Vol. 56, No. 4
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Apr. 1990, p. 1172-1174
0099-2240/90/041172-03$02.00/0 Copyright X 1990, American Society for Microbiology
Screening of Symbiotic Frankiae for Host Specificity by Restriction Fragment Length Polymorphism Analysis APHAKORN NITTAYAJARN,l BETH C. MULLIN,2 AND DWIGHT D. BAKER3* Department of Agriculture, Biological Nitrogen Fixation Resource Centre for Southeast Asia, Bangkhen 10900, Bangkok, Thailand'; Department of Botany and Plant Physiology and Genetics Program, University of Tennessee, Knoxville, Tennessee 37996-11002; and School of Forestry and Environmental Studies, Yale University, New Haven, Connecticut 065113 Received 20 September 1989/Accepted 9 January 1990
Restriction fragment length polymorphism analysis of numerous Frankia strains, using a nifDH probe, separated the strains into three distinct groups based on hybridization patterns. The groups identified in this study were well correlated with host specificity groups identified in earlier cross-inoculation studies.
After centrifugation of the cell lysate, the supernatant was decanted and deproteinated by a phenol-chloroform extraction. DNA was precipitated in ice-cold ethanol at -20°C overnight. The isolated DNA was suspended in TE buffer (10 mM Tris hydrochloride, 1 mM EDTA, pH 8.0) and stored until use at -20°C. Preparation of the DNA probe. The recombinant plasmid, pAS1, which carries K. pneumoniae nifD and -H genes, was obtained from transformed cells of E. coli JA221. Cells from an overnight broth culture were harvested and plasmid DNA was extracted by the boiling method of Holms and Quigley (7). The 3.2-kb HindIII-EcoRI DNA fragment containing nifD and -H was radiolabeled by using [32P]dCTP and a
Since the successful isolation of Frankia sp. strain Cpll (5), a large amount of research effort has been devoted to characterizing pure cultured strains by their symbiotic capabilities. Tedious cross-inoculation studies requiring large numbers of plants (2, 8) have been the only method to determine the host specificities of pure cultured strains. Recently, Lalonde et al. (9) described the utilization of DNA analyses to classify strains in concert with other laboratory characterization methods. Normand et al. (11) demonstrated that the genes encoding nitrogenase proteins were highly conserved in Frankia strains which they tested. Because of the overall high degree of DNA sequence divergence among Frankia strains (1, 4), we thought that restriction fragment length polymorphism analysis could be useful in categorizing strains into groups which would reflect host specificities in initial screenings. Bacteria and media. Twenty-nine Frankia strains (Table 1) representing the four host specificity groups of Baker (2) were used in this study. Each strain was cultured in an appropriate medium (Table 1) at 28°C. Escherichia coli JA221 containing the recombinant plasmid pAS1 was cultured in liquid LB medium on a gyratory shaker or on agar-solidified plates at 37°C. Ampicillin was added to the medium at a concentration of 50 pg ml-' when required. Plasmid pAS1 is a recombinant plasmid constructed in the laboratory of F. M. Ausubel by cloning a 3.2-kilobase (kb) HindlIl-EcoRI fragment containing Klebsiella pneumoniae nifD and nifH and part of nifK sequences into the corresponding sites of pBR322. Frankia DNA isolation. The isolation of DNA from Frankia sp. was essentially the same as described by Bloom et al. (4), which was originally adapted from a procedure described by Hintermann et al. (6). Cells from 2- to 3-week-old broth cultures were harvested by centrifugation and washed two times with distilled water, and the quantity of cells was measured by packed cell volume (10). The cells were then suspended in 3 volumes of TS buffer (50 mM Tris hydrochloride, 0.75 M sucrose, pH 8.0) and homogenized by grinding in a Potter-Elvehjem tissue grinder. The homogenized cells were lysed by adding EDTA (0.1 M) and lysozyme (5 mg ml-') and incubating at 37°C overnight. Proteinase K addition (50 ,ul ml-') followed by sodium dodecyl sulfate addition (0.5%) and incubation at 60°C were used to complete the
TABLE 1. Frankia strains used in this experiment Culture Trivial Geographical HSGc no. Registry Reity mediuMa originb designation
AllI1 ArI3 ArI5
AvcIl AvsI3 CcI3 CcOl Cellb CeI2 CpI2b CpI3 G2 JCT 287 McI2b MgI5 R2 R43 1935 55005 14 6 24 31 63 61 77 89
WgCcl.17
lysis.
DPM FB FB FB FB DPM DPM
YCz DPM FB FB YCz DPM DPM FB DPM DPM DPM DPMN DPMN S+TW L/2 L/2 S+TW S+TW L/2 L/2 DPM
Florida Oregon Oregon Ontario Washington Florida China Hawaii Florida Massachusetts New Hampshire Guadeloupe Australia Virginia New York Vermont Florida Thailand Florida New Jersey New Jersey New Jersey New Jersey New Jersey New Jersey New Jersey New Jersey The Netherlands
Culture media are as described by Baker (3). Indicates the original source of material for bacterial isolation. cTaken from Baker (2; and unpublished data).
a
b *
HFP 022801 HFP 013003 DDB 01310310 DDB 01020110 DDB 01360610 HFP 020203 Not defined DDB 02060120 DDB 02060210 DDB 07010210 DDB 07010310 ORS 020604 Not defined DDB 16060820 DDB 16110210 LLR 03013 LLR 02022 DAB 021002 DDB 02060510 RBR 162001 RBR 162002 RBR 162009 RBR 162013 RBR 162017 RBR 162019 RBR 162020 RBR 162021 Not defined
Corresponding author. 1172
2 1 1 1 1 2 2 4 3 1 1 3 2 3 4 3 3 2 3 3 3 ? 1 3 3 1 1 3
VOL. 56, 1990
NOTES
1173
FIG. 1. Composite autoradiograph of representative hybridization patterns obtained with the nifDH probe with numerous Frankia strains. Lanes: 1, CpI2b; 2, AvcIl; 3, ArI3; 4, AllIl; 5, JCT 287; 6, CeI2; 7, 55005; 8, MgI5; 9, Cellb; 10, RBR 162009; 11, RBR 162020; 12, AvcIl; 13, AllIl; 14, G2; 15, McI2b; 16, R100; 17, R43; 18, WgCcl.17. Lanes 8, 13, and 16 contained insufficient DNA to be resolved in this autoradiograph.
random primers labeling kit (Bethesda Research Laboratories). Membrane transfer and detection. Frankia DNA (3 to 5 ,ug well-) was digested with restriction endonuclease and electrophoresed on 0.8% agarose gels in TAE (40 mM Trisacetate, 2 mM EDTA, pH 8.0). DNA fragments were transferred to a Hybond-N nylon membrane (Amersham Corp.) by Southern blotting (12). Hybridization with the nifDH probe was performed overnight at 65°C. After hybridization and washing with lx SSC (150 mM NaCl, 15 mM sodium citrate, pH 7.0), the membrane was exposed to Kodak X-Omat film overnight at -70°C, using two intensifying screens.
Figure 1 illustrates typical DNA hybridization patterns when the restriction endonuclease Sall and the nifjH probe are used. A summary of our hybridization results is provided in Table 2. Three distinct hybridization patterns are evident. Pattern I, which corresponds to host specificity group 1 strains, had hybridization bands at 1.7 and 0.94 kb; pattern II, which corresponds to host specificity group 2 strains, had hybridization bands at 2.4 and 1.1 kb; and pattern III, which corresponds to host specificity group 3 and 4 strains, had a single hybridization band at approximately 3.5 kb with some higher or lower. Use of other restriction endonucleases (PstI, XhoI, or BamHI) resulted in hybridization patterns which, although useful in grouping strains, were more complex than the Sall patterns (not shown). Determination of host specificity among newly isolated Frankia strains is important for practical purposes. This is particularly so because, for unknown reasons, isolated strains may not have the capability of reinfecting the host genus from which they were derived (2, 3). From the Frankia strains currently held in pure culture, a limited understand-
ing of host specificity has been gained. Three major host specificity groups can be defined if the cross-inoculation data from the two promiscuous actinorhizal genera Gymnostoma and Myrica are disregarded. Briefly, these host specificity groups (HSG) are HSG 1, consisting of strains infecting Alnus and Comptonia spp.; HSG 2, composed of strains infecting genera of the family Casuarinaceae; and HSG 3, composed of strains infecting genera of the Elaeagnaceae. Gymnostoma and Myrica are infected by strains from all of these groups with the exception of a few strains which only nodulate members of the Elaeagnaceae. Baker (2) has defined the latter strains as "restrictive" and separated them into a fourth HSG. At present, the determination of host specificity for any new strain requires cross-inoculation tests TABLE 2. Groups defined by restriction fragment length
polymorphism (RFLP) analysis RFLP group I
(bands, 0.94 and 1.7 kb)
ArI3 ArI5 AvcIl AvsI3
CpI2b CpI3 31 77
89
RFLP group II (bands, 1.1 & 2.4 kb)
AllIl CcI3 CcOl JCT 287 1935
RFLP group III
(bands, 2.9-4.4 kb)
CelIlb CeI2 G2 McI2b MgI5 R2 R43
WgCcl.17 6 24 61 63
55005
1174
NOTES
conducted on a minimum of four host genera. Thus, a rapid in vitro assay to predict host specificity would be greatly beneficial. Frankia strains which belonged to HSG 1 and HSG 2 were particularly consistent in their hybridization patterns. Frankia strains belonging to HSG 3 or HSG 4 demonstrated hybridization of the probe to a single Sall fragment which varied slightly in size among the strains. The variability of fragment size observed with HSG 3 and HSG 4 strains would indicate greater genetic diversity among these strains than among strains within HSG 1 or HSG 2. A similar conclusion was drawn by Lalonde et al. (9) when Frankia strains were analyzed by using a number of chemical and physiological tests. Our work should have considerable practical value in initial screenings of newly isolated Frankia strains. Restriction fragment length polymorphism characterization could be conducted and necessary cross-inoculation tests could be reserved for only those strains likely to belong to a desired HSG. We thank Raanan Bloom and Dusty Ide for use of Frankia strains and DNA samples and F. M. Ausubel for plasmid pAS1. This research was supported by grant DPE-5542-G-SS-7707-00 from the U.S. Agency for International Development, Program for Science and Technology Cooperation.
LITERATURE CITED 1. An, C. S., W. S. Riggsby, and B. C. Mullin. 1985. Relationships of Frankia isolates based on deoxyribonucleic acid homology studies. Int. J. Syst. Bacteriol. 35:140-146. 2. Baker, D. D. 1987. Relationships among pure cultured strains of
APPL. ENVIRON. MICROBIOL.
Frankia based on host specificity. Physiol. Plant. 70:245-248. 3. Baker, D. D. 1989. Methods for the isolation, culture, and characterization of the Frankiaceae: soil actinomycetes and symbionts of actinorhizal plants, p. 213-236. In D. Labeda (ed.), Isolation of biotechnological organisms from nature. McGrawHill Book Co., New York. 4. Bloom, R. A., B. C. Mullin, and R. L. Tate. 1989. DNA restriction patterns and solution hybridization studies of Frankia isolates from Myrica pensylvanica (bayberry). Appl. Environ. Microbiol. 55:2155-2160. 5. Callaham, D., P. Del Tredici, and J. G. Torrey. 1978. Isolation and cultivation in vitro of the actinomycete causing root nodulation in Comptonia. Science 199:899-902. 6. Hintermann, G., R. Crameri, T. Kieser, and R. Hutter. 1981. Restriction analysis of the Streptomyces glaucescens genome by agarose gel electrophoresis. Arch. Microbiol. 130:218-222. 7. Holms, D. S., and M. Quigley. 1981. A rapid boiling method for the preparation of bacterial plasmids. Anal. Biochem. 114: 551-559. 8. Huang, J. B., Z. Y. Zhao, G. X. Chen, and H. C. Liu. 1985. Host range of Frankia endophytes. Plant Soil 87:61-65. 9. Lalonde, M., L. Simon, J. Bousquet, and A. Seguin. 1988. Advances in the taxonomy of Frankia: recognition of species alni and elaeagni and novel subspecies pommerii and vandijkii, p. 671-680. In H. Bothe, F. J. de Bruijn, and W. E. Newton (ed.), Nitrogen fixation: hundred years after. Gustav Fischer, Stuttgart, Federal Republic of Germany. 10. Nittayajarn, A., and D. D. Baker. 1989. Methods for the quantification of Frankia cell biomass. Plant Soil 118:199-204. 11. Normand, P., P. Simonet, and R. Bardin. 1988. Conservation of nif sequences in Frankia. Mol. Gen. Genet. 213:238-246. 12. Southern, E. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98: 503-517.
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Apr. 1990, p. 1172-1174
0099-2240/90/041172-03$02.00/0 Copyright X 1990, American Society for Microbiology
Screening of Symbiotic Frankiae for Host Specificity by Restriction Fragment Length Polymorphism Analysis APHAKORN NITTAYAJARN,l BETH C. MULLIN,2 AND DWIGHT D. BAKER3* Department of Agriculture, Biological Nitrogen Fixation Resource Centre for Southeast Asia, Bangkhen 10900, Bangkok, Thailand'; Department of Botany and Plant Physiology and Genetics Program, University of Tennessee, Knoxville, Tennessee 37996-11002; and School of Forestry and Environmental Studies, Yale University, New Haven, Connecticut 065113 Received 20 September 1989/Accepted 9 January 1990
Restriction fragment length polymorphism analysis of numerous Frankia strains, using a nifDH probe, separated the strains into three distinct groups based on hybridization patterns. The groups identified in this study were well correlated with host specificity groups identified in earlier cross-inoculation studies.
After centrifugation of the cell lysate, the supernatant was decanted and deproteinated by a phenol-chloroform extraction. DNA was precipitated in ice-cold ethanol at -20°C overnight. The isolated DNA was suspended in TE buffer (10 mM Tris hydrochloride, 1 mM EDTA, pH 8.0) and stored until use at -20°C. Preparation of the DNA probe. The recombinant plasmid, pAS1, which carries K. pneumoniae nifD and -H genes, was obtained from transformed cells of E. coli JA221. Cells from an overnight broth culture were harvested and plasmid DNA was extracted by the boiling method of Holms and Quigley (7). The 3.2-kb HindIII-EcoRI DNA fragment containing nifD and -H was radiolabeled by using [32P]dCTP and a
Since the successful isolation of Frankia sp. strain Cpll (5), a large amount of research effort has been devoted to characterizing pure cultured strains by their symbiotic capabilities. Tedious cross-inoculation studies requiring large numbers of plants (2, 8) have been the only method to determine the host specificities of pure cultured strains. Recently, Lalonde et al. (9) described the utilization of DNA analyses to classify strains in concert with other laboratory characterization methods. Normand et al. (11) demonstrated that the genes encoding nitrogenase proteins were highly conserved in Frankia strains which they tested. Because of the overall high degree of DNA sequence divergence among Frankia strains (1, 4), we thought that restriction fragment length polymorphism analysis could be useful in categorizing strains into groups which would reflect host specificities in initial screenings. Bacteria and media. Twenty-nine Frankia strains (Table 1) representing the four host specificity groups of Baker (2) were used in this study. Each strain was cultured in an appropriate medium (Table 1) at 28°C. Escherichia coli JA221 containing the recombinant plasmid pAS1 was cultured in liquid LB medium on a gyratory shaker or on agar-solidified plates at 37°C. Ampicillin was added to the medium at a concentration of 50 pg ml-' when required. Plasmid pAS1 is a recombinant plasmid constructed in the laboratory of F. M. Ausubel by cloning a 3.2-kilobase (kb) HindlIl-EcoRI fragment containing Klebsiella pneumoniae nifD and nifH and part of nifK sequences into the corresponding sites of pBR322. Frankia DNA isolation. The isolation of DNA from Frankia sp. was essentially the same as described by Bloom et al. (4), which was originally adapted from a procedure described by Hintermann et al. (6). Cells from 2- to 3-week-old broth cultures were harvested by centrifugation and washed two times with distilled water, and the quantity of cells was measured by packed cell volume (10). The cells were then suspended in 3 volumes of TS buffer (50 mM Tris hydrochloride, 0.75 M sucrose, pH 8.0) and homogenized by grinding in a Potter-Elvehjem tissue grinder. The homogenized cells were lysed by adding EDTA (0.1 M) and lysozyme (5 mg ml-') and incubating at 37°C overnight. Proteinase K addition (50 ,ul ml-') followed by sodium dodecyl sulfate addition (0.5%) and incubation at 60°C were used to complete the
TABLE 1. Frankia strains used in this experiment Culture Trivial Geographical HSGc no. Registry Reity mediuMa originb designation
AllI1 ArI3 ArI5
AvcIl AvsI3 CcI3 CcOl Cellb CeI2 CpI2b CpI3 G2 JCT 287 McI2b MgI5 R2 R43 1935 55005 14 6 24 31 63 61 77 89
WgCcl.17
lysis.
DPM FB FB FB FB DPM DPM
YCz DPM FB FB YCz DPM DPM FB DPM DPM DPM DPMN DPMN S+TW L/2 L/2 S+TW S+TW L/2 L/2 DPM
Florida Oregon Oregon Ontario Washington Florida China Hawaii Florida Massachusetts New Hampshire Guadeloupe Australia Virginia New York Vermont Florida Thailand Florida New Jersey New Jersey New Jersey New Jersey New Jersey New Jersey New Jersey New Jersey The Netherlands
Culture media are as described by Baker (3). Indicates the original source of material for bacterial isolation. cTaken from Baker (2; and unpublished data).
a
b *
HFP 022801 HFP 013003 DDB 01310310 DDB 01020110 DDB 01360610 HFP 020203 Not defined DDB 02060120 DDB 02060210 DDB 07010210 DDB 07010310 ORS 020604 Not defined DDB 16060820 DDB 16110210 LLR 03013 LLR 02022 DAB 021002 DDB 02060510 RBR 162001 RBR 162002 RBR 162009 RBR 162013 RBR 162017 RBR 162019 RBR 162020 RBR 162021 Not defined
Corresponding author. 1172
2 1 1 1 1 2 2 4 3 1 1 3 2 3 4 3 3 2 3 3 3 ? 1 3 3 1 1 3
VOL. 56, 1990
NOTES
1173
FIG. 1. Composite autoradiograph of representative hybridization patterns obtained with the nifDH probe with numerous Frankia strains. Lanes: 1, CpI2b; 2, AvcIl; 3, ArI3; 4, AllIl; 5, JCT 287; 6, CeI2; 7, 55005; 8, MgI5; 9, Cellb; 10, RBR 162009; 11, RBR 162020; 12, AvcIl; 13, AllIl; 14, G2; 15, McI2b; 16, R100; 17, R43; 18, WgCcl.17. Lanes 8, 13, and 16 contained insufficient DNA to be resolved in this autoradiograph.
random primers labeling kit (Bethesda Research Laboratories). Membrane transfer and detection. Frankia DNA (3 to 5 ,ug well-) was digested with restriction endonuclease and electrophoresed on 0.8% agarose gels in TAE (40 mM Trisacetate, 2 mM EDTA, pH 8.0). DNA fragments were transferred to a Hybond-N nylon membrane (Amersham Corp.) by Southern blotting (12). Hybridization with the nifDH probe was performed overnight at 65°C. After hybridization and washing with lx SSC (150 mM NaCl, 15 mM sodium citrate, pH 7.0), the membrane was exposed to Kodak X-Omat film overnight at -70°C, using two intensifying screens.
Figure 1 illustrates typical DNA hybridization patterns when the restriction endonuclease Sall and the nifjH probe are used. A summary of our hybridization results is provided in Table 2. Three distinct hybridization patterns are evident. Pattern I, which corresponds to host specificity group 1 strains, had hybridization bands at 1.7 and 0.94 kb; pattern II, which corresponds to host specificity group 2 strains, had hybridization bands at 2.4 and 1.1 kb; and pattern III, which corresponds to host specificity group 3 and 4 strains, had a single hybridization band at approximately 3.5 kb with some higher or lower. Use of other restriction endonucleases (PstI, XhoI, or BamHI) resulted in hybridization patterns which, although useful in grouping strains, were more complex than the Sall patterns (not shown). Determination of host specificity among newly isolated Frankia strains is important for practical purposes. This is particularly so because, for unknown reasons, isolated strains may not have the capability of reinfecting the host genus from which they were derived (2, 3). From the Frankia strains currently held in pure culture, a limited understand-
ing of host specificity has been gained. Three major host specificity groups can be defined if the cross-inoculation data from the two promiscuous actinorhizal genera Gymnostoma and Myrica are disregarded. Briefly, these host specificity groups (HSG) are HSG 1, consisting of strains infecting Alnus and Comptonia spp.; HSG 2, composed of strains infecting genera of the family Casuarinaceae; and HSG 3, composed of strains infecting genera of the Elaeagnaceae. Gymnostoma and Myrica are infected by strains from all of these groups with the exception of a few strains which only nodulate members of the Elaeagnaceae. Baker (2) has defined the latter strains as "restrictive" and separated them into a fourth HSG. At present, the determination of host specificity for any new strain requires cross-inoculation tests TABLE 2. Groups defined by restriction fragment length
polymorphism (RFLP) analysis RFLP group I
(bands, 0.94 and 1.7 kb)
ArI3 ArI5 AvcIl AvsI3
CpI2b CpI3 31 77
89
RFLP group II (bands, 1.1 & 2.4 kb)
AllIl CcI3 CcOl JCT 287 1935
RFLP group III
(bands, 2.9-4.4 kb)
CelIlb CeI2 G2 McI2b MgI5 R2 R43
WgCcl.17 6 24 61 63
55005
1174
NOTES
conducted on a minimum of four host genera. Thus, a rapid in vitro assay to predict host specificity would be greatly beneficial. Frankia strains which belonged to HSG 1 and HSG 2 were particularly consistent in their hybridization patterns. Frankia strains belonging to HSG 3 or HSG 4 demonstrated hybridization of the probe to a single Sall fragment which varied slightly in size among the strains. The variability of fragment size observed with HSG 3 and HSG 4 strains would indicate greater genetic diversity among these strains than among strains within HSG 1 or HSG 2. A similar conclusion was drawn by Lalonde et al. (9) when Frankia strains were analyzed by using a number of chemical and physiological tests. Our work should have considerable practical value in initial screenings of newly isolated Frankia strains. Restriction fragment length polymorphism characterization could be conducted and necessary cross-inoculation tests could be reserved for only those strains likely to belong to a desired HSG. We thank Raanan Bloom and Dusty Ide for use of Frankia strains and DNA samples and F. M. Ausubel for plasmid pAS1. This research was supported by grant DPE-5542-G-SS-7707-00 from the U.S. Agency for International Development, Program for Science and Technology Cooperation.
LITERATURE CITED 1. An, C. S., W. S. Riggsby, and B. C. Mullin. 1985. Relationships of Frankia isolates based on deoxyribonucleic acid homology studies. Int. J. Syst. Bacteriol. 35:140-146. 2. Baker, D. D. 1987. Relationships among pure cultured strains of
APPL. ENVIRON. MICROBIOL.
Frankia based on host specificity. Physiol. Plant. 70:245-248. 3. Baker, D. D. 1989. Methods for the isolation, culture, and characterization of the Frankiaceae: soil actinomycetes and symbionts of actinorhizal plants, p. 213-236. In D. Labeda (ed.), Isolation of biotechnological organisms from nature. McGrawHill Book Co., New York. 4. Bloom, R. A., B. C. Mullin, and R. L. Tate. 1989. DNA restriction patterns and solution hybridization studies of Frankia isolates from Myrica pensylvanica (bayberry). Appl. Environ. Microbiol. 55:2155-2160. 5. Callaham, D., P. Del Tredici, and J. G. Torrey. 1978. Isolation and cultivation in vitro of the actinomycete causing root nodulation in Comptonia. Science 199:899-902. 6. Hintermann, G., R. Crameri, T. Kieser, and R. Hutter. 1981. Restriction analysis of the Streptomyces glaucescens genome by agarose gel electrophoresis. Arch. Microbiol. 130:218-222. 7. Holms, D. S., and M. Quigley. 1981. A rapid boiling method for the preparation of bacterial plasmids. Anal. Biochem. 114: 551-559. 8. Huang, J. B., Z. Y. Zhao, G. X. Chen, and H. C. Liu. 1985. Host range of Frankia endophytes. Plant Soil 87:61-65. 9. Lalonde, M., L. Simon, J. Bousquet, and A. Seguin. 1988. Advances in the taxonomy of Frankia: recognition of species alni and elaeagni and novel subspecies pommerii and vandijkii, p. 671-680. In H. Bothe, F. J. de Bruijn, and W. E. Newton (ed.), Nitrogen fixation: hundred years after. Gustav Fischer, Stuttgart, Federal Republic of Germany. 10. Nittayajarn, A., and D. D. Baker. 1989. Methods for the quantification of Frankia cell biomass. Plant Soil 118:199-204. 11. Normand, P., P. Simonet, and R. Bardin. 1988. Conservation of nif sequences in Frankia. Mol. Gen. Genet. 213:238-246. 12. Southern, E. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98: 503-517.
