IL-23 and IL-1β Drive Human Th17 Cell Differentiation and Metabolic Reprogramming in Absence of CD28 Costimulation

Abstract

Th17 cells drive autoimmune disease but also control commensal microbes. A common link among antigens from self-proteins or commensal microbiota is relatively low activation of T cell receptor (TCR) and costimulation signaling. Indeed, strong TCR/CD28 stimulation suppressed Th17 cell differentiation from human naive T cells, but not effector/memory cells. CD28 suppressed the classical Th17 transcriptional program, while inducing known Th17 regulators, and acted through an Akt-dependent mechanism. Th17 cells differentiated without CD28 were not anergic: they showed robust proliferation and maintained Th17 cytokine production following restimulation. Interleukin (IL)-23 and IL-1β promoted glucose uptake and increased glycolysis. Although modestly increased compared to CD28 costimulation, glycolysis was necessary to support Th17 differentiation, indicating that cytokine-mediated metabolic shifts were sufficient to obviate the classical requirement for CD28 in Th17 differentiation. Together, these data propose that, in humans, strength of TCR/CD28/Akt activation serves as a rheostat tuning the magnitude of Th17 development driven by IL-23 and IL-1β.

Keywords: CD28; IL-1; IL-17; IL-23; Th17; differentiation; human; metabolism.

Figure 1. TCR/CD28 Costimulation Inhibits IL-23/IL-1β-Driven Th17 Differentiation

Human naive T cells were activated with anti-CD3 with or without anti-CD28 under Th0, Th17 (IL-23 + IL-1β), or Th1 (IL-12 + IL-2) conditions as indicated for 7 days before analysis of cytokines by flow cytometry, gating on live CD4⁺CD45RO⁺ cells. (A) Representative flow cytometry plots. (B) IL-17A-producing cells in indicated condition, each point represents a unique donor (Th0 n = 19; Th17 n = 28; Th17+TGFβ n = 7; Th0+IL-23 or IL-1β n = 3). (C) IFNγ expression under Th17 and Th1 conditions and with or without αCD28, each point represents a unique donor. (D) Live CD4⁺ cells identified as Ghostdye negative. (E and F) Th17 cells were stimulated with anti-CD3 (5 μg/mL), IL-23, IL-1β, and indicated concentration of anti-CD28, representative FACS plots (E), and pooled data below indicate frequency of IL-17⁺ cells as fold change over Th17 condition for each donor (F); n = 5 unique donors. (G) Th17 cells were stimulated with 5 μg/mL anti-CD3 or 1 μg/mL anti-CD28 and the indicated concentration of anti-CD3, IL-17⁺ cells shown as fold change over Th17 condition for each donor (n = 5 unique donors). Data are represented as mean ± SD.

Figure 2. Effect of CD28 Costimulation on Th17 Gene Signature

Th0 and Th17 cells were activated for 5 days before RNA sequencing gene expression analysis. (A) Heatmap shows gene expression (FPKM) values normalized relative to mean expression for that gene for each of two unique donors. (B) qPCR analysis of TCF7 mRNA on day 5 of Th17 cultures ± CD28 as indicated; n = 10 donors. (C) IL-22 assayed in supernatant on day 7; n = 11. (D) IL-2 assayed by ELISA on day 5; n = 10. (E–G) qPCR analysis of (E) GFI1, (F) OSM, and (G) LAG3 mRNA on day 5 of Th17 cultures ± CD28 as indicated; n = 8–10. (H) Foxp3+ cells analyzed by flow cytometry on day 7 of indicated cultures; n = 13–18 donors. (I and J) qPCR analysis of (I) JUN and (J) JUND mRNA on day 5 of Th17 cultures ± CD28 as indicated; n = 10 donors. Data are represented as mean ± SD.

Figure 3. Th17 Cells Generated without CD28 Costimulation Are Not Anergic

Th17 cells generated without anti-CD28 (Th17^CD28null) or with anti-CD28 (Th17^CD28) were rested with IL-2 on day 7, restimulated under original conditions or with changed anti-CD28 as indicated on day 19, and then rested for a further 5 days with IL-2. Cells were harvested, counted, and re-plated at a constant number at each rest or restimulation stage, and FACS analysis was performed to determine IL-17 and IFNγ production in live CD4⁺CD45RO⁺ cells at each time point. (A) Representative FACS plots for one donor. (B) Mean IL-17 production over time for n = 6 donors. (C) Mean IFNγ production over time for n = 6 donors. (D) T cell expansion calculated as cumulative cell numbers per starting cell in each condition for n = 6 donors; on days 19 and 24, the difference between Th17^CD28null-Th17^CD28null and Th17^CD28-Th17^CD28 is statistically significant; other groups not significantly different from each other. (E) Frequency of CD45RO⁺ cells at day 7 of culture in indicated conditions; n = 20–26 donors. (F) Mean CD45RO expression in restimulation cultures for n = 6 donors. Data are represented as mean ± SD.

Figure 4. Th17 Cells Become Resistant to CD28 Suppression

(A) Naive CD4⁺ T cells were cultured under Th17 conditions with anti-CD28 added on indicated days after initiation of culture and IL-17 expression analyzed on day 7 by flow cytometry and shown relative to Th17^CD28null condition for each donor (n = 5). (B) CD4⁺CD45RO⁺ memory T cells were isolated from PBMCs and cultured under indicated conditions for 7 days and IL-17 expression analyzed on day 7 by flow cytometry (n = 5). Data are represented as mean ± SD.

Figure 5. Akt Activity Modulates Th17 Development

(A) IL-17 expression in T cells cultured with IL-23, IL-1β, and presence or absence of anti-CD28 or IL-2 as indicated (n = 8 donors). (B and C) pSTAT5 representative FACS plots (B) and pooled data analyzed on day 7 (C); n = 5. (D and E) pSTAT3 representative FACS plots (D) and pooled data analyzed on day 7 (E); n = 9. (F) Th17 cells cultured in presence or absence of anti-CD28 with anti-IL-2 as indicated; IL-17 expression shown as fold change relative to Th17^CD28null (no anti-IL-2) condition for each donor; n = 13. (G and H) Akt inhibitor was added to cultures of Th17 cells ± anti-CD28 at indicated doses and IL-17 expression analyzed on day 7 by flow cytometry; (G) is pooled data from n = 7 donors represented as fold change over Th17^{CD28 null} condition for each donor; n = 7; (H) is representative FACS plots from one donor. Data are represented as mean ± SD.

Figure 6. IL-23 and IL-1β Drive Increased Metabolism Necessary for Th17 Differentiation

(A) Th0 and Th17 cells were activated with/without anti-CD28 for 5 days before RNA sequencing gene expression analysis; heatmap shows gene expression (FPKM) values normalized relative to Th0 for each of two unique donors. (B) Th0 and Th17 cells were activated with/without anti-CD28 and glucose uptake assessed by uptake of glucose analog 2NBDG added for final 30 min of culture on day 7; n = 5. (C) Extracellular acidification rate (ECAR) measured by Seahorse in indicated groups on day 7 of culture, with addition of glucose, oligomycin, and 2DG at indicated times. (D) Maximal glycolysis measured as increase in ECAR following addition of glucose; n = 10 individual donors. (E) Representative images of indicated cell culture conditions on day 7. (F) Basal oxygen consumption rate (OCR) measured by Seahorse in indicated groups on day 7 of culture; n = 5 individual donors. (G) Spare respiratory capacity assessed as increased OCR following addition of FCCP; n = 5 individual donors. (H) Th17 cells were activated with/without anti-CD28 for 7 days in presence of indicated concentration of the glycolysis inhibitor 2DG before cytokine production analysis by flow cytometry; n = 6 donors. Data are represented as mean ± SD.