Low diversity of planktonic bacteria in the tropical ocean.

The planktonic bacteria in the oceans comprise one of the largest and most active microbial communities of the planet1. These bacteria contain the enzymatic machineries that drive the global biogeochemical fluxes of the elements for life2. They are powering the cycles of carbon, sulfur and nitrogen3,4,5, with important consequences for the composition of greenhouse gases in the atmosphere, algal blooms, and ocean acidification. In view of these global challenges, understanding bacterioplankton biogeography is of utmost importance. One of the key questions in biogeography is the pattern of diversity. In macro-organisms, diversity peaks in the tropics, an observation which dates back to Alexander von Humboldt6. It is the most universal biogeographic pattern on the planet and is found, with very few exceptions, across taxa, habitats, body sizes and functional groups of organisms7. It is similarly strong in marine as in terrestrial habitats, and has been demonstrated for marine phytoplankton, protozoa and zooplankton8. More than 30 hypotheses have been debated to explain this pattern9. The two factors thought to be key are productivity (the larger pie can be divided in more pieces) and temperature (the red queen runs faster when she is hot)6,10. Given the intimate symbiotic interactions between prokaryotes and eukaryotes and the importance of understanding the ecological consequences of global warming, it is mandatory to know if the concepts established for macro-organisms hold for bacteria as well. The first study to investigate this question found a decrease of species richness with latitude11. Then, an analysis of 103 samples taken around the world and using ARISA (Automated Ribosomal Intergenic Spacer Analysis) to type the community members also found a negative correlation between diversity and latitude; moreover, a positive relationship with temperature was observed, suggesting that global macro-ecological patterns hold for bacteria too and that indeed the kinetics of biological processes might have a strong influence on diversity6. Biogeographic studies in the ocean, which were traditionally hampered by low sampling depth (in relation to the huge dimensions of this ecosystem) and lack of taxonomic resolution, have profited enormously from next generation sequencing and large international sampling efforts (e.g. International Census of Marine Microbes (ICoMM), Census of Antarctic Marine Life (CAML)). It has now become possible to directly observe microbial diversity on a global scale with OTU level taxonomic resolution. However, regarding the pattern of alpha diversity, the results are conflicting. A study of 277 epipelagic samples12 covering a range from 74.4 °N to 75.6 °S found a negative correlation between species richness and latitude both in the Northern and the Southern hemisphere, which was week but significant and confirmed the above described investigation6. A comparable study13 using a smaller sample set but including also winter and summer samples did not find a latitudinal diversity gradient. To take advantage of the numerous studies that have been undertaken by different groups with different methodologies, a modelling approach was recently applied covering 377 marine samples from 164 locations with depths <150 m14. Species distribution modelling (SDM) was used to predict the global distribution of taxa by extrapolating from the available samples and environmental data. This approach robustly predicted (1) an inverted latitudinal diversity gradient and (2) an extreme seasonality of this gradient, such that diversity peaked in the higher latitudes of the Northern hemisphere in winter and in the higher latitudes of the Southern hemisphere in summer. The authors suggest that bacterioplankton biogeography follows different rules than those found for macro-organisms. Accordingly there was no overlap between diversity hotspots for bacteria and macro-organisms predicted in this study14. Thus, it is still controversial if bacterioplankton diversity peaks in the tropics or in higher latitudes. The few studies that were undertaken do not provide consistent results. To directly address this important question, here we use a spatially highly resolved sample set from the Atlantic Ocean. The samples were obtained during a 5 week transect (10 April – 15 May 2012) and span a geographical distance of ~12.000 km (51 °S to 47 °N) (Fig. 1). For each of the 26 stations, 5 epipelagic depths between 20–200 m were sampled, and each sample was divided into three size classes by sequential filtering. Altogether 359 samples were obtained. Deep sequencing of the V5–V6 region of the 16S rRNA gene was carried out using Illumina amplicon sequencing. Thus we can clearly demonstrate the global patterns of species richness of the free-living and particle associated planktonic bacteria in the Atlantic Ocean. We then investigated the correlation of microbial diversity with temperature and, as a proxy of biomass, bacterial cell numbers, to determine if the mechanisms acting on macro-organisms are also shaping diversity patterns in marine bacteria. Figure 1 Sampling stations of cruise ANT 28-5 across the Atlantic Ocean.