Skin depletion of Kif3a resembles the pediatric atopic dermatitis transcriptome profile

Abstract

Skin deficiency of kinesin family member 3A causes disrupted skin barrier function and promotes development of atopic dermatitis (AD). It is not known how well Kif3aK14∆/∆ mice approximate the human AD transcriptome. To determine the skin transcriptomic profile of Kif3aK14∆/∆ mice and compare it with other murine AD models and human AD, we performed RNA-seq of full-thickness skin and epidermis from 3- and 8-week-old Kif3aK14∆/∆ mice and compared the differentially expressed genes (DEGs) with transcriptomic datasets from mite-induced NC/Nga, flaky tail (Tmem79ma/ma Flgft/ft), and filaggrin-mutant (Flgft/ft) mice, as well as human AD transcriptome datasets including meta-analysis derived atopic dermatitis [MADAD] and the pediatric atopic dermatitis [PAD]. We then interrogated the Kif3aK14∆/∆ skin DEGs using the LINCS-L1000 database to identify potential novel drug targets for AD treatment. We identified 471 and 901 DEGs at 3 and 8 weeks of age, respectively, in the absence of Kif3a. Kif3aK14∆/∆ mice had 3.5-4.5 times more DEGs that overlapped with human AD DEGs compared to the flaky tail and Flgft/ft mice. Further, 55%, 85% and 75% of 8-week Kif3aK14∆/∆ DEGs overlapped with the MADAD and PAD non-lesional and lesional gene lists, respectively. Kif3aK14∆/∆ mice spontaneously develop a human AD-like gene signature, which better represents pediatric non-lesional skin compared to other mouse models including flaky tail, Flgft/ft and NC/Nga. Thus, Kif3aK14∆/∆ mice may model pediatric skin that is a precursor to the development of lesions and inflammation, and hence may be a useful model to study AD pathogenesis.