Borrelia are gram-negative screw-like bacteria that can be transmitted to humans through the bite of hard-bodies ticks. The species Borrelia burgdorferi, B. afzelii, B. garinii, and B. spielmannii, collectively known as Borrelia burgdorferi sensu lato, can cause a disease called Borreliosis. The disease presents with highly variable symptoms and is therefore often misdiagnosed. The more severe symptoms include arthritis, carditis, and neurologic problems. Borrelia encodes at least 120 adhesive outer surface lipoproteins (adhesins) that are differentially expressed throughout the infection cycle. Many of these adhesins play essential roles in initiation and/or maintenance of infection. For ~ 30 adhesins, a human ligand has been identified through in vitro binding assays, whereas for only a small fraction an in vivo function has been assigned. Only a few adhesins have been structurally characterized so far. It is expected that most uncharacterized adhesins possess hitherto unknown folds. Many adhesins seem to be unique within the Borrelia proteome, since they do not share any sequence similarity to other Borrelia proteins. Within this project, our focus lies on the structural characterization of adhesins in complex with their human interaction partners as well as selected adhesins for which no binding partner has been identified yet, but with defined roles in the infection cycle.

Legionella pneumophila is a waterborne gram-negative bacterium that lives within free-living amoebae of different species. However, it can cause a life-threatening atypical pneumonia in humans and is usually transmitted via inhaled water droplets. To sustain its intracellular lifestyle in amoebae as well as in human cells, L. pneumophila injects more than 300 effector proteins into the host cell. To date the functions of many of these effectors are unknown due to the lack of structural data and the absence of sequence homology to other proteins. We are working on the structural and functional characterization of selected effectors, for which a function in the host cell has been already postulated. We have cloned and purified about 20 effectors, some of which have already been crystallized.

Sterols are essential components of eukaryotic membranes. Whereas mammals including humans utilize cholesterol as the principal membrane sterol, lower eukaryotes employ ergosterol, which differs from cholesterol in the architecture of its side chain. This difference is achieved by a divergence (branching) of the biosynthetic pathway after the last common intermediate lanosterol. Several enzymes upstream of lanosterol are already targeted by commonly used antifungal drugs. However, their use is often accompanied by unwanted side effects that may arise due to the similarity between host and pathogen enzymes. Thus, the side chain modifying enzymes acting downstream of lanosterol are attractive drug targets, since they do not possess human counterparts.
We currently focus on the enzyme 24-sterol methyltransferase (24-SMT), which adds a methyl group to the side-chain double bond between carbon atoms 24 and 25in a stereoselective manner. These enzymes have been subject of intensive research during the past three decades, but have eluded structural characterization so far. We have cloned the coding sequences of 24-SMTs from diverse fungi and protozoa and have recombinantly produced the corresponding proteins that are currently subject to crystallization experiments. Moreover, we have established an enzyme coupled activity assay that permits us to determine the isoform specificity of known inhibitors to validate their use in crystallization trials.