PHYSIOLOGICAL, SEROLOGICAL, AND PLASMID CHARACTERIZATION

of their relative tolerance to pH and NaCl. Previous studies of rhizobia (20,43,53,106) showed that fast-growers were relatively more alkali-tolerant and acid-sensitive, than the slow-growers. Also, Graham and Parker (53) showed that among rhizobia, tolerance to 2% NaCl was restricted to the fast-growing R. meliloti. The responses of rhizobia in litmus milk (44,53) and their relative resistance to antibiotics (128) have also been used to separate rhizobia into two growth-rate groups. Fast-growers tend to produce acid and peptonization reactions in litmus milk and are generally sensitive to antibiotics. The slow-growers on the other hand, do not possess these characteristics.

C. Deoxyribonucleic acid base ratios and homology

Rhizobia can also be separated into several groups based on DNA base ratios and nucleic acid hybridizations. While DNA base ratios, mole percent G+C, for Rhizobium are relatively broad ranged [values ranging from 59-66% (32)], they do indicate relationships between strains. De Ley and Rassel (32) indicated that the peritrichously flagellated fast-growing rhizobia, R. leguminosarum, R. phaseoli, and R. trifolii, tend to have low %G+C values, ranging from 59-63%, while the polar to subpolarly flagellated slow-growers, R. japonicum, R. lupini, and some of the cowpea miscellany, have relatively high %G+C values, 63-65%. Heberlein et al. (57) reported that with the exception of Agrobacterium pseudotsugae and R. japonicum, the agrobacteria and rhizobia which they examined had about the same %G+C content, about 59-63%. Rhizobium japonicum on the other hand, had G+C values of about 64-65%. However, Elkan (40) examined 25 strains of R. japonicum and found their mole percentage G+C to be relatively homogeneous, 61-64%. Elkan (40) also indicated that although %G+C values are significantly different between some strains of R.japonicum, these values fall within the range of other fast- and slow-growing rhizobia and thus, are of limited taxonomic value.

DNA-DNA homologies have been used by several investigators (57,47,64) to study the degree of genetic relatedness between strains and isolates of Rhizobium. The homology studies of Gibbons and Gregory (47) and Heberlein et al. (57) have indicated that there is a sharp line of demarcation between the fast-growing species (R. leguminosarum, R. trifolii, R. phaseoli, and R. meliloti) and the slow-growing ones (R. japonicum and R. lupini). Hollis et al. (64) have recently shown that the slow-growing strains of R. japonicum which they examined could be separated into at least three distinct DNA homology groups. These authors also indicated that there was little homology between DNA from several R. japonicum strains and R. leguminosarum, R. meliloti, R. phaseoli, R. trifolii, and Agrobacterium tumefaciens. However, there was substantial homology between the slow-growing R. lupini and R.japonicum. Similarly, Crowe et al. (29) showed that the 113 fast-growing and 9 slow-growing strains of rhizobia which they examined could be placed into 4 DNA homology groups, and that the fast-growers were clearly separated from the slow-growers. Interestingly, these authors also indicated that there was little homology between fast- and slow-growing lotus rhizobia. However, the two groups were more closely related to each other than to the other fast- and slow-growing rhizobia which were examined. Thus, fast- and slow-growing rhizobia can also be separated into two groups on the basis of DNA-DNA homologies.

In addition to the above mentioned characteristics, fast- and slow-growing rhizobia have also been separated into fast- and slow-growing groups on the basis of polyacrylamide gel electrophoresis of cell proteins (119) and nodule-bacteroid inclusion bodies (28).

D. Serological relationships

Serological techniques have been used routinely in the study of rhizobia: 1, to obtain information about their antigenic composition (52,68); 2, for strain identification (13,78,104,123); 3, to investigate the serological relatedness of strains and species of Rhizobium (36,37,46,69,71,136,137) and 4, for ecological studies (13,14,15,16,104,85,121,122,123,). Three techniques in particular have found wide acceptance in serological investigations of the Rhizobiaceae: agglutination, immunodiffusion (Ouchterlony Gel Diffusion), and immunofluorescence.

Agglutination techniques were the first of the serological methods applied to the study of rhizobia (38). In general, early serological studies were concerned mainly with the relationships between serological groupings and host-specificity. Results of most studies indicated that rhizobia are a serologically heterogeneous group of organisms. Stevens (127) and Wright (140) found that different strains isolated from the same host-plant species could be serologically unrelated. Hughes and Vincent (67) found that even strains isolated from different nodules on the same plant could be unrelated serologically. However, in their study, Bushnell and Sarles (23) found that some strains from different cross-inoculation groups were serologically related. Kleczkowski and Thornton (86) indicated that the ability of a rhizobial strain to nodulate a particular host-plant is not necessarily related to it serological characteristics as detected by agglutinations. Bushnell and Sarles (23) examined rhizobia from the soybean, cowpea, and lupin cross-inoculation groups and found no correlation between the ability of certain strains in one group to cross-inoculate another, and their ability to cross-agglutinate. In addition, Stevens (127) and Bushnell and Sarles (23) indicated that due to their serologic heterogeneity, all strains within the same Rhizobium species cannot be identified by agglutination reactions with a limited number of antisera.

Despite the inability to show correlations between host-specificity and serological groupings, agglutination reactions have been used to assess the serological relatedness of strains and species of Rhizobium. Kleczkowski and Thornton (86), using whole-cell antisera against 4 strains of R.trifolii and 2 strains of R. leguminosarum, examined the agglutination cross-reactions of antigens from 161 strains of R. trifolii, 29 R. leguminosarum, 5 strains of R. meliloti and R. lupini, and 13 non-Rhizobium soil isolates. Results of their study indicated that while no cross-reactions occurred outside of the clover and pea groups and with the 13 soil isolates, some cross-reactions occurred between the two groups. And, while some antisera were quite specific, others were relatively non-specific.

In their studies on the serological relationships of 25 strains of the slow-growing R. japonicum, Koontz and Faber (88) identified 6 somatic serogroups using agglutination adsorption reactions. Wright et al. (140) similarly found 6 serogroups among the R. japonicum they examined, however, these authors did not differentiate between somatic and flagellar antigens. Date and Decker (30) analyzed 28 strains of R. japonicum and found 17 somatic serogroups on the basis of cross-reactions and agglutination cross-adsorptions.

Graham (49) tested 113 strains of Rhizobium for agglutination by whole cell antisera produced against 58 strains of Rhizobium and 16 Agrobacterium strains. The results of his study of whole- and somatic-cell antigens indicated that the rhizobia could be separated into three serologically distinct groups: 1, R. trifolii, R. leguminosarum, and R. phaseoli; 2, R. japonicum, R. lupini, and Rhizobium spp. of the cowpea miscellany; and 3, R. meliloti. While there were no cross-reactions between the groups, there were cross-reactions within the groups. Graham (49) also indicated that strains of R. meliloti showed some serological relatedness to Agrobacterium tumefaciens and A. radiobacter and that agglutination cross-reactions were greater with whole-cell antigens than with somatic-cell antigens.

Immunodiffusion techniques, specifically Ouchterlony double-diffusion, have also been used extensively to investigate the serological relationships between strains and species of Rhizobium (36,37,51,69,70,71,125). The technique relies on the separation of soluble, diffusible antigens through an agar-gel matrix. Relationships between various antigens and antisera are determined by examining the nature of the interaction at the junction of precipitin bands from various wells. Gel diffusion methods have been used in the study of rhizobia because they permit the rapid enumeration of soluble antigens, the techniques are relatively simple, and they can be used to study serological relationships of strains at the single antigen level. Dudman (37) has indicated that while agglutination reactions can be used to separate rhizobia into serological groups, agglutination techniques lack the resolving power of immunodiffusion in distinguishing between antigenically identical and closely related, but not identical strains.

Dudman (36) was the first investigator to use immunodiffusion techniques to study the serological relatedness of strains and species of Rhizobium. In his study of the extracellular soluble antigens of 2 strains of R. meliloti, Dudman (36) found that the two strains examined shared all antigens accept several fast-moving ones. He proposed that since the strains did not cross-agglutinate, that these strain-specific antigens could be used for identification purposes.

Using gel immunodiffusion, Skrdleta (125) divided the 11 slow-growing R. japonicum which he examined into two basic somatic serogroups. While he detected the same serogroups using agglutination reactions, he found that immunodiffusion allowed him to show serological relationships between strains that were not agglutinated by the same antisera. Skrdleta (125) also indicated that the somatic antigens were more specific than flagellar ones in differentiating individual strains.

Dudman (37) in his study of seven strains of R. japonicum found that pretreatment of antigens (by boiling or ultrasonic disruption) was required for the proper immunodiffusion analysis of these slow-growing strains. Gibbins (46) found that while ultrasonic disruption prevented the formation of precipitin bands in immunodiffusion reactions, band formation could be restored by heating the sonicated antigen preparations.

While the use of somatic antigens (heat-stable antigens) have been more specific than flagellar ones (heat-labile) in differentiating individual strains of Rhizobium, internal antigens have also been reported (135) to provide some insight into the serological relatedness of fast- and slow-growing rhizobia. Using whole-cell antisera against three strains of R. japonicum, Vincent et al. (137) studied the internal antigens of sixty-nine strains of Rhizobium and 5 Agrobacterium strains.

Immunodiffusion reactions indicated that at least one common antigen was present in 13 strains of R. japonicum, 4 strains of R. lupini, and 4 strains of the slow-growing cowpea and lotus rhizobia. Their results also indicated that the forty-six fast-growing rhizobia examined were readily distinguished from the slow-growing strains and that the 5 strains of agrobacteria grouped with the fast-growing rhizobia. More recently, Pankhurst (110) studied the immunodiffusion cross-reactions of somatic and interal antigens from 62 fast- and slow-growing strains of lotus rhizobia. Results of his study indicated that while the fast- and slow-growers shared no common somatic antigens, internal antigens were shared by all of the fastgrowing strains, and with seven exceptions, by all of the slow-growing strains.

The fluorescent antibody technique is the method of choice for the direct examination and identification of strains of rhizobia in culture and nodules (13,14,123) and for the enumeration of specific strains directly in soi1 (14,85,122). The major advantages of immunofluorescence over other techniques is that only small amounts of antigen and antibody are needed (121), the procedures are relatively rapid, and its the only technique readily applicable to the study of rhizobia in situ (16).

Vincent (135) in his recent review of the literature has pointed out that when serological (137) and other taxonomic evidence [see (73) and (79)] are taken together, clear relationships among rhizobia can be recognized. That is; 1, there is a closer relationship between the fast-growing species of Rhizobium and Agrobacterium than there is between the fast- and slow-growing groups of rhizobia; 2, R. trifolii and R. phaseoli should be made separate biovars of the species R. leguminosarum; 3, R. meliloti is so different from other species of Rhizobium, that it requires its own species status; 4, among the slow-growers, the slow-growing soybean rhizobia should remain as a separate species group and be included in the new genus Bradyrhizobium; and 5, the fast- and slow-growing lotus rhizobia fall within the genera Rhizobium and Bradyrhizobium, respectively.

E. Current Rhizobium taxonomy

Due to the large number of differences existing between the fast- and slow-growing rhizobia, and to inadequacies in the cross-inoculation-plant-infection group scheme, a new classification scheme for rhizobia has been proposed [see Jordan (79) and Jarvis (73)]. This scheme (since adopted by the International Subcommittee on Agrobacterium and Rhizobium) divides rhizobia taxonomically into fast- and slow-growing groups. The first group consists of the fast-growing rhizobia. Those organisms previously designated as R. leguminosarum, R. trifolii, and R. phaseoli will be combined as one species, R. leguminosarum (Table I-2), comprising three biovars (trifolii, phaseoli, and viceae). Rhizobium meliloti [which differs significantly from other rhizobia (134)], is kept as a separate species group. The slow-growing rhizobia, R. japonicum, were transferred to a separate genus, Bradyrhizobium. Only one species is present in this genus, R. japonicum. Other slow-growing rhizobia are to be referred to as Bradyrhizobium spp., with the name of the designated-nodulated plant following in parentheses. In the remainder of this dissertation, those slow-growing organisms previously referred to as Rhizobium japonicum will be referred to as Bradyrhizobium japonicum and R.japonicum will be used only to refer to the fast-growing soybean-rhizobia from China. It should be noted that the slow-growing rhizobia formerly referred to as R. lupini are not included as a separate species in the new genus Bradyrhizobium, since their only major distinguishing characteristic is their nodulation affinity for Ornithopus and Lupinus. Subsequent to the adoption of this new scheme (which will be appearing in the forthcoming edition of Bergey’s Manual of Determinative Bacteriology), Jarvis et al. (73) have proposed that another species group be included in the new genus Rhizobium. This species, Rhizobium loti refers to the heterogeneous group of fast-growing rhizobia which effectively nodulate Lotus corniculatus (birds-foot treefoil), Lupinus densiflorus (lupines) and Anthyllis vulneraria (kidney vetch). It should be noted however, that some strains of the lotus rhizobia ineffectively nodulate a great variety of platns. In addition, the slow-growing lotus rhizobia will be included in the new genus Bradyrhizobium.

Introduction

Species within the genus Rhizobium have been divided into two groups (80,133) depending on their growth rate and effect on the pH of yeast extract-mannitol (YEM) medium under standard laboratory conditions. The fast-growing rhizobia have mean generation times of between two and four hours and produce a net decrease in the pH of YEM culture medium, while those referred to as slow-growing have mean generation times of six hours and longer and do not lower the pH of the medium (133).

The typical slow-growers that form nodules on the roots of soybeans (Glycine max) have in the past been referred to as Rhizobium japonicum (80). Recently, these rhizobia have been re-classified in a new genus, Bradyrhizobium, on the basis of their slow growth rate and other characteristics (79) to distinguish them from fast-growing, acid-producing root-nodule bacteria which now comprise the genus Rhizobium [see (73)].

Recently, Keyser et al (84) reported the isolation of fast-growing soybean-rhizobia from root nodules and soil collected in the provinces of Shansi, Honan, Shandong, and Shanghai in the People’s Republic of China (PRC). The isolates have been reported (84) to have mean doubling times between 2 and 4 hours and lower the pH of YEM culture medium (final pH ranging from 4.7 to 6.7). All strains were reported to form effective nitrogen-fixing nodules on wild perennial soybeans (Glycine soja) and on an unbred soybean cultivar from China (cv Peking), but formed ineffective symbioses with most commercial cultivars (84).

This report describes the taxonomic investigation of the fast-growing, acid-producing PRC isolates. Several diagnostic microbiological, physiological, and biochemical tests were performed in order to determine the degree of relatedness of the fast-growing soybean rhizobia to each other, to the "typical" slow-growing B. Japonicum, and to other fast-growing rhizobia. It was assumed, that in order to make any meaningful assessment of this newly described group of organisms, that their relative taxonomic relationship to other rhizobia needed to be determined.

Materials and Methods

A. Bacterial Strains and Growth Conditions

The fast-growing soybean-rhizobia, USDA 191, 192, 193, 194, 201, 205, 206, 208, 214, 217, and 257 were isolated from soil or nodules collected in the People’s Republic of China [Keyser, et al. (84)]. The slow-growing Bradyrhizobium Japonicum, Chinese strains PRC-005, 74, 113-2, 121-6, 2031, and B15 were obtained from T. S. Hu, Institute of Soils and Fertilizers, Chinese Academy of Agricultural Sciences, Beijing, People’s Republic of China. Bradyrhizobium japonicum strains USDA 6, 31, 34, 74, 94, 110, 122, 123, 136, 138, 142, and Y1, Yla, Y2, Y2a, Y3, K2, K2a, S1, S1a, were from the USDA Culture Collection, Beltsville, Maryland. Rhizobium lupini CC814s, NZP 2021, NZP 2238, SU 343, and NZP 2037, R. leguminosarum Nitragin 92A3, R. phaseoli NZP 5097, NZP 5253, and NZP 5260, R. trifolii NZP 560, and WU 95, and R. meliloti NZP 4013 were obtained from R. M. Greenwood, Department of Scientific and Industrial Research, Palmerston North, New Zealand. Rhizobium leguminosarum HI 5-0 was isolated in Hawaii [May and Bohlool (104)]. Rhizobium leguminosarum 6015(pJB5JI) was obtained from P. Hirsch, Max Planck Institute, Koln, FRG; Rhizobium phaseoli Bel 7.1 from E. L. Schmidt, University of Minnesota, Minneapolis; Rhizobium sp. (Leucaena) Tal-82, from the Niftal Project, Paia, Hawaii; and Rhizobium sp. (Leucaena) UMKL 19 and R. leguminosarum PRE from W. J. Broughton, Max Planck Institute, Koln, FRG. Agrobacterium tumefaciens 79 and 101 and Rhizobium sp. (Sesbania) 3F4a4 were from USDA Beltsville, Maryland. Rhizobium meliloti L530 was obtained from B. Rolfe, Australia National University, Canberra City, Australia. All Rhizobium cultures were maintained on yeast extract-mannitol (YEM) agar slants of the following composition in g/1: yeast extract, 1.0; mannitol, 10.0; K₂HPO₄·3H₂0, 0.65; MgSO₄·7H₂0, 0.2; NaCl, 0.1; pH 6.9 (132). Agar slants used for the maintenance of fast-growing rhizobia contained 0.05% CaCO₃. Agrobacterium cultures were maintained on Nutrient agar (Difco) slants. All cultures were incubated at 28-30⁰C, subcultured at least once every month and stored at 4⁰C.

B. Staining, Morphology, and Cultural Characteristics

Cultures were examined for cell morphology and Gram reaction after 3 d of growth in YEM liquid medium. Colony morphology was examined on cultures grown for 6 d on YEM agar containing brom-thymol-blue, BTB,(0.25 mg/1). Motility was estimated from both YEM liquid and agar cultures and on B5 agar medium (45). Cell size determinations were performed using a calibrated ocular micrometer. Fast- and slow-growing PRC isolates were identified using strain specific fluorescent antibodies prepared according to Schmidt et al. (123).

C. Biochemical Tests

Tolerance to pH extremes was determined by inoculating 10⁷ cells/ml from exponentially growing YEM liquid cultures into tubes containing 10 ml of YEM liquid medium which were adjusted to pH 4.5, 9.0, and 9.5. Tubes were incubated at 30⁰C for 14 d and scored for growth. Tests were performed in triplicate.

Tolerance to sodium chloride was determined on YEM agar plates containing 2.0% NaCl. Plates were spread with 10⁸ cells, and growth was scored after 14 d of incubation at 30⁰C. Tests were done in triplicate.

For growth reactions in litmus milk (Difco), tubes (10 ml/tube) were incubated in quadruplicate for 6 weeks at 30⁰C and were examined for pH changes, reduction of litmus, and peptonization (serum zone formation).

Production of 3-ketolactose was determined according to Bernaerts and De Ley (9). Agrobacterium tumefaciens was used as a positive control for this test.

For gelatinase activity, exponential phase cultures from YEM liquid medium were swabbed onto the surface of tryptone yeast extract (TY) agar plates (62) containing 0.4% (w/v) gelatin (Difco). Plates were incubated at 28⁰C for 7 d. A positive reaction was indicated by a clearing zone surrounding the growth of the organism. If no clearing zone was detected, the plates were flooded with a 10% solution of trichloroacetic acid and re-examined.

The pH reactions of isolates on agar plates were determined using YEM medium containing 0.25 mg/l bromthymol blue.

Urease activity was determined on urea agar slants (26) incubated for 7 d at 28⁰C.

Citrate utilization was determined on the solid medium of Koser (89). Plates were spread with 10⁸ cells, incubated at 30⁰C for 14 d, and examined for growth.

Penicillinase (B-lactamase) was detected by the method of Foley and Perret (42), oxidase by the method of Kovaks (90), and catalase by the method of Graham and Parker (53).

Hydrogen sulfide production was determined on agar slants [Hunter and Crecelius (72)]. Slants were inoculated and examined for H₂S after 14 d at 28⁰C.

Nitrate reduction was tested as described in the Manual of Methods for General Bacteriology (126), in the same medium used by Graham and Parker (53) and in Difco nitrate broth.

For carbohydrate utilization, the basal medium used (Bis) was that of Bishop et al. (11) with different carbohydrates substituted for mannitol and 0.6 g/l KNO₃ used as the nitrogen source. The medium was solidified with purified agar (Difco). All carbohydrates, with the exception of dextrin and starch, were filter sterilized (0.4 um Nuclepore) before addition to gooled, molten, agar medium. Dextrin and starch were added to the medium before autoclaving at 100⁰C for 15 min. Each carbohydrate was added to a final concentration of 1.0% (w/v). Inocula were prepared by removing cells from YEM agar slants (with a cotton swab) and suspending the cells to approximately 1X10⁷ cells/ml in sterile distilled water. A ten-fold dilution of each cell suspension was added to the wells of a multiple inoculator plate [Josey, et al. (81)] and inoculated onto the surface of carbohydrate containing agar plates. Bishop’s agar plates without carbohydrate served as controls. Duplicate plates of each carbohydrate were incubated at 28⁰C for 7 d and scored for growth.

D. Generation times in culture media and sterile soil

Growth and pH responses were determined in TY, Bis, YEM and PA (62) liquid media. All media were adjusted to pH 6.9 prior to autoclaving. Fast-growing isolates were pre-grown (in the medium into which they would be subsequently inoculated) for 3 d, while the slow-growers were pre-grown for 7 d. Inocula were added to an initial density of 10⁶cells/ml into 50 ml of the respective medium in 125 ml “side-arm” Erlenmeyer flasks. Flasks were agitated at 150 revolutions per minute (RPM) at 28⁰C in a water-bath shaker. Cell growth was monitored using a Klett-Summerson Photoelectric-Colorimeter (equipped with a # 66 red filter) and pH determined after four days using an Orion Research (model 501) pH meter and a glass combination-electrode.

Generation times in sterile soil were determined using 10 g samples of air-dried Kula loam soil (Typic Eutrandept, pH 6.5) in 70 ml screw-cap test tubes. Soil tubes were autoclaved for 45 min at 121⁰C, on two successive days, and inoculated with stationary phase cultures of B. japonicum strains USDA 110 and 136, or fast-growing PRC isolates USDA 193 and 205. Fifteen tubes of each organism were inoculated to obtain initial cell numbers of about 5x10⁵ cells/gm and a soil moisture tension of about 60% of water-holding capacity. Cell growth was monitored by plate counts on YEM agar using destructive samplings.

E. Intrinsic antibiotic resistances

Resistance to low levels of antibiotics was determined using the method of Josey et al. (81). Inocula were prepared as outlined above for carbohydrates. Freshly prepared, filter sterilized (0.4 um Nuclepore) solutions of antibiotics were added to cooled, molten TY agar to give the following concentrations (ug/ml): chloramphenicol 12.0, 25.0; kanamycin sulfate 10.0; naladixic acid 10.0; neomycin 2.5, polymyxin B sulfate 20.0, rifampicin 1.0, 6.0, streptomycin sulfate 2.5, 10.0; tetracycline-HCl 4.0; and vancomycin 1.5, 5.0. Controls consisted of TY agar plates without antibiotics. Isolates showing growth were scored as positive. Duplicate plates of each antibiotic were incubated (in the dark) at 28⁰C for 7 d and scored for growth.

F. Enzyme Assays

6-phosphogluconate dehydrogenase (EC 1.1.1.43) activity was determined in cultures grown for 72 h at 26⁰C in yeast extract-glucose medium (82). Cells were centrifuged at 6,000 X g for 10 min at 4^oC and washed twice in 0.05 M sodium phosphate buffer, pH 7.4. Cell pellets were resuspended in the same buffer, containing 2X10^-4 M 2-mercaptoethanol (100) to a final concentration of 1.0 g wet-weight cells/1.5 ml buffer and disrupted by two passages through a French pressure cell at 15,000 pounds per square inch (PSI) and cell debris removed by centrifugation at 14,000 X g for 30 min at 4^oC. The clear supernate was stored at -20⁰C until use. The activity of NADP-linked 6-phosphogluconate dehydrogenase was measured by following the reduction of NADP according to the method of Martinez-De Drets and Arias (101). Apparent endogenous enzyme activity was subtracted from the results. Specific activities were expressed in nanomoles of NADPH formed / min / mg of protein at 25⁰C. Protein was determined by the method of Lowry, et al. (98) using bovine serum albumin as the standard.

B-galactosidase (EC 3.2.1.23) activity was determined in cultures grown for 3 d at 26⁰C in TY medium containing 0.5% (w/v) lactose. Cells were centrifuged and washed as above and resuspended in 0.05 M sodium phosphate buffer, pH 7.2, to a final concentration of 1.0 g wet-weight / 3.0 ml buffer and broken by passage through a French pressure cell at 15,000 PSI. Cell debris was removed as before and supernatant fractions stored at 20⁰C until use. Enzyme activity was measured by following the appearance of a colored product (o-nitrophenol, ONP) at 420 nm (126). The incubation mixture (4.8 ml) contained: 2.7 ml enzyme reaction buffer [0.1 M sodium phosphate buffer (pH 7.0), 1X10^-3 M MgSO₄·7H₂0, 2X10^-4 M MnSO₄, and 0.05 M 2- mercaptoethanol]; 1.8 ml ONPG solution [0.1 M sodium phosphate buffer (pH 7.0) and 1.3X10^-2 M o-nitrophenyl-B-D-galactopyranoside (Sigma Chemical Co. St. Louis, MO.)]; and 0.3 ml of cell-free extract. Enzyme assays were done at 37⁰C and the reactions stopped by the addition of 1.3 ml of reaction stop buffer (8.0 M urea and 1.0 M Na₂CO₃, pH 12.0). Enzyme activities were expressed in micromoles of ONP produced per min per mg of protein at 37⁰C. Corrections were made for absorbance values obtained in controls without substrate.

G. Ethanol utilization

The basal medium of Bishop (Bis) (11), with or without added mannitol, and with 0.6 g/l KNO₃ as the nitrogen source, was used in all ethanol experiments. Inocula for all studies were prepared by gently washing the cells from YEM agar slants into 50 ml of Bis medium without any carbon source. All cultures were “starved” by incubation overnight at 28^oC prior to inoculation. To determine whether strains could utilize ethanol as the sole source of carbon and energy, starved cells were inoculated into 50 ml of Bis (initial concentration approximately 10⁶ cells/ml) supplemented with 0.1, 0.25, 1.0, 2.0, 3.0, or 4.0% (v/v) ethanol. Cultures were examined after 7 d of incubation at 28⁰C and scored for growth. Bishop’s medium without any carbon source served as control.

For growth yield and mean generation time determinations when growing on ethanol, starved USDA 191 cells were inoculated to approximately 10⁶ cells/ml into 50 ml of Bis supplemented with 0.1, 0.2, 0.4, or 1.0% ethanol and incubated at 28^oC. Final, total cell numbers were determined using a Petroff-Hausser counting chamber. Mean generation times were calculated from the linear portion of growth curves constructed by following absorbance at 600 nm. Growth yield values were determined after 7 d for 0.1 and 0.2% ethanol and after 13 d for 0.4 and 1.0% ethanol. Substrate conversion values were determined by dividing the final dry weight of cells in grams by the number of moles of substrate utilized. The disappearence of ethanol from the growth medium was determined by gas chromatography using a Porpak Q column at 185⁰C with N₂ as the carrier gas.

For growth and survival studies, 50 ml aliquots of Bis containing 0.2% ethanol were inoculated to about 1X10⁶ cells/ml and incubated at 25^oC. Samples from each culture flask were removed at 0, 9, 14, and 18 days and viable cell numbers determined by plate counts on YEM agar.

In experiments designed to determine whether the lag phase of cells growing on limiting concentrations of ethanol were affected by a subsequent addition of ethanol (at the same or at a greater concentration), a starved USDA 191 culture was inoculated to a final concentration of about 10⁶ cells/ml into three, 500 ml side-arm flasks each containing 100 ml of Bis medium. Ethanol was added to two of the flasks to a final concentration of 0.1% (a limiting concentration), while the remaining flask received 1.0% ethanol. Cultures were incubated at 25^oC and growth was monitored using a Klett-Summerson Electric Colorimeter equipped with a # 66 red filter. One of the two flasks which initially contained 0.1% ethanol, received an additional 0.1% ethanol at the beginning of each stationary phase of growth, while to the other flask 1.0% ethanol was added only once. The flask which originally contained 1.0% ethanol, received no further ethanol additions.

To determine if the fast-growing soybean rhizobia were capable of showing diauxy when growing under limiting concentrations of ethanol and mannitol, a starved culture of USDA 191 was inoculated into 75 ml of Bis supplemented with either 0.2% ethanol alone, 0.02% mannitol alone, or 0.02% mannitol plus 0.2% ethanol. Flasks were inoculated, in triplicate, with approximately 10⁶ cells/ml and two ml aliquots removed at different times to monitor growth (absorbance at 600 nm), ethanol utilization by quantitative gas chromatography (see above), and mannitol utilization by the periodate-3-methyl-2-benzothiazolinone hydrazone method of Johnson and Sieburth (75). For mannitol determinations, periodate-digested samples and controls (samples without prior periodate digestion), were analyzed in triplicate. All values are the averages of three replicates.

In experiments designed to determine whether the inoculum size affected the length of the observed lag phase, 125 ml side-arm flasks, containing 50 ml of Bis supplemented with 0.4% ethanol, were inoculated, in triplicate, with USDA 191 to an initial cell concentration of 1.6X10⁵, 3.2X10⁵, 6.4X10⁵, 1.2X10⁶, or 2.5X10⁶ cells/ml. Initial cell numbers were determined by viable counts on YEM agar and growth monitored spectrophotometrically using as Klett-Summerson PhotoElectric Colorimeter equipped with a # 66 red filter.

Results

Morphological and cultural characteristics. Both the fast- and slow-growing soybean-rhizobia were Gram negative, non-sporeforming rods. The fast-growing soybean isolates tended to be slightly larger than the slow-growers, with their average dimensions being 2 to 4 um by 0.5 to 1 um. Cells from late log phase to stationary phase YEM cultures of fast-growers tended to become enlarged and exhibited marked pleomorphism. Only a few cells (1-5%) of the fast-growing rhizobia were motile in young (1-2 d) YEM cultures whereas the majority of cells (80-90%) of the slow-growing rhizobia were motile. However, when growing on the surface of moist B5 agar medium, a larger percentage (up to 25%) of the fast-growing soybean rhizobia were found to be motile. On YEM agar plates containing brom thymol blue, both fastand slow-growing rhizobia formed circular, convex, entire colonies. After 6-7 d of growth, the fast-growers formed colonies with sizes between 1.0 and 5.0 mm in diameter and produced an acid-reaction, while the slow-growers had colony sizes of approximately 0.5 to 1.0 mm in diameter and produced an alkaline-reaction. Several of the fast-growing soybean strains had a dry-crusty (calcified) appearance when grown on YEM agar containing CaCO₃. In contrast to a large number of other fast-growing species of Rhizobium, the fast-growing soybean rhizobia do not produce much extracellular polysaccharides on YEM agar. Two of the isolates, USDA 191 and 192 produced “watery” colonies. Upon repeated restreaking, some of the fast-growers produced a second colony-type that was somewhat smaller than colonies produced by the parent cultures. These colonial variants were not always stable. Several of the variants that did appear stable were isolated from seven cultures of fast-growers (USDA 191, 192, 193, 208, 214, 217, and 257) and tested for serological affinity (using strain-specific fluorescent antibodies) and for effectiveness on soybeans (Glycine max) cultivar Peking. They were found to be identical to the parental cultures.

Generation times in culture medium and sterile soil

All of the fast-growing soybean rhizobia examined, had mean generation times much less than the “typical” slowgrowers, in the four culture media examined (Table III-1). For the fast- and slow-growers, the most rapid growth was attained in YEM medium, with an average generation time (MGT) for the fast- and slow-growers of 3.6 and 9.3 h, respectively. Results in Table III-1 also indicate that while TY medium is acceptable for the growth of fast-growing rhizobia (MGT ranging from 3.1-4.2 h), most of the slow-growers did not grow very well in this medium. The fast-growing PRC rhizobia acidified only the YEM culture medium (average pH 5.6). In the other three growth media, the pH either remained the same or increased substantially. The slow-growers on the other hand, raised the pH of all four culture media.

graph of the time to reach 15 Klett units vs Log inoculum size is linear (Figure III-4) with an R² = 0.885.

To determine whether the fast-growing soybean rhizobia would show diauxic growth when grown under limiting concentrations of ethanol and mannitol, USDA 191 was inoculated into 0.2 % ethanol, 0.02% mannitol, and 0.2% ethanol plus 0.02% mannitol. Results in Figure III-5A show that when cells were grown in limiting concentrations of mannitol alone, there was no appreciable lag phase and cells reached stationary phase about 2 d after inoculation. However, when cells were grown in ethanol alone (also at a limiting concentration), the growth rate was slower than that seen in mannitol, and it took about 5 d to reach stationary phase. Cells grown under limiting conditions of both mannitol and ethanol together, initially had a growth rate similar to cells growing in mannitol alone, however, when mannitol became limited, after about 2 d, the growth rate declined to about that seen for cells growing on ethanol alone. The results in Figure III-5B indicate that ethanol did not begin to disappear appreciably from the growth medium untill after 3 d of growth and that ethanol utilization was about the same for cells grown in ethanol alone or in ethanol plus mannitol. Similarly, mannitol utilization was about the same for cells grown in mannitol only, or cells grown in mannitol plus ethanol (Figure III-5C).

Discussion

Several diagnostic tests were performed to compare fast-and slow-growing rhizobia that nodulate soybeans. The results indicate that the fast-growers are quite different biochemically from the slow-growers. While the fast-growers share host specificity with the slow-growers [both nodulate soybeans as well as other hosts (84)], they appear more similar to other fast-growing Rhizobium species in their microbiological, physiological and biochemical characteristics. Some characteristics, however, are shared between the two groups of organisms.

All of the fast- and slow-growing soybean-rhizobia examined were typically positive for catalase, oxidase, urease, penicillinase, and nitrate reductase. Both groups were unable to utilize citrate as a sole source of carbon, produce H₂S, or produce 3-ketolactose. Graham and Parker (53) also found that most Rhizobium japonicum strains they tested possessed several of these attributes. They reported that the production of H₂S and utilization of citrate was restricted to some isolates of R. meliloti. Bernaerts and De Ley (9) found that the production of 3-ketolactose from lactose is limited to Agrobacterium species, a genus closely related to Rhizobium. Graham and Parker (53) found that the production of penicillinase seemed to be restricted mainly to the slow-growing rhizobia, R. japonicum, R. lupini and Rhizobium spp. of the “cowpea miscellany” group. However, these authors indicated that some of the R. leguminosarum strains also possess this attribute. Microbiologically, the fast and slow-growers appear to be related in that they produce similar colony types (although they do differ in size) and tend to produce morphological variants. However, as with other bacteria, the production of morphological variants is not uncommon in Rhizobium (58,134)

The PRC isolates were truly fast-growing in sterile soil and in all of the laboratory culture media examined. In most instances, when growing in laboratory media, the mean generation times of the fast-growers were 2 to 3 times that of the slow-growers. While the slow-growers grew from 1.5 to 2 times faster in sterile soil than in media, growth rates of the fast-growers in sterile soil was in the same range as that found in laboratory media. It is interesting to note that the fast-growing soybean isolates only reduced the pH of YEM medium, while in all other media, the pH was raised. This pH reduction-property can not only be due to the presence of mannitol, since Bis also contained the same mannitol concentration. The reason(s) why Bis medium became alkaline following growth of the fast-growers, is not understood. However, it is important to remember, that for taxonomic considerations, determinations as to whether rhizobia are alkaline producing or acid-producing should only be ascertained in YEM medium.

The results of the pH and salt tolerance tests place the fast-growing soybean-rhizobia with other fast-growing rhizobia. Previous studies of rhizobia (20,43,53), in pure culture and in soil, showed that the fast-growers were relatively more alkali-tolerant and acid-sensitive, than the slow-growers. Also, Graham and Parker (53) showed that among rhizobia, tolerance to 2% NaCl was restricted to the fast-growing R. meliloti. In addition, the responses of the fast-growing soybean-rhizobia in litmus milk were more typical of fast-growing Rhizobium, e.g. R. trifolii, R. leguminosarum, and R. meliloti (44,53).

Fast- and slow-growers also appeared to be separated on the basis of their utilization of carbohydrates and sensitivities to antibiotics. Graham and Parker (53) and Fred et al. (44) showed that fast-growing rhizobia tend to use a wider variety of carbohydrates than the slow-growers. This is clearly evident in Table III-5, with the fast-growers using the majority of the carbohydrates examined and the slow-growers relatively few. It should be noted that the types of carbohydrates utilized also varies amongst the soybean-rhizobia. As was pointed out by Glenn and Dilworth (48), slow-growing rhizobia tend to lack both uptake systems and catabolic enzymes for disaccharides. My results are in agreement with these authors in that the disaccharides cellobiose, lactose, maltose, trehalose, and sucrose and the trisaccharide raffinose clearly separated the fast- from slow-growing rhizobia. The results in Table III-7 show that only the fast-growing soybean rhizobia have appreciable B-galactosidase activities, similar to that of the other fastgrowers examined (data not shown). These results agree with those of Glenn and Dilworth (48) who found that among the rhizobia they examined, the inability of slow-growers to utilize several disaccharides was due to the lack of disaccharide uptake systems and the hydrolytic enzymes to cleave the disaccharides. The results in Table III-6 show that the fast-growing soybean rhizobia were sensitive to more antibiotics than the slow-growers. It should be noted that the fast- and slow-growing rhizobia could not be placed into any resistance groupings using the antibiotics and concentrations employed. Vincent (134) has indicated that while strains within a species of Rhizobium show differential sensitivity to antibiotics, the slow-growing species of Rhizobium are generally more resistant than the fast-growing ones.

Consistent with the division of rhizobia into fast- and slow-growing groups on the basis of carbohydrates utilized, is their division based on the presence and absence of enzymes of the pentose phosphate pathway. As was pointed out by Martinez-De Drets and Arias (101,102), although both fast- and slow-growing rhizobia have NAD-linked 6PGD activity, only the fast-growers have NADP-linked 6PGD activity. The results in Table III-7 demonstrate that the fast-growing soybean-rhizobia have levels of NADP-6PGD activity comparable to those of the other fast-growing rhizobia. These results are consistent with those of Martinez-De Drets and Arias (100,101) and Keele et al. (82) who found high levels of NADP-6PGD activity for only fast-growing species of Rhizobium.

Although mannitol and sucrose are the preferred carbon sources for fast-growing rhizobia (44) and arabinose for the slow-growers (1), both groups are capable of utilizing many different types of carbon compounds. While it has been shown that in B. japonicum bacteroids (the morphologic state of rhizobia within soybean root-nodules), aldehydes and alcohols can support acetylene reduction and oxygen consumption (112), and that soybean nodules contain acetaldehyde and ethanol (129), there have been no reports of ethanol utilization by free-living rhizobia. Results presented in this study indicate that both the fast- and slow-growing soybean rhizobia are capable of utilizing ethanol as the sole source of carbon and energy.

Although fast- and slow-growing soybean rhizobia could not be separated on the basis of ethanol utilization, there were some clear differences in the extent in which they used ethanol. While the fast-growers routinely grew to final cell densities of 1X10⁸-1X10⁹ on 0.2% ethanol, the slow-growers never reached densities greater than about 1X10⁷ cells/ml on any concentration of ethanol. It is apparent from results (Table III-10 and Figure III-1) showing decreased growth rates in higher ethanol concentrations, that ethanol most likely is toxic to the cells. The mean generation time of USDA 191 increased nearly 2-fold, when the ethanol concentration was raised from 0.1% to 1.0%. It is interesting to note (Figure III-4) that whether the cells were originally started in 1.0% ethanol or had their ethanol concentration raised from 0.1 to 1.0%, that their growth rates remained about the same. In the flask receiving multiple additions of 0.1% ethanol, growth rates also remained relatively constant following each ethanol addition. The results presented in Table III-10 and Figure III-1 indicate that growth is substrate limited with ethanol concentrations less than 0.4%. At ethanol concentrations of 0.1-0.2%, final cell numbers and dry weight are proportional to ethanol concentration. However, at an ethanol concentration of 0.4% and greater, growth did not appear substrate limited and growth yield values were not proportional to ethanol concentration. Results in Figure III-1 show that the maximum rate of ethanol utilization (when 191 was growing in 0.1 and 0.2% ethanol) occurred after 3 to 4 d of growth. In 0.4 or 1.0% ethanol, the rate of utilization was considerably less than that seen in lower concentrations.

Three of the fast-growing soybean rhizobia (USDA 192, 194, and 205) were also examined for their ability to use other low carbon-number compounds. While all three of the isolates utilized acetate as a sole source of carbon, only USDA 192 could use 1-propanol (data not presented). None of the organisms examined could utilize methanol, formate, or 1-butanol as a sole source of carbon (data not shown). While all of the fast-growing soybean rhizobia utilized ethanol as a sole source of carbon, I could not detect any NAD- or NADP- linked alcohol dehydrogenase (ADH) activity in cellfree extracts of the organisms (data not shown). Rigaud and Trinchant (118) have demonstrated a soluble alcohol dehydrogenase in R. meliloti. However, they did not examine whether cells were capable of growth in ethanol. Similarly, De Vries et al. (34) and Tajima and La Rue (129) demonstrated the presence of an NAD-linked ADH in pea and soybean root nodule tissue, respectively. The inabilty to detect NAD or NADP-linked ADH in cell-free extracts of the fast-growing soybean rhizobia may indicate that either the ADH of these organisms requires a different coenzyme, or that the enzyme may not be present in cell-free extracts (i.e., is membrane bound), or that the enzyme was present at undetectable levels in the cells. Since malate synthetase was detected in cell-free extracts of USDA 205 (data not shown), it is assumed that they at least have some enzymes of the glyoxylate cycle. With this cycle present, the organisms have a means for replenishing C-4 intermediates of the TCA cycle from acetyl CoA derived from acetate, or ethanol. While Johnson et al. (74) detected malate synthetase activity in culture-grown B. japonicum cells, these authors failed to detect any isocitrate lyase activity in these isolates. However, isocitrate lyase was detected in culture-grown R. meliloti, R. leguminosarum, R. trifolii, and R. phaseoli cells (64) and in bacteroids from senescent soybean nodules (131). Interestingly, the organisms Johnson et al. (74) examined could not utilize acetate as a sole source of carbon. Thus, it is apparent that more research is needed on the mechanism(s) by which fast- and slow-growing rhizobia are able to utilize ethanol as a source of carbon.

In summary, while the fast-growing soybean-rhizobia share symbiotic host-specificity with the typical slow-growers, they appear more closely related, on a biochemical basis, to other fast-growing species of Rhizobium. The typical soybean-rhizobia are now classified as Bradyrhizobium japonicum (79). Most of the evidence presented in this thesis indicates that the newly isolated fast-growers do not fit into this species. Therefore, the taxonomic position of these new isolates must logically be in the new genus Rhizobium.

temperature for 20 min, the lysate was adjusted to pH 8.5 with 2 M Tris-HCl (Sigma) and transferred to a 40 ml polycarbonate centrifuge tube. Cold (4^oC) 5 M NaCl was added to a final concentration of 1 M and the contents of the tube mixed by gentle inversion. The mixture was incubated at 4^oC for 4 h and the SDS/NaCl precipitate removed by centrifugation at 10,000 X g for 20 min at 4^oC. The supernatant was transferred to a 40 ml centrifuge tube and polyethylene glycol (PEG) 6000 (Sigma) [50% (w/v) in TEN] was added to a final concentration of 10% (v/v). The tube was mixed by inversion and incubated overnight at 4^oC and plasmid DNA collected by centrifugation at 7,000 X g for 15 min at 4^oC. The supernatant was gently decanted and the pellet was allowed to dissolve in 0.5 ml of TEN buffer. Plasmid samples were stored at 4^oC for up to 1 week.

For preparative plasmid extractions, all isolation steps were scaled up 10-fold and the plasmids were further purified by CsCl-ethidium bromide (EtBr) equilibrium density gradient centrifugation. Crude plasmid DNA, 4.5 ml, was added to 20 ml of CsCl-saturated TEN buffer in a 40 ml centrifuge tube, the solution gently mixed by inversion of the tube, and centrifuged at 10,000 X g for 20 min at 4^oC. The resultant PEG/SDS precipitate was removed from the upper portion of the solution and ethidium bromide (Sigma) (10 mg/ml in sterile distilled water) was added to a final concentration of 0.31 mg/ml. The refractive index of the solution was adjusted to 1.3925+0.001 and transferred to 13 ml nitrocellulose ultracentrifuge tubes (Beckman). The DNA preparation was centrifuged, 2 times, at 36,000 RPM and 17⁰C for 48 hours, in a Type 40 (Beckman) rotor (25). Ethidium bromide and CsCl were removed from the plasmid preparation as is outlined by Hirsch et al. (62) except that CsCl-saturated n-butanol was used as the ethidium bromide extractant.

For plasmid screening, samples were mixed with tracking dye [50%(w/v) glycerol, 0.125% (w/v) bromphenol blue, and 50 mM Na₂-EDTA, pH 8.0] and plasmids were resolved by electrophoresis on horizontal 0.7% agarose (Biorad, Richmond, CA; Standard Low) gels (18 by 13 by 0.6 em) at 60 mA for 6 h at 4⁰C with Tris-borate buffer (62). Gels were stained for 20 min in Tris-borate buffer containing 5 ug/ml ethidium bromide and bands were visualized by reflective, short-wave, UV light and photographed using Kodak Pan-X film and a yellow (Hoya, K2) filter.

C. Restriction endonuclease analysis

For restriction enzyme analysis, 60 ul of purified Plasmid DNA was digested with either BamHI, EcoRI, or HindIII (Bethesda Research Labs, Inc.) for 1 h at 37^oC according to the suppliers directions. Following digestion, samples were heated at 65^oC for 3 min to inactivate endonucleases, and 15 ul of tracking dye was added. Digested samples were electrophoresed on 0.7% agarose gels at 50 mA for 16 h at 4^o with Tris-acetate buffer (65). Gels were stained for 20 min in Tris-acetate containing 5 ug/ml ethidium bromide and bands were visualized and photographed as before. For molecular weight estimations, the electrophoretic mobilities of plasmid bands were compared to the reference strain R.leguminosarum 6015 (pJB5JI) (62,114) and the BamHI, EcoRI and HindIII digests of lambda DNA (31).

D. Acridine orange and heat-curing of plasmid DNA

Plasmid curing was done by the acridine orange procedure of Zurkowski et al. (141) or by the heat-treatment curing procedure of Zurkowski and Lorkiewicz (142). For acridine orange curing, overnight tryptone yeast extract (TY) (62) cultures were diluted to approximately 10⁴ cells/ml in YM medium (141) containing 1,3,7,15 or 20 ug/ml acridine orange (Sigma Chemical Co.,St. Louis, Mo.). Cultures were incubated at 25^oC in the dark for 6 d. After incubation, cultures were diluted and spread-plated onto YEM agar plates and the resulting colonies which formed were streaked for purification two consecutive times on the same medium. For heat curing, overnight TY cultures were inoculated to approximately 10⁶ cells/ml into PA medium. Cultures were incubated at 37, 40, and 42^oC and transferred at weekly intervals. After heat-treatment, cultures were spread-plated on YEM agar plates and isolates purified as is outlined above.

E. Bacterial matings

Crosses were done according to the membrane-filter method of Buchanan-Wollaston et al. (22) from overnight TY cultures. Aliquots from serial dilutions of the bacterial mixtures, harvested from the membrane filters (with aid of a magnetic stir bar), were spread-plated onto the appropriate selective media. To select for R. japonicum and R. trifolii transconjugants receiving plasmid pJB5JI (kan^r), and against the auxotrophic (Phe, Trp) donor 6015 (pJB5JI), the minimal Y-medium (3) supplemented with 50 or 100ug/ml kanamycin was used. In back-crosses to the Nod^- mutant, 6015 (Rif^r,Str^r), the same medium was supplemented with kanamycin (50 ug/ml), rifampiein (20 ug/ml), streptomycin (100 ug/ml), phenylalanine (50 ug/ml), and tryptophan (50 ug/ml). This medium was called Ysupp. In crosses between R.trifolii (pJB5JI) and a Nod^- USDA 205 (Str^r, Chlr^r) mutant, the medium used was Y-medium supplemented with kanamycin (100 ug/ ml), streptomycin (200 ug/ml), and chloramphenicol (10 ug/ml). All mixtures were also spread-plated onto nonselective TY medium to obtain total viable counts.

F. Construction of antibiotic resistant mutants

Spontaneous antibiotic resistant mutants were obtained by incubating cells, obtained from the surface of YEM agar slants, in 100 ml of TY broth medium supplemented with various concentrations and types of antibiotics.

G. Assessment of culture purity

Culture purity was ascertained for all donors, recipients, and transconjugants in the following manner: each culture was streaked two consecutive times on the appropriate selective medium and isolated colonies transferred to TY liquid medium. After 2 d of growth, cultures were restreaked on plates of selective media. Colonies were transferred to slants of TY or TY supplemented with 50 ug/ml kanamyein, as appropriate.

Each pure culture thus obtained was further identified by immunofluorescence (123) or immunodiffusion (132) using strain specific antiserum prepared according to Schmidt et al. (123).

H. Plant infection assays

Nodulation tests for soybeans and peas were performed in modified Leonard jar assemblies (96) consisting of a 250 ml wide-mouth Erlenmeyer flask with a 13 mm hole in the bottom. The flask was glued (silicone seal) onto the cover (also having a 13 mm hole) of a 500 ml short, wide-mouth screw-cap bottle, which served as the nutrient reservoir. An absorbant cotton wick was connected between the two reservoirs to facilitate transfer of nutrient solution from the lower reservoir into the plant growth vessel above. The growth vessel was filled with a Vermiculite:Perlite mixture (3:1) and the nutrient reservoir with quarter-strength Hoagland's plant nutrient solution (63). The modified flask assembly was capped with aluminum foil and autoclaved before use. Clover nodulation was tested in screw-cap tubes (2.5 by 20 cm) containing 25 ml of quarter-strength Hoagland's solution (63) with 1.0% agar. The tubes were slanted after sterilization. Peas (Pisum sativa var Wisconsin Perfection) and soybean (Glycine max vars Peking, Chippew a 64, and Lee) seeds were surface sterilized by immersion in a 4.0% (w/v) calcium hypochlorite solution for 20 min followed by exhaustive washings in sterile distilled water. Three sterile seeds of the same variety were aseptically transferred to each plant growth vessel and allowed to germinate and grow until they reached the aluminum foil cap. Plants were selected for uniform size and two of the three seedlings were aseptically removed from each vessel. Plants were inoculated with one ml aliquots of 2-3 day-old YEM cultures and the vessels topped-off with approximately 2 cm of sterilized Perlite and 2-3 cm of sterilized paraffin-coated sand (1.0 ml of paraffin:chloroform (1:100) per 10 gm silica sand). Clover seeds (Trifolium repens var ladino) were surface sterilized by immersion in a 2.0% solution of sodium hypochlorite for 10 min followed by exhaustive washings in sterile distilled water. Seeds were aseptically transferred to Petri dishes (150 by 25 cm) containing 1.0% water-agar and germinated at 25^oC in the dark. Seedlings were transferred to Hoagland's-agar slants and inoculated with 0.1 ml of 2-3 day-old YEM liquid cultures.

Plant vessels were covered with aluminum foil to exclude light from the root zone and incubated at 24+2^oC in a plant growth chamber with an average light intensity of 250 uEinsteins/m²/sec and a day length of 16 h. All plant tests were done in triplicate. After four weeks of growth (unless specified), plants were examined for the presence of nodules, for color, and for nitrogenase activity by the acetylene reduction method (56).

I. Isolation and identification of nodule occupants

For the recovery of isolates from nodules, root sections (with attached nodules) were excised from the plants and surface sterilized in a solution consisting of 75% ethanol and 8% H₂0₂. After exhaustive washings in sterile distilled water, nodules were macerated in 10% glycerol (peas and clover) or in distilled water (soybeans) and streaked onto YEM agar containing 0.25 mg/1 brom thymol blue. Macerates were also spread onto slides and nodule occupants examined by direct immunofluorescence using strain specific antibodies (123).

Results

Plasmid profiles

All of the ten fast-growing PRC isolates examined contained from 1 to 4 plasmids of approximately 100 Mdal and greater. Figure IV-1 shows the relative electrophoretic mobilities of plasmids from USDA isolates 217, 208, 206, 193, 191, 201, 192, 214, 205 and 194 (Lanes: A, D, E, F, G, H, J, K, L and M, respectively). The reference plasmids from R. leguminosarum 6015 (pJB5JI) appear in Fig. IV-1, Lanes C and I. All of the fast-growing PRC isolates with the exception of USDA 194 and USDA 191 shared a plasmid of similar size (approximately 200x10⁶ Mdal). Isolates 217, 191, 201, 192, 205, and 194 each had 2 to 3 demonstratable plasmids (Fig. IV-1, Lanes A, G, H, J, L and M, respectively). One isolate, USDA 206, contained 4 plasmids (Fig. IV-1, Lane E), while the other PRC isolates (Fig. IV-1, Lanes D, F and K) contained one plasmid. Isolate USDA 191 (Fig. IV-1, Lane G) contained two plasmids, one of which exhibited the highest electrophoretic mobility (lowest molecular weight) of all the PRC isolates examined. No plasmids could be detected in the slow-growing R. japonicum isolate USDA 110 (Fig. IV-1. Lane N). However,

ability to nodulate clover after they were subsequently isolated from pea nodules. However, these transconjugants only produced an ineffective symbiosis on clover.

Restoration of nodulating ability

To determine whether pea nodulating ability could be transferred to a fast-growing PRC isolate which has lost its own nodulation plasmid, an acridine-orange-cured, nodulation-deficient USDA 205 mutant [2051A03 (see above)] was used in crosses with R. leguminosarum 6015(pJB5JI) and R. trifolii 403-42 [0403 Chl^r, Str^r(pJB5JI)]. Despite repeated conjugation attempts, with either donor, kanamycin resistant 2051A03 transconjugants could not be isolated (data not shown).

Discussion

In this report, I have demonstrated the presence of from 1 to 4 plasmids in ten fast-growing strains of soybean-rhizobia. The recent results of Masterson et al. (103) have indicated that several of the fast-growing isolates contain plasmids which hybridize to the structural nifK and D genes of Klebsiella pneumoniae. However, these authors did not show the location of nodulation genes in these isolates. My results suggest that in at least one of these isolates, a large plasmid may also be the location of some of the nodulation genes. All of the acridine orange-generated, nodulation-deficient mutants of USDA 205 had lost their largest plasmid. Zurkowski and Lorkiewicz (143) showed a correlation between the loss of nodulation character and the elimination of a plasmid in Rhizobium trifolii T12 and 24. Morrison, et al. (105) have recently demonstrated that a Lablab purpureus isolate NGR234, which can nodulate some legumes and the nonlegume Parasponia, could be cured of its nodulating ability by treatment at elevated temperatures. In this isolate, the loss of nodulating ability was found to result from the elimination of its largest resident plasmid. In my study the heat-treated USDA 194 mutant, missing a smaller plasmid, was still able to effectively nodulate soybeans. Masterson et al. (103) indicated that in USDA 194, the structural nif genes, were not located on any of the plasmids.

Zurkowski and Lorkiewicz (142) have shown that R. trifolii strains can easily lose their nodulation plasmid at elevated temperatures. The fast-growing soybean rhizobia that I examined did not lose their nodulating ability at restrictive temperatures of 37^oC for up to 7 weeks, or at 40 and 42^oC for 7 d. These isolates seem to be resilient to the heat-curing-treatments used effectively for other species of Rhizobium. However, at levels of acridine orange that allowed slight growth, one isolate was cured of its nodulating plasmid, while in bacteriostatic concentrations, no curing of any plasmids could be demonstrated These results are in agreement with those of Pariiskaya (111) who found that R. meliloti L-1 was only significantly cured of modulating ability in subbaeteriostatic levels of acridine orange. However, Higashi (60) and Zurkowski et al. (141) cured different R. trifolii isolates of their nodulating ability in bacteriocidal concentrations of acridine orange.

The results of this study indicate that plasmid pJB5JI, the pea-modulation, host-range plasmid from R. leguminosarum 6015(pJB5JI) could be transferred to the fastgrowing soybean rhizobia and R.trifolii but was variably expressed in different genetic environments. When the plasmid was transferred to a modulation-deficient isolate of R. trifolii, #2, nitrogen-fixation and modulation genes were expressed, since the resulting transconjugants effectively nodulated peas. Initially, it was thought that plasmid pJB5JI did not restore the clover modulation deficiency present in the #2 transconjugants. However, after recovery from pea nodules, two of the transconjugants gained the ability to nodulate clover. When pJB5JI was transferred to a modulation-eom petent R. trifolii isolate, 0403, the resulting transconjugants were capable of nodulating both clover and peas. However, peas (with the exception of one plant) were only nodulated ineffectively (Nod⁺, Fix^-). It should be noted however, that after isolation from pea and clover nodules, there was a differential expression of nodulation and nitrogen fixation genes among some of the 403 transconjugants. While one out of the six transconjugants from clover nodules had lost the ability to nodulate peas (most likely indicating a lost or deleted plasmid) three gained the ability to effectively nodulate peas. While none of the five 403 transconjugants reisolated from pea nodules lost the ability to nodulate peas or clover, two of the reisolates could now nodulate peas effectively. Thus, those plasmid-borne genes required for an effective symbiosis, were differentially expressed in some of the transconjugants after passage through host plants. The mechanism for this phenomenon is unknown. Beynon et al. (10) noted that while some plasmids controlling host-range specificity may be able to coexist (and behave one way) in culture, they may not be able to do so within the nodules of a given host legume. That is, there may be some “physiological incompatibility” or functional interference between the resident nitrogen fixation and nodulation genes and those specified on plasmid pJB5JI. Likewise, there may be functional interference between the clover nitrogen fixation genes present in R. trifolii 0403, and the pea genes on pJB5JI, which accounts for their lack of expression in 403 transconjugants. Passage through the appropriate host plant may in some way modify this incompatability and allow for the expression of both sets of genes.

Transfer of plasmid pJB5JI to the fast-growing soybean isolate, USDA 201, did not bestow upon the transconjugants the ability to nodulate peas. Thepresence of the plasmid in the transconjugants did not interfere with their ability to nodulate their own host soybeans. However, in some transconjugants, a loss of, or a deletion in, a resident plasmid was evident on agarose gels. This did not, however, affect the nodulating ability of the transconjugants, implying that the lost or deleted fragments may not be involved in modulation. While plasmid pJB5JI was transferable to fast-growing soybean strain USDA 201, other isolates failed to accept the plasmid. One strain, USDA 205 was not able to receive plasmid pJB5JI from R. leguminosarum 6015(pJB5JI) or R. trifolii 0403(pJB5JI) even when cured of its largest plasmid.

Plasmid pJB5JI has been used by Brewin et al. (17, 18) and Hirsch et al. (62) to transfer pea-nodulating ability to non-nodulating mutants of R. leguminosarum. Similarly, Beynon et al. (10) used pJB5JI to transfer pea-nodulating ability to R. phaseoli and Djordjevet et al. (35) used the same plasmid to transfer pea-nodulating ability to R. trifolii. Even though the fast-growing USDA 201 transconJugants that received pJB5JI do not themselves nodulate peas, they are, nevertheless, capable of transferring this property to a non-nodulating strain of R. leguminosarum, 6015. This strain is a non-nodulating derivative of R. leguminosarum strain 300 (77), similar to 6015(pJB5JI), but it is non-nodulating because of a deletion in its largest Plasmid (62). The smallest plasmid which appeared in 201-11 and 201-14 transconjugants, may be due to either the co-transfer of another plasmid from R. leguminosarum 6015(pJB5JI), or a deletion in an incoming or resident plasmid. Since USDA 201 contains a plasmid which has a similar electrophoretic mobility as pJB5JI, deletions in, or recombinational events between, this and other plasmids and the major replicon are very probable. This may explain why transconjugants 201-8 and 201-13 did not show any additional plasm id bands despite their kanamycin resistance. However, the 6015 back-crosses, 6015-11-7 and 6015-14-8, appear to have received pJB5JI from 201-11 and 201-14 intact as judged by the appearence of a new plasmid (Figure IV-3) with an electrophoretic mobility similar to pJB5JI. Thus, the plasmid may have “recombined-out” of 201-11 and 201-14 during conjugation to R. leguminosarum 6015. As was indicated by Keyser et al. (84), the fast-growing soybean rhizobia only form effective symbiosis with Glycine Max var Peking and with Glycine soja, but form ineffective nodules on most New World commercial varieties. The results of this study show that despite the presence of the extra nitrogen fixation genes (present in pJB5JI), USDA 201 transconjugants still form ineffective nodules on the two commercial cultivars of soybeans examined.

In conclusion, the results of this study indicate that: 1, the loss of a large plasmid from the fast-growing R. japonicum leads to loss of nodulating ability. These nodulation-deficient Rhizobium japonicum mutants may prove useful in genetic manipulations of the soybean/Rhizobium symbiotic system; 2, although the pea SYH plasmid, pJB5JI, is transferrable to many different species of Rhizobium, the plasmid is differentially expressed in different genetic backgrounds. Despite the fact that the R. leguminosarum pea host-range plasmid, pJB5JI, can effectively function in closely-related rhizobia, my results suggest that it does not by itself carry all the genetic information necessary for the nodulation of peas in a genetically dissimilar environment. However, the plasmid is maintained intact in the different genetic backgrounds despite its lack of expression; and 3, the host may play some role in modifying the genetic make-up of the Rhizobium in the nodules, since some of the nodule-reisolates had a different pattern of nodulation than the parent cultures.

LITERATURE CITED

1. Allen, E. E., and O. N. Allen. 1950. Biochemical and symbiotic properties of the rhizobia. Bacteriol. Rev. 14:73-30.

2. Balassa, G. 1963. Genetic transformation of Rhizobium: a review of the work of R. Balassa. Bacteriol. Rev. 27:228-242.

3. Beringer, J. E. 1974. R factor transfer in Rhizobium leguminosarum. J. Gen. Microbiol. 84:188-198.

4. Beringer, J. E. and D. A. Hopewood. 1976. Chromosomal recombination and mapping in Rhizobium leguminosarum. J. Gen Microbiol. 98:291-293.

5. Beringer, J. E., A. W. B. Johnston, and B. Wells. 1977. The isolation of conditional ineffective mutants of Rhizobium leguminosarum. J. Gen. Microbiol. 98:339-343.

6. Beringer, J. E., J. L. Beynon, A. V. Buchanan-Wollaston, and A. W. B. Johnston. 1978. Transfer of drug resistance transposon Tn5 to Rhizobium. Nature 276:633-634.

7. Beringer, J. E., S. A. Hoggan, and A. W. B. Johnston. 1978. Linkage mapping in Rhizobium leguminosarum by means of R plasmid-mediated recombination. J. Gen. Microbiol. 104:201-207.

8. Beringer, J. E. 1980. The development of Rhizobium genetics. J. Gen. Microbiol. 116:1-7.

9. Bernaerts, M. J., and J. De Ley. 1963. A biochemical test for crown gall bacteria. Nature. 197:406-407.

10. Beynon, J. L., J. E. Beringer, and A. W. B. Johnston. 1980. Plasmids and bost-range in Rhizobium leguminosarum and Rhizobium phaseoli. J. Gen. Microbiol. 120:421-429.

11. Bishop, P. E., J. G. Guevara, J. A. Engelke, and H. J. Evans. 1976. Relation between glutamine synthetase and nitrogenase activities in the symbiotic association between Rhizobium jpoonicum and Glycine max. Plant Physiol. 57:542-546.

12. Bohlool B. B., and E. L. Schmidt. 1968. Nonspecific staining: its control in immunofluorescence examination of soil. Science. 162:1012-1014.

13. Bohlool, B. B., and E. L. Schmidt. 1970. Immunofluoreseent detection of Rhizobium japonicum in soils. Soil Sci. 110:229-236.

14. Bohlool, B. B., and E. L. Schmidt. 1973. A fluorescent antibody technique for determination of growth rates of bacteria in soil. Bull. Eco. Res. Comm. (Stockholm). 17:336-338.

15. Bohlool, B. B., and E. L. Schmidt. 1973. Persistance and competition aspects of Rhizobium japonicum observed in soil by immunofluorescence microscopy. Soil Sci. Soc. Am. Proc. 37:561-564.

16. Bohlool, B. B., and E. L. Schmidt. 1980. The immunofluorescence approach in microbial ecology, p.203-241. In M. Alexander (ed.), Advances in Microbial Ecology, Vol. 4. Plenum Press, New York.

17. Brewin, K. J., J. E. Beringer, and A. W. B. Johnston. 1980. Plasmid-mediated transfer of host-range specificity between two strains of Rhizobium leguminosarum. J. Gen. Microbiol. 120:413-420.

18. Brewin, K. J., J. E. Beringer, A. V. Buchanan-Wollaston, A. W. B. Johnston, and P. R. Hirsch. 1980. Transfer of symbiotic genes with bacteriocinogenic plasmids in Rhizobium leguminosarum. J. Gen. Microbiol. 116:261-270.

19. Brewin, K. J., T. M. Dejong, D. A. Phillips, and A. W. B. Johnston. 1980. Co-transfer of determinants for hydrogenase activity and nodulation ability in Rhizobium leguminosarum. Nature 288:77-79.

20. Bryan, U. C. 1923. Effects of acid soils on noduleforming bacteria. Soil Sci. 15:37-40.

21. Buchanan-Wollaston, V. 1979. Generalized transduction in Rhizobium leguminosarum. J. Gen. Microbiol. 112:135142.

22. Buchanan-Wollaston, A. V., J. E. Beringer, N. J. Brewin, P. R. Hirsch, and A. W. B. Johnston. 1980. Isolation of symbiotically defective mutants of Rhizobium leguminosarum by insertion of the transposon Tn5 into a transmissible plasmid. Mol. Gen. Genet. 178:185-190.

23. Bushnell, 0. A., and W. B. Sarles. 1939. Investigations upon the antigenic relationships among the root-nodule bacteria of the soybean, cowpea, and lupine cross-inoculation groups. J. Bacteriol. 38:401-410.

24. Caldwell, B. E. and G. Vest. 1968. Nodulation interactions between soybean genotypes and serogroups of Rhizobium japonicum. Crop Sci. 8:680-688.

25. Casse, F., C. Boucher, J. S. Julliot, M. Michel, and J. Denarie. 1979. Identification and characterization of large plasmids in Rhizobium meliloti using agarose gel electrophoresis. J. Gen. Microbiol. 113:229-242.

26. Christensen, W. B. 1946. Urea decomposition as a means of differentiating Proteus and paracolon organisms from each other and from Salmonella and Shigella types. J. Bacteriol. 52:461-466.

27. Cloves, R. C., E. E. M. Moody, and R. H. Pritchard. 1965. The elimination of extrachromosomal elements in thy mineless strains of Escherichia coli K12. Genet. Res., Camb. 6:147.

28. Craig, A. S., R. M. Greenwood, and K. I. Williamson. 1973. Ultrastructural inclusions of rhizobial bacteroid of Lotus nodules and their taxonomic significance. Arch. Mickrobiol. 89:23-32.

29. Crow, V. L., B. D. W. Jarvis, and R.M. Greenwood. 1981. Deoxyribonucleic acid homologies among acid-producing strains of Rhizobium. Int. J. Syst. Bacteriol. 31:152-172.

30. Date, R. A., and A. M. Decker. 1965. Minimal antigenic constitution of 28 strains of Rhizobium japonicum. Can. J. Microbiol. 11:1-8.

31. Davis, R. W., D. Botstein, and J. R. Roth. 1980. A manual for genetic engineering; Advanced bacterial genetics, p. 236. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.

32. De Ley, J., and A. Rassel. 1965. DNA base composition, flagellation, and taxonomy of the genus Rhizobium. J. Gen. Microbiol. 41:85-91.

33. Denarie, J., P. Boistard, F. Casse-Delbert, A. G. Atherly, J. O. Berry, and P. Russell. 1981. Indigenous plasmids of Rhizobium, p. 225-246. In K. L. Giles and A. G. Atherly (ed.), Biology of the Rhizobiaceae, Suppl. 13. Academic Press, New York.

34. De Vries, G. E., P. IN ‘T Veld, and J. W. Kijne. 1980. Production of organic acids in Pisum sativum root nodules as a result of oxygen stress. Plant Sci. Lett. 20:115-123.

35. Djordjevic, M. A., W. Zurkowski, and B. G. Rolfe. 1982. Plasmids and stability of symbiotic properties of Rhizobium trifolii. J. Bacteriol. 151:560-568.

36. Dudman, W. F. 1964. Immunodiffusion analysis of the extracellular soluble antigens of two strains of Rhizobium meliloti. J. Bacteriol. 88:782-794.

37. Dudman, W. F. 1971. Antigenic analysis of Rhizobium japonicum by immunodiffusion. Appl. Microbiol. 21:973-985.

38. Dudman, W. F. 1977. Serological methods and their application to dinitrogen-fixing organisms, p. 487-508. In R. W. F. Hardy and A. H. Gibson (ed.), A Treatise On Dinitrogen Fixation, Section IV. John Wiley and Sons, New York.

39. Dunican, M. J., and A. B. Tierney. 1974. Genetic transfer of nitrogen fixation, from Rhizobium trifolii to Klebsiella aerogenes. Biochem. Biophys. Res. Comm. 57:62-72.

40. Elkan, G. H. 1969. Deoxyribonucleic acid base compostion of isolates of Rhizobium japonicum. Can. J. Microbiol. 15:490-493.

41. Elkan, G. H. 1981. The taxonomy of the Rhizobiaceae, p. 1-14. In K. L. Giles and A. A., Atherly (ed.), Biology of the Rhizobiaceae, Suppl. 13. Academic Press, New York.

42. Foley, J. M., and C. J. Perret. 1962. Screening bacterial colonies for penicillinase production. Nature 195:287-288.

43. Fred, E. B., and A. Davenport. 1918. Influence of reaction on nitrogen-assimilating bacteria. J. Ag. Res. 14:317-336.

44. Fred, E. B., I. L. Baldwin, and E. McCoy. 1932. Root nodule bacteria and leguminous plants. University of Wisconsin Studies in Science, Number 5, Madison.

45. Gamborg, 0. L. 1975. Callus and cell cultures, p. 1-9. In 0. L. Gamborg and L. R. Wetter (ed.), Plant Tissue Culture Methods. National Research Council of Canada, Saskatchewan.

46. Gibbins, L. N. 1967. The preparation of antigens of Rhizobium meliloti by ultrasonic disruption: an anomaly. Can J. Microbiol. 13:1375-1379.

47. Gibbins, A. M., and K. F. Gregory. 1972. Relatedness among Rhizobium and Agrobacterium species determined by three methods of nucleic acid hybridization. J. Bacteriol. 111:129-141.

48. Glenn, A. R., and M. J. Dilworth. 1981. The uptake and hydrolysis of disaccharides by fast- and slow-growing species of Rhizobium. Arch. Microbiol. 129:233-239.

49. Graham, P. H. 1963. Antigenic affinities of the rootnodule bacteria of legumes. Antonie van Leeuwenhoek, J. Microbiol. Serol. 29:281-291.

50. Graham, P. H. 1964. The application of computer techniques to the taxonomy of the root-nodule bacteria of legumes. J. Gen. Microbiol. 35:511-517.

51. Graham, P. H. 1971. Serological studies with Agrobacterium radiobacter, A. tumefaciens, and Rhizobium strains. Arch. Mikrobiol. 78:70-75.

52. Graham, P. H., and M. O’Brien. 1968. Composition of lipopolysaccharides from Rhizobium and Agrobacterium. Antonie van Leeuwenhoek 34:326-330.

53. Graham, P. H., and C. A. Parker. 1964. Diagnostic features in the characterization of the root-nodule bacteria of legumes. Plant Soil 20:383-396.

54. Gross, D. C., A. K. Klucas, and R. V. Klucas. 1979. Plasmids, biological properties and efficacy of nitrogen fixation in Rhizobium japonicum strains indigenous to alkaline soils. J. Gen. Microbiol. 114:257-266.

55. Hadley, H. H., and T. Hymowitz. 1973. Chapter 3. Speciation and cytogenetics, p. 97-116. In B. E. Caldwell (ed.), Soybeans: Improvement, Production, and Uses. Am. Soc. Agron., Madison, Wisconsin.

56. Hardy, R. W. F., R. D. Holsten, E. K. Jackson, and R. C. Burns. 1968. The C₂H₂-C₂H₄ assay for nitrogen fixation; laboratory and field evaluation. Pl. Physiol. 43:1185-1207.

57. Heberlein, G. T., J. De Ley, and R. Tijtgat. 1967. Deoxyribonucleic acid homology and taxonomy of Agrobacterium, Rhizobium, and Chromobaeterium. J. Bacteriol. 94:116-124.

58. Herridge, D. F., and R. J. Roughley. 1975. Variation in colony characteristics and symbiotic effectiveness of Rhizobium. J. Appl. Bact. 38:19-27.

59. Heumann, W. 1968. Conjugation in starforming Rhizobium lupini. Mol. Gen. Genet. 102:132-144.

60. Higashi, S. 1967. Transfer of clover infectivity of Rhizobium trifolii to Rhizobium phaseoli as mediated by an episomic factor. J. Gen. Appl. Microbiol. 13:391403.

61. Hirsch, P. R. 1979. Plasmid-determined bacteriocin production by Rhizobium leguminosarum. J. Gen. Microbiol. 113:219-228.

62. Hirsch, P. R., M. Van Montagu, A. W. B. Johnston, N. J. Brewin, and J. Schell. 1980. Physical identification of bacteriocinogenic, nodulation, and other plasmids in strains of Rhizobium leguminosarum. J. Gen. Microbiol. 120:403-412.

63. Hoagland, D. R., and D. I. Arnon. 1938. The water culture method for growing plants without soil. California A9, Exp. Std. Circular Number 347.

64. Hollis, A. B., W. E. Kloos, and G. H. Elkan. 1981. DNA: DNA hybridization studies of Rhizobium japonicum and

65. Hombrecher, G., N. J. Brewin, and A. W. B. Johnston. 1981. Linkage of genes for nitrogenase and nodulation ability on plasmids in Rhizobium leguminosarum and R. phaseoli. Mol. Gen. Genet. 182:133-136.

66. Hooykaas, P. J. J., A. A. N. van Brussel, H. den Dulk-Ras, G. M. S. van Slogteren, and R. A. Schilperoort. 1981. Sym plasmid of Rhizobium trifolii expressed in different rhizobial species and Agrobacterium tumefaciens. Nature 291:351-353.

67. Hughes, D. Q., and J. M. Vincent. 1942. Serological studies of the root-nodule bacteria. III. Tests of neighboring strains of the same species. Proc. Linnean Soc. N.S. Wales. 67:142-152.

68. Humphrey, B. A., and J. M. Vincent. 1962. Calcium in cell walls of Rhizobium trifolii. J. Gen. Microbiol. 29:557-561.

69. Humphrey, B. A., and J. M. Vincent. 1965. The effect of calcium nutrition on the production of diffusible antigens by Rhizobium trifolii. J. Gen. Microbiol. 41:109-118.

70. Humphrey, B. A., and J. M. Vincent. 1969. The somatic antigens of two strains Rhizobium trifolii. J. Gen. Microbiol. 59:411-425.

71. Humphrey, B. A., and J. M. Vincent. 1975. Specific and shared antigens in strains of Rhizobium meliloti. Microbios. 13:71-76.

72. Hunter, C. A., and H. G. Crecelius. 1938. Detection of hydrogen sulphide in cultures. J. Bacteriol. 35:185-196.

73. Jarvis, B. D. W., C. E. Pankhurst, and J. J. Patel. 1982. Rhizobium loti a new species of legume root nodule bacteria. Int. J. Syst. Bacteriol. 32:378-380.

74. Johnson, G. V., H. J. Evans, and T. Ching. 1966. Enzymes of the glyoxylate cycle in rhizobia and nodules of legumes. Pl. Physiol. 41:1330-1336.

75. Johnson, K. M., and J. McN. Sieburth. 1977. Dissolved carbohydrates in seawater:I, a precise spectrophotometric analysis for monosaccharides. Marine Chem. 5:1-13.

76. Johnston, A. W. B., J. L. Beynon, A. V. Buchanan-Wollaston, S. M. Setchell, P. R. Hirsch., and J. E. Beringer. 1978. High frequency transfer of nodulation ability between strains and species of Rhizobium. Nature 276:634-636.

77. Johnston, A. W. B., G. Hombrecher, A. J. Brewin, and M. C. Cooper. 1982. Two transmissible plasmids in Rhizobium leguminosarum strain 300. J. Gen. Microbiol. 128:85-93

78. Jones, D. G., and P. E. Russel. 1972. The application of immunofluorescence techniques to host plant/nodule bacteria selectivity experiments using Trifolium repens. Soil Biol. Bioehem. 4:277-282.

79. Jordan, D. C. 1982. Transfer of Rhizobium japonicum Buchanan 1980 to Bradyrhizobium gen. nov., a genus of slow-growing root nodule bacteria from leguminous plants. Int. J. Syst. Bacteriol. 32:136-139.

80. Jordan, D. C., and O. N. Allen. 1974. Genus II. Rhizobium, p. 262-264. In R. E. Buchanan and N. E. Gibbons (ed.), Bergey’s Manual of Determinative Bacteriology, 8^th ed. The Williams and Wilkins Co., Baltimore.

81. Josey, D. P., J. L. Beynon, A. W. B. Johnston, and J. E. Beringer. 1979. Strain identification in Rhizobium using intrinsic antibiotic resistance. J. Appl. Bacteriol. 46:343-350.

82. Keele, B. B. Jr., P. B. Hamilton, and G. H. Elkan. 1969. Glucose catabolism in Rhizobium japonicum. J. Bacteriol. 97:1184-1191.

83. Kennedy, C., B. Dreyfus, and J. Brockwell. 1981. Transfer, maintenance and expression of P plasmids in strains of cowpea rhizobia. J. Gen. Microbiol. 125:233-240.

84. Keyser, H. H., B. B. Bohlool, T. S. Hu, and D. F. Weber. 1982. Fast-grwoing rhizobia isolated from root nodules of soybean. Science 215:1631-1632.

85. Kingsley, M. T., and B. B. Bohlool. 1981. Release of Rhizobium spp. from tropical soils and recovery for immunofluorescence enumeration. Appl. Environ. Microbiol. 42:241-248.

86. Kleczkowski, A., and H. G. Thorton. 1944. A serological study root-nodule bacteria from pea and clover inoculation groups. J. Bacteriol. 48:661-672.

87. Kondorosi, A., G. B. Kliss, T. Forrai, B. Vincze, and Z. Banfalvi. 1977. Circular linkage map of the Rhizobium meliloti chromosome. Nature 268:525-527.

88. Koontz, F. P., and J. E. Faber. 1961. Somatic antigens of Rhizobium japonicum. Soil Sci. 91:228-232.

89. Koser, S. A. 1923. Utilisation of the salts of organic acids by the colon-aerogenes group. J. Bacteriol. 8:493

90. Kovaks, N. 1956. Identification of Pseudomonas pyocyanea by the oxidase reaction. Nature 128:703.

91. Kowalski, M. 1967. Transduction in Rhizobium meliloti. Acta Microbiol. Polon. 16:7-12.

92. Krasilnikov, N. A. 1941. Variability of nodule bacteria. Doklady Akademii Nauk SSSR 31:90-92. (In Russian).

93. Krasilnikov, N. A. 1945. Induction of new virulence properties in nodule bacteria. Mikrobiol. 14:230-236. (In Russian).

94. Krol, A. J. N., J. G. G. Hontelez, R. C. Van Den Bos, and A. Van Kamen. 1981. Expression of large plasmids in the endosymbiotic form of Rhizobium leguminosarum. Nucleic Acid Res. 8:4337-4347.

95. Kuykendall, L. D. 1979. Transfer of R factors to and between genetically marked sublines of Rhizobium japonicum. Appl. Environ. Microbiol. 37:862-866.

96. Leonard, L. T. 1943. A simple assembly for use in the testing of cultures of rhizobia. J. Bacteriol. 45:523-527.

97. Lohnis, F., and R. Hansen. 1921. Nodule bacteria of leguminous plants. J. Agr. Res. 20:543-555.

98. Lowry, 0. H., N. J. Rosenbrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275.

99. Maier, R. J., and W. J. Brill. 1976. Ineffective and non-nodulatiog mutant strains of Rhizobium japonicum. J. Bacteriol. 127:763-769.

100. Martinez-de Drets, G., and A. Arias. 1970. Metabolism of some polyols by Rhizobium meliloti. J. Bacteriol. 103:97-103.

101. Martinez-de Drets, G., and A. Arias. 1972. Enzymatic basis for the differentiation of Rhizobium into fast- and slow-growing groups. J. Bacteriol. 109:467-470.

102. Martinez-de Drets, G., and A. Arias. 1977. 6-Phospho-D-gluconate: NAD⁺ 2-oxidoreductase (decarboxylating) from slow-growing rhizobia. J. Bacteriol. 130:1139-1143.

103. Masterson, R. V., P. R. Russell, and A. G. Atherly. 1982. Nitrogen fixation (nif) genes and large plasmids of Rhizobium japonicum. J. Bacteriol. 152:928-931.

104. May, S. N., and B. B. Bohlool. 1983. Competition among Rhizobium leguminosarum strains for nodulation of lentils (Lens esculenta). Appl. Environ. Microbiol. 45:960-965.

105. Morrison, N. A., Y. H. Cen, M. J. Trinick, J. Shine, and B. G. Rolfe. 1983. Heat curing of a Sym plasmid in a fast-growing Rhizobium sp. that is able to nodulate legumes and the nonlegume Parasponia sp. J. Bacteriol. 153:527-531.

106. Norris, D. 0. 1965. Acid production by Rhizobium: an unifying concept. Pl. Soil 22:143-166.

107. Nuti, M., A. M. Ledeboer, A. A. Lepidi, and R. A. Schilperoort. 1977. Large plasmids in different Rhizobium species. J. Gen. Microbiol. 100:241-248.

108. Nuti, M. P., A. A. Lepidi, R. K. Prakash, R. A. Schilperoort, and F. C. Cannon. 1979. Evidence for nitrogen fixation (nif) genes on indigenous Rhizobium plasmids. Nature 282:533-535.

109. Page, W. J. 1978. Transformation of Azotobacter vinelandii strains unable to fix nitrogen with Rhizobium spp. DNA. Can J. Microbiol. 24:209-214.

110. Pankhurst, C. E. 1979. Some antigenic properties of cultured cells and bacteroid forms of fast- and slow-growing strains of Lotus rhizobia. Microbios. 24:19-28.

111. Pariiskaya, A. N. 1973. The effect of acridine orange and mitomycin C on the symbiotic properties of Rhizobium meliloti. Mikrobiol. 42:119-121.

112. Peterson, J. B., and T. LaRue. 1981. Utilization of aldehydes and alcohols by soybean bacteroids. Pl. Physiol. 68:489-493.

113. Pilicinski, W. P., and E. L. Schmidt. 1981. Plasmid transfer within and between serologieally distinct strains of Rhizobium japonicum, using antibiotic resistance mutants and auxotrophs. J. Bacteriol. 145:1025-1030.

114. Prakash, R. K., P. J. J. Hooykaas, A. M. Ledeboer, J. W. Kijne, R. A. Schilperoort, M. P. Nuti, A. A. Lepidi, F. Casse, C. Boucher, J. S. Julliot, and J. Denarie. 1980. Detection, isolation, and characterization of large plasmids in Rhizobium, p. 139-163. In W. E. Newton and W. H. Orme-Johnson (ed.), Nitrogen Fixation. University Park Press, Baltimore.

115. Raina, J. L., and V. V. Modi. 1969. Genetic transformation in Rhizobium. J. Gen. Microbiol. 57:125-130.

116. Raina, J. L., and V. V. Modi. 1971. Further studies on genetic transformation in Rhizobium. J. Gen. Microbiol. 65:161-165.

117. Raina, J. L., and V. V. Modi. 1972. Deoxyribonucleate binding and transformation in Rhizobium japonicum. J. Bacteriol. 111:356-360.

118. Rigaud, J., and J. Trinchant. 1973. Isolement d’une alcool deshydrogenase chez Rhizobium et role dans le metabolisme indolique. Physiol. Plant. 23:160-165.

119. Roberts, G. P., W. T. Leps, L. E. Silver, and W. J. Brill. 1980. Use of two-dimensional polyacrylamide gel electrophoresis to identify and classify Rhizobium strains. Appl. Environ. Microbiol. 39:414-422.

120. Ruvkin, G. B., and F. M. Ausubel. 1980. Interspecies homology of nitrogenase genes. Proc. Nat. Acad. Sci. U.S.A. 77:191-195.

121. Schmidt, E. L. 1973. Fluorescent antibody techniques for the study of microbial ecology. Bull. Eco. Res. Comm. (Stockholm). 19:67-76.

122. Schmidt, E. L. 1974. Quantitative autecological study of microorganisms in soil by immunofluorescence. Soil Sci. 118:141-149.

123. Schmidt, E. L., R. 0. Bankole, and B. B. Bohlool. 1968. Fluorescent antibody approach to the study rhizobia in soil. J. Bacteriol. 95:1987-1992.

124. Skrdleta, V. 1965. Somatic sero-groups of Rhizobium japonicum. Plant Soil 23:43-48.

125. Skrdleta, V. 1969. Serological analysis of eleven strains of Rhizobium japonicum. Antonie. van Leeuwenhoek 35:77-83.

126. Smibert, R. M., and N. R. Krieg. 1981. General characterizations, p. 409-443. In P. Gerhardt (chief ed.), Manual of Methods for General Bacteriology. American Society for Microbiology, Washington, D. C.

127. Stevens, J. W. 1923. Can all strains of a specific organism be recognized by agglutination? J. Infectious Diseases. 33:557-566.

128. Strzelcowa, A. 1968. The use of the technique of van Schreven for the taxonomy of Rhizobium strains. Acta Microbiol. Polon. 17:263-268.

129. Tajima, S., and T. A. LaRue. 1982. Enzymes for acetaldehyde and ethanol formation in legume nodules. Plant Physiol. 70:388-392.

130. Trinick, M. J. 1965. Medicago sativa nodulation with Leucaena leucocephala root nodule bacteria. Aust. J. Sci. 27:263-264.

131. Van Straten, J., and E. L. Schmidt. 1974. Volatile compounds produced during acetylene reduction by detached soybean nodules. Soil Biol. Biochem. 6:347-351.

132. Vincent, J. M. 1970. A Manual for the Practical Study of Root-Nodule Bacteria. IBP Handbook 15. Blackwell Scientific Publications, Oxford.

133. Vincent, J. M. 1974. Root-nodule symbiosis with Rhizobium, p. 266-341. In A. Quispel (ed.), Biology of Nitrogen Fixation. North Holland Publishing Co., Amsterdam.

134. Vincent, J. M. 1977. Rhizobium: General microbiology, p. 277-366. In R. W. F. Hardy and W. S. Silver (ed.), A Treatise on Dinitrogen Fixation, Sec. III Biology. John Wiley and Sons, New York.

135. Vincent, J. M. 1982. The basic serology of rhizobia, p. 13-26. In J. M. Vincent (ed.), Nitrogen Fixation in Legumes. Academic Press, Sydney.

136. Vincent, J. M., and B. Humphrey. 1970. Taxonomically significant group antigens in Rhizobium. J. Gen. Microbiol. 63:379-382.

137. Vincent, J. M., B. Humphrey, and V. Skrdleta. 1973. Group antigens in slow-growing rhizobia. Arch. Mikrobiol. 89:79-82.

138. Vincent, J. M., P. S. Nutman, and F. A. Skinner. 1979. The identification and classification of Rhizobium, p. 49-69. In F. A. Skinner and D. W. Lovelock (ed.), Identification Methods for Microbiology, 2^nd ed. Academic Press, London.

139. Vintikova, H., M. Srogl, and V. Skrdleta. 1961. A contribution to the serological typization of the rhizobia. Folia Microbiol. 6:243-249.

140. Wright, W. H. 1925. The nodule bacteria of soybeans: I. Bacteriology of strains. Soil Sci. 20:95-141.

141. Zurkowski, W., M. Hoffman, and Z. Lorkiewicz. 1973. Effect of acriflavin and sodium dodecyl sulphate on infectiveness of Rhizobium trifolii. Acta Microbiol. Polon. 5:55-60.

142. Zurkowski, W., and Z. Lorkiewicz. 1978. Effective method for the isolation of non-nodulatiog mutants of Rhizobium trifolii. Genet. Res. 32:311-314.

143. Zurkowski, Y., and Z. Lorkiewicz. 1979. Plasmid-mediated control of nodulation in Rhizobium trifolii. Arch. Microbiol. 123:195-201.