Cytotoxicity of naphthylisothiocyanates in rat hepatocyte-neutrophil cocultures.

1-Naphthylisothiocyanate (ANIT) produces cholangiolitic hepatitis in rats. This injury is characterized by periportal bile duct and hepatic parenchymal cell necrosis with inflammatory cell involvement. In contrast, 2-naphthylisothiocyanate (BNIT) does not induce cholangiolitic hepatitis. The mechanism(s) involved in ANIT-induced hepatic injury remain to be elucidated. To investigate this difference in toxicity further, we examined the cytotoxicity of ANIT and BNIT in primary rat hepatocyte cultures. Since neutrophils (PMNs) are required for the development of ANIT-induced cholangiolitic hepatitis in vivo, we also examined the potential for PMNs to modulate ANIT and BNIT cytotoxicity in rat hepatocyte-PMN cocultures. Both ANIT and BNIT injured rat hepatocytes within the range of concentrations examined (0-100 microM); however, BNIT was more potent. The presence of PMNs did not significantly influence the hepatocellular injury produced by either naphthylisothiocyanate (NIT). In an attempt to clarify the disparity between these results in vitro and observations reported in vivo, we examined, in hepatocyte PMN cocultures, the cytotoxic potential of bile collected from NIT-treated rats. Bile from BNIT-treated rats was markedly more cytotoxic to hepatocytes than was bile from ANIT-treated rats. As was observed in earlier experiments, the inclusion of PMNs had no effect on the hepatocellular toxicity of bile from NIT-treated rats. These findings prompted evaluation of the effect of NITs on rat PMNs. ANIT (1 and 10 microM) had no effect on phorbal myristate acetate (PMA)-induced superoxide production by PMNs, whereas BNIT (1 and 10 microM) markedly inhibited PMA-induced superoxide production. In contrast, ANIT and BNIT were equally effective at inhibiting f-met-leu-phe (fMLP)-induced PMN degranulation (beta-glucuronidase release). Altogether, the relative NIT toxicity observed in hepatocyte primary cultures is contrary to reports of hepatotoxic potential of these NITs in vivo. The PMN-dependence of ANIT hepatotoxicity in vivo was not reproduced in hepatocyte-PMN cocultures exposed to ANIT, suggesting that the PMN dependence in vivo involves factors not present in hepatocyte PMN cocultures. The greater PMN inhibitory effect of BNIT may, in part, underlie its inability to elicit the PMN-dependent liver injury in vivo that characterizes ANIT-induced cholangiolitic hepatitis.