Peroxisome proliferator-activated receptor gamma ligands enhance human B cell antibody production and differentiation

Abstract

Protective humoral immune responses critically depend on the optimal differentiation of B cells into Ab-secreting cells. Because of the important role of Abs in fighting infections and in successful vaccination, it is imperative to identify mediators that control B cell differentiation. Activation of B cells through TLR9 by CpG-DNA induces plasma cell differentiation and Ab production. Herein, we examined the role of the peroxisome proliferator-activated receptor (PPAR)gamma/RXRalpha pathway on human B cell differentiation. We demonstrated that activated B cells up-regulate their expression of PPARgamma. We also show that nanomolar levels of natural (15-deoxy-Delta(12,14)-prostaglandin J(2)) or synthetic (rosiglitazone) PPARgamma ligands enhanced B cell proliferation and significantly stimulated plasma cell differentiation and Ab production. Moreover, the addition of GW9662, a specific PPARgamma antagonist, abolished these effects. Retinoid X receptor (RXR) is the binding partner for PPARgamma and is required to produce an active transcriptional complex. The simultaneous addition of nanomolar concentrations of the RXRalpha ligand (9-cis-retinoic acid) and PPARgamma ligands to CpG-activated B cells resulted in additive effects on B cell proliferation, plasma cell differentiation, and Ab production. Furthermore, PPARgamma ligands alone or combined with 9-cis-retinoic acid enhanced CpG-induced expression of Cox-2 and the plasma cell transcription factor BLIMP-1. Induction of these important regulators of B cell differentiation provides a possible mechanism for the B cell-enhancing effects of PPARgamma ligands. These new findings indicate that low doses of PPARgamma/RXRalpha ligands could be used as a new type of adjuvant to stimulate Ab production.

Figure 1. PPARγ expression is up-regulated by B cell activation

A. Highly purified human B lymphocytes isolated from peripheral blood were left untreated or were treated for 48 hr with 2 µg/ml anti-IgM Ab, 1 µg/ml of CpG DNA alone, or a combination of CpG plus anti-IgM. Western blots from three individual donors shows immunoreactivity of PPARγ in B cells. PPARγ expression was detectable in untreated cells, with inter-individual variability in expression noted. Upon B cell stimulation, there was an increase in PPARγ expression in all three donors, with Donor 1 exhibiting the greatest increase in protein expression. Total actin expression was used as a protein-loading control. B. Densitometry of the Western blot for all three human B cell donors shows that PPARγ protein levels increase up to 9-fold for anti-IgM, 14.3-fold for CpG and 70-fold with CpG+anti-IgM compared to untreated B cells. C. Flow cytometric analysis of intracellular PPARγ expression confirmes that the level of PPARγ increases upon B cell activation (from ≈ 47% in untreated B cells to 68% in B cells activated with anti-IgM+CpG).

Figure 1. PPARγ expression is up-regulated by B cell activation

A. Highly purified human B lymphocytes isolated from peripheral blood were left untreated or were treated for 48 hr with 2 µg/ml anti-IgM Ab, 1 µg/ml of CpG DNA alone, or a combination of CpG plus anti-IgM. Western blots from three individual donors shows immunoreactivity of PPARγ in B cells. PPARγ expression was detectable in untreated cells, with inter-individual variability in expression noted. Upon B cell stimulation, there was an increase in PPARγ expression in all three donors, with Donor 1 exhibiting the greatest increase in protein expression. Total actin expression was used as a protein-loading control. B. Densitometry of the Western blot for all three human B cell donors shows that PPARγ protein levels increase up to 9-fold for anti-IgM, 14.3-fold for CpG and 70-fold with CpG+anti-IgM compared to untreated B cells. C. Flow cytometric analysis of intracellular PPARγ expression confirmes that the level of PPARγ increases upon B cell activation (from ≈ 47% in untreated B cells to 68% in B cells activated with anti-IgM+CpG).

Figure 2. Normal B cell proliferation and antibody production is enhanced by PPARγ ligands

A. Purified human B cells (0.5×10⁶ cells/ml) were labeled with CFSE and were left untreated (non-stimulated B cells) or were cultured with CpG (1 µg/ml) with or without Rosiglitazone (0.5 µM) or 15d-PGJ₂ (0.2 µM) (CpG-stimulated B cells). Cell division was analyzed by flow cytometry at day 5. A total of 25,000 events were collected for each sample and the data were gated on the live cell population based on forward and side-scatter. The results are representative of three separate experiments. B. The percent cell division for three separate donors is shown. Note that a similar trend was observed with all three donors; PPARγ ligands increased the percentage of cell division from 8–40%. C. Purified B cells were stimulated with CpG (1 µg/ml) for 5 days in the presence and absence of 0.5 µM Rosiglitazone or 0.2 µM of 15d-PGJ₂ and IgM and IgG levels were analyzed by ELISA. Vehicle (DMSO) was included as a negative control. Low doses of both PPARγ ligands significantly induced both IgM and IgG levels. D. Purified human B cells were transfected as described in Materials and Methods with a PPRE-Luciferase construct. Eighteen hours post-transfection, cells were treated with PPARγ ligands in the presence or absence of CpG (1 µg/ml). Twenty-four after treatments, cells were lysed and a luciferase assay was performed. CpG-activated B cells showed increased luciferase activity upon PPARγ ligand treatment.

Figure 3. PPARγ ligands enhance 9-cis-RA induced B cell proliferation

Human B cells were stimulated with CpG (1 µg/ml) and treated with vehicle or with PPARγ ligands (0.5 µM Rosiglitazone or 0.2 µM of 15d-PGJ₂), 9-cis-RA (100 nM) alone or a combination of a PPARγ ligand plus 9-cis-RA for 5 days. CFSE results were expressed graphically as mean percent division at 5 days. Results from three donor preparations are shown.

Figure 4. PPARγ ligands enhance the ability of 9-cis-RA to induce plasma cell differentiation

Peripheral blood B cells were treated with CpG (1 µg/ml) plus vehicle (a), Rosiglitazone at 0.5 µM (b), 15d-PGJ₂ (0.2 µM) (c), GW9662 at (500 nM) alone (d) or in combination with Rosiglitazone (e) or 15d-PGJ₂ (f). Some cells were treated with 9-cis-RA at a 100 nM (g–i) alone (g) or in combination with Rosiglitazone (h) or 15d-PGJ₂ (i). The cells were harvested at 5 days and the frequency of cells with CD38^highCD27^high (Upper right quadrants) and CD38^highCD27^neg/low (Lower right quadrants) phenotype was determined. The values are representative of three separate experiments.

Figure 5. PPARγ ligands and 9-cis-RA enhance antibody production

Purified B cells were stimulated with CpG (1 µg/ml) for 6 days in the presence and absence of 0.5 µM Rosiglitazone or 0.2 µM 15d-PGJ₂ and both IgG (A) and IgM (B) levels were analyzed by ELISA. Vehicle (DMSO) was added as a negative control (left bars). Some cells were also treated in the presence of the PPARγ antagonist GW9662 (500 nM, middle bars) or in the presence of 9-cis-RA (100 nM, right bars). PPARγ ligands significantly induced both IgM and IgG levels. GW9662 abrogated PPARγ ligand-induced IgG, but not IgM, levels. 9-cis-RA also induced both IgM and IgG levels, and when combined with PPARγ ligands, further enhanced IgM and IgG production. *p<0.05; **p< 0.01; ***p<0.001 vs. vehicle treated. $, p<0.05; $$, p<0.01, $$$, p<0.001 and ns (non significance) vs. respective PPARγ ligand alone. ##, p<0.01 vs. 9-cis-RA.

Figure 6. PPARγ ligands increase CpG-induced COX-2 and BLIMP-1 expression

A. Purified B cells were either left untreated (panel i), or were treated with 1µg/ml of CpG and vehicle (panels ii and v), 0.5 µM of Rosiglitazone (panels iii and vi) or 0.2 µM of 15d-PGJ₂ (panels iv and vii). Flow cytometry analysis of purified B cells shows that the percentage of CD19⁺ B cells expressing Cox-2 protein (upper right quadrants) was induced upon activation (27% on CpG+Vehicle vs. 3% on untreated). Cells treated with Rosiglitazone or 15d-PGJ₂ further increased the percentage of Cox-2 positive cells (36% and 43%, respectively, compared to 27% of CpG+vehicle control). B. Results are expressed as Cox-2 mean fluorescence intensity (MFIs).**p<0.01 versus untreated; C. Purified B cells were stimulated with CpG (1 µg/ml) for 6 days in the presence and absence of 0.5 µM Rosiglitazone or 0.2 µM 15d-PGJ₂ and both IgM and IgG levels were analyzed by ELISA. Vehicle (DMSO) was added as a negative control (left bars). Some cells were also treated in the presence of the Cox-2 selective inhibitor SC-58125 at a concentration of 10 µM (right bars). PPARγ ligands significantly induced both IgM and IgG levels. SC-58125 abrogated PPARγ ligand-induced IgG and IgM levels. *p<0.05, **, p< 0.01 and ***, p<0.001 vs. vehicle treated; ###p<0.001 vs. respective PPARγ ligand. D. Normal B cells were lysed immediately after isolation, were left untreated for 72 hr or were treated with CpG (1 µg/ml) alone or with PPARγ ligands for 72hrs. BLIMP-1 expression was analyzed by Western blot as indicated; representative western blot is shown. Total actin was used to normalize protein loading. BLIMP-1 levels were up-regulated upon CpG activation and PPARγ ligands further increased CpG-induced BLIMP-1 expression. Unstimulated B cells treated with PPARγ ligands had no effect on BLIMP-1 expression (data not shown). Graph: Densitometry of the Western blots show that the CpG-activated B cells increased BLIMP-1 protein levels. Treatment with either Rosiglitazone (Rosi) or 15d-PGJ₂ significantly increased BLIMP-1 expression compared to CpG (*p<0.05). E. GW9662 attenuates BLIMP-1 protein expression. Expression of BLIMP-1 was assessed by western blot in B cells that were freshly isolated, untreated, or were activated by CpG in conjunction with Rosiglitazone (Rosi; 0.5 µM) or 15d-PGJ₂ (0.2 µM); some cells were also exposed to the PPARγ antagonist GW9662 (500 nM). Treatment with GW9662 reduced BLIMP-1 expression in B cells that were treated with CpG+Vehicle, as well as those treated with Rosiglitazone or 15d-PGJ₂.

Figure 6. PPARγ ligands increase CpG-induced COX-2 and BLIMP-1 expression

A. Purified B cells were either left untreated (panel i), or were treated with 1µg/ml of CpG and vehicle (panels ii and v), 0.5 µM of Rosiglitazone (panels iii and vi) or 0.2 µM of 15d-PGJ₂ (panels iv and vii). Flow cytometry analysis of purified B cells shows that the percentage of CD19⁺ B cells expressing Cox-2 protein (upper right quadrants) was induced upon activation (27% on CpG+Vehicle vs. 3% on untreated). Cells treated with Rosiglitazone or 15d-PGJ₂ further increased the percentage of Cox-2 positive cells (36% and 43%, respectively, compared to 27% of CpG+vehicle control). B. Results are expressed as Cox-2 mean fluorescence intensity (MFIs).**p<0.01 versus untreated; C. Purified B cells were stimulated with CpG (1 µg/ml) for 6 days in the presence and absence of 0.5 µM Rosiglitazone or 0.2 µM 15d-PGJ₂ and both IgM and IgG levels were analyzed by ELISA. Vehicle (DMSO) was added as a negative control (left bars). Some cells were also treated in the presence of the Cox-2 selective inhibitor SC-58125 at a concentration of 10 µM (right bars). PPARγ ligands significantly induced both IgM and IgG levels. SC-58125 abrogated PPARγ ligand-induced IgG and IgM levels. *p<0.05, **, p< 0.01 and ***, p<0.001 vs. vehicle treated; ###p<0.001 vs. respective PPARγ ligand. D. Normal B cells were lysed immediately after isolation, were left untreated for 72 hr or were treated with CpG (1 µg/ml) alone or with PPARγ ligands for 72hrs. BLIMP-1 expression was analyzed by Western blot as indicated; representative western blot is shown. Total actin was used to normalize protein loading. BLIMP-1 levels were up-regulated upon CpG activation and PPARγ ligands further increased CpG-induced BLIMP-1 expression. Unstimulated B cells treated with PPARγ ligands had no effect on BLIMP-1 expression (data not shown). Graph: Densitometry of the Western blots show that the CpG-activated B cells increased BLIMP-1 protein levels. Treatment with either Rosiglitazone (Rosi) or 15d-PGJ₂ significantly increased BLIMP-1 expression compared to CpG (*p<0.05). E. GW9662 attenuates BLIMP-1 protein expression. Expression of BLIMP-1 was assessed by western blot in B cells that were freshly isolated, untreated, or were activated by CpG in conjunction with Rosiglitazone (Rosi; 0.5 µM) or 15d-PGJ₂ (0.2 µM); some cells were also exposed to the PPARγ antagonist GW9662 (500 nM). Treatment with GW9662 reduced BLIMP-1 expression in B cells that were treated with CpG+Vehicle, as well as those treated with Rosiglitazone or 15d-PGJ₂.