Comparative analysis of the conventional and novel pmo (particulate methane monooxygenase) operons from methylocystis strain SC2

Abstract

In addition to the conventional pmoA gene (pmoA1) encoding the active site polypeptide of particulate methane monooxygenase, a novel pmoA gene copy (pmoA2) is widely distributed among type II methanotrophs (methane-oxidizing bacteria [MOB]) (M. Tchawa Yimga, P. F. Dunfield, P. Ricke, J. Heyer, and W. Liesack, Appl. Environ. Microbiol. 69:5593-5602, 2003). Here we report that the pmoA1 and pmoA2 gene copies in the type II MOB Methylocystis strain SC2 are each part of a complete pmoCAB gene cluster (pmoCAB1, pmoCAB2). A bacterial artificial chromosome (BAC) library of strain SC2 genomic DNA was constructed, and BAC clones carrying either pmoCAB1 or pmoCAB2 were identified. Comparative sequence analysis showed that these two gene clusters exhibit low levels of identity at both the DNA level (67.4 to 70.9%) and the derived protein level (59.3 to 65.6%). In contrast, the secondary structures predicted for PmoCAB1 and PmoCAB2, as well as the derived transmembrane-spanning regions, are nearly identical. This suggests that PmoCAB2 is, like PmoCAB1, a highly hydrophobic, membrane-associated protein. A total of 190 of the 203 amino acid residues representing a highly conserved consensus sequence of the currently known PmoCAB1 and AmoCAB sequence types could be identified in PmoCAB2. The amoCAB gene cluster encodes ammonia monooxygenase and is evolutionarily related to pmoCAB. Analysis of a set of amino acid residues that allowed differentiation between conventional PmoA and AmoA provided further support for the hypothesis that pmoCAB2 encodes a functional equivalent of PmoCAB1. In experiments in which we used 5' rapid amplification of cDNA ends we identified transcriptional start sites 320 and 177 bp upstream of pmoC1 and pmoC2, respectively. Immediately upstream of the transcriptional start sites of both pmoCAB1 and pmoCAB2, sequence motifs similar to Escherichia coli sigma(70) promoters were identified.

FIG. 1.

Sequence alignment of experimentally determined (Methylocystis sp. strain M, pmoCAB1 and pmoCAB2 of Methylocystis strain SC2) and computationally predicted (M. trichosporium OB3b) promoter regions of pmo operons. The alignment was computed by using a ClustalW algorithm. The shading indicates the −35 and −10 sequence motifs. Nucleotides identical to the corresponding nucleotides of the E. coli σ⁷⁰ promoter consensus sequences (−35 sequence, TTGACA; −10 sequence, TATAAT) are indicated by boldface type, and the putative transcriptional start sites (positions A1 and C1) are underlined.

FIG. 2.

Primary structures of derived PmoCAB2 expressed in one-letter code. Predicted transmembrane helices are indicated by shading. Residues that are highly conserved in PmoCAB1 and AmoCAB (consensus sequence) are indicated by boldface type (37). Amino acids that are located at conserved positions but differ from the amino acids in the consensus sequence determined by Tukhavatullin et al. (37) are underlined. The N-terminal helix of PmoB, as predicted by toppred and shown in Fig. 3, is not indicated because the residues are thought to constitute a leader sequence (26, 37).

FIG. 3.

Predicted topologies of derived PmoCAB1 and PmoCAB2 from Methylocystis strain SC2. (Top) Hydrophobicity plot of PmoCAB1 versus PmoCAB2. The y axis indicates the relative hydrophobicity value at a given position in PmoCAB. (Bottom) Schematic presentation of predicted transmembrane helices of PmoCAB2. N-terminal sequences are predicted to be located in the cytosol. Exact locations of transmembrane-spanning regions are shown in Fig. 2.