HM Medical Clinic

 

研_07_小澤先生.indd

Bull. Hiroshima Inst. Tech. Research Vol.49（2015）27-33 Inhibitory effect of cadmium on competitive nodulation ability of Bradyrhizobium japonicum Takashi OZAWA* and Kenji IJIRI** （Received Oct. 23, 2014） Competitive nodulation abilities of Bradyrhizobium japonicum strains USDA110ET and A1017ET were significantly depressed by growing the strains in yeast extract-mannitol broth supplemented with 2 μM CdCl2. Soybean seedlings were co-inoculated with each test strain and its competitor strain 138NR, and the bacteria in nodules formed on 21 days old plant roots were identified by the antibiotic resistant markers. Cell surface hydrophobicity of each test strain increased by growing the strains with 2 μM CdCl2, though definite increase in bacterial attachment to plant root surface was hindered by excessive secretion of exopolysaccharides of the strains. Polyacrylamide gel electro-phoresis revealed that growing the test strains in the medium with 2 μM CdCl2 induced the produc-tion of lipopolysaccharides of small molecular sizes. The results in this study suggest that cadmium of low concentration causes the weakening of competitive nodulation ability of rhizobia through in-hibition of the lipopolysaccharide synthesis.
Key Words: cadmium, competitiveness, hydrophobicity, legume, lipopolysaccharide, rhizobium
manner affected markedly the CNA of the indigenous Bradyrhizobium strains (Ozawa et al. 2000). These Rhizobium is a soil bacterium that infects legumi- results suggest that some environmental factors af- nous plants and produces nodules in the root to fix fect the expression of the genes involving the com- atmospheric nitrogen symbiotically with their host plant. There is a difference in the root nodule forma- Cadmium (Cd) is one of the heavy metals gener- tion ability between strains of rhizobia (Dowling and ally existing as a minor element in soils. Cd is a Broughton 1986). A rhizobium strain should compete highly toxic metal that causes deleterious effects at with other strains to form a root nodule on their host higher concentrations on soil bacteria as well as plant and occupy it.
plants (Sanitá di Toppi and Gabbrielli 1999). Degree The competitive nodulation ability (CNA) of a of sensitivity of rhizobia to Cd differs from strain to rhizobial strain is genetically determined, but not strain. Consequently, contamination with Cd at high- constant. Environmental factors such as temperature, er concentrations could change a population struc- soil pH, and soil nitrate can affect the CNA of ture of rhizobia in the soil (Kinkle et al. 1987). Thus, Rhizobium and Bradyrhizobium (Triplett 1990). We the strain more tolerant to Cd would have more op- demonstrated that Bradyrhizobium japonicum portunity to form nodules on the host plant. Effect of strains changed their CNA if the strains grown in a Cd on the CNA of a rhizobial cell, however, has not nutrient broth were introduced into soil environment been elucidated. (Ozawa 1988). We have also reported that fertilization Cell surface hydrophobicity of a rhizobial strain ＊＊＊ Department of Food Sciences and Biotechnology, Hiroshima Institute of Technology＊＊＊ Department of Life and Environmental Sciences, Osaka Prefecture University Takashi OZAWA and Kenji IJIRI is one of the key characteristics that determine the fold with distilled water, and then mixed with the CNA of the strain (Araujo et al. 1994). We have dem- competitor strain 138NR to prepare five varieties of onstrated that Bradyrhizobium japonicum strains had inoculum. Bacterial cell densities of the strains were a positive correlation between the cell surface hydro- examined by the dilution plate count method. Densi- phobicity and the CNA (Ozawa et al. 1991, Ozawa ties of the test strains and the competitor in the inoc- et al. 1992). Cell surface hydrophobicity of Gram- ulum were approximately as follows: 106 and 108, 107 negative bacteria was determined by the amount and 108, 108 and 108, 108 and 107, 108 and 106 cfu mL-1, and nature of exopolysaccharide (EPS) and lipopoly- saccharide (LPS) (Nikaido 1976, Nikaido 1994, Triplett 1990). Schue et al. (2011) reported that extra- 2.2. Bacterial inoculation to plants
cellular Cd induces the over-production of EPS. Soybean (Glycine max cv. Tamanishiki) seeds sur- These results suggest that Cd could change the face sterilized with 5% calcium hypochlorite were CNA of a rhizobial strain by affecting bacterial sown on vermiculite (approximate particle size: 5 metabolic processes of polysaccharides. The objective mm) with a depth of 10 cm in test tubes (25 x 200 of this study is to confirm the possible role of extra- mm), one seed per tube. Each tube was previously cellular Cd in controlling the cell surface hydro- supplied with 30 mL of Jensen＇s mineral solution phobicity and the CNA of Bradyrhizobium japonicum (Gibson 1980), and sterilized by autoclaving (120°C, 20 min). After 3 days at 30°C in the dark, each seed-ling was inoculated with 1.0 mL of the mixed cell . Materials and Methods
suspension of a test strain and strain 138NR, which 2.1. Bacterial strains
was prepared as mentioned above.
B. japonicum USDA110 and USDA138 were ob- The inoculated plants were grown for 21 days in tained from the U.S. Department of Agriculture, a growth chamber (23°C; 110 μmol m-2 s-1, 16 h Beltsville, Md. B. japonicum A1017 was obtained light-8 h dark). Forty to 50 nodules (> 1 mmφ) were from Dr. K. Minamisawa, Tohoku University, Japan. detached from 6 plants for each treatment. The nod- Spontaneous mutants of these strains resistant to 100 ules were surface sterilized with 5% calcium hypo- mg L-1 nalidixic acid and 100 mg L-1 rifampicin chlorite, and then nodule occupancy was determined (strain 138NR) or to 100 mg L-1 erythromycin and 10 by examining the antibiotic resistance of bacteria in mg L-1 tetracycline (strains 110ET and A1017ET) each nodule as described elsewhere (Ozawa 1988).
were isolated in our laboratory. The parental strains N50 value, which is the inoculum ratio of a test and the mutants were maintained on yeast extract- strain to the competitor strain when it yields 50% mannitol (YEM) agar slants and stored at 4°C (Mac- nodule occupancy, was used to express the competi- Gregor and Alexander 1971). Prior to inoculation to tive nodulation ability of each strain. N50 was calcu- soybean, the strains were grown on a shaker (110 lated in the same way for estimating ED50 (effective rpm) for 5 days at 30°C in 5 mL of YEM broth in a dose 50%) by the Probit method (Finney 1952).
glass tube (18 x 200 mm). One hundred μL of the culture was transferred to another YEM broth (50 2.3. Bacterial adhesion to plant root and vermiculite
mL) in a 200-mL flask, and grown for 7 days under Strains 110ET and A1017ET were examined on com- the same conditions as above. In case of examining petitive adhesion to soybean root surface against the effect of Cd, 10 μL of 10 mM CdCl2 was added to strain 138NR. Each strain was cultured for 7 days in the 50 mL of YEM just before the transfer. Concen- YEM broth with or without 2 μM CdCl2 at 30°C, and tration of CdCl2 in YEM was 2 μM. then washed with distilled water as described above. We evaluated here competitive nodulation abili- Each of the washed cell suspension of former strains ties of the strains USDA110, A1017, 110ET, and was mixed with the washed 138NR cell suspension. A1017ET against the strain 138NR as a competitor. Cell density of each strain in the mixture was adjust- Each culture of the test strains was diluted 10- to 103- ed to 1 x 108 cfu mL-1. Three days old soybean seed- Inhibitory effect of cadmium on competitive nodulation ability of Bradyrhizobium japonicum lings, which were aseptically germinated on paper in where ODa and ODb are optical density at 660 nm of a petri dish at 30°C in the dark, were cut with a razor the lower phase after and before agitation, respec- blade at position about 3 cm above the root tip. Sets of 4 excised root segments were soaked in 50 mL of the mixed cell suspension to remove loosely bound 2.5. Determination of EPS
bacteria. After 90 min of incubation at 30°C, the root Bacteria grown for 7 days at 30°C in 50 mL of YEM segments were washed 4 times with distilled water. broth were harvested by centrifugation at 14,000 rpm Bacterial cells attached to plant roots were removed for 3 min, and the precipitate was washed 4 times by ultrasonic cleaning with the Bronson Sonifier with 50 mM potassium phosphate (pH 6.5). First two model 250 (output level 4, 2 min on ice). Bacteria re- washes were collected, and assayed for hexose by leased from the root segments were counted by the the phenol-H2SO4 method (Dubois et al. 1956).
dilution plate method with YEM agar containing the antibiotics.
2.6. Polyacrylamide gel electrophoresis (PAGE) of LPS
Adhesion ability of strains USDA110, A1017, The pelleted cells from 1 mL culture were washed 110ET, A1017ET, and 138NR to vermiculite used in once with 10 mM Tris-HCl (pH 7.5), 5 mM MgCl2, this study for soybean cultivation was evaluated in 10 mM 2-mercaptoethanol, and then suspended in 130 the following way using vermiculite columns. Opti- μL of sodium dodecyl sulfate (SDS) sample buffer (12 cal density (OD) at 660 nm of cell suspension of each mM Tris [pH 6.8], 4% SDS, 10% glcerol, 0.2% bro- strain obtained as described above was adjusted to mophenol blue, 4% 2-mercaptoethanol). The total 0.2, and then the cell suspension was passed through SDS extracts of bacterial strains were heated at 100°C a column of vermiculite (1.5 x 20 cm) at room tem- for 10 min followed by digestion with proteinase K perature. We used in this study the rate of decrease in (150 μg mL-1) at 60°C for 60 min (Cava et al. 1989). OD of cell suspension during the filtration as an indi- LPS in the crude preparations was electrophoresed cation of adhesion ability of the bacterial strain to through 4% stacking and 12.5% separating acryl- amide gel with 0.4% SDS. The gels were silver stained by the method of Tsai and Frasch (1982).
2.4. Cell surface hydrophobicity
. Results and Discussion
Cell surface hydrophobicity of bacterial cells was measured by the method of Rosenberg et al. (1980). 3.1. Cd tolerance
This method is based on partitioning behavior of We assessed minimum inhibitory concentration bacterial cells between oil and water. Bacteria with (MIC) of Cd against the Bradyrhizobium strains in more hydrophobic surfaces tend to be attracted to the this study by using YEM broth containing CdCl2. oil-water phase boundary (Marshal 1976). The rhizo- The strains from stock cultures were incubated for 7 bial cells grown in the YEM broth for 7 days were days at 30°C in 5 mL of YEM supplemented with 0 washed twice with distilled water or 50 mM potassi- － 100 μM CdCl2 on a shaker (110 rpm), then OD at um phosphate (pH 6.5). To 3.0 mL of the washed 660 nm of each culture was measured. MIC of Cd cell suspension of approximately 108 cfu mL-1 in a was 30 μM against every strains tested. Two μM test tube (12 x 100 mm) with a screw cap, 0.3 mL n- CdCl2 had no effect on growth rate of the strains octane as the water-insoluble phase was added, used in this study (data not shown).
and then the tube was vertically agitated on a shaker (300 rpm, amplitude of 5 cm) for 2 min. After standing 3.2. Effect of Cd on CNA
for 2 min, optical density of the lower aqueous A rhizobial strain competes with other strains to in- phase was determined at 660 nm. fect their host plant and multiply in nodules formed Cell surface hydrophobicity index (CHI) was in the host roots. Figure 1 shows nodule-occupancy expressed as the following equation: rates of the test strains USDA110, A1017, 110ET, and CHI = (ODb － ODa)/ODb A1017ET in competition with strain 138NR after the Takashi OZAWA and Kenji IJIRI co-inoculation to soybean. In case of 110ET and Effect of cultivation with CdCl2 on the competitive nodulation index (N50) of B. japonicum strains. A1017ET, proportion of the nodules containing both mU and mL are upper and lower limit of confidence interval at 95 %, respectively.
the test strain and the reference strain 138NR ranged from 0 to 14%. The double-infected nodules were (mU, mL) excluded to calculate nodule occupancy rates in Fig. (0.20, 2.9 × 10-8) 1. The rate of nodules occupied by a strain increased (560, 3.1 × 10-3) with the increase in the ratio of the strain to the com- (18, 3.1 × 10-2) petitor strain in an inoculum. 3.3. Effect of Cd on adhesion of bacterial cells to host
roots and vermiculite
As a rhizobial cell density on the host root surface is one of the major factors for competitive nodule for- mation by the strain (Dowling and Broughton 1986), higher ability to adhere to host roots and lower abili- ty to adhere to vermiculite would both give higher CNA to the rhizobial strain. Cultivation of A1017ET strain in YEM broth with 2 μM CdCl2 caused signif- log of test strain/138NR ratio in inoculum
icantly higher ratio of A1017ET cells to 138NR cells that adhered to soybean root segments, but in case of Fig. 1. Nodule occupancy rates of B. japonicum strains USDA110,
110ET, A1017, and A1017ET in competition with B. 110ET there was no effect (Table 2). Adhesion ability japonicum strain 138NR.
of 110ET to vermiculite particles declined by culti-vating them in YEM with 2 μM CdCl2 (Table 3).
From the dose-response curve, we can obtain a 50 value, the proportion of a strain to its competing Effect of cultivation with CdCl2 on the adhesion of B. japonicum strains to soybean root segments.
strain in an inoculum when 50% of nodules are Root-attached cells (cfu per root segment) formed by the strain (Table 1). A strain showing smaller value of N 50 is more competitive for nodule 3.6×105±3.0×104 2.4×105 ± 3.8×104 formation. Growing in YEM supplemented with 2.4×105±4.5×104 1.6×105 ± 1.5×104 2μM CdCl2 significantly decreased the CNA of 110ET/138NR ratio strains 110ET and A1017ET. Strains USDA110 and A1017, showing more competitive than the antibiotic resistant mutants, also showed a decreasing tendency A1017/138NR ratio in CNA by growing with 2 μM CdCl2, though statis-tical significance was low.
As both strains USDA110 and A1017 have no Table 3. Effect of cultivation with CdCl2 on the adhesion of B.
japonicum strains to vermiculite. Values with asterisk antibiotic marker, we were not able to distinguish are significantly different (n = 3, p < 0.05) from the nodules occupied by single strain from double-in- control value.
Adhesion rate (%) of cells cultivated with fected nodules. We did not use strains USDA110 and 0 μM Cd (Control) A1017 in the following experiments.
Inhibitory effect of cadmium on competitive nodulation ability of Bradyrhizobium japonicum Table 4. Effect of cultivation with CdCl2 on the cell surface
Araujo et al. (1994) reported that a Tn-5 mutant hydrophobicity of B. japonicum strains. of Rhizobium etli produced EPS that was indistin- Values (mean±S.E., n = 3) with asterisk are significantly different from the control (0 μM CdCl2) value of each guishable from that of its parent. However, the mu- washed cells (p < 0.05).
tant was highly hydrophobic, while greatly reduced CHI of rhizobium cultivated with in competitive nodulation. Bittinger et al. (1997) es- tablished that a single gene homologous to a family of transcriptional regulators affected both competi-tive nodulation and cell surface hydrophobicity of the strain. The results in this study also indicate the correlation between cell surface hydrophobicity and # DW: distilled water, PB: 50 mM K-phosphate (pH 6.5) Other strains tested did not show significant change LPS is one of the major components of the outer in the adhesion ability after growing in YEM with membrane of Gram-negative bacteria, and is a het- CdCl2. These results indicate that 2μM CdCl2 in a erogeneous molecule that varies O-antigen length growth medium does not cause a decrease in rhizobi- (Jann et al. 1975). LPS that has shorter length of the al cell density on the host root surface to bring about polysaccharide chain would make more hydrophobic the decrease of CNA.
of bacterial cell surface (Nikaido 1976). Growing in the YEM broth supplemented with 3.4. Effect of Cd on cell surface hydrophobicity
2μM CdCl2 resulted in an alteration in LPS-band pat- Adhesiveness of bacterial cells to solid surface terns of SDS-PAGE (Fig. 3). LPSs of small molecu- would be affected by the cell surface hydrophobicity lar size appeared in the cells of 110ET and A1017ET (van Loosdrecht et al. 1987). Vesper et al. (1987) re- when they were grown in YEM broth with 2μM ported that transposon mutants of a Bradyrhizobium CdCl2. This explains the increase in cell surface hy- japonicum strain reduced both attachment to soybean drophobicity by growing with 2μM CdCl2 as de- roots and cell hydrophobicity. Surface hydrophobici- scribed above.
ty of strains 110 ET and A1017ET in this study that were washed with distilled water significantly de-creased by cultivation in YEM broth with 2 μM CdCl2 (Table 4). On the contrary, the Cd-cultured cells of both strains washed with phosphate buffer increased hydrophobicity.
3.5. Effect of Cd on production of EPS and LPS
B. japonicum 110ET and A1017ET produced a large
Fig. 2. Production of gel-like substances by B. japonicum strains.
amount of extracellular substances when grown in Strain 110ET (a, b) and A1017ET (c, d) were grown for 7 days in YEM broth supplemented with 0μM (a, c) and 2 μM CdCl2 (Fig. 2). When both 2μM CdCl2 (b, d), then 1.5 mL of the culture was strains were grown in YEM supplemented with 2 μM centrifuged (10,000 rpm, 3 min).
CdCl2, significantly larger amount of EPS was de-tected in the culture than in YEM without CdCl Effect of cultivation with CdCl2 on the production of EPS by B. japonicum strains. ble 5). As EPS is a hydrophilic molecule, the large Values are means ± S.E. (n =3). Values with asterisk of amount production of EPS would be responsible for each strain are significantly different (p < 0.05) from the the decrease in cell surface hydrophobicity as shown EPS (μg Glc/mL) in Table 4. When the EPSs loosely bound to cell sur- face were washed out with 50 mM phosphate, the 0μM (Control) Cd-cultured cells showed larger values of CHI than Takashi OZAWA and Kenji IJIRI nodulation of legumes. Annu. Rev. Microbiol., 40,
131-157.
Dubois M, Gilles KA, Hamilton, JK, Rebers, PA, and Smith, F 1956: Colorimetric method for determination
of sugars and related substances. Anal. Chem., 28,
350-356.
Finney DJ 1952: Probit Analysis, Cambridge University.
Gibson AH 1980: Methods for legumes in glasshouses and controlled environment cabinets. In Methods for Evaluating Biological Nitrogen Fixation, Ed, Bergersen FJ, p. 139-184, John Wiley & Sons, New Fig. 3. SDS-PAGE of LPS of B. japonicum 110ET and A1017ET.
Both strains were grown in the YEM supplemented with Jann B, Reske K, Jann, K. 1975: Heterogeneity of 0μM (Cd -) or 2μM (Cd +) CdCl2. The total SDS-extracts from 10 mg (dry weight) of cells of each strain were lipopolysaccharides. Analysis of polysaccharide chain loaded on each well, and electrophoresed. Slanting arrows lengths by sodium dodecylsulfate-polyacrylamide gel indicate characteristically appeared bands in the cells grown with Cd.
electrophoresis. Eur. J. Biochem., 60, 239-245.
Kinkle BK, Angle JS, Keyser HH 1987: Long-term effects LPS molecules have been thought to play vari- of metal-rich sewage sludge application on soil ous roles in establishing the symbiosis between rhi- populations of Bradyrhizobium japonicum. Appl. zobia and their host plants (Stacey et al. 1982). O-an- Environ. Microbiol., 53, 315-319.
tigen polysaccharide portion of a LPS molecule MacGregor AN, Alexander M 1971: Formation of tumor- would be responsible for successful infection of the like structures on legume roots by Rhizobium. J. host plant (Cava et al. 1989). Noel et al. (1986) re- Bacteriol. 105, 728-732.
ported that reduction or defect of the polysaccharide Marshal KC 1976: Interfaces in Microbial Ecology, chains resulted in abortion of infection thread devel- Harvard University Press, Cambridge.
opment. The observations in the present study pro- Nikaido H 1976: Outer membrane of Salmonella vide a hypothesis that a very small amount (2μM) of typhimurium transmembrane diffusion of some Cd causes the decrease in CNA of B. japonicum by hydrophobic substances. Biochim. Bioph. Acta, 433,
disturbing the synthesis of polysaccharide chains in Nikaido H 1994: Prevention of drug access to bacterial targets: Permeability barriers and active efflux. Science, 264, 382-388.
Araujo RS, Robleto EA, Handelsman J 1994: A Noel KD, Vandenbosch KA, Kulpaca B 1986: Mutations in hydrophobic mutant of Rhizobium etli altered in Rhizobium phaseoli that lead to arrested development nodulation competitiveness and growth in the of infection threads. J. Bacteriol. 168, 1392-1401.
rhizosphere. Appl. Environ. Microbiol., 60, 1430-1436.
Ozawa T 1988: Competitive nodulation ability of Bittinger MA, Milner JL, Saville BJ, Handelsman J 1997: Bradyrhizobium japonicum strains incubated in soil. rosR, a determinant of nodulation competitiveness in Soil Biol. Biochem., 20, 315-318.
Rhizobium etli. Mol. Plant Microbe Interact., 10, 180-
Ozawa T, Matsui E, Kimura Y 2000: Effect of the application of rice straw and (NH4)2SO4 to a paddy Cava JR, Elias PM, Turowski DA, Noel KD 1989: field on the competitive nodulation ability of Rhizobium leguminosarum CFN42 genetic regions indigenous Bradyrhizobium strains. Sci. Rep. Coll. encoding lipopolysaccharide structures essential for Agric. Osaka Pref. Univ., 52, 7-11.
complete nodule development on bean plants. J. Ozawa T, Ogata H, Doi R, Komai Y 1992: Isolation of Bacteriol., 171, 8-15.
transposon Tn5-induced hydrophobic mutants of a Dowling DN, Broughton, WJ 1986: Competition for Bradyrhizobium japonicum strain with improved Inhibitory effect of cadmium on competitive nodulation ability of Bradyrhizobium japonicum competitive nodulation abilities. Soil Sci. Plant Nutr., 6, e26771.
38, 545-552.
Stacey G, Paau AS, Noel KD, Maier RJ, Silver LE, Brill Ozawa T, Tokuda S, Komai Y 1991: Correlation between WJ. 1982: Mutants of Rhizobium japonicum defective competitive nodulation ability and cell surface in nodulation. Arch. Microbiol. 132, 219-224.
hydrophobicity of Bradyrhizobium japonicum. Triplett EW 1990: The molecular genetics of nodulation Bulletin of Japanese Society of Microbial Ecology, 6,
competitiveness in Rhizobium and Bradyrhizobium. Mol. Plant-Microbe Interactions, 3, 199-206.
Rosenberg M, Gutnick D, Rosenberg E 1980: Adherence of Tsai CM, Frasch CE 1982: A sensitive silver strain for bacteria to hydrocarbons: A simple method for detecting lipopolysaccharides in polyacrylamide gels. measuring cell-surface hydrophobicity. FEMS Anal. Biochem. 119, 115-119.
Microbiol. Lett., 9, 29-33.
van Loosdrecht MCM, Lyklema J, Norde W, Schraa G, Sanitá di Toppi L, Gabbrielli R 1999: Response to Zehnder AJB 1987: The role of bacterial cell wall cadmium in higher plants. Environ. Exp. Bot., 41, 105-
hydrophobicity in adhesion. Appl. Environ. Microbiol. Schue M, Fekete A, Ortet P, Brutesco C, Heulin T, Vesper SJ, Malik NSA, Bauer WD 1987: Transposon Schmitt-Kopplin P, Achouak W, Santaella C 2011: mutants of Bradyrhizobium japonicum altered in Modulation of metabolism and switching to biofilm attachment to host roots. Appl. Environ. Microbiol. prevail over exopolysaccharide production in the response of Rhizobium alamii to cadmium. PLoS One,

Source: http://www.it-hiroshima.ac.jp/institution/library/pdf/research49_027-033.pdf