Lack of TYK2 signaling enhances host resistance to Candida albicans skin infection

Abstract

Candida albicans is the most common human fungal pathogen, causing diseases ranging from local to life-threating systemic infections. Tyrosine kinase 2 (TYK2), a crucial mediator in several cytokine signaling pathways, has been associated with protective functions in various microbial infections. However, its specific contribution in the immune response to fungal infections has remained elusive. In this study, we show that mice lacking TYK2 or its enzymatic activity exhibit enhanced resistance to C. albicans skin infections, limiting fungal spread and accelerating wound healing. Impaired TYK2-signaling prompted the formation of a distinctive layer of necrotic neutrophils around the fungal pathogens. Transcriptomic analysis revealed TYK2's pivotal role in regulating interferon-inducible genes in neutrophils, thereby impacting their antifungal capacity during infection. Furthermore, we show that TYK2-dependent interferon-gamma (IFNγ) production contributes to fungal dissemination from the skin to the kidneys. Our study uncovers a hitherto unrecognized detrimental role of TYK2 in cutaneous C. albicans infections.

Fig. 1. Absence of enzymatically active TYK2 results in less severe candidiasis.

a WT, Tyk2^-/-, and Tyk2^K923E mice were intradermally (i.d) injected in the back skin with 1 × 10⁸ CFU of C. albicans or PBS (control). Fungal load in the skin (b) and in the kidneys (c) was measured on days 2 and 4 post-infection (p.i). For each time-point, pooled data from 2 independent experiments are shown, and median values are given; n: biological replicates; the dotted line indicates the assay detection limit (b, c). Skin day 2: n = 10 (WT), n = 12 (Tyk2^-/-) and n = 9 (Tyk2^K923E); skin day 4: n = 11 (WT, Tyk2^-/-) and n = 12 (Tyk2^K923E) (b); kidney day 2: n = 10 (WT, Tyk2^K923E) and n = 12 (Tyk2^-/-); kidney day 4: n = 11/genotype (c). d Blood cell composition on day 4 p.i was determined with a Vet ABC analyzer. The percentage of granulocytes and lymphocytes out of total white blood cells (WBC) is shown. Pooled data from 2 independent experiments are shown. Mean values ± SEM are given; PBS: n = 7 (WT) and n = 6 (Tyk2^-/-, Tyk2^K923E); C. albicans: n = 12 (WT, Tyk2^K923E) and n = 11 (Tyk2^-/-); n: biological replicates. e Representative pictures of the infected skin on day 2 p.i. A Gomori Methenamine-Silver (GMS) staining of the skin sections is shown. Scale bar: 200 μm (top), 50 μm (bottom). Data are representative for sections from 7 mice per genotype. f WT, Tyk2^-/- and Tyk2^K923E mice were infected as described above and wounds were visually monitored for 25 days. g The percentage of mice with wounds is shown. Pooled data from 2 independent experiments are shown. Error bars indicate the standard error (SE), centers for the error bars indicate mean values; n = 13 (WT), n = 11 (Tyk2^-/-) and n = 10 (Tyk2^K923E); n: biological replicates. Statistical analysis was conducted using One-way ANOVA followed by Tukey’s multiple comparison test (b–d) and the Log-rank (Mantel-Cox) test (g). Statistical significance is only given for the comparison between the genotypes (b–d, g). Source data are provided as a Source Data file.

Fig. 2. C. albicans infection induces TYK2-independent infiltration of neutrophils, which form a Ki-67-negative cell layer around the pathogen in TYK2 mutants but not in WT mice.

a Representative IHC pictures from the infected skin of WT, Tyk2^-/-, and Tyk2^K923E mice on day 2 p.i. Sections were stained with an anti-NIMP-R14 antibody (which stains Ly6C⁺ and Ly6G⁺ cells). Scale bar: 200 μm. Data are representative of sections from 3 mice per genotype (PBS) and from 7 mice per genotype (day 2). b–e The percentages of skin-infiltrating neutrophils (gated as CD45⁺CD11b⁺Ly6C⁺Ly6G⁺ cells) (b, c) and monocytes (gated as CD45⁺CD11b⁺Ly6G^-Ly6C^highF4/80^- cells) (d, e) out of CD45⁺CD11b⁺ cells were determined by flow cytometry analysis on days 1, 2 and 4 p.i. Representative contour plots (b, d) and pooled data from 2 independent experiments for each time-point (c, e) are shown. PBS: n = 7/genotype (c, e), C. albicans day 1: n = 8/genotype (c) and n = 7/genotype (e); C. albicans day 2: n = 8 (WT) and n = 7 (Tyk2^-/-, Tyk2^K923E) (c, e); C. albicans day 4: n = 9 (WT, Tyk2^K923E), n = 8 (Tyk2^-/-) (c) and n = 9 (WT, Tyk2^K923E) and n = 7 (Tyk2^-/-) (e); Mean values ± SEM are given; n: biological replicates. Statistical analysis was conducted using One-way ANOVA followed by Tukey’s multiple comparison test; statistical significance is only given for the comparison between the genotypes (c, e). Representative IHC pictures from the infected skin on day 1 (f) and 2 (g) p.i. Sections were stained with an anti-Ki-67 antibody. Scale bar: 200 μm (top), 50 μm (bottom). Data are representative of sections from 5 mice per genotype (f) and from 7 mice per genotype (g). Source data are provided as a Source Data file.

Fig. 3. Depletion of neutrophils results in increased fungal dissemination to the kidneys.

a WT and Tyk2^-/- mice were intraperitoneally (i.p) injected with an antibody against Ly6G (anti-Ly6G) or PBS (control) one day before being intradermally (i.d) injected in the back skin with 1×10⁸ CFU of C. albicans. The percentages of neutrophils (gated as CD45⁺Ly6C⁺Ly6G⁺ cells) out of CD45⁺ cells in the blood (b) and the skin (c) were determined by flow cytometry analysis on day 2 p.i. Representative contour plots and pooled data from 2 independent experiments are shown and mean values ± SEM are given; PBS: n = 11 (WT), n = 10 (Tyk2^-/-); anti-Ly6G: n = 10 (WT) and n = 6 (Tyk2^-/-); n: biological replicates (b, c). d Representative IHC pictures from the infected skin on day 2 p.i. Sections were stained with an anti-NIMP-R14 antibody (which stains Ly6C⁺ and Ly6G⁺ cells). PBS: Data are representative for sections from 8 (WT) and 11 (Tyk2^-/-) mice. Anti-Ly6G: Data are representative for sections from 9 (WT) and 6 (Tyk2^-/-) mice. Scale bar: 200 μm (top), 50 μm (bottom). e Representative pictures of the infected skin on day 2 p.i. A GMS staining of the skin sections is shown. PBS: Data are representative for sections from 8 (WT) and 11 (Tyk2^-/-) mice. Anti-Ly6G: Data are representative for sections from 9 (WT) and 6 (Tyk2^-/-) mice. Scale bar: 50 μm. f Fungal load in the kidneys of infected mice previously treated with anti-Ly6G antibody or PBS was measured on day 2 p.i. Pooled data from 2 independent experiments are shown and median values are given; PBS: n = 11/genotype; anti-Ly6G: n = 11 (WT) and n = 6 (Tyk2^-/-); ns: not significant; n: biological replicates. The dotted line indicates the assay detection limit. Statistical analysis was conducted using One-way ANOVA followed by Tukey’s multiple comparison test and statistical significance is only given for the comparison between the genotypes (b, c, f). Source data are provided as a Source Data file.

Fig. 4. Absence of TYK2 or its kinase activity alters the transcriptome of skin-infiltrating myeloid cells.

a WT, Tyk2^-/- and Tyk2^K923E mice were infected as described in the legend to Fig. 1. RNA from skin-sorted CD45⁺CD3^-CD19^-NK1.1^- myeloid cells was isolated for subsequent RNA-sequencing. b Volcano plots showing the differentially expressed genes between Tyk2^-/- and WT and Tyk2^K923E and WT cells. Genes were identified from DESeq2-normalized read counts of genes with a threshold of padj < 0.05 between Tyk2^-/- and WT and Tyk2^K923E and WT cells. Data are from one experiment (n = 3 mice per genotype). c Heatmap showing the differentially expressed genes between WT and TYK2-mutant cells with an absolute log2-fold change > 1 (padj <0.05). IFN-stimulated genes (ISGs) are highlighted in orange. d Pathway analysis of deregulated pathways between WT and ^Tyk2-/- cells was performed using Reactome. The top 10 upregulated and downregulated pathways are shown. e mRNA levels of Ifng in the infected skin on day 4 p.i were measured by RT-qPCR. Data were normalized to the housekeeping gene Ube2d2. Pooled data from 2 independent experiments are shown and mean values ± SEM are given; PBS: n = 2/genotype; C. albicans: n = 12 (WT), n = 7 (Tyk2^-/-) and n = 9 (Tyk2^K923E); n.d not detectable; n: biological replicates. f IFNγ in the skin on day 2 p.i was measured using a Luminex assay. Pooled data from 2 independent experiments are shown. Mean values ± SEM are given; PBS: n = 2/genotype; C. albicans: n = 8 (WT) and n = 9 (Tyk2^-/-, Tyk2^K923E); n.d, not detectable; n: biological replicates. Statistical analysis was conducted using One-way ANOVA followed by Tukey’s multiple comparison test and statistical significance is only given for the comparison between the genotypes (e, f). Source data are provided as a Source Data file.

Fig. 5. Ifngr1^-/- mice are less sensitive to cutaneous candidiasis than WT mice and grossly phenocopy TYK2-mutants.

a WT and Ifngr1^-/- were infected as described in the legend to Fig. 1. Fungal load in the skin (b) and in the kidneys (c) was measured on day 4 p.i. Pooled data from 2 independent experiments are shown. Mean values ± SEM are given; skin: n = 12 (WT) and n = 13 (Ifngr1^-/-); kidneys: n = 13/genotype; n: biological replicates. The dotted line indicates the assay detection limit (b, c). d Blood cell composition on day 4 p.i. was determined with a VetABC analyzer. The percentage of granulocytes and lymphocytes out of total white blood cells (WBC) and the total numbers of these populations in the blood are shown. Pooled data from 2 independent experiments are shown. Mean values ± SEM are given; PBS: n = 4/genotype, C. albicans: n = 13/genotype; n: biological replicates. Representative pictures of the infected skin on day 4 p.i. Sections were stained with GMS (e) or an anti-Ki-67 antibody (f). Scale bar: 200 μm (left), 50 μm (right). Data are representative of sections from 8 (WT) and 7 (Ifngr1^-/-) mice. g The percentages of skin-infiltrating neutrophils (gated as CD45⁺CD11b⁺Ly6C⁺Ly6G⁺ cells), monocytes (gated as CD45⁺CD11b⁺Ly6G^-Ly6C^highF4/80^- cells) and macrophages (gated as CD45⁺CD11b⁺Ly6G^-F4/80⁺ cells) out of CD45⁺CD11b⁺ cells were determined by flow cytometry analysis on day 1 and 4 p.i. Representative contour plots and pooled data from 2 independent experiments for each time-point are shown. Mean values ± SEM are given; PBS: n = 10 (WT) and n = 14 (Ifngr1^-/-); day 1: n = 11 (WT) and n = 13 (Ifngr1^-/-); day 4: n = 8 (WT) and n = 10 (Ifngr1^-/-) for neutrophils and macrophages and n = 9 (WT) and n = 11 (Ifngr1^-/-) for monocytes; n: biological replicates. Statistical analysis was conducted using a two-tailed Mann Whitney test (b, c) and One-way ANOVA followed by Tukey’s multiple comparison test (d, g). Statistical significance is only given for the comparison between the genotypes (b–d, g). Source data are provided as a Source Data file.

Fig. 6. RAG2 deficiency leads to diminished C. albicans tissue dissemination and impaired IFNγ production.

a WT and Rag2^-/- mice were infected as described in the legend to Fig. 1. b mRNA levels of Ifng in the skin of WT and Rag2^-/- mice on day 4 p.i were measured by RT-qPCR. Data were normalized to the housekeeping gene Ube2d2 and fold changes were calculated relative to the uninfected WT controls. Pooled data from 2 independent experiments are shown. Mean values ± SEM are given; PBS: n = 2 (WT) and n = 3 (Rag2^-/-); C. albicans: n = 7/genotype; n: biological replicates. Statistical significance is only given for the comparison between the genotypes). Fungal load in the skin (c) and in the kidneys (d) was determined on day 4 p.i by CFU assay. Pooled data from 2 independent experiments are shown. Median values are given; n = 11 (WT) and n = 12 (Rag2^-/-); n: biological replicates. The dotted line indicates the assay detection limit. e Representative pictures of the infected skin on day 4 p.i. A GMS staining of the skin sections is shown. Data are representative of sections from 4 mice per genotype. Scale bar: 200 μm (left), 50 μm (right). f The percentages of skin-infiltrating neutrophils (gated as CD45⁺CD11b⁺Ly6C⁺Ly6G⁺ cells), monocytes (gated as CD45⁺CD11b⁺Ly6G^-Ly6C^highF4/80^- cells) and macrophages (gated as CD45⁺CD11b⁺Ly6G^-F4/80⁺ cells) out of CD45⁺CD11b⁺ cells were determined by flow cytometry analysis on day 4 p.i. Representative contour plots and pooled data from 2 independent experiments are shown. Mean values ± SEM are given; PBS: n = 6 (WT) and n = 5 (Rag2^-/-); C. albicans: n = 9 (WT) and n = 12 (Rag2^-/-) for neutrophils and macrophages; n = 9 (WT) and n = 11 (Rag2^-/-) for monocytes; n: biological replicates. Statistical analysis was conducted using a two-tailed Mann Whitney test (c, d) and One-way ANOVA followed by Tukey’s multiple comparison test (b, f). Statistical significance is only given for the comparison between the genotypes (b–d, f). Source data are provided as a Source Data file.

Fig. 7. Deletion of TYK2 in αβ T cells and NKT cells is not sufficient to protect mice against candidiasis.

a Tyk2^ΔT and Tyk2^fl/fl mice were infected as described in the legend to Fig. 1. Fungal load in the skin (b) and in the kidneys (c) was measured on day 4 p.i. Pooled data from 2 independent experiments are shown. Median values are given; skin: n = 7 (Tyk2^fl/fl) and n = 11 (Tyk2^ΔT); kidneys: n = 8 (Tyk2^fl/fl) and n = 11 (Tyk2^ΔT); n: biological replicates. The dotted line indicates the assay detection limit. d Blood cell composition on day 4 p.i was determined with a Vet ABC analyzer. The percentage of granulocytes and lymphocytes out of total white blood cells (WBC) and total numbers of these populations in the blood are shown. Pooled data from 2 independent experiments are shown. PBS: n = 3 (Tyk2^fl/fl) and n = 5 (Tyk2^ΔT); C. albicans: n = 11 (Tyk2^fl/fl) and n = 16 (Tyk2^ΔT); n: biological replicates. e Ifng mRNA levels in the skin on day 4 p.i were measured by RT-qPCR. Pooled data from 2 independent experiments are shown. PBS: n = 2/genotype; C. albicans: n = 10 (Tyk2^fl/fl) and n = 9 (Tyk2^ΔT); n: biological replicates. f IFNγ protein levels on day 4 p.i were measured using a Luminex assay. Pooled data from 2 independent experiments are shown. PBS: n = 2/genotype; C. albicans: n = 4/genotype; n.d, not detectable; n: biological replicates. Representative pictures of the infected skin on day 4 p.i. Sections were stained with anti-NIMP-R14 (g) or anti-Ki-67 (h) antibodies or with GMS staining (i). Scale bar: 200 μm (left), 50 μm (right). Data are representative of sections from 4 (Tyk2^fl/fl) and 3 (Tyk2^ΔT) mice (g–i). j The percentages of skin-infiltrating neutrophils (gated as CD45⁺CD11b⁺Ly6C⁺Ly6G⁺ cells), monocytes (gated as CD45⁺CD11b⁺Ly6G^-Ly6C^highF4/80^- cells) and macrophages (gated as CD45⁺CD11b⁺Ly6G^-F4/80⁺ cells) out of CD45⁺CD11b⁺ cells were determined by flow cytometry analysis on day 1 and 4 p.i. Representative contour plots and pooled data from 2 independent experiments for each time-point are shown. PBS - n = 12 (Tyk2^fl/fl), n = 9 (Tyk2^ΔT) for Neutrophils; n = 11 (Tyk2^fl/fl), n = 7 (Tyk2^ΔT) for Monocytes; n = 5 (Tyk2^fl/fl), n = 6 (Tyk2^ΔT) for Macrophages; C. albicans day 1 - n = 14 (Tyk2^fl/fl), n = 10 (Tyk2^ΔT); C. albicans day 4 - n = 11/genotype; n: biological replicates; Mean values ± SEM are given (d–f, j). Statistical analysis was conducted using a two-tailed Mann Whitney test (b, c) and One-way ANOVA followed by Tukey’s multiple comparison test (d–f). Statistical analysis between genotypes has not been done for PBS controls in (e, f), as the sample size is <3. Statistical significance is only given for the comparison between the genotypes (b–f, j). Source data are provided as a Source Data file.

Fig. 8. γδ T cells produce IFNγ upon C. albicans stimulation.

a The skin of WT and Tyk2^-/- mice was collected, digested and single cell suspensions were prepared. 2 × 10⁶ cells were incubated for 24 h with heat-killed C. albicans (HK C.a), IL-12 or IL-2 and stained for flow cytometry analysis. The percentage of IFNγ⁺ cells out of γδ T cells (b, c) and the total numbers of these cells (d) are shown. Representative contour plots (b) and pooled data from 2 independent experiments are shown (c, d). Mean values ± SEM are given; IL-2: n = 5 (WT) and n = 4 (Tyk2^-/-); IL-12: n = 6 (WT) and n = 7 (Tyk2^-/-); HK C.a: n = 8 (WT) and n = 7 (Tyk2^-/-); n: biological replicates. Statistical analysis was conducted using One-way ANOVA followed by Tukey’s multiple comparison test and statistical significance is only given for the comparison between the genotypes (c, d). Source data are provided as a Source Data file.