Toxicological effects of palytoxin after cutaneous exposure

Palytoxin (PLTX) is a marine toxin identified in Palythoa zoanthid corals and Ostreopsis dinoflagellates, representing an increasing hazard for human health. Human poisonings attributed to PLTX exposure are usually associated to ingestion of contaminated seafood and to marine aerosol exposure during Ostreopsis blooms. However, also dermatological problems have been recently associated to PLTX cutaneous exposure during Ostreopsis blooms as well as after handling of Palythoa corals. Despite the increasing human cases of dermotoxicity attributed to PLTX, very few data about its dermal toxicity are presently available. Hence, the aim of this study is to investigate the cutaneous effects of PLTX characterizing its mechanism of action. Thus, this toxicological in vitro study has been carried out on spontaneously immortalized human keratinocytes (HaCaT cells), as a first-round screening of dermotoxicity. The entity of cytotoxicity induced by PLTX has been firstly investigated. A short time exposure (4 h) to PLTX reduces mitochondrial activity (MTT assay), cell mass (SRB assay) and plasma membrane integrity (LDH leakage) with different potencies (EC50 values of 6.1±1.3x10-11, 4.7±0.9x10-10 M and 1.8±0.1x10-8 M, respectively). All these effects are ouabain-sensitive corroborating the dependency of PLTX effects on the interaction with Na+/K+-ATPase. These results indicate that among the chain of intracellular events following the interaction of PLTX with the Na+/K+-ATPase the earliest is mitochondrial damage. This sustained cytotoxic effect can be explained by the high affinity of binding to HaCaT cells. Indeed, saturation experiment reveals a Kd affinity constant of 3.0±0.4x10-10 M after an exposure time as short as 10 minutes. A possible mechanism of mitochondrial dysfunction can be reactive oxygen species (ROS) overproduction. Among all, only superoxide anion (O2-) seems to be produced by the toxin after only 1 h, whereas neither nitric oxide nor peroxynitrite formation are detected. Hence, the mechanism of O2- production has been investigated. Real time PCR analysis together with western blot analysis suggest a possible involvement of NADPH oxidase (NOX) and inducible nitric oxide synthetase (iNOS) since an early increase of their gene and protein expression was observed after short (1 †" 4 h) but not longer (24 h) exposure times. On the contrary, other enzymes involved in ROS production (i.e. COX-1, COX-2, XOD) seem to be not involved in PLTX effects. Moreover, using selective inhibitors of these enzymes, we found that only DPI, a nonspecific inhibitor of both NOX and NOS, is able to inhibit by 15%, 26% and 43% O2- production induced by 10-10, 10-9 and 10-8 M PLTX, respectively. However, NMMA, inhibitor of NOS, significantly reduces only O2- produced by high (10-8 M) but no by low (10-9 and 10-10 M) PLTX concentrations, whereas the selective inhibitor of NOX apocynin is totally ineffective. Moreover, since their co-administration does not reproduce DPI effect, a prominent role of these enzymes in causing PLTX-induced oxidative stress seems unlikely. Another feasible source of O2- is mitochondria itself and its production is regulated by H+ fluxes through mitochondrial membranes. Indeed, in presence of nigericin, an ionophore that reduces the H+ imbalance, PLTX-induced O2- is significantly reduced by 23% (10-9 M PLTX) and 24% (10-8 M PLTX). Furthermore, the co-administration with rotenone, a complex I inhibitor, that per se is ineffective, results in a further inhibition of O2- production (-32% and -43% in the presence of 10-9 and 10-8 M PLTX, respectively). Moreover, O2- production turned out to be ouabain-sensitive and Na+-dependent but Ca2+-independent. Thus, on the basis of these results it has been hypothesized that PLTX binding to Na+/K+-ATPase induces intracellular overload of Na+ followed by intracellular increase of H+ with a consequent ?pH increase across H+-impermeable mitochondrial inner membrane and O2- overproduction by reverse electron transports through mitochondrial chain. Under oxidative stress conditions, mitochondrial dysfunction can be mediated by mitochondrial permeability transition pore (MPTP), which opening, indeed, is induced by PLTX already after only 5 minutes exposure. MPTP opening, which turned out to be cyclosporine A-independent, seems to be mainly induced by the sustained ionic imbalance, since in Na+-free, Ca2+-free medium and in presence of nigericin PLTX effect is strongly inhibited. The very rapid Na+-dependent opening of MPTP suggests that this is the peculiar mechanism of PLTX cytotoxicity and cell death primum movens. Cell death induced by the toxin seems to occur with necrotic-like features. PLTX, indeed, induces a concentration- and time-dependent as well as irreversible uptake of PI after only 1 h exposure and confocal images revealed dramatic morphological alterations such as plasma membrane ruptures and leakage of cytolpasmic content after 4 h. By contrast, caspasis 3/7, 8 and 9 are not activated by PLTX up to 24 h, neither under recovery conditions. Moreover, apoptotic bodies formation is not observed, discarding apoptosis occurrence. Finally, PLTX effects on some pro-inflammatory mediators such as cytokines (IL-1?, IL-6, IL-8 and TNF-?) and arachidonic acid metabolism products (PGE2 and LTB4) have been evaluated. The toxin (10-11 M) induces an early release of PGE2 that is time-dependent after 2 h exposure. On the contrary, even if an early gene expression (1†"4 h) is observed, the toxin induces a delayed release of IL-6 and IL-8 (24 h), whereas no effects have been observed evaluating IL-1? and TNF-?. In conclusion, this study highlights the toxic in vitro properties of PLTX on human keratinocytes. The intracellular pathway of the sustained PLTX cytotoxicity leading to cell death has been characterized, as well as the inflammatory mediators involved in skin irritant properties of the toxin. These results can corroborate the use of non steroidal anti-inflammatory drugs in association with anti-inflammatory corticosteroids.