Activation of poly [ADP-Ribose] polymerase in endothelial cells and keratinocytes: role in an in vitro model of sulfur mustard-mediated vesication.

Although endothelial cells and keratinocytes appear to be the primary cellular targets of sulfur mustard (SM), the role of the nuclear enzyme poly (ADP-ribose) polymerase (PARP) in SM-induced vesication has not been clearly defined. PARP is thought to play a crucial role in DNA repair mechanisms following exposure to alkylating agents like SM. Using a combination of fluorescence microscopy and biochemical assays, we tested the hypothesis that SM causes activation of PARP in endothelial cells and keratinocytes with subsequent loss of nicotinamide adenine dinucleotide (NAD) and depletion of adenosine triphosphate (ATP) levels. To determine if PARP activation accounts for SM-induced vesication, keratinocyte adherence and permeability of endothelial monolayers were measured as in vitro correlates of vesication. As early as 2 to 3 h after exposure to SM concentrations as low as 250 microM, dramatic changes were induced in keratinocyte morphology and microfilament architecture. Exposure to 500 microM SM induced a fourfold increase in PARP activity in endothelial cells, and a two- to threefold increase in keratinocytes. SM induced a dose-related loss of NAD+ in both endothelial cells and keratinocytes. ATP levels fell to approximately 50% of control levels in response to SM concentrations >/=500 microM. SM concentrations >/=250 microM significantly reduced keratinocyte adherence as early as 3 h after exposure. Endothelial monolayer permeability increased substantially with concentrations of SM >250 microM. These observations support the hypothesis that the pathogenic events necessary for SM-induced vesication (i.e., capillary leak and loss of keratinocyte adherence) at higher vesicating doses of SM (>/=500 microM) may depend on NAD loss with PARP activation and subsequent ATP-dependent effects on microfilament architecture. Vesication developing as a result of exposure to lower concentrations of SM presumably occurs by mechanisms that do not depend on loss of cellular ATP (e.g., apoptosis and direct SM-mediated damage to integrins and the basement membrane).