Inflammatory mediator TAK1 regulates hair follicle morphogenesis and anagen induction shown by using keratinocyte-specific TAK1-deficient mice

Abstract

Transforming growth factor-beta-activated kinase 1 (TAK1) is a member of the NF-kappaB pathway and regulates inflammatory responses. We previously showed that TAK1 also regulates keratinocyte growth, differentiation, and apoptosis. However, it is unknown whether TAK1 has any role in epithelial-mesenchymal interactions. To examine this possibility, we studied the role of TAK1 in mouse hair follicle development and cycling as an instructive model system. By comparing keratinocyte-specific TAK1-deficient mice (Map3k7(fl/fl)K5-Cre) with control mice, we found that the number of hair germs (hair follicles precursors) in Map3k7(fl/fl)K5-Cre mice was significantly reduced at E15.5, and that subsequent hair follicle morphogenesis was retarded. Next, we analyzed the role of TAK1 in the cyclic remodeling in follicles by analyzing hair cycle progression in mice with a tamoxifen-inducible keratinocyte-specific TAK1 deficiency (Map3k7(fl/fl)K14-Cre-ER(T2)). After active hair growth (anagen) was induced by depilation, TAK1 was deleted by topical tamoxifen application. This resulted in significantly retarded anagen development in TAK1-deficient mice. Deletion of TAK1 in hair follicles that were already in anagen induced premature, apoptosis-driven hair follicle regression, along with hair follicle damage. These studies provide the first evidence that the inflammatory mediator TAK1 regulates hair follicle induction and morphogenesis, and is required for anagen induction and anagen maintenance.

Figure 1. Impaired hair follicle morphogenesis in keratinocyte-specific TAK1-deficient mice.

(A) Histological analysis of hair follicle development from E14.5 to P6. Map3k7 ^fl/flK5-Cre were keratinocyte-specific TAK1-deficient mice . Map3k7 ^fl/fl mice were used as controls. Arrows indicate hair germs or hair pegs. Scale bar, 100 µm. (B) The hair morphogenesis stage of each hair follicle in each mouse group was evaluated as described previously . Then, the rate (%) of a certain hair morphogenesis stage in the total hair follicles (left panel) and the median score in each mouse group (right panel) were determined. The score of Map3k7 ^fl/flK5-Cre mice was compared with Map3k7 ^fl/fl mice. Statistical significance was determined using a Mann-Whitney's U-test. *P<0.01.

Figure 2. Southern blot analysis.

Genomic DNA prepared from the ear skin of mice treated with tamoxifen solution (15 µL/ear) or ethanol for 5 consecutive days was digested with XbaI and EcoRI. Southern blot analysis for the deletion of the floxed Tak1 allele was performed as described previously . Cre expression resulted in excision of the floxed allele (flox) and generated the deleted allele (Δ) of Map3k7.

Figure 3. Keratinocyte-specific TAK1 deletion results in hair loss in adolescent mice.

Tamoxifen was topically applied to the dorsal skin of the Map3k7 ^fl/flK14-Cre-ER^T2 mice for 5 consecutive days to delete TAK1. The clinical appearance of the mice 4 weeks after the application is shown. As controls, Map3k7 ^fl/fl mice or Map3k7 ^fl/flK14-Cre-ER^T2 mice were treated with tamoxifen or ethanol, respectively.

Figure 4. Keratinocyte-specific TAK1 deletion delays hair cycle progression in adolescent mice.

(A) Schedule of tamoxifen application. The hair cycle was synchronized to anagen phase by wax depilation, and tamoxifen was topically applied to the dorsal skin of Map3k7 ^fl/flK14-Cre-ER^T2mice to delete TAK1. As controls, Map3k7 ^fl/fl mice or Map3k7 ^fl/flK14-Cre-ER^T2 mice were treated with tamoxifen or ethanol, respectively. (B) Clinical appearance at the indicated time point after the depilation. At 2 weeks, hair shaft formation was noted in the control mice, while hair shaft growth was not seen in the tamoxifen-treated Map3k7 ^fl/flK14-Cre-ER^T2 mice. (C) Histological analysis at the indicated time point after the depilation. Anagen progression was severely delayed in TAK1-deleted mice at 1–3 weeks. Scale bar, 100 µm.

Figure 5. Quantitative hair cycle analysis.

Hair cycle stage of each hair follicle in each mouse group in Fig. 4 was evaluated according to our previous report . Then, the rate (%) of a certain hair cycle stage in the total hair follicles (left panel) and the median score in each mouse group (right panel) were determined. The score of tamoxifen-treated Map3k7 ^fl/flK14-Cre-ER^T2 mice was compared with tamoxifen-treated Map3k7 ^fl/fl mice or ethanol-treated Map3k7 ^fl/flK14-Cre-ER^T2 mice. Statistical significance was determined using a Mann-Whitney's U-test. *P<0.01.

Figure 6. Keratinocyte-specific TAK1 deletion causes a transition from anagen to dystrophic catagen in adolescent mice.

(A) Schedule of tamoxifen application. The hair cycle was synchronized to anagen phase by wax depilation. At 7 days after depilation, tamoxifen was applied for 5 days. (B) Clinical appearance of the mice 2 weeks after depilation. (C) Histological analysis of the mice 2 weeks after depilation. (D) Higher magnification of the tamoxifen-treated Map3k7 ^fl/flK14-Cre-ER^T2 mice in (C). Scale bar, 100 µm. (E) Dystrophic catagen was defined according to a previous report . Then, the rate (%) of a certain hair follicle stage per total hair follicles was calculated. Quantitative analyses confirmed that most of the hair follicles in tamoxifen-treated Map3k7 ^fl/flK14-Cre-ER^T2 mice were in late dystrophic catagen, while those of controls were in anagen.

Figure 7. Model for the signaling pathways involved in hair follicle development.

In the development of primary hair placodes, TAK1 is involved in the Edar/NF-κB pathway. In secondary hair placodes, NF-κB (TAK1)-independent pathways regulate hair placode development.