Carboxylic acids are available in abundance from renewable resources and they could serve as economic precursors for bio-based products such as polymers, aldehyde building blocks and alcohols. However, carboxylic acid reduction is a challenge for chemical as well as enzyme catalyzed transformations due to the thermodynamic stability of carboxylates: Whereas chemical reduction methods typically require the use of organic solvents, toxic reagents, undesired reaction conditions, multiple steps or excessive use of reducing equivalents, the enzymatic reduction is poorly understood and only two enzymes are currently available to be used at a preparative scale. A significant number of carboxylate reductase sequences has not yet been elucidated and remains to be explored for biocatalysis. One goal is therefore to identify the primary sequences of enzymes that have already been reported to possess carboxylate reductase activity by a model assisted database mining approach. The putative new carboxylate reductases will then be heterologously expressed in Escherichia coli or Pichia pastoris and their activity will be assessed with a chromatography based in vitro assay. Active carboxylate reductases will be characterized biochemically using a spectrophotometric assay and subjected to crystallization trials in order to solve the first crystal structure of a carboxylate reductase. In addition, molecular model- or x-ray structure-guided site directed mutagenesis will be used to elucidate key residues for substrate binding and catalysis. In summary, we will gain understanding of the fundamentals of enzyme catalyzed reduction of carboxylic acids on the molecular level and provide new enzymes for biocatalysis.