Microbial communities in surface waters of temperate seas are exposed to recurring annual seasonal variation in temperature and nutrient concentrations. To what extent bacterioplankton populations in natural communities alter their functional repertoire as a result of seasonal succession has not been thoroughly investigated. Here we use metatranscriptomics and leverage a comprehensive catalogue of metagenome-assembled genomes (MAGs) to follow gene expression in individual populations over a two-year time period at an offshore station in the Baltic Sea. We show that the collective expression of the MAGs changed in a consistent manner across seasons in the two years, forming clusters representing the four seasons, and that more than 80% of these displayed a recurring seasonal pattern. Furthermore, we found that the changes in expression could partly be explained by modulation of expression within the prokaryotic populations, since intra-population expression patterns also changed with season. Taken together, our results demonstrate how natural microbial populations alter their expression on the gene level, and how these changes drive large scale changes on both population and community level. This work aims to broaden the understanding of how microbes respond and adapt to their environment by preferentially altering their expressed genetic repertoire, and how microbial community dynamics can be explained through the gene expression of various populations constituting the community. 

Aphanizomenon, together with Dolichospermum and Nodularia, constitute the major genera of bloom forming filamentous nitrogen fixing cyanobacteria in the Baltic Sea. Like the other genera, Aphanizomenon displays summer blooms that are highly variable in magnitude and duration but unlike the others it is considered a holoplanktonic species. Still, the molecular mechanisms enabling Aphanizomenon year-round presence in surface waters are currently unknown. Here we combine analysis of Aphanizomenon population dynamics at the Linnaeus Microbial Observatory (LMO) station in the Baltic Proper over nine years (2011-2019) with associated gene expression patterns during 2016-2017 to identify annual abundance, and metabolic and life cycle strategies. Aphanizomenon biomass showed large annual variability and a consistent biovolume peak in summer, with bloom intensity ranging from 78-1334 mm3 m-3. 16S rRNA gene amplicon sequence data showed that one Aphanizomenon amplicon sequence variant (ASV) dominated, and its relative abundance correlated with biovolume measurements. Metatranscriptomic reads that mapped to an Aphanizomenon metagenome- assembled genome (MAG) revealed annually repeating gene expression patterns, resulting in distinct gene expression profiles during different meteorological seasons. Genes encoding proteins involved in several important functional classes, e.g. carbon fixation, photosynthesis, and associated photopigments showed seasonal variation, but were detected year round. Other genes, particularly those involved in nitrogen fixation, were highly expressed in summer, while absent in winter. Vitamin metabolism and phosphorus scavenging genes were preferentially expressed during the colder periods of the annual cycle. Together, these data show that Aphanizomenon regulates the molecular machinery on the seasonal scale, providing context to the observed dynamics of Aphanizomenon in the Baltic Proper and a foundation for understanding the ecology of these cyanobacteria. 

The distribution of prokaryotic metabolism between maintenance and growth activities has a profound impact on the transformation of carbon substrates to either biomass or CO₂. Knowledge of key factors influencing prokaryotic maintenance respiration is, however, highly limited. This mesocosm study validated the significance of prokaryotic maintenance respiration by mimicking temperature and nutrients within levels representative of winter and summer conditions. A global range of growth efficiencies (0.05-0.57) and specific growth rates (0.06-2.7 d(-1)) were obtained. The field pattern of cell-specific respiration versus specific growth rate and the global relationship between growth efficiency and growth rate were reproduced. Maintenance respiration accounted for 75% and 15% of prokaryotic respiration corresponding to winter and summer conditions, respectively. Temperature and nutrients showed independent positive effects for all prokaryotic variables except abundance and cell-specific respiration. All treatments resulted in different taxonomic diversity, with specific populations of amplicon sequence variants associated with either maintenance or growth conditions. These results validate a significant relationship between specific growth and respiration rate under productive conditions and show that elevated prokaryotic maintenance respiration can occur under cold and oligotrophic conditions. The experimental design provides a tool for further study of prokaryotic energy metabolism under realistic conditions at the mesocosm scale.

Temperature and dissolved organic matter (DOM) are important drivers of marine microbial activity, but their effects, alone or in combination, on the physiological responses of sub-arctic prokaryotic assemblages remain poorly understood. In a one-month mesocosm experiment initiated in early March in the northern Baltic Sea, we thus exposed a coastal microbial community to temperature and nutrient regimes representative of winter and early summer (i.e., 1°C and 10°C, with and without DOM additions) in a 2x2 factorial design. Prokaryotic abundance and heterotrophic production increased until around day 17 in the 10°C mesocosms. Yet, mid through the experiment (days 10 and 17, when samples for metatranscriptomics analyses were analyzed), estimates of growth rates were highest for the 10°C plus DOM treatment (TN; ~2.5 day-1), comparable for the 1°C plus DOM (N) and the 10°C treatments (T; ~1.0 day-1), and low for the control (C; 0.2 day-1). PCA analysis showed that samples for prokaryotic transcription in the 1°C plus DOM and the 10°C treatments clustered in different directions from the control, and the combined 10°C plus DOM treatment triggered even further changes. Taxonomic analysis of the transcripts uncovered broad treatment specific responses. This included a dominance of Nitrosopumilus (Archaea) in the 1°C mesocosms (with and without DOM), an increase in the relative expression of Alphaproteobacteria (both Rhodobacterales and SAR11) in the 10°C mesocosms without DOM addition, and an increase in Oceanospirillales in the 10°C plus DOM treatment. Burkholderiales (Betaproteobacteria) maintained a high relative expression (up to 25%) in all mesocosms. A PERMANOVA on the total of 182,618 transcribed genes revealed statistically significant effects of both temperature and DOM, and also a significant interaction effect between the two factors. EdgeR analysis identified significant differential transcription for up to 10% of the genes in each of the tested contrasts. Prominent features among the significant genes included Nitrosopumilus genes for ammonium uptake and ammonia oxidation in the 1°C mesocosms (C, N), membrane transporters for small organic acids in the N-treatment, genes for N and P assimilation along with molecular chaperones in the T-treatment, and dominance of Oceanospirillales genes for energy and growth metabolism in the TN treatment. These metatranscriptomics responses were associated with changes in ecologically important characteristics of the prokaryotic communities, such as growth rates and growth efficiency, providing clues to how successional changes in community composition and metabolism are induced by environmental conditions linked with seasonality.