Analysis and comparison of nucleotide sequences encoding the genes for [NiFe] and [NiFeSe] hydrogenases from Desulfovibrio gigas and Desulfovibrio baculatus

Abstract

, DNA 6:539-551, 1987) were analyzed by the codon usage method of Staden and McLachlan. The reported reading frames were found to contain regions of low codon probability which are matched by more probable sequences in other frames. Renewed nucleotide sequencing showed the probable frames to be correct. The corrected sequences of the two small and large subunits share a significant degree of sequence homology. The small subunit, which contains 10 conserved cysteine residues, is likely to coordinate at least 2 iron-sulfur clusters, while the finding of a selenocysteine codon (TGA) near the 3' end of the [NiFeSe] large-subunit gene matched by a regular cysteine codon (TGC) in the [NiFe] large-subunit gene indicates the presence of some of the ligands to the active-site nickel in the large subunit.Our knowledge of the structure and function of hydrogenases is increasing rapidly because of work on the molecular biology of the genes encoding these proteins. These investigations have provided the primary structure of the [Fe] hydrogenase from Desulfovibrio vulgaris subsp. vulgaris Hildenborough (12,23,24) [Fe] hydrogenase from D. vulgaris (63% G+C) was used as the standard in initial calculations, which indicated that the gene for the large subunit of D. baculatus (57% G+C) and both genes for the small and large subunits of D. gigas (63% G+C) contained regions of low coding probability. These were matched by regions of high coding probability in other frames, indicating the possibility of frameshifts due to errors in the original nucleotide sequence data. Comparable results were obtained when the codon usage table for the smallsubunit gene of the [NiFeSe] hydrogenase from D. baculatus, which displayed a high coding probability throughout its reading frame in these initial calculations, was used as the standard. Use of this table is preferred for calculations on the D. baculatus genes in view of the different G + C contents indicated above, and the results of the calculations on the previously published sequences (7,9) are shown in Fig. 1A through F.The large-subunit gene of D. baculatus hydrogenase, which contained three regions of codon improbability (Fig. 1C, I through III), was next resequenced, and this confirmed the correctness of the codon probability analysis. An erratum with the complete, corrected sequence has been pub-

The quinone‐reactive Ni/Fe‐hydrogenase of Wolinella succinogenes

“…There is 35-45% sequence identity among the HydA and 26-34% among the HydB subunits. The degrees of similarity are smaller with the periplasmic Ni-hydrogenases of Desulfovibrio gigas and Desulfbvihrio haculutus [50] and there is no significant similarity with the cytoplasmic Ni-hydrogenases of A . eutrophus [51] and Methanobacterium thermoautotrophicum [52] or the Fehydrogenase of Desulfovibrio vulguris [53].…”

Nickel hydrogenases: in search of the active site

“…Also the EPR spectra of the reduced D. gigas enzyme [33,199] do not show the typical g = 1.94 type of signals, but are unusually broad. The amino-acid sequence [217] shows that this enzyme belongs to the standard hydrogenases (Fig. 6).…”

“…The codon for cysteine is usually a TGC triplet, but in the Se-containing nickel hydrogenases from Desulfovibrio baculatus [217] and Methanococcus voltae [90] the codon for the first cysteine in this motif is replaced by a TGA triplet, usually a stop-codon, but here coding for selenocysteine as in formate dehydrogenase from E. coli [26,230] and glutathione peroxidase from mouse ceils [38]. There is now conclusive evidence from EPR measurements on 77Se-enriched (I = 1/2) hydrogenases from D. baculatus [96] and M. voltae [184], as well as from EXAFS measurements on the D. baculatus enzyme [57] that Se is a ligand to nickel in these hydrogenases.…”

Three-dimensional structure of the nickel-containing hydrogenase from Thiocapsa roseopersicina

“…The amino acid sequences of the two subunits of the [NiFe] hydrogenase from Desulfovibrio gigas and the corresponding subunits of the [NiFeSe] hydrogenase from Desulfovibrio baculatus show a high degree of homology (16).…”

“…The molecular masses of the two subunits in D. gigas were calculated from the amino acid sequences to be 62 and 26 kDa (16). These may also be used as the best estimates for the molecular masses of the hydrogenase subunits in T. roseopersicina, since the hydrogenase subunits from these two organisms have nearly identical retention values by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (8).…”