In Ta mice it was not expressed during primary guard hair follicle induction stages.
Sweat glands were normally formed in WTDk4TG mice, suggesting their development, like primary guard hair, is Dkk4-independent (Fig. 3A).
In Map3k7fl/flK5-Cre mice, the number of hair germs was significantly reduced at E15.5, indicating that the development of primary guard hair follicles was greatly impaired.
Primary guard hair germs were induced normally in WT and WTDk4TG at E14.5, but not in Ta or TaDk4TG littermates (Fig. 5C).
In mice, "primary" guard hairs, constituting less than 5% of mouse hair on the back skin, overlay and protect the 95% of "secondary" hairs.
Wnt signaling activates Eda and Edar [2], [41], and Wnt inactivation suppresses the EDA pathway in mice, especially during primary guard hair follicle induction [2], [14], [16], [42].
To see if supplementation of Dkk4 in Ta mice was able to restore Ta phenotypes, we further analyzed development of the two major target appendages of Eda, primary guard hair and sweat gland germs, in TaDk4TG and WTDk4TG embryos.
In a recent study, analyses of Eda and EdaR homologue Troy double-mutant mice revealed that, in addition to primary guard hair follicles, awl/auchene hair follicles were defective in these mice [38].
In the absence of NF-κB activity, downstream events, such as maintenance of Wnt signaling and an increase of Wnt10a, Wnt10b, and Dkk4 expression, are impaired [8] and further placode down-growth does not occur in primary guard hair follicles [36].
Impaired development of primary guard hair follicles at E15.5 can be explained by the absence of NF-κB activity, due to TAK1 deficiency, consistent with a model in which TAK1 is involved in the EdaA1/EdaR/NF-κB pathway [35] (see Fig. 7).
