Genomic Evidence for Simultaneous Optimization of Transcription and Translation through Codon Variants in the pmoCAB Operon of Type Ia Methanotrophs

Abstract

Understanding the interplay between genotype and phenotype is a fundamental goal of functional genomics. Methane oxidation is a microbial phenotype with global-scale significance as part of the carbon biogeochemical cycle and a sink for greenhouse gas. Microorganisms that oxidize methane (methanotrophs) are taxonomically diverse and widespread around the globe. In methanotrophic bacteria, enzymes in the methane oxidation metabolic module (KEGG module M00174, conversion of methane to formaldehyde) are encoded in four operons (pmoCAB, mmoXYZBCD, mxaFI, and xoxF). Recent reports have suggested that methanotrophs in Proteobacteria acquired methane monooxygenases through horizontal gene transfer. Here, we used a genomic meta-analysis to infer the transcriptional and translational advantages of coding sequences from the methane oxidation metabolic modules of different types of methanotrophs. By analyzing isolate and metagenome-assembled genomes from phylogenetically and geographically diverse sources, we detected an anomalous nucleotide composition bias in the coding sequences of particulate methane monooxygenase genes (pmoCAB) from type Ia methanotrophs. We found that this nucleotide bias increases the level of codon bias by decreasing the GC content in the third base of codons, a strategy that contrasts with that of other coding sequences in the module. Further codon usage analyses uncovered that codon variants of the type Ia pmoCAB coding sequences deviate from the genomic signature to match ribosomal protein-coding sequences. Subsequently, computation of transcription and translation metrics revealed that the pmoCAB coding sequences of type Ia methanotrophs optimize the usage of codon variants to maximize translation efficiency and accuracy, while minimizing the synthesis cost of transcripts and proteins.IMPORTANCE Microbial methane oxidation plays a fundamental role in the biogeochemical cycle of Earth's system. Recent reports have provided evidence for the acquisition of methane monooxygenases by horizontal gene transfer in methane-oxidizing bacteria from different environments, but how evolution has shaped the coding sequences to execute methanotrophy efficiently remains unexplored. In this work, we provide genomic evidence that among the different types of methanotrophs, type Ia methanotrophs possess a unique coding sequence of the pmoCAB operon that is under positive selection for optimal resource allocation and efficient synthesis of transcripts and proteins. This adaptive trait possibly enables type Ia methanotrophs to respond robustly to fluctuating methane availability and explains their global prevalence.

Keywords: codon usage; methane monooxygenase; resource allocation; synthesis cost; translation efficiency.

FIG 1

Nucleotide composition bias of pmoCAB coding sequences. An extended analysis of all coding sequences is shown in Fig. S1c.

FIG 2

Analysis of the codon usage bias. (a) Analysis of the codon adaptation index (CAI). CAI indexes were converted to percentile ranks based on the relative distribution of CAIs in each methanotroph. Dashed lines with various slopes delineate the variations between the percentile ranks of both indexes. (b) Analysis of the distribution of the effective numbers of codons (ENC) of coding sequences from type Ia methanotrophs. The two-tailed Wilcoxon signed-rank test was applied to test the difference of ENC values between the pmoCAB and other coding sequences. A nonsignificant result of the tests is denoted by ns (P > 0.01). **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001. (c) Analysis of the ENC as a function of GC₃ content, with the line representing the fit of a linear model between the ENC and the GC₃ content.

FIG 3

Strategies for optimal translation. (a) The tRNA adaptation index (tAI) was used to estimate the translation efficiencies of coding sequences. tAI values were converted to percentile ranks based on the relative distribution of tAIs in each methanotroph. Coding sequences were ordered by the mean of the tAI percentile rank, and the two-tailed t test was applied to test whether the percentile ranks of tAIs are significantly different between the pmoC coding sequence (highest mean) and other coding sequence. ns, P > 0.01; ***, P ≤ 0.001. (b) Network of the interaction between codons and tRNA in type Ia methanotrophs. The two toy examples illustrate the expected networks when pmoCAB coding sequences have a codon usage bias that grants an advantage (marked by an arrow) in the competition for the tRNA pool (I) or does not grant any advantage (II). The observed network for threonine is shown in panel III, and all other amino acids are shown in Fig. S5a. (c) Quantitative analysis of the codon-tRNA interaction network. The figure shows how many copies of tRNA a gene can access according to an RSCU cutoff varying from 0.0 to 2.0, with a step size of 0.01. The reference line at an RSCU of 1.0 indicates the change from the region of nonpreferred codon usage (RSCU < 1.0) to the region of high preference for individual codons (RSCU > 1.0). The colored lines show the fit of a polynomial surface using local fitting. The chi-square test of proportions was applied to test whether the proportions of tRNAs available to coding sequences in the methane oxidation metabolic module are different between an RSCU of 0 and an RSCU of 2. (d) Median usage of each amino acid in the translated coding sequence as a function of the median RSCU value of the codon exhibiting the highest preference for each amino acid for type Ia methanotrophs. The P value corresponds to the significance of the linear regression. (e) Usage of prebiotic (inexpensive) and modern (expensive) amino acids in the proteins of the methane oxidation metabolic module of type Ia methanotrophs. The t test was used to evaluate the difference between the two groups.

FIG 4

Strategies for optimal transcription in type Ia methanotrophs. (a, b) Elemental compositions of transcribed coding sequences in M. tundripaludum 31/32. The main panel shows the number of element atoms per codon in each transcribed coding sequence, with the dashed lines representing the mean of all transcribed coding sequences in the genome. The top and right panels show the mean number of element atoms per codon of the pmoCAB transcribed coding sequences and the mean number of element atoms per codon of 1,000 random combinations of three coding sequences. (c) Summary of the carbon and oxygen atom demands per codon of transcribed coding sequences. The difference (Δ) was calculated using the genome-scale mean demand as the reference (dashed line). M. buryatense 5G possesses genes encoding both the pMMO and the sMMO. An extended analysis for each methanotroph is shown in Fig. S6.

FIG 5

Three dimensions of molecular optimization encoded in the coding sequences of the methane oxidation metabolic module of type Ia methanotrophs.