Degradation by proteases Lon, Clp and HtrA, of Escherichia coli proteins aggregated in vivo by heat shock; HtrA protease action in vivo and in vitro

Thermally aggregated, endogenous proteins of Escheri-chia coli form a distinct fraction, denoted S, which is separable by sucrose-density-gradient centrifugation. It was shown earlier that DnaK, DnaJ, IbpA and IbpB heat-shock proteins are associated with the S fraction. Comparison of the rise and decay of the S fraction in mutants defective for heat-shock proteases Lon (La), Clp, HtrA (DegP, Do) and in wild-type strains made studies of proteolysis and the function of the heat-shock response possible in vivo. Different timing and the extent of action of particular proteases was revealed by the initial size and decay kinetics of the S fraction. The proteases Lon, Clp, and HtrA all participated in removal of the aggregated proteins. Mutation in the gene encoding ClpB caused the most prominent effect (47% stabilization of the S fraction). The correlation between the disappearance of the S fraction and proteolytic activity was supported by the result of the in vitro reaction. Approximately one third of the isolated S fraction was converted to trichloroacetic acid-soluble products by the purified HtrA protease. Mg²⁺ ions stimulated the reaction, in contrast to the reaction of the HtrA protease with casein. The digestion of the aggregated proteins, unlike the digestion of casein, by HtrA protease in vitro was inhibited by added DnaJ, which might reflect protection of the aggregated proteins in vivo by DnaJ from excessive degradation. One might expect that such an activity of DnaJ would promote denatured protein renaturation versus proteolysis. Moreover, among the aggregated proteins that are discernible by electrophoresis, none could be identified as being more susceptible than any other to HtrA degradation. The separation pattern of these proteins before and after the in vitro digestion did not show a difference corresponding to the loss of about 30% of constituting proteins. This was interpreted as recognition by the HtrA protease of a state of protein denaturation rather than specific amino acid sequences in particular proteins. We conclude that the fraction consisting of proteins heat- aggregated in vivo (i.e. the S fraction) contains endogenous substrates for the heat-shock proteases tested. Their use for in vitro reaction reveals information that is in some respects different from that obtained with exogenous substrates such as casein.