More than just pattern recognition: Prediction of uncommon protein structure features by AI methods.

The CASP14 experiment demonstrated the extraordinary structure modeling capabilities of artificial intelligence (AI) methods. That result has ignited a fierce debate about what these methods are actually doing. One of the criticisms has been that the AI does not have any sense of the underlying physics but is merely performing pattern recognition. Here, we address that issue by analyzing the extent to which the methods identify rare structural motifs. The rationale underlying the approach is that a pattern recognition machine tends to choose the more frequently occurring motifs, whereas some sense of subtle energetic factors is required to choose infrequently occurring ones. To reduce the possibility of bias from related experimental structures and to minimize the effect of experimental errors, we examined only CASP14 target protein crystal structures determined to a resolution limit better than 2 Å, which lacked significant amino acid sequence homology to proteins of known structure. In those experimental structures and in the corresponding models, we track cis peptides, π-helices, 310-helices, and other small 3D motifs that occur in the PDB database at a frequency of lower than 1% of total amino acid residues. The best-performing AI method, AlphaFold2, captured these uncommon structural elements exquisitely well. All discrepancies appeared to be a consequence of crystal environment effects. We propose that the neural network learned a protein structure potential of mean force, enabling it to correctly identify situations where unusual structural features represent the lowest local free energy because of subtle influences from the atomic environment.

Cryo-EM structure of the ancient eukaryotic ribosome from the human parasite Giardia lamblia.

Giardiasis is a disease caused by the protist Giardia lamblia. As no human vaccines have been approved so far against it, and resistance to current drugs is spreading, new strategies for combating giardiasis need to be developed. The G. lamblia ribosome may provide a promising therapeutic target due to its distinct sequence differences from ribosomes of most eukaryotes and prokaryotes. Here, we report the cryo-electron microscopy structure of the G. lamblia (WB strain) ribosome determined at 2.75 Å resolution. The ribosomal RNA is the shortest known among eukaryotes, and lacks nearly all the eukaryote-specific ribosomal RNA expansion segments. In contrast, the ribosomal proteins are typically eukaryotic with some species-specific insertions/extensions. Most typical inter-subunit bridges are maintained except for one missing contact site. Unique structural features are located mainly at the ribosome's periphery. These may be exploited as target sites for the design of new compounds that inhibit selectively the parasite's ribosomal activity.

Computational models in the service of X-ray and cryo-electron microscopy structure determination.

Critical assessment of structure prediction (CASP) conducts community experiments to determine the state of the art in computing protein structure from amino acid sequence. The process relies on the experimental community providing information about not yet public or about to be solved structures, for use as targets. For some targets, the experimental structure is not solved in time for use in CASP. Calculated structure accuracy improved dramatically in this round, implying that models should now be much more useful for resolving many sorts of experimental difficulties. To test this, selected models for seven unsolved targets were provided to the experimental groups. These models were from the AlphaFold2 group, who overall submitted the most accurate predictions in CASP14. Four targets were solved with the aid of the models, and, additionally, the structure of an already solved target was improved. An a posteriori analysis showed that, in some cases, models from other groups would also be effective. This paper provides accounts of the successful application of models to structure determination, including molecular replacement for X-ray crystallography, backbone tracing and sequence positioning in a cryo-electron microscopy structure, and correction of local features. The results suggest that, in future, there will be greatly increased synergy between computational and experimental approaches to structure determination.

Discovery and Preclinical Development of Antigiardiasis Fumagillol Derivatives.

Giardiasis, caused by the intestinal parasite Giardia lamblia, is a severe diarrheal disease, endemic in poverty-stricken regions of the world, and also a common infection in developed countries. The available therapeutic options are associated with adverse effects, and G. lamblia resistance to the standard-of-care drugs is spreading. Fumagillin, an antimicrosporidiosis drug, is a therapeutic agent with potential for the treatment of giardiasis. However, it exhibits considerable, albeit reversible, toxicity when used to treat immunocompromised microsporidiosis patients. Fumagillin is also a highly unstable compound. To address these liabilities, we designed and synthesized stable fumagillol derivatives with lower levels of permeation across polarized epithelial Caco-2 cells and better potency against G. lamblia trophozoites than fumagillin. Metronidazole-resistant G. lamblia strains were also susceptible to the new fumagillol derivatives. In addition, these compounds were more potent against the amebiasis-causing parasite Entamoeba histolytica than fumagillin. Two compounds exhibited better thermal and acid stability than fumagillin, which should prolong the drug shelf life and reduce compound degradation in the stomach. Studies with a mouse model of giardiasis with the most stable compound, 4-(((((3R,4S,5S,6R)-5-methoxy-4-((2R,3R)-2-methyl-3-(3-methylbut-2-en-1-yl)oxiran-2-yl)-1-oxaspiro[2.5]octan-6-yl)oxy)carbonyl)amino)benzoic acid (compound 9), revealed that it had better efficacy (effective dose [ED]) than fumagillin at both the fully curative dose (the 100% ED) of 6.6 mg/kg of body weight and a 50% ED of 0.064 mg/kg. Plasma pharmacokinetics revealed the slow absorption of compound 9 through the gut, consistent with the in vitro characterization in Caco-2 cells. An acute-dose study yielded a maximum tolerated dose (MTD) of 1,500 mg/kg, 227-fold higher than the fully curative dose. Thus, along with improved stability, compound 9 also exhibited an excellent therapeutic window.

Gene polymorphism linked to increased asthma and IBD risk alters gasdermin-B structure, a sulfatide and phosphoinositide binding protein.

The exact function of human gasdermin-B (GSDMB), which regulates differentiation and growth of epithelial cells, is yet to be elucidated. In human epidermal growth factor receptor 2 (HER2)-positive breast cancer, GSDMB gene amplification and protein overexpression indicate a poor response to HER2-targeted therapy. Genome-wide association studies revealed a correlation between GSDMB SNPs and an increased susceptibility to Crohn's disease, ulcerative colitis, and asthma. The N- and C-terminal domains of all gasdermins possess lipid-binding and regulatory activities, respectively. Inflammatory caspases cleave gasdermin-D in the interdomain linker but not GSDMB. The cleaved N-terminal domain binds phosphoinositides and cardiolipin, forms membrane-disrupting pores, and executes pyroptosis. We show that both full-length GSDMB and the N-terminal domain bind to nitrocellulose membranes immobilized with phosphoinositides or sulfatide, but not with cardiolipin. In addition, the GSDMB N-terminal domain binds liposomes containing sulfatide. The crystal structure of the GSDMB C-terminal domain reveals the structural impact of the amino acids encoded by SNPs that are linked to asthma and inflammatory bowel disease (IBD). A loop that carries the polymorphism amino acids corresponding to healthy individuals (Gly299:Pro306) exhibits high conformational flexibility, whereas the loop carrying amino acids found in individuals with increased disease risk (Arg299:Ser306) exhibits a well-defined conformation and higher positive surface charge. Apoptotic executioner caspase-3, -6, and -7, but not the inflammatory caspases, cleave GSDMB at 88DNVD91 within the N-terminal domain. Selective sulfatide binding may indicate possible function for GSDMB in the cellular sulfatide transport.

Target highlights from the first post-PSI CASP experiment (CASP12, May-August 2016).

The functional and biological significance of the selected CASP12 targets are described by the authors of the structures. The crystallographers discuss the most interesting structural features of the target proteins and assess whether these features were correctly reproduced in the predictions submitted to the CASP12 experiment.

Structural basis for the binding specificity of human Recepteur d'Origine Nantais (RON) receptor tyrosine kinase to macrophage-stimulating protein.

Recepteur d'origine nantais (RON) receptor tyrosine kinase and its ligand, serum macrophage-stimulating protein (MSP), play important roles in inflammation, cell growth, migration, and epithelial to mesenchymal transition during tumor development. The binding of mature MSPαß (disulfide-linked α- and ß-chains) to RON ectodomain modulates receptor dimerization, followed by autophosphorylation of tyrosines in the cytoplasmic receptor kinase domains. Receptor recognition is mediated by binding of MSP ß-chain (MSPß) to the RON Sema. Here we report the structure of RON Sema-PSI-IPT1 (SPI1) domains in complex with MSPß at 3.0 Å resolution. The MSPß serine protease-like ß-barrel uses the degenerate serine protease active site to recognize blades 2, 3, and 4 of the ß-propeller fold of RON Sema. Despite the sequence homology between RON and MET receptor tyrosine kinase and between MSP and hepatocyte growth factor, it is well established that there is no cross-reactivity between the two receptor-ligand systems. Comparison of the structure of RON SPI1 in complex with MSPß and that of MET receptor tyrosine kinase Sema-PSI in complex with hepatocyte growth factor ß-chain reveals the receptor-ligand selectivity determinants. Analytical ultracentrifugation studies of the SPI1-MSPß interaction confirm the formation of a 1:1 complex. SPI1 and MSPαß also associate primarily as a 1:1 complex with a binding affinity similar to that of SPI1-MSPß. In addition, the SPI1-MSPαß ultracentrifuge studies reveal a low abundance 2:2 complex with ∼ 10-fold lower binding affinity compared with the 1:1 species. These results support the hypothesis that the α-chain of MSPαß mediates RON dimerization.

Structural basis for inactivation of Giardia lamblia carbamate kinase by disulfiram.

Carbamate kinase from Giardia lamblia is an essential enzyme for the survival of the organism. The enzyme catalyzes the final step in the arginine dihydrolase pathway converting ADP and carbamoyl phosphate to ATP and carbamate. We previously reported that disulfiram, a drug used to treat chronic alcoholism, inhibits G. lamblia CK and kills G. lamblia trophozoites in vitro at submicromolar IC50 values. Here, we examine the structural basis for G. lamblia CK inhibition of disulfiram and its analog, thiram, their activities against both metronidazole-susceptible and metronidazole-resistant G. lamblia isolates, and their efficacy in a mouse model of giardiasis. The crystal structure of G. lamblia CK soaked with disulfiram revealed that the compound thiocarbamoylated Cys-242, a residue located at the edge of the active site. The modified Cys-242 prevents a conformational transition of a loop adjacent to the ADP/ATP binding site, which is required for the stacking of Tyr-245 side chain against the adenine moiety, an interaction seen in the structure of G. lamblia CK in complex with AMP-PNP. Mass spectrometry coupled with trypsin digestion confirmed the selective covalent thiocarbamoylation of Cys-242 in solution. The Giardia viability studies in the metronidazole-resistant strain and the G. lamblia CK irreversible inactivation mechanism show that the thiuram compounds can circumvent the resistance mechanism that renders metronidazole ineffectiveness in drug resistance cases of giardiasis. Together, the studies suggest that G. lamblia CK is an attractive drug target for development of novel antigiardial therapies and that disulfiram, an FDA-approved drug, is a promising candidate for drug repurposing.

Challenging the state of the art in protein structure prediction: Highlights of experimental target structures for the 10th Critical Assessment of Techniques for Protein Structure Prediction Experiment CASP10.

For the last two decades, CASP has assessed the state of the art in techniques for protein structure prediction and identified areas which required further development. CASP would not have been possible without the prediction targets provided by the experimental structural biology community. In the latest experiment, CASP10, more than 100 structures were suggested as prediction targets, some of which appeared to be extraordinarily difficult for modeling. In this article, authors of some of the most challenging targets discuss which specific scientific question motivated the experimental structure determination of the target protein, which structural features were especially interesting from a structural or functional perspective, and to what extent these features were correctly reproduced in the predictions submitted to CASP10. Specifically, the following targets will be presented: the acid-gated urea channel, a difficult to predict transmembrane protein from the important human pathogen Helicobacter pylori; the structure of human interleukin (IL)-34, a recently discovered helical cytokine; the structure of a functionally uncharacterized enzyme OrfY from Thermoproteus tenax formed by a gene duplication and a novel fold; an ORFan domain of mimivirus sulfhydryl oxidase R596; the fiber protein gene product 17 from bacteriophage T7; the bacteriophage CBA-120 tailspike protein; a virus coat protein from metagenomic samples of the marine environment; and finally, an unprecedented class of structure prediction targets based on engineered disulfide-rich small proteins.

Discovery of novel antigiardiasis drug candidates.

Giardiasis is a severe intestinal parasitic disease caused by Giardia lamblia, which inflicts many people in poor regions and is the most common parasitic infection in the United States. Current standard care drugs are associated with undesirable side effects, treatment failures, and an increasing incidence of drug resistance. As follow-up to a high-throughput screening of an approved drug library, which identified compounds lethal to G. lamblia trophozoites, we have determined the minimum lethal concentrations of 28 drugs and advanced 10 of them to in vivo studies in mice. The results were compared to treatment with the standard care drug, metronidazole, in order to identify drugs with equal or better anti-Giardia activities. Three drugs, fumagillin, carbadox, and tioxidazole, were identified. These compounds were also potent against metronidazole-resistant human G. lamblia isolates (assemblages A and B), as determined in in vitro assays. Of these three compounds, fumagillin is currently an orphan drug used within the European Union to treat microsporidiosis in immunocompromised individuals, whereas carbadox and tioxidazole are used in veterinary medicine. A dose-dependent study of fumagillin in a giardiasis mouse model revealed that the effective dose of fumagillin was ∼ 100-fold lower than the metronidazole dose. Therefore, fumagillin may be advanced to further studies as an alternative treatment for giardiasis when metronidazole fails.

Structures of p63 DNA binding domain in complexes with half-site and with spacer-containing full response elements.

Transcription factor p63, a p53 family member, plays a role in epithelial cell development, cell cycle arrest, apoptosis, and tumorigenesis. Point mutations, primarily in the DNA binding domain (p63DBD), lead to malformation syndromes. To gain insight into differences between p63 and p53 and the impact of mutations on the structure, we have determined two crystal structures of p63DBD in complex with A/T-rich response elements. One complex contains a 10-bp DNA half-site response element (5'AAACATGTTT3') and the other contains a 22-bp DNA full response element with a 2-bp spacer between two half-sites (5'AAACATGTTTTAAAACATGTTT3'). In both structures, each half-site binds a p63DBD dimer. The two p63DBD dimers do not interact in the presence of the DNA spacer, whereas they interact with one another in the p63DBD/10-bp complex where the DNA simulates a full response element by packing end-to-end. A unique dimer-dimer interaction involves a variable loop region, which differs in length and sequence from the counterpart loop of p53DBD. The DNA trajectories in both structures assume superhelical conformations. Surface plasmon resonance studies of p63DBD/DNA binding yielded K(d) = 11.7 µM for a continuous full response element, whereas binding was undetectable with the 22-bp DNA, suggesting an important contribution of a p63DBD interdimer interface to binding and establishing that p63DBD affinity to the response element is approximately 1,000-fold lower than that of p53DBD. Analyses of the structural consequences of p63DBD mutations that cause developmental defects show that, although some mutations affect DNA binding directly, the majority affects protein stability.

Pliable DNA conformation of response elements bound to transcription factor p63.

We show that changes in the nucleotide sequence alter the DNA conformation in the crystal structures of p63 DNA-binding domain (p63DBD) bound to its response element. The conformation of a 22-bp canonical response element containing an AT spacer between the two half-sites is unaltered compared with that containing a TA spacer, exhibiting superhelical trajectory. In contrast, a GC spacers abolishes the DNA superhelical trajectory and exhibits less bent DNA, suggesting that increased GC content accompanies increased double helix rigidity. A 19-bp DNA, representing an AT-rich response element with overlapping half-sites, maintains superhelical trajectory and reveals two interacting p63DBD dimers crossing one another at 120°. p63DBD binding assays to response elements of increasing length complement the structural studies. We propose that DNA deformation may affect promoter activity, that the ability of p63DBD to bind to superhelical DNA suggests that it is capable of binding to nucleosomes, and that overlapping response elements may provide a mechanism to distinguish between p63 and p53 promoters.

Understanding the Molecular Basis for Homodimer Formation of the Pneumococcal Endolysin Cpl-1.

The rise of multi-drug-resistant bacteria that cannot be treated with traditional antibiotics has prompted the search for alternatives to combat bacterial infections. Endolysins, which are bacteriophage-derived peptidoglycan hydrolases, are attractive tools in this fight. Several studies have already demonstrated the efficacy of endolysins in targeting bacterial infections. Endolysins encoded by bacteriophages that infect Gram-positive bacteria typically possess an N-terminal catalytic domain and a C-terminal cell-wall binding domain (CWBD). In this study, we have uncovered the molecular mechanisms that underlie formation of a homodimer of Cpl-1, an endolysin that targets Streptococcus pneumoniae. Here, we use site-directed mutagenesis, analytical size exclusion chromatography, and analytical ultracentrifugation to disprove a previous suggestion that three residues at the N-terminus of the CWBD are involved in the formation of a Cpl-1 dimer in the presence of choline in solution. We conclusively show that the C-terminal tail region of Cpl-1 is involved in formation of the dimer. Alanine scanning mutagenesis generated various tail mutant constructs that allowed identification of key residues that mediate Cpl-1 dimer formation. Finally, our results allowed identification of a consensus sequence (FxxEPDGLIT) required for choline-dependent dimer formation─a sequence that occurs frequently in pneumococcal autolysins and endolysins. These findings shed light on the mechanisms of Cpl-1 and related enzymes and can be used to inform future engineering efforts for their therapeutic development against S. pneumoniae.

A unique borrelial protein facilitates microbial immune evasion.

IMPORTANCE: Lyme disease is a major tick-borne infection caused by a bacterial pathogen called Borrelia burgdorferi, which is transmitted by ticks and affects hundreds of thousands of people every year. These bacterial pathogens are distinct from other genera of microbes because of their distinct features and ability to transmit a multi-system infection to a range of vertebrates, including humans. Progress in understanding the infection biology of Lyme disease, and thus advancements towards its prevention, are hindered by an incomplete understanding of the microbiology of B. burgdorferi, partly due to the occurrence of many unique borrelial proteins that are structurally unrelated to proteins of known functions yet are indispensable for pathogen survival. We herein report the use of diverse technologies to examine the structure and function of a unique B. burgdorferi protein, annotated as BB0238-an essential virulence determinant. We show that the protein is structurally organized into two distinct domains, is involved in multiplex protein-protein interactions, and facilitates tick-to-mouse pathogen transmission by aiding microbial evasion of early host cellular immunity. We believe that our findings will further enrich our understanding of the microbiology of B. burgdorferi, potentially impacting the future development of novel prevention strategies against a widespread tick-transmitted infection.

Correlation of structure and function in the human hotdog-fold enzyme hTHEM4.

Human THEM4 (hTHEM4) is comprised of a catalytically active hotdog-fold acyl-CoA thioesterase domain and an N-terminal domain of unknown fold and function. hTHEM4 has been linked to Akt1 regulation and cell apoptosis. Herein, we report the X-ray structure of hHTEM4 bound with undecan-2-one-CoA. Structure guided mutagenesis was carried out to confirm the catalytic residues. The N-terminal domain is shown to be partially comprised of irregular and flexible secondary structure, reminiscent of a protein-binding domain. We demonstrate direct hTHEM4-Akt1 binding by immunoprecipitation and by inhibition of Akt1 kinase activity, thus providing independent evidence that hTHEM4 is an Akt1 negative regulator.

Structure of Escherichia coli O157:H7 bacteriophage CBA120 tailspike protein 4 baseplate anchor and tailspike assembly domains (TSP4-N).

Four tailspike proteins (TSP1-4) of Escherichia coli O157:H7 bacteriophage CBA120 enable infection of multiple hosts. They form a branched complex that attaches to the tail baseplate. Each TSP recognizes a different lipopolysaccharide on the membrane of a different bacterial host. The 335 N-terminal residues of TSP4 promote the assembly of the TSP complex and anchor it to the tail baseplate. The crystal structure of TSP4-N335 reveals a trimeric protein comprising four domains. The baseplate anchor domain (AD) contains an intertwined triple-stranded ß-helix. The ensuing XD1, XD2 and XD3 ß-sheet containing domains mediate the binding of TSP1-3 to TSP4. Each of the XD domains adopts the same fold as the respective XD domains of bacteriophage T4 gp10 baseplate protein, known to engage in protein-protein interactions via its XD2 and XD3 domains. The structural similarity suggests that XD2 and XD3 of TSP4 also function in protein-protein interactions. Analytical ultracentrifugation analyses of TSP4-N335 and of domain deletion proteins showed how TSP4-N335 promotes the formation of the TSP quaternary complex. TSP1 and TSP2 bind directly to TSP4 whereas TSP3 binding requires a pre-formed TSP4-N335:TSP2 complex. A 3-dimensional model of the bacteriophage CBA120 TSP complex has been developed based on the structural and ultracentrifuge information.

Structure of oxalacetate acetylhydrolase, a virulence factor of the chestnut blight fungus.

Oxalacetate acetylhydrolase (OAH), a member of the phosphoenolpyruvate mutase/isocitrate lyase superfamily, catalyzes the hydrolysis of oxalacetate to oxalic acid and acetate. This study shows that knock-out of the oah gene in Cryphonectria parasitica, the chestnut blight fungus, reduces the ability of the fungus to form cankers on chestnut trees, suggesting that OAH plays a key role in virulence. OAH was produced in Escherichia coli and purified, and its catalytic rates were determined. Oxalacetate is the main OAH substrate, but the enzyme also acts as a lyase of (2R,3S)-dimethyl malate with approximately 1000-fold lower efficacy. The crystal structure of OAH was determined alone, in complex with a mechanism-based inhibitor, 3,3-difluorooxalacetate (DFOA), and in complex with the reaction product, oxalate, to a resolution limit of 1.30, 1.55, and 1.65 A, respectively. OAH assembles into a dimer of dimers with each subunit exhibiting an (alpha/beta)(8) barrel fold and each pair swapping the 8th alpha-helix. An active site "gating loop" exhibits conformational disorder in the ligand-free structure. To obtain the structures of the OAH.ligand complexes, the ligand-free OAH crystals were soaked briefly with DFOA or oxalacetate. DFOA binding leads to ordering of the gating loop in a conformation that sequesters the ligand from the solvent. DFOA binds in a gem-diol form analogous to the oxalacetate intermediate/transition state. Oxalate binds in a planar conformation, but the gating loop is largely disordered. Comparison between the OAH structure and that of the closely related enzyme, 2,3-dimethylmalate lyase, suggests potential determinants of substrate preference.

High-throughput Giardia lamblia viability assay using bioluminescent ATP content measurements.

The human pathogen Giardia lamblia is an anaerobic protozoan parasite that causes giardiasis, one of the most common diarrheal diseases worldwide. Although several drugs are available for the treatment of giardiasis, drug resistance has been reported and is likely to increase, and recurrent infections are common. The search for new drugs that can overcome the drug-resistant strains of Giardia is an unmet medical need. New drug screen methods can facilitate the drug discovery process and aid with the identification of new drug targets. Using a bioluminescent ATP content assay, we have developed a phenotypic drug screen method to identify compounds that act against the actively growing trophozoite stage of the parasite. This assay is homogeneous, robust, and suitable for high-throughput screening of large compound collections. A screen of 4,096 pharmacologically active small molecules and approved drugs revealed 43 compounds with selective anti-Giardia properties, including 32 previously reported and 11 novel anti-Giardia agents. The most potent novel compound was fumagillin, which showed 50% inhibitory concentrations of 10 nM against the WB isolate and 2 nM against the GS isolate.

Structural basis for the mechanism and substrate specificity of glycocyamine kinase, a phosphagen kinase family member.

Glycocyamine kinase (GK), a member of the phosphagen kinase family, catalyzes the Mg(2+)-dependent reversible phosphoryl group transfer of the N-phosphoryl group of phosphoglycocyamine to ADP to yield glycocyamine and ATP. This reaction helps to maintain the energy homeostasis of the cell in some multicelullar organisms that encounter high and variable energy turnover. GK from the marine worm Namalycastis sp. is heterodimeric, with two homologous polypeptide chains, alpha and beta, derived from a common pre-mRNA by mutually exclusive N-terminal alternative exons. The N-terminal exon of GKbeta encodes a peptide that is different in sequence and is 16 amino acids longer than that encoded by the N-terminal exon of GKalpha. The crystal structures of recombinant GKalphabeta and GKbetabeta from Namalycastis sp. were determined at 2.6 and 2.4 A resolution, respectively. In addition, the structure of the GKbetabeta was determined at 2.3 A resolution in complex with a transition state analogue, Mg(2+)-ADP-NO(3)(-)-glycocyamine. Consistent with the sequence homology, the GK subunits adopt the same overall fold as that of other phosphagen kinases of known structure (the homodimeric creatine kinase (CK) and the monomeric arginine kinase (AK)). As with CK, the GK N-termini mediate the dimer interface. In both heterodimeric and homodimeric GK forms, the conformations of the two N-termini are asymmetric, and the asymmetry is different than that reported previously for the homodimeric CKs from several organisms. The entire polypeptide chains of GKalphabeta are structurally defined, and the longer N-terminus of the beta subunit is anchored at the dimer interface. In GKbetabeta the 24 N-terminal residues of one subunit and 11 N-terminal residues of the second subunit are disordered. This observation is consistent with a proposal that the GKalphabeta amino acids involved in the interface formation were optimized once a heterodimer emerged as the physiological form of the enzyme. As a consequence, the homodimer interface (either solely alpha or solely beta chains) has been corrupted. In the unbound state, GK exhibits an open conformation analogous to that observed with ligand-free CK or AK. Upon binding the transition state analogue, both subunits of GK undergo the same closure motion that clasps the transition state analogue, in contrast to the transition state analogue complexes of CK, where the corresponding transition state analogue occupies only one subunit, which undergoes domain closure. The active site environments of the GK, CK, and AK at the bound states reveal the structural determinants of substrate specificity. Despite the equivalent binding in both active sites of the GK dimer, the conformational asymmetry of the N-termini is retained. Thus, the coupling between the structural asymmetry and negative cooperativity previously proposed for CK is not supported in the case of GK.