Transcription Factor KLF10 Constrains IL-17-Committed Vγ4+ γδ T Cells

Abstract

γδ T cells, known to be an important source of innate IL-17 in mice, provide critical contributions to host immune responses. Development and function of γδ T cells are directed by networks of diverse transcription factors (TFs). Here, we examine the role of the zinc finger TFs, Kruppel-like factor 10 (KLF10), in the regulation of IL-17-committed CD27^- γδ T (γδ^27--17) cells. We found selective augmentation of Vγ4⁺ γδ^27- cells with higher IL-17 production in KLF10-deficient mice. Surprisingly, KLF10-deficient CD127^hi Vγ4⁺ γδ^27--17 cells expressed higher levels of CD5 than their wild-type counterparts, with hyper-responsiveness to cytokine, but not T-cell receptor, stimuli. Thymic maturation of Vγ4⁺ γδ^27- cells was enhanced in newborn mice deficient in KLF10. Finally, a mixed bone marrow chimera study indicates that intrinsic KLF10 signaling is requisite to limit Vγ4⁺ γδ^27--17 cells. Collectively, these findings demonstrate that KLF10 regulates thymic development of Vγ4⁺ γδ^27- cells and their peripheral homeostasis at steady state.

Keywords: IL-17; Innate-like γδ-17; KLF10; homeostasis; γδ T cells.

Figure 1

Kruppel-like factor 10 (KLF10) deficiency promotes peripheral γδ²⁷⁻ cell expansion. (A,B) Overlayed histogram of CD27 expression on γδ T cells [(A), left] and frequency of γδ²⁷⁻ cells [(A), right] and absolute number of γδ²⁷⁻ and γδ²⁷⁺ cells (B) in pooled peripheral lymph nodes (pLN; cervical, axillary, brachial, and inguinal) from wild-type (WT) and KLF10 knockout (KO) mice (n = 5 per group), as determined by flow cytometry. (C) Frequency of CD44^hiCD62L⁻ cells among γδ T cells (left) and of CD44^hiCD62L⁻ γδ T cells among total cells (right) in pLN, spleen (SPL), lung, and peritoneal exudate cells (PEC) from WT and KO mice (n = 5 per group), as determined by flow cytometry. Data are the mean ± SD. (D,E) The ratio of KO to WT γδ²⁷⁻ cells (D), homeostatic expansion of WT versus KO γδ²⁷⁻ cells (E) from pLN cells of Rag-1-deficient mice, which had been co-injected with CTV-labeled CD45.1 WT and CD45.2 KO pLN cells at ratio of 2:1, were gated on CTV^+ γδTCR⁺CD27⁻ cells and then analyzed by flow cytometry after 5 days. Numbers in quadrants of the plot indicate percent of cells in each. Each symbol (A,B) represents an individual mouse; error bars are the mean ± SD. ns, non-significant; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. Data are representative of at least three (A–C) or two (E) independent experiments, or are pooled from two independent experiments (D).

Figure 2

KLF10 deficiency selectively expands innate-like IL-17-committed Vγ4⁺ γδ²⁷⁻ cells. (A) Mean fluorescence intensity (MFI) of each molecule expressed on γδ²⁷⁻ (blue) and γδ²⁷⁺ (red) cells in peripheral lymph nodes (pLN) from wild-type (WT) and knockout (KO) mice (n = 5 per group). (B) Frequency (left) and absolute number (right) of Vγ1⁺, Vγ4⁺, and Vγ1⁻Vγ4⁻ cells among γδ²⁷⁻ and γδ²⁷⁺ cells obtained as in (A). (C) Intracellular IL-17 or IFN-γ expression in pooled pLN cells from WT and KO mice (n = 3 per group) after stimulation with phorbol 12-myristate 13-acetate (PMA) plus ionomycin, gated on γδ²⁷⁺ or γδ²⁷⁻ cells. Number adjacent to the gating indicates the percent of each population. (D) Frequency of IL-17⁺ or IFN-γ⁺ cells among γδ²⁷⁺ and γδ²⁷⁻ cells (left) and their absolute number (right) in WT and KO mice determined as in (C). (E) Frequency of IL-17⁺ cells among CCR6⁺ (top, left) or Vγ4⁺ (top, right) γδ T cells and their absolute number (bottom) in pooled pLN cells from WT (n = 6) and KO mice (n = 8) after stimulation with PMA plus ionomycin. (F) IL-17 MFI of IL-17⁺Vγ4⁺CCR6⁺ γδ T cells determined as in (E). (G) Real-time reverse transcription PCR analysis of Rorc expression in sorted Vγ4⁺ γδ²⁷⁺, Vγ4⁻ γδ²⁷⁺, Vγ4⁺ γδ²⁷⁻, and Vγ4⁻ γδ²⁷⁻ cells from pooled pLN and spleen of WT and KO mice (n = 10 per group), normalized to the housekeeping gene Efa1. Expression level of Rorc in WT Vγ4⁺ γδ²⁷⁺ cells was set to 1. Data in (B,D,G) are mean ± SD. Each symbol (A,E,F) represents an individual mouse; error bars are the mean ± SD. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. Data are representative of at least three independent experiments (A–F) or two independent experiments (G).

Figure 3

KLF10 inhibits γδ²⁷⁻ cell responsiveness to cytokine but not TCR stimuli. (A) The change in numbers of Vγ1⁺, Vγ4⁺, and Vγ1⁻Vγ4⁻ of γδ²⁷⁺ or γδ²⁷⁻ cells from wild-type (WT) or knockout (KO) peripheral lymph node (pLN) cells (n = 3 per group) cultured with IL-7 (20 ng/ml) for 5 days, as determined by flow cytometry. The cell number at start was 5 × 10⁶. (B,C) Half-offset histogram of proliferation (B) and intracellular IL-17 expression [(C); after stimulation with PMA plus ionomycin] in γδ T cells sorted from pooled pLN and spleen of WT and KO mouse (n = 10 per group), CTV-labeled and cultured with IL-7 (20 ng/ml) for 5 days. (D) Overlayed histogram of phosphorylated STAT3 or STAT5 in Vγ4⁺ γδ²⁷⁻ or Vγ4⁻ γδ²⁷⁻ cells from pooled WT or KO pLN cells (n ≥ 3 per group) cultured with IL-7 (20 ng/ml) for 30 min and assessed by flow cytometry. Numbers indicate mean fluorescence intensity variation between cells cultured with or without IL-7. Con, control; FMO, fluorescence minus one. (E) Percent of dividing cells among Vγ4⁺ γδ²⁷⁻ or Vγ4⁻ γδ²⁷⁻ cells from CD45.2 KO pLN cells, injected into Rag-1-deficient mouse and then analyzed by flow cytometry after 5 days as in Figure 1D, gated on CD45.2⁺CD27⁻CTV^+ γδTCR⁺ cells. (F) Real-time reverse transcription PCR analysis of Klf10 and Rorc expression (normalized to the housekeeping gene Efa1) in sorted γδ²⁷⁻ cells from pooled pLN and spleen of WT mice (n ≥ 10), cultured with or without IL-7 (20 ng/ml), IL-1β (10 ng/ml) plus IL-23 (20 ng/ml), or TGF-β (10 ng/ml) plus IL-6 (20 ng/ml) for 17 h. Expression levels of Klf10 and Rorc in control (Con) were set to 1. (G) Half-offset histogram of proliferation of γδ T cells sorted as in (B), CTV-labeled and cultured on plated-bound anti-CD3ε (0.1 µg/ml) and anti-CD28 (10 µg/ml) for 3 days. (H) Real-time reverse transcription PCR analysis of Klf10 and Rorc expression (normalized to the housekeeping gene Efa1) in γδ T cells sorted as in (F), cultured on plated-bound anti-CD3ε (αCD3ε; 0, 0.1, 1, and 10 µg/ml) for 3 h. Expression levels of Klf10 and Rorc in cells without anti-CD3ε mAb were set to 1. (I) Intracellular Ca²⁺ mobilization in pLN cells obtained from WT and KO mice (n = 4), stained for surface markers to identify the indicated γδ subsets, stimulated via the TCR with biotinylated anti-CD3ε (20 µg/ml) followed by crosslinkage of the TCR with streptavidin (40 µg/ml) and then assayed over 5 min. Numbers in (B,C,G) indicate the percent of dividing cells (B,G) or of the gated population among IL-17 positive or negative cells in each panel (C). Data in (A,F,H) are mean ± SD. Each symbol (E) represents an individual mouse; error bars are the mean ± SD ns, non-significant; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. Data are representative of at least three independent experiments (A–C,E,I) or two independent experiments (D,F,H).

Figure 4

CD5^loCD127^hi γδ²⁷⁻ cells are IL-17-competent γδ T cells. (A) Overlayed histogram of CD5 and CD127 expression on γδ²⁷⁺ versus γδ²⁷⁻ cells in peripheral lymph node (pLN) cells pooled from wild-type (WT) mice (n ≥ 4), as determined by flow cytometry. ISO, isotype control. (B) Pseudocolor plot of CD5 versus CD127 expression in pLN γδ²⁷⁻ cells from WT and knockout (KO) mice (n ≥ 4 per group). (C) Percent of CD5^loCD127^hi and CD5^hiCD127^lo cells [gated as in (B)] among γδ²⁷⁻ cells (left) and their absolute numbers (right) in pLN cells from WT and KO mice (n ≥ 4 per group). (D,E) MFI of surface CD28, CD25, CD132, CD103, and Ly6C expression (D) and Vγ distribution (E) on CD5^loCD127^hi γδ²⁷⁻, CD5^hiCD127^lo γδ²⁷⁻ [gated as in (B)] and γδ²⁷⁺ cells obtained as in (A) (n = 4). (F,G) Pseudocolor plot of CD5 and CD127 on γδ²⁷⁻ cells (F) and density plot of CD5 and CD127 [(G), left] or Vγ1 and Vγ4 [(G), right] on CD24^hi or CD24^lo γδ²⁷⁻ cells in pLN cells from 2-week-old (young) and 8-week-old (adult) WT mice (n ≥ 4 per each). (H) Overlayed histogram of CD5 and CD127 on IL-17⁺ or IL-17⁻ γδ²⁷⁻ cells obtained as in (A) (n = 3), after stimulation with PMA plus ionomycin. Numbers in outlined areas or quadrants of plots indicate percent of cells in each. Data in (C) are mean ± SD. Each symbol (D) represents an individual mouse; error bars are the mean ± SD. *P ≤ 0.05; ***P ≤ 0.001. Data are representative of at least three independent experiments.

Figure 5

Vγ4⁺ γδ²⁷⁻ cells developmentally express relatively higher CD5 under KLF10-deficient condition. (A) Density plot of CD5 and CD127 on peripheral lymph node (pLN) γδ²⁷⁻ cells from wild-type (WT) mice (n ≥ 4). (B–D) Percent of CD5^loCD127^hi, CD5^intCD127^hi and CD5^hiCD127^lo [gated as in (A)] among γδ²⁷⁻ cells [(B), top], their absolute numbers [(B), bottom] and Vγ distribution (C), and frequency of IL-17⁺ cells among the each gated group [(D), top; after stimulation with phorbol myristate acetate plus ionomycin] and their absolute numbers [(D), bottom] in pLN cells from WT and knockout (KO) mice (n ≥ 5 per group). (E) Density plot of CD5 and Vγ4 on CD127^hi γδ²⁷⁻ cells (left) and percent of Vγ4⁺CD5^int cells among them (right) in pLN cells from WT and KO mice (n = 4). (F,G) Half-offset histogram of CD5 and CD127 (F) and their mean fluorescence intensity (MFI) (G) on CD24^hi or CD24^lo Vγ4⁺ γδ²⁷⁻ cells in thymocytes from neonatal (day 4 after birth) WT and KO mice (n = 5 per group). (H) Density plot of phosphorylated ERK in CD24^hi or CD24^lo cells of Vγ4⁺ γδ²⁷⁻ and γδ²⁷⁺ cells from neonatal thymocytes obtained as in (F), stimulated with soluble anti-CD3ε (1 µg/ml) for 3 min. Con, control. Numbers adjacent outlined areas of plots (E,H) indicate percent of cells in each. Data in (B–D) are mean ± SD. Each symbol (E,G) represents an individual mouse; error bars are the mean ± SD ns, non-significant; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. Data are representative of at least three independent experiments (A–E) or two independent experiments (F–H).

Figure 6

KLF10 deficiency leads to enhanced thymic maturation of Vγ4⁺ γδ²⁷⁻-17 cells in neonates. (A,B) Density plot and percent of CD24^loCD44^hi cells (A) or IL-17⁺ cells [(B); after stimulation with PMA plus ionomycin] among Vγ1⁻Vγ4⁻ or Vγ4⁺ γδ²⁷⁻ cells in thymocytes from neonatal (day 4 after birth) wild-type (WT) and knockout (KO) mice (n ≥ 4 per group). (C) Absolute numbers of CD24^hiCD44^lo and CD24^loCD44^hi cells in Vγ1⁻Vγ4⁻ or Vγ4⁺ γδ²⁷⁻ cells obtained as in (A). (D,E) Percent of IL-17⁺ cells among CD44^lo or CD44^hi cells of Vγ1⁻Vγ4⁻ or Vγ4⁺ γδ²⁷⁻ cells (D) and their absolute numbers (E) obtained as in (B). Numbers adjacent outlined areas of plots indicate percent of cells in each. Each symbol represents an individual mouse; error bars are the mean ± SD ns, non-significant; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. In (C–E), two-way ANOVA followed by a Bonferroni posttest was used to determine significance. Data are representative of three independent experiments.

Figure 7

The uniqueness of Vγ4⁺ γδ²⁷⁻ thymic development. (A) Half-offset histogram of CD5 and CD127 on Vγ1⁺, Vγ4⁺, and Vγ1⁻Vγ4⁻ subsets of γδ²⁷⁺ or γδ²⁷⁻ cells in peripheral lymph node cells from adult (6-week-old) WT mice. (B–D) MFI of CD5 and CD127 [(B); n = 10], robust coefficient of variation (CV) of CD5 MFI [(C); n = 5], and MFI of phosphorylated Zap70 [(D); n = 4] on CD24^hi or CD24^lo cells of each subset [indicated as in (A)] from neonatal (day 3 after birth) thymocytes of WT mice. (E–G) MFI of CD5 on TCRβ^intCD69⁺, TCRβ^hiCD69⁺ and TCRβ^hiCD69⁻ cells of CD4⁺CD8⁺ double-positive (DP) cells (E), density plot of CD4 and CD8 expression on TCRβ^intCD69⁺, TCRβ^hiCD69⁺ and TCRβ^hiCD69⁻ cells (F), and percent of CD24^loQa-2^hiCD62L⁺CD69⁻ cells among CD4⁺ or CD8⁺ TCRβ^hi cells (G) in thymocytes from WT and KO mice (n ≥ 4). Data (E,G) are mean ± SD. Each symbol (B–D) represents an individual mouse; error bars are the mean ± SD. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. Data are representative of at least three independent experiments.

Figure 8

KLF10 restraint of innate-like Vγ4⁺ γδ²⁷⁻-17 cells is cell-intrinsic. (A) Dot plot and percent of Vγ4⁺Vγ1⁻ cells (top) and CD5^loCD127^hi cells (bottom) among peripheral lymph node (pLN) γδ²⁷⁻ cells of CD45.2 WT or KO mice (n = 4), irradiated and reconstituted with CD45.1 WT bone marrow (BM) cells and then assessed by flow cytometry at least 12 weeks later, gated on CD45.1⁺ γδ²⁷⁻ cells. (B,C) CD45.1/2 WT mice (n ≥ 3) were irradiated and reconstituted with a mixture of CD45.1 WT BM plus CD45.2 KO BM cells at ratio 1:1 and then pLN cells were analyzed by flow cytometry at least 12 weeks later. (B) Percent of Vγ4⁺ γδ²⁷⁻ cells (left) or γδ²⁷⁻ cells (right) among WT and KO total γδ T cells or, γδ²⁷⁻ cells (middle) among WT and KO Vγ4⁺ cells. (C) Dot plot (left) and percent (right) of Vγ4⁺ (top) or CD5^loCD127^hi cells (middle) among WT and KO γδ²⁷⁻ cells, or IL-17⁺ cells (bottom) among WT and KO Vγ4⁺ γδ²⁷⁻ cells. (D) Real-time reverse transcription PCR analysis of Klf10 expression in sorted Vγ4⁺ γδ²⁷⁺, Vγ4⁻ γδ²⁷⁺, Vγ4⁺ γδ²⁷⁻, and Vγ4⁻ γδ²⁷⁻ cells from pooled pLN and spleen of WT mice (n = 10 per group), normalized to the housekeeping gene Efa1 and presented as the percent of maximum expression of Klf10. Data (A,D) are mean ± SD. Each symbol represents an individual recipient mouse. A line connects WT-derived cells to KO-derived cells that developed within the same recipient mouse. ns, non-significant; **P ≤ 0.01; ***P ≤ 0.001. Data are representative of two independent experiments (A,D) or are pooled from two independent experiments (B,C).