Diel transcriptional response of a California Current plankton microbiome to light, low iron, and enduring viral infection

Phytoplankton and associated microbial communities provide organic carbon to oceanic food webs and drive ecosystem dynamics. However, capturing those dynamics is challenging. Here, an in situ, semi-Lagrangian, robotic sampler profiled pelagic microbes at 4 h intervals over ~2.6 days in North Pacific high-nutrient, low-chlorophyll waters. We report on the community structure and transcriptional dynamics of microbes in an operationally large size class (>5 μm) predominantly populated by dinoflagellates, ciliates, haptophytes, pelagophytes, diatoms, cyanobacteria (chiefly Synechococcus), prasinophytes (chiefly Ostreococcus), fungi, archaea, and proteobacteria. Apart from fungi and archaea, all groups exhibited 24-h periodicity in some transcripts, but larger portions of the transcriptome oscillated in phototrophs. Periodic photosynthesis-related transcripts exhibited a temporal cascade across the morning hours, conserved across diverse phototrophic lineages. Pronounced silica:nitrate drawdown, a high flavodoxin to ferredoxin transcript ratio, and elevated expression of other Fe-stress markers indicated Fe-limitation. Fe-stress markers peaked during a photoperiodically adaptive time window that could modulate phytoplankton response to seasonal Fe-limitation. Remarkably, we observed viruses that infect the majority of abundant taxa, often with total transcriptional activity synchronized with putative hosts. Taken together, these data reveal a microbial plankton community that is shaped by recycled production and tightly controlled by Fe-limitation and viral activity.

. Ab initio ORF predictions were called on assembled large fraction contigs and directly on small fraction ORFs due to the longer read length, lower coverage nature of 454 sequencing. This less-restrictive amino acid space approach allowed us to map 7x more reads than traditional nucleotide space mapping to known references. Still, despite mapping 107 million reads, 158 million reads could not be mapped to ab initio ORFs, and those that did only averaged 67.2% identity to their best BLAST hit. Transcriptomes of reference organisms were chosen based on similarity and abundance of closely related species. Large and small fraction reads were mapped to reference transcriptomes using nucleotide Burrows Wheeler Aligner (BWA; (2). Reference transcriptomes were hierarchically clustered together with large and small fraction ORFs to gene ortholog groups. The resultant clusters, reference transcriptomes, and de-novo ORFs from both fractions were annotated taxonomically and functionally and used for downstream analyses, including pattern recognition algorithms such as Harmonic Regression Analysis (HRA) and Weighted Gene Network Correlation Analysis (WGCNA).  Orange: modeled total Fe < .04 umol/m3 based on global oceanographic data (4) Yellow: June 1996/ June 1997, measured DPSCSV reactive Fe ≤ 0.1 nM, total Fe ≤ 0.1 nM, Feenrichment experiments confirm moderate to severe Fe limitation (5) Green: Monterey Bay moorings M1 and M2, seasonal Fe limitation documented (e.g. June 1999/August 1999, measured total Fe < 1 nM) (6) Blue: September-October 2009, total Fe < 1nM for at least 1 depth at given coordinates (7) Figure S5 Expression of major nutrient cycling genes across size classes. Pies represent annotated functional clusters of ab initio ORFs and are colored by relative taxonomic contribution. The biogeochemical pathway each cluster participates in is noted in blue; asterisks denote ORFs previously observed to be transcriptionally sensitive to iron limitation. Clusters are grouped by modules of similar expression as given by WGCNA.

Figure S7
Average nucleotide percent identity of large fraction reads mapping to reference transcriptomes in the large (orange) and small (blue) size classes.

Figure S8
Comparison of average percent identity of reads mapping to reference transcriptomes in nucleotide space (red) and reads mapping to ab initio ORFs in amino acid space (blue).    S12 Total library (time point) normalized activity of large fraction genera that exhibit significant 24-h periodicity (HRA; FDR p ≤ 0.1). Two photosynthetic eukaryotes, Pelagodinium, a photosynthetic dinoflagellate symbiotic with foraminifera (8), and the centric diatom Skeletonema, had peak activity during the day. The remaining genera were non-photosynthetic bacteria with aggregate gene expression peaking at night: Loktanella, Mesoflavibacter, Oceanibulbus, Pseudovibrio, Roseobacter, Roseovarius, Tenacibaculum, and Unclassified candidate division WWE1. Several are known phytoplankton associates (e.g. Loktanella spp. (9,10)) and early particle colonizers (11) not previously known to operate on a diel cycle.

Figure S13
A comparison of functional diversity across fractions by mapping reads to transcriptomes of cultured representatives. Pies represent most abundant functional clusters of reference ORFs. Pies are colored by relative taxonomic contribution and grouped by modules of similar expression as given by WGCNA. Note that reads mapping to Favella ehrengbergii Strain Fehren 1 (e.g. those involved in photosynthesis) may be hitting remnants of its photosynthetic food source.

Figure S16
Peak expression time of large fraction ORFs involved in (A) metabolism, (B) signaling and nutrient transport, (C) transcription and (D) translation and protein synthesis. Night is indicated by grey shading. Significantly periodic ORFs (HRA; FDR adjusted p ≤ 0.1) are colored by functional annotation; insignificant ORFs are shown in grey. Several eukaryotic translation elongation factor 3 (eEF3) ab initio ORFs detected in the large size class were significantly periodic (dark red). eEF3 presents a novel peptide synthesis mechanism for phytoplankton. eEF3 was previously thought to be unique to fungi, but homologs have been recently discovered in various phytoplankton lineages, and one haptophyte (Phytopthora infestans) eEF3 was proven capable of restoring function in yeast (19).

Figure S17
Comparison of peak expression time of photosynthesis related ORFs between the current data and previously studied ESP drift tracks (1,20,21). Taxa groups are distinguished by shape (legend, top right) and radius (innermost: Prochlorococcus, outermost: "Other Photosynthetic Eukaryotes"). Colors indicate dataset of origin. Night (as observed for current data) is indicated by grey shading. In some cases, addition of picoplankton data from other environments revealed a difference in timing between prokaryotic and eukaryotic photosynthetic proteins. For example, cyanobacterial PSII, CP43/CP47, and PSII OEC peak earlier than their equivalents in photosynthetic eukaryotes, with Ostreococcus and other chlorophytes peaking last, and cyanobacterial FtsH and B6F peak earlier than equivalents in photosynthetic eukaryotes.

Figure S18
Continuation of Figure 6: virus/host dynamics in the large size class. Viruses and hosts are annotated as the closest reference available in our database, as determined by LPI. Library normalized expression of ORFs classified as ssRNA (yellow) and dsDNA (pink) viruses and their putative hosts by LPI are shown. Putative host expression is represented by solid lines and corresponds to left y-axes; virus expression is represented by dashed lines and corresponds to the right y-axes. Phaeocystis globulosa virus virophage expression was multiplied by 10 3 for better visualization. Night hours are shaded in grey.