Open ocean and coastal strains of the N2-fixing cyanobacterium UCYN-A have distinct transcriptomes

Abstract

Decades of research on marine N2 fixation focused on Trichodesmium, which are generally free-living cyanobacteria, but in recent years the endosymbiotic cyanobacterium Candidatus Atelocyanobacterium thalassa (UCYN-A) has received increasing attention. However, few studies have shed light on the influence of the host versus the habitat on UCYN-A N2 fixation and overall metabolism. Here we compared transcriptomes from natural populations of UCYN-A from oligotrophic open-ocean versus nutrient-rich coastal waters, using a microarray that targets the full genomes of UCYN-A1 and UCYN-A2 and known genes for UCYN-A3. We found that UCYN-A2, usually regarded as adapted to coastal environments, was transcriptionally very active in the open ocean and appeared to be less impacted by habitat change than UCYN-A1. Moreover, for genes with 24 h periodic expression we observed strong but inverse correlations among UCYN-A1, A2, and A3 to oxygen and chlorophyll, which suggests distinct host-symbiont relationships. Across habitats and sublineages, genes for N2 fixation and energy production had high transcript levels, and, intriguingly, were among the minority of genes that kept the same schedule of diel expression. This might indicate different regulatory mechanisms for genes that are critical to the symbiosis for the exchange of nitrogen for carbon from the host. Our results underscore the importance of N2 fixation in UCYN-A symbioses across habitats, with consequences for community interactions and global biogeochemical cycles.

Fig 1. The heat map shows that expression is driven first by location (Stn. ALOHA, Scripps Pier, Tara SAtl), then by sublineage.

Each column is a specific sublineage at a location and is standardized (quantiles). Cell colors are the quantile for the median intensity of the genes in the pathway. Sample and pathway clusters with solid discs were strongly supported (Methods). In the Location bar, samples from South Atlantic Tara Oceans stations 76 (brown) and 78 (orange) are shown.

Fig 2. Shown are all 760 genes (n_UCYN-A1 = 368, n_UCYN-A2 = 392) that were detected at both Stn. ALOHA and Scripps Pier.

Axes are the standardized transcript levels (Methods) and the dashed reference lines indicate equal standardized transcript levels at both sites. Genes are above the reference lines which is consistent with the hypothesis that the coastal study favored the detection of highly transcribed UCYN-A genes but missed less transcribed genes (main text). Genes are colored by pathway and the legend includes in brackets the number of cross-site detected genes for UCYN-A1 and A2. Genes that had transcript levels in the top 10% at both sites for UCYN-A1 are named and shown as open triangles. Top genes for UCYN-A2 are shown with filled triangles, and names shared with UCYN-A1 indicate orthologs.

Fig 3. All detected genes at Stn. ALOHA and Scripps Pier are shown at the time of day when their transcript levels peaked.

For each habitat and sublineage, the histogram shows the time of day when each detected gene had its highest average transcript level (Methods). The y axes show the percentage of detected genes from the sublineage. Black arrows indicate genes important to nitrogen fixation (nif genes and ATP synthases) which always peaked near sunrise (D12).

Fig 4. Abundance of UCYN-A1 and A2 diel genes at Stn. ALOHA (blue) and Scripps Pier (green).

The x axis shows the number of diel genes detected from each pathway. For legibility only pathways with ≥3 diel genes are shown (450 of 459 total diel genes at either site).