INTERNATIONAL
JOURNAL OF SYSTEMATIC
BACTERIOLOGY,
Oct. 1988, p. 358-361
0020-7713/88/040358-04$02.OO/O
Copyright 0 1988, International Union of Microbiological Societies
Vol. 38, No. 4
Fatty Acids, Antibiotic Resistance, and Deoxyribonucleic Acid
Homology Groups of Bradyrhizobiurn japonicum
L. D. KUYKENDALL,'" M. A. ROY,2 J. J. O'NEILL,' AND T. E. DEVINE'
Plant Sciences Institute, Agricultural Research Service, U . S . Department of Agriculture, Beltsville, Maryland 20705,l
and Microbial ID, Inc., Newark, Delaware 197112
The fatty acid compositions and multiple antibiotic resistance patterns of 32 strains of Bradyrhizobium
japonicum correlated with two major deoxyribonucleic acid homology groups. In group I, the fatty acid
composition was 1.3% 16:l cis9 acid, 3.6% 16:lC acid, 8.8% 16:O acid, 1.2% 19:0 cyclopropane acid, and
81.2% 18:l acid. Group I1 contained 0.5% 16:lC acid, 11.1% 16:O acid, 0.8% 17:O cyclopropane acid, 24.7%
19:O cyclopropane acid, and 62.3% 18:l acid. Group I strains were susceptible to rifampin (500 pglml),
tetracycline (100 pglml), streptomycin (100 pg/ml), chloramphenicol (500 pg/ml), erythromycin (250 pglml),
carbenicillin (500 pg/ml), and nalidixic acid (50 pg/ml), whereas group I1 strains were resistant to these
antibiotics. Both groups were resistant to trimethoprim (50 pg/ml) and vancomycin (100 pglml).
taining these antibiotics (250 and 100 pg/ml, respectively).
The remaining strains were purified following dilution in HM
salts containing 0.01% Tween 20 (14).
Fatty acid analysis. (i) Growth conditions. Bacteria were
grown on AIE agar (13) amended with 0.1% gluconate at
28°C for 6 days.
(ii) Fatty acid extraction. Fatty acids were extracted by the
method of Miller and Berger (16), with minor modifications.
Portions (40 mg) of wet cells were removed from the surfaces
of the agar plates and placed in test tubes (13 by 100 mm)
with Teflon-lined screw caps. To each tube, 1 ml of 1.2 N
NaOH in 50% methanol was added, and the tubes were
sealed, vortexed, and placed in boiling water for 30 min. The
tubes were cooled to room temperature, 2 ml of 6 N HC1 in
methanol (325:275, vol/vol) was added, and the tubes were
capped and placed in an 80°C water bath for 10 min. After
cooling, 1.25 ml of hexane-methyl tert-butyl ether ( l : l , vol/
vol) was added to each tube. The tubes were rotated on a
hematology mixer (Fisher Scientific Co., Pittsburgh, Pa.) for
10 min. The lower aqueous phase was removed and discarded, and 3 ml of 0.3 N NaOH was added. The tubes were
returned to the mixer for 5 min. The organic phase was then
transferred to a chromatography vial and sealed with a
Teflon-lined septum.
(iii) Chromatography. The fatty acid analyses were performed by using a model 5898A microbial identification
system (Hewlett-Packard Co., Palo Alto, Calif.) consisting
of Hewlett-Packard model 5890 gas chromatograph equipped
with a 5% phenyl methyl silicone capillary column (0.2 mm
by 25 m), a flame ionization detector, a Hewlett-Packard
model 3392 integrator, a Hewlett-Packard model 7673A
automatic sampler, and a Hewlett-Packard model 216 computer. Peaks were automatically integrated, and fatty acid
names and percentages were calculated.
The gas chromatographic parameters were as follows:
carrier gas, ultra-high-purity hydrogen; column head pressure, 10 lb/in2; injection volumn, 2 pl; column head split
ratio, 1OO:l; septum purge, 5 mumin; column temperature,
170 to 270°C at S"C/rnin; injection port temperature, 250°C;
and detector temperature, 300°C.
Antibiotic resistance. The same medium that was used for
fatty acid analysis was used for the antibiotic resistance
experiments. Stationary-phase broth cultures were used as
inocula in the antibiotic resistance tests. Plates containing an
antibiotic, as well as a nonselective control plate, were
Bradyrhizobium japonicum (Jordan 1982) (formerly Rhizobium japonicum) is the slow-growing symbiont of soybeans. This species is comprised of two highly divergent
taxonomic groups which should be phenotypically distinguishable. The deoxyribonucleic acid (DNA)-DNA hybridization studies of Hollis et al. (8) established DNA homology
groups I, Ia, and 11. These groups were characterized with
respect to acidic extracellular polysaccharide composition
and phenotypic ex planta nitrogenase expression. Groups I
and Ia were reported to be quite distinct from DNA homology group I1 according to both of these criteria (9). Using
some of the same strains, Stanley et al. (21) characterized
the DNA nucleotide sequence divergences in and proximal
to known symbiotic genes and concluded that strains of B .
japonicum comprise two symbiotic genotypes (sTI and sTII)
consistent with separate species designations.
Whole-cell fatty acid profiles of strains grown under
standardized conditions have been used for identification (1,
4, 18) and classification of many bacteria (10, 17), but not in
the genus Bradyrhizobium.
Pankhurst et al. (20) correlated rifampin resistance in
some slow-growing rhizobial strains with in vitro nitrogenase
activity. This observation, together with that of Huber et al.
(9) that only DNA homology group I1 strains express nitrogenase in vitro, suggested to us that patterns of antibiotic
resistance might also distinguish the groups. In this paper we
report that whole-cell fatty acid profiles correlate with DNA
homology groups I and I1 of Hollis et al. (8) and that patterns
of high-level resistance to antibiotics distinguish the two
DNA homology groups.
MATERIALS AND METHODS
Bacterial strains. All USDA strains were obtained from
H. H. Keyser and D.Weber, U. S. Department of Agriculture (USDA), Beltsville, Md.; strain ATCC 10324 was obtained from the American Type Culture Collection, Rockville, Md.; strains 5631, 5633, D193, D209, THA6, and 8'
were obtained from G. H. Elkan, North Carolina State
University, Raleigh; and strains 61A.50 and 61A76 were
obtained from S. Smith, Nitragin Co., Milwaukee, Wis.
Strains that exhibit rifampin resistance and tetracycline
resistance were purified by streaking onto a medium con-
* Corresponding author.
358
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 00:03:31
VOL. 38, 1988
PHENOTYPIC DIFFERENCES AMONG B. JAPONZCUM STRAINS
359
TABLE 1. Fatty acid contents of the two distinct groups of Bradyrhizohiurn which nodulate soybeans
Fatty acid composition (%)
Group
1"
2h
16:lcis9
16:1C
1.3 -+ 0.8
0.0
3.6 t 2.5
0.5 t 0.5
17:O
c yclopropane
16:O
19:O
cyclopropane
18:1
~
8.8 k 1.0
11.1 k 1.7
0.0
0.8 t 0.4
1.2 t 1.1
24.7 +- 4.2
81.2 t 5.7
62.3 4.8
*
Fatty acid group 1 consists of B. japonicum strains ATCC 10324, 1-110, ARS, L1-110, 5631, 8", 61A50, USDA 62, D193, USDA 24, USDA 38, USDA 58,
USDA 115, 5633, THA 6, and D209 (all in DNA homology group I).
Fatty acid group 2 consists of strains USDA 29, USDA 31, USDA 46, USDA 76, USDA 86, USDA 94, USDA 130, and 61A76. These eight strains are in DNA
homology group I1 of Hollis et al. (8). Strains USDA 23, USDA 40, USDA 61, USDA 83, USDA 84, USDA 85, USDA 100, and USDA 119 fell into fatty acid
group I1 although these strains have not been characterized as to DNA homology.
'I
streaked with portions (approximately 10 pl) of inoculum,
and growth was scored after 10 days. Resistance to a
particular concentration of antibiotic was defined as the
ability of a strain to form colonies at that concentration. It is
important to make the distinction between natural resistance
of a wild-type strain and the selection of mutants. Intrinsic
resistance is clearly distinguished from mutant selection
since in the latter case only a few colonies are formed per lo7
to 10' cells plated. Stock solutions of streptomycin, chloramphenicol, carbenicillin, and vancomycin were prepared in
distilled water; nalidixic acid and rifampin were prepared in
0.35 N NaOH; tetracycline and erythromycin were prepared
in 50% ethanol: and trimethoprim was prepared in 70%
ethanol. All preparations were filter sterilized by using
0.45-wm membrane filters. Antibiotics were added to molten
agar after sterilization and cooling to 50°C.
less, one of these antibiotics is useful in combination with
any two of the five others which are not inoculum dependent. For example, it is sufficient to test for resistance to
erythromycin, rifampin, and tetracycline.
Data on the antibiotic resistance profiles of 10 strains
classified as B . japonicum which were not included in the
DNA hybridization studies of Hollis et al. (8) are also shown
in Table 2. Nine of these strains possessed antibiotic resistance patterns that are characteristic of DNA homology
TABLE 2. Antibiotic resistance, fatty acid composition groups,
and DNA homology groups
DNA
homologv
group
1-
Strain
I
RESULTS
When grown under standardized conditions, strains of
DNA homology group I1 were distinguishable from strains of
DNA homology group I or Ia on the basis of fatty acid
content. The mean percent compositions of the 19:O cyclo
and 18:l fatty acids were the most distinctive (Table 1).In
our grouping of 32 Bradyrhizobium strains according to
whole-cell fatty acid content, fatty acid group 1 contains all
strains of DNA homology groups I and Ia. Fatty acid group
2 consists of eight strains of DNA homology group 11, as well
as eight other USDA strains that were previously undefined.
Table 2 lists the bacterial strains with their known DNA
homology groups and their growth responses on media
containing the antibiotics used. Of the nine antibiotics, seven
were useful for distinguishing groups. All of the strains in
DNA homology group 11, with the exception of USDA 94,
exhibited high levels of resistance to carbenicillin (500 pg/
ml), chloramphenicol(500 pg/ml), erythromycin (250 pghl),
nalidixic acid (50 pg/ml), rifampin (500 pg/ml), tetracycline
(100 pg/ml), and streptomycin (100 pg/ml). Strain USDA 94
was resistant to all of the antibiotics tested except carbenicillin and nalidixic acid. However, all of the B . japonicum
strains in DNA homology groups I and Ia except strain
USDA 24 were susceptible to these antibiotics at the concentrations used; strain USDA 24 (group I) was resistant to
streptomycin. Neither trimethoprim nor vancomycin was
useful in distinguishing groups since all of the Brudyrhizobium strains tested in both DNA homology groups except
strains 5633 and USDA 123 were resistant to these antibiotics at concentrations of 50 and 100 Fg/ml, respectively;
strains 5633 and USDA 123 were resistant only to trimethoprim. Although erythromycin and nalidixic acid could be
used to distinguish between strains of DNA homology group
I1 and strains of groups I and Ia, resistance to these two
antibiotics was somewhat inoculum dependent. Nonethe-
THA 6
USDA 58
USDA115
ATCC10324
USDA 24
D193
D209
USDA123
USDA 38
5633
Ia
USDA 62
USDA110
61A50
8"
5631
I1
61A76
USDA29
USDA 31
USDA46
USDA76
USDA86
USDA94
USDA117
USDA130
Undefined USDA 7
USDA 23
USDA40
USDA 61
USDA 83
USDA84
USDA85
USDA91
USDA100
USDA119
Fatty
Antibiotic resistanceU
acid
group tmp van rif tet str chl ery car nal
1
1
1
1
1
1
1
ND"
1
1
1
1
1
1
1
2
2
2
2
2
2
2
ND
2
ND
2
2
2
2
2
2
ND
2
2
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
b
r
r
r
r
r
r
s
r
s
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
s
s
s
s
s
s
s
s
s
s
s
s
s
s
r
r
r
r
r
r
r
r
r
s
r
r
r
r
r
r
r
r
r
s
s s
s s
s s
s s
s r
s s
s s
s s
s s
s s
s s
s s
s s
s s
s s
r r
r r
r r
r r
r r
r r
r r
r r
r r
s s
r r
r r
r r
r r
r r
r r
r r
r r
r r
s
s
s
s
s
s
s
s
s
s
s
s
s
s
r
r
r
r
r
r
r
r
r
s
r
r
r
r
r
r
r
r
r
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
r
r
r
r
r
r
r
r
r
s
r
r
r
r
r
r
r
r
r
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
r
r
r
r
r
r
s
r
r
s
r
r
r
r
r
r
r
r
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
r
r
r
r
r
r
s
r
r
s
r
r
r
r
r
r
r
r
s
tmp, Trimethoprim (50 pg/ml); van, vancomycin (100 pg/rnl);rif, rifampin
(500 pg/ml); tet, tetracycline (100 pg/ml); str, streptomycin (100 pg/ml); chl,
chloramphenicol(500 ~ g / m l )ery,
; erythromycin (250 p,g/ml); car, carbenicillin
(500 pglml); nal, nalidixic acid (50 pglrnl).
r. Resistant: s. susceDtible.
' ND, Not determined:
'
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 00:03:31
360
INT. J . SYST. BACTERIOL.
KUYKENDALL ET AL.
group 11, and their fatty acid contents were characteristic of
group I1 strains. Included in this group of strains with
multiple resistance markers was USDA 119, which is susceptible to carbenicillin and nalidixic acid. Strain USDA 7
lacked resistance to each of the antibiotics except trimethoprim and vancomycin and therefore has characteristics of
strains in DNA homology groups I and Ia.
DISCUSSION
Hollis, et al. (8) interpreted DNA homology groups I and
Ia as subspecies of B. japonicum, whereas DNA homology
group I1 was considered dissimilar enough to warrant separate species designation. Stanley et al. (21) found that
several of the strains fell into markedly different groups
based on estimated DNA nucleotide sequence divergence in
nif and nod genes and designated these groups “symbiotic
types sTI and sTII.” Since we found that strains of DNA
homology group 11, as well as strains of type sTII, possess a
broad spectrum of high-level antibiotic resistance, these
results support the proposed classification systems of Hollis
et al. (8) and Stanley et al. (21).
Four of the nine strains in DNA homology group I1 are
strains of bradyrhizobia which produce foliar chlorosis ( 5 )
caused by the anti-metabolite rhizobitoxine (19). Of the 10
strains tested which are not characterized by relative DNA
homology, 8 are chlorosis producers (6, 7). The chlorosisproducing strains produced nodulelike swellings on peanuts
(6) whereas group I strains did not, and we found that the
chlorosis-producing strains can be differentiated from B.
juponicum based on their antibiotic resistance profiles. Rifampin resistance, in vitro nitrogenase activity, and chlorosis-producing ability are properties shared by group I1
strains and Bradyrhizobium sp. (cowpea). This phenotypic
similarity suggests a closer genetic relatedness of these
organisms to each other than to B. japonicum.
Fatty acid contents and antibiotic resistance profiles
clearly distinguish the two major DNA homology groups of
Bradyrhizobium strains which nodulate soybeans and which
are currently classified as B. japonicum. This is in contrast to
the results of the fatty acid analysis of MacKenzie et al. (15),
in which intergeneric differences among rhizobia were
readily found, but interspecific differences within a genus
were not identified. Standardized conditions are important
for analysis of fatty acid composition, and these authors (15)
used different cultural conditions (mannitol-yeast broth,
growth for 5 days at 25°C) than those used in this study
(AlEG agar plates, growth for 6 days at 28°C). The two
studies also used radically different chemical extraction
procedures, which may affect fatty acid recovery. In addition, MacKenzie et al. (15) used packed-column chromatography, which does not have the high resolution of capillary
column chromatography to show the differences found in the
current study.
Previous reports have shown that antibiotic resistance in
root nodule bacteria is common (2, 12, 20) although there is
a considerable amount of variation within and between
species. Cole and Elkan (3) suggested that multiple antibiotic
resistance is characteristic of B. japonicum. However, our
data show that strains in DNA homology group I1 possess a
characteristic profile of multiple antibiotic resistance at
levels not found in strains of DNA homology groups I and Ia.
Thus, differentiation based on multiple resistance markers
provides a rapid and efficient method for distinguishing
between strains of Bradyrhizobium with distinct genetic and
phenotypic backgrounds and should prove to be valuable for
surveying soybean microsymbiont populations in field assays of symbiotic competence or competitiveness. Strains
USDA 94 and USDA 119 apparently lack the carbenicillin
and nalidixic acid resistance profiles of other group I1
strains, and this may be interpreted as divergence within the
group. Thus, resistance to the following five antibiotics is
perhaps the most useful for differentiating the two DNA
homology groups: rifampin, tetracycline, streptomycin,
chloramphenicol, and erythromycin. Testing any three of
these should suffice. These characteristics may also prove to
be useful as selective markers in interspecific matings between B . japonicum strains of DNA homology groups I and
Ia and strains of DNA homology group 11. Analysis of
potential linkage between a gene(s) conditioning specific
resistance to these antibiotics and other markers, including
genes encoding symbiotic functions, would result in a better
understanding of the genetic organization of Brudyrhizobium.
ACKNOWLEDGMENTS
We express our appreciation to Karl Whitley and Seanne Udell
for their assistance.
LITERATURE CITED
1. Abel, K., H. deschmertzing, and J. I. Peterson. 1963. Classification of microorganisms by analysis of chemical composition.
I. Feasibility of utilizing gas chromatography. J . Bacteriol. 85:
1039-1044.
2. Cole, M. A., and G. H. Elkan. 1973. Transmissible resistance to
penicillin G, neomycin, and chloramphenicol in Rhizobium
japonicum. Antimicrob. Agents Chemother. 4:24&253.
3. Cole, M. A., and G. H. Elkan. 1979. Multiple antibiotic resistance in Rhizobium juponicum. Appl. Environ. Microbiol. 37:
867-870.
4. De Boer, S. H., and M. Sasser. 1986. Differentiation of Erhhiu
carotovora ssp. carotovoru and E . carotovora ssp. utroseptica
on the basis of cellular fatty acid composition. Can. J. Microbiol. 32:796-800.
5. Devine, T. E. 1985. Host range and compatibility of soybean
with rhizobial microsymbionts, p. 484-492. In R. Shibles (ed.),
World Soybean Research Conference 111: Proceedings. Westview Press, London.
6. Devine, T. E., L. D. Kuykendall, and B. H. Briethaupt. 1983.
Nodule-like structures induced on peanut by chlorosis producing strains of Rhizobium classified as R . japonicum. Crop Sci.
23:394-397.
7. Devine, T. E., and D. F. Weber. 1977. Genetic specificity of
nodulation. Euphytica 26527-535.
8. Hollis, A. B., W. E. Kloos, and G. H. Elkan. 1981. DNA:DNA
hybridization studies of Rhizobiurn juponicum and related Rhizobiaceae. J. Gen. Microbiol. 123:215-222.
9. Huber, T. A., A. K. Agarwal, and D. L. Keister. 1984. Extracellular polysaccharide composition, ex planta nitrogenase activity, and DNA homology in Rhizobium japonicum. J. Bacterial. 158:1168-1171.
10. Jackwood, M. W., M. Sasser, and Y. M. Saif. 1985. Contribution
to the taxonomy of the turkey coryza agent: cellular fatty acid
analysis of the bacterium. Avian Dis. 30:172-178.
11. Jordan, D. C. 1982. Transfer of Rhizobium japonicum Buchanan
1980 to Brudyrhizobium gen. nov., a genus of slow-growing,
root-nodule bacteria from leguminous plants. Int. J. Syst. Bacteriol. 32:136-139.
12. 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.
13. Kuykendall, L. D. 1979. Transfer of R factors to and between
genetically marked sublines of Rhizobium juponicum. Appl.
Environ. Microbiol. 375362466.
14. Kuykendall, L. D., and G. H. Elkan. 1976. Rhizobiumjuponi-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 00:03:31
VOL.38, 1988
PHENOTYPIC DIFFERENCES AMONG B. JAPONZCUM STRAINS
cum derivatives differing in nitrogen-fixing efficiency and carbohydrate utilization. Appl. Environ. Microbiol. 32511-519.
15. MacKenzie, S. L., M. S. Lapp, and J. J. Child. 1979. Fatty acid
composition of Rhizobium spp. Can. J. Microbiol. 2568-74.
16. Miller, L., and T. Berger. 1985. Bacteria identification by gas
chromatography of whole cell fatty acids. Hewlett-Packard
Application Note 228-38. Hewlett-Packard Co., Palo Alto,
Calif.
17. Moore, L. V. H., J. L. Johnson, and W. E. C. Moore. 1987.
Selenomonas noxia sp. nov., Selenomonas fiueggei sp. nov.,
Selenomonas infelix sp. nov., Selenomonas dianae sp. nov.,
and Selenomonas artemidis sp. nov., from human gingival
crevices. Int. J. Syst. Bacteriol. 36:271-280.
361
18. Moss, C. W. 1981. Gas-liquid chromatography as an analytical
tool in microbiology. J. Chromatogr. 203:337-347.
19. Owens, L. D., J. F. Thompson, R. G. Pitcher, and T. Williams.
1972. Structure of rhizobitoxine, an antimetabolic enol-ether
amino acid from Rhizobium japonicum. J. Chem. SOC.Chem.
Commun. 1972:714.
20. Pankhurst, C. E., D. B. Scott, and C. W. Ronson. 1982.
Correlation between rifampicin-resistance of slow-growing Rhizobium strains and their ability to express nitrogenase activity in
culture. FEMS Microbiol. Lett. 15137-139.
21. Stanley, J., G. G. Brown, and D. P. S. Verma. 1985. Slowgrowing Rhizobium japonicum comprises two highly divergent
symbiotic types. J. Bacteriol. 163:148-154.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 00:03:31