenic, inguinal and MLN B cells were analyzed by flow cytometry using anti-CD19, anti-CD45.1, and anti-CD45.2 mAbs. CD19-positive cells were gated and the Ly5.2/Ly5.1 ratio was plotted.

FIGURE 7.

Cell autonomous regulation of Splenic B cells by TGF-β1. Mixed BM chimera with either TGF-β1 WT (•) or KO B cells (Δ) were tested. BM, recirculating, splenic, inguinal and MLN B cells were analyzed by flow cytometry using anti-CD19, anti-CD45.1, and anti-CD45.2 mAbs. CD19-positive cells were gated and the Ly5.2/Ly5.1 ratio was plotted.

Close modal
FIGURE 8.

Identical proliferation and renewal of B1 cells in the absence of TGF-β1. Mixed BM chimera with either TGF-β1 WT (•) or KO B cells (Δ) were analyzed. PC B1 and B2 cells from either Ly5.1- or Ly5.2-positive cells were gated. For each mouse, the percentage of BrdU-positive cells among WT Ly5.1 (▪) and WT Ly5.2 (•) or TGF-β1 KO Ly5.2 (Δ) is plotted.

FIGURE 8.

Identical proliferation and renewal of B1 cells in the absence of TGF-β1. Mixed BM chimera with either TGF-β1 WT (•) or KO B cells (Δ) were analyzed. PC B1 and B2 cells from either Ly5.1- or Ly5.2-positive cells were gated. For each mouse, the percentage of BrdU-positive cells among WT Ly5.1 (▪) and WT Ly5.2 (•) or TGF-β1 KO Ly5.2 (Δ) is plotted.

Close modal

TGF-β1 deficiency leads to loss of tolerance and appearance of autoantibody. TGF-β1 signaling in B cells is required for preventing autoantibody production as TGF-βR deficiency in B cells induces such production. Furthermore, T cells are necessary, as their depletion prevents autoantibody appearance (5). Given the importance of autonomous effects in the TGF-β1 control of B cell homeostasis and immune responses, we wondered whether TGF-β1 autonomous production could be important in preventing autoantibody appearance. Despite increased serum IgG content, there was no evidence for increased levels of IgG anti-DNA in mice with TGF-β1-deficient B cells (Fig. 9). Therefore, prevention of autoantibody production uses other sources of TGF-β1. Because the specific depletion of TGF-β1 in T cells is sufficient to promote autoantibody production, T cells are probably the predominant source. Tregs could be implicated, as their absence in Foxp3 mutant mice led to autoantibody production (14, 15). This population depends on TGF-β1 for their maintenance in the periphery, but the source of TGF-β1 responsible is not known. Because our chimeras showed normal levels of Tregs, B cell-produced TGF-β1 does not seem to play a role in maintaining this population (Fig. 9).

FIGURE 9.

No role for B cell -roduced TGF-β1 in regulating anti-DNA production and Treg cell homeostasis. CD3KO and T-suff chim, with either TGF-β1 WT (•) or KO cells (○), were analyzed. Left panel, The amount of IgG anti-DNA Abs in the serum. Positive controls corresponding to New Zealand Black/White mice sera are included (♦). Right panel, The percentage of Treg cells in the Sp, LNs, and MLNs.

FIGURE 9.

No role for B cell -roduced TGF-β1 in regulating anti-DNA production and Treg cell homeostasis. CD3KO and T-suff chim, with either TGF-β1 WT (•) or KO cells (○), were analyzed. Left panel, The amount of IgG anti-DNA Abs in the serum. Positive controls corresponding to New Zealand Black/White mice sera are included (♦). Right panel, The percentage of Treg cells in the Sp, LNs, and MLNs.

Close modal

Our data indicates that B cell-produced TGF-β1 plays a major role in controlling B cell biology including B cell responses, the Ig isotype switch, and homeostasis of B1 and splenic B cells. We show that TGF-β1 regulates serum IgA and IgG1 levels in unimmunized mice. These Abs are produced mainly by BM-derived B2 cells during T cell-dependent responses. The analysis of chimeric mice with normal T cells and TGF-β1-deficient B cells showed that B cell-derived TGF-β1 played a preponderant role in this regulation. When we immunized those mice with Ag in CFA, the classical T-dependent isotypes were not affected. By contrast, IgG3, which is more associated with T-independent responses, was boosted in the absence of B cell-derived TGF-β1. Consequently, normal isotype switch is affected by B cell-derived TGF-β1, inhibiting IgG3 while promoting IgA. Alternatively, it could suggest that B cell-derived TGF-β1 limits T-independent responses during T-dependent immunizations.

That TGF-β1 plays a major role in regulating T-independent responses is further documented in CD3/TGF-β1 double KO mice. These mice had increased IgG1 and decreased IgA as compared with their WT counterparts, and chimeras showed that this control depended on the autonomous production of TGF-β1 by B cells. Finally, active immunization with T-independent Ags further evidenced that B cell-produced TGF-β1 autonomously controls those responses.

A major phenotype induced by TGF-β1 deficiency in B cells was the accumulation of B1 cells in the PC. Similar levels of BrdU accumulation failed to show a role for TGF-β1 in controlling their homeostatic proliferation. We hypothesize that the autonomous production of TGF-β1 could promote B1 cell death. This hypothesis is consistent with a role of TGF-β1 in inducing apoptosis. Also, transgenic mice that constitutively expressed the antiapoptotic protein Bcl-xL or Bcl-2 exhibited increased peritoneal B1 cells (16). Alternatively, the accumulation of B1 cells could simply reflect their failure to differentiate into gut IgA-producing B cells.

The cell autonomous control by TGF-β1 might be the consequence of multiple mechanisms. Cell stickiness might restrict available TGF-β1 to the close vicinity of its production site. This proposal is best exemplified by experiments in which TGF-β1 was added to cell cultures and tested for its availability in the supernatant (17). These authors reported that TGF-β1 became undetectable in the supernatant minutes after its addition. Furthermore, TGF-β1 can be detected at the cell surface of B lymphocytes (18); it could be envisaged that this location limits its action to these cells or those in close interaction. Any of these processes could result in a cell autonomous effect of TGF-β1. It should be noted that the complexity of TGF-β1 action, its potential production as a propeptide and/or a latent form allows for complex control mechanisms (19). TGF-β1 activation by integrins could restrict its action where cognate interaction occurs, while the action of proteases oxidative stress and pH could result in a cell autonomous activation during internalization processes. Interestingly, we noted that the transfection of TGF-β1 in B cell lines induced membrane expression and secretion of mainly pro-TGF-β1, which could support the notion that its bioavailability would require the action of furin enzymes. Interestingly, furin has been shown to be expressed in Th1 cells and in dendritic cells upon proinflammatory conditions (20). Therefore, the cognate interaction of B cells with Th1 cells during T-dependent responses or with dendritic cells during T-dependent or independent responses could allow processing of B cell-derived pro-TGF-β1 and bioavailability. This scenario would be compatible with a B cell autonomous effect of TGF-β1 even though the decision process, to use TGF-β1 or not, might be taken by either B cell or its cognate partners.

Finally, B cells with regulatory functions have been described that control experimental autoimmune encephalomyelitis or gastritis via IL-10 secretion (21, 22). The autonomous regulation by TGF-β1 presented here opens up new avenues for new types, or mode of action, of regulatory B cells. This situation is clearly reminiscent of the one established for T cells where Tr1, Treg, and Th3 cells prevent autoimmunity by provision of either IL-10 or TGF-β1 (23).

We thank Lee Leserman for critical reading of the manuscript, the CIML cytometry, and animal and transgene facilities.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by institutional funding from Institut National de la Santé et de la Recherche Médicale and Centre National de la Recherche Scientifique. M.J.G. was financed by a grant from the Ministère de la Recherche and from Fondation pour la recherche médicale.

3

Abbreviations used in this paper: KO, knockout; LN, lymph node; MLN, mesenteric LN; Sp, spleen; PC, peritoneal cavity; BM, bone marrow; Treg, regulatory T cell; WT, wild type.

1
Li, M. O., Y. Y. Wan, S. Sanjabi, A. K. Robertson, R. A. Flavell.
2006
. Transforming growth factor-β regulation of immune responses.
Annu. Rev. Immunol.
24
:
99
-146.
2
Shull, M. M., I. Ormsby, A. B. Kier, S. Pawlowski, R. J. Diebold, M. Yin, R. Allen, C. Sidman, G. Proetzel, D. Calvin, et al
1992
. Targeted disruption of the mouse transforming growth factor-β1 gene results in multifocal inflammatory disease.
Nature
359
:
693
-699.
3
Letterio, J. J., A. G. Geiser, A. B. Kulkarni, H. Dang, L. Kong, T. Nakabayashi, C. L. Mackall, R. E. Gress, A. B. Roberts.
1996
. Autoimmunity associated with TGF-β1-deficiency in mice is dependent on MHC class II antigen expression.
J. Clin. Invest.
98
:
2109
-2119.
4
Kobayashi, S., K. Yoshida, J. M. Ward, J. J. Letterio, G. Longenecker, L. Yaswen, B. Mittleman, E. Mozes, A. B. Roberts, S. Karlsson, A. B. Kulkarni.
1999
. β2-microglobulin-deficient background ameliorates lethal phenotype of the TGF-β 1 null mouse.
J. Immunol.
163
:
4013
-4019.
5
Bommireddy, R., S. J. Engle, I. Ormsby, G. P. Boivin, G. F. Babcock, T. Doetschman.
2004
. Elimination of both CD4+ and CD8+ T cells but not B cells eliminates inflammation and prolongs the survival of TGFβ1-deficient mice.
Cell. Immunol.
232
:
96
-104.
6
Li, M. O., S. Sanjabi, R. A. Flavell.
2006
. Transforming growth factor-β controls development, homeostasis, and tolerance of T cells by regulatory T cell-dependent and -independent mechanisms.
Immunity
25
:
455
-471.
7
Li, M. O., Y. Y. Wan, R. A. Flavell.
2007
. T cell-produced transforming growth factor-β1 controls T cell tolerance and regulates Th1- and Th17-cell differentiation.
Immunity
26
:
579
-591.
8
Cazac, B. B., J. Roes.
2000
. TGF-β receptor controls B cell responsiveness and induction of IgA in vivo.
Immunity
13
:
443
-451.
9
Van Ginkel, F. W., S. M. Wahl, J. F. Kearney, M. N. Kweon, K. Fujihashi, P. D. Burrows, H. Kiyono, J. R. McGhee.
1999
. Partial IgA-deficiency with increased Th2-type cytokines in TGF-β1 knockout mice.
J. Immunol.
163
:
1951
-1957.
10
Tanaka, Y., L. Ardouin, A. Gillet, S. Y. Lin, A. Magnan, B. Malissen, M. Malissen.
1995
. Early T-cell development in CD3-deficient mice.
Immunol. Rev.
148
:
171
-199.
11
Macpherson, A. J., D. Gatto, E. Sainsbury, G. R. Harriman, H. Hengartner, R. M. Zinkernagel.
2000
. A primitive T cell-independent mechanism of intestinal mucosal IgA responses to commensal bacteria.
Science
288
:
2222
-2226.
12
Tesseur, I., K. Zou, E. Berber, H. Zhang, T. Wyss-Coray.
2006
. Highly sensitive and specific bioassay for measuring bioactive TGF-β.
BMC Cell Biol.
7
:
15
13
Hardy, R. R..
2006
. B-1 B cell development.
J. Immunol.
177
:
2749
-2754.
14
Nguyen, L. T., J. Jacobs, D. Mathis, C. Benoist.
2007
. Where FoxP3-dependent regulatory T cells impinge on the development of inflammatory arthritis.
Arthritis Rheum.
56
:
509
-520.
15
Wan, Y. Y., R. A. Flavell.
2007
. Regulatory T-cell functions are subverted and converted owing to attenuated Foxp3 expression.
Nature
445
:
766
-770.
16
Chattopadhyay, G., A. Q. Khan, G. Sen, J. Colino, W. Dubois, A. Rubtsov, R. M. Torres, M. Potter, C. M. Snapper.
2007
. Transgenic expression of Bcl-xL or Bcl-2 by murine B cells enhances the in vivo antipolysaccharide, but not antiprotein, response to intact Streptococcus pneumoniae.
J. Immunol.
179
:
7523
-7534.
17
Menoret, A., L. M. Myers, S. J. Lee, R. S. Mittler, R. J. Rossi, A. T. Vella.
2006
. TGFβ protein processing and activity through TCR triggering of primary CD8+ T regulatory cells.
J. Immunol.
177
:
6091
-6097.
18
Parekh, V. V., D. V. Prasad, P. P. Banerjee, B. N. Joshi, A. Kumar, G. C. Mishra.
2003
. B cells activated by lipopolysaccharide, but not by anti-Ig and anti-CD40 antibody, induce anergy in CD8+ T cells: role of TGF-β1.
J. Immunol.
170
:
5897
-5911.
19
Annes, J. P., J. S. Munger, D. B. Rifkin.
2003
. Making sense of latent TGFβ activation.
J. Cell Sci.
116
:
217
-224.
20
Pesu, M., L. Muul, Y. Kanno, J. J. O'Shea.
2006
. Proprotein convertase furin is preferentially expressed in T helper 1 cells and regulates interferon γ.
Blood
108
:
983
-985.
21
Mizoguchi, A., E. Mizoguchi, H. Takedatsu, R. S. Blumberg, A. K. Bhan.
2002
. Chronic intestinal inflammatory condition generates IL-10-producing regulatory B cell subset characterized by CD1d upregulation.
Immunity
16
:
219
-230.
22
Fillatreau, S., C. H. Sweenie, M. J. McGeachy, D. Gray, S. M. Anderton.
2002
. B cells regulate autoimmunity by provision of IL-10.
Nat. Immunol.
3
:
944
-950.
23
Wan, Y. Y., R. A. Flavell.
2006
. The roles for cytokines in the generation and maintenance of regulatory T cells.
Immunol. Rev.
212
:
114
-130.