Diatom-Associated Bacteria Report

Exploring the diversity, abundance, and variability of diatom-associated bacteria in the oligotrophic ocean

I. Abstract
The ecology of diatoms may be better explained by conceptualizing them as composite organisms consisting of the host cell and its bacterial associates. Our previous investigated diatom-bacterial interactions at the single-cell level found that bacterial assemblages varied substantially even among closely related individual host cells. The bacterial assemblages associated with single cells could be separated into three distinct groups, but these groups occurred irrespective of host cell identity. Instead, the distinct groups were best explained by strong interactions among host-associated bacteria; for example, in one group,
a single bacterial genus (Arthrobacter) occurred to the near-exclusion of any other bacteria on the same host cell. This led to questioning whether diatom-bacterial associations were persistent; how are these associations affected by abiotic (e.g. nutrients) or biotic (e.g. host-bacteria or bacteria-bacteria interactions) factors. To test the effect of various stressors on host-bacterial interactions, a multi-factor experiment was designed to examine changes in the phylogenetic composition of Chaetoceros-associated bacteria. How the host-associated bacteria responded to changes in the host growth state, induced by manipulations of nutrient concentration and viral infection, was monitored using 16S rDNA. Marinobacter and Alteromonas phylotypes dominated the bacterial consortia attached to a Chaetoceros host in all treatments, regardless of host growth stage, but nutrient concentration and host growth stage had a statistically significant effect on the phylogenetic composition of the attached bacteria. We again found evidence that interactions between attached bacteria affect the composition of host-associated bacterial consortia. These results led to the exploration of whether manipulations of the relative proportions of co-cultured bacteria could affect a xenic diatom host.
Multiple strains of Alteromonas and Marinobacter were isolated from the same Chaetoceros host. Individual strains were added to three different xenic diatom hosts (the origin host, a naïve Chaetoceros host, and a naïve Amphipora host), to evaluate whether perturbations in their bacterial consortia could affect host growth, carrying capacity, and decline. Additionally, inoculations were repeated in vitamin-rich and vitamin-poor media to test whether the added bacteria provided B-vitamins to their host. Manipulating the bacterial consortia had a strong effect on the naïve Chaetoceros host cell, but minimal effects were observed for the origin Chaetoceros host or the more distantly related Amphipora host. These results demonstrate that the relationship may differ between congeners, and that host-associated bacterial consortia may have limited resilience or resistance to perturbation. Finally, a metagenomic analysis of the original amplified genomes of single diatom cells plus their associated bacteria was performed to gain a better understanding of the functional capabilities contributed by bacteria to a diatom-bacterial association in nature. Six diatom cells derived from two of the three distinct
groups mentioned above were selected for analysis. Their bacterial associates vary in phylogenetic diversity and composition. Genes of particular interest may be essential to the diatom-bacterial interaction, including nutrient pathways, antibiotic production, quorum sensing, motility, and others. Similarities between the functional capacities of the bacterial consortia found on Thalassiosira cells will also be compared to other single-cell isolates as well as particle-associated bacteria. The overarching objective is to gain insight into how bacteria may contribute to the health, success, or failure of their diatom host in marine systems.

II. Introduction
Diatom health and growth
Diatoms are a vital part of the oceanic carbon cycle and are responsible for nearly 40% of total marine primary productivity.
The carbon fixed by oceanic diatoms is equivalent to the organic carbon produced by all of the terrestrial rainforests combined (Nelson et al. 1995). Diatoms not only generate organic carbon from carbon dioxide, but also play a major role in the ‘biological pump’, wherein nutrients (N, P, Fe, Si, trace metals) are taken up in the euphotic zone and sink to the benthos as biological material incorporated into fecal pellets or marine snow. Nearly half of the sinking organic carbon produced by diatoms is consumed by bacteria and remineralized into the upper ocean as inorganic nutrients, a process referred to as the microbial loop (Azam et al. 1983). Microbial remineralization is essential to maintain nutrients in this system (Williams and Yentsch 1976; Cole 1982), including silica, which otherwise may limit diatom growth (Bidle et al.
2003).

Like all algae, diatoms thrive when they have suitable environmental conditions. Diatoms are most active in the euphotic zone, which at station ALOHA typically extends to 200m, with light levels of 10-100 µE m-2s-1 and temperatures between 17-25 °C (Cortés et al. 2001). Diatoms are thought to be relatively resilient to changing light levels (Mitrovic et al. 2003), however their ability to thrive under varying light depends on other factors including salinity and temperature (Miller and Kamykowski 1986). Nutrient requirements depend on the specific nutrient and are highly variable among diatom species; the ranges compared to the nutrient concentrations commonly seen at station ALOHA can be seen in Table 1 below. Maximum growth rates are also highly variable among diatom species, ranging from 0.2 d-1 to 3.3 d-1 with an average of 1.5 ± 0.8 d-1 (n=67). The most likely cause of this variation is diatom cell volume, which can be from 13 µm3 to 7x105 µm3 (Sarthou et al. 2005). Larger cells have a lower surface to volume ratio than smaller cells of the same morphology, thereby limiting diffusion of nutrients into the larger cells.