Clinical treatment for, and research into, the structural and functional repair and regeneration of cutaneous tissue after severe burn injury is a huge challenge. Recently, cell therapy has emerged as a new tool to repair the injured skin structure and re-establish the sweating function [1, 3, 6, 15].

With easy accession and culture, a strong differentiation potency, and lower immune resistance, BM-MSCs are considered to be promising cells for cutaneous regeneration in severe burn patients. In the present study, we have demonstrated that direct activation of the EDA gene has great potential for facilitating the formation of sweat gland-like cells via the doxycycline-induced dCas9-E system, giving us a chance to improve the current methods of sweat gland cell generation.

Previous studies have demonstrated that protein encoded by the EDA gene could regulate the development of ectodermal tissues such as sweat glands, hair, and teeth [20]. People who lack the EDA gene have sparse or absent hair and teeth. These results and phenomenon indicate that EDA acts as a key regulator in the initiation stage of cutaneous appendage development [18, 21]. Thus, we hypothesized that activation of the EDA gene could facilitate BM-MSC differentiation into sweat glands.

The sgRNA-guided dCas9-E system is a newly developed system to activate transcription of the target gene [7, 22]. In the present study, three different sgRNAs were designed to target the –244 bp to –58 bp region upstream of the TSS in the human EDA promoter. Interestingly, individual sgRNAs targeting the EDA promotor around the –244 bp to –58 bp region could effectively increase the transcriptional activation in the presence of dCas9-E (Fig. 3), and the sgRNA2 around the –244 bp to –225 bp region showed higher activity than sgRNA1 (from –131 bp to –112 bp) and sgRNA3 (from –78 bp to –59 bp), suggesting that target position is involved in the activity of the dCas9-E.

Sweat gland epithelial cells have been identified as essential for the construction of skin substitutes and regeneration of sweat glands [23]. Several markers, including CEA, CK7, CK8, CK14, CK15, CK18, and CK19, were identified as the sweat gland-specific markers during SG development. In normal adult skin tissue, these markers (CK7, CK8, CK14, CK15, CK18, and CK19) are expressed in the secretory portion [5, 24]. In addition, CEA, a biomarker for colorectal cancer, is also expressed in some normal adult tissue, such as mucous neck cells and pyloric mucous cells in the stomach, and secretory epithelia and duct cells of sweat glands [25]. Therefore, the high expression of CEA, CK7, CK8, CK14, CK15, CK18, and CK19 is considered as an index for the identification of SGs [15, 23, 26]. After transfection with sgRNA-guided dCas9-E, the BM-MSCs were identified as showing high expression of CEA, CK7, CK14, and CK19 (Fig. 4). These results indicated that activation of the EDA gene could induce BM-MSCs into sweat gland-like cells. The function of the transfected BM-MSCs was measured between treated paws and sham paws in vivo. Results showed that paws treated with Dox-induced dCas9-E BM-MSCs had accelerated wound healing with less collagen deposition (Fig. 5a, b). Primary research has shown that the wound healing process depends on cell proliferation in the basement membrane of the wound edge [27]. Immunofluorescence staining of Ki67 in the Dox-induced dCas9-E BM-MSC treatment group showed higher proliferation in the basement membrane cells compared with the sham group after newly complete re-epithelialization (Fig. 5c), while the number of Ki67-positive cells decreased in the stabilized healing site (Additional file 4), in accordance with changes in MSC-based wound healing therapy [28]. Therefore, we concluded that the transfected cells showed partly preserved MSC features. In addition, hematoxylin and eosin staining and immunofluorescence staining also indicated that the scalded paws treated with Dox-induced dCas9-E BM-MSCs showed sweat gland duct structure and positive markers (CEA and CK19) for sweat glands (Fig. 6). However, the mechanism for EDA regulation of sweat gland development was still unclear.

Previous studies have revealed that multiple signaling pathways, such as ERK-MAPK, EDA/EDAR/NF-κB, and Wnt/β-catenin, were involved in sweat gland development [29, 30]. EDA, a member of the TNF superfamily, is one of the functional genes that regulate the development of sweat glands, and mutations in EDA can cause ectodermal dysplasia in humans and lead to sweat-free syndrome [31]. In the present study, therefore, we focused on the EDA/EDAR/NF-κB pathway. The EDA pathway could activate NF-κB through the IKK pathway, and the activated NF-κB can enter the nucleus to promote the expression of cyclin D1, Shh, Fox family genes, and keratins [15], which play an important role in the development of sweat glands [32]. We further found that the EDA/EDAR/NF-κB signaling pathway was activated in the reprogrammed sweat gland-like cells and the injured site in vivo. The activated NF-κB then activates the expression of Shh and cyclin D1 downstream (Fig. 8). In combination with the in vivo experimental results, we have shown a new therapeutic direction in sweat gland regeneration by revealing a promising technology for direct and effective reprogramming.