Primary structure of hydrogenase I from Clostridium pasteurianum

Peptides obtained by cleavage of Clostridium pasteurianum hydrogenase I have been sequenced. The data allowed design of oligonucleotide probes which were used to clone a 2310-bp Sau3A fragment containing the hydrogenase encoding gene. The latter has been sequenced and was found to translate into a protein composed of 574 amino acids (Mr = 63,836), including 22 cysteines. C. pasteurianum hydrogenase is homologous to, but longer than, the large subunit of Desulfovibrio vulgaris (Hildenborough) [Fe] hydrogenase. It includes an additional N-terminal domain of ca. 110 amino acids which contains eight cysteine residues and which therefore could accommodate two of its postulated four [4Fe-4S] clusters. C. pasteurianum hydrogenase is most similar in length, cysteine positions, and sequence altogether to the translation product of a putative hydrogenase encoding gene from D. vulgaris (Hildenborough). Comparisons of the available [Fe] hydrogenase sequences show that these enzymes constitute a structurally rather homogeneous family. While they differ in the length of their N-termini and in the number of their [4Fe-4S] clusters, they are highly similar in their C-terminal halves, which are postulated to harbor the hydrogen-activating H cluster. Five conserved cysteine residues occurring in this domain are likely ligands of the H cluster. Possible ligation by other residues, and in particular by methionine, is discussed. The comparisons carried out here show that the H clusters most probably possess a common structural framework in all [Fe] hydrogenases. On the basis of the available data on these proteins and on the current developments in iron-sulfur chemistry, the H clusters possibly contain six to eight iron atoms.(ABSTRACT TRUNCATED AT 250 WORDS)