The cytokines IL-17A and IL-17F have 50% amino-acid identity and bind the same receptor; however, their functional differences have remained obscure. Here we found that Il17f–/– mice resisted chemically induced colitis, but Il17a–/– mice did not, and that Il17f−/− CD45RBhiCD4+ T cells induced milder colitis in lymphocyte-deficient Rag2–/– mice, accompanied by an increase in intestinal regulatory T cells (Treg cells). Clostridium cluster XIVa in colonic microbiota capable of inducing Treg cells was increased in both Il17f−/− mice and mice given transfer Il17f−/− T cells, due to decreased expression of a group of antimicrobial proteins. There was substantial production of IL-17F, but not of IL-17A, not only by naive T cells but also by various colon-resident cells under physiological conditions. Furthermore, antibody to IL-17F suppressed the development of colitis, but antibody to IL-17A did not. These observations suggest that IL-17F is an effective target for the treatment of colitis.
Access optionsAccess options
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Subscribe to Journal
Get full journal access for 1 year
only $18.75 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We thank Y. Shinkai (Kyoto University) for Rag2−/− mice. Supported by the Science and Technology Research Promotion Program for Agriculture, Forestry, Fisheries and Food Industry (Y.I.); by CREST (Y.I.); and by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (Y.I. and C.T.).
Integrated supplementary information
Supplementary Figure 1 Over-expansion of Tregs in Il17f−/− mice accounts for the attenuated colitis.
(a, b) Colitis was induced by the method described in Fig. 1c for 42 days, and histological analysis for colonic inflammation (a, H-E staining) and gross appearance (b) were performed (n=3/group). (c) CD25-CD45RBhiCD4+ T cells harvested from BALB/c WT or Il17f−/− mouse drain lymph nodes were transferred into BALB/c Rag2−/− mice for 70 days, and gross appearance were performed (n=4/group). (d-g) The effect of anti-CD25 antibody on the development of DSS-colitis in Il17f–/– mice were examined. (d) The protocol for antibody and DSS treatment. Il17f−/− mice were i.p. administrated with anti-CD25 antibody or control IgG followed by DSS administration. (e) Mice were sacrificed on day 15 after DSS treatment, and colon length and macroscopic histology were examined (n=4/group). *P=0.0476. (f) Gating strategy for the analyzed by flow cytometry. (g) Proportions of total Foxp3+ CD4 T cells in cLP cells, CD25+Foxp3+ and CD25-Foxp3+ Tregs in cLP CD4 T cells were shown (n=5/group). ***P=0.0007, ****P<0.0001. Data in (e, g) are expressed as means ± SD (unpaired two-tailed Student’s t-test).
Supplementary Figure 2 Fecal microbiota compositions are different between Il17f−/− and Il17a−/− mice.
(a) 16S ribosomal RNA sequences of fecal commensal microbiota from WT, Il17f−/− and Il17a−/− mice were analyzed, and proportions of major phyla in the whole microflora were presented (n=9/group). *, #, $: P<0.05. *P=0.0438, **P=0.0054, ***P=0.0009, #P=0.0342, ##P=0.0012, ###P=0.0007, $$P=0.0032.(b, c) Fresh feces from WT, Il17f−/− and Il17a−/− or Il17a−/−Il17f−/− mice were collected and the relative content of total commensal C. XIVa and C. XIVa sp. (b) or commensal Lactobacillus (c) was determined by qPCR (WT n=11, Il17f−/− n=10, Il17a−/− n=4in b; WT n=7, Il17f−/− n=6, Il17a−/−Il17f−/− n=6 in c). *, #: P<0.05.b, *P=0.0435, #P=0.0493, **P=0.0014, ##P=0.0017. c, *P=0.033, #P=0.0106, **P=0.0032. (d) Analysis in (a) were performed, and proportions of indicated species in C. XIVa in the whole microflora are shown. *, #, $: P<0.05. *P=0.028, **P=0.0062, ##P=0.0013, $$P=0.0041. (e) Il17f−/− mice were treated with antibiotics mixture VAMN or vancomycin only for 3 weeks, and fecal microbiota from vancomycin-treated mice were harvested and mixed with C. bolteae (JCM 12243) and C. clostridioforme (JCM 1291) and were transferred into VAMN-treated mice for 3 weeks. Clearance of C. XIVa in vancomycin-treated mouse feces were confirmed, and cLP Treg population in recipient mice after microbiota transfer was examined by flow cytometry (n=3/group). *, #, $, &: P<0.05. *P=0.0476, #P=0.0306, $P=0.0482, &P=0.0455. (f) SPF WT mice were orally administrated with a mixture of C. bolteae (JCM 12243) and C. clostridioforme (JCM 1291) or P. loescheii (JCM 12249) and P. shahii (JCM 12083) every 2 days for 2 weeks, and relative levels of indicated mRNA in the colon were determined by qPCR (n=4/group). *, #, $, &: P<0.05. *P=0.0251, **P=0.0052, #P=0.035, $P=0.0154, &P=0.0211. (g) Analysis in (a) were performed, and proportions of a part of C. IV species in the whole microflora are shown (n=9/group). Data in (b, e, f) are expressed as means ± SD, and measure of center in (a, c, d, g) is mean (for a-f, one-way ANOVA followed by Tukey’s multiple-comparisons test).
(a) Three week-old WT and Il17f–/– mice were weaned and co-housed in the same cages for 4 weeks. Then, they were separated into different cages. After the indicated period, fecal microbiota was harvested and the relative content of C. XIVa and Prevotella was determined by qPCR (WT n=7, Il17f−/− n=5). *P=0.0119, **P=0.0057 (unpaired two-tailed Student’s t-test). (b) The expression of indicated genes encoding AMPs in WT, Il17f−/− or Il17a−/− mouse colon was examined by qPCR (in panels from Lcn2 to Defa4, n=4/group; in other panels, WT n=12, Il17f–/– n=8, Il17a–/– n=8). *, #: P<0.05. *P=0.0333, **P=0.007, ***P=0.0005, #P=0.0433 (one-way ANOVA followed by Tukey’s multiple-comparisons test). Data are expressed as means ± SD.
(a) C. bolteae, C. clostridioforme, P. loescheii or P. shahii was cultured in the present of indicated AMPs with different concentrations for 48 hr, and bacterial growth/recovery was measured with spectrophotometer (3 replicates for each sample). *, #, $, &: P<0.05. In +Angionenin4 panels, **P=0.0068, ***P=0.0001, ##P=0.0096, $$P=0.0056; in +PhospholipaseA2 panels, **P=0.0049, ***P=0.001, ##P=0.0038, $$P=0.0026; in +β-denfensin1 panels, **P=0.0037, ***P=0.0001, ##P=0.0017, ###P=0.0006, $$$P=0.0002, &&&P=0.0001; in +β-denfensin4 panels, *P=0.0262, **P=0.0011, ***P=0.0002, #P=0.0142, ##P=0.0017, ###P=0.0007. (b-d) C. bolteae, C. clostridioforme, P. loescheii, P. shahii, Lactobacillus murinus or Alcaligenes faecalis was cultured in the presence of indicated AMPs (5 mg/ml of each AMP), and bacterial growth/recovery at indicated time points was measured with spectrophotometer (3 replicates for each sample). b, *P=0.0275. c, *P=0.0123, **P=0.0021. Data are expressed as means ± SD (one-way ANOVA followed by Tukey’s multiple-comparisons test).
(a, b) Proportion of IgA+ B cells in the colon of indicated mice were examined by flow cytometry (n=3–6/group). (a) Cells are pre-gated on cLP 7AAD-Lymphocytes (left) or 7AAD-CD19+Lymphocytes (right) and numbers in dot plot panels indicate percentages of IgA+ B cells in total cLPLs (left) or total cLP B cells (right). (b) Percentages of IgA+ B cells in cLP total B cells in indicated mice are shown (WT n=4, l17f–/– n=3, Il17a–/– n=5, l17f–/–Il17a–/– n=3). WT vs Il17a–/– *P=0.0158, Il17f–/– vs Il17a–/–Il17f–/– *P=0.0356. (c) Expression of Pigr in the colon of indicated mice were examined by qPCR (WT n=12, l17f–/– n=8, Il17a–/– 7). ***P=0.0006. (d-f) Fecal microflora were harvested from WT, Il17f−/− or Il17a−/− mice and were stained by anti-IgA Ab. The IgA+ or IgA- commensal bacteria were purified by FACS sorting, and were further identified as IgA-specific/nonspecific commensals. (d) IgA-bound/non-bound bacterial population before and after the purification was confirmed by flow cytometry. (e) Indicated IgA-specific/nonspecific commensals in indicated groups of mice are shown by the ratio of % in purified total IgA+ bacteria / % in purified total IgA- bacteria. (f) IgA-specific/nonspecific commensals in indicated individual mouse are shown in heat map by using the same data described in (e) (n=3/group). Data in (a, d) are representative of two independent experiments with similar results. Data in (b, c, e) are expressed as means ± SD (for b-c, one-way ANOVA followed by Tukey’s multiple-comparisons test).
(a) The content of Prevotellaceae species the whole microflora is shown (n=9/group). **P=0.0006. Measure of center in each group is mean. (b) Experiment described in Fig. 5e was carried out. Mice were transferred with a mixture of C. bolteae and C. clostridioforme and the proportion of fecal Actinobacteria, Proteobacteria and Lactobacillus was examined by qPCR (n=4/group). (c) Colon pieces from WT mice were stimulated with a mixture of C. XIV (equal CFU of C. clostridioforme + C. bolteae) with indicated doses for 11 hr, and the expression of indicated AMPs was determined by qPCR (3 replicates for each sample). Data in (c) are expressed as means ± SD (for a&c, one-way ANOVA followed by Tukey’s multiple-comparisons test).
Supplementary Figure 7 Correlation between Il17a and Il17f expression and their associated transcription factors.
(a) Lymphocytes isolated from WT or Il17a−/−Il17f−/− mouse spleens were stimulated with or without PMA + Ionomycin for 5 h, and IL-17F- or IL-17A-producing CD4+ T cell subsets were determined by flow cytometry. Data in are representative of two independent experiments with similar results. (b) CD25-CD45RBhi CD4 T cells were sorted from WT or Il17f−/− mouse spleen and the mRNA expression of Il17f, Il17a, Il17ra and Il17rc was examined by real-time qPCR (n=4/group). (c) CD62L+CD4+ naïve T cells purified from WT spleen and LNs were stimulated with anti-CD3 and anti-CD28 Abs, and Il17a and Il17f expression were determined at indicated time points by qPCR. Relative values to naïve T cells are shown (2 replicates for each sample). (d, e) Messenger RNA levels of Il17a and Il17f and their associated transcription factors, Irf4, IκBζ, RoRγt, Runx1 and Batf, in colonic cells from lamina propria or intraepithelial layer described in Fig. 7c were examined by real-time qPCR, and the correlations between Il17a or Il17f and each transcription factor are shown (11 samples in each panel). In the right column of (e), correlation was calculated using colonic cells excluding cLP NK and DCs (8 samples in each panel). Data in (b, c) are expressed as means ± SD.