First genome data from uncultured upland soil cluster alpha methanotrophs provide further evidence for a close phylogenetic relationship to Methylocapsa acidiphila B2 and for high-affinity methanotrophy involving particulate methane monooxygenase

Abstract

Members of upland soil cluster alpha (USC alpha) are assumed to be methanotrophic bacteria (MB) adapted to the trace level of atmospheric methane. So far, these MB have eluded all cultivation attempts. While the 16S rRNA phylogeny of USC alpha members is still not known, phylogenies constructed for the active-site polypeptide (encoded by pmoA) of particulate methane monooxygenase (pMMO) placed USC alpha next to the alphaproteobacterial Methylocapsa acidiphila B2. To assess whether the pmoA tree reflects the evolutionary identity of USC alpha, a 42-kb genomic contig of a USC alpha representative was obtained from acidic forest soil by screening a metagenomic fosmid library of 250,000 clones using pmoA-targeted PCR. For comparison, a 101-kb genomic contig from M. acidiphila was analyzed, including the pmo operon. The following three lines of evidence confirmed a close phylogenetic relationship between USC alpha and M. acidiphila: (i) tetranucleotide frequency patterns of 5-kb genomic subfragments, (ii) annotation and comparative analysis of the genomic fragments against all completely sequenced genomes available in public domain databases, and (iii) three single gene phylogenies constructed using the deduced amino acid sequences of a putative prephenate dehydratase, a staphylococcal-like nuclease, and a putative zinc metalloprotease. A comparative analysis of the pmo operons of USC alpha and M. acidiphila corroborated previous reports that both the pmo operon structure and the predicted secondary structure of deduced pMMO are highly conserved among all MB.

FIG. 1.

Genome signature-based phylogenetic map calculated for 5-kb subfragments of all publicly available bacterial genome sequences (only the relevant section of the total bacterial map is shown). Symbols (×) indicate the positions calculated for the 5-kb genomic subfragments of USCα (black), M. acidiphila B2 (white), and Methylosinus/Methylocystis spp. (red). Single data points may be indicative of several genomic subfragments that were mapped at the same position. The green arrow points to the position of a genomic subfragment of USCα that contains ORF16. The black arrows indicate the positions of genomic subfragments of Methylosinus/Methylocystis spp. that were placed outside of the alphaproteobacterial map area.

FIG. 2.

Schematic overview of gene arrangements in a 101-kb fragment of Methylocapsa acidiphila B2, the complete genomes of Beijerinckia japonicum and Rhodopseudomonas palustris, and a 42-kb fragment of USCα. Corresponding genomic regions are linked by grayish interconnections, except for those between USCα and M. acidiphila B2. These are highlighted in yellow. The ORF numbering corresponds to that obtained by the automatic ORF prediction program ORPHEUS (20). Color code: green, pmoCAB and, only on the USCα fragment, the associated ORF45 (OrfD [21]); blue (only on the M. acidiphila B2 fragment), genes encoding enzymes involved in H₄MPT/MFR-mediated C₁ metabolism (57); red, genes encoding other proteins with assigned functions; white, genes encoding hypothetical proteins; gray, genes encoding conserved hypothetical proteins.

FIG.3.

Phylogenetic dendrograms constructed for the deduced amino acid sequences of ORFs 42, 54, and 56 (Table 1). The shaded boxes indicate the highly similar branching patterns observed between the three phylogenies for USCα, M. acidiphila B2, B. japonicum, and R. palustris. α indicates the branch leading to the alphaproteobacterial clade. The numbers of amino acid positions used for tree construction were 220 (ORF42), 135 (ORF54), and 324 (ORF56). Phylogenies were constructed using a maximum frequency filter of 25% (ORFs 42 and 54) or 35% (ORF56). The dendrograms are consensus trees of phylogenies constructed using different approaches, including distance-based and position-specific algorithms (neighbor joining, Treepuzzle, and ProtML). An exception is the ORF42 gene homolog that was obtained from M. palustris K. Due to its shorter length, its deduced amino acid sequence was inserted into the tree according to maximum parsimony criteria, without allowing changes in the existing tree topology, using the appropriate function of the ARB software. Bar, 0.1 substitutions per amino acid position.

FIG. 4.

Phylogenetic dendrogram constructed for the deduced amino acid sequences of concatenated pmoCAB/amoCAB sequences. The tree was derived by maximum likelihood analysis but was also confirmed by TreePuzzle and neighbor-joining methods. Bar, 0.1 substitutions per amino acid position.