Crystal structures of a halophilic archaeal malate synthase from Haloferax volcanii and comparisons with isoforms A and G
- Colten D Bracken†1,
- Amber M Neighbor†1,
- Kenneth K Lamlenn1, 3,
- Geoffrey C Thomas1, 4,
- Heidi L Schubert2,
- Frank G Whitby2 and
- Bruce R Howard1Email author
© Bracken et al; licensee BioMed Central Ltd. 2011
Received: 20 December 2010
Accepted: 10 May 2011
Published: 10 May 2011
Malate synthase, one of the two enzymes unique to the glyoxylate cycle, is found in all three domains of life, and is crucial to the utilization of two-carbon compounds for net biosynthetic pathways such as gluconeogenesis. In addition to the main isoforms A and G, so named because of their differential expression in E. coli grown on either acetate or glycolate respectively, a third distinct isoform has been identified. These three isoforms differ considerably in size and sequence conservation. The A isoform (MSA) comprises ~530 residues, the G isoform (MSG) is ~730 residues, and this third isoform (MSH-halophilic) is ~430 residues in length. Both isoforms A and G have been structurally characterized in detail, but no structures have been reported for the H isoform which has been found thus far only in members of the halophilic Archaea.
We have solved the structure of a malate synthase H (MSH) isoform member from Haloferax volcanii in complex with glyoxylate at 2.51 Å resolution, and also as a ternary complex with acetyl-coenzyme A and pyruvate at 1.95 Å. Like the A and G isoforms, MSH is based on a β8/α8 (TIM) barrel. Unlike previously solved malate synthase structures which are all monomeric, this enzyme is found in the native state as a trimer/hexamer equilibrium. Compared to isoforms A and G, MSH displays deletion of an N-terminal domain and a smaller deletion at the C-terminus. The MSH active site is closely superimposable with those of MSA and MSG, with the ternary complex indicating a nucleophilic attack on pyruvate by the enolate intermediate of acetyl-coenzyme A.
The reported structures of MSH from Haloferax volcanii allow a detailed analysis and comparison with previously solved structures of isoforms A and G. These structural comparisons provide insight into evolutionary relationships among these isoforms, and also indicate that despite the size and sequence variation, and the truncated C-terminal domain of the H isoform, the catalytic mechanism is conserved. Sequence analysis in light of the structure indicates that additional members of isoform H likely exist in the databases but have been misannotated.
The glyoxylate cycle, originally described by Kornberg and Krebs , is essential for microorganisms surviving on two-carbon compounds as sole carbon sources. A variant on the tricarboxylic acid cycle (TCA), it allows conversion of two-carbon compounds such as acetate into TCA cycle intermediates, to supply necessary metabolite building blocks such as amino acids and carbohydrates. Two enzymes, isocitrate lyase and malate synthase are unique to the glyoxylate cycle. First, isocitrate lyase cleaves isocitrate to form succinate and glyoxylate, thereby bypassing steps in the TCA cycle that would normally evolve two molecules of CO2. These two carbon atoms instead are maintained as the two-carbon compound glyoxylate, which can then react in a Claison condensation with acetyl-coenzyme A (acetyl-CoA) to form a malyl-CoA intermediate that is subsequently hydrolyzed to produce malate and CoA. This condensation and subsequent hydrolysis are catalyzed by malate synthase. Thus the glyoxylate cycle allows the conversion of one TCA cycle intermediate to two, using two acetyl groups from CoA to form the second. This pathway therefore allows organisms to utilize acetyl groups for net biosynthesis such as in the conversion of oils stored within plant seeds to carbohydrates for the construction of plant tissues during germination. Importantly, the glyoxylate cycle has been shown to contribute to the virulence of several human pathogens including Mycobacterium tuberculosis and Candida albicans, and its absence in humans makes it an attractive target for the development of novel antibacterial and antifungal drugs [4, 5]. Interestingly, this pathway has recently been implicated in the process of fruit ripening .
Malate synthase activity was initially discovered in E. coli. Since then it has been found in a wide range of organisms including many bacteria, plants, and fungi; and even in some animals. Although there is a report that gene sequences coding for malate synthase have been identified in the genome sequences of platypus and opossum , a UniProt database search  shows that there currently are no malate synthase sequences deposited for any reptiles, birds or mammals. While it has been long appreciated that the glyoxylate cycle is distributed widely in bacteria and eukaryotic organisms, it wasn't until recently that it became clear this metabolic pathway is also found in the domain Archaea , and therefore spans all three domains of life.
There are two main isoforms of malate synthase: MSA and MSG, originally identified in E. coli grown on either acetate (A) or glycolate (G) respectively [11, 12]. These two isoforms differ significantly in both size and sequence homology. Members of isoform A comprise ~530 amino acid residues, while those belonging to isoform G comprise ~730 residues. Although the sequence conservation among MSA isoform members, and among MSG members is high (27-99% and 49-98% sequence identity respectively), the sequence identity for structurally conserved regions between these two isoforms is only ~18% . More recently, two examples of malate synthase representing novel isoforms have been found in Archaea [10, 13]. The first example of an archaeal malate synthase was purified from Haloferax volcanii, a halophile originally isolated from the mud of the Dead Sea . This malate synthase, encoded by the aceB gene, comprises only 433 residues, shares very little sequence identity with either the A or G isoform (estimated at 10.2-14.1% and 10.5-12.0% respectively) , and therefore belongs to a third isoform of this enzyme. A BLAST search against the current UniProt database using the H. volcanii sequence as a query indicates a 23% identity with some MSA members found in bacteria of the order actinomycetales, suggesting a closer relationship with the MSA isoform than previously thought. Other examples of isoform H have been identified in genome sequencing projects of halophilic Archaea, including an additional variant in H. volcanii[16, 17]. Since this new isoform is found thus far only in halophilic archaeal organisms, it has been proposed to denote it as isoform H (MSH) , a convention we will continue to use here for comparisons with the other two well-characterized isoforms MSA and MSG.
Comparison of the H isoform with MSA and MSG offers potential insight into the adaptation of this enzyme to a high-salt environment such as found in the Dead Sea. Halophilic archaea have been shown to accumulate KCl to concentrations as high as 4.2 M in order to maintain turgor pressure in such an environment [19, 20]. Proteins within organisms like Haloferax volcanii have acquired characteristics that allow them to be soluble, stable and functional at these high ionic strengths. H. volcanii MSH, for example, displays optimal activity in 3 M KCl , which is similar to levels expected in vivo . One common characteristic of proteins functioning in these high-salt environments is a drastic increase in the number of acidic residues, especially aspartate, and a corresponding decrease in lysine [21–23]. Other characteristics have been described including a decrease in overall hydrophobic content [24–26], increased ion binding , ordered water networks and intermolecular ion pairs [28–30]. Although much attention has focused on the role of increased surface acidity in protein stabilization due to increased binding of water and ions, a recent study in which surface residues of an obligate halophilic protein were systematically mutated to convert it to a non-halophilic protein and also the reverse, indicated that overall protein charge was not vital . Rather it was concluded that halophilicity is directly related to a decrease in the solvent accessible surface. It has been proposed that the increase in aspartate and decrease in lysine residues may be the result of genetic drift with the increased GC content of genomic DNA in halophilic organisms . However, the high GC content among halophiles is not universal [26, 32], and reshuffling of halophilic proteomes at the DNA level demonstrates that the amino acid bias found in halophiles is not a consequence of mononucleotide composition bias .
A fourth isoform of malate synthase has been found in crenarchaeal species which is approximately 100 residues larger than the MSG isoform and shares only low levels of sequence identity with the other three isoforms of malate synthase. The Sulfolobus acidocaldarius malate synthase, for example, is composed of 824 residues and shares only 31% identity with E. coli MSA [13, 33]. Intriguingly, there is no magnesium requirement for catalytic activity of this fourth isoform, and it therefore may function via a mechanism distinct from the other three isoforms .
Structure determinations of MSG [34–37] and MSA  by X-ray crystallography and MSG by nuclear magnetic resonance  have revealed structural and functional similarities and differences. While previously solved structures for MSA and MSG have revealed monomeric enzymes, MSH from H. volcanii has been reported to exist in the native state as a trimer , but was later revised to a tetramer . In order to understand how MSH relates structurally to the larger MSA and MSG isoforms; to clarify the native oligomeric state and understand how it relates to previously solved monomeric versions, and to gain insight into its mechanisms of haloadaptation, we have determined crystal structures of H. volcanii malate synthase in complex with glyoxylate, and also as a ternary complex with acetyl-coenzyme A and pyruvate.
Results and Discussion
H. volcanii MSH Structure
Crystallographic Data and Refinement Statistics
Unit cell dimensions
a = b (Å)
α = β (°)
Number of observations
Wilson B factor (Å2)
Figure of Merit from SOLVE
Figure of Merit after RESOLVE
Cullis R factor
Bond lengths (Å)
Bond angles (°)
Mean isotropic B factor (Å2)
Most favored (%)
Additional allowed (%)
Generously allowed (%)
0.3 (Thr 276)
0.3 (Glu 24)
0.3 (Glu 24)
Two lead binding sites were used for phasing, and a third lower-occupancy site was identified during model building and refinement of the lead derivative. Two of the sites were found at or near intersubunit interfaces, which may explain the higher-resolution diffraction of the derivative. Unfortunately, lead substitution of the required magnesium ion within the active site precluded its use as a higher-resolution pseudo-native structure.
The native structure (3 mM glyoxylate) was refined to an Rfactor of 0.202 and an Rfree of 0.263. Two substantial loops, residues 283-330 and 355-386 were not ordered in the crystal and have not been included in the refined model, which comprises residues 2-22, 25-282, 331-354, and 387-432, one glyoxylate molecule, one magnesium ion, four potassium ions, four chloride ions, and 71 water molecules. The stereochemistry is satisfactory with no residues in the generously allowed or disallowed regions of the Ramachandran plot (Table 1). Distorted magnesium coordination geometry, apparently caused by binding of an adjacent potassium ion, coupled with high B-factors for Mg2+ and glyoxylate (78 and 72-79 respectively), prompted us to modify conditions to obtain a high-occupancy glyoxylate complex.
The high-occupancy glyoxylate complex was prepared by increasing the concentrations of MgCl2 from 13 mM to ~0.1 M and glyoxylate from 3 mM to ~0.1 M in mother liquor after crystal growth was complete and one week prior to data collection. The final model has a crystallographic Rfactor of 0.195 and an Rfree of 0.248, and is comprised of residues 5-283, 331-353, and 387-432, two glyoxylate molecules, one magnesium ion, three potassium ions, five chloride ions and 134 water molecules. One glyoxylate molecule is bound to the Mg2+ at the active site and the other is bound weakly in the position at which the acetyl-CoA thioester resides in the ternary complex (below). The structure is in good agreement with expected stereochemistry, with only one residue in the generously allowed region (Thr 276) and one residue in the disallowed region (Glu 24) of the Ramachandran plot (Table 1).
The pyruvate/acetyl-CoA ternary complex was prepared by soaking crystals in mother liquor containing 50 mM MgCl2, and supplemented with ~70 mM pyruvate and ~0.15 M acetyl-CoA one week before data collection. This structure was refined to a crystallographic Rfactor of 0.205 and an Rfree of 0.239. One loop not visible in the glyoxylate complex becomes significantly ordered in the pyruvate/acetyl-CoA complex, and the refined model comprises residues 5-284, 328-371, 381-432, one molecule of pyruvate, one molecule of acetyl-CoA, one magnesium ion, three potassium ions, four chloride ions, a phosphate ion and 176 water molecules. There is only one residue, Glu 24, in the generously allowed region of the Ramachandran plot, and none in the disallowed regions (Table 1).
The native oligomerization state of hvMSH was initially reported to be a trimer based on gel-filtration mobility and SDS PAGE analysis with estimated molecular weights of 200 ± 30 kDa and 67 ± 4 kDa for the native enzyme and individual subunits respectively . But after the aceB gene was cloned, it became clear that individual subunits were actually only 47.9 kDa leading to a revised prediction of a tetrameric assembly . This abnormally slow SDS PAGE mobility is a common characteristic of halophilic proteins which have an excess of acidic residues .
Comparison of H. volcanii MSH overall structure with E. coli MSA and MSG
Molecular overlays of the hvMSH ternary complex onto the corresponding ternary complexes of ecMSA [PDB:3CV2]  and ecMSG [PDB:1P7T]  were performed with SSM . The overlays used the entire model for each structure involved, and resulted in 271 residues aligning between hvMSH and ecMSA, with an 18.8% identity and a root-mean-square deviation of 1.90 Å for aligned alpha carbons. A similar number of residues aligned between hvMSH and ecMSG: 262 residues with a 17.2% identity and rmsd = 1.85 Å for structurally equivalent Cα positions.
The ends of the protein segment connecting the TIM barrel and the C-terminal domain that are visible in the hvMSH ternary complex suggest this connection is quite different from those of MSA and MSG. The last structurally equivalent residue in the TIM barrel among these three isoforms is found at the completion of the eighth and final helix: Leu 272, Asn 380, and His 549 in hvMSH, ecMSA and ecMSG respectively. Two of the next three residues in hvMSH are proline (PPK) with the trajectory of the backbone in essentially the opposite direction as those of the comparable segments in MSA and MSG. The next residue that is structurally common to all three is near the beginning of the first α-helix in the C-terminal domain of the MSA and MSG isoforms: Ser 342, Gly 417, and Glu 595 in hvMSH, ecMSA and ecMSG respectively. Again, the direction of the trajectory of the preceding segment in hvMSH is quite different from those of MSA and MSG, essentially orthogonal (Figure 5b). The connection between the last common structure in the TIM barrel, and that of the first common structure in the C-terminal domain among these three isoforms consists of 69 residues in hvMSH, 36 in ecMSA and 45 in ecMSG. Of these 69 residues in hvMSH, 43 are disordered in the crystal structure, preventing a more detailed comparison for this region.
The overlays reveal that the C-terminal domain of hvMSH, which caps the active site of the TIM barrel, is quite different from those found in ecMSA and ecMSG. This C-terminal domain, consisting largely of a bundle of five α-helices, is closely related in MSA and MSG. However, only two of these α-helices are structurally conserved in hvMSH, connected by a β-hairpin which is also conserved among all three isoforms (Figure 5b). While the β-hairpin is conserved, the length of each hairpin varies substantially, with that of hvMSH fifteen residues longer and ecMSG five residues longer than the hairpin in ecMSA. Only the two ends of each β-strand at the base of the hairpin superimpose closely in all three structures, along with the C-terminal end of the preceding α-helix (helix 1), and the N-terminal end of the following helix (helix 2) (Figure 5b). This close structural alignment is an important region of the C-terminal domain since it contributes the catalytic base to the active site. This catalytic base, Asp 388 in hvMSH, resides at the junction between the second strand of the β-hairpin, and helix 2 of the C-terminal domain. The backbone carbonyl of the preceding residue, Trp 387, is involved in the last H-bond in the β-hairpin, while the backbone carbonyl of Asp 388 accepts the first H-bond of the helix. It is in a position which might be expected to cap this helix, but the backbone geometry prevents it from forming an H-bond to the amide NH (N-O distances are 3.6, 3.8 and 3.7 Å for hvMSH, ecMSA and ecMSG respectively). While the N-terminus of this helix aligns fairly well in all three structures, they eventually diverge at the C-terminal end with the helix in hvMSH having a drastic bend in the middle due to Pro 398. Helix 2 in ecMSA is also slightly bent although no proline residues are present, but is not bent in ecMSG. A comparison of the remaining segment of each protein following their roughly common departure point from this helix, shows that while ecMSA and ecMSG are quite similar, the structure adopted by the C-terminal residues of hvMSH is radically different and is also significantly shorter (Figure 5c). This final segment of the protein in hvMSH is 41 residues shorter than ecMSA, and 47 residues shorter than ecMSG, contributing to its smaller overall size.
These observations suggest the possibility that the N-terminal deletion in MSH and oligomerization are related. One possible scenario would be that an ancestral monomeric enzyme acquired mutations that destabilized the interactions between the N-terminal sequences and barrel and at the same time favored a weak intersubunit aggregation. A displaced N-terminal domain would have then become expendable since exposed regions of the barrel surface would be buried and any potentially stabilizing effects to the enzyme structure could have been satisfied by interactions with neighboring subunits. These interactions, fine-tuned by natural selection, would then allow for a functional, soluble enzyme in the event of an N-terminal deletion in the gene. Of course, this is only one possible scenario, and the reverse can also be imagined where an oligomeric enzyme acquired an N-terminal extension which was able to compete for and replace subunit interactions to become a stable monomer. Regardless of the actual process involved, the structural comparisons are consistent with an evolutionary model in which N-terminal deletion and oligomerization are coupled. It will be interesting to see future structural determinations of oligomeric forms of MSA, which presumably still have N-terminal domains yet form stable multimers [43–45], to understand how they have adapted to interact with neighboring subunits.
Comparison of the active site of H. volcanii MSH with those of E. coli MSA and MSG
Acetyl-CoA binding site
The acetyl-CoA binding sites in MSH, MSA and MSG are located at structurally equivalent positions. Acetyl-CoA binds to hvMSH in a bent conformation similar to that seen in the ecMSG ternary complex (Figure 9b). In both cases an intramolecular hydrogen bond forms between adenine N7 and the hydroxyl group of the pantothenate moiety. There are also two hydrogen bonds between the exocyclic N6 of the adenine ring and two backbone carbonyls that are structurally conserved in all three isoforms. One of these carbonyls in the hvMSH structure also forms a hydrogen bond (3.0 Å O-N bond distance) to the amide NH of the pantothenate moiety of the acetyl-CoA. The comparable interaction is not observed in the ecMSG ternary complex (3.7 Å O-N distance). Unfortunately the pantothenate, β-mercaptoethylamine and acetyl portions of acetyl-CoA were not visible in the ecMSA ternary complex , precluding a direct comparison in these regions of the acetyl-CoA binding site. Additionally, there is a hydrogen bond between adenine N1 and the side chain hydroxyl of Ser 17 in the hvMSH structure which is absent in both the ecMSA and ecMSG adenine binding sites. The structurally equivalent positions are Gly 96 and Val 119 in ecMSA and ecMSG respectively. The adenine ring binds in a hydrophobic pocket against a helix-capping proline as seen in both ecMSA and ecMSG (Pro 261, Pro 369, and Pro 538 in hvMSH, ecMSA and ecMSG respectively). Adjacent to Pro 261, the side chain of Phe15 contributes to the hydrophobic pocket on the same side of the adenine ring. The structurally equivalent position is Ile 94 in ecMSA and Leu 117 in ecMSG. Met 30 packs against the opposite side of the adenine ring, and is structurally equivalent to Met 102 in ecMSA and Tyr 126 in ecMSG. The terminal carbonyl of the pantothenate moiety forms a hydrogen bond to the side chain of Thr 16 in HvMSH. This same position in ecMSA is Thr 95 and therefore may form a similar interaction, but is Val 118 in ecMSG. Met 508 of ecMSG (Met 330 in ecMSA), which forms a hydrophobic interaction for the β-mercaptoethylamine portion of acetyl-CoA, is replaced by Pro 231 in hvMSH, which also interacts with the methyl group of the pyruvate molecule bound in the active site. The hydrophobic surface formed by Met 508 in ecMSG is partially formed by Leu 259 in the hvMSH structure.
Cys 617 in ecMSG and Cys 438 in ecMSA have both been observed to be oxidized to sulfenic acid in crystal structures of these enzymes, suggesting a potential catalytic and/or regulatory function [4, 36]. The equivalent position in hvMSH is Val 119, with no cysteine residues occurring in the active site. There is only one cysteine residue in hvMSH (Cys 225), which is located on the opposite end of the TIM barrel from the active site. Even this single cysteine is not conserved among MSH isoform members, apparently eliminating the possibility of a potentially similar type of redox regulatory function in this isoform.
Structural Overlay of HvMSH glyoxylate and ternary complexes
A plausible catalytic mechanism for malate synthase was first proposed based on the crystal structure of the glyoxylate complex of MSG from E. coli [PDB:1D8C] , with Asp 631 acting as a catalytic base to deprotonate the methyl group of acetyl-CoA to form an enol(ate) intermediate stabilized by Arg 338 (corresponding residues in hvMSH are Asp 388 and Arg 84). The enol(ate) was proposed to swing down to attack the aldehyde carbon of glyoxylate to form the malyl-CoA intermediate which is subsequently hydrolyzed. This mechanism is consistent with the observed inversion of configuration of the acetyl-CoA methyl group during the reaction [47, 48], and has been supported by subsequent crystal structures of malate synthases in complex with substrates and inhibitors [4, 35–37]. Additionally, site directed mutagenesis has confirmed the importance of both Asp 631 and Arg 338 for catalytic activity . Asp 631 was shown to be absolutely essential, with a D631N mutation rendering the enzyme activity unmeasurable, while Arg 338 could be replaced by lysine with activity reduced to 6.6% of wild type. The structures reported here are also consistent with this proposed mechanism, having identical catalytic and magnesium-coordinating residues observed in all previously determined malate synthase structures. While the structure of the active site of the glyoxylate complex reported here is very similar to other previously determined glyoxylate complexes [34, 35], the structure of the hvMSH ternary complex appears to add a novel observation addressing the catalytic mechanism. There is only one previously determined malate synthase structure in which the terminal region of acetyl-CoA has been seen in electron density maps, to allow the position of the acetyl group in the active site to be identified: the E. coli MSG ternary complex . In this structure [PDB:1P7T] the terminal methyl group of acetyl-CoA is making a close contact with the proposed catalytic base Asp 631, refining to a C-O distance of 2.78 Å. The distance from this methyl carbon to the ketone carbonyl carbon of pyruvate refined to 3.16 Å. The close contact with the catalytic base supports the proposal that Asp 631 acts to deprotonate the terminal methyl group in the enolization step of the reaction. In the ternary complex reported here for hvMSH, however, the acetyl group is seen to bind in a different relative position in the active site. The distance between the carboxylate oxygen of the catalytic aspartate (Asp 631 and Asp 388, in ecMSG and hvMSH respectively) and the ketone carbonyl carbon of pyruvate is similar in both complexes: ~5.9 Å and ~6.2 Å respectively for ecMSG and hvMSH. But rather than forming a close contact with the catalytic base as seen in the ecMSG ternary complex, the terminal methyl carbon of acetyl-CoA instead forms a close contact (2.46 Å refined distance) with the electrophilic keto carbon in pyruvate, and is ~4.1 Å from the reactive oxygen of the catalytic base (Figure 8b, 11b). It appears that the hvMSH ternary complex is well along the reaction coordinate of a nucleophilic attack on pyruvate, with the contact distance intermediate between that of a van der Waals interaction and a covalent bond.
Enolization of acetyl-CoA has been demonstrated for yeast malate synthase which, in the presence of pyruvate, catalyzes isotopic exchange between acetyl-CoA and tritiated water [44, 49]. Therefore, the enol(ate) form of acetyl-CoA is expected to exist at least transiently in the ternary complex. But if the structure does indeed show the enolate intermediate in the process of bond formation with the carbonyl carbon, why is it arrested along the reaction pathway? Pyruvate, while able to stimulate the enolization of acetyl-CoA, is in fact an inhibitor of malate synthase, unable to complete the reaction . The forced expansion of the active site described in the previous section and the close contacts the methyl group of pyruvate makes with Pro 231 and Trp 257 appear to prevent the formation of the tetrahedral geometry required for the condensation reaction. Whereas there is plenty of space for a hydrogen atom attached to the electrophilic carbonyl in glyoxylate to drop below the plane to form the tetrahedral transition state, the methyl group appears constrained. This is analogous to the situation seen in complexes of bovine pancreatic trypsin inhibitor and trypsin where the active site serine oxygen makes a close contact (~2.6 Å) to the peptide carbonyl carbon of BPTI, but is prevented from completely reacting by the constraints imposed by the enzyme and inhibitor, thus freezing the process at an intermediate state of the nucleophilic addition reaction [50, 51]. Similar reaction intermediates interpreted as nucleophilic addition reactions proceeding to varying extents have been observed in small molecule crystal structures containing nucleophilic nitrogen atoms and electrophilic carbonyl groups, with nitrogen-carbonyl carbon distances ranging from 2.9 to 1.5 Å . As in the analysis of the trypsin/protein inhibitor complexes, these cases were interpreted to arise from the constraints imposed by the crystal environment which froze the addition reaction at intermediate points along the reaction coordinate. An analysis of these structures led to insight into the reaction pathway that was confirmed by theoretical calculations and improved understanding of the process . Thus, we interpret the close contact in our ternary complex to represent the enolate intermediate of acetyl-CoA caught in the process of bond formation with the carbonyl carbon of pyruvate, but unable to complete the process due to steric hindrance. This implies that removal of atoms responsible for the steric hindrance would allow the reaction to proceed to completion. Therefore, we would expect the double mutant W257H, P231A, if still folding competent, to acquire the ability to catalyze acetyl transfer from acetyl-CoA to pyruvate.
As expected, hvMSH exhibits characteristics similar to those seen in other halophilic proteins. It has a marked increase in acidic amino acids with 95 of the 433 residues being either glutamic or aspartic acid making the protein 21.9% acidic. This is consistent with other halophilic proteins , however hvMSH contains a greater amount of glutamic acid residues (55) than aspartic acid residues (40). By comparison, ecMSA and ecMSG are 13.5% and 12.3% acidic respectively. Utilizing PISA (Protein Interactions, Surface, and Assembly)  to analyze the trimeric assembly, it was determined that of the 78 acidic residues per subunit that are ordered in the ternary complex, 41 are solvent accessible, 35 are buried at intersubunit interfaces, and two are inaccessible, making ~52% of all ordered acidic residues accessible to solvent. Of the 159 total residues in each monomer accessible to solvent in the trimeric assembly of hvMSH, 25.8% are acidic. The single cysteine and the nine lysine residues found in hvMSH are also consistent with what is seen in proteome surveys of halophilic organisms, which show that halophilic proteins have an underrepresentation of cysteine and lysine . The number of expected cysteine and lysine residues for a protein of this size, based on the average occurance typically found in proteins (1.9 and 5.9% respectively)  would be approximately 8 and 25.
H. volcanii MSH also demonstrates a substantial number of intermolecular ion pairs. An analysis of the three different protein interfaces present in the trimeric and hexameric assemblies showed that the interface between monomers of the trimer contains six intermolecular salt bridges. Of the two interfaces per subunit between the two trimers in the hexameric assembly, one has no salt bridges, while the other has eight. Thus the total number of intermolecular salt bridges stabilizing the trimer is 18 (six at each interface). The hexamer is stabilized by an additional 24 intermolecular salt bridges (eight at each pair of subunits across the interface) for a total of 60 in the hexameric assembly. H. volcanii MSH also is seen to bind a number of solvent ions: three potassium and 4 or 5 chloride ions per subunit in the pyruvate/acetyl-CoA ternary complex and the high-occupancy glyoxylate complex respectively, with one K+ and one Cl- ion bound at the trimer interface. Interestingly, the ternary complex binds a phosphate ion along the three-fold axis of the trimer at the same position of the fifth chloride ion that is observed in the glyoxylate complex.
Sequence analysis in light of the structure
Protein sequences closely related to H. volcanii MSH [UniProt:Q977U4]
Halobacterium volcanii (Haloferax volcanii)
Haloferax volcanii (strain ATCC 29605/DSM 3757/IFO 14742/NCIMB 2012/DS2)
Haloarcula marismortui (Halobacterium marismortui)
Citrate Lyase Beta Subunit
Halogeometricum borinquense (strain ATCC 700274/DSM 11551/JCM 10706/PR3)
Haloterrigena turkmenica (strain ATCC 51198/DSM 5511/NCIMB 13204/VKM B-1734) (Halococcus turkmenicus)
Haloferax volcanii (strain ATCC 29605/DSM 3757/IFO 14742/NCIMB 2012/DS2)
Haloquadratum walsbyi (strain DSM 16790)
Homolog to citryl-CoA lyase
Natronomonas pharaonis (strain DSM 2160/ATCC 35678)
Natrialba magadii (strain ATCC 43099/DSM 3394/NCIMB 2190/MS3) (Natronobacterium magadii)
The structures reported here for the glyoxylate and the pyruvate/acetyl-CoA complexes of Haloferax volcanii malate synthase represent the first examples of an H isoform member. Instead of the expected tetramer , a trimer is found to be the major state in solution, although an equilibrium with a significant hexamer population is evident.
The overall structure of hvMSH reveals that, like MSA and MSG, this halophilic isoform is based on a TIM barrel and indicates that deletion in hvMSH of an N-terminal domain distinguishes this isoform from those of MSA and MSG; and that the surface of the barrel normally buried by this domain and connecting loops is instead involved in trimeric and hexameric interfaces, suggesting a potential evolutionary coupling of the N-terminal deletion and oligomerization.
Despite the sequence divergence and overall smaller size of hvMSH compared to MSA and MSG, the active site and catalytic mechanism are conserved in all three isoforms. In the ternary complex of hvMSH, however, the position of the terminal methyl group of acetyl-CoA is found to differ considerably from that seen in the ecMSG ternary complex. Instead of a structure corresponding to the deprotonation step by the catalytic aspartate as seen in ecMSG , the ternary complex of hvMSH reveals this methyl group interacting closely with the carbon atom of the electrophilic carbonyl of pyruvate, in an apparent nucleophilic attack arrested by steric hindrance. Therefore, the ternary complexes of ecMSG and hvMSH are complementary, revealing the active site configurations for two important steps in the catalytic mechanism: proton abstraction by the catalytic base, and nucleophilic attack of the enolate intermediate on the electrophilic substrate.
Haloferax volcanii Malate synthase encoded by the aceB1 gene was produced and purified as previously described [10, 18]. Briefly, a lyophilized sample of Haloferax volcanii was obtained from the American Type Culture Collection , and grown at 37°C in a chemically-defined medium with acetate as a sole carbon source to induce expression of glyoxylate cycle enzymes. Cells were lysed by sonication on ice. Protein purification was performed at 4°C using three chromatographic steps: reverse phase, anion-exchange and gel-filtration as previously described . Calibration of the Sephacryl-300 sizing column (Pharmacia) was performed with gel-filtration standards from Bio-Rad: Thyroglobulin (bovine), 670 kDa; γ-globulin (bovine), 158 kDa; Ovalbumin (chicken), 44 kDa; Myoglobin (horse), 17 kDa; and Vitamin B12, 1.35 kDa. Progress was monitored by silver-stained SDS PAGE analysis and enzyme activity assays. Average yield was 0.5 mg of ~90% pure enzyme per liter of cell culture.
Enzyme Activity Assay
Malate synthase activity was measured by monitoring the loss of absorbance at 232 nm upon acetyl-CoA thioester cleavage as previously described [18, 49]. The reaction conditions were 0.34 mM acetyl-CoA, 1.1 mM glyoxylate, 20 mM Tris pH 8.0, 2 mM EDTA, 3 M KCl, and 5 mM MgCl2. The reaction was initiated by the addition of 10 μL of enzyme solution into a 1 mL total reaction volume.
Crystallization, and heavy atom derivatization
Crystals of H. volcanii MSH were grown at room temperature in sitting drops as previously described . The protein solution contained H. volcanii malate synthase at 7 mg/mL, 13 mM MgCl2, 3 mM glyoxylate, 50 mM Tris·HCl pH 8.0, and 2 M KCl. The well solution contained 0.17 M ammonium acetate, 24.5-27% w/v PEG 4500, 15% glycerol, and 0.085 M sodium acetate trihydrate at a pH of 4.4-5.0. Two microliters of protein solution were mixed with an equal volume of well solution, and allowed to equilibrate at room temperature. Crystals grew over a period of approximately two weeks. A lead derivative was prepared by addition of 0.4 μl of a saturated lead (II) acetate solution to an equilibrated drop after crystal growth was complete. A high-occupancy glyoxylate complex was produced by increasing the concentrations of MgCl2 and glyoxylate to ~0.1 M in drops of mother liquor containing fully grown crystals. The ternary complex of pyruvate and acetyl-CoA was produced using the same well solution as above, and a protein solution containing H. volcanii malate synthase at 7 mg/mL, 50 mM MgCl2, 50 mM Tris·HCl pH 8.0, and 2 M KCl. Pyruvate and acetyl-CoA were added to equilibrated drops of mother liquor following crystal growth to ~70 mM and ~0.15 M respectively.
Data collection, processing, phasing and structure determination
Crystals were suspended in nylon loops and cryocooled by plunging into liquid nitrogen. Data were collected at 100 K on an R-axis IV detector using Copper Kα radiation produced by a Rigaku 007 HF rotating anode generator equipped with Osmic confocal X-ray optics. Data were indexed, integrated and scaled with the HKL2000 package . Phasing was carried out with SOLVE  using the single isomorphous replacement with anomalous scattering (SIRAS) method using the lead derivative data and the 2.7 Å native data (both at 3 mM glyoxylate) (Table 1), with subsequent density modification using RESOLVE . Model building into the experimental map was performed manually with COOT  and model refinement with REFMAC5 [65, 66]. High B-factors for Mg2+ and glyoxylate and a distorted magnesium coordination sphere instigated a pursuit of conditions to drive a high-occupancy complex. The partially refined protein model (3 mM glyoxylate) comprising virtually all the ordered residues (6-281, 331-353, 387-432) was used for molecular replacement using PHASER [66, 67] followed by cycles of manual rebuilding and refinement to solve both the high-occupancy glyoxylate complex and the pyruvate/acetyl-CoA ternary complex. The atomic coordinates and structure factors have been deposited in the PDB  with accession numbers 3PUG for the native structure (3 mM glyoxylate), 3OYX for the high-occupancy glyoxylate complex, and 3OYZ for the ternary pyruvate/acetyl-CoA complex.
Figures were made with PyMol (DeLano Scientific; http://www.pymol.org). Analysis of protein interfaces and buried surface area calculations were carried out with PISA . Sequence alignments were conducted with ClustalW . Structural alignments were performed using SSM  and least squares superposition (LSQ) in COOT [46, 64].
E. coli malate synthase isoform A
E. coli malate synthase isoform G
H. volcanii malate synthase isoform H
Protein data bank
Universal protein resource
- SDS PAGE:
Sodium dodecyl sulfate polyacrylamide gel electrophoresis
Basic local alignment search tool
Tricarboxylic acid or citric acid cycle
Triose phosphate isomerase
Data were collected at the University of Utah Macromolecule Crystallography Core Facility. We thank Chris Hill for helpful comments on the manuscript, and the provost's faculty development program at Southern Utah University for financial support of this project.
- Kornberg HL, Krebs HA: Synthesis of cell constituents from C2-units by a modified tricarboxylic acid cycle. Nature 1957, 179(4568):988–991. 10.1038/179988a0View ArticlePubMed
- McKinney JD, Honer zu Bentrup K, Munoz-Elias EJ, Miczak A, Chen B, Chan WT, Swenson D, Sacchettini JC, Jacobs WR Jr, Russell DG: Persistence of Mycobacterium tuberculosis in macrophages and mice requires the glyoxylate shunt enzyme isocitrate lyase. Nature 2000, 406(6797):735–738. 10.1038/35021074View ArticlePubMed
- Lorenz MC, Fink GR: The glyoxylate cycle is required for fungal virulence. Nature 2001, 412(6842):83–86. 10.1038/35083594View ArticlePubMed
- Lohman JR, Olson AC, Remington SJ: Atomic resolution structures of Escherichia coli and Bacillus anthracis malate synthase A: Comparison with isoform G and implications for structure-based drug discovery. Protein Science 2008, 17(11):1935–1945. 10.1110/ps.036269.108PubMed CentralView ArticlePubMed
- Smith CV, Sharma V, Sacchettini JC: TB drug discovery: addressing issues of persistence and resistance. Tuberculosis (Edinb) 2004, 84(1–2):45–55. 10.1016/j.tube.2003.08.019View Article
- Pua EC, Chandramouli S, Han P, Liu P: Malate synthase gene expression during fruit ripening of Cavendish banana ( Musa acuminata cv. Williams). J Exp Bot 2003, 54(381):309–316. 10.1093/jxb/54.381.309View ArticlePubMed
- Wong DTO, Ajl SJ: Conversion of acetate and glyoxylate to malate. J Am Chem Soc 1956, 78: 3230–3231. 10.1021/ja01594a079View Article
- Kondrashov FA, Koonin EV, Morgunov IG, Finogenova TV, Kondrashova MN: Evolution of glyoxylate cycle enzymes in Metazoa: evidence of multiple horizontal transfer events and pseudogene formation. Biol Direct 2006, 1: 31. 10.1186/1745-6150-1-31PubMed CentralView ArticlePubMed
- Consortium TU: The Universal Protein Resource (UniProt) in 2010. Nucleic Acids Research 2009, (38 Database):D142-D148.
- Serrano JA, Camacho M, Bonete MJ: Operation of glyoxylate cycle in halophilic archaea: presence of malate synthase and isocitrate lyase in Haloferax volcanii . FEBS Lett 1998, 434(1–2):13–16. 10.1016/S0014-5793(98)00911-9View ArticlePubMed
- Falmagne P, Vanderwinkel E, Wiame JM: Demonstration of 2 Malate Synthases in Escherichia Coli . Biochim Biophys Acta 1965, 99: 246–258.View ArticlePubMed
- Vanderwinkel E, De Vlieghere M: Physiology and genetics of isocitritase and the malate synthases of Escherichia coli . Eur J Biochem 1968, 5(1):81–90. 10.1111/j.1432-1033.1968.tb00340.xView ArticlePubMed
- Uhrigshardt H, Walden M, John H, Petersen A, Anemuller S: Evidence for an operative glyoxylate cycle in the thermoacidophilic crenarchaeon Sulfolobus acidocaldarius . FEBS Lett 2002, 513(2–3):223–229. 10.1016/S0014-5793(02)02317-7View ArticlePubMed
- Mullakhanbhai MF, Larsen H: Halobacterium volcanii spec. nov., a Dead Sea halobacterium with a moderate salt requirement. Arch Microbiol 1975, 104(3):207–214.View ArticlePubMed
- Serrano JA, Bonete MJ: Sequencing, phylogenetic and transcriptional analysis of the glyoxylate bypass operon (ace) in the halophilic archaeon Haloferax volcanii . Biochim Biophys Acta 2001, 1520(2):154–162.View ArticlePubMed
- Hartman AL, Norais C, Badger JH, Delmas S, Haldenby S, Madupu R, Robinson J, Khouri H, Ren Q, Lowe TM, Maupin-Furlow J, Pohlschroder M, Daniels C, Pfeiffer F, Allers T, Eisen JA: The Complete Genome Sequence of Haloferax volcanii DS2, a Model Archaeon. PLoS One 2010, 5(3):e9605. 10.1371/journal.pone.0009605PubMed CentralView ArticlePubMed
- Baliga NS, Bonneau R, Facciotti MT, Pan M, Glusman G, Deutsch EW, Shannon P, Chiu Y, Weng RS, Gan RR, Hung P, Date SV, Marcotte E, Hood L, Ng WV: Genome sequence of Haloarcula marismortui : a halophilic archaeon from the Dead Sea. Genome Res 2004, 14(11):2221–2234. 10.1101/gr.2700304PubMed CentralView ArticlePubMed
- Thomas G, Lamlenn K, Howard BR: Crystallization and preliminary x-ray diffraction of a halophilic archaeal malate synthase. American Journal of Undergraduate Research 2009, 8(2 & 3):15–23.
- Christian JH, Waltho JA: Solute concentrations within cells of halophilic and non-halophilic bacteria. Biochim Biophys Acta 1962, 65: 506–508. 10.1016/0006-3002(62)90453-5View ArticlePubMed
- Ginzburg M, Sachs L, Ginzburg BZ: Ion metabolism in a Halobacterium. I. Influence of age of culture on intracellular concentrations. J Gen Physiol 1970, 55(2):187–207. 10.1085/jgp.55.2.187PubMed CentralView ArticlePubMed
- Lanyi JK: Salt-dependent properties of proteins from extremely halophilic bacteria. Bacteriol Rev 1974, 38(3):272–290.PubMed CentralPubMed
- Fukuchi S, Yoshimune K, Wakayama M, Moriguchi M, Nishikawa K: Unique amino acid composition of proteins in halophilic bacteria. J Mol Biol 2003, 327(2):347–357. 10.1016/S0022-2836(03)00150-5View ArticlePubMed
- Kastritis PL, Papandreou NC, Hamodrakas SJ: Haloadaptation: insights from comparative modeling studies of halophilic archaeal DHFRs. Int J Biol Macromol 2007, 41(4):447–453. 10.1016/j.ijbiomac.2007.06.005View ArticlePubMed
- Eisenberg H, Mevarech M, Zaccai G: Biochemical, structural, and molecular genetic aspects of halophilism. Adv Protein Chem 1992, 43: 1–62.View ArticlePubMed
- Jaenicke R: Protein stability and molecular adaptation to extreme conditions. Eur J Biochem 1991, 202(3):715–728. 10.1111/j.1432-1033.1991.tb16426.xView ArticlePubMed
- Paul S, Bag SK, Das S, Harvill ET, Dutta C: Molecular signature of hypersaline adaptation: insights from genome and proteome composition of halophilic prokaryotes. Genome Biol 2008, 9(4):R70. 10.1186/gb-2008-9-4-r70PubMed CentralView ArticlePubMed
- Ebel C, Costenaro L, Pascu M, Faou P, Kernel B, Proust-De Martin F, Zaccai G: Solvent interactions of halophilic malate dehydrogenase. Biochemistry 2002, 41(44):13234–13244. 10.1021/bi0258290View ArticlePubMed
- Richard SB, Madern D, Garcin E, Zaccai G: Halophilic adaptation: novel solvent protein interactions observed in the 2.9 and 2.6 Å resolution structures of the wild type and a mutant of malate dehydrogenase from Haloarcula marismortui . Biochemistry 2000, 39(5):992–1000. 10.1021/bi991001aView ArticlePubMed
- Dym O, Mevarech M, Sussman JL: Structural features that stabilize halophilic malate dehydrogenase from an archaebacterium. Science 1995, 267(5202):1344–1346. 10.1126/science.267.5202.1344View ArticlePubMed
- Frolow F, Harel M, Sussman JL, Mevarech M, Shoham M: Insights into protein adaptation to a saturated salt environment from the crystal structure of a halophilic 2Fe-2S ferredoxin. Nat Struct Biol 1996, 3(5):452–458. 10.1038/nsb0596-452View ArticlePubMed
- Tadeo X, Lopez-Mendez B, Trigueros T, Lain A, Castano D, Millet O: Structural basis for the aminoacid composition of proteins from halophilic archea. PLoS Biol 2009, 7(12):e1000257. 10.1371/journal.pbio.1000257PubMed CentralView ArticlePubMed
- Bolhuis H, Palm P, Wende A, Falb M, Rampp M, Rodriguez-Valera F, Pfeiffer F, Oesterhelt D: The genome of the square archaeon Haloquadratum walsbyi : life at the limits of water activity. BMC Genomics 2006, 7: 169. 10.1186/1471-2164-7-169PubMed CentralView ArticlePubMed
- Chen L, Brügger K, Skovgaard M, Redder P, She Q, Torarinsson E, Greve B, Awayez M, Zibat A, Klenk HP, Garrett RA: The genome of Sulfolobus acidocaldarius , a model organism of the Crenarchaeota. J Bacteriol 2005, 187(14):4992–4999. 10.1128/JB.187.14.4992-4999.2005PubMed CentralView ArticlePubMed
- Howard BR, Endrizzi JA, Remington SJ: Crystal structure of Escherichia coli malate synthase G complexed with magnesium and glyoxylate at 2.0 Å resolution: mechanistic implications. Biochemistry 2000, 39(11):3156–3168. 10.1021/bi992519hView ArticlePubMed
- Smith CV, Huang CC, Miczak A, Russell DG, Sacchettini JC, Honer zu Bentrup K: Biochemical and structural studies of malate synthase from Mycobacterium tuberculosis . J Biol Chem 2003, 278(3):1735–1743. 10.1074/jbc.M209248200View ArticlePubMed
- Anstrom DM, Kallio K, Remington SJ: Structure of the Escherichia coli malate synthase G:pyruvate:acetyl-coenzyme A abortive ternary complex at 1.95 Å resolution. Protein Science 2003, 12(9):1822–1832. 10.1110/ps.03174303PubMed CentralView ArticlePubMed
- Anstrom DM, Remington SJ: The product complex of M. tuberculosis malate synthase revisited. Protein Science 2006, 15(8):2002–2007. 10.1110/ps.062300206PubMed CentralView ArticlePubMed
- Grishaev A, Tugarinov V, Kay LE, Trewhella J, Bax A: Refined solution structure of the 82-kDa enzyme malate synthase G from joint NMR and synchrotron SAXS restraints. J Biomol NMR 2008, 40(2):95–106. 10.1007/s10858-007-9211-5View ArticlePubMed
- Yonezawa Y, Tokunaga H, Ishibashi M, Tokunaga M: Characterization of nucleoside diphosphate kinase from moderately halophilic eubacteria. Biosci Biotechnol Biochem 2001, 65(10):2343–2346. 10.1271/bbb.65.2343View ArticlePubMed
- Krissinel E, Henrick K: Inference of macromolecular assemblies from crystalline state. J Mol Biol 2007, 372(3):774–797. 10.1016/j.jmb.2007.05.022View ArticlePubMed
- Winzor DJ: Analytical exclusion chromatography. J Biochem Biophys Methods 2003, 56(1–3):15–52. 10.1016/S0165-022X(03)00071-XView ArticlePubMed
- Krissinel E, Henrick K: Secondary-structure matching (SSM), a new tool for fast protein structure alignment in three dimensions. Acta Crystallogr D Biol Crystallogr 2004, 60(Pt 12 Pt 1):2256–2268.View ArticlePubMed
- Woodcock E, Merrett MJ: Purification and immunochemical characterization of malate synthase from Euglena gracilis . Biochem J 1978, 173(1):95–101.PubMed CentralView ArticlePubMed
- Durchschlag H, Biedermann G, Eggerer H: Large-scale purification and some properties of malate synthase from baker's yeast. Eur J Biochem 1981, 114(2):255–262. 10.1111/j.1432-1033.1981.tb05144.xView ArticlePubMed
- Beeckmans S, Khan AS, Kanarek L, Van Driessche E: Ligand binding on to maize ( Zea mays ) malate synthase: a structural study. Biochem J 1994, 303(Pt 2):413–421.PubMed CentralView ArticlePubMed
- Emsley P, Lohkamp B, Scott WG, Cowtan K: Features and development of Coot. Acta Crystallographica Section D Biological Crystallography 2010, 66(4):486–501. 10.1107/S0907444910007493PubMed CentralView Article
- Cornforth JW, Redmond JW, Eggerer H, Buckel W, Gutschow C: Asymmetric methyl groups, and the mechanism of malate synthase. Nature 1969, 221(5187):1212–1213. 10.1038/2211212a0View ArticlePubMed
- Luthy J, Retey J, Arigoni D: Preparation and detection of chiral methyl groups. Nature 1969, 221(5187):1213–1215. 10.1038/2211213a0View ArticlePubMed
- Eggerer H, Klette A: On the catalysis principle of malate synthase. Eur J Biochem 1967, 1(4):447–475. 10.1111/j.1432-1033.1967.tb00094.xView ArticlePubMed
- Huber R, Bode W: Structural Basis of the Activation and Action of Trypsin. Accounts of Chemical Research 1978, 11: 114–122. 10.1021/ar50123a006View Article
- Huber R, Kukla D, Bode W, Schwager P, Bartels K, Deisenhofer J, Steigemann W: Structure of the complex formed by bovine trypsin and bovine pancreatic trypsin inhibitor. II. Crystallographic refinement at 1.9 Å resolution. J Mol Biol 1974, 89(1):73–101. 10.1016/0022-2836(74)90163-6View ArticlePubMed
- Burgi HB, Dunitz JD, Shefter E: Geometrical Reaction Coordinates. II. Nucleophilic Addition to a Carbonyl Group. Journal of the American Chemical Society 1973, 95(15):5065–5067. 10.1021/ja00796a058View Article
- Burgi HB, Dunitz JD, Lehn JM, Wipff G: Stereochemistry of reaction paths at carbonyl centres. Tetrahedron 1974, 30: 1563–1572. 10.1016/S0040-4020(01)90678-7View Article
- Fasman GD, (ed): Prediction of protein structure and the principles of protein conformation. New York: Plenum Press; 1989.
- Kanehisa M, Goto S: KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res 2000, 28(1):27–30. 10.1093/nar/28.1.27PubMed CentralView ArticlePubMed
- Khomyakova M, Bukmez O, Thomas LK, Erb TJ, Berg IA: A methylaspartate cycle in haloarchaea. Science 2011, 331(6015):334–337. 10.1126/science.1196544View ArticlePubMed
- Malfatti S, Tindall BJ, Schneider S, Fähnrich R, Lapidus A, Labuttii K, Copeland A, Glavina Del Rio T, Nolan M, Chen F, Lucas S, Tice H, Cheng JF, Bruce D, Goodwin L, Pitluck S, Anderson I, Pati A, Ivanova N, Mavromatis K, Chen A, Palaniappan K, D'haeseleer P, Göker M, Bristow J, Eisen JA, Markowitz V, Hugenholtz P, Kyrpides NC, Klenk HP, Chain P: Complete genome sequence of Halogeometricum borinquense type strain (PR3). Stand Genomic Sci 2009, 1(2):150–159.PubMed CentralPubMed
- Falb M, Pfeiffer F, Palm P, Rodewald K, Hickmann V, Tittor J, Oesterhelt D: Living with two extremes: conclusions from the genome sequence of Natronomonas pharaonis . Genome Res 2005, 15(10):1336–1343. 10.1101/gr.3952905PubMed CentralView ArticlePubMed
- Thompson JD, Gibson TJ, Higgins DG: Multiple sequence alignment using ClustalW and ClustalX. Curr Protoc Bioinformatics 2002, Chapter 2(Unit 2):3.PubMed
- American Type Culture Collection (ATCC). P.O. Box 1549, Manassas, VA 20108, USA[http://www.atcc.org/]
- Otwinowski Z, Minor W: Processing of X-ray Diffraction Data Collected in Oscillation Mode. In Methods in Enzymology. Volume 276. Edited by: Carter CWJ, Sweet RM. New York: Academic Press; 1997:307–327.
- Terwilliger TC, Berendzen J: Automated MAD and MIR structure solution. Acta Crystallogr D Biol Crystallogr 1999, 55(Pt 4):849–861.PubMed CentralView ArticlePubMed
- Terwilliger TC: Maximum-likelihood density modification. Acta Crystallogr D Biol Crystallogr 2000, 56(Pt 8):965–972.PubMed CentralView ArticlePubMed
- Emsley P, Cowtan K: Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr 2004, 60(Pt 12 Pt 1):2126–2132.View ArticlePubMed
- Murshudov GN, Vagin AA, Dodson EJ: Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr 1997, 53(Pt 3):240–255.View ArticlePubMed
- The CCP4 suite: programs for protein crystallography Acta Crystallogr D Biol Crystallogr 1994, 50(Pt 5):760–763.
- McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn MD, Storoni LC, Read RJ: Phaser crystallographic software. J Appl Crystallogr 2007, 40(Pt 4):658–674.PubMed CentralView ArticlePubMed
- Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, Bourne PE: The Protein Data Bank. Nucleic Acids Res 2000, 28(1):235–242. 10.1093/nar/28.1.235PubMed CentralView ArticlePubMed
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.