Lon protease homologues contain a poorly conserved N-terminal region of variable length. To better understand the role of the N-terminal region of Lon in the complicated reaction cycle of ATP-dependent protein degradation, we expressed and characterized mutants of the Lon protease from Mycobacterium smegmatis (Ms-Lon) lacking 90, 225, and 277 N-terminal residues (N-G91, N-E226, and N-I278, respectively). N-I278 displayed neither peptidase nor ATPase activity despite the fact that it was stable and soluble in vivo, had a near-wild-type CD spectrum, and the deleted residues included neither the catalytic nucleophile for peptide bond hydrolysis (S675) nor the ATP binding regions. N-G91 and N-E226 retained peptidase activities against small unstructured peptides that were stimulated, to near-wild-type levels, by the Ms-Lon substrate protein alpha-casein. By contrast, N-G91 and N-E226 retained basal ATPase activities, but these activities were only stimulated weakly by alpha-casein. Ms-Lon, N-E226, and N-G91 all exhibited low-level peptidase activity in assays containing nonhydrolyzed nucleotide analogues. However, these peptidase activities were stimulated strongly by alpha-casein in the case of Ms-Lon but weakly by alpha-casein in the cases of N-G91 and N-E226. Strikingly, despite the near-wild-type peptidase activities of N-G91 and N-E226, both were severely impaired in their degradation of the Ms-Lon protein substrates alpha-casein in vitro and RcsA in vivo. Overall, N-G91 and N-E226 displayed catalytic properties similar to Escherichia coli Lon (Ec-Lon) in the presence of the PinA inhibitor, suggesting that PinA inhibits Ec-Lon protease by inhibiting the function of Ec-Lon's N-terminal region. In vivo protease assays further revealed that, in contrast to the inactive Ms-Lon point mutant S675A, N-G91 and N-E226 did not reduce the cellular activity of RcsA. This same defect was observed previously for Ms-Lons with multiple mutations in their peptidase active sites. We conclude that proteolytically inactive mutants of Ms-Lon retain the ability to reduce the cellular activity of RcsA but that both the N-terminal region and the peptidase active site region of Ms-Lon are required for this activity of wild-type Ms-Lon. The inabilities of N-G91 and N-E226 to degrade larger protein substrates and to reduce the cellular activity of RcsA were not the result of drastic alterations in their quaternary structures. Gel filtration profiles of N-G91 and N-E226 revealed that each was primarily tetrameric, with an increased percentage of dimeric species and a decreased percentage of trimeric species relative to Ms-Lon. The observed shifts in the dimer/trimer ratios of the N-terminal truncation mutants suggest that the Ms-Lon tetramer contains two types of subunit-subunit interactions.

We have charterized a Mycobacterium smegmatis gene encoding a homolog of the ATP-dependent protease Lon (La). Our identification of a Lon homolog, in conjunction with our previous work, identifies M. smegmatis as the first known example of a eubacterium containing both Lon and a complete 20S proteasome (containing both alpha- and beta-subunits). Despite the significant primary sequence divergence between M. smegmatis Lon (Ms-Lon) and E. coli Lon (Ec-Lon), expression of Ms-Lon was only moderately toxic to E. coli cells. The ability of E. coli cells to tolerate expression of Ms-Lon reveals that Ms-Lon does not recognize and degrade essential E. coli proteins. We conclude that discrimination against nonsubstrate proteins is broadly conserved between Ec-Lon and Ms-Lon. Additional conservation of substrate recognition was demonstrated by the ability of Ms-Lon to degrade efficiently RcsA, a natural substrate of Ec-Lon. Purified Ms-Lon displays chymotrypsin-like specificity in peptidase assays that are stimulated by unfolded protein and supported by nonhydrolyzed nucleotide analogs. Maximal peptidase activity requires ATP or dATP. Replacement of Ms-Lon's catalytic Ser with Ala (S675A), Thr (S675T), or Cys (S675C) reduced to background levels Ms-Lon's in vitro peptidase activity. However, by employing a sensitive in vivo assay, based on the degradation of RcsA, we demonstrated that the S675C variant retained specific protease activity. Finally, variants of Ms-Lon, with substututions at or near S675, reduce the enzyme's basal ATPase activity, suggesting a structural interaction between the peptidase and ATPase active sites of Ms-Lon.

No abstract