Proteorhodopsins dominate the expression of phototrophic mechanisms in seasonal and dynamic marine picoplankton communities
- Published
- Accepted
- Subject Areas
- Biodiversity, Bioinformatics, Ecology, Microbiology, Biological Oceanography
- Keywords
- phototrophy, proteorhodopsin, metatranscriptome, metagenome, spectral tuning, marine, picoplankton, expression
- Copyright
- © 2018 Sieradzki et al.
- Licence
- This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Preprints) and either DOI or URL of the article must be cited.
- Cite this article
- 2018. Proteorhodopsins dominate the expression of phototrophic mechanisms in seasonal and dynamic marine picoplankton communities. PeerJ Preprints 6:e26950v2 https://doi.org/10.7287/peerj.preprints.26950v2
Abstract
The most abundant and ubiquitous microbes in the surface ocean use light as an energy source, capturing it via complex chlorophyll-based photosystems or simple retinal-based rhodopsins. Studies in various ocean regimes compared the abundance of these mechanisms, but few investigated their expression. Here we present the first full seasonal study of abundance and expression of light-harvesting mechanisms (proteorhodopsin, PR; aerobic anoxygenic photosynthesis, AAnP; and oxygenic photosynthesis, PSI) from deepsequenced metagenomes and metatranscriptomes of marine picoplankton (< 1 μm) at three coastal stations of the San Pedro Channel in the Pacific Ocean. We show that, regardless of season or sampling location, the most common phototrophic mechanism in metagenomes of this dynamic region was PR (present in 65-104% of the genomes as estimated by single-copy recA), followed by PSI (5-104%) and AAnP (5-32%). Furthermore, the normalized expression (RNA to DNA ratio) of PR genes was higher than that of oxygenic photosynthesis (average±standard deviation 26.2±8.4 vs. 11±9.7), and the expression of the AAnP marker gene was significantly lower than both mechanisms (0.013±0.02). We demonstrate that PR expression was dominated by the SAR11-cluster year-round, followed by other Alphaproteobacteria, unknown-environmental clusters and Gammaproteobacteria. This highly dynamic system further allowed us to identify a trend for PR spectral tuning, in which blue-absorbing PR genes dominate in areas with low chlorophyll-a concentrations (< 0.25 μgL-1). This suggests that PR phototrophy is not an accessory function but instead a central mechanism that can regulate photoheterotrophic population dynamics.
Author Comment
This is a revised preprint submission to PeerJ Preprints which addresses the reviewer comments.
Supplemental Information
Nutrients and Chlorophyll-A concentrations measured in samples collected in this study
All Chl-A measurements as well as nutrient measurements in October through April are courtesy of the Caron lab, USC. Values in bold are below the instrument detection limit.
Comparison between read recruitment to functional genes using a protein HMM vs. reciprocal blast
Total reads per sample for normalization purposes are indicated in the bottom row.
Placement of assembled proteorhodopsin ORFs in the MicRhoDE tree clusters indicates that locally assembled rhodopsins represent much of the Prd diversity
The number of reads (pooled from all samples) that mapped to the assembled ORFs from metatranscriptomes (MT) and metagenomes (MG) was significant compared to number of pooled reads mapped to the curated proteins from MicRhoDE.
Evenness of rhodopsins by clade across samples
Each bar represents the Shannon index of evenness based on relative abundance of rhodopsin clades by site and month. Evenness in metagenomes is denoted by full bars, and in metatranscriptomes by striped bars.
A comparison of phototrophic mechanisms expanding on figure 2 (main text)
Normalized gene abundance (A) and expression (B) of phototrophic mechanisms per sample
Normalized relative abundance of phototrophic mechanisms per sampling site expanding on figure 2 (main text)
Variability of phototrophic mechanisms per genome in metagenomes from the three sampling locations of the San Pedro Channel (n=4, shown in Figure 1: the Y-axis denotes the percentage of genomes with the particular phototrophic mechanism normalized to recA gene (see methods). psaA for oxygenic photosynthesis, prd for proteorhodopsin and pufM for aerobic anoxygenic photosynthesis.
Maximum-likelihood tree of PSI protein (psaA) sequences used to build the HMM and to recruit short reads with pplacer
Bacterial sequences appear in blue, viral in red, eukaryotic in green and assembled ORFs in black.
Read recruitment to functional genes using a Hidden Markov Model (HMM) vs. reciprocal blast
Comparison of relative abundance (number of reads recruited over total number of reads per sample) of functional genes using reads recruited by HMM or reciprocal blast in (A) metagenomes and (B) metatranscriptomes
Maximum-likelihood tree of PSII protein sequences (pufM and psbA) used to build the HMM and to recruit short reads with pplacer
pufM leaves are colored red and psbA leaves in black.
