Microbes Environ Vol 23, No 3, 215–220, 2008 http://wwwsoc.nii.ac.jp/jsme2/ doi:10.1264/jsme2.23.215 NAD-Malic Enzyme Affects Nitrogen Fixing Activity of Bradyrhizobium japonicum USDA 110 Bacteroids in Soybean Nodules TAN VAN DAO1,2, MIKA NOMURA2, RIE HAMAGUCHI2, KENSUKE KATO2, MANABU ITAKURA3, KIWAMU MINAMISAWA3, SUPHAWAT SINSUWONGWAT4, HOA THI-PHUONG LE5, TAKAKAZU KANEKO6, SATOSHI TABATA6, and SHIGEYUKI TAJIMA1* 1Centre for Ressources and Environmental Studies, Vietnam National University, No 22 Ngo Luong Su B, Quoc Tu Giam Str Hanoi, Vietnam; 2Faculty of Agriculture, Kagawa University, Miki-cho, Kita-gun, Kagawa 761–0795, Japan; 3Graduate School of Life Sciences, Tohoku University, 2–1–1, Katahira, Aoba-ku, Sendai 980–8577, Japan; 4Department of Biotechnology, Faculty of Agro-Industry, Chiangmai University, Chiangmai, 50100, Thailand; 5Faculty of Biology, Hanoi University of Education, 136 XuanThuy Road, Hanoi Vietnam; and 6Kazusa DNA Research Institute, 2–6–7 Kazusa-Kamatari, Kisarazu, Chiba 292–0812, Japan (Received May 23, 2008—Accepted June 18, 2008) The NAD+-dependent malic enzyme (DME) has been reported to play a key role supporting nitrogenase activity in bacteroids of Sinorhizobium meliloti Genetic evidence for a similar role in Bradyrhizobium japonicum USDA110 was obtained by constructing a dme mutant Soybean plants inoculated with a dme mutant did not show delayed nodulation, but formed small root nodules and exhibited significant nitrogen-deficiency symptoms Nodule numbers and the acetylene reducting activity per nodule as a dry weight value 14 and 28 days after inoculation with the dme mutant were comparable to those of plants inoculated with wild-type B japonicum However, shoot dry weight and acetylene reducting activity per nodule decreased to ca 30% of the values in plants with wild-type B japonicum The sucrose and organic acid (malate, succinate, acetate, α-ketoglutarate and lactate) contents of the nodules were investigated Amounts of sucrose, malate and α-ketoglutarate increased on inoculation with the dme mutant, suggesting that the decreased DME and nitrogenase activities in the bacteroids resulted in a reduction in the consumption of these respiratory metabolites by the nodules The data suggest that the DME activity of B japonicum bacteroids plays a role in nodule metabolism and supports nitrogen fixation Key words: Bradyrhizobium japonicum, NAD-malic enzyme, symbiotic nitrogen fixation, soybean nodules Rhizobia can infect plant roots and form symbiotic tissue (nodules) This symbiosis contributes to agriculture because the bacteria differentiate into bacteroids expressing nitrogenase, an enzyme that converts nitrogen gas to ammonia In the nodules, bacteroids are surrounded by a peribacteroid membrane and form a stable metabolic unit This structure is believed to function like an intracellular organelle, and is designated the symbiosome4,31) Nitrogenase can be expressed in various soil bacteria, but it is the endo-symbiotic systems connecting the supply of photosynthetic products from higher plants to bacteroid nitrogenase systems that make this kind of system efficient and important to agriculture Since the nitrogenase reaction consumes a large amount of ATP, there have been various reports on the unique carbon metabolism under endo-symbiotic conditions4,31) The bacteroids have been reported to use glucose as a respiratory substrate inefficiently and only C4 dicarboxylic acids and amino acids can be respiratory substrates for supplying ATP to nitrogenease in bacteroids Since biochemical and radio-respiratory experiments strongly suggested operation of the TCA cycle in nitrogen-fixing bacteroids, anaplerotic reactions supplying acetyl-CoA from C4 dicarboxylic acids are necessary3,4,26,31) To generate a continuous supply of acetyl-CoA, malic enzymes (EC 1.1.1.38) are believed to be operative in bacteroids as anaplerotic enzymes3,4,26,31) Genetic evidence of the role in nitrogenase activity of a malic enzyme in the bacteroids has been reported by Driscoll and Finan5,6) They found that dme mutants of Sinorhizobium meliloti were Nod+ and Fix−, and conversely, that a NADP+dependent malic enzyme (TME)-defective mutant was Nod+ and Fix+ In a recent report14), they examined the biochemical background of S meliloti bacteroids by preparing various types of dme mutants Although the conversion of pyruvate and malate was similar for DME and TME, their data showed that the high expression level of TME did not restore DME activity In addition, Escherichia coli NAD+-dependent malic enzyme restored wild-type N2 fixation by the dme mutant14) In Bradyrhizobium japonicum, however, there has been no genetic evidence for roles of DME and TME in the nitrogenase activity of bacteroids in nodules Radio-respirometric experiments of succinate degradation suggested operation of the TCA cycle in isolated bacteroids26), and purification and characterization of DME and TME of B japonicum also suggested biochemical characteristics in soybean nodule bacteroids1) Currently, genome sequence data shows that there are three malic enzyme genes in B japonicum USDA11011); TME (blr3726), DME (blr4145), and a malic enzyme (bll6469), although a Blast2 search of bll6469 suggested that it is TME Of note, blr3726 (TME) and blr4145 (DME) both have PTA domains, whereas bll6469 does not, that is the former two are related in structure and highly homologus whereas the third enzyme appears to be related to different NADP-dependent malic enzymes If both 3726 and 6469 are TMEs, it would seem logical to incorrectly assume 216 that they are duplicate alleles, but they are very different This might also be more useful information than EC For this investigation, we prepared a dme (blr4145) mutant of B japonicum USDA110 and analyzed the symbiotic phenotype of inoculated soybean plants Materials and Methods Bacterial strains, plasmids and culture conditions The cells of Bradyrhizobium japonicum USDA110 were grown at 30°C in HM salts medium2) supplemented with 0.1% (w/v) arabinose and 0.025% (w/v) yeast extract (Difco, Detroit, MI, USA) Escherichia cells were grown at 37°C in Luria-Bertani medium19) Antibiotics were added to media at the following concentrations, unless otherwise indicated; spectinomycin 50 µg mL−1, streptomycin 50 µg mL−1 for B japonicum: and spectinomycin 50 µg mL−1, streptomycin 50 µg mL−1, kanamycin 50 µg mL−1 and ampicillin 100 µg mL−1 for E coli DNA manipulations The isolation of plasmid, ligation of DNA and bacterial transformation of E coli were performed as described by Sambrook et al.19) All DNA samples were prepared from B japonicum USDA110 aerobically grown in HM broth medium as described previously2,10,13) Construction of a B japonicum USDA110 dme mutant The cloning of the dme gene of B japonicum USDA110 (blr4145) was performed by PCR Primers for the PCR were prepared based on the genome sequence of B japonicum USDA110 (http://www.kazusa.or.jp/rhizobase/, NADME5Fw: 5'-CCCGAATTCATGTCGTCTTATTCT-3', NADME3Rv: 5'-GGGAAGCTTTCAGGTAAACAACGT-3')10), and the PCR product, once sequenced, was ligated with a pGEM-T easy vector (Promega Corporation, Madison, WI) Mutagenesis targeting the dme gene was performed as reported by Sameshima-Saito et al.20,21) The dme gene was excised as a 3.3kb EcoRI fragment, and inserted into the EcoRI site of pK18mob23) to generate the plasmid pKdme (7.0 kb) For the deletion of dme, the Ω cassette, which was excised from pHP45Ω17) digested with SmaI, was inserted into the EcoRV sites of pKDdme to generate pKDdme::Ω (9.0 kb) Then, pKDdme::Ω was introduced into B japonicum USDA110 by triparental mating using pRK2013 as a helper plasmid7) Putative double-crossover mutants resistant to spectinomycin (50 µg mL−1), streptomycin (50 µg mL−1), polymixin (50 µg mL−1) and chloramphenicol (50 µg mL−1), and sensitive to kanamycin were selected The double crossover event was further verified by Southern hybridization using 714-bp fragments carrying the dme gene (blr4145) and 972-bp Ω gene as a probe These fragments were amplified by PCR using the following primers (NAD6 Fw: 5'-CAAGCCCGTGATGGAGGGCA-3', NAD7 Rv: 5'-AGGGAAAGCACAGGACGTTGTTGA-3' Omega-Fw: 5'-CTTGACCTGATAGTTTGGCTGTGAG-3', Omega-Rv: 5'-GGGTCGATGTTTGATGTTATGGAGC-3') The resulting dme mutant (BjDME::Ω; B japonicum strain USDA110 Ω cassette; Smr Spr) was used for elucidation of the DME in nitrogen-fixing bacteroids Plant materials Soybean seeds (Glycine max cv Akishirome) were sterilized in a 70% ethanol solution for min, and then in a 0.5% sodium hypochlorite solution containing 0.02% Tween 20 for The seeds were rinsed with distilled water five times They were then planted in a sterilized vermiculite bed in a sealed plastic box After germination of the seeds at 25°C in the dark, the seedlings were transplanted to a Leonard-jar plastic bag16) carrying sterilized vermiculite A B&D nitrogen-free nutrient was supplied at half strength every week16) The plants were grown in incubators at 25°C (12,000 lux; 16 hrs light, hrs dark) The height and dry weight of DAO et al sampled plants were measured after nodules were collected The nodule dry weight and nodule number per plant were measured after the acetylene-reducting activity wad assayed Inoculation of rhizobia and preparation of bacteroids B japonicum USDA110 was cultured at 28°C for days with HM medium BjDME::Ω was cultured with HM medium containing streptomycin (50 µg mL−1) and spectinomycin 50 µg mL−19) For the isolation of bacteroids, freshly harvested nodules (1 g) were gently ground with times their volume of 50 mM Tris-HCl The homogenate was subjected to step-wise centrifugations and the bacteroid fraction was stored in a −80°C freezer9) Biochemical assays Preparation of a crude enzyme solution and the malic enzyme assay were performed as we have reported11,14) Malate dehydrogenase activity was determined as reported by Smith25) Protein concentrations were determined with a Bio-Rad protein assay reagent (Bio-Rad, Hercules, CA) The organic acid content of nodules was determined using a L7100 HPLC system (Hitachi High-Technologies Co., Hitachi, Japan) with two RSpak KC-811 columns (Showa Denko KK., Tokyo, Japan) as reported16) The sucrose content of the nodules was analyzed with a Shimazu LCMS-2010 system with a SIM-pack SPR-Ca column (Shimazu Co., Kyoto, Japan) The acetylene-reducting activity of the nodules was analyzed as reported7) The nodules detached from roots were analyzed within hr at 25°C Preparation of DME antibody and immunoblot analysis DME protein was overexpressed using pET32a(+)(Novagen, San Diego, CA) The DME fusion protein was purified by ammonium sulfate precipitation and a HisTrap Chelating column The protein was used to prepare a polyclonal antibody in a rabbit The antiserum was used for the immuno-blot analysis16) Microscopic observation of root nodules The nodules were fixed with a Technovit 7100 (Heraeus Kulzer, Germany) The embedded plant materials were cut to a thickness of µm with a microtome (Leica Camera AG, Solms, Germany) Dried sections were stained in a solution of 1% Toloudine Blue for then rinsed with water16) Results Expression of malic enzymes in wild-type and the dme mutant of B japonicum USDA110 Fig shows an immunoblot analysis for detecting DME protein using the extracts of wild-type cultured cells and of bacteroids The wild-type cells showed one positive band at 82.4 kD which corresponds to the molecular weight of the DME protein of B japonicum (blr4145) In contrast, no positive band was detected in the extract from the cultured cells or bacteroids of the dme mutant (BjDME::Ω), demonstrating that DME was absent in free-living and bacteroid cells The growth rates in the culture medium were almost the same for the wild type and the dme mutant When three malate-metabolizing enzymes (DME, TME and malate dehydrogenase; MDH) were analyzed in cultured cells and bacteroids, the DME activity of the dme mutant was found to have decreased significantly in both cells and bacteroids, in comparison to that of the wild-type (Table 1) DME activity per mg protein in the dme mutant bacteroids was ca 22% of that in thewild-type B japonicum bacteroids In contrast, no significant differences were observed between NAD-Malic Enzyme Gene in B japonicum 217 Fig Immunoblot analysis to detect DME protein in extracts of freeliving cells and bacteroids Proteins of DME mutant (BjDME::Ω) and wild type (WT) of B japonicum USDA110 were extracted from cultured cells and from bacteroids in the nodules Nodules were harvested at 28 days after the inoculation of B japonicum USDA110 wild-type or dme mutant Total bacterial protein (20 µg) was applied to each lane Molecular weight (MW) markers are indicated by short bars the wild type and the dme mutant in either TME or MDH activity in cultured cells and bacteroids, although the MDHspecific activity of wild-type B japonicum was increased significantly in the bacteroids compared to the free-living cells The remaining DME activity in BjDME::Ω cells might be the result of a compensative expression of other malatedegrading proteins, such as NADH generating dehydrogenase activity3) This reduction of DME activity in bacteroids influenced the values in the extract obtained using whole nodules The data strongly suggests that BjDME::Ω is a successful dme mutant of B japonicum which would cause a significant reduction of in vivo DME activity in nodule bacteroids The symbiotic phenotype of soybean plants inoculated with the dme mutant Soybean plants inoculated with BjDME::Ω showed a Nod+ phenotype and a slight increase in the number of nodules per plant at 28 days post-inoculation (Fig 2A) When the seedlings were supplied nitrogen-free B&D medium, the leaves turned yellow at the second to third main leaf development growth phase (10–14 days after inoculation), and the plant Table Various malate-metabolizing enzymatic activities in nodules, cultured cells and bacteroids of B japonicum USDA110 Enzymatic activities (nmol min−1 mg protein−1) DME TME MDH Wild type BjDME::Ω Wild type BjDME::Ω Wild type BjDME::Ω Cultured cells Nodules Bacteroids 117.3±13.0 15.7±2.7 10.3±1.5 28.9±8.6 7.3±1.5 2.3±0.4 39.6±6.3 32.1±3.3 21.6±5.8 37.2±4.3 27.6±5.6 22.0±2.9 309.1±34.8 375.5±107.3 1632.7±484.3 281.2±54.1 357.7±61.5 1058.3±133.6 Soybean nodules infected with wild-type B japonicum USDA110 or the dme mutant (BjDME::Ω) were harvested at 28 days post inoculation Malate dehydrogenase (MDH), NAD-malic enzyme (DME) and NADP-malic enzyme (TME) activities (means±SDs) were assayed as described in Materials and Methods Fig Comparison of symbiotic phenotypes of soybean nodules which were infected with wild-type B japonicum or the dme mutant Nodule number per plant (A), nodule dry weight (B) per plant and acetylene-reducting activity per gram dry weight nodules at 14 (C) and 28 (D) days after inoculation with wild-type B japonicum USDA110 (open bars) and the DME mutant (BjDME::Ω, solid bars) are presented Values are means±SDs for four plants Asterisks denote significant differences from wild type based on t test (P