UBE2N Is Essential for Maintenance of Skin Homeostasis and Suppression of Inflammation

Abstract

UBE2N, a Lys63 ubiquitin-conjugating enzyme, plays critical roles in embryogenesis and immune system development and function. However, its roles in adult epithelial tissue homeostasis and pathogenesis are unclear. We generated conditional mouse models that deleted Ube2n in skin cells in a temporally and spatially controlled manner. We found that Ube2n knockout in the adult skin keratinocytes induced a range of inflammatory skin defects characteristic of psoriatic and actinic keratosis. These included inflammation, epidermal and dermal thickening, parakeratosis, and increased immune cell infiltration as well as signs of edema and blistering. Single-cell transcriptomic analyses and RT-qPCR showed that Ube2n-knockout keratinocytes expressed elevated myeloid cell chemoattractants such as Cxcl1 and Cxcl2 and decreased the homeostatic T lymphocyte chemoattractant Ccl27a. Consistently, the infiltrating immune cells were predominantly myeloid-derived cells, including neutrophils and M1-like macrophages, which expressed high levels of inflammatory cytokines such as Il1β and Il24. Pharmacological blockade of the IL-1 receptor associated kinases (IRAK1/4) alleviated inflammation, epidermal and dermal thickening, and immune infiltration of the Ube2n-mutant skin. Together, these findings highlight a key role of keratinocyte UBE2N in maintenance of epidermal homeostasis and skin immunity and identify IRAK1/4 as potential therapeutic target for inflammatory skin disorders.

Keywords: Cutaneous biology; IL-1 signaling; Inflammation; Skin immunology; UBE2N.

Figure 1.. Topically induced deletion of Ube2n in the skin leads to inflammation.

(a) Experimental scheme of induction of Ube2n deletion in 6-10 weeks old Rosa26^CreER.Ube2n^fl/fl mice. 40 μL of 5 mg/mL of 4-hydroxytamoxifen (4-OHT) was applied topically every 48 hours for four times. (b) Appearance of mouse back skin 15 days post-induction. (c) H&E histological analysis of the mouse skin sections. Scale bar: 100μm. (n = 3–7) (d) UMAP of skin cells identifying 7 cell populations. Single-cell RNA-seq analysis of control and Ube2n KO skin samples (n = 2–3 per group pooled). (e) split UMAPs of control and Ube2n KO skin samples. (f) Dot plot of cell identification with key genes. (g) Percent distribution of the control and the Ube2n KO skin cells. (h) Violin plot and box plot of Ube2n.

Figure 2.. Ube2n KO epidermal cells display abnormal growth and differentiation and are highly inflammatory.

(a) UMAP of subclustering epidermal cells identifying 8 keratinocyte subgroups. (b) split UMAPs of epithelial cells of control and Ube2n KO skin samples. (c) Dot plot depicting expressions of key identifying genes. (d) Percent distribution of the control and the Ube2n KO epidermal cells. (e) Violin and bar plot of Ube2n expression between the control and the Ube2n KO group per subgroup. (f) Volcano pot of upregulated (red) and downregulated (blue) genes in the Ube2n KO epidermal cells. (adjusted p-value < 0.05) (g) Gene ontology analysis of top 100 differentially increased genes using Biological Process 2023 reference. (h) Relative mRNA expressions of Cxcl1, Cxcl2, Il1β, Il24, and Ccl27a in the mutant epidermis compared to the vehicle ethanol (EtOH) control counterparts. Fold change is log transformed. (n = 2–5) (i) Relative mRNA expressions of Krt15 and Krt6a in the mutant epidermis compared to the EtOH control. 18S is used for internal control. (n = 2–9) * p< 0.05; ** p< 0.01; *** p < 0.001; **** p < 0.0001 (Unpaired Student’s t-test).

Figure 3.. Subclustering of immune cell populations reveals myeloid cell enrichment in the Ube2n KO skin.

(a) UMAP of immune cells identifying 7 subgroups. (b) split UMAPs of immune cells of control and Ube2n KO skin samples. (c) Dot plot depicting expressions of key identifying genes. (d) Percent distribution of different cell types of the control and the Ube2n KO skin. (e) Gating strategy of flow cytometry analysis of Rosa26^CreER.Ube2n^fl/fl back skin treated with 4-OHT or EtOH vehicle control. Graph represents percent of CD45⁺ cells over live cells ± standard deviation (SD). (n = 2–3) (f) Gating strategy of CD45⁺/CD11b⁺ and CD45⁺/CD11b⁺/Ly6C⁺ myeloid cells. Graphs represents percent of CD11b⁺ and Ly6C⁺ over live CD45⁺ and CD45⁺/CD11b⁺ cells, respectively ± SD. * p< 0.05; ** p< 0.01 (Unpaired Student’s t-test). (n = 2–3)

Figure 4.. Epidermal keratinocyte-specific deletion of Ube2n is sufficient to induce inflammation.

(a) Experimental scheme of induction of Ube2n deletion of the Krt5^CreER.Ube2n^fl/fl mice. (b) Relative mRNA level of Ube2n in the Krt5^CreER.Ube2n^fl/fl epidermis (KO) upon 4-OHT treatment compared to the Ube2n^fl/fl control (WT). (n = 6) (c) Western blotting of protein lysates isolated from the WT and the KO epidermis. (n = 3) (d) Quantification of western blot band intensity of UBE2N relative to β-Actin. (e) Appearance of adult mice 25 days post-induction, (f) H&E histological analysis of the mouse back skin sections. Scale bar: 100μm. (n = 3–7) (g) Relative mRNA expressions of Krt15 and Krt6a in the KO epidermis compared to the WT control mice. (h–i) Representative images of immunostaining of (h) KRT15 (red), KRT6A (red), ITGA6 (green) (n = 3) and (i) CD45 (red) and Hoechst (nuclei, blue) (n = 2) in the Ube2n^fl/fl and the KO skin. Scale bars: 100μm and 50μm. (j) Relative mRNA expressions of Cxcl1, Cxcl2, Il1β, Il24, and Ccl27a. 18S is used for internal control. * p< 0.05; ** p< 0.01; *** p < 0.001; **** p < 0.0001 (Unpaired Student’s t-test). (n = 6)

Figure 5.. Pharmacological inhibition of IRAK1/4 alleviates skin inflammation caused by Ube2n deletion.

(a) IRAK1 protein level of the Ube2n^fl/fl and the Krt5^CreER.Ube2n^fl/fl mutant epidermis. β-Actin was used as internal control. (n = 1) (b) Experimental scheme. The control and the inhibitor R509 formulated chows were provided 8 days prior to the start of 4-OHT treatment. (c) Appearance of the control and the drug-treated skin 18 days post 4-OHT treatment. (n = 7–10) (d) H&E staining of the mouse back skin sections. Scale bar: 100μm. (n = 2–4) (e–f) Representative images of immunostaining of (e) p-IRAK1 (Thr 209) (red), ITGA6 (green) (n = 2) and (f) KRT6A (red), CD45 (green), Hoechst (nuclei, blue). (n = 2) Scale bars: 100μm and 50μm.