Molecular analysis of the Rhodobacter capsulatus B10 porin (porCa) gene; Purification and biochemical characterisation of the porin protein

The pore-forming outer-membrane protein from Rhodobacter (R.) capsulatus (wild-type B10 strain) was isolated and purified under non-denaturing conditions. The monomer unit of the isolated porin has a molecular mass of about 28 kDa, as judged by SDS-PAGE, whereas the native protein migrates at 75 kDa. This suggests that the native porin from R. capsulatus B10 exists in a trimeric form. The N-terminal amino acid sequence was used to design an oligonucleotide which was utilised to screen a pBluescript library containing EcoRI fragments of R. capsulatus B10 DNA. A 5.3-kb DNA fragment, which included the entire structural porin gene (named porCa) and its flanking regions was identified. A 945-bp open reading frame, coding for a mature protein of 295 amino acid residues (molecular mass 30 586 Da) plus a presequence of 20 amino acids, was found. The directly determined sequence of the amino-terminus and of four tryptic peptides of the purified porin matched perfectly with the deduced amino acid sequence. Northern blot analysis showed that the porin gene encodes an RNA transcript of 1050 nucleotides. In addition, there is no differential response in terms of either the size or abundance of the mRNA under different environmental conditions. Primer extension experiments confirmed a putative promoter upstream of the porin gene, and localised the RNA transcription start site 73 bp upstream of the ATG start codon, which is close to the putative promoter (-10/-35). As shown using a plasmid-borne porin-lacZ gene fusion, [in R. capsulatus] expression of the lacZ gene under the control of the pol Ca promoter is not regulated under the two different environmental conditions tested. The promoter of the porin gene was localised within 305 bp upstream of the ATG start codon. A model of the 3-D structure of porin B10 was deduced by comparative modelling with the R. capsulatus 37b4 and Rhodopseudomonas blastica porin crystallographic structures using the ProMod program on the Swiss-Model protein modelling e-mail Server. Analysis of the B10 sequence and comparison of a model of the B10 porin structure with the crystallographic structure of porin from the capsuleless strain 37B4 has revealed some important differences at the level of the protein surface, the pore and the putative ligand binding site.