STAT4 isoforms differentially regulate Th1 cytokine production and the severity of inflammatory bowel disease

Abstract

STAT4, a critical regulator of inflammation in vivo, can be expressed as two alternative splice forms, a full-length STAT4alpha, and a STAT4beta isoform lacking a C-terminal transactivation domain. Each isoform is sufficient to program Th1 development through both common and distinct subsets of target genes. However, the ability of these isoforms to mediate inflammation in vivo has not been examined. Using a model of colitis that develops following transfer of CD4(+) CD45RB(high) T cells expressing either the STAT4alpha or STAT4beta isoform into SCID mice, we determined that although both isoforms mediate inflammation and weight loss, STAT4beta promotes greater colonic inflammation and tissue destruction. This correlates with STAT4 isoform-dependent expression of TNF-alpha and GM-CSF in vitro and in vivo, but not Th1 expression of IFN-gamma or Th17 expression of IL-17, which were similar in STAT4alpha- and STAT4beta-expressing T cells. Thus, higher expression of a subset of inflammatory cytokines from STAT4beta-expressing T cells correlates with the ability of STAT4beta-expressing T cells to mediate more severe inflammatory disease.

FIGURE 1

T-cells expressing STAT4 isoforms have differential TNF-α production. A, CD4+CD62L+ T-cells from mice of the indicated genotypes were cultured under Th1 priming conditions (IL-12, anti-IL-4, α-CD3, α-CD28) with irradiated APCs (30 Gy) for five days. Every 24 hours, supernatants of the developing Th1 cells were collected from each genotype. Cell free supernatants were analyzed for IFN-γ production using ELISA. Results are represented as mean ± SD. B, Cells cultured as in (A) for five days were stimulated for 24 hours and cell-free supernatants were analyzed for IFN-γ using ELISA. Results are represented as mean ± SD and are representative of 3 independent experiments. C, CD4+CD62L+ T-cells were cultured as in (A) for five days. Cells were collected, washed, and stimulated with PMA and ionomycin in the presence of Golgi-Plug before intracellular staining for the indicated cytokines. Data shown are gated on CD4+ cells. Numbers represent % of cells in the respective quadrant while numbers in parentheses represent the MFI of the x-axis. Results are representative of 3 independent experiments. D, RNA was isolated from Th1 cells cultured as in (A) following 4 hours of treatment with anti-CD3. Quantitative PCR was performed for Tnfa mRNA and normalized for β2m expression. Results are relative to WT cells. E, Cells cultured under Th1 priming conditions for five days were stimulated in the indicated condition for 24 hours before cell-free supernatants were collected for analysis of TNF-α. F, Cells cultured as in (A) for five days were stimulated for 24 hours and cell-free supernatants were analyzed by ELISA for TNF-α and IL-2. Results are represented as mean ± SD and are representative of 2–4 independent experiments. *, significantly different (p<0.05) from wild-type, Stat4α, and Stat4−/− Th1 cultured cells using unpaired Student’s T-test.

FIGURE 2

Activation kinetics of the STAT4 isoforms during Th1 differentiation. A, Naïve CD4+ T cells freshly isolated (0 time point) or cultured in Th1 conditions for 24, 48, or 72 hours were collected for intracellular staining with anti-pStat4. Results are representative of 2 independent experiments. Numbers in quadrants represent % of pSTAT4+ T cells. B, Total cell extracts from WT and STAT4 transgenic T cells were immunoblotted for STAT4 protein levels at day 1, 2 and 5 of Th1 differentiation. Data is presented as arbitrary units of densitometry normalized to actin expression and relative to WT day 1 in the left panel or WT day 5 in the right panel. C, Th1 cells cultured for 5 days were washed and stimulated with IL-12 for the indicated time points before being intracellular stained for pSTAT4. Data are shown for the averages of duplicate samples of representative data. D, Th1 cells were stimulated with IL-12 and IL-18 for 24 hours and cell-free supernatants were analyzed for IFN-γ using ELISA. Results are shown as mean ± SD.

FIGURE 3

Effects of STAT4α and STAT4β expression on Th17 differentiation. A, CD4+CD62L+ T-cells were cultured in the presence of TGF-β1, IL-6, IL-23, anti-IL-4, and anti-IFN-γ for 5 days. Cells were collected, washed, and stimulated with plate-bound α-CD3 or PMA and Ionomycin (P+I) in the presence of Golgi-Plug before intracellular staining for the indicated cytokines. CD4+ cells were gated and the results were plotted as indicated. Numbers represent % of cells in the respective quadrant. Results are representative of 2 independent experiments. B, Total CD4 T cells were cultured for five days in the presence of IL-23 before restimulation with anti-CD3 and assessing production of IL-17A using ELISA. Results are shown as mean ± SD and are representative of 3 independent experiments. C, T cells cultured as in (B) were stimulated with IL-23 and IL-18 for 24 hours and cell-free supernatants were analyzed by ELISA for IL-17A. Results are shown as mean ± SD of results from 2–4 independent experiments.

FIGURE 4

STAT4α and STAT4β mediate inflammatory bowel disease. A, The change of weight over time is expressed as percent of the original weight. Data represent the mean ± SEM of each group (7–10 mice per group). Mice were sacrificed 14 weeks after T-cell reconstitution. *, CD45RB^low cells are significantly different (p<0.05) from CD45RB^high WT, STAT4α or STAT4β using 2-way ANOVA and unpaired Student’s T-test post-hoc. B, MLN single-cell suspensions were counted and surface stained for CD4 and analyzed by FACS. Absolute cell numbers were calculated from % of CD4+ cells and cell counts (left panel). QPCR was performed for STAT4 using cDNA made from MLN RNA (right panel). C, Gross appearance of representative colon from each group as indicated. D, Representative photomicrographs (100x) of colon from mice of the indicated group were stained with H&E. E, The mean histological scores ± SEM for the SCID mice reconstituted with the CD4+ T-cells as indicated with STAT4α or STAT4β signifying histological scores from the SCID mice reconstituted with the CD45RB^high subset and the low signifying histological scores from the SCID mice reconstituted with the CD45RB^low subset. *, p<0.05 where STAT4β is significantly different from both STAT4α and the CD45RB^low subset using the Mann-Whitney U-test.

FIGURE 5

Cytokine production from STAT4α- and STAT4β-expressing T cells ex vivo. (A and B) Cells were isolated and stimulated as described in Materials and Methods and concentration of cytokines were determined by ELISA and are displayed as mean ± SEM (Stat4α n=9; Stat4β n=10). *, p<0.05, ** p<0.02 using Unpaired Student’s T-test. C, Cells were isolated and stimulated as described in Materials and Methods. The concentration of cytokines were determined by ELISA and are displayed as mean ± SD of pooled MLNs from the SCID mice reconstituted with the CD45RB^high subset of the indicated STAT4 isoform. *, p<0.05 using unpaired Student’s T-test.

FIGURE 6

Increased lamina propria neutrophil infiltration correlates with increased GM-CSF levels seen in the SCID mice reconstituted with the STAT4β isoform. A, PMN scores were determined as described in Materials and Methods. Data are presented as mean ± SEM. *, p<0.05 using Mann-Whitney U test. B, Single cell suspensions from MLNs were pooled from the indicated mice, stimulated with anti-CD3 for 72 hours and cell-free supernatants were analyzed using ELISA for GM-CSF. Data are presented as mean ± SD. *, p<0.05 using unpaired Student’s T-test.

FIGURE 7

STAT4β Th1 cells are programmed to secrete more GM-CSF than STAT4α Th1 cells. A, CD4+CD62L+ T-cells were primed for Th1 differentiation using the same conditions as in Fig. 1. After five days, cells were stimulated for 24 hours and cell-free supernatants were analyzed by ELISA for GM-CSF. Results are represented as mean ± SD and are representative of 3 independent experiments. *, p<0.05 using unpaired Student’s T-test. B, Cells cultured under Th17 conditions as in Fig. 3A for five days were stimulated for 24 hours and analyzed by ELISA for GM-CSF production. Results are presented as mean ± SD and are representative of 2 independent experiments.