Calmodulin mediates sulfur mustard toxicity in human keratinocytes

Sulfur mustard (SM) causes blisters in the skin through a series of cellular changes that we are beginning to identify. We earlier demonstrated that SM toxicity is the result of induction of both death receptor and mitochondrial pathways of apoptosis in human keratinocytes (KC). Because of its importance in apoptosis in the skin, we tested whether calmodulin (CaM) mediates the mitochondrial apoptotic pathway induced by SM. Of the three human CaM genes, the predominant form expressed in KC was CaM1. RT-PCR and immunoblot analysis revealed upregulation of CaM expression following SM treatment. To delineate the potential role of CaM1 in the regulation of SM-induced apoptosis, retroviral vectors expressing CaM1 RNA in the antisense (AS) orientation were used to transduce and derive stable CaM1 AS cells, which were then exposed to SM and subjected to immunoblot analysis for expression of apoptotic markers. Proteolytic activation of executioner caspases-3, -6, -7, and the upstream caspase-9, as well as caspase-mediated PARP cleavage were markedly inhibited by CaM1 AS expression. CaM1 AS depletion attenuated SM-induced, but not Fas-induced, proteolytic processing and activation of caspase-3. Whereas control KC exhibited a marked increase in apoptotic nuclear fragmentation after SM, CaM1 AS cells exhibited normal nuclear morphology up to 48h after SM, indicating that suppression of apoptosis in CaM1 AS cells increases survival and does not shift to a necrotic death. CaM has been shown to activate the phosphatase calcineurin, which can induce apoptosis by Bad dephosphorylation. Interestingly, whereas SM-treated CaM1-depleted KC expressed the phosphorylated non-apoptotic sequestered form of Bad, Bad was present in the hypophosphorylated apoptotic form in SM-exposed control KC. To determine if pharmacological CaM inhibitors could attenuate SM-induced apoptosis via Bad dephosphorylation, KC were pretreated with the CaM-specific antagonist W-13 or its less active structural analogue W-12. Following SM exposure, KC exhibited Bad dephosphorylation, which was inhibited in the presence of W-13, but not with W-12. Consequently, W-13 but not W-12 markedly suppressed SM-induced proteolytic processing and activation of caspase-3, as well as apoptotic nuclear fragmentation. Finally, while the CaM antagonist W-13 and the calcineurin inhibitor cyclosporin A attenuated SM-induced caspase-3 activation, inhibitors for CaM-dependent protein kinase II (KN62 and KN93) did not. These results indicate that CaM, calcineurin, and Bad also play a role in SM-induced apoptosis, and may therefore be targets for therapeutic intervention to reduce SM injury.