Stress-mediated increases in systemic and local epinephrine impair skin wound healing: potential new indication for beta blockers

Abstract

Background: Stress, both acute and chronic, can impair cutaneous wound repair, which has previously been mechanistically ascribed to stress-induced elevations of cortisol. Here we aimed to examine an alternate explanation that the stress-induced hormone epinephrine directly impairs keratinocyte motility and wound re-epithelialization. Burn wounds are examined as a prototype of a high-stress, high-epinephrine, wound environment. Because keratinocytes express the beta2-adrenergic receptor (beta2AR), another study objective was to determine whether beta2AR antagonists could block epinephrine effects on healing and improve wound repair.

Methods and findings: Migratory rates of normal human keratinocytes exposed to physiologically relevant levels of epinephrine were measured. To determine the role of the receptor, keratinocytes derived from animals in which the beta2AR had been genetically deleted were similarly examined. The rate of healing of burn wounds generated in excised human skin in high and low epinephrine environments was measured. We utilized an in vivo burn wound model in animals with implanted pumps to deliver beta2AR active drugs to study how these alter healing in vivo. Immunocytochemistry and immunoblotting were used to examine the up-regulation of catecholamine synthetic enzymes in burned tissue, and immunoassay for epinephrine determined the levels of this catecholamine in affected tissue and in the circulation. When epinephrine levels in the culture medium are elevated to the range found in burn-stressed animals, the migratory rate of both cultured human and murine keratinocytes is impaired (reduced by 76%, 95% confidence interval [CI] 56%-95% in humans, p < 0.001, and by 36%, 95% CI 24%-49% in mice, p = 0.001), and wound re-epithelialization in explanted burned human skin is delayed (by 23%, 95% CI 10%-36%, p = 0.001), as compared to cells or tissues incubated in medium without added epinephrine. This impairment is reversed by beta2AR antagonists, is absent in murine keratinocytes that are genetically depleted of the beta2AR, and is reproduced by incubation of keratinocytes with other beta2AR-specific agonists. Activation of the beta2AR in cultured keratinocytes signals the down-regulation of the AKT pathway, accompanied by a stabilization of the actin cytoskeleton and an increase in focal adhesion formation, resulting in a nonmigratory phenotype. Burn wound injury in excised human skin also rapidly up-regulates the intra-epithelial expression of the epinephrine synthesizing enzyme phenylethanolamine-N-methyltransferase, and tissue levels of epinephrine rise dramatically (15-fold) in the burn wounded tissue (values of epinephrine expressed as pg/ug protein +/- standard error of the mean: unburned control, 0.6 +/- 0.36; immediately postburn, 9.6 +/- 1.58; 2 h postburn, 3.1 +/- 1.08; 24 h post-burn, 6.7 +/- 0.94). Finally, using an animal burn wound model (20% body surface in mice), we found that systemic treatment with betaAR antagonists results in a significant increase (44%, 95% CI 27%-61%, p < 0.00000001) in the rate of burn wound re-epithelialization.

Conclusions: This work demonstrates an alternate pathway by which stress can impair healing: by stress-induced elevation of epinephrine levels resulting in activation of the keratinocyte beta2AR and the impairment of cell motility and wound re-epithelialization. Furthermore, since the burn wound locally generates epinephrine in response to wounding, epinephrine levels are locally, as well as systemically, elevated, and wound healing is impacted by these dual mechanisms. Treatment with beta adrenergic antagonists significantly improves the rate of burn wound re-epithelialization. This work suggests that specific beta2AR antagonists may be apt, near-term translational therapeutic targets for enhancing burn wound healing, and may provide a novel, low-cost, safe approach to improving skin wound repair in the stressed individual.

Conflict of interest statement

Competing Interests: RRI and CEP are named in a patent application submitted by the University of California for the use of beta adrenergic agents in modulating wound repair.

Figures

Figure 1. Epinephrine Decreases Keratinocyte Migratory Speed and In Vitro Scratch Wound Healing by Activation of the β2AR

(A) Neonatal human keratinocytes (NHK) migratory speeds were measured for 1 h after treatment with epinephrine (10 nM), epinephrine (10 nM) and timolol (10 μM, nonspecific βAR antagonist), or no added drug (control). Error bars represent standard error of the mean, n = 80; * p < 0.05; ** p < 0.001. (B) Epidermal keratinocytes were cultured from FVB β2AR knockout mice (β2AR-KO, β2AR −/−) and wild-type controls (WT, β2AR +/+) and assessed for their migratory speeds, as in Figure 1A. Error bars represent standard error of the mean, n = 31; * p < 0.01. (C) NHK in vitro scratch wound closure after control, epinephrine, and timolol treatments. Error bars represent standard error of the mean, n = 12; * p < 0.05; ** p < 0.001. (D) Representative pictures from the scratch wound assay are shown. Scale bar = 100 μm.

Figure 2. The β2AR Pathway Involves PI3K/AKT in Controlling Keratinocyte Migration

(A) (Top) Keratinocytes were treated with procaterol (10 nM, specific β2AR agonist), and the cells were collected at sequential time points and assayed for the expression of phosphorylated AKT (p-AKT) and total AKT. A representative western blot (of two different human keratinocyte strains) is shown. Untreated keratinocytes served as control. (Bottom) p-AKT expression was significantly decreased at 15, 30, and 60 min after procaterol (10 nM) treatment. Error bars represent standard error of the mean, n = 4; * p < 0.05. (B) Keratinocyte migration was measured after exposure to either procaterol (10 nM), ICI-118,551 (20 μM, specific β2AR antagonist), procaterol with ICI-118,551, LY294002 (50 μM, PI3K inhibitor), or LY294002 (50 μM) with ICI-118,551 (20 μM). Untreated keratinocytes served as controls. Data were pooled from results from two different human keratinocyte strains. Error bars represent standard error of the mean, n = 67–106; *, p < 0.05; **, p < 0.001. (C) Keratinocytes immunostained for the expression of actin and vinculin after they were treated with either procaterol (10 nM), ICI-118,551 (20 μM), or LY294002 (50 μM). Procaterol and LY294002 both induced similar changes in the cytoskeletal organization of vinculin whereas ICI-118,551 did not. Untreated keratinocytes served as controls. Scale bar = 15 μm

Figure 3. Ex vivo Burn Wound Re-epithelialization Is Inhibited by Epinephrine

(A) Ex vivo human skin burn wounds (2 mm wide) were treated with either epinephrine (10 nM) (Epi), or with the combination of epinephrine (10 nM) and timolol (10 μM) (EpiTim). Control (Con) treatment consisted of DMEM with 10% FBS. Re-epithelialization was monitored by histology followed over 7 d. Error bars represent standard error of the mean, n = 3; * p < 0.05; ** p < 0.01. (B) Representative day 7 burn wound histological cross-sections are shown. Arrows represent the wound edge and the arrowheads represent the edge of re-epithelialization. Scale bar = 100 μm.

Figure 4. Human Skin Generates Local Epinephrine Ex Vivo in Response to Burn Injury

(A) Ex vivo human burn wounds (2.5 mm wide) in one piece of donor skin were divided into three groups that received either no treatment (CON) or timolol (10 μM ) (TIM). It should be noted that these burn wounds are larger in width than those in Figure 2, but the relative rate at which the timolol treated group re-epithelializes is consistent at approximately twice the speed of the control by day 5 (Figure 2A) or day 6 (Figure 3A). Error bars represent standard error of the mean, n = 5; * p < 0.01. (B) Representative day 10 burn wound histological sections are shown. Arrows represent the wound edge and the arrowheads represent the edge of re-epithelialization. Scale bar = 100 μm. (C) Human skin burn wounds were examined for the expression of PNMT by immunohistochemistry at day 1 after burn wounding. Increased expression of PNMT was found in the epidermis immediately adjacent to the burn wound margins. Scale bar = 100 μm. (D) Western blot for PNMT expression in lysates of peri-wound skin. Lane 1, unburned tissue control; lane 2, 30 min postburn; lane 3, 2 h postburn; and lane 4, 24 h postburn. (E) Epinephrine levels were measured in unburned skin (control), immediately (time 0), 2 h, and 24 h after burn wounding. Error bars represent standard error of the mean, n = 3, * p < 0.001.

Figure 5. In Vivo Burn Wound Re-epithelialization Is Accelerated by β2AR Blockade

Mice were subjected to a 20% body surface area full thickness burn and were treated using continuous infusion pumps with either saline (control), 3 kg/mg/d salbutamol (β2AR specific agonist), or 0.7 kg/mg/d ICI-118,551 (β2AR specific antagonist). (A) Plasma epinephrine levels were measured in each animal at day 10 postburn wounding. The mean preburn level of epinephrine (dashed line) was determined from plasma samples taken immediately following burn wounding. Error bars represent standard error of the mean, n = 5. (B) Wound re-epithelialization was measured by image analysis of histologic sections of the wounds excised on either day 5 or day 10 postburn wounding. Error bars represent standard error of the mean, n = 5, * p < 0.001. (C) Representative histological sections from day 10 burn wounds are shown. Arrows represent the wound edge and the arrowheads represent the edge of re-epithelialization. The blacks lines indicate the zone of re-epithelialization. The overall wound images are a montage compiled from multiple images as indicated by vertical bars. Scale bar = 1 mm. (D) Magnified images showing the re-epithelializing edge from the right side of the pictures in (C). Arrowheads represent the edge of re-epithelialization and the black lines indicate the zone of re-epithelialization. Scale bar = 4 mm.