Skin β-endorphin mediates addiction to UV light

Abstract

UV light is an established carcinogen, yet evidence suggests that UV-seeking behavior has addictive features. Following UV exposure, epidermal keratinocytes synthesize proopiomelanocortin (POMC) that is processed to melanocyte-stimulating hormone, inducing tanning. We show that, in rodents, another POMC-derived peptide, β-endorphin, is coordinately synthesized in skin, elevating plasma levels after low-dose UV. Increases in pain-related thresholds are observed and reversed by pharmacologic opioid antagonism. Opioid blockade also elicits withdrawal signs after chronic UV exposure. This effect was sufficient to guide operant behavioral choices to avoidance of opioid withdrawal (conditioned place aversion). These UV-induced nociceptive and behavioral effects were absent in β-endorphin knockout mice and in mice lacking p53-mediated POMC induction in epidermal keratinocytes. Although primordial UV addiction, mediated by the hedonic action of β-endorphin and anhedonic effects of withdrawal, may theoretically have enhanced evolutionary vitamin D biosynthesis, it now may contribute to the relentless rise in skin cancer incidence in humans.

Fig. 1. Plasma β-endorphin increases with chronic UV exposure and parallels Naloxone reversible changes in pain tolerance

A) Plasma β-endorphin in C57Bl6 mice receiving daily UV or Mock irradiation. Mice were treated twice a week with either Naloxone or Saline as indicated. Data are represented as the mean +/− SEM, 2way ANOVA analysis with Bonferroni’s multiple comparisons test gives p<0.05 for both UV treated groups compared to both Mock treated groups (during UV treatment, days 14–42), and no significant effect of Naloxone treatment within either group. B) von Frey thresholds and C) Hot Plate thresholds in chronically UV-irradiated and mock-irradiated C57Bl6 mice (mean +/− SEM). Half of each group was pre-treated with naloxone (10mg/kg) 15 minutes prior to nociceptive testing, while the remainder received saline (n=10 per group). Analgesic thresholds were further monitored for 2 additional weeks after cessation of UV/Mock treatment. 2way ANOVA with Bonferroni’s multiple comparisons test reveals p<0.0001 for the UV/Saline treated group compared to all other groups during UV treatment, days 9 to 39).

Fig. 2. Straub Tail in UV-irradiated mice is reversed by naloxone

A) Straub Tail in C57Bl6 mice over the course of 42 days of UV irradiation (n=13) or mock-irradiation (n=6). Data are represented as the mean +/− SEM, for days 10–37 p<0.0001 by 2way Anova analysis. B) Straub Tail in at day 17 before (Pre) and 15 minutes after (Post) injection of naloxone (n=7) or saline (n=6). Data are represented as the mean +/− SEM, p<0.001 by Student’s t-test. C) Representative animals from each group in part (B). The beginning of black fur re-growth produces a patchy appearance.

Fig. 3. Chronically UV-exposed mice show symptoms of opioid dependence

A) Signs of opioid withdrawal in mice under experimental conditions described in Figure 3. UV / saline (n=9), mock / saline (n=7), UV / naloxone (n=15), and mock / naloxone (n=7). Data are represented as the mean +/− SEM, *p<0.05 compared to UV/Saline group by 2way ANOVA with Bonferroni’s multiple comparisons test. B) Conditioned place aversion testing in UV treated mice conditioned to the naloxone-paired box (black box) with an injection of naloxone or saline (white box) following 42 days of UV or mock treatment. Mice were permitted to freely move between naloxone-paired and saline-paired boxes prior to (pretest, n=8) and after 4 days of conditioning (test), and place preferences were assessed as change in time spent in the naloxone-paired box (postconditioning – preconditioning). Data are represented as the mean +/− SEM, p values were generated by 2way ANOVA with Bonferroni’s multiple comparisons test. C) Morphine dose-response curves in mice following 42 days UV irradiation (n=31) or mock exposure (n=29). Data are represented as the mean +/− SEM, p<0.0001 by 2way ANOVA. D) Conditioned place preference testing in mice administered i.v. β-endorphin or saline through the tail vein. Mice were conditioned to β-endorphin (6) or Saline (8) in the white box and Saline in the black box. Place preferences were assessed as change in time spent in the white (β-endorphin -paired box), postconditioning – preconditioning. Data are represented as the mean +/− SEM, p=0.0145 by Student’s t-test.

Fig. 4. Genetic lack of β-endorphin abolishes changes in pain tolerance and opioid dependence with chronic UV exposure

A) von Frey test and B) thermal analgesic thresholds in wild type (n=11) and β-endorphin −/− (n=13) mice over 35 day UV exposure. Data are represented as the mean +/− SEM, *p<0.05 by 2way ANOVA with Bonferroni’s multiple comparisons test. C) Signs of naloxone precipitated opioid withdrawal in control and β-endorphin null mice after 6 weeks of UV. Data are represented as the mean +/− SEM, *p<0.0001 compared to β-endorphin^−/−/Naloxone group by 2way ANOVA with Bonferroni’s multiple comparisons test. D) Conditioned place aversion testing in UV treated control and β-endorphin null mice conditioned to the naloxone-paired box (black box) with an injection of naloxone or saline. All mice were conditioned to saline in the white box, n>10 for all groups. Mice were permitted to freely move between naloxone-paired and saline-paired boxes prior to and after 4 days of conditioning, and place preferences were assessed as change in time spent in the naloxone-paired box (postconditioning – preconditioning). Data are represented as the mean +/− SEM, p values were generated by 2way ANOVA with Bonferroni’s multiple comparisons test.

Fig. 5. Keratinocyte expression of p53 is required for elevated β-endorphin levels and pain thresholds

A) Representative K14cre and p53fl/fl K14cre mice after 4 weeks of daily UV treatment. B) Plasma β-endorphin in mice in K14cre and p53fl/fl K14cre mice receiving daily UV irradiation. Data are represented as the mean +/− SEM, *p<0.05 by 2way ANOVA analysis with Bonferroni’s multiple comparisons test. C) Mechanical analgesic thresholds in K14cre (n=10) and p53fl/fl K14cre (n=9) mice over 13 days UV exposure. Data are represented as the mean +/− SEM, *p<0.05 by 2way ANOVA with Bonferroni’s multiple comparisons test. D) K14cre and p53fl/fl K14cre mice were conditioned to naloxone in the black box after 3 weeks of daily UV. Place preferences were assessed as change in time spent in the naloxone-paired box. Data are represented as the mean +/− SEM, p=0.0317 by Student’s t-test. The change in time spent in the black box was not significant when the postconditioning and preconditioning times were compared by Student’s t-test (p=0.26).