Peroxisome Proliferator-Activated Receptor γ Ligands Enhance Human B Cell Antibody Production and Differentiation¹

The differentiation of B cells into Ig-secreting plasma cells is crucial for protective humoral immune responses to combat infection (1). The innate immune system recognizes microorganisms through pattern recognition receptors, such as TLRs. Activation of human B cells by unmethylated CpG-DNA motifs, a TLR9 ligand, induces B cell differentiation as well as increased cytokine and Ab production (1). During humoral immune responses, naive B cells that become activated first proliferate and secrete IgM, followed by IgG. Some B cells become long-lived plasma cells that secrete copious amounts of Ab or further differentiate into memory B cells (2). Activation of B cells also results in the expression of key transcription factors, such as BLIMP-1, that lead to the expression of genes necessary for terminal B cell differentiation (3).

We recently published that peroxisome proliferator-activated receptor γ (PPARγ) overexpression and knockdown influence BLIMP-1 expression in Burkitt’s lymphoma (4). PPARs belong to the nuclear hormone receptor superfamily of transcription factors (5), of which there are three isoforms: PPARα, PPARβ/δ, and PPARγ. PPARγ and its ligands are involved in regulating proliferative, inflammatory, and in some cases differentiating properties of immune and cancer cells (6, 7). We previously demonstrated that normal and malignant B lymphocytes express PPARγ and that exposure to micromolar levels of certain types of electrophilic PPARγ ligands inhibits B cell proliferation (8, 9, 10). PPARγ ligands are diverse and at high concentrations (μM) can have PPARγ-independent effects. Endogenous ligands include 15-deoxy-Δ^12,14 prostaglandin J₂ (15d-PGJ₂), as well as fatty acid derivatives (i.e., oxidized low-density lipoproteins). PGD₂ and 15d-PGJ₂ are derived from arachidonic acid by the catalytic activities of cyclooxygenase-2 (COX-2)³ and prostaglandin D synthase (11, 12, 13, 14). PGD₂ spontaneously undergoes a series of dehydration reactions to form the PGJ family of prostaglandins, including 15d-PGJ₂, and 15d-PGD₂, which can also transactivate PPARγ (12, 15, 16, 17). PPARγ is also activated by synthetic ligands belonging to the thiazolidinediones class of antidiabetic drugs, which includes rosiglitazone. Following ligand binding, PPARγ forms a heterodimer with retinoid X receptors (RXRs) and subsequently binds to the peroxisome proliferator response element (PPRE) found in target gene promoters. RXR is an obligate partner of PPARγ. It is required to induce transcription (18) and is activated by 9-cis-retinoic acid (9-cis-RA), a vitamin A metabolite (19).

It is unknown whether PPARγ/RXR regulates B cell differentiation. We hypothesized that during human B cell activation, PPARγ protein levels would increase and would stimulate differentiation and Ab production. We also proposed that PPARγ would interact with RXR to increase plasma cell formation and Ab production. Herein, we report our studies on PPARγ expression and how low doses of PPARγ ligands enhance B cell function.

CpG oligodeoxynucleotides 2395 (5′-TCGTCGTTTTCGGCGCGCGCCG-3′) were purchased from the Coley Pharmaceutical Group and used at a concentration of 1 μg/ml. A rabbit anti-human F(ab′)₂ anti-IgM Ab (Jackson ImmunoResearch Laboratories) was used at 2 μg/ml to crosslink the BCR. Rosiglitazone and the irreversible PPARγ antagonist GW9662 were purchased from Cayman Chemical and 15d-PGJ₂ was purchased from BIOMOL. 9-cis-RA was obtained from Sigma-Aldrich. The anti-BLIMP-1 Ab was purchased from Novus Biologicals. The anti-PPARγ Abs were purchased from Abcam and Santa Cruz (Santa Cruz Biotechnology). Total actin (CP-01) Ab was from Oncogene. The COX-2 selective inhibitor SC-58125 was purchased from Cayman Chemical.

Normal B lymphocytes were isolated from a unit of whole blood from healthy donors with ethical permission from the Research Subjects Review Board at the University of Rochester. The isolation of normal B lymphocytes has been previously described (20). Briefly, buffy coats were obtained from whole blood, and PBMCs were obtained using Ficoll-Paque (Amersham Biosciences) gradient centrifugation. PBMCs were then incubated with anti-CD19 Ab-coated Dynabeads (Dynal Biotech) and subjected to a magnetic field to separate B lymphocytes; negatively selected cells were washed out. B lymphocytes were then detached from the beads using an equal volume of CD19 Detachabeads (Dynal Biotech). B lymphocyte purity was >98% CD19⁺ (as determined by flow cytometry; data not shown). Purified B cells were cultured in RPMI 1640 tissue culture medium (Invitrogen) supplemented with 10% FBS, 5 × 10⁻⁵ M 2-ME (Eastman Kodak), 10 mM HEPES (US Biochemical), 2 mM l-glutamine (Invitrogen), and 50 μg/ml gentamicin (Invitrogen). All experiments were conducted with B cells from at least three different donors.

Transient transfections of normal B lymphocytes with a PPRE-luciferase reporter plasmid containing three copies of the ACO-PPRE (PPAR response element) from rat acyl CoA oxidase (a gift from Dr. B. Seed, Massachusetts General Hospital, Boston, MA) (21, 22) were conducted using the Nucleofector protocol from Amaxa Biosystems. Eighteen hours posttransfection cells were left untreated or were treated with 1 μg/ml CpG in the presence or absence of rosiglitazone (0.5 μM) or 15d-PGJ₂ (0.2 μM). These optimal doses were chosen based on pilot experiments. Twenty-four hours after treatments, luciferase activity was assayed using the Promega luciferase assay system. Relative light units (RLU) were determined with a Lumicount Microplate luminometer (Packard Instrument). RLU were normalized to transfection efficiency that was monitored by cotransfection of GFP expression vector. Transfection efficiency was ∼40% (data not shown).

For cell division, a CellTrace CFSE cell proliferation kit (Invitrogen) was used according to the manufacturer’s protocol. Briefly, cells were labeled with 0.5 μM CFSE for 15 min at 37°C, followed by two washes with 1× PBS, and then resuspended in RPMI 1640 culture media containing 10% FBS. Cells were then plated at a density of 1 × 10⁵ cells/well in a 96-round-bottom plate. Five days later, cells were acquired using a BD Biosciences FACSCalibur flow cytometer and analyzed using FlowJo software (Tree Star).

B cells were incubated with mouse anti-human CD19-allophycocyanin (BD Biosciences), anti-human CD38-PE (BD Biosciences), and/or anti-human CD27-allophycocyanin (BD Biosciences) in cold PBS with sodium azide (0.02%) and BSA (0.3%) for 20 min at 20°C. COX-2 intracellular staining was performed as described previously (8). All samples were acquired on a BD Biosciences FACSCalibur flow cytometer and analyzed using FlowJo software (Tree Star).

For intracellular staining for PPARγ, untreated or activated B lymphocytes were surface stained with 20 μl of allophycocyanin anti-human CD19 mAb (BD Biosciences) for 30 min in the dark at room temperature. Cells were then fix and permeabilized with BD Cytofix/Cytoperm Fixation/Permeabilization kit (BD Biosciences) following the manufacturer’s instructions. A FITC-conjugated anti-human PPARγ Ab was used at a 1/100 dilution. An equal amount of IgG1 FITC mAb was used as an isotype control.

Purified human B lymphocytes (5 × 10⁵ cells/ml) were cultured in 96-well round-bottom microtiter plates. Cells were treated for 5–6 days with activating agents in the presence and absence of PPARγ ligands and/or 9-cis-RA (100 nM). Pilot experiments were performed to optimize the doses of PPARγ and RXR ligands. For some experiments, cells were also treated with an optimal dose of GW9662 (500 nM). Supernatants were harvested and the concentrations of IgM and IgG were analyzed using human-specific ELISAs (Bethyl Laboratories).

Whole-cell extracts were collected using ELB buffer (50 mM HEPES (pH 7), 250 mM NaCl, 0.1% Nonidet P-40, 5 mM EDTA, 10 mM NaF, 0.1 mM Na₃VO₄, 50 μM ZnCl₂, supplemented with 0.1 mM PMSF, 1 mM DTT, and a mixture of protease and phosphatase inhibitors) and total protein was quantified using bicinchoninic acid protein assay (BCA assay kit; Pierce). Twenty-five micrograms of protein was electrophoresed on 8–16% Precise protein gels (Pierce) and transferred to polyvinylidene fluoride membrane (Millipore). The membranes were analyzed for immunoreactivity with the indicated primary Ab, washed, and then incubated with an appropriate HRP-conjugated secondary Ab. The membranes were visualized by chemiluminescence using an ECL kit (Pierce).

Statistical analysis was performed using GraphPad Prism (GraphPad Software). For comparison between groups of three or more, an ANOVA with Newman-Keuls multiple comparison test was used to determine differences between treatments. A t test was used to compare vehicle and PPARγ ligand. Results are expressed as the means ± SEM. Values of p < 0.05 were considered significant. All experiments were repeated at least three times.

We have previously shown that freshly isolated normal human B cells express low levels of PPARγ protein (8). However, it is unknown if PPARγ expression changes after provocation with stimulatory agents. Therefore, normal peripheral blood B cells were left untreated or activated with unmethylated CpG-DNA (a TLR9 ligand) (23), with or without anti-IgM, and PPARγ expression was examined by Western blot. PPARγ is expressed in human B cells (molecular mass of ∼54 kDa), with PPARγ levels in untreated B cells being variable (undetectable to low) between three individual donors (Fig. 1,A). However, PPARγ expression was increased in B cells that were activated with anti-IgM or CpG or a combination of anti-IgM plus CpG (Fig. 1, donors 1–3). Densitometric analysis demonstrates that the range of induction in PPARγ protein levels in activated B cells after 48 h was between 10- and 70-fold (Fig. 1,B). We also performed intracellular staining for PPARγ in activated B cells. Here, treatment with anti-IgM, CpG, or anti-IgM plus CpG increased intracellular PPARγ levels compared with B cells that were untreated (Fig. 1 C). Collectively, these results indicate that PPARγ expression is increased by agents that trigger B cell activation and differentiation.

The natural PPARγ ligand 15d-PGJ₂ is derived from its precursor, PGD₂, by a series of dehydration steps (24). Physiological concentrations of 15d-PGJ₂ are estimated to reach at least nanomolar concentrations (25). Additionally, therapeutic blood levels of the synthetic PPARγ ligand rosiglitazone reach low micromolar levels (26). To examine the role of PPARγ in B cell function, we first examined the effects of physiologically relevant concentrations of rosiglitazone and 15d-PGJ₂ on B cell proliferation. B cells were labeled with the cell division-tracking dye CFSE and activated for 5 days with CpG in the presence or absence of rosiglitazone (0.5 μM) or 15d-PGJ₂ (0.2 μM). These doses were chosen based on pilot experiments; these concentrations did not adversely affect cell viability (based on 7-aminoactinomycin D incorporation, cell size and [³H]thymidine incorporation; data not shown). Five days after activation, cells were analyzed by flow cytometry. Nonactivated B cells treated with PPARγ ligands did not proliferate (Fig. 2,A, left panel). However, activated B cells incubated with either rosiglitazone (0.5 μM) or 15d-PGJ₂ (0.2 μM) increased cell division (Fig. 2,A, dotted histograms) compared with vehicle control (Fig. 2,A, shaded histograms). Fig. 2,B illustrates the percentage of cell division for three different B cell donors. A similar trend was observed with all three donors (Fig. 2 B), where there was an increase in the percentage of cell division (∼8–40%) in activated B cells treated with PPARγ ligands compared with vehicle control. These results indicate that low doses of PPARγ ligands enhance B cell proliferation.

Next, we evaluated whether activation of PPARγ influenced the differentiation of B cells into Ab-secreting plasma cells. Fig. 2,C shows that CpG significantly induced both IgM and IgG production. Moreover, both rosiglitazone (0.5 μM) and 15d-PGJ₂ (0.2 μM) further enhanced IgM and IgG production, by up to 2-fold, over CpG alone (Fig. 2 C). Thus, activation of PPARγ significantly increases Ab production.

To test whether the PPARγ ligand concentrations used here activated PPARγ, normal B cells were transfected with a PPARγ luciferase reporter construct. Eighteen hours posttransfection, cells were either treated with rosiglitazone (0.5 μM) or with 15d-PGJ₂ (0.2 μM) in the presence or absence of CpG. Nonactivated B cells did not increase luciferase activity when treated with PPARγ ligands (Fig. 2,D). CpG-activated B cells with no exogenous PPARγ ligand also did not induce PPARγ activity. However, both rosiglitazone and 15d-PGJ₂ increased luciferase activity, indicating activation of PPARγ (Fig. 2 D). Therefore, activated B cells, which have higher PPARγ levels, can respond to PPARγ ligands, while nonactivated B cells, with low PPARγ expression, were not able to activate PPARγ upon low-dose PPARγ ligand exposure.

PPARγ forms a heterodimer with the 9-cis-RA receptor, RXRα (27). We hypothesized that 9-cis-RA, in conjunction with PPARγ activation, would enhance B cell proliferation. Normal B cells activated with CpG were treated with vehicle, rosiglitazone (Rosi) or 15d-PGJ₂ with or without 9-cis-RA (9-RA), and proliferation was measured at 5 days postactivation using CFSE labeling (Fig. 3). Five days post-CpG activation, cells that were treated with PPARγ ligands alone or 9-cis-RA alone increased the percentage of dividing cells (Fig. 3). Moreover, results from three separate donors indicate that there was a 2–3-fold increase in the percentage of cells dividing with 9-cis-RA plus rosiglitazone or 15d-PGJ₂ (Fig. 3). These results show that PPARγ ligands and RXRα ligands enhance B cell proliferation.

Peripheral blood B lymphocytes include both naive and memory B cell populations. These two B cell subsets can be distinguished based on CD27 expression, which is a marker of memory B cells (28). Since CD38 up-regulation is a marker of B cell differentiation (29, 30), we evaluated whether PPARγ ligands had an effect on CD38 surface expression in both naive (CD27⁻) and memory (CD27⁺) B cells. Nonstimulated B cells have no changes in differentiation markers upon PPARγ ligand treatment (data not shown). CpG treatment alone yielded 7.0 ± 1.7% CD38^highCD27^high B cells, indicative of plasma cells (Fig. 4,A, see upper right quadrant). The percentage of CD38^highCD27^high cells increased to 10.7 ± 1.6% with rosiglitazone (Rosi, ∼1.7-fold over vehicle) and to 12.5 ± 1.4% with 15d-PGJ₂ (∼2-fold over vehicle) (Fig. 4, B and C). In contrast, PPARγ ligands had little effect on CD38 expression in naive (CD27⁻) B cells (Fig. 4 A–C, see bottom right quadrants). This suggests that PPARγ ligands increase memory B cell differentiation to plasma cells.

To assess whether the effects of the PPARγ ligands were PPARγ dependent, a widely used small molecule PPARγ irreversible antagonist, GW9662, was used. GW9662 covalently modifies the PPARγ ligand-binding site and acts as an irreversible antagonist (12, 31). The results indicate that PPARγ ligand-induced CD38 expression in memory B cells is attenuated with GW9662 (Fig. 4,D–F). Treatment with 9-cis-RA increased the percentage of CD38^highCD27^high to 10.3 ± 1.9% (∼1.7-fold vs vehicle) and increased the percentage of CD38-expressing naive B cells (CD38^highCD27^low) from 9.1 ± 0.6% in CpG plus vehicle to 30.2 ± 5.5% in CpG plus 9-cis-RA (∼3.5-fold vs CpG plus vehicle) (Fig. 4, A and G, bottom right quadrants). Strikingly, the combination of PPARγ ligands plus 9-cis-RA further induced CD38 expression in both naive (bottom right quadrants) and memory (upper right quadrants) B cells by ∼2-fold compared with 9-cis-RA alone (compare Fig. 4,G with Fig. 4, H and I). Thus, PPARγ ligands enhance B cell differentiation of CpG-stimulated memory B cells, but not naive B cells, in a PPARγ-dependent manner. This suggests that activation of PPARγ/RXR heterodimers is a novel regulatory pathway for stimulating B cell differentiation.

Since the effects of PPARγ ligands on B cell differentiation markers was PPARγ dependent (Fig. 4), we next asked whether PPARγ ligand-induced Ab production was also PPARγ dependent. Again, PPARγ ligands significantly induced IgM and IgG production over vehicle control in CpG-activated cells (Fig. 5). Treatment with the PPARγ antagonist GW9662 abolished the effects of PPARγ ligands on IgG (Fig. 5,A), but not IgM, production (Fig. 5 B). This suggests that activation of PPARγ is necessary for PPARγ ligand-induced IgG, but not IgM, production.

We also asked whether PPARγ ligands, in combination with 9-cis-RA, would further enhance Ab production. CpG-activated B cells were treated with 9-cis-RA (9-RA) alone or in combination with rosiglitazone (Rosi) or 15d-PGJ₂. Treatment with 9-cis-RA significantly induced both IgM and IgG production (Fig. 5). Addition of rosiglitazone with 9-cis-RA significantly enhanced IgG (Fig. 5,A), but not IgM (Fig. 5,B), production compared with 9-cis-RA alone. However, when combined treatment (9-RA plus Rosi) was compared with rosiglitazone alone, both IgM and IgG were significantly induced. Addition of 15d-PGJ₂, together with 9-cis-RA, also resulted in a significant increase in both IgM and IgG production compared to15d-PGJ₂ alone (Fig. 5). These results indicate that combining PPARγ and RXR ligands further enhances Ab production.

Our laboratory has demonstrated that CpG induces COX-2 expression in B cells (32), which is important for B cell differentiation (20, 32). Therefore, we evaluated the levels of COX-2 expression following PPARγ ligand treatment. Nonactivated B cells treated with PPARγ ligands showed no increase in COX-2 expression (data not shown). However, CpG activation increased the percentage of B cells, which express COX-2, from 3% (untreated) to 27% (Fig. 6,A, compare ii with i). The percentage of COX-2-positive B cells was further increased by rosiglitazone (36%) and by 15d-PGJ₂ (43%) (Fig. 6,A, iii and iv, upper right quadrants). Fig. 6,B shows the mean fluorescence intensity (MFI), indicative of the intensity of COX-2 expression. Both CpG and PPARγ ligand treatment increased the levels of COX-2 expression. These results further confirm that PPARγ stimulates COX-2 expression in activated B cells. To determine whether the PPARγ-induced COX-2 expression was responsible for increased Ab production, normal B cells were treated with PPARγ ligands in the presence or absence of the COX-2 selective inhibitor SC-58125. We have previously shown that COX-2 inhibitors reduce CpG-induced IgM and IgG production (32). Addition of SC-58125 to CpG-stimulated B cells significantly reduced IgM and IgG production (Fig. 6,C, open bars). Moreover, the increased Ab production upon PPARγ ligand treatment was also significantly reduced by the addition of SC-58125 (Fig. 6 C). These results suggest that PPARγ ligand-induced COX-2 expression is at least partially responsible for the increase in Ab production.

Lastly, we evaluated BLIMP-1 expression, a transcription factor important in B cell differentiation (33). BLIMP-1 protein levels were significantly up-regulated in response to CpG treatment in normal B cells compared with untreated or freshly isolated B cells (Fig. 6, D and E). When B cells were treated with a combination of CpG and PPARγ ligands, there was a further increase in BLIMP-1 expression. Densitometric analysis shows an induction of ∼6- and ∼9-fold with CpG plus rosiglitazone and CpG plus 15d-PGJ₂ treatment, respectively, over CpG-treated cells (Fig. 6 D).

Finally, we assessed if the increase in BLIMP-1 by rosiglitazone and 15d-PGJ₂ was PPARγ dependent. In B cells treated with CpG and GW9662, there was a decrease in the expression of BLIMP-1 compared with CpG alone (Fig. 6,E, compare lanes 3 and 6). The increase in BLIMP-1 by treatment of CpG-activated B cells with rosiglitazone and 15d-PGJ₂ was dramatically attenuated by GW9662 (Fig. 6 E, compare lanes 4 and 5 with lanes 7 and 8). Collectively, these results support our hypothesis that PPARγ ligands enhance B cell differentiation.

The differentiation of B lymphocytes into Ab-producing plasma cells is necessary for protection against invading microorganisms and for successful vaccination. Augmenting Ab responses not only could improve normal humoral immune responses, but could also improve the outcome of patients with immune deficiencies or those who are immunosuppressed, elderly, or very young. In this study, we present new evidence that PPARγ is a novel regulator of B cell differentiation and Ab production. We demonstrate that PPARγ levels increase in B cells upon TLR9 activation and BCR cross-linking. Since these mitogenic stimuli induce B cell differentiation, our results suggest that PPARγ plays an important role in B cell function. Moreover, physiological (nM) doses of PPARγ ligands alone or in combination with RXRα ligands accelerated the differentiation of B cells into plasma cells and increased Ig synthesis. This supports the concept that, in normal B cells, PPARγ activation is an important pathway that can be exploited to boost humoral immune responses.

Certain PPARγ ligands are recognized as having antiinflammatory properties and can be antiproliferative agents in immune cells (34). In most studies, including our own (8, 9, 10, 35, 36), the effects of PPARγ ligands have been studied at high micromolar concentrations, at which PPARγ-independent effects can be observed (10), especially with electrophilic PPARγ ligands such as 15d-PGJ₂. Herein, we demonstrated that nanomolar concentrations of both an endogenous PPARγ ligand (15d-PGJ₂) and a synthetic ligand (rosiglitazone) enhance B cell proliferation and Ig production. Many of the effects observed on B cell differentiation at nanomolar concentrations of 15d-PGJ₂ and rosiglitazone were reversible upon treatment with a highly specific PPARγ antagonist, GW9662 (31) (Figs. 4–6). These observations agree with findings on nonimmune cells, such as epithelial cells, where nanomolar concentrations of PPARγ ligands increase cell proliferation in a PPARγ-dependent manner, whereas (high) micromolar concentrations inhibit proliferation in a PPARγ-independent manner (37, 38, 39). Additionally, the ability of a cell to respond to PPARγ ligands may be a direct reflection of the level of PPARγ protein expression. In the present study, we demonstrated that PPARγ levels increase upon B cell activation. These results mirror our previous studies on T cells, where PPARγ levels increased upon T cell activation (40). PPARγ expression also increases during the differentiation of monocytes to macrophages, and PPARγ/RXR signaling induces macrophage differentiation (41, 42). This increase in PPARγ expression may help normal B cells respond to endogenous PPARγ ligands (e.g., 15d-PGJ₂). Indeed, we did not observe any change in B cell function in nonactivated B cells (i.e., those with low PPARγ expression) that were exposed to PPARγ ligands (data not shown). Taken together, these results indicate that physiologically relevant concentrations of PPARγ ligands induce differentiation of B lymphocytes through a PPARγ-dependent process.

The transcriptional actions of PPARγ depend on its dimerization partner RXR. Peripheral blood B lymphocytes express RXRα (43). Vitamin A is important for optimal humoral immune responses (44, 45, 46, 47, 48, 49). Treatment of peripheral blood B cells with the vitamin A metabolite all-trans-retinoic acid induces CD38 expression and increases Ab production (45). Although all-trans-retinoic acid does induce B cell differentiation, it only binds to the retinoid acid receptor (RAR). In contrast, 9-cis-RA, a vitamin A metabolite, is a ligand for both RAR and RXR (19, 50). RXR can heterodimerize with other receptors, including RAR (51). Thus, the ability of 9-cis-RA to robustly increase Ab production (Fig. 5), compared with PPARγ ligands alone, may be a reflection of its ability to activate both RAR and PPARγ signaling pathways.

Moreover, certain studies have shown synergistic effects with RXR and PPARγ ligands on cell differentiation (52, 53). Herein, we observed additive effects on B cell differentiation when PPARγ ligands were used in combination with 9-cis-RA (Figs. 3–5). The combined effect observed with PPARγ ligands and 9-cis-RA suggests that activation of the PPARγ/RXR pathway enhances B cell differentiation. TLR signals such as CpG are sufficient to induce BLIMP-1 expression (54). We found that PPARγ ligands enhanced CpG-induced BLIMP-1 expression by 6- to 9-fold over CpG alone (Fig. 6, D and E). This increase in BLIMP-1 was PPARγ-dependent, as GW9662 reduced BLIMP-1 expression in CpG-activated B cells that were treated with rosiglitazone or 15d-PGJ₂ (Fig. 6 E). Thus, BLIMP-1 induction may be due to a direct transcriptional regulation by PPARγ on BLIMP-1.

The ability of the PPARγ ligands rosiglitazone or 15d-PGJ₂ to regulate Ab production is partially PPARγ dependent. This was demonstrated by the fact that the PPARγ antagonist GW9662 significantly decreased PPARγ ligand-induced IgG (Fig. 5,A) but not IgM (Fig. 5,B). This suggests that PPARγ may not regulate the primary immune response, in which IgM is the first Ig class produced, but rather may regulate the ability of B cells to class-switch. Additionally, these ligands also increased CpG-induced COX-2 expression (Figs. 6, A and B). We have previously published that COX-2 is increased after B cell activation and its activity is crucial for optimal Ab production (20, 32, 55). This increase in COX-2 may permit more B cells to differentiate to Ab-secreting cells. Indeed, the addition of a COX-2 selective inhibitor attenuated IgM and IgG induction by PPARγ ligands (Fig. 6 C). Despite the fact that Ab production by PPARγ ligands is only partially PPARγ dependent, our data clearly demonstrate that COX-2 activity is essential for the enhanced Ab production elicited by rosiglitazone or 15d-PGJ₂. Thus, activation of PPARγ, in concert with COX-2, may be a novel mechanism for regulating B cell differentiation and class switching during an immune response.

The ability of a B cell to mature, differentiate, and produce Ab is a complex process and often involves cells within the periphery, particularly T cells. We and others have shown that PPARγ profoundly affects T cell function (56, 57, 58). Interestingly, a reduction in PPARγ expression increases T cell proliferation and skews toward Th1 immune response (59), which includes increased IFN-γ and IL-12 production (59, 60). These cytokines can directly influence B cell function, including plasma cell formation (61), proliferation (62), and Ab production (63). The alteration in T cell function caused by reduced PPARγ expression may account for the results obtained by Setoguchi and colleagues (60). Here, utilizing B cells derived from PPARγ haploinsufficient (PPARγ^+/−) mice, where PPARγ expression is reduced by 50% (64), they demonstrated that this reduction in PPARγ expression resulted in enhanced B cell proliferation and serum IgG and IgM levels (60). In their study, it seemed that the loss of PPARγ in B cells, rather than its activation (as described herein in Figs. 2, 5, and 6) exerts control over B cell function, particularly Ab production. However, the contribution of reduced PPARγ expression in mouse T cells (and other APCs) could not be excluded. It is possible that, in the PPARγ^+/− mice, T cell activation (caused by reduced PPARγ expression) and subsequent interaction with primed B cells account for the heightened B cell proliferation and Ab production observed in the PPARγ^+/− mice.

B cells are a critical component of both innate and adaptive immunity. Activation and subsequent differentiation of B cells in response to antigenic challenge is required for successful clearance of a pathogen. Our new findings show that activation of normal human B cells increases PPARγ protein levels, and that PPARγ activation increases cell differentiation. The concomitant use of PPARγ ligands plus 9-cis-RA greatly enhances B cell differentiation. Up-regulation of PPARγ, together with its activation by prostaglandins and RXRα ligands, represents a novel regulatory pathway for B cell differentiation. This new pathway could be exploited to enhance desirable Ab responses.

The authors have no financial conflicts of interest.