Structural studies depend on high quality protein samples. A major effort in protein structural studies therefore concerns the development of reliable protein expression and purification protocols. This is more difficult to achieve in case the studied protein is a complex of two different polypeptide chains, like in our projects on the trifunctional enzyme (TFE) and the collagen prolyl 4-hydroxylases (C-P4H). We study three different TFEs: Mycobacterium tuberculosis TFE (MtTFE), human (mitochondrial) TFE (HsTFE) and (anaerobic) Escherichia coli TFE (anEcTFE). MtTFE is a soluble enzyme for which a suitable expression and purification protocol exists. HsTFE and anEcTFE are actually membrane associated enzymes, which requires that a suitable detergent has to be found for solubilising these complexes. In the TFE and C-P4H projects two chains, a and b, need to be expressed simultaneously. Using codon-optimised synthetic genes and selecting appropriate plasmids and E. coli host strains it is now possible to express about 0.5 mg HsTFE and 0.4 mg anEcTFE per liter of culture medium, using the best possible detergent. CryoEM studies with the Butcher group (University of Helsinki) have been initiated. A particular complexity of the C-P4H project concerns the presence of SS-bridges. According to current knowledge there are possibly 8 SS-bridges in the full length tetramer. Another problem concerns aspecific (?) proteolytic cleavage, via which truncated complexes are purified. In this project, in collaboration with the Myllyharju group, we work on C-P4Hs of human (type I, II, and III) as well as on Caenorhabditis elegans and Brugia malayi C-P4H. With help of the Ruddock group (FBMM) we have developed a more efficient expression protocol, which works currently the best for human C-P4H-II of which we can now purify 2 mg full length tetramer per liter of culture. Together with Bergmann (FBMM) and Janis (University of Eastern Finland, Joensuu) we are trying to determine the exact nature of the SS-bridges of C-P4H by mass spectroscopy approaches.
Several structures of a Mycobacterium smegmatis thiolase (MSM13-thiolase) were determined at the IISc (Bangalore, India) in the research group of Murthy. This thiolase is a tetrameric thiolase of unknown function. Three different crystal forms have been obtained and in each of the crystal forms the tetramer has been crystallised as an asymmetric tetramer, unexpectedly (Janardan et al., 2015). The shape of the tetrameric thiolases has unique features, such that the active site is located at the end of a bulk solvent accessible cleft that is formed by loops of each of the four subunits. The asymmetry of the tetramer generates a “narrow” cleft and a “wide” cleft, as schematically visualised in Figure 1. In the structure of the CoA bound complex the highest occupancy of bound CoA is found in the active sites facing the “narrow” cleft. MSM13-thiolase has been classified as a thiolase with unknown function (Anbazhagan, Harijan et al., 2013), but pairwise sequence comparisons show that it has much sequence similarity with the mitochondrial T1-thiolase, which is a tetrameric, degradative thiolase. The study of the active site geometry of the MSM-13 thiolase and the human T1-thiolase shows nevertheless significant structural differences and also the enzyme kinetic assays suggest different kinetic properties for MSM13- and T1-thiolase. The substrate specificity, and therefore the function, of the MSM-13 thiolase remains unknown, despite the large amount of available data.
In another project, two structures of the periplasmic part of CdsD of the human pathogen Chlamydia trachomatis (CdsD, Figure 2) were completely refined and deposited (Meriläinen et al., manuscript submitted for publication). CdsD is a building block of the basal body of the Type-III Secretion System (T3SS) of Chlamydia trachomatis. The T3SS system translocates proteins from within the bacterium into the eukaryotic host cell. The basal body of the T3SS crosses the inner membrane (IM) and the outer membrane (OM) of this gram-negative bacterium. The basal body is built from three structural proteins, CdsD, CdsJ and CdsC. CdsD and CdsJ form two concentric rings of 24 proteins which cross the IM. CdsD forms the outer ring, CdsJ forms the inner ring. The CdsC protein forms a 15-mer ring structure that crosses the OM. The determination of the crystal structure of chlamydial CdsD has been challenging as the more routine molecular replacement procedures for obtaining initial phase information failed. Therefore experimental phases had to be obtained by making a suitable heavy atom derivative, which in the end was successful after generating a new crystal form in a mother liquor that was suitable for soaking experiments with heavy atom compounds. The structure adopts an extended shape of three domains each having an abbab-fold (Figure 2), which are assembled in a linear fashion, generating a molecule of about 90Å in length. Two crystal forms have been obtained showing the same (extensive) crystal packing interactions, which might visualise the packing of the CdsD molecule in the outer ring assembly of the basal body of the chlamydial T3SS injectisome.
The recently completed structures of MFE1 suggest a hypothesis for the channelling mechanism of this multifunctional enzyme that has two active sites. The N-terminal domain harbours the hydratase active site and the C-terminal region harbours the second, dehydrogenase active site. In one of these structures, the second active site is complexed with a 3-ketodecanoyl-CoA molecule as well as with NAD+ (or NADH). The crystal form of this ternary complex has been obtained by a complicated protocol in which the crystal is equilibrated with 2-trans-decenoyl-CoA, provided by Dr Schmitz, University of Wurzberg, Germany). This compound is hydrated in the crystal by the first active site, then diffuses to the second active site and, in the presence of NAD+, is then converted to 3-ketodecanoyl-CoA. The preliminary analysis of this ternary complex suggest that the linker helix between the two domains might be involved in allosteric communication between the two active sites, which then could explain the substrate channelling, as observed for the 2-trans,4-trans-decadienoyl-CoA substrate (Yang et al., 1986).
Viimeksi päivitetty: 28.10.2016