Methanoarchaeal sulfolactate dehydrogenase: prototype of a new family of NADH-dependent enzymes

Abstract

The crystal structure of the sulfolactate dehydrogenase from the hyperthermophilic and methanogenic archaeon Methanocaldococcus jannaschii was solved at 2.5 A resolution (PDB id. 1RFM). The asymmetric unit contains a tetramer of tight dimers. This structure, complexed with NADH, does not contain a cofactor-binding domain with 'Rossmann-fold' topology. Instead, the tertiary and quaternary structures indicate a novel fold. The NADH is bound in an extended conformation in each active site, in a manner that explains the pro-S specificity. Cofactor binding involves residues belonging to both subunits within the tight dimers, which are therefore the smallest enzymatically active units. The protein was found to be a homodimer in solution by size-exclusion chromatography, analytical ultracentrifugation and small-angle neutron scattering. Various compounds were tested as putative substrates. The results indicate the existence of a substrate discrimination mechanism, which involves electrostatic interactions. Based on sequence homology and phylogenetic analyses, several other enzymes were classified as belonging to this novel family of homologous (S)-2-hydroxyacid dehydrogenases.

Figure 1

Tight dimer of MjSLDH and topology of the monomer. (A) Ribbon representation of the tight dimer. The α-helices are shown in thick helical coil, the β-strands are represented by arrows and the coil regions by thin ribbon. The NADH cofactor is shown as black sticks. The monomers A and B are colored in red and mauve, respectively. The surface of the dimer is shown by a dotted Van der Waals surface calculated using all nonhydrogen atoms. (B) Ribbon representation of the secondary structure elements at the interface between two monomers within the tight dimer. The dotted Van der Waals surface is shown for monomer A only (depicted in red ribbon). The hydrophobic external surfaces of the central mixed β-sheets of each monomer face each other. This hydrophobic core is surrounded by the two α-helices αH and αI from monomers A and B. (C) Stereo view of the MjSLDH monomer showing the secondary structure elements. The 11 α-helices are labelled αA–αK, the 17 β-strands are named β1–β17 and four α-helical turns are labelled αB′, αH′, αH″ and αK′. The central mixed β-sheet is made up of strands β2, 3, 4, 15, 5, 9 and 10. There, only strands β3 and 4 are parallel. The NADH cofactor is shown as dark sticks. (D) Projection topology cartoon of the MjSLDH subunit. α-Helices are represented by circles and β-strands are shown as triangles. The lines connecting the secondary structure elements indicate the direction taken by these secondary structure elements. The two chain termini are indicated by N and C.

Figure 2

Active site of MjSLDH. (A) The NADH cofactor is shown in sticks inside the final omit map (magenta chicken wire) contoured at 1.0σ. The residues involved in cofactor binding, and the presumed residues of the catalytic site are labelled, with an indication of their subunit of origin. Broken lines represent the hydrogen bonds between cofactor atoms and those of the protein. A stereo version of this figure is available as Supplementary Figure 2S. (B) Close-up view of the active site in monomer A. Monomers A and B are colored in red and mauve, respectively, except for the mobile domain of monomer A, which is shown in cyan. The NADH and several potential active site residues are shown in stick representation. (C) Schematic representation of the active site. Hydrogen bonds are represented by thick dashed lines, while thin dashed lines show the interaction of T156A with the C4 and C5 atoms of the nicotinamide ring of NADH.

Figure 3

Comparison of the glycine-rich motif in the 3-D structure of MjSLDH to the ‘classical' motif present in dehydrogenases with a Rossman-fold motif. Hydrogen bonds between cofactor and protein are shown by dashed lines. (A) Stereo view of NADH interacting at the level of the diphosphate moiety with residue K225 in the glycine-rich motif of MjSLDH. This glycine-rich sequence signature is located at the amino-terminal end of helix αH. (B) Stereo view of NAD interacting with the glycine-rich motif found in a representative member of Rossman-fold dehydrogenases, the (R207S, R292S) mutant of HmMalDH (PDB id. 1O6Z; Irimia et al, 2003).

Figure 4

Unrooted neighbor-joining tree of sequences homologous to that of MjSLDH. The scale bar represents 0.5 amino-acid substitution per site. Bootstrap values are indicated. Archaeal sequences are shown underlined in bold, eucaryotic sequences are in italics and those from bacteria are in nonbold roman. Numbering is according to the sequence numbers listed in Supplementary Figure 3S.