Cloning and sequencing of the genes encoding the large and small subunits of the periplasmic (NiFeSe) hydrogenase of Desulfovibrio baculatus

Intermediates in the catalytic cycle of [NiFe] hydrogenase: functional spectroscopy of the active site

The [NiFe] hydrogenase from the anaerobic sulphate reducing bacterium Desulfovibrio vulgaris Miyazaki F is an excellent model for constructing a mechanism for the function of the so-called 'oxygen-sensitive' hydrogenases. The present review focuses on spectroscopic investigations of the active site intermediates playing a role in the activation/deactivation and catalytic cycle of this enzyme as well as in the inhibition by carbon monoxide or molecular oxygen and the light-sensitivity of the hydrogenase. The methods employed include magnetic resonance and vibrational (FTIR) techniques combined with electrochemistry that deliver information about details of the geometrical and electronic structure of the intermediates and their redox behaviour. Based on these data a mechanistic scheme is developed.

Distribution of Hydrogenase Genes in Desulfovibrio spp. and Their Use in Identification of Species from the Oil Field Environment

The distribution of genes for [Fe], [NiFe], and [NiFeSe] hydrogenases was determined for 22 Desulfovibrio species. The genes for [NiFe] hydrogenase were present in all species, whereas those for the [Fe] and [NiFeSe] hydrogenases had a more limited distribution. Sulfate-reducing bacteria from 16S rRNA groups other than the genus Desulfovibrio (R. Devereux, M. Delaney, F. Widdel, and D. A. Stahl, J. Bacteriol. 171:6689-6695, 1989) did not react with the [NiFe] hydrogenase gene probe, which could be used to identify different Desulfovibrio species in oil field samples following growth on lactate-sulfate medium.

Genomic investigation of the system for selenocysteine incorporation in the bacterial domain

The elucidation of nearly 100 bacterial genomes has made it possible to categorize them into two groups, according to the presence or absence of a selenocysteine (Sec) tRNA. In the group with the tRNA, a Sec incorporation system like that of Escherichia coli would be expected. However, for the other group, the following question has been left unsolved. Is it reasonable to assume that bacteria without the tRNA lack the entire Sec system, and do such bacteria exist commonly? To explore it experimentally, we chose Bacillus subtilis, a representative eubacterium for which a Sec tRNA has not been found. First, we reviewed the genome to search for the tRNA gene. Second, we examined the possible expression of an unknown tRNA. Third, we examined Sec-related enzymes and proteins in B. subtilis cell extracts. Fourth, the B. subtilis and E. coli seryl-tRNA synthetases were expressed, and the specificity was analyzed. Consequently, all of the data showed negative results about the existence of the Sec system in B. subtilis. Additionally, we discuss the possibility of predicting the existence or absence of the system in each bacterial organism by using the Sec tRNA and seryl-tRNA synthetase as indicators.

Structural Studies of the Carbon Monoxide Complex of [NiFe]hydrogenase from Desulfovibrio vulgaris Miyazaki F: Suggestion for the Initial Activation Site for Dihydrogen

The carbon monoxide complex of [NiFe]hydrogenase from Desulfovibrio vulgaris Miyazaki F has been characterized by X-ray crystallography and absorption and resonance Raman spectroscopy. Nine crystal structures of the [NiFe]hydrogenase in the CO-bound and CO-liberated forms were determined at 1.2-1.4 A resolution. The exogenously added CO was assigned to be bound to the Ni atom at the Ni-Fe active site. The CO was not replaced with H(2) in the dark at 100 K, but was liberated by illumination with a strong white light. The Ni-C distances and Ni-C-O angles were about 1.77 A and 160 degrees, respectively, except for one case (1.72 A and 135 degrees ), in which an additional electron density peak between the CO and Sgamma(Cys546) was recognized. Distinct changes were observed in the electron density distribution of the Ni and Sgamma(Cys546) atoms between the CO-bound and CO-liberated structures for all the crystals tested. The novel structural features found near the Ni and Sgamma(Cys546) atoms suggest that these two atoms at the Ni-Fe active site play a role during the initial H(2)-binding process. Anaerobic addition of CO to dithionite-reduced [NiFe]hydrogenase led to a new absorption band at about 470 nm ( approximately 3000 M(-1)cm(-1)). Resonance Raman spectra (excitation at 476.5 nm) of the CO complex revealed CO-isotope-sensitive bands at 375/393 and 430 cm(-1) (368 and 413 cm(-1) for (13)C(18)O). The frequencies and relative intensities of the CO-related Raman bands indicated that the exogenous CO is bound to the Ni atom with a bent Ni-C-O structure in solution, in agreement with the refined structure determined by X-ray crystallography.

Evolutionarily different RNA motifs and RNA-protein complexes to achieve selenoprotein synthesis

A wealth of RNAs or RNA motifs are instrumental in controlling a variety of post-transcriptional or post-translational regulations. In this regard, selenocysteine incorporation in response to a redefined UGA stop codon certainly constitutes an intriguing and fascinating process. Translation elongation factors specialized for selenocysteine are needed to decode UGA selenocysteine codons. Discrimination between UGA selenocysteine and UGA stop codons also necessitates selenoprotein mRNA hairpins, called SECIS, that are internal to the coding frame in eubacteria or contained in the 3' untranslated regions in archaea/eukaryotes. This dichotomy leads to SECIS RNAs with distinct sequences and structures that tether the specialized translation elongation factor in a direct or indirect fashion, depending on the location of the SECIS RNA. The scope of this review is to bring a sharper focus on the SECIS RNA structures and SECIS RNA-protein complexes involved. Obviously, the examples described here highlight once again the versatility in form and function of RNA.

A universal system for the transport of redox proteins: early roots and latest developments

The transport of proteins binding redox cofactors across a biological membrane is complicated by the fact that insertion of the redox cofactor is often a cytoplasmic process. These cytoplasmically assembled redox proteins must thus be transported in partially or completely folded form. The need for a special transport system for redox proteins was first recognized for periplasmic hydrogenases in gram-negative bacteria. These enzymes, which catalyze the reaction H2 <--> 2H+ + 2e, are composed of a large and a small subunit. Only the small subunit has an unusually long signal sequence of 30-50 amino acid residues, characterized by a conserved motif (S/T)-R-R-x-F-L-K at the N-terminus. This sequence directs export of the large and small subunit complex to the periplasm. Sequencing of microbial genes and genomes has shown that signal sequences with this conserved motif, now referred to as twin-arginine leaders, occur ubiquitously and export different classes of redox proteins, containing iron sulfur clusters, molybdopterin cofactors, polynuclear copper sites or flavin adenine dinucleotide. Mutations in an Escherichia coli operon referred to as mtt (membrane targeting and translocation) or tat (twin arginine translocation) are pleiotropic, i.e. these prevent the expression of a variety of periplasmic oxido-reductases in functional form. The Mtt or Tat pathway is distinct from the well-known Sec pathway and occurs ubiquitously in prokaryotes. The fact that its component proteins share sequence homology with proteins of the delta pH pathway for protein transport associated with chloroplast thylakoid assembly, illustrates the universal nature of this novel protein translocation system.

Characterization of the cytochromes C from Desulfovibrio desulfuricans G201

A monoheme cytochrome c553 and a hexadecaheme high molecular weight cytochrome (Hmc) have been isolated and characterized from the sulfate-reducing bacteria Desulfovibrio desulfuricans G201, in addition to the tetraheme cytochrome c3 (Mr 13000) that has been previously described. Both cytochromes are homologous with respect to several biochemical properties to the corresponding cytochromes found in other Desulfovibrio species. However, they are acidic proteins while the corresponding molecules, isolated from other Desulfovibrio species, are relatively more basic. The D. desulfuricans cytochrome content appears identical to that of D. vulgaris Hildenborough. Isolation of these cytochromes from a Desulfovibrio desulfuricans strain is of great interest in order to get more insight on the physiological function of these molecules.

Barriers to heterologous expression of a selenoprotein gene in bacteria

The specificity parameters counteracting the heterologous expression in Escherichia coli of the Desulfomicrobium baculatum gene (hydV) coding for the large subunit of the periplasmic hydrogenase which is a selenoprotein have been studied. hydV'-'lacZ fusions were constructed, and it was shown that they do not direct the incorporation of selenocysteine in E. coli. Rather, the UGA codon is efficiently suppressed by some other aminoacyl-tRNA in an E. coli strain possessing a ribosomal ambiguity mutation. The suppression is decreased by the strA1 allele, indicating that the hydV selenocysteine UGA codon has the properties of a "normal" and suppressible nonsense codon. The SelB protein from D. baculatum was purified; in gel shift experiments, D. baculatum SelB displayed a lower affinity for the E. coli fdhF selenoprotein mRNA than E. coli SelB did and vice versa. Coexpression of the hydV'-'lacZ fusion and of the selB and tRNA(Sec) genes from D. baculatum, however, did not lead to selenocysteine insertion into the protein, although the formation of the quaternary complex between SelB, selenocysteyl-tRNA(Sec), and the hydV mRNA recognition sequence took place. The results demonstrate (i) that the selenocysteine-specific UGA codon is readily suppressed under conditions where the homologous SelB protein is absent and (ii) that apart from the specificity of the SelB-mRNA interaction, a structural compatibility of the quaternary complex with the ribosome is required.

HupUV proteins of Rhodobacter capsulatus can bind H2: evidence from the H-D exchange reaction

The H-D exchange reaction has been measured with the D2-H2O system, for Rhodobacter capsulatus JP91, which lacks the hupSL-encoded hydrogenase, and R. capsulatus BSE16, which lacks the HupUV proteins. The hupUV gene products, expressed from plasmid pAC206, are shown to catalyze an H-D exchange reaction distinguishable from the H-D exchange due to the membrane-bound, hupSL-encoded hydrogenase. In the presence of O2, the uptake hydrogenase of BSE16 cells catalyzed a rapid uptake and oxidation of H2, D2, and HD present in the system, and its activity (H-D exchange, H2 evolution in presence of reduced methyl viologen [MV+]) depended on the external pH, while the H-D exchange due to HupUV remained insensitive to external pH and O2. These data suggest that the HupSL dimer is periplasmically oriented, while the HupUV proteins are in the cytoplasmic compartment.

Sequences and characterization of hupU and hupV genes of Bradyrhizobium japonicum encoding a possible nickel-sensing complex involved in hydrogenase expression

A 2.7-kb DNA fragment of Bradyrhizobium japonicum previously shown to be involved in hydrogenase expression has been sequenced. The area is located just upstream of the hupSLCDF operon and was found to contain two open reading frames, designated hupU and hupV; these encode proteins of 35.4 and 51.8 kDa, respectively. These proteins are homologous to Rhodobacter capsulatus HupU, a possible repressor of hydrogenase expression in that organism. B. japonicum HupU is 54% identical to the N terminus of R. capsulatus HupU, and HupV is 50% identical to the C terminus of R. capsulatus HupU. HupU and HupV also show homology to the [Ni-Fe] hydrogenase small and large subunits, respectively. Notably, HupV contains the probable nickel-binding sites RxCGxC and DPCxxCxxH, which are located in the N- and C-terminal portions, respectively, of the large subunit of hydrogenases. Hydrogenase activity assays, immunological assays for hydrogenase subunits, and beta-galactosidase assays on mutant strain JHCS2 (lacking a portion of HupV) were all indicative that HupV is necessary for transcriptional activation of hydrogenase. A physiological role as a possible nickel- or other environmental (i.e., oxygen or hydrogen)-sensing complex is proposed for HupU and HupV.

Genes coding for the selenocysteine-inserting tRNA species from Desulfomicrobium baculatum and Clostridium thermoaceticum: structural and evolutionary implications

The genes (selC) coding for the selenocysteine-inserting tRNA species (tRNA(Sec)) from Clostridium thermoaceticum and Desulfomicrobium baculatum were cloned and sequenced. Although they differ in numerous positions from the sequence of the Escherichia coli selC gene, they were able to complement the selC lesion of an E. coli mutant and to promote selenoprotein formation in the heterologous host. The tRNA(Sec) species from both organisms possess all of the unique primary, secondary, and tertiary structural features exhibited by E. coli tRNA(Sec) (C. Baron, E. Westhof, A. Böck, and R. Giegé, J. Mol. Biol. 231:274-292, 1993). The structural and functional properties of the tRNA(Sec) species from prokaryotes analyzed thus far support the notion that tRNA(Sec) may be an evolutionarily conserved structure whose function in the primordial genetic code was to decode UGA with selenocysteine.

In vivo and in vitro nickel-dependent processing of the [NiFe] hydrogenase in Azotobacter vinelandii

H2 oxidation in Azotobacter vinelandii is catalyzed by a membrane-bound, alpha beta dimeric [NiFe] hydrogenase. Maturation of the enzyme involves cleavage of a putative N-terminal signal sequence in the beta subunit and removal of 15 amino acids from the C terminus of the alpha subunit. Cells limited for nickel exhibited low hydrogenase activities and contained an apparently large form of the alpha subunit. Addition of nickel to such cells increased hydrogenase activities fivefold over 2 h. The increase in the first hour did not require transcription and translation and correlated with processing of the large form of the alpha subunit (pre-alpha) to the small form (alpha) resembling the alpha subunit from the purified enzyme. In vivo, pre-alpha appeared soluble whereas the majority of alpha was membrane bound. Processing of pre-alpha to alpha was reproduced in vitro in membrane-depleted extracts of nickel-limited cells. Processing specifically required the addition of Ni2+, whereas Co2+, Cu2+, Ca2+, Fe2+, Mn2+, and Zn2+ were ineffective. However, Zn2+, Co2+, and Cu2+ inhibited nickel-dependent processing. Mg-ATP and Mg-GTP stimulated processing, whereas anaerobic conditions and/or the addition of dithiothreitol and sodium dithionite was unnecessary. Processing was not inhibited by the protease inhibitors phenylmethylsulfonyl fluoride, E64, and pepstatin.

Sequence analysis and interposon mutagenesis of the hupT gene, which encodes a sensor protein involved in repression of hydrogenase synthesis in Rhodobacter capsulatus

The hupT gene, which represses hydrogenase gene expression in the purple photosynthetic bacterium Rhodobacter capsulatus, has been identified and sequenced. The nucleotide sequence of hupT and of the contiguous downstream open reading frame, hupU, is reported. The HupT protein of 456 amino acids (48,414 Da) has sequence similarity with the FixL, DctB, NtrB, and ArcB proteins and is predicted to be a soluble sensor kinase. Insertional inactivation of the hupT gene led to deregulation of transcriptional control, so that the hydrogenase structural operon hupSLC became overexpressed in cells grown anaerobically or aerobically. The HupT- mutants were complemented in trans by a plasmid containing an intact copy of the hupT gene. The hupU open reading frame, capable of encoding a protein of 84,879 Da, shared identity with [NiFe]hydrogenase subunits; the strongest similarity was observed with the periplasmic hydrogenase of Desulfovibrio baculatus.

Intimate relationships of the large and the small subunits of all nickel hydrogenases with two nuclear-encoded subunits of mitochondrial NADH: ubiquinone oxidoreductase

The sequence pattern CxxCxnGxCxxxGxmGCPP, thus far found in the small subunits from 21 different nickel hydrogenases, appears also to be present in the PSST polypeptide from NADH:ubiquinone oxidoreductase (Complex I) of beef-heart mitochondria. There is only one difference: the first cysteine residue is a leucine in the PSST subunit. The large nickel-binding subunit of nickel hydrogenases shows a surprising homology with the 49 kDa subunit of mitochondrial Complex I.

In Azotobacter vinelandii hydrogenase, substitution of serine for the cysteine residues at positions 62, 65, 294, and 297 in the small (HoxK) subunit affects H2 oxidation [corrected]

The essential role of the small (HoxK) subunit of hydrogenase of Azotobacter vinelandii in H2 oxidation was established. This was achieved by modification of the two Cys-X2-Cys amino acid motifs at the N and C termini of the HoxK subunit (Cys-62, -65, -294, and -297). The Cys codons were individually mutated to Ser codons. Modifications in these two motifs resulted in loss of hydrogenase activity. At the N terminus, the mutations of the codons for the motif Cys-62-Thr-Cys-64-Cys-65 decreased the activity of hydrogenase to levels no higher than 30% of those of the parental strain. H2 oxidation with the alternate electron acceptors methylene blue and benzyl viologen was decreased. H2 evolution and exchange activities were also affected. Cys-64 possibly substitutes for either Cys-62 or Cys-65, allowing for partial activity. Mutation of the codons for Cys-294 and Cys-297 to Ser codons resulted in no hydrogenase activity. The results are consistent with alterations of the ligands of FeS clusters in the HoxK subunit of hydrogenase [corrected].

Overproduction of prismane protein in Desulfovibrio vulgaris (Hildenborough): evidence for a second S = 1/2-spin system in the one-electron reduced state

The gene encoding the prismane protein from Desulfovibrio vulgaris (Hildenborough) was inserted into broad-host-range vector pSUP104. The recombinant plasmid, pJSP104, was transferred to D. vulgaris by conjugal plasmid transfer. In the transconjugant D. vulgaris cells the prismane protein was 25-fold overproduced. The overproduced prismane protein was characterized by molecular mass, isoelectric point, iron content and spectroscopical properties. Both the iron content and the ultraviolet/visible spectrum are identical to the wild-type protein indicating that iron incorporation in the overproduced protein is complete. EPR spectra of the dithionite-reduced form of the overproduced protein indicated that the Fe-S cluster might occur in a similar structure as found in inorganic model compounds containing a [6Fe-6S] prismane core. The as-isolated overproduced protein showed the presence of a second S = 1/2 spin system that was also detected in the corresponding prismane protein from D. desulfuricans (ATCC 27774), but not in the protein from wild-type D. vulgaris. This additional signal was irreversibly transformed to the 'wild-type' high-spin and low-spin systems upon two reduction/re-oxidation cycles. It is shown that the EPR spectroscopy of the overproduced prismane protein is very similar to that of the D. desulfuricans enzyme and, with the exception of the second S = 1/2 spin system, to that of the prismane protein from wild-type D. vulgaris. Contrary to claims for the D. desulfuricans protein, it is shown here that all data can be fully explained assuming a single [6Fe-6S] cluster, that might be titrated into four different redox states and occurs in up to three different spin systems in the one-electron reduced state.

Maturation of membrane-bound hydrogenase of Alcaligenes eutrophus H16

The formation of the catalytically active membrane-bound hydrogenase (MBH) of Alcaligenes eutrophus H16 requires the genes for the small and large subunits of the enzyme (hoxK and hoxG, respectively) and an accompanying set of accessory genes (C. Kortl ke, K. Horstmann, E. Schwartz, M. Rohde, R. Binsack, and B. Friedrich, J. Bacteriol. 174:6277-6289, 1992). Other genes located in the adjacent pleiotropic region are also required. In the absence of these genes, MBH is synthesized but is catalytically inactive. Immunological analyses revealed that cells containing active MBH produced the small and large subunits of the enzyme in two distinct conformations each; only one of each, presumably the immature form, occurred in cells devoid of MBH activity. The results suggest that the conversion of the two subunits into the catalytically active membrane-associated heterodimer depends on specific maturation processes mediated by hox genes.

Carboxyl-terminal processing may be essential for production of active NiFe hydrogenase in Azotobacter vinelandii

The NiFe hydrogenase from Azotobacter vinelandii is a membrane-bound alpha beta heterodimer that can oxidize H2 to protons and electrons and thereby provide energy. Genes encoding the alpha and beta subunits, hoxG and hoxK respectively, followed by thirteen contiguous accessory genes potentially involved in H2 oxidation, have been previously sequenced. Mutations in some of these accessory genes give rise to inactive enzyme containing an alpha subunit with decreased electrophoretic mobility. Mass spectral analysis of the subunits demonstrated that the alpha subunit had a molecular weight 1,663 Da less than that predicted from hoxG. Since the N-terminal sequence of the purified alpha subunit matches the sequence predicted from hoxG we suggest this difference is due to removal of the C-terminus of the alpha subunit which may be an important step linked to metal insertion, localization, and formation of active hydrogenase.

Mutational analysis of the operon (hyc) determining hydrogenase 3 formation in Escherichia coli

In-frame deletions were introduced into each of the eight genes of the hyc operon coding for products required for the formation of the formate hydrogenlyase (FHL) system. The deletions were transferred to the chromosome and the resulting mutants were analysed for development of formate dehydrogenase H and hydrogenase 1, 2 and 3 activity. It was found that hycA, the promoter-proximal gene, is a regulatory gene and that it codes for a product counteracting transcriptional activation by FhlA. Deletions within the hycB to hycH genes specifically affected formate dehydrogenase H activity or hydrogenase 3 activity, or both. None of the mutations affected hydrogenase 1 or 2 activity. A model is proposed for the functional interaction of the different hyc operon gene products in the formate hydrogenlyase complex, which is based on the results of the mutational analysis, on the determination of the subcellular localization of the FdhF, HycE, HycF and HycG polypeptides and on the similarity of hyc gene product sequences with those from other hydrogenase systems. HycH, the product of the most promoter-distal gene, does not seem to form part of the functional FHL complex but rather is required for the conversion of a precursor form of the large subunit of hydrogenase 3 into the mature form.

Purification and properties of a F420-nonreactive, membrane-bound hydrogenase from Methanosarcina strain Gö1

The distribution of the F420-reactive and F420-nonreactive hydrogenases from the methylotrophic Methanosarcina strain Gö1 indicated a membrane association of the F420-nonreactive enzyme. The membrane-bound F420-nonreactive hydrogenase was purified 42-fold to electrophoretic homogeneity with a yield of 26.7%. The enzyme had a specific activity of 359 mumol H2 oxidized.min-1.mg protein-1. The purification procedure involved dispersion of the membrane fraction with the detergent Chaps followed by anion exchange, hydrophobic and hydroxylapatite chromatography. The aerobically prepared enzyme had to be reactivated anaerobically. Maximal activity was observed at 80 degrees C. The molecular mass as determined by native gel electrophoresis and gel filtration was 77,000 and 79,000, respectively. SDS gel electrophoresis revealed two polypeptides with molecular masses of 60,000 and 40,000 indicating a 1:1 stoichiometry. The purified enzyme contained 13.3 mol S2-, 15.1 mol Fe and 0.8 mol Ni/mol enzyme. Flavins were not detected. The amino acid sequence of the N-termini of the subunits showed a higher degree of homology to eubacterial uptake-hydrogenases than to F420-dependent hydrogenases from other methanogenic bacteria. The physiological function of the F420-nonreactive hydrogenase from Methanosarcina strain Göl is discussed.

Recent evidence for evolution of the genetic code

The genetic code, formerly thought to be frozen, is now known to be in a state of evolution. This was first shown in 1979 by Barrell et al. (G. Barrell, A. T. Bankier, and J. Drouin, Nature [London] 282:189-194, 1979), who found that the universal codons AUA (isoleucine) and UGA (stop) coded for methionine and tryptophan, respectively, in human mitochondria. Subsequent studies have shown that UGA codes for tryptophan in Mycoplasma spp. and in all nonplant mitochondria that have been examined. Universal stop codons UAA and UAG code for glutamine in ciliated protozoa (except Euplotes octacarinatus) and in a green alga, Acetabularia. E. octacarinatus uses UAA for stop and UGA for cysteine. Candida species, which are yeasts, use CUG (leucine) for serine. Other departures from the universal code, all in nonplant mitochondria, are CUN (leucine) for threonine (in yeasts), AAA (lysine) for asparagine (in platyhelminths and echinoderms), UAA (stop) for tyrosine (in planaria), and AGR (arginine) for serine (in several animal orders) and for stop (in vertebrates). We propose that the changes are typically preceded by loss of a codon from all coding sequences in an organism or organelle, often as a result of directional mutation pressure, accompanied by loss of the tRNA that translates the codon. The codon reappears later by conversion of another codon and emergence of a tRNA that translates the reappeared codon with a different assignment. Changes in release factors also contribute to these revised assignments. We also discuss the use of UGA (stop) as a selenocysteine codon and the early history of the code.

Structure-function relationships among the nickel-containing hydrogenases

The enzymology of the heterodimeric (NiFe) and (NiFeSe) hydrogenases, the monomeric nickel-containing hydrogenases plus the multimeric F420-(NiFe) and NAD(+)-(NiFe) hydrogenases are summarized and discussed in terms of subunit localization of the redox-active nickel and non-heme iron clusters. It is proposed that nickel is ligated solely by amino acid residues of the large subunit and that the non-heme iron clusters are ligated by other cysteine-rich polypeptides encoded in the hydrogenase operons which are not necessarily homologous in either structure or function. Comparison of the hydrogenase operons or putative operons and their hydrogenase genes indicate that the arrangement, number and types of genes in these operons are not conserved among the various types of hydrogenases except for the gene encoding the large subunit. Thus, the presence of the gene for the large subunit is the sole feature common to all known nickel-containing hydrogenases and unites these hydrogenases into a large but diverse gene family. Although the different genes for the large subunits may possess only nominal general derived amino acid homology, all large subunit genes sequenced to date have the sequence R-X-C-X-X-C fully conserved in the amino terminal region of the polypeptide chain and the sequence of D-P-C-X-X-C fully conserved in the carboxyl terminal region. It is proposed that these conserved motifs of amino acids provide the ligands required for the binding of the redox-active nickel. The existing EXAFS (Extended X-ray Absorption Fine Structure) information is summarized and discussed in terms of the numbers and types of ligands to the nickel and the various redox species of nickel defined by EPR spectroscopy. New information concerning the ligands to nickel is presented based on site-directed mutagenesis of the gene encoding the large subunit of the (NiFe) hydrogenase-1 of Escherichia coli. Based on considerations of the biochemical, molecular and biophysical information, ligand environments of the nickel in different redox states of the (NiFe) hydrogenase are proposed.

Mutational analysis and characterization of the Escherichia coli hya operon, which encodes [NiFe] hydrogenase 1

Deletion mutants of Escherichia coli specific for hydrogenase isoenzyme 1 (HYD1) have been constructed and characterized. The hya operon, which contains genes for the two HYD1 structural subunits and four additional genes, was mapped at 22 min on the E. coli chromosome. The total hydrogenase activities of the HYD1-negative mutant and wild-type strains were similar. However, the formate dehydrogenase activity associated with the formate hydrogen lyase pathway was lower in the mutant. The hya mutant (strain AP1), complemented with only the hydrogenase structural genes (hyaAB), produced antigenically identifiable but inactive HYD1 protein. The first five genes of hya (hyaA to hyaE) were required for the synthesis of active HYD1, but wild-type levels of HYD1 activity were restored only when mutant cells were transformed with all six genes of the operon. When AP1 was complemented with hya carried on a high-copy-number plasmid, the HYD1 structural subunits were overexpressed, but the excess protein was unprocessed and localized in the soluble fraction of the cell. The products of hyaDEF are postulated to be involved in the processing of nascent structural subunits (HYAA and HYAB). This processing takes place only after the subunits are inserted into the cell membrane. It is concluded that the biosynthesis of active HYD1 is a complex biochemical process involving the cellular localization and processing of nascent structural subunits, which are in turn dependent on the insertion of nickel into the nascent HYD1 large subunit.

Interspecies compatibility of selenoprotein biosynthesis in Enterobacteriaceae

Several species of Enterobacteriaceae were investigated for their ability to synthesize selenium-containing macromolecules. Seleniated tRNA species as well as seleniated polypeptides were formed by all organisms tested. Two selenopolypeptides could be identified in most of the organisms which correspond to the 80 kDa and 110 kDa subunits of the anaerobically induced formate dehydrogenase isoenzymes of E. coli. In those organisms possessing both isoenzymes, their synthesis was induced in a mutually exclusive manner dependent upon whether nitrate was present during anaerobic growth. The similarity of the 80 kDa selenopolypeptide among the different species was assessed by immunological and genetic analyses. Antibodies raised against the 80 kDa selenopolypeptide from E. coli cross-reacted with an 80 kDa polypeptide in those organisms which exhibited fermentative formate dehydrogenase activity. These organisms also contained genes which hybridised with the fdhF gene from E. coli. In an attempt to identify the signals responsible for incorporation of selenium into the selenopolypeptides in these organisms we cloned a portion of the fdhF gene homologue from Enterobacter aerogenes. The nucleotide sequence of the cloned 723 bp fragment was determined and it was shown to contain an in-frame TGA (stop) codon at the position corresponding to that present in the E. coli gene. This fragment was able to direct incorporation of selenocysteine when expressed in the heterologous host, E. coli. Moreover, the E. coli fdhF gene was expressed in Salmonella typhimurium, Serratia marcescens and Proteus mirabilis, indicating a high degree of conservation of the seleniating system throughout the enterobacteria.

Type I iodothyronine deiodinase is a selenocysteine-containing enzyme

Although thyroxine (3,5,3',5'-tetraiodothyronine, T4) is the principal secretory product of the vertebrate thyroid, its essential metabolic and developmental effects are all mediated by 3,5,3'-triiodothyronine (T3), which is produced from the prohormone by 5'-deiodination. The type-I iodothyronine deiodinase, a thiol-requiring propylthiouracil-sensitive oxidoreductase, is found mainly in liver and kidney and provides most of the circulating T3(1) but so far this enzyme has not been purified. Using expression cloning in the Xenopus oocyte, we have isolated a 2.1-kilobase complementary DNA for this deiodinase from a rat liver cDNA library. The kinetic properties of the protein expressed in transient assay systems, the tissue distribution of the messenger RNA, and its changes with thyroid status, all confirm its identity. We find that the mRNA for this enzyme contains a UGA codon for selenocysteine which is necessary for maximal enzyme activity. This explains why conversion of T4 to T3 is impaired in experimental selenium deficiency and identifies an essential role for this trace element in thyroid hormone action.

Conservation of primary structure in prokaryotic hydrogenases

All prokaryotic (NiFe)-hydrogenases so far studied at the primary sequence level appear to have evolved from a common ancestral sequence. Highly conserved cysteinyl and histidinyl residues indicate regions likely to be essential for enzyme activity, ligand and co-factor binding. There is a very highly conserved sequence over 100 basepairs (bp) in length within the intergenic region upstream of the methyl-viologen hydrogenase encoding genes in several different strains of Methanobacterium thermoautotrophicum, indicating that a sequence of this length is needed to direct and regulate the expression of these genes.

Cloning, sequencing and characterization of the [NiFe]hydrogenase-encoding structural genes (hoxK and hoxG) from Azotobacter vinelandii

The Azotobacter vinelandii [NiFe]hydrogenase-encoding structural genes were isolated from an A. vinelandii genomic cosmid library. Nucleotide (nt) sequence analysis showed that the two genes, hoxK and hoxG, which encode the small and large subunits of the enzyme, respectively, form part of an operon that contains at least one other gene. The hoxK gene encodes a polypeptide of 358 amino acids (aa) (39,209 Da). The deduced aa sequence encodes a possible 45-aa N-terminus extension, not present in the purified A. vinelandii hydrogenase small subunit, which could be a cellular targeting sequence. The hoxG gene is downstream form, and overlaps hoxK by 4 nt and encodes a 602-aa polypeptide of 66,803 Da. The hoxK and hoxG gene products display homology to aa sequences of hydrogenase small and large subunits, respectively, from other organisms. The hoxG gene lies 16 nt upstream from a third open reading frame which could encode a 27,729-Da (240-aa) hydrophobic polypeptide containing 53% nonpolar and 11% aromatic aa. The significance of this possible third gene is not known at present.

Localization of membrane-associated (NiFe) and (NiFeSe) hydrogenases of Desulfovibrio vulgaris using immunoelectron microscopic procedures

The intracellular location of membrane-associated (NiFe) and (NiFeSe) hydrogenases of Desulfovibrio vulgaris was determined using pre-embedding and post-embedding immunoelectron microscopic procedures. Polyclonal antisera directed against the purified (NiFe) and (NiFeSe) hydrogenases were raised in rabbits. One-day-old cultures of D. vulgaris, grown on a lactate/sulfate medium, were used for all experiments in these studies. For post-embedding labeling studies cells were fixed with 0.2% glutaraldehyde and 0.3% formaldehyde, dehydrated with methanol, and embedded in the low-temperature resin Lowicryl K4M. Our post-embedding studies using antibody-gold or protein-A-gold as electron-dense markers revealed the location of the two hydrogenases exclusively at the cell periphery; the precise membrane location was then demonstrated by pre-embedding labeling. Spheroplasts were incubated with the polyclonal antisera against (NiFe) and (NiFeSe) hydrogenase followed by ferritin-linked secondary antibodies prior to embedding and sectioning. The observed labeling pattern unequivocally revealed that the antigenic reactive sites of the (NiFe) hydrogenase are located in the near vicinity of the cytoplasmic membrane facing into the periplasmic space, whereas the (NiFeSe) hydrogenase is associated with the cytoplasmic side of the cytoplasmic membrane.

Cloning and nucleotide sequences of the genes for the subunits of NAD-reducing hydrogenase of Alcaligenes eutrophus H16

The genes hoxF, -U, -Y, and -H which encode the four subunit polypeptides alpha, gamma, delta, and beta of the NAD-reducing hydrogenase (HoxS) of Alcaligenes eutrophus H16, were cloned, expressed in Pseudomonas facilis, and sequenced. On the basis of the nucleotide sequence, the predicted amino acid sequences, and the N-terminal amino acid sequences, it was concluded that the structural genes are tightly linked and presumably organized as an operon, denoted hoxS. Two pairs of -24 and -12 consensus sequences resembling RpoN-activatable promoters lie upstream of hoxF, the first of the four genes. Primer extension experiments indicate that the second promoter is responsible for hoxS transcription. hoxF and hoxU code for the flavin-containing dimer (alpha and gamma subunits) of HoxS which exhibits NADH:oxidoreductase activity. A putative flavin-binding region is discussed. The 26.0-kilodalton (kDa) gamma subunit contains two cysteine clusters which may participate in the coordination of two [4F3-4S]centers. The genes hoxY and hoxH code for the small 22.9-kDa delta subunit and the nickel-containing 54.8-kDa beta subunit, respectively, of the hydrogenase dimer of HoxS. The latter dimer exhibits several conserved regions found in all nickel-containing hydrogenases. The roles of these regions in coordinating iron and nickel are discussed. Although the deduced amino acid sequences of the delta and beta subunits share some conserved regions with the corresponding polypeptides of other [NiFe] hydrogenases, the overall amino acid homology is marginal. Nevertheless, significant sequence homology (35%) to the corresponding polypeptides of the soluble methylviologen-reducing hydrogenase of Methanobacterium thermoautotrophicum was found. Unlike the small subunits of the membrane-bound and soluble periplasmic hydrogenases, the HoxS protein does not appear to be synthesized with an N-terminal leader peptide.

Features of the formate dehydrogenase mRNA necessary for decoding of the UGA codon as selenocysteine

The fdhF gene encoding the 80-kDa selenopolypeptide subunit of formate dehydrogenase H from Escherichia coli contains an in-frame TGA codon at amino acid position 140, which encodes selenocysteine. We have analyzed how this UGA "sense codon" is discriminated from a UGA codon signaling polypeptide chain termination. Deletions were introduced from the 3' side into the fdhF gene and the truncated 5' segments were fused in-frame to the lacZ reporter gene. Efficient read-through of the UGA codon, as measured by beta-galactosidase activity and incorporation of selenium, was dependent on the presence of at least 40 bases of fdhF mRNA downstream of the UGA codon. There was excellent correlation between the results of the deletion studies and the existence of a putative stem-loop structure lying immediately downstream of the UGA in that deletions extending into the helix drastically reduced UGA translation. Similar secondary structures can be formed in the mRNAs coding for other selenoproteins. Selenocysteine insertion cartridges were synthesized that contained this hairpin structure and variable portions of the fdhF gene upstream of the UGA codon and inserted into the lacZ gene. Expression studies showed that upstream sequences were not required for selenocysteine insertion but that they may be involved in modulating the efficiency of read-through. Translation of the UGA codon was found to occur with high fidelity since it was refractory to ribosomal mutations affecting proofreading and to suppression by the sup-9 gene product.

The identification, characterization, sequencing and mutagenesis of the genes (hupSL) encoding the small and large subunits of the H2-uptake hydrogenase of Azotobacter chroococcum

The structural genes (hupSL) of the membrane-bound NiFe-containing H2-uptake hydrogenase (Hup) of Azotobacter chroococcum were identified by oligonucleotide screening and sequenced. The small subunit gene (hupS) encodes a signal sequence of 34 amino acids followed by a 310-amino-acid, 34156D protein containing 12 cysteine residues. The large subunit gene (hupL) overlaps hupS by one base and codes for a predicted 601-amino-acid, 66433D protein. There are two regions of strong homology with other Ni hydrogenases: a Cys-Thr-Cys-Cys-Ser motif near the N-terminus of HupS and an Asp-Pro-Cys-Leu-Ala-Cys motif near the carboxy-terminus of HupL. Strong overall homology exists between Azotobacter, Bradyrhizobium japonicum and Rhodobacter capsulatus Hup proteins but less exists between the Azotobacter proteins and hydrogenases from Desulfovibrio strains. Mutagenesis of either hupS or hupL genes of A. chroococcum yielded Hup- phenotypes but some of these mutants retained a partial H2-evolving activity. Hybridization experiments at different stages of gene segregation confirmed the multicopy nature of the Azotobacter genome.

Cloning and sequencing of a putative Escherichia coli [NiFe] hydrogenase-1 operon containing six open reading frames

DNA encompassing the structural genes of an Escherichia coli [NiFe] hydrogenase has been cloned and sequenced. The genes were identified as those encoding the large and small subunits of hydrogenase isozyme 1 based on NH2-terminal sequences of purified subunits (kindly provided by K. Francis and K. T. Shanmugam). The structural genes formed part of a putative operon that contained four additional open reading frames. We have designated the operon hya and the six open reading frames hyaA through F. hyaA and hyaB encode the small and large structural subunits, respectively. The nucleotide-derived amino acid sequence of hyaC has a calculated molecular mass of 27.6 kilodaltons, contains 20% aromatic residues, and has four potential membrane-spanning regions. Open reading frames hyaD through F could encode polypeptides of 21.5, 14.9, and 31.5 kilodaltons, respectively. These putative peptides have no homology to other reported protein sequences, and their functions are unknown.