Rhizobitoxine inhibition of hydrogenase synthesis in free-living Bradyrhizobium japonicum.

Rhizobitoxine produced by Bradyrhizobium species strongly prevented derepression of hydrogenase expression in free-living Bradyrhizobium japonicum, although the toxin had no effect on the activity of cells which had already synthesized hydrogenase protein. Dihydrorhizobitoxine, a structural analog of rhizobitoxine, proved to be a less potent inhibitor of hydrogenase derepression. Rhizobitoxine did not cause cell death at a concentration sufficient to eliminate hydrogenase expression. The large subunit of hydrogenase was not detectable with antibody after derepression in the presence of rhizobitoxine. The general pattern of proteins synthesized from 14C-labeled amino acids during derepression was not significantly different in the presence or absence of rhizobitoxine. These results indicated that rhizobitoxine inhibited hydrogenase synthesis in free-living B. japonicum. Cystathionine and methionine strongly prevented the inhibition of hydrogenase derepression by rhizobitoxine, suggesting that the inhibition involves the level of sulfur-containing amino acids in the cell.     

The HypB protein from Bradyrhizobium japonicum can store nickel and is required for the nickel-dependent transcriptional regulation of hydrogenase

The HypB protein from Bradyrhizobium japonicum is a metal-binding GTPase required for hydrogenase expression. In-frame mutagenesis of hypB resulted in strains that were partially or completely deficient in hydrogenase expression, depending on the degree of disruption of the gene. Complete deletion of the gene yielded a strain (JHΔEg) which lacked hydrogenase activity under all conditions tested, including the situation as bacteroids from soybean nodules. Mutant strain JHΔ23H lacking only the N-terminal histidine-rich region (38 amino acids deleted, 23 of which are His residues) expressed partial hydrogenase activity. The activity of strain JHΔ23H was low in comparison to the wild type in 10–50 nM nickel levels, but could be cured to nearly wild-type levels by including 50 μM nickel during the derepression incubation. Studies on strains harbouring the hup promoter–lacZ fusion plasmid showed that the complete deletion of hypB nearly abolished hup promoter activity, whereas the histidine deletion mutant had 60% of the wild-type promoter activity in 50 μM NiCl₂. Further evidence that HypB is required for hup promoter-binding activity was obtained from gel-shift assays. HypB could not be detected by immunoblotting when the cells were cultured heterotrophically, but when there was a switch to microaerobic conditions (1% partial pressure O₂, 10% partial pressure H₂) HypB was detected, and its expression preceded hydrogenase synthesis by 3–6 h. ⁶³Ni accumulation by whole cells showed that both of the mutant strains accumulate less nickel than the wild-type strain at all time points tested during the derepression incubation. Wild-type cultures that received nickel during the HypB expression-specific period and were then washed and derepressed for hydrogenase without nickel had activities comparable to those cells that were derepressed for hydrogenase with nickel for the entire time period. In contrast to the wild type, strain JHΔ23H cultures supplied with nickel only during the HypB expression period achieved hydrogenase activities that were 30% of those cultures supplied with nickel for the entire hydrogenase derepression period. These results indicate that the loss of the metal-binding area of HypB causes a decrease in the ability of the cells to sequester and store nickel for later use in one or more hydrogenase expression steps.     

Nickel-dependent reconstitution of hydrogenase apoprotein in Bradyrhizobium japonicum Hupc mutants and direct evidence for a nickel metabolism locus involved in nickel incorporation into the enzyme

A double mutant (JH103K10) was created from hydrogenase constitutive mutant (JH103) by replacement of a chromosomal 0.60 kb nickel metabolism related locus with a kanamycin resistance gene. The double mutant required 10 to 20 times more nickel (Ni) to achieve near parental strain levels of hydrogenase activity. In the absence of nickel, both JH103K10 and JH103 synthesized high levels of (inactive) hydrogenase apoprotein (large subunit, 65 kDa). With nickel, the double mutant JH103K10 synthesized the same level of hydrogenase apoenzyme (65-kDa subunit) as the JH103 parent strain; however, whole cell hydrogenase activity in JH103K10 was less than half of that in JH103, and the CPM (due to ⁶³Ni in hydrogenase) of membranes and the calculated ratio of nickel per unit of hydrogenase enzyme of the double mutant were 40% of that in JH103. Therefore, the difference in hydrogenase activities between the double mutant and the Hup^ck strain can be accounted for by different abilities of the strains to incorporate nickel into the hydrogenase apoenzyme. The addition of nickel ions to previously Ni-starved and then chloramphenicol-treated Bradyrhizobium japonicum whole cells (JH103 and JH103K10) resulted in (an in vivo) restoration of hydrogenase activity, suggesting that the apoprotein synthesized in the Ni-free cultures could be activated by addition of nickel even in the absence of protein synthesis. The extent of reconstitution of active hydrogenase by nickel was greater in the absence of chloramphenicol. Hydrogenase apoprotein could not be activated by nickel in vitro even with the addition of ATP. The successful in vivo but not in vitro results suggest that enzymatic but cell-disruption labile factors are required for Ni incorporation into hydrogenase.     

Feedback regulation of the Bradyrhizobium japonicum nodulation genes

Lipochitin Nod signals are produced by rhizobia and are required for the establishment of a nitrogen-fixing symbiosis with a legume host. The nodulation genes encode products required for the synthesis of this signal and are induced in response to plant-produced flavonoid compounds. The addition of chitin and lipo-chitin oligomers to Bradyrhizobium japonicum cultures resulted in a significant reduction in the expression of a nod–lacZ fusion. Intracellular expression of NodC, encoding a chitin synthase, also reduced nod gene expression. In contrast, expression of the ChiB chitinase increased nod gene expression. The chain length of the oligosaccharide was important in feedback regulation, with chitotetraose molecules the best modulators of nod gene expression. Feedback regulation is mediated by the induction of nolA by chitin, resulting in elevated levels of the repressor protein, NodD2.     

Lack of carbon substrate repression of uptake hydrogenase activity in Bradyrhizobium japonicum SR.

The expression of ex planta uptake hydrogenase (Hup) activity in Bradyrhizobium japonicum SR induced in the absence or presence of carbon substrates was compared. Hup activity was influenced by pH, indicating that acidification of induction medium with low buffering capacity resulting from carbon substrate metabolism inhibited Hup activity. Cell suspensions in medium with adequate buffering capacity and carbon substrate were limited in O2; increasing O2 availability to cells during induction stimulated Hup expression. The data showed a lack of carbon substrate repression of Hup activity in cell suspensions provided with adequate O2 and buffering capacity.     

Investigation of the form of selenium in the hydrogenase from chemolithotrophically cultured Bradyrhizobium japonicum

It has been established that the hydrogenase from autotrophically cultured Bradyrhizobium japonicum contains selenium as a bound constituent. About 80% of the enzyme selenium remains bound during precipitation with 5% trichloroacetic acid (TCA). However, 85% of the selenium bound to the enzyme is released by a combined treatment of urea, heat and TCA. Neither selenomethionine nor selenocysteine could be detected on analysis of anaerobically hydrolyzed enzyme. These results are consistent with the report showing that the structural genes for this enzyme do not contain a TGA codon (Sayavedra-Soto et al. 1988) which has been reported to code for selenocysteine incorporation into several proteins (Chambers et al. 1986; Zinoni et al. 1986; Stadtman 1987). We have demonstrated that ⁷⁵Se from the labeled hydrolyzed enzyme forms the derivative' selenodicysteine. The form of selenium resulting in the synthesis of this derivative apparently is SeO _inf3 ^sup= or a compound such as Se⁼ which is easily oxidized to SeO _inf3 ^sup= . In a separate approach it was established that 12–16% of the total ⁷⁵Se in the native enzyme reacted with 2,3-diaminonaphthalene indicating that this fraction was present as SeO _inf3 ^sup= . The remaining ⁷⁵Se was bound to the enzyme protein. From this research, we concluded that Se in Bradyrhizobium japonicum hydrogenase is present in a labile bound form. In this respect, this enzyme is similar to xanthine dehydrogenase and nicotinic acid hydroxylase, both of which contain labile Se constituents that have not been defined.Technical paper no. 8980 from the Oregon Agricultural Experiment Station