Search

The uptake hydrogenase of chemolithotrophically grown Rhizobium japonicum was purified to apparent homogeneity with a final specific activity of 69 mumol of H2 oxidized per min per mg of protein. The procedure included Triton extraction of broken membranes and DEAE-cellulose and Sephacryl S-200 chromatographies. The purified protein contained two polypeptides separable only by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. They comigrated on native polyacrylamide gels and sucrose density gradients. The molecular weights were ca. 60,000 and 30,000. Densitometric scans of the sodium dodecyl sulfate gels indicated a molar ratio of 1.03 +/- 0.03. Antiserum was developed against the 60-kilodalton polypeptide for use in hydrogenase detection by an enzyme-linked immunosorbent assay. The antiserum did not cross-react with the 30-kilodalton polypeptide. Native gel electrophoresis of Triton-extracted cells grown in the presence of 63Ni showed comigration of the hydrogenase and radioactive Ni.

Soybean plants and Rhizobium japonicum 122 DES, a hydrogen uptake-positive strain, were cultured in media purified to remove Ni. Supplemental Ni had no significant effect on the dry matter or total N content of plants. However, the addition of Ni to both nitrate-grown and symbiotically grown plants resulted in a 7- to 10-fold increase in urease activity (urea amidohydrolase, EC 3.5.1.5) in leaves and significantly increased the hydrogenase activity (EC 1.18.3.1) in isolated nodule bacteroids. When cultured under chemolithotrophic conditions, free-living R. japonicum required Ni for growth and for the expression of hydrogenase activity. Hydrogenase activity was minimal or not detectable in cells incubated either without Ni or with Ni and chloramphenicol. Ni is required for derepression of hydrogenase activity and apparently protein synthesis is necessary for the participation of Ni in hydrogenase expression. The addition of Cr, V, Sn, and Pb in place of Ni failed to stimulate the activity of hydrogenase in R. japonicum and urease in soybean leaves. The evidence indicates that Ni is an important micronutrient element in the biology of the soybean plant and R. japonicum.

The survival of pure cultures of Escherichia coli, Streptococcus faecalis, Clostridium perfringens, and Vibrio parahaemolyticus under simulated deep-sea conditions of low temperature (4 C), seawater, and hydrostatic pressures ranging from 1 to 1,000 atm was determined over a period exceeding 300 h. The viability of E. coli and total aerobic bacteria in seawater-diluted raw sewage subjected to these deep-sea conditions was also measured. There was a greater survival of both E. coli and S. faecalis at 250 and 500 atm than at 1 atm at 4 C. S. faecalis was quite insensitive to 1,000 atm, whereas with E. coli there was a 10-fold die-off per 50-h exposure to 1,000 atm. In contrast, V. parahaemolyticus and C. perfringens were quite sensitive to pressures exceeding 250 atm, and with both of these species there was a total loss of viability of approximately 10(8) cells per ml within 100 h at 1,000 arm and within 200 h at 500 atm. The viability of the naturally occurring fecal coliforms in sewage exposed to moderate pressures at 4 C was found to be similar to the survival patterns demonstrated with pure cultures of E. coli. The total numbers of aerobic bacteria in these sewage samples, however, stabilized at 500 and 1,000 atm after 100 h, and at 1 and 250 atm there was significant growth of sewage-associated bacteria, which apparently utilized the organic compounds in the seawater-diluted sewage samples. A preliminary classification of some of these bacteria indicated that approximately 90% (160 isolates) of the organisms that survived over a 400-h exposure to 500 and 1,000 atm were Arthrobacter/Corynebacterium species, and the representative organisms capable of growing at 1 and 250 atm in seawater at 4 C were gram-positive cellulose digesters and an unidentified gram-negative coccus. The significance of these results with respect to the contamination of the deep ocean with human pathogens and the possibility of sewage-associated microorganisms growing and competing with indigenous marine microbial flora in situ is discussed.

Cosmids containing hydrogen uptake genes have previously been isolated in this laboratory. Four new cosmids that contain additional hup gene(s) have now been identified by conjugal transfer of a Rhizobium japonicum 122DES gene bank into a Tn5-generated Hup(-) mutant and screening for the acquisition of Hup activity. The newly isolated cosmids, pHU50-pHU53, contain part of the previously isolated pHU1 but extend as far as 20 kilobases beyond its border. pHU52 complements five of six Hup(-) mutants and confers activity on several Hup(-) wild-type R. japonicum strains in the free-living state and where tested in nodules. Transconjugants obtained from interspecies transfer of pHU52 to Rhizobium meliloti 102F28, 102F32, and 102F51 and Rhizobium leguminosarum 128C53 showed hydrogen-dependent methyleneblue reduction, performed the oxyhydrogen reaction, and showed hydrogen-dependent autotrophic growth by virtue of the introduced genes. The identity of the presumptive transconjugants was confirmed by antibiotic-resistance profiles and by plant nodulation tests.

Recombinant cosmids from a gene library of the DNA from Hup+ Rhizobium japonicum 122DES previously have been shown to restore hydrogenase activity when transferred by conjugation into certain Hup- mutants of R. japonicum. We generated a restriction map covering 32.2 kilobases of this cosmid DNA. At least 25.3 kilobases of the cosmid pHU1 were shown to have the same arrangement as those in the genome of strain 122DES. Analysis of Tn5 insertions into the 122DES genome indicates that hup-specific sequences occur in a region spanning about 15 kilobases of insert DNA within pHU1. Introduction of pHU1 into five out of six R. japonicum Hup- mutants resulted in a Hup+ phenotype in some transconjugants. Three of the mutations appear to be in transcriptional units completely contained within pHU1, whereas the other two must be in genes that are at least partially contained within pHU1. pBR235 derivatives containing fragments of hup DNA can be transferred into the R. japonicum Hup- mutant PJ18nal if the derivatives contain a region of homology with the R. japonicum genome. The hup mutation in strain PJ18nal appears to be dominant. The hup genes in R. japonicum strain 122DES appear to be organized in at least two, and probably three, transcriptional units.

The use of the reverse-flow filtration technique to quantitatively concentrate marine bacteria was evaluated using both a pure culture and seawater samples. Our data indicate that the cells are altered during the filtration procedure and that a significant and inconsistent number of cells are lost on the membrane filter. The results obtained indicate that data on marine bacteria concentrated in this manner should be interpreted with caution.

We adapted a method for the rapid screening of colonies of free-living Rhizobium japonicum for hydrogenase activity to determine the hydrogenase status of individual soybean nodules. Crude bacteroid suspensions from nodules containing strains known to be hydrogen uptake positive (Hup) caused a localized decolorization of filter paper disks, whereas suspensions from nodules arising from inoculation with hydrogen uptake-negative (Hup) mutants or strains did not decolorize the disks. The reliability of the method was demonstrated by its successful application to 29 slow-growing rhizobia. The Hup phenotype on methylene blue filters agreed with that determined amperometrically with either methylene blue or oxygen as the electron acceptor.

Recently reported research from this laboratory has demonstrated the autotrophic growth of certain hydrogen-uptake-positive strains of Rhizobium japonicum and defined minimal conditions for such growth. Ribulose 1,5-bisphosphate carboxylase has been detected in autotrophically growing cells, but at low specific activity. Moreover, growth rates were low, and growth ceased at low cell densities. We report here improved autotrophic growth rates of R. japonicum SR through the use of a modified mineral salts/vitamins medium and a programmed increase in oxygen tension as autotrophic growth proceeds. Under these conditions, ribulose, 1,5-biphosphate carboxylase activity increased greater than 10-fold and crude-extract-uptake-hydrogenase activities were from 20 to 47 times those heretofore reported for free-living R. japonicum. It is likely that previous assays for these enzymes were done on preparations of cells in which their synthesis had been partially repressed. The contribution of CO2 fixation to organic carbon accumulation in autotrophic cells was assessed as sufficient to support observed growth. Enzymological determination of the product of carbon fixation has established a stoichiometric ratio of 1.9 mol of 3-phosphoglycerate per mol of CO2 fixed and unequivocally assigns the role of carbon fixation catalysis to ribulose 1,5-bisphosphate carboxylase. Ammonium served best as a nitrogen source, nitrate was less effective, and gaseous nitrogen would not support autotrophic growth. Ecological, evolutionary, and practical considerations of autotrophy in the rhizobia are briefly discussed in the light of our findings.

Previous research from this laboratory has demonstrated CO(2)-fixing and H(2)-uptake capacities of certain strains of Rhizobium japonicum. In this report we have shown that SR, a H(2)-uptake-positive (Hup(+)) strain of R. japonicum, is capable of autotrophic growth with H(2) as the energy source. Growth occurred on mineral salts/vitamins/Noble agar, mineral salts/vitamins liquid medium (0.27 mug of C as vitamins per ml), and in mineral salts liquid medium with no added vitamins when cultures were provided with NH(4)Cl and incubated in an atmosphere containing H(2), CO(2), O(2), and N(2). Little or no growth occurred when either H(2) or CO(2) was omitted from the atmosphere or when the culture was inoculated with SR3, a Hup(-) mutant of SR. Growth was measured by protein synthesis, fixed organic carbon, and increase in cell number in liquid cultures. The organism that grew autotrophically was verified as R. japonicum by (i) apparent purity on streak plates; (ii) retention of the double antibiotic resistance markers; and (iii) its capability to nodulate soybeans. H(2)- and CO(2)-supported growth was demonstrated for three additional Hup(+) wild-type R. japonicum strains (USDA 136, 3I1b 6, and 3I1b 143), while three Hup(-) wild-type strains (USDA 120, 3I1b 144, and USDA 117) were incapable of growth on the Noble agar medium containing mineral salts/vitamins in the H(2)/CO(2)/O(2)/N(2) atmosphere. This demonstrated capability of Hup(+)R. japonicum strains to grow autotrophically requires revision of current concepts regarding conditions for survival and competition of these bacteria in the soil and their relationships to other microorganisms.

Some Rhizobium strains synthesize a unidirectional hydrogenase system in legume nodule bacteroids; this system participates in the recycling of hydrogen that otherwise would be lost as a by-product of the nitrogen fixation process. Soybeans inoculated with Rhizobium japonicum strains that synthesized the hydrogenase system fixed significantly more nitrogen and produced greater yields than plants inoculated with strains lacking hydrogen-uptake capacity. Rhizobium strains used as inocula for legumes should have the capability to synthesize the hydrogenase system as one of their desirable characteristics.

Eight strains of Rhizobium lacking hydrogenase uptake (Hup) activity and 17 transconjugant strains carrying the hup cosmids pHU1, pHU52, or pHU53 (G. R. Lambert, M. A. Cantrell, F. J. Hanus, S. A. Russell, K. R. Haddad, and H. J. Evans, Proc. Natl. Acad. Sci. USA, 82:3232-3236, 1985) were screened for Hup activity and the presence of immunologically detectable hydrogenase polypeptides. Crude extracts of these strains were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot analysis with affinity-purified antibodies against the two subunits of purified hydrogenase (Mr 60,000 and 30,000). Derepressed transconjugants carrying the cosmid pHU52 were Hup+ and contained detectable levels of both hydrogenase subunit polypeptides. Non-derepressed strains, Hup- parent strains, and strains carrying cosmids other than pHU52 did not express Hup activity and contained no immunologically detectable protein. These data provide further evidence for the essential involvement of the smaller (Mr 30,000) subunit in the expression of hydrogenase activity in Rhizobium japonicum and suggest that the determinants for hydrogenase subunit synthesis are present on pHU52.

All aerobic biological systems, including N(2)-fixing root nodules, are subject to O(2) toxicity that results from the formation of reactive intermediates such as H(2)O(2) and free radicals of O(2). H(2)O(2) may be removed from root nodules in a series of enzymic reactions involving ascorbate peroxidase, dehydroascorbate reductase, and glutathione reductase. We confirm here the presence of these enzymes in root nodules from nine species of legumes and from Alnus rubra. Ascorbate peroxidase from soybean nodules was purified to near homogeneity. This enzyme was found to be a hemeprotein with a molecular weight of 30,000 as determined by sodium dodecyl sulfate gel electrophoresis. KCN, NaN(3), CO, and C(2)H(2) were potent inhibitors of activity. Nonphysiological reductants such as guaiacol, o-dianisidine, and pyrogallol functioned as substrates for the enzyme. No activity was detected with NAD(P)H, reduced glutathione, or urate. Ascorbate peroxidation did not follow Michaelis-Menten kinetics. The substrate concentration which resulted in a reaction rate of (1/2) V(max) was 70 micromolar for ascorbate and 3 micromolar for H(2)O(2). The high affinity of ascorbate peroxidase for H(2)O(2) indicates that this enzyme, rather than catalase, is responsible for most H(2)O(2) removal outside of peroxisomes in root nodules.

Factors that regulate the expression of an H2 uptake system in free-living cultures of Rhizobium japonicum have been investigated. Rapid rates of H2 uptake by R. japonicum were obtained by incubation of cell suspensions in a Mg-phosphate buffer under a gas phase of 86.7% N2, 8.3% H2, 4.2% CO2, and 0.8% O2. Cultures incubated under conditions comparable with those above, with the exception that Ar replaced H2, showed no hydrogenase activity. When H2 was removed after initiation of hydrogenase derepression, further increase in hydrogenase activity ceased. Nitrogenase activity was not essential for expression of hydrogenase activity. All usable carbon substrates tested repressed hydrogenase formation, but none of them inhibited hydrogenase activity. No effect on hydrogenase formation was observed from the addition of KNO3 or NH4Cl at 10 mM. Oxygen repressed hydrogenase formation, but did not inhibit activity of the enzyme in whole cells. The addition of rifampin or chloramphenicol to derepressed cultures resulted in inhibition of enzyme formation similar to that observed by O2 repression. The removal of CO2 during derepression caused a decrease in the rate of hydrogenase formation. No direct effect of CO2 on hydrogenase activity was observed.

We have investigated the effect of added selenite on autotrophic growth and the time course of hydrogen oxidation derepression in Bradyrhizobium japonicum 122DES cultured in a medium purified to remove selenium compounds. In addition, hydrogenase was purified to near homogeneity and examined for the specific incorporation of Se into the enzyme. The addition of Se at 0.1 microM significantly increased total cell protein and hydrogenase specific activity of harvested cells. Also, the addition of SeO3(2-) enhanced the time course of hydrogenase derepression by 133%, whereas VO3, AsO2(2-), SO2(2-), and TeO3(2-) failed to substantially affect hydrogenase derepression. During the final chromatographic purification of hydrogenase, a striking coincidence in peaks of protein content, Se radioactivity, and hydrogenase activity of fractions was obtained. The total Se content expressed per milligram of protein increased manyfold during the purification procedure. The mean Se content of the purified hydrogenase was 0.56 +/- 0.13 mol of Se per mol of enzyme. These results indicate that Se is an important element in the H2 metabolism of B. japonicum and that hydrogenase from B. japonicum is a seleno protein.

The expression of cosmid-borne Bradyrhizobium japonicum hydrogenase genes in alfalfa, clover, and soybean nodules harboring Rhizobium transconjugants was studied. Cosmid pHU52 conferred hydrogen uptake (Hup) activity in both free-living bacteria and in nodules on the different plant hosts, although in nodules the instability of the cosmid resulted in low levels of Hup activity. In contrast, cosmid pHU1, which does not confer Hup activity on free-living bacteria, gave a Hup phenotype in nodules on alfalfa and soybean. Nodules formed by B. japonicum USDA 123Spc(pHU1) recycled about 90% of nitrogenase-mediated hydrogen evolution. Both subunits of hydrogenase (30- and 60-kilodalton polypeptides) were detected in enzyme-linked immunosorbent assays of bacteroid preparations from nodules harboring B. japonicum strains with pHU1 or pHU52. Neither pHU53 nor pLAFR1 conferred detectable Hup activity in either nodules or free-living bacteria. Based on the physical maps of pHU1 and pHU52, it is suggested that a 5.5-kilobase EcoRI fragment unique to pHU52 contains a gene or part of a gene required for Hup activity in free-living bacteria but not in nodules. This conclusion is supported by the observation that two Tn5 insertions in the chromosome of B. japonicum USDA 122DES obtained by marker exchange with Tn5-mutagenized pHU1 abolished Hup activity in free-living bacteria but not in nodules.

The critical problem of oxygen toxicity for nitrogen-fixing organisms may be related to damage caused by oxygen radicals and peroxides. An enzymatic mechanism is described for removal of peroxides in root nodules of soybean (Glycine max). The system utilizes ascorbate as an antioxidant and glutathione as a reductant to regenerate ascorbate. The enzymes involved are ascorbate peroxidase (ascorbate:hydrogen-peroxide oxidoreductase, EC 1.11.1.7), dehydroascorbate reductase (glutathione:dehydroascorbate oxidoreductase, EC 1.8.5.1), and glutathione reductase (NADPH:oxidized-glutathione oxidoreductase, EC 1.6.4.2). The reactions are essentially the same as those involving scavenging of H(2)O(2) in chloroplasts. Glutathione peroxidase (glutathione:hydrogenperoxide oxidoreductase, EC 1.11.1.9) was not detected. During the course of early nodule development, ascorbate peroxidase and dehydroascorbate reductase activities and total glutathione contents of nodule extracts increased strikingly and were positively correlated with acetylene reduction rates and nodule hemoglobin contents. The evidence indicates an important role of glutathione, ascorbate, ascorbate peroxidase, dehydroascorbate reductase, and glutathione reductase as components of a peroxide-scavenging mechanism in soybean root nodules.

A medium is described on which selected Rhizobium japonicum strains express hydrogenase (H(2) uptake) activity under free-living conditions. Low concentrations of carbon substrates, decreased oxygen tension, and the quantity of combined nitrogen in the medium were major factors influencing hydrogenase expression. Hydrogenase activity was dependent upon a preincubation period in the presence of H(2) under conditions such that the cells did not exhibit nitrogenase activity. H(2) uptake rates were easily measured amperometrically in aerobically or anaerobically prepared suspensions from free-living cultures. Six R. japonicum strains that formed nodules with the ability to utilize H(2) oxidized this gas when grown in free-living cultures. In comparison six randomly chosen strains forming nodules that lost H(2) in air either showed no or low capacity to take up H(2) under free-living conditions. The reduction of triphenyltetrazolium chloride in an agar medium was used to detect strains capable of oxidizing H(2). This method has enabled us to isolate a spontaneous R. japonicum mutant strain that has lost the ability to utilize H(2). This mutant strain forms nodules that evolve H(2) but other symbiotic characteristics appear normal. This strain will be useful in evaluating the importance of the hydrogenase system in the nitrogen-fixing process of legumes.

A method has been developed for the rapid screening of Rhizobium japonicum colonies for hydrogenase activity based on their ability to reduce methylene blue in the presence of respiratory inhibitors and hydrogen. Hydrogen uptake-positive (Hup) colonies derepressed for hydrogenase activity were visualized by their localized decolorization of filter paper disks impregnated with the dye. Appropriate responses were seen with a number of Hup and Hup wild-type strains of R. japonicum as well as Hup mutants. Its specificity was further confirmed in selected strains on the basis of comparisons with chemolithotrophic growth and the presence of other genetic markers. Utilization of the method in identifying Hup colonies among 16,000 merodiploid derivatives of the Hup mutant strain PJ17nal containing cloned DNA fragments of the Hup strain 122 DES has demonstrated its applicability as a screening procedure in the genetic analysis of the R. japonicum hydrogen uptake system.

Agar has been shown to interfere with the activity of some antibiotics against Staphylococcus aureus. This interference has been observed as an increase in the minimal inhibitory concentration and in the diameter of the zone of inhibition. Purifying the agar with water extractions substantially reduced this adverse effect.