Identification, characterization, and structure analysis of the cyclic di-AMP-binding PII-like signal transduction protein DarA.

The cyclic dimeric AMP nucleotide c-di-AMP is an essential second messenger in Bacillus subtilis. We have identified the protein DarA as one of the prominent c-di-AMP receptors in B. subtilis. Crystal structure analysis shows that DarA is highly homologous to PII signal transducer proteins. In contrast to PII proteins, the functionally important B- and T-loops are swapped with respect to their size. DarA is a homotrimer that binds three molecules of c-di-AMP, each in a pocket located between two subunits. We demonstrate that DarA is capable to bind c-di-AMP and with lower affinity cyclic GMP-AMP (3'3'-cGAMP) but not c-di-GMP or 2'3'-cGAMP. Consistently the crystal structure shows that within the ligand-binding pocket only one adenine is highly specifically recognized, whereas the pocket for the other adenine appears to be promiscuous. Comparison with a homologous ligand-free DarA structure reveals that c-di-AMP binding is accompanied by conformational changes of both the fold and the position of the B-loop in DarA.

Control of the diadenylate cyclase CdaS in Bacillus subtilis: an autoinhibitory domain limits cyclic di-AMP production.

The Gram-positive bacterium Bacillus subtilis encodes three diadenylate cyclases that synthesize the essential signaling nucleotide cyclic di-AMP. The activities of the vegetative enzymes DisA and CdaA are controlled by protein-protein interactions with their conserved partner proteins. Here, we have analyzed the regulation of the unique sporulation-specific diadenylate cyclase CdaS. Very low expression of CdaS as the single diadenylate cyclase resulted in the appearance of spontaneous suppressor mutations. Several of these mutations in the cdaS gene affected the N-terminal domain of CdaS. The corresponding CdaS mutant proteins exhibited a significantly increased enzymatic activity. The N-terminal domain of CdaS consists of two α-helices and is attached to the C-terminal catalytically active diadenylate cyclase (DAC) domain. Deletion of the first or both helices resulted also in strongly increased activity indicating that the N-terminal domain serves to limit the enzyme activity of the DAC domain. The structure of YojJ, a protein highly similar to CdaS, indicates that the protein forms hexamers that are incompatible with enzymatic activity of the DAC domains. In contrast, the mutations and the deletions of the N-terminal domain result in conformational changes that lead to highly increased enzymatic activity. Although the full-length CdaS protein was found to form hexamers, a truncated version with a deletion of the first N-terminal helix formed dimers with high enzyme activity. To assess the role of CdaS in sporulation, we assayed the germination of wild type and cdaS mutant spores. The results indicate that cyclic di-AMP formed by CdaS is required for efficient germination.

Alternating sites reactivity is a common feature of thiamin diphosphate-dependent enzymes as evidenced by isothermal titration calorimetry studies of substrate binding.

Thiamin diphosphate (ThDP)-dependent enzymes play vital roles in cellular metabolism in all kingdoms of life. In previous kinetic and structural studies, a communication between the active centers in terms of a negative cooperativity had been suggested for some but not all ThDP enzymes, which typically operate as functional dimers. To further underline this hypothesis and to test its universality, we investigated the binding of substrate analogue methyl acetylphosphonate (MAP) to three different ThDP-dependent enzymes acting on substrate pyruvate, namely, the Escherichia coli E1 component of the pyruvate dehydrogenase complex, E. coli acetohydroxyacid synthase isoenzyme I, and the Lactobacillus plantarum pyruvate oxidase using isothermal titration calorimetry. The results unambiguously show for all three enzymes studied that only one active center of the functional dimers accomplishes covalent binding of the substrate analogue, supporting the proposed alternating sites reactivity as a common feature of all ThDP enzymes and resolving the recent controversy in the field.

A δ38 deletion variant of human transketolase as a model of transketolase-like protein 1 exhibits no enzymatic activity.

Besides transketolase (TKT), a thiamin-dependent enzyme of the pentose phosphate pathway, the human genome encodes for two closely related transketolase-like proteins, which share a high sequence identity with TKT. Transketolase-like protein 1 (TKTL1) has been implicated in cancerogenesis as its cellular expression levels were reported to directly correlate with invasion efficiency of cancer cells and patient mortality. It has been proposed that TKTL1 exerts its function by catalyzing an unusual enzymatic reaction, a hypothesis that has been the subject of recent controversy. The most striking difference between TKTL1 and TKT is a deletion of 38 consecutive amino acids in the N-terminal domain of the former, which constitute part of the active site in authentic TKT. Our structural and sequence analysis suggested that TKTL1 might not possess transketolase activity. In order to test this hypothesis in the absence of a recombinant expression system for TKTL1 and resilient data on its biochemical properties, we have engineered and biochemically characterized a "pseudo-TKTL1" Δ38 deletion variant of human TKT (TKTΔ38) as a viable model of TKTL1. Although the isolated protein is properly folded under in vitro conditions, both thermal stability as well as stability of the TKT-specific homodimeric assembly are markedly reduced. Circular dichroism and NMR spectroscopic analysis further indicates that TKTΔ38 is unable to bind the thiamin cofactor in a specific manner, even at superphysiological concentrations. No transketolase activity of TKTΔ38 can be detected for conversion of physiological sugar substrates thus arguing against an intrinsically encoded enzymatic function of TKTL1 in tumor cell metabolism.

Recombinant expression, in vitro refolding, and biophysical characterization of the human glucagon-like peptide-1 receptor.

Activation of the glucagon-like peptide-1 receptor (GLP-1R) upon ligand binding leads to the release of insulin from pancreatic cells. This strictly glucose-dependent process renders the receptor and its ligands useful in the treatment of type II diabetes mellitus. To enable a biophysical characterization in vitro, we expressed the human full-length GLP-1R in the cytosol of Escherichia coli as inclusion bodies. After purification, refolding of the SDS-solubilized receptor was achieved by the exchange of SDS against the detergent Brij78 using an artificial chaperone system. Far-UV circular dichroism spectroscopic studies revealed that the receptor adopts a characteristic alpha-helical structure in Brij78 micelles. Ligand binding of the renatured protein was quantified by fluorescence quenching and surface plasmon resonance spectroscopy. In the presence of Brij micelles, the refolded receptor binds the agonist exendin-4 with an apparent dissociation constant of approximately 100 nM in a reversible one-step mechanism. To demonstrate that the detected ligand binding activity is not only due to an autonomously functional N-terminal domain (nGLP-1R) but also due to additional contacts with the juxtamembrane part, we separately expressed and refolded the extracellular domain relying on identical protocols established for the full-length GLP-1R. In support of the suggested multidomain binding mode, the nGLP-1R binds exendin-4 with a lower affinity (K(app) in the micromolar range) and a different kinetic mechanism. The lower ligand affinity of the nGLP-1R results entirely from a decreased kinetic stability of the receptor-ligand complex, dissociation of which is approximately 40-fold faster in the case of the nGLP-1R compared to the full-length GLP-1R. In summary, a framework was developed to produce functional human full-length GLP-1R by recombinant expression in E. coli as a prerequisite for eventual structure determination and a rigorous biophysical characterization including protein variants.

The crystal structure of human transketolase and new insights into its mode of action.

The crystal structure of human transketolase (TKT), a thiamine diphosphate (ThDP) and Ca(2+)-dependent enzyme that catalyzes the interketol transfer between ketoses and aldoses as part of the pentose phosphate pathway, has been determined to 1.75 Å resolution. The recombinantly produced protein crystallized in space group C2 containing one monomer in the asymmetric unit. Two monomers form the homodimeric biological assembly with two identical active sites at the dimer interface. Although the protomer exhibits the typical three (α/ß)-domain structure and topology reported for TKTs from other species, structural differences are observed for several loop regions and the linker that connects the PP and Pyr domain. The cofactor and substrate binding sites of human TKT bear high resemblance to those of other TKTs but also feature unique properties, including two lysines and a serine that interact with the ß-phosphate of ThDP. Furthermore, Gln(189) spans over the thiazolium moiety of ThDP and replaces an isoleucine found in most non-mammalian TKTs. The side chain of Gln(428) forms a hydrogen bond with the 4'-amino group of ThDP and replaces a histidine that is invariant in all non-mammalian TKTs. All other amino acids involved in substrate binding and catalysis are strictly conserved. Besides a steady-state kinetic analysis, microscopic equilibria of the donor half-reaction were characterized by an NMR-based intermediate analysis. These studies reveal that formation of the central 1,2-dihydroxyethyl-ThDP carbanion-enamine intermediate is thermodynamically favored with increasing carbon chain length of the donor ketose substrate. Based on the structure of human transketolase and sequence alignments, putative functional properties of the related transketolase-like proteins TKTL1 and -2 are discussed in light of recent findings suggesting that TKTL1 plays a role in cancerogenesis.

Direct kinetic evidence for half-of-the-sites reactivity in the E1 component of the human pyruvate dehydrogenase multienzyme complex through alternating sites cofactor activation.

Recent kinetic and structural studies on various thiamin-dependent enzymes, including the bacterial E1 component of the pyruvate dehydrogenase complex (PDHc), suggested an active center communication between the cofactors in these multimeric enzymes. This regulatory mode has been inferred from the dissymmetry of active sites in proteolytic patterns and X-ray structures and from a complex macroscopic kinetic behavior not being consistent with independently working active sites. Here, direct microscopic kinetic evidence for this hypothesis is presented for the alpha2beta2-type E1 component of the human pyruvate dehydrogenase complex. Only one of the two thiamin molecules bound to the two active sites is in a chemically activated state exhibiting an apparent C2 ionization rate constant of approximately 50 s(-1) at pH 7.6 and 30 degrees C, whereas the thiamin in the "inactive site" ionizes with a rate that is at least 3 orders of magnitude smaller. The chemical nonequivalence is also exhibited in the ability to bind the substrate analogue methyl acetylphosphonate and in the catalytic turnover of the substrate pyruvate in the E1-only reaction. In the activated active site, pyruvate is rapidly bound and decarboxylated with apparent forward rate constants of covalent pyruvate binding of 2 s(-1) and decarboxylation of the formed 2-lactyl-thiamin intermediate of 5 s(-1). In the dormant site, these steps are as slow as 0.03 s(-1). Under the conditions that were used, only the heterotetramer can be detected by analytical ultracentrifugation, thus ruling out the possibility that multiple oligomeric species with different reactivities cause the observed kinetic effects. The results are consistent with the recently suggested model of an active site synchronization in PDHc-E1 via a proton wire that keeps the two active sites in an alternating activation state [Frank, R. A., et al. (2004) Science 306, 872]. Kinetic studies on the related thiamin enzymes transketolase, pyruvate oxidase, and bacterial pyruvate decarboxylase are not consistent with a chemical and/or functional nonequivalence of the active sites as observed in the E1 component of hsPDHc. We hypothesize that the alternating sites reaction in PDHc-E1 aids in the synchronized acyl transfer to the E2 component in the highly organized multienzyme complex.