Homology between the HrpO protein of Pseudomonas solanacearum and bacterial proteins implicated in a signal peptide-independent secretion mechanism

Abstract

A region of approximately 22 kb of DNA defines the large hrp gene cluster of strain GMI1000 of Pseudomonas solanacearum. The majority of mutants that map to this region have lost the ability to induce disease symptoms on tomato plants and are no longer able to elicit a hypersensitive reaction (HR) on tobacco, a non-host plant. In this study we present the complementation analysis and nucleotide sequence of a 4772 bp region of this hrp gene cluster. Three complete open reading frames (ORFs) are predicted within this region. The corresponding putative proteins, HrpN, HrpO and HpaP, have predicted sizes of 357, 690 and 197 amino acids, respectively, and predicted molecular weights of 38,607, 73,990 and 21,959 dalton, respectively. HrpN and HrpO are both predicted to be hydrophobic proteins with potential membrane-spanning domains and HpaP is rich in proline residues. A mutation in hpaP (for hrp associated) does not affect the HR on tobacco or the disease on tomato plants. None of the proteins is predicted to have an N-terminal signal sequence, which would have indicated that the proteins are exported. Considerable sequence similarities were found between HrpO and eight known or predicted prokaryotic proteins: LcrD of Yersinia pestis and Y. enterocolitica, FlbF of Caulobacter crescentus, FlhA of Bacillus subtilis, MxiA and VirH of Shigella flexneri, InvA of Salmonella typhimurium and HrpC2 of Xanthomonas campestris pv. vesicatoria. These homologies suggest that certain hrp genes of phytopathogenic bacteria code for components of a secretory system, which is related to the systems for secretion of flagellar proteins, Ipa proteins of Shigella flexneri and the Yersinia Yop proteins. Furthermore, these homologous proteins have the common feature of being implicated in a distinct secretory mechanism, which does not require the cleavage of a signal peptide. The sequence similarity between HrpO and HrpC2 is particularly high (66% identity and 81% similarity) and the amino acid sequence comparison between these two proteins presented here reveals the first such sequence similarity to be shown between Hrp proteins of P. solanacearum and X. campestris. An efflux of plant electrolytes was found to be associated with the interactions between P. solanacearum and both tomato and tobacco leaves. This phenomenon may be part of the mechanism by which hrp gene products control and determine plant-bacterial interactions, since hrpO mutants induced levels of leakage which were significantly lower than those induced by the wild type on each plant.