The phototrophic bacterium Rhodobacter sphaeroidesDSM 158 is able to reduce nitrate to nitrite by means of a periplasmic nitrate reductase which is induced by nitrate and is not repressed by ammonium or oxygen. Recently, a 6.8 kb PstI DNA fragment carrying the napABCgenes coding for this periplasmic nitrate-reducing system was cloned [Reyes, Roldán, Klipp, Castillo and Moreno-Vivián (1996) Mol. Microbiol. 19, 1307–1318]. Further sequence and genetic analyses of the DNA region upstream from the napABCgenes reveal the presence of four additional napgenes. All these R. sphaeroidesgenes seem to be organized into a napKEFDABCtranscriptional unit. In addition, a partial open reading frame similar to the Azorhizobium caulinodans yntCgene and the Escherichia coli yjcCand yhjKgenes is present upstream from this napgene cluster. The R. sphaeroides napKgene codes for a putative 6.3 kDa transmembrane protein which is not similar to known proteins and the napEgene codes for a 6.7 kDa transmembrane protein similar to the Thiosphaera pantotrophaNapE. The R. sphaeroides napFgene product is a 16.4 kDa protein with four cysteine clusters that probably bind four [4Fe-4S] centres. This iron–sulphur protein shows similarity to the NapF and NapG proteins of E. coliand Haemophilus influenzae.Finally, the napDgene product is a 9.4 kDa soluble protein which is also found in E. coliand T. pantotropha. The 5´ end of the naptranscript has been determined by primer extension, and a δ70-like promoter has been identified upstream from the napKgene. The same transcriptional start site is found for cells growing aerobically or anaerobically with nitrate. Different mutant strains carrying defined polar and non-polar insertions in each napgene were constructed. Characterization of these mutant strains demonstrates the participation of the napgene products in the periplasmic nitrate reduction in R. sphaeroides.
Present address: Nitrogen Fixation Laboratory, John Innes Centre, Norwich NR4 7UH, U.K.
Contributed equally to this work.
The nucleotide sequence data reported in this paper have been deposited in the EMBL, GenBank and DDBI Nucleotide Sequence Databases under accession number Z46806.