One mechanism by which non-steroidal anti-inflammatory drugs (NSAIDs) cause intestinal injury is by inducing matrix metalloproteinases (MMPs) that degrade and remodel the extracellular matrix. In addition to the intestinal mucosa, MMPs are expressed in the skin and can be activated by mast cell-secreted tryptase. We therefore investigated whether intestinal injury resulting from treatment with the NSAID indomethacin induced MMPs in the skin of mice and caused an associated disruption of skin function. Hairless mice and mast cell-deficient mice were administered indomethacin, after which damage to the jejuna and skin was assessed with immunohistochemistry and Western blotting. The plasma concentration of inflammatory mediators was assessed to evaluate potential pathways for signalling skin disruption in response to intestinal injury. In hairless mice with intestinal injury, transepidermal water loss (TEWL) was higher and skin hydration was lower than in control mice. The expression levels of mast cells, tryptase, MMP-1 and MMP-9 were also increased, with concurrent degradation of types I and IV collagen. In contrast, no changes in skin TEWL or skin hydration were observed in mast cell-deficient mice with indomethacin-induced intestinal injury. In all mice evaluated, the plasma concentrations of IgE, IgA, histamine and TNF-α were increased in response to indomethacin treatment. Skin disruption was strongly associated with indomethacin-induced small intestinal injury, and the activation of mast cells and induction of tryptase, MMP-1 and MMP-9 are critical to this association.

Dry skin has been clinically associated with visceral diseases, including liver disease, as well as for our previously reported small intestinal injury mouse model, which have abnormalities in skin barrier function. To clarify this disease-induced skin disruption, we used a dextran sulphate sodium (DSS)-induced colitis mouse model. Following treatment with DSS, damage to the colon and skin was monitored using histological and protein analysis methods as well as the detection of inflammatory mediators in the plasma. Notably, transepidermal water loss was higher, and skin hydration was lower in DSS-treated mice compared to controls. Tumor necrosis factor-alpha (TNF-α), interleukin 6 and NO2-/NO3- levels were also upregulated in the plasma, and a decrease in body weight and colon length was observed in DSS-treated mice. However, when administered TNF-α antibody or an iNOS inhibitor, no change in skin condition was observed, indicating that another signalling mechanism is utilized. Interestingly, the number of tryptase-expressing mast cells, known for their role in immune function via cholinergic signal transduction, was elevated. To evaluate the function of cholinergic signalling in this context, atropine (a muscarinic cholinoceptor antagonist) or hexamethonium (a nicotinic cholinergic ganglion-blocking agent) was administered to DSS-treated mice. Our data indicate that muscarinic acetylcholine receptors (mAChRs) are the primary receptors functioning in colon-to-skin signal transduction, as DSS-induced skin disruption was suppressed by atropine. Thus, skin disruption is likely associated with DSS-induced colitis, and the activation of mast cells via mAChRs is critical to this association.

We have previously reported that impaired skin barrier function was induced by small intestinal injury in mice. Therefore, we postulated that other intestinal diseases might also influence skin barrier function. In this study, we evaluated the skin barrier function of hairless mice with colon carcinoma that was induced by azoxymethane (AOM) and dextran sodium sulfate (DSS). In mice treated with these drugs, we observed elevated transepidermal water loss and reduced skin hydration levels, compared to those in the control mice. In addition, plasma nitrogen di/trioxide (NO 2 /NO 3 ) levels were significantly elevated, and expression of type I collagen was significantly reduced in the treated mice, compared to those in control. These results suggest that impaired skin barrier function occurs in mice when colon carcinoma is present.

We propose a novel mechanism whereby plasma TNF-α, via TNFR2 alone or with TNFR1, plays an important role in skin barrier function during chronic liver disease in these mouse models.