Novel signaling pathways mediating reciprocal control of keratinocyte migration and wound epithelialization through M3 and M4 muscarinic receptors

Abstract

To test the hypothesis that keratinocyte (KC) migration is modulated by distinct muscarinic acetylcholine (ACh) receptor subtypes, we inactivated signaling through specific receptors in in vitro and in vivo models of reepithelialization by subtype-selective antagonists, small interfering RNA, and gene knockout in mice. KC migration and wound reepithelialization were facilitated by M4 and inhibited by M3. Additional studies showed that M4 increases expression of "migratory" integrins alpha5beta1, alphaVbeta5, and alphaVbeta6, whereas M3 up-regulates "sedentary" integrins alpha2beta1 and alpha3beta1. Inhibition of migration by M3 was mediated through Ca2+-dependent guanylyl cyclase-cyclic GMP-protein kinase G signaling pathway. The M4 effects resulted from inhibition of the inhibitory pathway involving the adenylyl cyclase-cyclic AMP-protein kinase A pathway. Both signaling pathways intersected at Rho, indicating that Rho kinase provides a common effector for M3 and M4 regulation of cell migration. These findings offer novel insights into the mechanisms of ACh-mediated modulation of KC migration and wound reepithelialization, and may aid the development of novel methods to promote wound healing.

Figure 1.

In vitro experiments with anti-mAChR siRNA. (A) Representative results of Western blot analysis of the effects of siRNA-M₃ and siRNA-M₄ on M₃ and M₄ mAChR expression in human KCs transfected with anti-mAChR siRNAs or negative control (NC) siRNA. On the third day after transfection, the cellular proteins were analyzed by Western blotting. The results were compared with those obtained with control (C) KCs exposed to TransIT-TKO (Mirus Corp.) transfection reagent mixed in Opti-MEM medium (GIBCO BRL) alone. The numbers underneath the bands are ratios of the densitometry value of each mAChR receptor to that of β-actin compared with the values obtained in control KCs (taken as 1). (B) IF analysis of the effects of anti-M₃ and anti-M₄ siRNAs on the expression of the M₃ and M₄ mAChR subtypes on the cell membrane of human KCs. The cells were cultured and treated with the siRNA-M₃ (black bars), siRNA-M₄ (white bars), or negative control siRNA (not depicted) as described in A, fixed to avoid cell membrane permeabilization, and stained with anti-M₃ or anti-M₄ antibodies. (C) Dose-dependent effects of receptor-selective siRNA on KC migration. Second passage human KCs were loaded into AGKOS plates, transfected with 1.2, 12, or 120 nM of negative control M₃ or M₄ siRNA, or exposed to TransIT-TKO transfection reagent mixed in Opti-MEM medium alone (control), and then used in AGKOS assay. (D) Opposing alterations in KC crawling locomotion due to M₃ and M₄ receptor gene silencing. (B–D) Asterisk indicates significant (P < 0.05) differences from untreated control. NC, negative control siRNA.

Figure 2.

Effects of mAChR gene KO on crawling locomotion of murine KCs. (A) The representative images of wound tissue of WT and mAChR KO mice harvested on the third and fifth day after wounding. The edge of the full thickness wound extending from the epidermis down to the dermis is depicted by arrowheads. The direction of forward movement of the epithelial tongue from the excision site (arrow) is indicated with a dotted line with an arrowhead. Bar, 150 μm. Insets show immunolocalization of M₃ and M₄ on the epithelial tongues of M₄ ^−/− and M₃ ^−/− mice, respectively, on the fifth day after wounding. Bar, 20 μm. (B) Alterations in epithelialization rate of skin wounds in M₃ ^−/− and M₄ ^−/− mice. The data are means ± SD of the rate of epithelialization computed in six M₃ ^−/− and M₄ ^−/− mice compared with that determined in six WT animals, taken as 100%. *, significant (P < 0.05) differences from the epithelization rate in WT mice. #, P = 0.05. (C) Representative images of hair follicles in the skin of 1-d-old WT and mAChR KO mice. An apparent stage of morphogenesis is shown as numbers underneath each hair follicle. Hematoxylin and eosin staining of cryostat sections of the longitudinal sections harvested from the upper back skin. Bar, 50 μm. (D) Quantitative morphometric analysis of hair follicle development in mAChR KO mice. The percentage of hair follicles in each stage of morphogenesis (x axis) was determined in 1-d-old WT, M₃ ^−/−, and M₄ ^−/− mice. (E) Alterations of migration rate of KCs from M₃ ^−/− and M₄ ^−/− mice in AGKOS plates. The results are means ± SD of the migration distance of intact WT KCs (taken as 100%). *, significant (P < 0.05) differences from the migration distance of WT KCs; #, significant (P < 0.05) differences from the migration distances of nontreated KCs in each subgroup.

Figure 3.

Effects of the mAChR signaling modifiers on migration of human KCs. The migration assays of intact or siRNA-M₃– or siRNA-M₄–transfected KCs were performed in AGKOS plates. The cells were exposed to 1 μM thapsigargin (TG), 10 μM LY-83,583 (LY), 1 mM SQ 22536 (SQ), 50 μM Sp-8-Br-cAMPS (Sp), 5 μM 8-pCPT-cGMP (8p),1 μM Rp-8-pCPT-cGMPS (Rp), 25 nM KT5720 (KT), 10 μg/ml C3 exoenzyme (C3), 5 μM Y-27632 (Y), or 10 nM MT3 dissolved in KGM. Some cells were pretreated for 30 min with thapsigargin, Rp-8-pCPT-cGMPS, or KT5720 before exposing them to a second test agent versus no additional (NA) treatment. The results are expressed as means ± SD of nontreated control, taken as 100%. *, significant (P < 0.05) differences from control. Significant differences between specific experimental conditions are indicated in the graph with square brackets with arrows.

Figure 4.

Muscarinic effects on integrin expression in human KCs. Relative amounts of α₂, α₃, α₅, α_V, and β₅ integrins were analyzed by Western blotting of total protein isolated from human KCs 72 h after transfection with siRNA-M₃, siRNA-M₄, or negative control siRNA (siRNA-NC), and from control KCs treated with TransIT-TKO transfection reagent mixed in Opti-MEM medium alone (control). Changes in the expression of integrins in human cultured KCs treated with mAChR signaling modifiers were similarly analyzed after incubation for 6 h or 10 d in the absence (control) or presence of 1 μM Rp-8-pCPT-cGMPS (Rp), 25 nM KT5720 (KT) or 10 nM MT3, or a combinations of these agents shown in the table. The images show typical bands appearing at the expected molecular weights (MW). *, significant (P < 0.05) differences from control.

Figure 5.

Role of integrins in mediating effects of M₃ and M₄ mAChRs on KC migration in vivo and in vitro. (A) Western blotting analysis of integrin gene expression. Equal samples of normal (N), i.e., intact, and wound (W) tissues were harvested from two wounds inflicted in six mice of each subgroup 24 h after wounding. Monolayers of murine KCs were established from the skin of neonatal WT and M₃ ^−/− and M₄ ^−/− mutant mice, and grown to ∼70% confluence in KGM. The results are expressed as ratios of the densitometry value of each integrin to that of β-actin in the same lane, compared with the values obtained in either intact skin or KCs grown from WT mice (taken as 1). (B) Semiquantitative IF analysis of KC integrin expression in the skin of WT and M₃ ^−/− and M₄ ^−/− mutant mice. Cryostat sections of intact and wounded skin from killed mice were stained with the antibodies targeting integrins indicated on the x axis. Fluorescence intensities of normal (intact) and wounded skin in the KO mice are expressed relative to levels determined in the intact and wounded skin, respectively, of WT mice (taken as 100%). Data are means ± SD from experiments performed with the tissue from two wounds in each of six mice in each subgroup. All changes are statistically significant (P < 0.05). (C) Immunolocalization of integrin receptors in KCs comprising the epithelialization tongue of WT mice on the third day after wounding. Bar, 25 μm. (D) Effects of the mAChR gene null mutation on the ability of murine KCs to migrate over ECM proteins. The migration distance was measured using a scratch assay. The M₃ ^−/− and M₄ ^−/− versus WT (control) KCs were freshly seeded in the culture dishes either uncoated (uc) or coated with fibronectin (fn), vitronectin (vn), laminin (lm), collagen I (cI) or IV (cIV), and the formed monolayers were wounded with a 100-μl pipette tip. The amount of migration was quantitated, and the results were expressed relative to migration distance of WT KCs (taken as 100%). (E) Effects of antiintegrin antibodies on migration distance of mAChR KO and WT KCs. The migration distance was measured using a scratch assay. The M₃ ^−/− and M₄ ^−/− KCs were grown to confluence in uncoated culture dishes, and the monolayers were wounded with a 100-μl pipette tip and incubated with (experiment) or without (control) antibodies targeting integrins identified on the x axis. The amount of migration was quantitated, and the results were expressed relative to migration distance of control KCs (taken as 100%). (A, D, and E) Asterisks indicate significant (P < 0.05) differences from control.

Figure 6.

Hypothetical scheme of the signaling mechanisms linking M₃ and M₄ to regulation KC motility and integrin expression. The signaling cascades originating at M₃ and M₄ include distinct stimulatory (→) and inhibitory (•••) steps and converge at the common effector pathway involving Rho proteins.