Animal studies were performed according to the protocol approved by the Institutional Animal Care and Use Committee of Shanghai Jiaotong University School of Medicine, Renji Hospital (#2021-120). To establish a DSS-induced IBD model, 2.5% DSS (MP Biomedicals) was added to the drinking water of C57BL/6J mice for one week. Meanwhile, to establish a 2,4,6-trinitrobenzenesulfonic acid (TNBS)-induced IBD model, 100 ug/g TNBS in 50% ethanol solution was gradually injected into the colon of C57BL/6J mice. IL-10 KO mice were purchased from Wukong biotechnology company. The weight of the mice were measured every day. At the endpoint of the studies, mice were sacrificed, and colon and CD4+ T cells were collected for further experiments.

Human samples

Colonic and CD4+ T cells of healthy donors, UC (including stable and activated status) and CD (including stable and activated status) used in this study were obtained from the Biological Sample Bank of Renji Hospital. This study was approved by the Shanghai Jiaotong University School of Medicine, Renji Hospital Ethics Committee (#KY2021-120-B).

In situ hybridization assay was used to detect the amount of fungi in colonic specimens of IBD patients, according to previous reports (Montone et al., 2011). Briefly, the colonic specimens were fixed, dewaxed, and rehydrated. Then, 18S DNA probes labeled with biotin probes were incubated with the section for one hour, with further incubation using fluorescence streptavidin. Finally, signaling was detected using a microscope.

For histopathology, the colon of IBD mice was collected and fixed after embedding in paraffin. Then, 5 μm tissue sections were stained with hematoxylin and eosin, and data were acquired using a microscope. For immunohistochemistry, the human and mice slides were de-paraffinized, and epitopes were retrieved following endogenous peroxidase blocking. Next, the tissue section was blocked with 10% BSA following primary antibody incubation: Interferon-gamma (abcam, ab231036), with 1:100 dilution; IL-17A (abcam, ab79056), with 1:100 dilution; and TNF-alpha (abcam, ab215188). Secondary antibodies were then incubated at room temperature for two hours. Finally, DAB regent was applied for signaling presentation.

The scores of histopathology and immunohistochemistry were evaluated by two blinded pathologists from Renji Hospital.

ChIP assay was conducted using a ChIP kit (abcam, ab500) following the manufacturer’s guidelines. Briefly, CD4+ T cells isolated from TNBS-induced IBD mice administrated with fungi were fixed, lysed, and sonicated. Then, the cell lysates were incubated with NF-κB p65 (CST, #8242) or immunoglobulin G (IgG). Finally, the GLS2 DNA was detected using RT-PCR.

Quantitative real-time PCR

CD4+ T cells were collected and lysed in TRIzol Plus (Takara) to extract total RNA, and 500 ng total RNA was reverse transcribed into cDNA using PrimeScript RT Master Mix (Takara). The cDNA was further mixed with SYBR Green PCR Master Mix and specific primers (Takara) to detect relative mRNA expression of specific genes.

The culture medium of CD4+ T cells was collected and centrifuged, and the supernatants were transferred into new tubes for enzyme-linked immunosorbent assay. Plasma levels of cytokines IFN-γ, IL-17A and TNF-α were detected using enzyme-linked immunosorbent assay (ELISA) assay kits (eBioscience) following the instructor’s guidelines.

The siRNA was used to knock down genes. Briefly, siRNA targeting dectin-1 and dectin-3 were mixed with jetPRIME (Polyplus) for 10 min at room temperature and transferred into the CD4+ culture medium for 48 h. Cytokine expression levels were then measured by ELISA.

OCR analysis

The OCR analysis was detected using Seahorse XFe96 extracellular flux analyzer, and the experiments were conducted as previously reported (Hu et al., 2021). Briefly, human and mice CD4+ T cells were seeded into the seahorse 96-well culture plate (in triplicate) for 16 h. Before measurement, the cells were washed twice with a seahorse assay medium. Oligomycin, FCCP, and rotenone/antimycin A were added in sequence. The data were normalized with the total protein of the cells.

Statistical analysis

Data were presented as means ± SD. Data were collected from at least three independent experiments. No statistical method was used to pre-determine sample size. Two-tailed unpaired student t-test was used compare two groups, while one-way analysis of variance (ANOVA) was used to compare multiple groups. Statistical significance was set based on the P value as follows: n.s., P > 0.05; *P < 0.05, **P < 0.01, and ***P < 0.001

Given that gut microbiota play an important role in the pathogenesis of IBD and fungi are an integral part of microbial environments, we investigated the changes in the amount of fungi in IBD. As expected, we found that the amount of fungi derived from feces of CD and UC patients was substantially more than that from healthy controls (Figure 1A). In addition, considering the status of IBD, we investigated whether the amount of fungi was associated with disease activities. We collected the stable and inflamed status fungi of CD and UC patients. Consistently, the amount of fungi in the activated status dramatically increased compared to the stable status (Figures 1B, C). To further confirmed this, we established experimental colitis models in wild-type (WT) C57BL/6 mice using TNBS and DSS and generated IL-10 KO mice. In line with the above results, the amount of fungi in the inflamed DSS-treated, TNBS-treated, and IL10 KO mice was substantially higher than in the initial state (Figures 1D–F). To further corroborate our preceding findings, in situ hybridization was performed to detect the fungi in the human and mice inflamed colon mucosa. Immunofluorescence staining results showed stronger fluorescence intensities in human and mice inflamed colon mucosa than their control counterparts (Figure 1G). Taken together, these findings indicate that fungi were increased in the activated status of IBD and contributed to the development of IBD.

To investigate whether fungi dysbiosis affects the development of IBD, we depleted fungi in the TNBS-derived colitis mouse model. The IBD mice were injected intraperitoneally with 10mg/kg of the anti-fungi drug terbinafine (TB) every three days. As expected, the abundance of fungi dramatically decreased upon TB treatment (Figure 2A). The body weight change curve showed that the weight of both groups of mice decreased upon TNBS administration; however, the terbinafine treated group exhibited substantially lesser weight loss from days 4–9, with about a 12% weight difference at the end time point compared to the control group (Figure 2B). The disease activity index (DAI) was measured during the development of colitis. We found the control mice exhibited irregular shape stools and more severe bleeding than terbinafine-treated mice, resulting in a 45% difference in DAI between the two groups (Figure 2C). Furthermore, terbinafine treatment alleviated colon shortening due to the progression of colitis (Figure 2D). The length of the colon derived from the control group was about one centimeter shorter than that of their counterpart (Figure 2E). Next, we treated the germ-free mice with fungi, which promotes the IBD progression. After that, terbinefine was administrated to the fungi transferred germ-free mice. The result showed that terbinefine could reverse the progression of IBD induced by fungi. Collectively, these data suggested terbianfine exerted their function by depleting the fungi in mice but not the alteration of gut bacterial community after antifungal therapy (Figure 2F). Additionally, colon tissues were collected, fixed, and stained with hematoxylin and eosin (H&E). Histopathological results showed that the colon sections from the control group of mice exhibited more severe ulceration and erosion, inflammatory cell infiltration, and mucosa thickening compared with the terbinafine-treated group, resulting in higher disease scores (Figure 2G). In summary, these results indicate that fungi depletion could prevent the development of inflammation in the colon.

Next, we aimed to elucidate the mechanisms through which fungi promote the pathogenesis of IBD. Given that CD4+ T cells are greatly involved in the initiation and development of IBD, we firstly detected whether the function of fungi in IBD depending on CD4+ T cells. The Cd4 knock out mice that induced IBD by DSS were treated with terbinafine, the results showed that fungi depletion mediated by terbinfine could not alleviate the IBD progression, indicating that the effect of fungi relying on the CD4+ cells (Supplementary Figure 1). Further, we evaluated the effects of CD4+ T cell differentiation under fungi treatment. The CD4+ T cells in the peripheral blood were collected from active IBD patients and healthy controls following different doses of fungi infection. To uncover the response of CD4+ T cells, the levels of mRNA expression of typical inflammation cytokines IFN-γ, IL-17A, and TNF-α were detected in both control and fungi-infected CD4+ T cells. Intriguingly, the RT-PCR results showed that fungi treatment induced IFN-γ, IL-17A, and TNF-α expression in CD4+ T cells isolated from healthy controls (Figures 3A–C). In line with this, the expression of all three cytokines was significantly up-regulated in CD4+ T cells isolated from peripheral blood of activated CD and UC patients compared with their control counterparts. Next, we measured three transcription factors, including T-bet, RORC, and Foxp3, that determine the direction of CD4+ T cells differentiation to Th1, Th17, and Treg, respectively. The results showed that T-bet and RORC expression sharply increased upon fungi infection (Figures 3D–F). Consistently, the up-regulation of T-bet and RORC was much higher in IBD CD4+ T cells than in healthy control subjects.

However, Foxp3 expression remained unchanged in CD4+ T cells derived from healthy donors and IBD patients. These results indicate that fungi infection promotes CD4+ T cell differentiation into Th1 and Th17 but not T reg cell. To further confirm this, we measured the protein level of IFN-γ, IL-17A, and TNF-α under fungi infection in both healthy donors and IBD patients. The supernatant of CD4+ T cells culture medium was collected, followed by ELISA. The results indicated that protein expression levels of the inflammation cytokines IFN-γ, IL-17A, and TNF-α were significantly up-regulated in culture supernatants of fungi-infected CD4+ T cells compared with control CD4+ T cells (Figures 3G–I). Finally, colon specimens derived from IBD patients treated with terbinafine or those not treated with terbinafine were collected following IHC staining assays to detect IFN-γ, IL-17A, and TNF-α. As expected, IFN-γ, IL-17A, and TNF-α expression in the fungi-depleted colon sections were markedly reduced compared to those not treated with terbinafine (Figure 3J). These results suggest that fungi infection in CD4+ T cells enhances CD4+ T cell pro-inflammatory response in IBD patients.

Further, we aim to investigate the mechanism through which fungi promote CD4+ T cell pro-inflammatory immune responses in IBD patients. T cells enhance glucose and glutamine metabolism to elevate the level of mitochondrial OXPHOS for cytokine production. To investigate whether OXPHOS is involved in CD4+ T cell response under fungi infection, we detected OXPHOS levels in both fungi-infected and control humans as well as murine CD4 T cells by measuring OCR levels using seahorse analysis. As expected, fungi infection significantly induced the up-regulation of OCR levels in human CD4 T cells. Consistently, this enhanced OXPHOS was also observed in murine CD4 T cells, showing that OXPHOS increased by a maximum of about 40% (Figures 4A, B). Moreover, we found that suppressing OXPHOS with FCCP could block cytokine production induced by fungi infection.

Previous studies reported that glutamine metabolism plays a critical role in regulating the differentiation of T cells into different T helper subtypes. To assess whether glucose or glutamine mediate enhanced OXPHOS, we measured the production of inflammatory cytokines by CD4+ T cells under glucose or glutamine-deficiency culture conditions. The results showed that the glucose-free culture medium had almost no effect on IFN-γ production. However, the glutamine-free culture medium could reverse cytokine production by CD4+ T cells induced by fungi infection(Figure 4C). Moreover, IFN-γ production was highest in both glutamine and glucose-supplemented culture media. To further confirm that glutaminolysis in the CD4+ T cells determines cytokines production, we further blocked glutaminolysis in CD4+ T cells using glutaminolysis inhibitors, 6-Diazo-5-oxo-L-norleucine (DON) or L-Glutamic acid γ-monohydroxamate (GAH) (Figure 4D). The two inhibitors reversed the up-regulated cytokines production. Finally, we detected the glutamine metabolism enzymes GLS, GLS2, and GLUL mRNA expression under fungi infection in murine CD4+ T cells and CD4+ T cells obtained from CD and UC patients. The Q-PCR results showed that the expression of GLS2, which converts glutamine to glutamate, was markedly enhanced in activated CD4+ T cells (Figures 4E, F). In summary, fungi infection enhanced glutaminolysis in CD4+ T cells to support the production of cytokines.

In addition, we investigated the mechanism through which fungi enhance glutaminolysis in CD4+ T cells. Firstly, to investigate receptors that mediate fungi sensing in CD4+ T cells, we evaluated the expression of common receptors for fungi in the CD4+ T cells, including Mincle, Dectin-1, Dectin-2, and Dectin-3 ligands. The real-time PCR results indicate that Mincle and Dectin-2 were expressed on CD4+ T cells. To assess which receptor exerts the function, siRNA was introduced to knock down Mincle and Dectin-2 in CD4 T cells. The results showed that suppressed Mincle, but not Dectin-2, expression reduced glutaminolysis in CD4+ T cells. To further confirm this, trehalose-6,6-dimycolate (TDM) and α-mannans, and Mincle and Dectin-2/3 ligands were applied to treat the CD4+ T cells, respectively. We found that stimulation with α-mannans failed to induce glutaminolysis (Figures 5A–D). However, TDM stimulation significantly induced glutaminolysis (Figures 5A,B), and the deletion of Mincle in BMDCs completely blocked TDM-induced glutaminolysis (Figures 5E, F). Moreover, treating BMDCs or human peripheral blood mononuclear cells (PBMCs) with Syk inhibitor significantly impaired glutaminolysis in CD4+ T cells.

Previous studies have shown that Syk signaling could activate the NF-κB pathway. To reveal whether NF-κB is involved in this process, CD4+ T cells derived from human and mice IBD colon were treated with NF-κB inhibitors (Figures 5G–I). Consistently, NF-κB blockage significantly suppressed the glutaminolysis and glutamine metabolism genes expression. These findings suggested that fungi enhance glutaminolysis in CD4+ T cells by activating the dectin-1-Syk- NF-κB signaling pathway.