Expression of heat shock proteins in chronic pancreatitis: protective or pathogenic roles?

The tissue-distributions of heat shock proteins (HSP) identified by monoclonal antibodies ML-30, TB-78, CA-Str7-1, and MAB 72/73 have been examined in formalin-fixed and paraffin wax-embedded tissues from 10 normal pancreatic specimens and 92 cases of chronic pancreatitis. The chronic pancreatitis cases were divided by probable etiology into alcohol-related, postacute pancreatitis, or idiopathic. The HSP identified by ML-30 and CA-Str7-1 were constitutively expressed, with a regional distribution, by duct and ductular epithelial cells in all normal pancreatic tissues. The HSP detected by ML-30 were expressed uniformly throughout the cytoplasm of the majority of ductular epithelial cells in all cases of chronic pancreatitis, irrespective of suspected etiology, and in the ducts of all but two cases of alcohol-associated chronic pancreatitis. The HSP defined by CA-Str7-1 were identified in the majority of duct or ductular epithelial cells in most of the cases of chronic pancreatitis, although possible differential expression was observed with respect to etiology (ie, there seemed to be less HSP in cases of postalcohol pancreatitis). The HSP identified by TB-78 were not constitutively expressed by normal pancreatic tissues but were found in a few scattered epithelial cells in two of these cases. However, significant expression of these HSP were observed in most epithelial cells in a population of cases regarded as either alcohol-associated (0.05 > p > 0.02) or postacute chronic pancreatitis (0.01 > p > 0.001). The HSP identified by monoclonal antibody MAB 72/73 were either not demonstrable or were expressed at very low levels in both the normal and inflamed pancreatic tissues. Expression of the four groups of HSP molecules appeared to be differentially regulated both in normal pancreatic and in chronic pancreatitis tissues. These differences in expression may indicate different functions in normal tissues, with either a protective or a pathogenic role for these proteins in the diseased state. Our current findings support the hypothesis that expression of certain HSP, particularly those identified by TB-78, may be involved in the pathogenesis of distinct subtypes of chronic pancreatitis. Our data do not suggest that HSP are the primary targets of immune-mediated cytotoxic activity; nevertheless, enhanced expression of these molecules by pancreatic ductular epithelial cells does provide an environment in which increased amounts of endogenous intracellular peptides may be transported to the cell surface, thereby becoming potential targets of immune-surveillance and cell-mediated cytotoxicity. Conversely, HSP may play a protective role in such a manner that selected groups of pancreatic ductular epithelial cells withstand cytotoxic damage of chemical, metabolic, or immune origin, and for significantly increased periods of time than they would otherwise; HSP thereby conserve a population of "reserve" epithelial cells from which pancreatic regeneration might occur. Identification of distinct pathogenic groups, defined according to precise immunohistochemical criteria, might provide the basis of a functional assessment and hence allow development of biologically appropriate strategies for managing individual patients with chronic pancreatitis.