Recent observations have raised the hypothesis that biochemical pathways could be involved in cloud chemistry and physics. Indeed, cloud water carries about ~104 microorganisms mL-1, and a metabolic activity has been detected in cloud water samples. Since 2003, more than 34 cloud events have been sampled from the puy de Dôme summit (1465 m a.s.l., France) using droplet samplers. These have been characterized for their chemical and microbiological contents, and viable bacterial strains have been isolated by culture. In laboratory, we are currently focusing on the capacity of isolates in (i) initiating the formation of ice crystals and act as ice nuclei (IN), (ii) producing biosurfactants and initiate the formation of cloud droplets (cloud condensation nuclei (CCN)) and (iii) transforming the organic compounds present in the atmosphere (formate, acetate, lactate, succinate, oxalate, formaldehyde and methanol). Experimental setups (microcosms) consist of bacterial suspensions incubated in liquid solutions of which the chemical composition reflects that of our cloud water samples: these have been designed for mimicking the conditions encountered under either oceanic or anthropic influence. The capacity of selected isolates to catalyze the formation of ice was tested by immersion freezing and in cloud chamber (AIDA, Karlsruhe). Ice nucleation active bacteria at -5°C were found in cloud samples; in cloud water, such cells could be responsible for the freezing of droplets into crystals, which often initiates precipitations. Preliminary analysis of the data showed that the IN efficiency of the isolates was dependent on the chemical composition of cloud water. Isolates have also been screened for biosurfactant production, by emulsification, oil spreading and surface tension meniscus tests. A majority of strains were positive to at least one test, suggesting that such cells would be efficient CCN in the atmosphere and could participate to the formation of liquid clouds. In order to investigate the implication of microbes in the chemistry of cloud water, the rates of biotransformation of atmospheric organic compounds by isolated strains under microcosm conditions have been determined. They were in the range of ~10-2 pmol-1 cell-1 d-1 at 5°C and were in general decreased in the “anthropic” versus “oceanic” medium. The presence of free radicals such as OH● was also detrimental. A basic numeric model of atmospheric chemistry taking into account realistic concentrations of organic compounds, of bacteria, and of OH● and NO3● radicals in a liquid cloud at 5°C for daytime and nighttime and their respective reaction rates towards the tested organic acids was applied on the results. This confirmed that the presence of OH● radicals was largely responsible for the oxidation of organic species during the day, but it is noteworthy that during the night, bacterial activity would dominate the reactivity of formic, acetic and succinic acids by contributing to more than 90 % of their degradation. Overall, our results suggest that cloudborne microorganisms are actors of the physico-chemical processes occurring in the atmosphere. The composition of cloud water modulates these roles, and this raises the question of the negative impact that human emissions could have on the capacity of microbes to interact with atmospheric processes and disseminate by the atmosphere.