Meso-and bathypelagic ecosystems represent the most common marine ecological niche on Earth and contain complex communities of microorganisms that are for the most part ecophysiologically poorly characterized. Gradients of physico-chemical factors (for example, depth-related gradients of light, temperature, salinity, nutrients and pressure) constitute major forces shaping ecosystems at activity 'hot spots' on the ocean floor, such as hydrothermal vents, cold seepages and mud volcanoes and hypersaline lakes, though the relationships between community composition, activities and environmental parameters remain largely elusive. We report here results of a detailed study of primary producing microbial communities in the deep Eastern Mediterranean Sea. The brine column of the deep anoxic hypersaline brine lake, L'Atalante, the overlying water column and the brine-seawater interface, were characterized physico-and geochemically, and microbiologically, in terms of their microbial community compositions, functional gene distributions and [ 14 C]bicarbonate assimilation activities. The depth distribution of genes encoding the crenarchaeal ammonia monooxygenase a subunit (amoA), and the bacterial ribulose-1,5-biphosphate carboxylase/ oxygenase large subunit (RuBisCO), was found to coincide with two different types of chemoautotrophy. Meso-and bathypelagic microbial communities were enriched in ammoniaoxidizing Crenarchaeota, whereas the autotrophic community at the oxic/anoxic interface of L'Atalante lake was dominated by Epsilonproteobacteria and sulfur-oxidizing Gammaproteobacteria. These autotrophic microbes are thus the basis of the food webs populating these deep-sea ecosystems.

[1] Total prokaryotic abundance, prokaryotic heterotrophic production and enzymatic activities were investigated in epi-, meso-and bathypelagic waters along a longitudinal transect covering the entire Mediterranean Sea. The prokaryotic production and enzymatic activities in deep waters were among the highest reported worldwide at similar depths, indicating that the peculiar physico-chemical characteristics of the Mediterranean Sea, characterized by warm temperatures (typically 13 C also at abyssal depths), support high rates of organic carbon degradation and incorporation by prokaryotic assemblages. The higher trophic conditions in the epipelagic waters of the Western basin resulted in significantly higher prokaryotic production and enzymatic activities rates than in the Central-Eastern basin. While all of the variables decreased significantly from epi-to meso-and bathypelagic waters, cell-specific hydrolytic activity and cell-specific carbon production significantly increased. In addition, the deep-water layers were characterized by low half-saturation constants (K m ) of all enzymatic activities. These findings suggest that prokaryotic assemblages inhabiting the dark portion of the Mediterranean Sea are able to channel degraded carbon into biomass in a very efficient way, and that prokaryotic assemblages of the deep Mediterranean waters work as a "bioreactor" of organic matter cycling. Since prokaryotic production and enzymatic activities in deep water masses were inversely related with oxygen concentration, we hypothesize a tight link between prokaryotic metabolism and oxygen consumption. As climate change is increasing deep-water temperatures, the predicted positive response of prokaryotic metabolism to temperature increases may accelerate oxygen depletion of deep Mediterranean waters, with cascade consequences on carbon cycling and biogeochemical processes on the entire deep basin.

In the Mediterranean Sea, the contribution of picophytoplankton to primary product~on varies from 31 % (Straits of Messina, with an average total production of 0.18 mg C m-3 h-') to 9 2 % (Ionian Sea, with an average of 1.64 mg C m-3 h-'). Although studies have used different methodologies, the contribution of this small phototrophic fractlon to whole " C assimilation seems to be more important in oligotrophic off-shore waters (71 %) compared to the neritic ones (44 %). Diluting effects of fluvial waters (Adriatic Sea) and hydrodynamic phenomena (Straits of Messina) probably influence the physiologic activity of these populations. Picophytoplankton contribution to primary production in brackish and eutrophic environment is clearly lower than in marine waters both in terms of chl a and of 14C assimilation. Reports of cellular abundance (Cyanobactena and eucaryotic cells) ranged from 105to 10' cells 1-' with a mean value of 10' showing no difference between seasons and suggesting a state of particular equilibrium for this population in the planktonic community. Recent data on prochlorophyte cellular density are of the same order of magnitude (107 cells 1-l) of other picophytoplanktonic procaryotic and eucaryotic organisms.

The EU Marine Strategy Framework Directive 2008/56/EC (MSFD) defines a framework for Community actions in the field of marine environmental policy in order to achieve and/or maintain the Good Environmental Status (GES) of the European seas by 2020. Microbial assemblages (from viruses to microbial-sized metazoa) provide a major contribution to global biodiversity and play a crucial role in the functioning of marine ecosystems, but are largely ignored by the MSFD. Prokaryotes are only seen as "microbial pathogens," without defining their role in GES indicators. However, structural or functional prokaryotic variables (abundance, biodiversity and metabolism) can be easily incorporated into several MSFD descriptors (i.e. D1. biodiversity, D4. food webs, D5. eutrophication, D8. contaminants and D9. contaminants in seafood) with beneficial effects. This review provides a critical analysis of the current MSFD descriptors and illustrates the reliability and advantages of the potential incorporation of some prokaryotic variables within the set of indicators of marine environmental quality. Following a cost/benefit analysis against scientific and economic criteria, we conclude that marine microbial components, and particularly prokaryotes, are highly effective for detecting the effects of anthropogenic pressures on marine environments and for assessing changes in the environmental health status. Thus, we recommend the inclusion of these components in future implementations of the MSFD.