Enpp2/Autotaxin in dermal papilla precursors is dispensable for hair follicle morphogenesis

Abstract

Systematic ablation of previously identified dermal papilla (DP) signature genes in embryonic DP precursors will reveal their functional roles during hair follicle morphogenesis. In this study, we validate Enpp2/Autotaxin as one of the highest expressed signature genes in postnatal DP, and demonstrate specific expression of this lysophosphatidic acid (LPA)-generating enzyme in embryonic dermal condensates. We further identify dermal and epidermal expression of several LPA receptors, suggesting that LPA signaling could contribute to follicle morphogenesis in both mesenchymal and epithelial compartments. We then use the recently characterized Cre-expressing Tbx18 knock-in line to conditionally ablate Enpp2 in embryonic DP precursors. Despite efficient gene knockout in embryonic day 14.5 (E14.5) dermal condensates, morphogenesis proceeds regularly with normal numbers, lengths, and sizes of all hair follicle types, suggesting that Enpp2 is not required for hair follicle formation. To interrogate DP signature gene expression, we finally isolate control and Enpp2-null DP precursors and identify the expression and upregulation of LIPH, an alternative LPA-producing enzyme, suggesting that this gene could functionally compensate for the absence of Enpp2. We conclude that future coablation of both LPA-producing enzymes or of several LPA receptors may reveal the functional role of LPA signaling during hair follicle morphogenesis.

Figure 1. Expression of Enpp2 in postnatal dermal papilla (DP) and embryonic DP precursors

(a) Real-time PCR analysis of Enpp2 expression in DP and other hair follicle cells isolated by fluorescence activated cell sorting (FACS) from P5 Lef1-RFP/K14-H2BGFP skin (n=2). (b) In situ hybridization for Enpp2 mRNA in P5 back skin section. (c) Immunofluorescence for ENPP2 in P5 back skin section. ENPP2 is expressed in DP cells (arrows). Asterisks mark autofluorescence in hair shafts. (d) Whole-mount in situ hybridization for Enpp2 in E13.5 and E14.5 embryos. (e) Back skin of sectioned embryos at E14.5 and E15.5 show specific Enpp2 expression in dermal condensates. (f) Real-time PCR analysis of Enpp2 and other DP genes in FACS isolated DP precursors from E14.5 Tbx18^H2BGFP embryos (n=2). (g) Whole-mount immunofluorescence for ENPP2 and SOX2 in E14.5 embryo. (h) ENPP2 immunofluorescence staining on back skin section at E15.5. ITGB4 marked the basement membrane. ENPP2 is expressed in DP precursors in dermal condensates (arrows). Scale bars, 25μm. Data are mean ± SD.

Figure 2. Efficient Tbx18^Cre-mediated ablation of Enpp2 in embryonic dermal condensates

(a) Schematic of Enpp2 ablation in embryonic DP precursors. Red arrows indicate primers for generating riboprobes. (b) Whole-mount in situ hybridization of WT (Enpp2^fl/fl) and cKO (Tbx18^Cre;Enpp2^fl/fl) embryos at E14.5. Note absence of Enpp2 mRNA in cKO. (c) Whole-mount immunofluorescence for ENPP2 and SOX2 in Het (Tbx18^Cre;Enpp2^fl/+) and cKO (Tbx18^Cre;Enpp2^fl/fl) back skin at E14.5 confirms robust ENPP2 protein ablation. (d) Confocal imaging and 3D-reconstruction of z-dimension of whole-mount immunofluorescence at E14.5. Note absence of ENPP2 in cKO (bottom). (e) ENPP2 immunofluorescence in WT and cKO sections at E15.5. SDC1 marks DP precursors. (f) Quantification of ENPP2 and SDC1 double-labeled dermal condensates. Note that the few positive cKO condensates only had a single cell labeled (n=2). Scale bars, 25 μm. Data are mean ± SD.

Figure 3. Enpp2 ablation in embryonic DP precursors is dispensable for hair follicle formation

(a) Hematoxylin and eosin staining of WT and cKO back skin at E18.5. (b) Quantification of hair follicle types in Enpp2 WT and cKO back skin at E18.5 (n=2). (c) Immunofluorescence for DP marker SOX2 and Alkaline Phosphatase (AP) activity at E18.5. (d) Immunofluorescence for proliferation marker KI67 in WT and cKO embryos at E18.5 and quantification of positive cells in all three hair types. (e) Side view of WT and cKO back skin with outgrowing hair shafts at postnatal day P8. (f) Quantification of hair shaft lengths (n=2). (g) Hematoxylin and eosin staining of WT and cKO back skin at P8. (h) Quantification of hair follicle lengths (n=2). (i) KI67 immunofluorescence of back skin at P8 and quantification of proliferating cells (n=2). (j–l) Immunofluorescence for DP markers SOX2 (j), HHIP (k) and GFRA1 (l). Scale bars, 25μm (b,c), 50 μm (i), 200μm (g). Data are mean ± SD.

Figure 4. Upregulated LIPH and LPA receptors in isolated cKO DP precursors

(a) Schematic of crosses for Het and cKO DP precursor isolation by FACS. (b) FACS profile of embryo back skin at E14.5 from ACTB Cre reporter. Tbx18^Cre-positive DP precursors were isolated as GFP positive cells (mG⁺). (c) Real-time PCR analysis of Sox2 and Tbx18 expression in isolated epidermal (E-cadherin⁺), total dermal (E-cadherin⁻) and mG⁺ cells. (d) Real-time PCR analysis of Enpp2 in isolated Het and cKO DP precursors. (e) Real-time PCR analysis of known DP signature genes in dermal condensates. (f) Real-time PCR for LIPH in FACS-isolated epidermal and total dermal cells, and in DP precursors sorted from E14.5 Tbx18^H2BGFP embryos as GFP^hi cells compared to negative dermal cells. (g) LIPH expression in FACS-isolated Het and cKO dermal condensates (mG) and epidermis. (h) Real-time PCR of LPA receptor expression in FACS-isolated Het and cKO dermal condensates (mG) and epidermis. All data are n=2 and represented as mean ± SD. *, p<0.05; **, p<0.01.