Identification of claudin-4 as a marker highly overexpressed in both primary and metastatic prostate cancer.

In the quest for markers of expression and progression for prostate cancer (PCa), the majority of studies have focussed on molecular data exclusively from primary tumours. Although expression in metastases is inferred, a lack of correlation with secondary tumours potentially limits their applicability diagnostically and therapeutically. Molecular targets were identified by examining expression profiles of prostate cell lines using cDNA microarrays. Those genes identified were verified on PCa cell lines and tumour samples from both primary and secondary tumours using real-time RT-PCR, western blotting and immunohistochemistry. Claudin-4, coding for an integral membrane cell-junction protein, was the most significantly (P<0.00001) upregulated marker in both primary and metastatic tumour specimens compared with benign prostatic hyperplasia at both RNA and protein levels. In primary tumours, claudin-4 was more highly expressed in lower grade (Gleason 6) lesions than in higher grade (Gleason >or=7) cancers. Expression was prominent throughout metastases from a variety of secondary sites in fresh-frozen and formalin-fixed specimens from both androgen-intact and androgen-suppressed patients. As a result of its prominent expression in both primary and secondary PCas, together with its established role as a receptor for Clostridium perfringens enterotoxin, claudin-4 may be useful as a potential marker and therapeutic target for PCa metastases.

Silent genes in bacteria: the previously designated 'cryptic' ilvHI locus of 'Salmonella typhimurium LT2' is active in natural isolates.

Gene ilvG in Escherichia coli K-12 and ilvI in 'Salmonella typhimurium LT2' (S. enterica serotype Typhimurium, strain LT2) are inactive due to frameshift or nonsense mutations, respectively. These inactive genes have been suggested to be part of 'cryptic' genetic systems which are defined as being of long-term regulatory and evolutionary significance. We have shown that the nonsense mutation in ilvI is present only in derivatives of the laboratory strain 'S. typhimurium LT2'. All natural isolates of Salmonella examined have an arginine codon at the corresponding location of their ilvI sequences. Further, two randomly selected natural isolates of serotype Typhimurium are shown to each have an active ALS III isozyme. Our findings strongly suggest that the only Salmonella strains which lack a functional ilvHI locus are LT2 isolates. We suggest that the mutations leading to inactivation of both ilvI in 'S. typhimurium LT2' and ilvG in E. coli K-12 are more likely to have been acquired during laboratory storage and/or cultivation, rather than representing cryptic systems of gene regulation.