Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Genetic diversity and symbiotic effectiveness of rhizobium isolated from perennial trifolium species Tesfaye, Mesfin 1997

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1997-250547.pdf [ 11.45MB ]
Metadata
JSON: 831-1.0088128.json
JSON-LD: 831-1.0088128-ld.json
RDF/XML (Pretty): 831-1.0088128-rdf.xml
RDF/JSON: 831-1.0088128-rdf.json
Turtle: 831-1.0088128-turtle.txt
N-Triples: 831-1.0088128-rdf-ntriples.txt
Original Record: 831-1.0088128-source.json
Full Text
831-1.0088128-fulltext.txt
Citation
831-1.0088128.ris

Full Text

GENETIC DIVERSITY AND SYMBIOTIC EFFECTIVENESS OF RHIZOBIUM ISOLATED FROM PERENNIAL TRIFOLIUM SPECIES by Mesfin Tesfaye B.Sc, Addis Ababa University, 1981 M.Ag.Sc, Lincoln University, 1989 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Plant Science We accept this thesis as conforming to the required standard THE UNIVERSItY^OF BRITISH COLUMBIA July 1997 © Mesfin Tesfaye, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of C V 7ZlVCZcr The University of British Columbia Vancouver, Canada •ate / f A^^.y^j- / <? ^ DE-6 (2/88) II Abstract Soil bacteria that form functional nodules on the genus Trifolium (clover), a taxa of about 240 plant species, are included in a single species, Rhizobium leguminosarum bv trifolii. The taxonomic criterion for this bacterial group is their ability to interact symbiotically with a narrow range of plant species originating from temperate regions, although several effectiveness groups which differ in their cross-inoculation patterns have been described. T. semipilosum Fresen (Kenya white clover), is one of the perennial clover species of agricultural importance in tropical and sub-tropical farming systems. The objectives of this study were to use nodulation characteristics, substrate utilization patterns and DNA analyses to determine the phylogenetic relationships of Rhizobium isolated from temperate and tropical perennial Trifolium species. Nodulation and symbiotic effectiveness of compatible and incompatible bacterial strains were investigated using the T. semipilosum host. The symbiotic association of 22 R. I. bv trifolii strains appeared to be highly specific: no single R. I. bv trifolii strain was able to nodulate all six perennial Trifolium species included as hosts in cross-inoculation experiments. Bacterial strains that were effective on temperate perennial species including T. repens, T. pratense, T. hybridum and T. fragiferum produced NodVFix" phenotypes on T. semipilosum and vice versa. This anomalous nodulation was characterized by the formation of significantly greater numbers of nodules, which were white and variable in size, but generally smaller than wild-type nodules. Substrate utilization analysis using the Biolog™ system suggested that R. I. bv trifolii strains effective on T. semipilosum may have a broader metabolic profile than strains effective on other Trifolium species. Genetic relationships of R. I. bv trifolii strains were obtained by DNA analyses using four polymerase chain reaction (PCR) based techniques: Randomly Amplified Polymorphic DNA (RAPD-PCR), Enterobacterial Repetitive Intergeneric Consensus (ERIC-PCR), and PCR- based nucleotide sequence analysis of 16S and 23S rDNA regions. A considerable level of genetic diversity was found using RAPD- and ERIC-PCR. Rhizobium I. bv trifolii strains that are effective on the tropical perennial clover, T. semipilosum, formed a tight cluster, especially with ERIC-PCR, that was distinct from R. I. bv trifolii effective on Ill temperate hosts. 16S rDNA nucleotide sequences were found to be highly conserved among R. I. bv trifolii. Comparative nucleotide sequence analysis of 23S rDNA regions clustered R. I. bv trifolii effective on T. semipilosum, T. repens, T. pratense, T. hybridum and T. fragiferum into two distinct groups, which were consistent with the pattern of symbiotic effectiveness observed in cross-inoculation experiments. The DNA sequences used as PCR primers for 23S rDNA analysis were found to be conserved among a wide range of rhizosphere bacterial species. Unique features identified by secondary structure analysis of the sequenced 23S rDNA region were used to design two 20-bp primers that provided group-specific differentiation and detection of Rhizobium by PCR. Detailed analyses using extracted DNA from many rhizosphere bacterial species confirmed the ultimate value of such group-specific primers for phylogenetic and ecological analyses. In compatible interactions, both T. semipilosum and T. repens were infected via root hairs. Furthermore, the anatomy of nodules induced by effective strains on T. semipilosum was analogous to that reported for other indeterminate nodules including T. repens. Nodules induced by effective strains on T. semipilosum were localized near the upper region of the tap root where fewer root hairs are located; nodules of T. repens were distributed largely on lateral roots. A Rhizobium strain ANU843, effective on T. repens, caused root hair branching and twisting on T. semipilosum, although infection threads were not detected in any of the plants examined. Microsymbionts for T. semipilosum and T. repens were transformed with a constitutively expressed gusA gene to provide a visual assay of rhizobial infection and nodulation. Strain ANU843 was shown to enter the root system of T. semipilosum mainly at the epidermal sites of emerging lateral roots. However, only 33% of the nodules from this incompatible interaction showed a positive GUS reaction. Of these, 70% were localized at the junction between the tap and lateral roots. iv TABLE OF CONTENTS Abstract ii LIST OF TABLES vi LIST OF FIGURES vii ACKNOWLEDGMENTS viii CHAPTER ONE General Introduction 1 1.1. Rhizobium classification 2 1.2. Molecular methods of classifying Rhizobium 4 1.3. The \egume-Rhizobium symbiosis 5 1.3.1. Plant factors 9 1.3.2. Infection and nodule formation 10 1.4. The clover symbiosis 13 1.5. Rationale 15 CHAPTER TWO Phenotypic and Genetic analysis of Rhizobium from perennial Trifolium species 17 2.1. Introduction 18 2.2. Materials and Methods 19 2.2.1. Plant species 19 2.2.2. Seed-surface sterilization and seed germination 20 2.2.3. Rhizobium strains and inoculum preparation 20 2.2.4. Cross-nodulation assays: Test for nodulation ability and symbiotic compatibility 24 2.2.5. Pattern of substrate utilization 24 2.2.6. DNA extraction 26 2.2.7. RAPD- and ERIC-PCR 26 2.2.8. rDNA sequence analysis 28 2.2.9. Nucleotide accession numbers 30 2.3. Results 30 2.3.1. Nodulation phenotypes 30 2.3.2. Pattern of substrate utilization 32 2.3.3. RAPD and ERIC-PCR 35 2.3.4. 16S rDNA sequence analysis 39 2.3.5. Comparative sequence analysis of 23S rDNA 43 2.4. Discussion 51 CHAPTER THREE Differentiation of clover Rhizobium groups by PCR amplification of 23S rDNA areas 57 3.1. Introduction 58 3.2. Materials and Methods 59 3.2.1. Bacterial Strains 59 3.2.2. DNA preparation 59 3.2.3. Direct amplification from Rhizobium cell cultures 59 3.2.4. Preparation of nodule extracts for PCR amplification 60 3.2.5. PCR Amplification 60 3.3. Results and Discussion 61 3.3.1. Design of group specific oligonucleotide primers 61 3.3.2. Specificities of oligonucleotide primers 64 3.3.3. Conservation of 20-bp rDNA sequence among Rhizobium species 67 3.3.4. Detection of clover Rhizobium from intact cells 71 CHAPTER FOUR Nodulation characteristics and symbiotic effectiveness of 7. semipilosum inoculated with compatible and incompatible microsymbionts 75 4.1. Introduction 76 4.2. Materials and Methods 76 4.2.1. Plant culture and inoculation using Leonard jar assemblies 76 4.2.2. Acetylene Reduction Assay 77 4.2.3. Plant DM and Nitrogen Analysis 78 4.2.4. Relative Effectiveness 78 4.2.5. Marking Rhizobium with the transposon GUS gene 79 4.2.6. Symbiotic Effectiveness Study 80 4.2.7. Microscopic Analysis 81 4.2.8. Nodule Number and Distribution 83 4.2.9. Data Analyses 84 4.3. RESULTS 85 4.3.1. Cross-inoculation experiments 85 4.3.2. Symbiotic effectiveness of T. repens and T. pratense 85 4.3.3. Symbiotic effectiveness of T. semipilosum 91 4.3.4. Nodulation responses of T. semipilosum by strains TAL910 andANU843 95 4.3.5. Assessment of marked Rhizobium strains for nodulation ability . 100 4.3.6. Root hair deformation and infection thread formation 102 4.3.7. Nodule number and distribution 107 4.3.8. Nodule anatomy 110 4.4. Discussion 114 vi CHAPTER FIVE Summary 120 References 127 LIST OF TABLES vii Table 2.1a. Bacterial strains used in this study 21 Table 2.1b. Nucleotide sequences used in this study 23 Table 2.2. Nodulation and effectiveness response of Trifolium spp. to inoculation with R. leguminosarum bv trifolii 31 Table 2.3. Substrate richness of Rhizobium strains included in this study 33 Table 2.4. List of substrates utilized by Rhizobium strains included in this study 34 Table 2.5. Average number of substrates used by Rhizobium from perennial clovers 36 Table 2.6. Sequence similarity values for the 23S rDNA sequences [%] 49 Table 3.1. PCR amplification reactions of Rhizobium and root-associated bacterial species 68 Table 4.1. Shoot nitrogen and carbon composition of nodulated T. repens and T. pratense plants at 35 days after inoculation. 89 Table 4.2. Dry matter and shootroot ratio of T. semipilosum plants at 35 days after inoculation 93 Table 4.3. Shoot nitrogen and carbon composition of nodulated T.semipilosum plants at 35 days after inoculation 99 VIII LIST OF FIGURES Figure 2.1. The relationship among R. leguminosarum bv trifolii strains based on their utilization pattern of 25 carbohydrate substrates 21 Figure 2.2. The relationship among R. leguminosarum bv trifolii strains based on their utilization pattern of nine amino acid substrates 38 Figure 2.3. Dendrogram derived from a matrix of RAPD-PCR 40 Figure 2.4. Dendrogram derived from a matrix of ERIC-PCR 41 Figure 2.5. Nucleotide sequence comparison of a 260-bp 16S rDNA region from Rhizobium and related bacteria 42 Figure 2.6. Phylogenetic relationship of Rhizobium and related bacteria based on 16S rDNA sequences 44 Figure 2.7. Sequence alignment of a 23S rDNA region from Rhizobium and related bacteria 46 Figure 2.8. Phylogenetic tree derived from the aligned sequences of a 23S rDNA region from Rhizobium and related bacteria 50 Figure 3.1.Potential secondary structure model of the 23S rDNA gene segment from consensus sequence of Rhizobium 62 Figure 3.2. Predicted secondary structures of the species- and strain-specific 23S rRNA sequences from Rhizobium 63 Figure 3.3. Schematic representation of the position of PCR primers along the 23S rDNA segment 65 Figure 3.4. Group-specific detection of clover Rhizobium by PCR using extracted DNA. 66 Figure 3.5. Restriction digestion analysis of the PCR product amplified from extracted DNA 66 Figure 3.6. Group-specific detection of clover Rhizobium by PCR using cell suspension directly from liquid culture 72 Figure 3.7. PCR amplification of clover Rhizobium using nodule tissue 73 ix Figure 4.1. Host-specificity and symbiotic effectiveness of Rhizobium strains 86 Figure 4.2. Effect of Rhizobium inoculation on shoot DM, root DM, whole plant DM and shoot:root ratio of T. repens and T. pratense 87 Figure 4.3. Total nitrogenase activity and specific nitrogenase activity of T. repens and T. pratense 88 Figure 4.4. Relative effectiveness of Rhizobium strains with T. repens and T. pratense 90 Figure 4.5. Canonical variate analysis 92 Figure 4.6. Relative effectiveness of Rhizobium strains with T. semipilosum. . . 94 Figure 4.7. Effect of Rhizobium inoculation on whole plant DM of T. semipilosum 96 Figure 4.8. Effect of Rhizobium inoculation on total nitrogenase activity of T. semipilosum 97 Figure 4.9. Effect of Rhizobium inoculation on specific nitrogenase activity of T. semipilosum 98 Figure 4.10. Host-specificity and symbiotic performance of Rhizobium strains marked with Tn5gusA11 and their parent strains (ANU843 andTAL910) 101 Figure 4.11. T. repens inoculated with GUS marked Rhizobium strain ANU843::gusA11 as seen under a dissecting microscope 103 Figure 4.12. Early steps of host microsymbiont interactions can be followed by chemical staining of ^-glucuronidase using GUS-marked strains 104 Figure 4.13. T. semipilosum inoculated with ANU843;;gusA11 and viewed by light microscopy after histochemical staining 106 Figure 4.14. Nodulation characteristics of T. semipilosum and T. repens inoculated with their respective compatible Rhizobium strains 108 Figure 4.15. Nodule number and distribution of T. semipilosum inoculated with a compatible and incompatible Rhizobium 109 Figure 4.16. Light micrographs of longitudinal section nodules induced by wild-type strain TAL910 on T. semipilosum 111 Figure 4.17. Light micrographs of longitudinal section nodules induced by incompatible strain ANU843 on T. semipilosum: at 16 and 21 days after inoculation 112 Figure 4.18. Light micrographs of longitudinal section nodules induced by incompatible strain ANU843 on T. semipilosum: at 21 and 30 days after inoculation 113 Figure 4.19. Light micrographs of longitudinal section nodules induced by incompatible strain ANU843 on T. semipilosum: nodules formed on tap roots, lateral roots and junction between tap roots 115 xi ACKNOWLEDGEMENT Grateful acknowledgement is expressed to the World Bank Graduate Scholarship Program for awarding me a fellowship for the first two years of my study. I would like to thank Dr. Brian Holl for allowing me to join his research group, for guidance during the research work and for helpful comments on the preparation and completion of this thesis without whom my thesis would have been but a dream. His help and support over the last four years was immense and also very grateful for the financial support over the years. I would also like to thank the other members of the supervisory committee, Dr. B. Ellis, Dr. J . Kronstad and Dr. C. Chanway, for their helpful suggestions throughout the program. Thanks to Dr. Ellis for the use of random primers, UBC Oligonucleotide set 100/1. I would also like to express my gratitude to Dr. R. Copeman for his help during photography using light microscopy and use of his lab. facility. I would also like to acknowledge the expert assistance and guidance from Dr. D. Petersen on DNA sequencing work, Dr. Naza Azizbekov on nodule sectioning, and Mr Leroy Scrubb on High Performance Liquid Chromatography. I owe many thanks to them. I am indebted to Dr. M. Djordjevic of the Australian National University, Australia, Dr. W. Rice and Dr. N. Lupwayin of the Agriculture and Agri-Food Canada, Beaverlodge Research Station, Canada, Dr. P. vanBerkum of the United States Department of Agriculture, Maryland, U.S.A., Dr. S. Long of the University of California, U.S.A., and Mr B. Martin, the Natrogen Fixation for Tropical Agriculture Legume Project, Maui, Hawaii, for providing Rhizobium strains. I would also like to thank Mr. W. Williams of the Margot Forde Germplasm Center, New Zealand and Dr. A. Bomkin of the Soil Science Department, University of British Columbia, Canada, for providing clover seed. I also thank the contribution from departmental staff: Ashley Herath, Derek White, Bev Busch, Robyne Alan, Seane Trehearne and Christia Roberts. To all friends and colleagues for their help and discussion; My thanks goes to Murali Srinivasan, Stefanie Butland, Suprayogi, Sam, Pat Warne, Bjorn Orvar, Shawn Wang, Debbie Wheeler, Solomon Shibairo, Upenyu Mazarura, Kevin Spoolen, Grant McKengey, Mirej Smudja, Amrita Singh, Nancy Furness and Daryl Louie. I also wish to thank Ato Getachew Worku, former Vice Minster of the Ministry of Agriculture, Ethiopia, for his encouragement and faith in my ablities. I would also like to thank my parents, in-laws, brothers and sisters for their encouragement and support during my study period. Most of all, I owe the most thanks to my wife Emawaysh Zike for her patience, understanding, love and support. CHAPTER ONE General Introduction 2 Rhizobium species are soil bacteria which are well known for their ability to infect the roots of leguminous plants to elicit the formation of a nodule, a symbiotic organ that can fix atmospheric nitrogen and make it available for plant growth (Brewin 1991; Hirsch 1992). The symbiotic nitrogen fixation process depends on the genetic compatibility between the rhizobia and the legume; effectively nodulated plants can fix substantial amounts of nitrogen. The most important symbiotic associations are those between Rhizobium and herbaceous legumes, which are important components of pastoral agricultural systems. Among the herbaceous legumes, the genus Trifolium (clover) is comprised of at least 240 species which are diverse in morphology, habitat and ecology (Williams 1987; Zohary and Heller 1984). Many Trifolium species have been widely cultivated in temperate and sub-tropical ecosystems and make a major contribution both to soil fertility and to the supply of high quality forage. 1.1. Rhizobium classification The term rhizobia refers to soil bacteria which are genetically diverse and physiologically heterogeneous. Taxonomically, rhizobia belong to the cr-sub-division of Proteobacteria consisting of three genera: Rhizobium, Bradyrhizobium (Jordan 1984) and Azorhizobium (Dreyfus et al. 1988). Species designations have been primarily based on effective nodulation of host plants. Alfalfa (Medicago sativa), Melilotus and Trigonella species are nodulated effectively by R. meliloti (Martinez 1994), although some isolates from alfalfa were shown to have the capacity to nodulate P. vulgaris (Eardly et al. 1985). Lotus species are 3 primarily compatible with R. loti (Jarvis et al. 1982), while the three groups of hosts for R. leguminosarum are clovers {Trifolium) for biovar trifolii, beans (Phaseolus) for biovar phaseoli, and peas (Pisum, Lathyrus), vetches (Vicia), and lentils (Lens) for biovar viciae. Additional species of Rhizobium have more recently been proposed including: R. fredii (Scholia and Elkan 1984), R. galegae (Lindstrom 1989), and R. huakuii (Chen et al. 1991). The host plants for these species are Glycine max and G. sq/'a (R. fredii), Galega officinalis and G. orientalis (R. galegae), and Astragalus sinicus (R. huakuii) respectively. The classification of rhizobia into groups and/or species based on their host range and specificity has provided a reasonably stable practical basis for taxonomy of Rhizobium. However, systematics based on cross-inoculation groups are gradually falling into disrepute, as the relationship between host range and phylogeny of Rhizobium has been shown to be more complex than previously reported (Lucas et al. 1995; Young and Johnston 1989). For example, Rhizobium isolated from Phaseolus nodules were found to be genetically diverse, and two additional species, R. etli (formerly R. leguminosarum bv phaseoli type I strains) (Segovia et al. 1993) and R. tropici (formerly R. leguminosarum bv phaseoli type II strains) (Martinez-Romero et al. 1991), have been proposed. Strains of both species have been shown to effectively nodulate Leucaena leucocephala in addition to Phaseolus vulgaris (Hernandez-Lucas et al. 1995). A variety of improved methods are now available to study genetic relationships among 4 Rhizobium strains. 1.2. Molecular methods of classifying Rhizobium Molecular techniques that evaluate both genetic variation and relatedness among Rhizobium and Bradyrhizobium species have been well established (Graham et al. 1991; Martinez-Romero 1994; Selenska-Pobell 1994). Approaches that have been used for the study of Rhizobium include PCR-based analysis of DNA sequences such as repetitive extragenic palindromic (REP) (de Bruijn 1992), and enterobacterial repetitive intergeneric consensus (ERIC) sequences (de Bruijn 1992; Judd et al. 1993), and restriction fragment length polymorphism (RFLP) analysis of chromosomal or 16S rRNA genes (Laguerre et al. 1994). de Bruijn (1992) showed that results from REP-and ERIC-PCR were in agreement with phylogenies derived from multilocus enzyme electrophoresis. Classification of genetically related strains of Bradyrhizobium japonicum serocluster 123 by the pattern of repetitive sequences was correlated with RFLP's (Judd et al. 1993). However, since repetitive DNA sequences are believed to be involved in recombination and amplification events, they may not be sufficiently stable to provide an effective source for taxonomic analysis (Flores et al. 1988). The amplification of genomic DNA using single primers of arbitrary sequences (AP-PCR and RAPD techniques) has also been applied to characterize several Rhizobium species (Harrison et al. 1992). However, the reproducibility of DNA fingerprints generated by RAPD analysis is highly sensitive to variations in reaction conditions (Abed et al. 1995), severely limiting the consistency of RAPD-5 derived phylogenetic trees and the comparative classification of new rhizobial isolates by different laboratories. Sequencing of large subunit ribosomal genes has provided not only alternative and reliable molecular markers to establish phylogenetic relationships, but also facilitated the identification of characteristic sequences which may be used to construct strain-specific probes (Lane 1991; Ludwig and Schleifer 1994). The polymerase chain reaction and DNA sequencing to determine complete or partial 16S rRNA (rDNA) followed by sequence alignment and comparison have been used to determine the phylogenies of many representative Rhizobium strains (Eardly et al. 1992; Graham et al. 1991; Huber and Slenska-Pobell 1994; Laguerre et al. 1993; Young et al. 1991). The increased size and complexity of 23S rDNA gene sequences have also been useful in providing specific probes for discrimination of other bacterial genera including Bradyrhizobium (Ludwig et al. 1994; Ludwig and Schleifer 1994; Springer et al. 1993). However, 23S rDNA sequence information from Rhizobium species is limited. Although these molecular methods are available for Rhizobium analysis, plant tests to determine the nodulation phenotype are nevertheless important to verify the host range predicted by characterization of the molecular relationships of nodule-forming bacteria. 1.3. The \egume-Rhizobium symbiosis The symbiotic interaction between rhizobia and host plants has been extensively reviewed (Brewin 1991; Denarie et al. 1992; Hirsch 1992), and the nodulation 6 process has been described in detail. The RhizobiumAegume symbiosis shows considerable specificity in that each rhizobial species has a defined host range (section 1.1.), and an effective symbiosis is predicated on the genetic compatibility between the host plant and the rhizobial partner. An exchange of signals between the legume plant and Rhizobium leads to root hair deformation, initiation of meristematic activity in the root cortex, entry of the bacteria into the plant and the ultimate formation of a root nodule (Brewin 1991). There appear to be at least two different phases in the early stages of the interaction at which nodulation specificity is determined. The process of nodule formation is initiated when mizobia detect and interact with compounds that are released by the roots of plants well before infection occurs (Innes et al. 1985; Mulligan and Long 1985; Redmond et al. 1986). The rhizobial nodD gene encodes a constitutively expressed regulatory protein that, in conjunction with plant factors, induces nod gene expression (Horvath et al. 1987; Mclver et al. 1989; Redmond et al. 1986; Rossen et al. 1985; Spaink et al. 1987). This interaction appears to mediate a first level of host-specificity in that different plants secrete different inducer substances (Djordjevic et al. 1987b; Hartwig et al. 1990; 1991; Hungaria et al. 1991a;b; Maxwell et al. 1989; Peters et al. 1986; Peters and Long 1988), and that nodD genes from different Rhizobium species recognize plant factors in a species-specific manner (Djordjevic et al. 1987b; Mclver et al. 1989; Peters and Long 1988). Compatible interaction leads to activation of Rhizobium nod genes, 7 commonly called common and host-specific nod genes, that are needed for the synthesis of return signals (Denarie et al. 1992; Long 1989). These return signals have been identified as Nod factors that stimulate root cortical cells to divide and prepare for the entry of the bacterial symbiotic partner (Spaink 1992). Mutations in the common nod genes abolish the deformation of root hairs, infection thread formation and induction of cortical cell division, indicating that these genes are essential for the early stages of nodule initiation in many systems (Djordjevic et al. 1985; Debelle et al. 1986; Long 1989). Mutations in the common nod genes can be complemented by DNA from other Rhizobium species. In contrast, mutations in individual host-specific nod genes do not abolish the ability to nodulate all compatible hosts and such mutations can not be complemented by DNA from another Rhizobium species (Denarie et al. 1992; Fisher and Long 1992; Spaink 1992). These genes appear to be required for nodulation to proceed at normal efficiency and/or modulate the host range. Several results indicate that a second level of host-specificity may be determined at the level of transcription of host-specific nod genes. Purified Nod factors have been shown to have a general type structure, although molecules from different Rhizobium species have structural modifications (different substituents) that influence their host specificity and biological activity (Denarie et al. 1992; Roche et al. 1991; Spaink et al. 1991). For example, wild-type R. meliloti produces sulphated Nod factors which regulate infection and nodulation of alfalfa. Mutations in nodH abolish the sulphation, resulting in strains which have lost the 8 ability to nodulate alfalfa, but have acquired the ability to infect and nodulate the normally incompatible host, Vicia (Roche et al. 1991). In contrast, R. meliloti nodQ" mutant strains can infect both M. sativa and Vicia sativa subsp. nigra (Roche et al. 1991). In another example, R. leguminosarum bv viciae strain 300, effective on pea (Pisum sativum L.) (Johnston and Beringer 1975), was able to cause root hair curling on white clover roots, but more slowly than a compatible R. leguminosarum bv trifolii strain ANU843 (Huang et al. 1995). However, in the incompatible interaction, no infection threads were formed and only limited cortical cell division was initiated 72 hours after inoculation. The addition of host-specific nod genes from R. leguminosarum bv trifolii to R. leguminosarum bv viciae strain 300 enabled the latter strain to induce infection threads and a normal pattern of nodulation on white clover roots, suggesting that the variety of Nod metabolites produced by these Rhizobium species may have been the key factor determining the process of nodule formation (Huang et al. 1995). Root nodules are classified into two developmental types: indeterminate and determinate (Brewin 1991; Hirsch 1992). The type of nodule development is dependent on the host plant (Bergersen 1978; 1982; Dart 1977; Newcomb 1981). Indeterminate nodules are characteristics of the genera Trifolium, Medicago and Pisum; they develop from the initial division of inner cortical root cells and have a persistent meristem (Bergersen 1978; 1982). In contrast, determinate nodules initiate from the division of subepidermal cortical cells and meristematic activity is limited to the early stages of nodule development. Plants of the genera Glycine, 9 Phaseolus, Lotus, Arachis and Vigna form determinate nodules (Bergersen 1978; 1982). 1.3.1. Plant factors The role of the plant in the \egume-Rhizobium association is to provide a surface for rhizobial development and to produce water-soluble and volatile carbon compounds as root exudates that may be used by the bacteria as carbon and energy sources for growth. Some compounds are produced at low concentrations and play a role as chemoattractants (Aguilar et al. 1988; Armitage et al. 1988; Caetano-Anolles et al. 1988; Dharmatilake and Bauer 1992) and as transcriptional signals in the communication between host plants and rhizobia (Djordjevic et al. 1987b; Hartwig et al. 1990; 1991; Hungaria et al. 1991a;b; Maxwell et al. 1989; Peters et al. 1986; Recourt et al. 1991; Redmond et al. 1986). Many of these regulatory compounds have been identified as low molecular-weight phenolic compounds (mainly flavonoids) that are synthesized via the phenylpropanoid pathway (Heller and Forkman 1988; Vickery and Vickery 1981). Different plant species secrete different flavonoid inducers that vary in their ability to induce or inhibit nod genes from different Rhizobium strains (section 1.3). Root exudation may occur as a result of mechanical injury from cultivation, abrasion by soil particles and/or by the rupture of tissues of the main root by the lateral or secondary roots as they emerge from the pericycle towards the surface (Hale et al. 1978). Contents of sloughed-off root cells can also supplement this 10 exudate (Hale et al. 1978). It has been suggested that Rhizobium inoculation stimulates production of nod-gene inducing substances contained in the root exudate of legumes (Dakora et al. 1993; Recourt et al. 1991; Rolfe et al. 1988). Nod-gene inducing compounds have been recovered from seed extracts, germinating seed rinses (Hartwig et al. 1990; 1991; Hungaria et al. 1991a), and exudates or extracts of plants grown primarily in axenic hydroponic culture (Djordjevic et al. 1987b; Hartwig et al. 1990; Hungaria et al. 1991b; Maxwell et al. 1989; Recourt et al. 1991; Redmond et al. 1986). Root exudates include a diverse array of naturally-occurring amino acids, organic acids, pentoses and hexoses, pyridmidines and puridines, vitamins and enzymes. The mechanism by which plants control and regulate the release of nod-inducing compounds into the rhizosphere is largely unclear. However, the increased accumulation of stimulatory signal molecules both in their number and concentration due to rhizobial inoculation may signify an attempt by the host to promote infection and nodulation by specific nodule-forming bacteria. 1.3.2. Infection and nodule formation In the early stages of host-symbiont interaction, Rhizobium causes root hair deformation (Bhuvaneswari and Solheim 1985; Dart 1977; Hirsch 1992), which may appear as tip swelling, lateral bulging, branching and curling, as observed in white clover (Bhuvaneswari and Solheim 1985); it may also be reflected in the formation of twists and spirals (Hirsch 1992). Root hair deformation usually occurs 6-18 hours after inoculation (Hirsch 1992); higher inoculum density, 5 x 108 11 bacteria ml"1, hastened the process to as early as 90 minutes after inoculation (Bhuvaneswari and Solheim 1985). Root hair deformation is a characteristic feature of most \egume-Rhizobium interactions (Turgeon and Bauer 1982; Yao and Vincent 1976), and successful nodulation is usually correlated with the ability of the Rhizobium to induce a marked root hair curling response (designated Hac+) on a compatible host plant. Across the spectrum of legume hosts, the route of infection by rhizobia may vary and includes entry through root hairs, through intercellular spaces in the epidermis or through crack entry (Bergersen 1978; 1982; Sprent 1979). The host plant appears to control the route of infection, because the same rhizobia can infect different host species by different routes, whereas a given host plant is normally infected by the same route regardless of the rhizobial strain (Young et al. 1989). In peanut (Arachis hypogea), root hairs are formed at the emergence points of lateral roots; rhizobia enter at the junction of the root hair and epidermal cells and nodules arise at lateral root emergence points (Bergersen 1982; Sprent 1979). Moreover, root hair curling in peanut is not accompanied by the formation of infection threads (Sprent 1979). In contrast, rhizobial infection in Trifolium and Medicago plants takes place through root hairs (Bergesen 1982). In the latter species, Rhizobium are guided to the emerging nodule tissue via infection threads (Djordjevic et al. 1985), whose cylindrical wall is deposited by the plant. Infection threads in plants forming indeterminate nodules are initiated from curled root hairs and grow towards the root cortex. Up to 100 infection threads 12 per plant have been reported for white clover (Djordjevic et al. 1985), although only a few such infections (10%-20%) result in nodule formation (Djordjevic et al. 1985; Hirsch 1992). The R. meliloti-M. sativa interaction also follows a similar pattern (Vasse and Truchet 1984). Inner cortical cells which are located in the path of growing infection threads dedifferentiate and a nodule primordium is initiated (Dudley et al. 1987; Libbenga and Harkes 1973), although this process begins before the infection thread is formed and can occur in the absence of root hair curling. Many reports have clearly indicated that diffusible signal molecules (Nod factors) produced by the Rhizobium are responsible for determining which host plant can be infected by a particular strain and for initiating the cascade of events in the host that lead to nodule organogenesis (Ardourel et al. 1994; Debelle et al. 1986; Roche et al. 1991; Spaink et al. 1991). The application of Nod factors in appropriate concentration to the roots of a compatible host induced root hair deformation (Roche et al. 1991; Spaink et al. 1991), formation of infection threads (Ardourel et al. 1994), as well as cortical cell division and the formation of nodule primordia (Spaink et al. 1991). In some host plants, Nod factor-induced development proceeds to the formation of nodules (Truchet et al. 1991), indicating that root nodule structure can develop without infection by the bacterial partner. It has been shown by Ardourel et al. (1994) that Nod factor structural requirements are more stringent for bacterial entry into root hairs than for the elicitation of plant developmental responses, such as tip growth in root hair and epidermal cells, and 13 the activation of the dedifferentiation program in cortical cells. The activity of the nodule meristem which is initiated distal to the nodule primordium accounts for further growth of the nodule (Libbenga and Harkes 1973). Proximal to the meristematic zone is an invasion zone where ramifying infection threads initiate a continual process of cell invasion. Bacteria are released into nodule tissue cells by endocytosis and occupy an organelle-like compartment, termed a symbiosome, surrounded by the peribacteroid membrane (Bergersen 1982). The bacteria inside these structures continue to proliferate into nitrogen-fixing bacteroids (Bergersen 1982). The formation of a subcellular compartment to house the bacteria inside the infected cell is essential for symbiosis; failure to form this membrane compartment or its premature disintegration, renders the association pathogenic (Djordjevic et al. 1987a). The enzymatic conversion of atmospheric nitrogen to ammonium is under the control of the rhizobial partner and nodule structures devoid of Rhizobium are ineffective. Anomalous nodulation resulting in the formation of ineffective nodules has also been described for mutant rhizobia that carry functional common nod genes but lack the ability to synthesize extracelluar polysaccharides (Kapp et al. 1990). 1.4. The clover symbiosis Rhizobium leguminosarum bv trifolii is the designation for all soil bacteria that are capable of forming highly specific symbiotic associations with plants of the genus Trifolium. The taxonomic criterion for this Rhizobium group is their ability to 14 nodulate effectively a narrow range of host species originating from temperate regions (Vincent 1974), although several effectiveness groups which differ in their cross-nodulation patterns (Burton 1980; Vincent 1974), and DNA relatedness (Jarvis et al. 1980) have been described. Trifolium repens L. (white clover) and T. pratense L. (red clover), predominant constituents of temperate pastures, respond similarly to rhizobial inoculation, (Brockwell and Katznelson 1976), although highly specific symbiotic interactions have been reported among cultivars (Brockwell and Katznelson 1976; Mytton 1975; Mytton and Livesey 1983). Mytton (1975), working with rhizobia isolated from four white clover cultivars, reported 25% more plant yield on average when cultivars were inoculated with strains isolated from their own nodules compared with strains isolated from different cultivars. Brockwell and Katznelson (1976), using 25 isolates of R. leguminosarum bv trifolii from Israel have also reported significant yield differences among ten Trifolium species and their respective microsymbionts. Rhizobium strains isolated from T. pratense were less effective on T. subterraneum than on T. pratense (Ferreira and Marques 1992; Robinson 1969) or T. fragiferum (Ferreira and Marques 1992). In contrast, clover species originating from sub-Saharan Africa appear to be incompatible with Rhizobium strains isolated from temperate clovers (Burton 1980; Vincent 1974). Furthermore, nodulation responses among clover species originating from sub-Saharan Africa were considerably more heterogeneous than those observed for temperate hosts (Burton 1980; Vincent 1974). 15 1.5. Rationale Burton (1980) and Vincent (1974) reviewed previous work on cross-inoculation and proposed up to 12 'effectiveness groups' or host sub-groups for Rhizobium-Trifolium nodulation responses. These groupings and other inoculation studies (Brockwell and Katznelson 1976; Mytton 1975) challenge the concept of homogeneity within the R. leguminosarum bv fr/'fo//7-clover symbiosis group, suggesting that distinct strain-host relationships may be present within the clover cross-inoculation group. Significant advances have been made in recent years in our knowledge and understanding of the symbiotic Rhizobium-Trifolium associations. Most of the information on nodule development in clovers has been derived almost exclusively from Trifolium species of temperate or Mediterranean origin:!, repens, T. pratense and T. subterraneum (Weinman et al. 1991). There are many agriculturally-important plant species for tropical and sub-tropical farming systems which have not been investigated. One example of such species is T. semipilosum Fresen (Kenya white clover), a perennial clover species native to Sub-Saharan Africa (Zohary and Heller 1984). There are published indications that Rhizobium strains capable of nodulating clover species belong to a genetically diverse group; DNA-DNA hybridization analysis by Jarvis et al. (1980) indicated that strains from T. semipilosum Fresen were less closely related to standard inoculant strains from Australia and New Zealand (effective with T. repens and T. subterraneum) than 16 are most strains from temperate clover species. More effective classification of clover rhizobia and increased understanding of their genetic relationships would enhance our ability to use these organisms appropriately in agriculture. The Rhizobium-T. semipilosum interaction is an attractive symbiotic system because of the apparent similarity of this clover to white clover. 7 semipilosum is reputed to be more drought tolerant than white clover (Mackay 1973), tolerant of acidic soils (Jones and Jones 1982), and represents a unique host sub-group classification (Burton 1980; Vincent 1974). Seed is also available commercially as a cultivar 'Safari' which has been developed in Australia. The objectives of this study were to examine Rhizobium from perennial Trifolium species in their host range and symbiotic effectiveness and to determine the phylogenetic position of Rhizobium from T. semipilosum relative to other rhizobia that are capable of nodulating temperate clover species. The specific objectives of the project were: * to screen c\over-Rhizobium compatibility using a range of Trifolium species and Rhizobium isolates, and to identify effective associations for the different c\o\/er-Rhizobium systems under controlled growth conditions, * to characterize the substrate utilization pattern of Rhizobium from 7. semipilosum, * to determine the phylogenetic position of Rhizobium isolates that are effective on 7. semipilosum relative to other Rhizobium isolates, and * to study the nodulation characteristics of 7. semipilosum with a compatible and an incompatible rhizobial strain. 17 CHAPTER TWO Phenotypic and Genetic analysis of Rhizobium from perennial Trifolium species 18 2.1. Introduction Rhizobium strains isolated from temperate clovers appear to be incompatible with Trifolium species from the tropics (Burton 1980; Vincent 1974). Despite these observations, all Rhizobium bacteria that nodulate Trifolium are included in a single species, R. leguminosarum bv trifolii. Within that classification several effectiveness groups which differ in their cross-inoculation patterns (Burton 1980; Vincent 1974), and DNA relatedness (Jarvis et al. 1980) have been described. A more definitive classification of this diverse and variable species would enhance our ability to use these organisms appropriately as agricultural inoculants. Molecular techniques that evaluate both genetic variation and relatedness among Bradyrhizobium and Rhizobium species have been well established (Laguerre et al. 1996; Martinez-Romero 1994). Genetic characterization of Rhizobium on the basis of PCR-based DNA banding patterns using RAPD and amplified repetitive DNA sequences (ERIC- and REP-PCR) is gaining acceptance in many laboratories. Nucleotide sequence comparisons of complete or partial 16S rRNA (rDNA) have also been applied to construct phylogenetic relationships of many Rhizobium species (Eardly et al. 1992; Huber and Selenska-Pobell 1994; Jarvis et al. 1992; Laguerre et al. 1993; van Berkum et al. 1996; Willems and Collins 1993; Young et al. 1991). Sequencing of ribosomal genes can also facilitate the identification of specific characteristic sequences which may be used in the construction of strain specific probes (Lane 1991; Ludwig and Schleifer 1994). In this regard, the 19 increased size and complexity of 23S rDNA gene sequences have provided specific probes for discrimination of other bacterial genera (Ludwig et al. 1994; Ludwig and Schleifer 1994; Springer et al. 1993). However, 23S rDNA sequence information from Rhizobium species is limited. This chapter contains details of studies on the phenotypic and genetic relationships of Rhizobium isolates from Trifolium species including strains effective on T. semipilosum, T. repens, T. pratense, T. hybridum, and T. fragiferum. Phenotypic characterization included an examination of patterns of cross-inoculation and patterns of substrate utilization by Rhizobium strains isolated from perennial Trifolium species. RAPD- and ERIC-PCR DNA fingerprinting as well as 16S/23S rDNA sequencing constitute the genetic approaches employed in this study to compare Rhizobium isolates from different Trifolium species with each other and with strains of R. meliloti and R. etli. Finally, the efficacy of the 16S/23S rDNA sequences to construct phylogenetic relationships among closely related strains was also evaluated using the data generated in this study and previously published 16S & 23S rDNA sequences from other members of the a-subdivision of Proteobacteria. 2.2. Materials and Methods 2.2.1. Plant species The host range and symbiotic associations of Rhizobium strains were studied using the following clover species as host plants: T. semipilosum Fresen cv. 20 Safari (Frank Sauer and Sons Ltd., Australia), T. repens L. cv. Ladino and T. pratense L. cv. Pacific (Richardson Seed Company Ltd., Canada), T. ambiguum L. cv. Monaro and T. fragiferum L. cv. Upward (Margot Forde germplasm centre, New Zealand) and T. hybridum L. (Dawson Seed Company Ltd., Canada). 2.2.2. Seed-surface sterilization and seed germination T. semipilosum seed was scarified by abrasion with sandpaper. Seed of other species was not scarified. Seed of each plant species was surface sterilized by soaking in 95% ethanol for 30 seconds, in 5% sodium hypochlorite solution for 4 minutes, followed by rinsing in several changes of sterile distilled water. Surface-sterilized seed was germinated aseptically on water agar plates (0.75% w/v) (Somasegaran and Hoben 1994). Two-day-old seedlings were placed aseptically onto growth medium for nodulation studies. 2.2.3. Rhizobium strains and inoculum preparation Rhizobial strains used throughout this study are listed in Table 2.1a. Except for those strains obtained from the ATCC, NifTAL Project or the USDA, strains used in this study were isolated from several nodulated host plants following procedures described by Vincent (1970). Presumptive tests, including Congo red and bromothymol blue, were carried out using single colony isolates (Somasegaran and Hoben 1994). Single colony isolates were maintained on yeast extract mannitol agar (YMA) slants stored at 4°C (Vincent 1970) or in 40% glycerol stored at -70°C. Table 2.1a. Bacterial strains used in this study. 21 Species Strain/Host Origin/Reference R leguminosarum bv trifolii TAL91 OITrifolium semipilosum TAL909/T. semipilosum TAL798/T. semipilosum TAL577/T. semipilosum ICMP1313/7". semipilosum ICMP2173/7. semipilosum ICMP4936/7". semipilosum ANU843/T. repens BCRC02/7; pratense BCSC826/7. subterraneum BCWC01/7". repens ATCC 14480/T. pratense E227077". burchellianum E2099/7". burchellianum E2154/7". semipilosum E2315/7". semipilosum E2228/7". semipilosum E2167/7". semipilosum E2260/7. semipilosum E2264/7". semipilosum E2169/7. semipilosum USDA2213/T. hybridum USDA2060/7". hybridum USDA2181/T, ambiguum Zimbabwe/ NifTAL3 Project / / H II II II II Kenya/ICMPb New Zealand 11 11 11 11 n 11 Tasmania Rolfe et al. 1980 B.CC.,Canada/ This study 11 11 11 I I ATCC d Ethiopia/ This study 11 11 11 11 11 11 11 I I 11 11 11 11 I I 11 USDA e l l 22 Table 2.1a. continued... Species Strain/Host Origin/Reference R. leguminosarum bv trifolii USDA2180/7". ambiguum USDA USDA2182/7". ambiguum I I n USDA2043/T. medium I I ANU845 nonnodulating variant of ANU843 Djordjevic et al. 1985 TAL910::gusA11 derivative of TAL910 This study ANU843::gusA11 derivative of ANU843 I I I I R.meliloti ATCC99307MeaYcao/o sativa ATCC I I Rm1021/M sativa Long 1985 I I NRG34 AAFC f n NRG85 I I R.etli TAL182/Priaseo/us vulgaris Hawaii/NifTAL Project I I USDA9032 (CFN42) USDA R.leguminosarum bv viciae NRG457 AAFC I I NRG480 n I I USDA2370 USDA R.leguminosarum bv phaseoli USDA2671 (RCR3644) l l R. tropicii USDA9030 (CIAT899) I I I I USDA9039 (CFN299) I I Root-associated bacteria B. polymyxa L6 B.C./Holl etal. 1988 B. polymyxa PW2 B.C./Shishido etal. 1995 B. megaterum A15 B.C./Srinivasan 1997 B. pumilus Bet 10a I I n B. azotoformans S08 I I M B. brevis S72 I I I I E. coli S17-1 /l-pir Wilson et al. 1995 a Nitrogen-fixation for Tropical Agriculture b International Collection of Micro-organisms from Plants c British Columbia dAmerican Type Culture Collection e United States Department of Agriculture ' Agriculture and Agri-Food Canada 23 Table 2.1b. Nucleotide sequences used in the study. Species Strain/Host Origin/Reference R. leguminosarum bv trifolii ATCC 14480 /T. pratense X672273 Willems & Collins 1993 I I T24 U31074 Briel & Triplett 1996 R.leguminosarum bv viciae LmZ/Lathyrus japonicus U08100 Drouin et al. 1994 I I Lp1013/L japonicus U08101 Drouin et al. 1994 RJeguminosarum bv phaseoli 8002/P. vulgaris M55494 Young et al. 1991 II RCR3644/P. vulgaris U29388 van Berkum et al. 1996 R. etli TAL182/P. vulgaris U28939 van Berkum et al. 1996 n CFN42/P. vulgaris U28916 van Berkum et al. 1996 R. meliloti ATCC 9930/M saf/Va M55242 Eardly et al. 1992 R. fredii USDA 205/Glycine max X67231 Willems & Collins 1993 R. loti NZP2213/lotos corniculatus X67229 Willems & Collins 1993 R. galegae HAMBI 540/Ga/egae orientalis X67226 Willems & Collins 1993 A. caulinodans ORS 571/Sesban/a rostrata M55491 Young et al. 1991 B.japonicum (a) USDA 110/G. max Z35330 Kundig et al. 1995 B.japonicum (b) DSM30131/G. max X71840 Springer et al. 1993 A.vitis (a) U45329 Otten & Ruff ray 1996 A.vitis (b) U28505 Otten et al. 1996 E. coli V00331 Brosius et al. 1980 a Genbank accession numbers 24 For inoculum preparation, strains were grown on yeast extract mannitol broth (YMB)(Vincent 1970) at 30°C on an orbital shaker. An early stationary phase culture in YMB at densities of approximately 1x108 cfu/ml was centrifuged and resuspended in sterile distilled water for inoculum preparation. 2.2.4. Cross-nodulation assays: Test for nodulation ability and symbiotic compatibility The objective of the initial study was to screen a large number of Rhizobium isolate-Trifolium species combinations for compatibility in nodulation and N-fixation. Two-day-old seedlings were grown aseptically on Petri plates (Bender and Rolfe 1985) or in NifTAL tubes (Somasegaran and Hoben 1994) into which deposit-free seedling medium (Jensen 1962) with 15 g agar L"1 was dispensed. Seedlings were inoculated by adding 0.25 ml of a liquid culture of the appropriate strain. After inoculation, both Petri plates and tubes were maintained horizontally overnight to allow the liquid inoculant to be absorbed into the agar. Seedlings were then oriented vertically in a growth chamber with a light intensity of 350 //mol-m"2-s"1, a 22°C:20°C day:night temperature cycle, and a 14-h photoperiod. Plates and tubes were arranged in a completely randomized design with 10 replicates for each treatment. Four weeks after inoculation, nodulation phenotypes were classified as Nod- when plants had no visible root nodules; Fix+ was assigned to plants which were healthy and contained deeply red pigmented nodules. All other responses including nodules which appeared white were categorized as Nod+. 25 2.2.5. Pattern of substrate utilization The Biolog GN™ microtiter plate (Biolog, Hayworth, California), which contains a set of 95 different carbon substrates and one control well with water, was used to study the substrate utilization pattern of Rhizobium isolates effective on T. semipilosum, T. repens, T. pratense, T. fragiferum, and T. hybridum. In this study, Rhizobium isolates were maintained on an R2A-agar (DIFCO laboratories, U.S.A.) minimal medium and stored at 4°C. Samples were prepared from an overnight culture on R2A-agar medium grown at 30°C and suspended in 1x A liquid media (Ausubel 1989) without sugar. Cell suspensions were standardized to an absorbance value of 0.2 to 0.22 at A 5 9 5 nm. A 150 pL sample was dispensed into each well and the microtiter plate was incubated covered for 72 hours at 30°C under static conditions. The degree of substrate oxidation was recorded by photometric analyses at A 5 9 5 nm using a Titertek®-Multiskan reader (Flow Laboratories). This study was repeated twice. The absorbance value of the control-well was subtracted from substrate values to obtain substrate-specific oxidation data. Negative values were treated as non-active and were assigned absorbance values of '0' for data analysis. The following parameters were evaluated using these substrate data: Substrate richness as defined by Vahjen et al. (1995) is the total number of substrate-wells with positive absorbance values after correction for the control-well. Total plate absorbance is the sum of the absorbance of all 95 substrate-containing wells and therefore an indicator of the overall metabolic activity of the 26 rhizobial isolate under consideration (Vahjen et al. 1995). A threshold value was calculated for each strain representing the total plate absorbance divided by the substrate richness (Vahjen et al. 1995). Utilized substrate-wells were those substrate-wells with absorbance values above the threshold (Vahjen et al. 1995). Cluster analysis was carried out on a binary matrix of presence (1) or absence (0) of utilized substrates. Dendrograms were constructed based on the normalized percent dissimilarity values as calculated by the average linkage method using SYSTAT for Windows (SYSTAT Inc., 1992). 2.2.6. DNA extraction Genomic DNA was extracted from Rhizobium isolates grown for 48 hours in YMB at 30° C on an orbital shaker. Template DNA for RAPD-and ERIC-PCR was extracted using the microwave technique described by Abed et al. (1995). The DNA purification procedure described by Jagadish and Szalay (1984) was used for rDNA sequence analysis. The resulting concentration and purity of DNA were assessed both spectrophotometrically (Genequant, Pharmacia, LKB) and visually (agarose gel electrophoresis) (Maniatis et al. 1982). 2.2.7. RAPD- and ERIC-PCR Primers (10-Mers) for the RAPD analysis were obtained from the UBC-RAPD primer synthesis project, Oligonucleotide set 100/1. ERICR (5'-ATGTAAGCTCCT-GGGGATTCAC-3') and ERIC2 (5-AAGTAAGTGACTGGGGTGAGCG-3') primer sequences were synthesized using a Beckman 1000M Oligo synthesizer (Versalovic et al. 1991). PCR amplification reactions (25 //I) for RAPD- and ERIC-27 PCR consisted of 100 ng DNA, 2.5 units Tag DNA polymerase with the corresponding 1x Taq buffer (Appligene), 1.2 mM of each dNTP and primer concentrations of: 250 pmol of the arbitrary primer or 50 pmol of each opposing ERIC primer. All PCR amplifications were performed in a Techne PHC-3 thermal cycler (Mandel Scientific Company Ltd.) at 40 cycles of 1-min denaturation at 94°C, 1-min annealing at 36°C and 2-min extension at 72°C for RAPD analysis (Harrison et al. 1992); or 35 cycles of 1-min denaturation at 94°C, 1-min annealing at 52°C and 1-min extension at 72°C for ERIC-PCR. An initial denaturation at 94°C for 1 min and a final extension for 5 min at 72°C were included for each reaction. RAPD- and ERIC-PCR amplification products were analyzed by electrophoresis in 1.5% (w/v) agarose gel using 1x Tris-acetate EDTA (TAE) buffer. The gels were stained in an aqueous ethidium bromide solution (0.5 mg/ml) and photographed under UV light. Bands with a similar molecular size were considered homologous and the presence (1) or absence (0) of a band at any position on the agarose gel was used to construct a binary matrix of RAPD or ERIC markers as described by Judd et al. (1993). Genetic distances between Rhizobium strains were calculated using the algorithm of Nei and Li (1979) as provided in the RAPDistance software package (Armstrong et al. 1994). Dendrograms were constructed based on the neighbor-joining method and NJTREE program (Saitou and Nei 1987). 28 2.2.8. rDNA sequence analysis 23S rDNA sequences from Rhizobium were amplified by using primers B A C 1 1 (5'-AGAGTGCGTAATAGCTCAC-3') and BAC19 (5'-CGGGTCTAGAACTTACCGACAA-GG-3') (Petersen et al. 1995). These primers correspond to positions 1086-1104 and 1941-1964, respectively, of the Escherichia coli 23S rDNA (Brosius et al. 1980) and were synthesized using a Beckman 1000M Oligo synthesizer. The PCR was carried out using the same reaction conditions described for the RAPD-and ERIC-PCR except that 20 pmol of each opposing primer were used for rDNA analysis. Amplification conditions were: 30 cycles of 1-min denaturation at 94°C, 1-min annealing at 52°C and 1-min extension at 72°C. An initial denaturation at 94°C for 1 min, and a final extension for 5 min at 72°C were included. PCR products generated by amplification of the 23S rDNA were subjected to electrophoresis in 1% (w/v) low-melt agarose gel in 1x TAE buffer containing ethidium bromide (0.2 //g/ml). The gels were photographed with UV illumination immediately after electrophoresis and the gel band containing the amplified DNA fragment was excised with a sterile scalpel. Excised gel slices were eluted by freeze/thaw extraction (Qian and Wilkinson 1993), followed by direct nucleotide sequencing in both directions using the B A C 1 1 and B A C 1 9 primers. Nucleotide sequencing was performed by the Nucleic Acid and Protein Sequencing unit of UBC using the Taq DyeDeoxy terminator cycle sequencing kit (Applied Biosystems, Inc.) and a model 371A DNA sequencer (Applied Biosystems, Inc.). 29 A 260-bp DNA sequence that encodes the 16S rDNA segment was determined by directly sequencing PCR products from Rhizobium bacteria using primers 27f (5'-AGAGTTTGATCMTGGCTCAG-3') and 342r (5-CTGCTGCSYCCCGTAG-3') (Lane 1991). These primer sequences were within the 16S rDNA region from position 8-27 and 327-342 of the E.coli 16S rDNA (Lane 1991) and were synthesized using a Beckman 1000M oligo synthesizer. Procedures for the determination of the 16S rDNA region were essentially the same as the 23S rDNA amplification and sequencing, except that the annealing temperature during the PCR amplification was increased to 56°C. Direct sequencing of the 16S rDNA fragment was performed using primer 342r. The resulting rDNA sequences and those of reference strains obtained from the GenBank databases were aligned and compared using the CLUSTAL option of the PC/GENE® software (IntelliGenetics Inc.). A matrix of Jukes-Cantor distances (Jukes and Cantor 1969) was used to construct unrooted phylogenetic trees using the neighbor-joining method of Saito and Nei (1987) with the NJTREE program (Armstrong et al. 1994) and the NEIGHBOR option provided with the PHYLIP software (Felsenstein 1995). In order to address topological errors in the unrooted trees by the neighbor-joining method, a series of 1000 bootstrap data sets of the same size as the original data were created using the SEQBOOT option and were analyzed by DNADIST, NEIGHBOR, and CONSENSE programs provided with the PHYLIP software (Felsenstein 1995). Phylogenetic relationships were also constructed by the unweighted pairwise grouping method of arithmetic 30 average (UPGMA) method using the PHYLIP software (Felsenstein 1995). 2.2.9. Nucleotide accession numbers The 23S rDNA sequences of strains in this study have been deposited in the GenBank database under accession numbers U47348 (TAL910), U47349 (Rm1021), U47350 (TAL909), U47351 (TAL798), U47352 (BCSC826), U47353 (ANU843), U47354 (BCRC02), U47355 (ATCC 9930), and U47356 (TAL182). GenBank accession numbers for 16S rDNA sequences are U76341 (ANU843) and U76342 (BCSC826). 2.3. Results 2.3.1. Nodulation phenotypes Table 2.2 summarizes the results of cross-inoculation experiments with 22 clover rhizobia on six Trifolium species, Medicago sativa, Pisum sativum and Glycine max. No nodules were formed on the non-clover hosts (Medicago, Pisum and Glycine). Among the clover hosts, degrees of specificity were reflected in nodulation responses ranging from no nodulation to completely effective nodulation. These results clearly indicated that specificity in the clover symbiosis was not limited to early nodulation but might also influence later stages of nodulation and/or establishment of effective nitrogen fixation. In associations involving T. semipilosum and T. ambiguum, the relationship was highly specific. Trifolium ambiguum was never nodulated by rhizobia isolated from other clover species. Rhizobia that were effective on T. ambiguum often 31 Table 2.2. Nodulation and effectiveness response of six clover species to inoculation with isolates of Rhizobium leguminosarum bv trifolii. Clover host1 Strain Semipilosum White Red Alsike Strawberry Caucasian TAL910 Fix+ Nod+ Nod* Nod+ Nod+ 0 TAL909 Fix+ Nod+ Nod+ Nod* Nod* 0 TAL798 Fix+ Nod* Nod* Nod* Nod* 0 TAL577 Fix+ Nod* Nod+ Nod+ Nod+ 0 ICMP1313 Fix+ Nod* Nod* Nod* Nod+ 0 ICMP2173 Fix+ Nod* Nod* Nod+ Nod* 0 ICMP4936 Fix+ Nod* Nod* Nod+ Nod+ 0 ANU843 Nod* Fix* Fix* Fix+ Fix* 0 ATCC 14480 Nod+ Fix+ Fix+ Fix+ Fix* 0 BCSC826 0 Fix+ Fix+ Fix+ Fix* 0 ATCC 14481 0 Fix+ Fix+ Fix+ Fix+ 0 BCRC02 0 Fix+ Fix+ Fix* Fix* 0 BCWC01 0 Fix+ Fix+ Fix+ Fix+ 0 E2270 0 . Fix+ Fix+ 0 0 0 E2093 0 Fix+ Fix+ 0 0 0 E2099 0 Fix+ Fix+ 0 0 0 USDA2060 0 Fix+ Nod+ Fix+ Nod+ 0 USDA2043 0 Nod+ Nod+ Nod+ Nod+ 0 USDA2126 Nod+ Nod+ Nod+ Nod+ Nod+ Fix+ USDA2134 Nod+ Nod* Nod* Nod+ Nod* Fix+ USDA2181 Nod+ Nod+ Nod* Nod+ Nod+ Fix+ USDA2182 Nod+ Fix+ Nod+ Nod+ Nod+ Fix+ Fix* = nodules contained deeply red-pigmented nodules. Nod+ = nodule-like structures lacked leghemoglobin. 0 = no nodules. The following species did not form nodules with any of the strains: Medicago sativa, Pisum sativum, Glycine max. 1Clover hosts: Semipilosum = T. semipilosum cv. Safari, White clover = T. repens cv. Ladino, Red clover = 7. pratense cv. Pacific, Alsike clove r= 7. hybridum, Strawberry clover = 7. fragiferum cv. Upward, and Caucasian clover = 7. ambiguum cv. Monaro. 32 produced small, white nodules on the other clover species tested. Strains that were effective on T. semipilosum showed the capacity to form ineffective nodules on all clover species except T. ambiguum. One Rhizobium strain originally isolated from T. medium formed nodules on T. repens, T. pratense, T. hybridum and T. fragiferum, but all of these associations were ineffective. This strain was obtained from the USDA and due to the lack of seed, the effectivity of this strain could not be authenticated with the original host. Nodulation phenotypes for T. repens, T. pratense, T. hybridum and T. fragiferum showed a high degree of cross-inoculation. 2.3.2. Pattern of substrate utilization The pattern of substrate utilization by 11 Rhizobium strains effective on T. semipilosum, T. repens, T. pratense, T. fragiferum and T. hybridum was studied by using the Biolog™ system. Results showing substrate richness and a list of substrates utilized by each strain are shown in Tables 2.3 and 2.4. A total of 77 substrate-wells produced positive absorbance values after correction for the control-well (Table 2.3), of which 57 were scored as "utilized" (Table 2.4). Overall, Rhizobium strains effective on T. semipilosum showed a greater number of substrates scored as utilized compared to strains effective on the other clover-hosts (Table 2.4). This was due to an increase in substrate richness by strains from T. semipilosum (Table 2.3). 33 Table 2.3. Substrate richness of Rhizobium strains included in this study. Carboxylic Amino Amines/ Carbohydrates acids acids Polymers Amides Misc. Total Strain (30) (24) (20) (5) (6) (10) (95) ICMP2173 3 29 11 16 4 6 10 76 ICMP4936 3 25 8 8 3 5 10 58 ICMP1313 3 24 8 13 5 5 10 65 TAL910 3 25 7 9 4 4 10 59 TAL909 3 21 6 7 4 2 6 46 TAL798 3 24 6 5 4 1 8 47 TAL577 3 22 4 5 4 2 7 44 BCSC826 11 6 4 5 3 8 37 BCRC02 8 3 2 2 1 8 24 BCWC01 24 4 3 4 1 7 43 ANU843 14 4 2 4 2 8 34 Numbers in parenthesis indicate the total number of substrates tested in each category. a- Rhizobium strains effective on T. semipilosum. 34 Table 2.4. List of substrates utilized by Rhizobium included in this study. ICMP ICMP ICMP Substrate /Category 4936 2173 1313 TAL910 TAL909 TAL798 TAL577 BCSC826 BCRC02 BCWC01 ANU8 Carbohydrates a-D-glucose + + + + + + + - - + -a-D-lactose lactulose - - - - - + - - - - -a-lactose - - - - - + - - - - -p-methyl D-glucoside - - - - - + - - - - -celloblose - - - - - + - - - - -D-arabltol + + - - - - - - - - -D-fructose + + + + + + + - - + -D-galactose - - - - - + - - - - -D-mannitol + - - - - + - + - - -D-mannose - + + - - + - + - + -D-melibiose - - - - - + - - - - -D-raffinose - - - - - + - - - - -D-sorbitol - - - - - - - - - + -D-trehalose - - - + - + - - - - -i-erythritol •- - - - - - - - - + -L-arabinose + + + + + + + - + + + L-fucose + + + + + - + - - + -L-rhamnose + - + + + - - - - - -maltose - - - + + + - - - + -methyl pyruvate + + + + + - + - + + + mono-methyl succinate + + + + - - - + + + + psicose - - - - - - - - - + -sucrose - - - - - + + - - - -turanose + - + + + - - - - - -xylitol + - - + + - + - - - -Amino Acids D-alanine + + - - - - - - - - + glycyl-L-glutamic acid - - - - - - - - + + -L-alanine + + + + + - + - - - -L-alanyl-glycine - - + - - - - - - - -L-asparagine - - + + - + - - - - -L-aspartic acid + + + + + + - - - - -L-histidine + + + + + - - - - - -L-pyroglutamic acid - - - - - - - - - + -L-serine - - - - - - - + - - -Carboxylic Acids p-hydroxy butyric acid + + - - • - - - - - - -D,L-lactic acid + - + - - - - - - - -D-galacturonic acid lactone + - + - + - - - - + -D-gluconic acid - - - - - + - - - - -formic acid + - + + + - + - - - + y-hydroxy butyric acid - - - - - - - + - - -succinic acid + + + + + + + - + + + Amines / Amides alaninamide + + + + - - - - - - -succinamic acid + + + + + + + + + + + Polymers a-cyclodextrin - + - - - - - - - - + dextrin + + + + + + + - + + -tween40 - - - - - - + + - + -tween80 - - - - - - - + - - -glycogen + - + - - + - - - - -Miscellaneous 2,3-butanediol - - - - - - - - - - + bromosuccinic acid + + + + + + + + + + + D,L-a-glycerol phosphate - - - - - - - + - - -glycerol + - - - - + - + - - + inosine - + + + - + - - - - -thymidine - - - - - + - - - - -uridine + + - + - + - - - - -urocanic acid + + + + + - + - + - -35 The 95 substrates were categorized into six guilds following the classification by Zak et al. (1994). The pattern of greater substrate utilization by strains effective on T. semipilosum was reflected in most of the six substrate groups (Table 2.5). Cluster analyses constructed based on the use pattern of twenty-five carbohydrate substrates showed four phenotypic clusters (Figure 2.1). Clusters I and IV contain only strains effective on T. semipilosum, whereas clusters II and III contain all strains effective on T. repens, T. pratense, T. fragiferum and T. hybridum. A dendrogram derived from the use pattern of nine amino acid substrates also produced four phenotypic clusters (Figure 2.2). Clusters I and II contain all but one of the strains effective on T. semipilosum; although TAL577 was clustered with the strains ANU843 and BCSC826. Cluster IV contained strains BCRC02 and BCWC01. 2.3.3. RAPD and ERIC-PCR Preliminary screening of twenty arbitrary primers (10-Mers) for their ability to amplify DNA from representative strains that are effective on T. pratense, T. repens, T. hybridum and T. fragiferum (ATCC 14480), T. semipilosum (TAL910) and Medicago sativa (ATCC 9930) identified a primer sequence "R82" (5-GGGCCCGAGG-3') , which has 90% G+C content, as the most useful distinguishing primer. This primer was subsequently used to amplify DNA from all strains used in the RAPD analysis. 36 Table 2.5. Average number of substrates used by Rhizobium isolates from perennial clovers. Numbers in parenthesis indicate substrate richness (average). Rhizobium isolates effective on Substrate category T. semipilosum other clovers* Carbohydrates 10.0(24.3) 5.0(14.3) Amino acids 3.3(9.0) 1.3(2.8) Carboxylic acids 2.9 ( 7.1) 1.5 ( 4.3) Amines/Amides 1.6 ( 3.6) 1.0 ( 1.8) Polymers 1.7(4.0) 1.5(3.8) Miscellaneous 3.4 (8.7) 2.3 (7.8) *T. repens cv. Ladino, T. pratense cv. Pacific, T. hybridum, and T. fragiferum cv. Upward. 37 D i s s i m i l a r i t y [%] I " l 1 1 1 100 75 50 25 0 I ICMP2173 I TAL910 I TAL909 • ICMP1313 ' ICMP4936 i- I TAL577 • BCWC01 r- ANU843 L. BCRC02 I BCSC826 TAL798 Figure 2.1. The relationship among Rhizobium leguminosarum bv trifolii strains based on their pairwise utilization patterns of 25 carbohydrate substrates using the Biolog™ system. The dendrogram was based on cluster analysis of normalized percent dissimilarities using the average linkage method. 38 D i s s i m i l a r i t y [%] I 1 1 50 25 0 BCWC01 BCRC02 BCSC826 TAL577 ANU843 TAL798 — ICMP1313 — TAL910 — TAL909 i - ICMP2173 L ICMP4936 Figure 2.2. The relationship among Rhizobium leguminosarum bv trifolii strains based on their pairwise utilization patterns of nine amino acid substrates using the Biolog™ system. The dendrogram was based on cluster analysis of normalized percent dissimilarities using the average linkage method. 39 The genetic relationship among eight clover rhizobia and one R. meliloti strain were determined with banding patterns produced by RAPD- and ERIC-PCR. The products of the amplification reactions consisted of 14 bands of different sizes ranging from 202 to 2500 bp long. On the basis of the presence/absence matrix from the RAPD-PCR data, the isolates were divided into three clover rhizobia groups (Figure 2.3). Analysis of the ERIC-PCR data produced a similar number of groups (Figure 2.4). However, the composition of the groups defined by the two techniques was not identical. Furthermore, the intensity of bands produced by RAPD-or ERIC-PCR varied between different reactions. 2.3.4. 16S rDNA sequence analysis Partial 16S rDNA sequences containing 260 nucleotides were obtained from three strains (TAL910, TAL909 and ICMP2173) that nodulate T. semipilosum and two strains (ANU843 and BCSC826) that nodulate T. repens, T. pratense, T. fragiferum and T. hybridum. The sequences which were identical for the five strains tested (Figure 2.5) were located within a region previously used to determine phylogenetic relationships among rhizobia (Eardly et al. 1992; Jarvis et al. 1992;Young et al. 1991). The 16S rDNA sequences for previously described strains of R. leguminosarum bv trifolii ATCC 14480 (GenBank accession number X67227), R. leguminosarum bv trifolii T24 (U31075), R. leguminosarum bv viciae (U08100 & 1)08101), R. leguminosarum bv phaseoli (M55494, U29388 & M55234), R. meliloti ATCC 9930 (M55242), R. ef//TAL182 (U28939), R. loti (X67229), R. fredii (X67231), R. galegae (X67226), Azorhizobium caulinodans 40 P a t r i s t i c distance 0.00 0.135 0.265 0.395 0.530 ATCC9930 BCRC02 TAL90 9 BCWC01 E2093 Is o l a t e d from Medicago sativa r BCSC826 T. subterranei T. pratense T. semipilosum T. repens ATCC144 8 0 T. repens TAL798 T. semipilosum TAL577 T. semipilosum T. burchellianum Figure 2.3. Dendrogram derived from a matrix of RAPD-PCR, constructed by the neighbor-joining method (Saitou and Nei 1987) using RAPDistance software (Armstrong et al. 1994). The patristic distance (Nei and Li 1979) provides a horizontal scale that may be used to calculate the distance between species in the tree; the shorter the distance between selected pairs the closer the relationship. 41 P a t r i s t i c distance ( | I 1 I I I 0.00 0.14 0.28 0.42 0.56 r- BCSC826 '— BCRC02 E2093 TAL9 0 9 TAL7 98 L TAL577 BCWC01 ATCC14480 Iso l a t e d from T. subterraneum T. pratense T. burchellianum T. semipilosum T. semipilosum T. semipilosum ATCC9930 Medicago sativa T. repens T. repens Figure 2.4. Dendrogram derived from a matrix of ERIC-PCR, constructed by the neighbor-joining method (Saitou and Nei 1987) using RAPDistance software (Armstrong et al. 1994). The patristic distance (Nei and Li 1979) provides a horizontal scale that may be used to calculate the distance between species in the tree; the shorter the distance between selected pairs the closer the relationship. 42 CONSENSUS AGGCITAA.CACATGaW3TCGAG03**CcccgCAAgggG* TAL910 **-CCCG---GGG-** TAL909 **_CCCG---GGG-** ICMP2173 **_CCCG—-GGG-** ANU843 **_CCCG---GGG-** BCSC826 **-CCCG---GGG-** ATCC14480 **-CCCG—-GGG-** Rlt T24 **_CCC*---GGG-** ATCC9930 **_CCCG---GGG-** TAL182 **_CCCG---GGG-** B.japo(a) GG-GTAG TAC-TC A.vitis (a) **_CTCG---GAG-** CONSENSUS AAACTtGtgcTAATACCGtATgtGtCCIT^ TAL910 T-TGC T--GT-T CG—A—- GTCA-G GA —-TG—- Q A TAL909 T-TGC T--GT-T CG—A—- GTCA-G GA —-TG—- Q A ICMP2173 T-TGC T--GT-T CG--A GTCA-G GA TG Q A ANU843 T-TGC T--GT-T CG--A GTCA-G---GA TG Q A BCSC826 T-TGC T--GT-T CG--A GTCA-G---GA TG ; Q A ATCC14480 -- T-TGC T--GT-T CG—A--- -GTCA-G GA —-TG—- Q A Rlt T24 ---T-TGC—- T--GT-T----CG- -A--- GTCA-G GA —-TG—- Q A ATCC9930 T-TGC T--AA-C CG—G—- GGAA-G GA —-TG—- Q A TAL182 T-TGC T- -GT-C TG- -G GTAA-G CG —-TG—- Q A B.japo.(a) -- T-TGC G--AA-C AC--G--- CCGA-A CG ---CT—- c A.vitis(a) -- G-AAT T--AC-C CX3--G--- GGGT-T GA —-TG— - Q A CONSENSUS AGGOGACX^ TCcaTAGCIX3GTCTGAGAG TAL910 TAL909 ICMP2173 ANU843 BCSC826 ATCC14480 Rlt T24 ATCC9930 TAL182 B.japo.(a) A.vitis(a) Figure 2.5. Nucleotide sequence comparison of a 260-bp 16S rDNA region from seven ft leguminosarum bv trifolii strains, ft meliloti, ft etli, B. japonicum and Agrobacterium vitis. Nucleotide sequences that are conserved are indicted as (-), while sequence gaps are shown as (*). Previously published nucleotide sequences are included for reference. B. japo (a) = Bradyrfiizobium japonicum and A. vitis (a) = Agrobacterium vitis. 43 (M55491), B. japonicum (Z35330 & X71840) and two A. vitis strains (U45329 & U28505) were aligned and compared pairwise with the five sequences obtained in this study. Interestingly, the 16S rDNA sequences in this study were also identical with a nucleotide sequence of R. leguminosarum bv trifolii strain ATCC 14480 determined previously by Willems and Collins (1993). Sequences from R. etli TAL182 showed nine nucleotide differences from all clover rhizobia. The sequence determined for the R. meliloti strain ATCC 9930 (Eardly et al. 1992) also showed nine nucleotide differences from the clover rhizobia. The unrooted phylogenetic tree based on the Jukes-Cantor distance matrix was constructed by the neighbor-joining method (Saito and Nei 1987). Nucleotide sequences from 22 strains were included in the analysis. The five clover rhizobial sequences from this study were clustered with other biovars of Rhizobium leguminosarum (trifolii, viciae and phaseoli) (Figure 2.6). This cluster was clearly separated from other Rhizobium, Bradyrhizobium and Agrobacterium species (Figure 2.6). The phylogenetic tree constructed by the UPGMA method (Felsenstein 1995) produced similar tree topology to that of the neighbor-joining method (data not shown). 2.3.5. Comparative sequence analysis of 23S rDNA The partial nucleotide sequences of the amplified 23S rDNA region, which included corresponding portions of domains II, III and IV of the E. coli 23S rDNA gene (Ludwig and Schleifer 1994), were determined for ten Rhizobium strains, of which four strains were effective on T. semipilosum and three were effective on T. repens, T. pratense, T. hybridum and T. fragiferum. 44 P a t r i s t i c distance 0.000 0.009 100 94 96 66 57 I 0.018 59 100 89 95 98 0 . 027 0 . 036 •B.japonicum(a) •B.japonicum(b) Azorhizobium caulinodans • R.loti R.galgae C A.vitis (a) A.vitis (b) R.meliloti (ATCC9930) R.fredii 47 t R.l.bv trifolii (T24) R.l. bv phaseoli (8002) R.l.bv t r i f o l i i (ATCC14480) R.l.bv viciae ICMP2173 * TAL909 * TAL910 * R.l.bv t r i f o l i i (ANU843) * R.l.bv t r i f o l i i (BCSC826) * R.l.bv viciae R.l.bv phaseoli (RCR3644) R . e t l i (TAL182) R.etli (CFN42) Figure 2.6. Phylogenetic relationship of seventeen Rhizobium species and selected Bradyrhizobium, Azorhizobium and Agrobacterium species, based on a 260-bp fragment of the 16S rDNA sequence. Strains included in the 23S rDNA sequence analysis in this study are indicated by stars (*). The tree was constructed from a Jukes-Cantor distance matrix by the neighbor-joining method (Saitou and Nei 1987). The numbers at the nodes are the bootstrap values for the nodes (based on 1000 bootstrap samplings). The patristic distance (Nei and Li 1979) provides a horizontal scale that may be used to calculate the distance between species in the tree; the shorter the distance between selected pairs the closer the relationship. Previously published sequences included in the analysis are listed in Table 2.1b. 45 The sequences were compared pairwise with each other, with an E. coli sequence, and with sequences from two strains each of Bradyrhizobium japonicum (Kundig et al. 1995; Springer et al. 1993) and Agrobacterium vitis (Otten et al, 1996; Otten and de Ruffray, 1996). The terminal nucleotides of each aligned sequence correspond to positions 1125 and 1929, respectively, of the E. coli 23S rDNA sequence (Brosius et al. 1980). However, the amplified product from all Rhizobium strains was only 674 nucleotides long, and when compared with E. coli, all sequences from this study, including Bradyrhizobium and Agrobacterium, revealed large alignment gaps within domains III and IV (Figure 2.7). Nucleotide sequences of the seven clover Rhizobium were polymorphic, clustering into two main groups. The sequences obtained for strains TAL910, TAL909, TAL798 and ICMP2173 were identical (clover Rhizobium group A), as were the sequences for strains BCRC02 and BCSC826 (clover Rhizobium group B). The other R. leguminosarum bv trifolii strain ANU843, had five polymorphic sites with clover Rhizobium group B compared to 23 polymorphic sites with clover Rhizobium group A. When ANU843 was included in group B, the two clover Rhizobium groups had an average 96.9% sequence similarity (Table 2.6). With 95.4% sequence homology, R. ef//TAL182 was more closely related to the clover Rhizobium group A (TAL910, TAL909, TAL798, and ICMP2173) compared to an average of 93.6% sequence similarity with the clover Rhizobium group B (BCSC826, BCRC02, and ANU843). ! 6 6 6 L * * -X * . t, t, t* t. E H fn ^  ^ * * * * * * * * * * * * I g g g i ^ ^ ^ f ^ l l ^ l 6 6 ^ ^ § 6 < < E H E H J 6 6 6 6 6 S 8 6 * * * * * i * * <d <C < i«C i « <q <; <q < i eg ! * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * i * 0 i—I ro w> ro 01 H O M Cl CN H ( T l C N r H O O ' t f O O O H O H ro w> co m H o o j r n N H O K N H C O ^ O O O H U o m U c o H n J O H ro ro o> H o CN] ro C N H C T i C N l H O O ^ O O O H U o c h U o p H r o i - q ^ 8 1 ^ 8 O H ro <£> ro (T\ H O O J ro CN H O K N r l C O ^ C D O H p o c f t U c p H r o J < £ H PQ E H P W O rH ro vo ro CT\ H O CN ro CN H O \ ( N H 0 0 < t 0 0 O H p o o i u r a d n J § g ^ 8 i S 5 8 <SHCQ<eEHPW 47 ATCC9930 GTCAGGTAGAGTATACGAAGGCHI^^ Rml021 GTGOITAGAGIATACCAAGGQ^^ TAL910 GTCAGCTAGACTATACO^AGGCGCIT^ BCSC826 GTO^GGTAGAGTATAO^AGGO^^ ANU843 GTCAGGTAGSVGTTATACXIAACGQGC^^ TAL182 GICAGCTAGAGTATACOiAGGGGCITGAG?^ CCJK3ACCCCATA******* DSM3 0131 GACTGCTAGAGTATACCAAGGa^^ ECOLI GTCAGGTAGAGAATACCAAGGQGCIT^ ATCC993 0 ******CT*AOGCAACTAGGATGGGCT**^ Rml021 ******CT*A0GCAAGTAG<3ATG^ TAL910 ******TC*AGGCAACTCTIGI03GCT***GGCAC^ BCSC826 ******CT*AGGCAACTATTATGC«^^ ANU843 ******TT*ACX3CAACTATGATGGG^ TAL182 ******TG*AGGCAACTCTIGTG^ DSM30131 ******CTQ3CGCAAGOGGGC^^ ECOLI TGAC<JTCOCTOGOoG^ ATCC9930 TAGGCTCTG^OGCCiraX!GGn^^ Rml021 TAGGGTCTG&OGCCTGCCC^^ TAL910 TAGQGTCTGACmJimXQGTGC^^ BCSC826 TAGGGTCTGACX^OCTGCCQ^^ ANU843 TAGGGTCTGACGOCIGGCOGffl^ TAL182 TAGGGTCICACGCCTGC^^ DSM30131 TAGGGICTGACGCOXKICC^ ECOLI TACGGTGTG&QGCCnmXG^^ ATCC9930 Rml021 TAL910 BCSC826 ANU843 TAL182 DSM30131 ECOLI CGGTTCCTAAG CGGTTCCTAAG CGGTTCCTAAG CG*TTCTTAAG GG*TTCTTAAG CGGT*CCTAAG CGGT*CCTAAG CX3GTCCTAAGG Figure 2.7. Sequence alignment of a 23S rDNA region from ten Rhizobium strains. The sequences for TAL910, TAL909, TAL798 and ICMP2173 were identical as were sequences for BCSG826 and BCRC02. The published homologous sequences from Bradymizobium (DSM30131), and £ coli (V00331) are included for references. Sequence gaps as identified by the PC/GENE® software are shown as (*). 48 There was only one nucleotide difference between the two R. meliloti strains. Overall, the three taxa represented in this study, R. leguminosarum bv trifolii, R. meliloti and R. etli, can be differentiated by their 23S rDNA sequences (Figure 2.7), although highly homologous regions (92-96% sequence identity) were evident among them (Table 2.6). Not surprisingly, the Rhizobium sequences showed a relatively low degree of relatedness with Bradyrhizobium (79% sequence identity) and Agrobacterium (81% sequence identity). A phylogenetic tree based on the Jukes-Cantor distance matrix was constructed by the neighbor-joining method (Saito and Nei 1987) using nucleotide sequences from 14 strains. Rhizobium species formed a tight cluster, but with sufficient variation to classify the strains used in the analysis into five major groupings: two groups of clover Rhizobium, a R. meliloti group, a Bradyrhizobium group, and an Agrobacterium group (Figure 2.8). A phylogenetic tree from the Jukes-Cantor distance matrix and UPGMA (Felsenstein 1995) from nucleotide sequences of those 14 strains produced a similar tree topology as that of the neighbor-joining method; it indicated a first branch comprised of B. japonicum that was distinct from other strains and a second branch diverged into two branches, with Agrobacterium strains being distinct from the rest of the Rhizobium species (data not shown). 49 Table 2.6. Sequence similarity values for the 23S rDNA sequences [%]. Strain 2 3 4 5 6 7 8 9 10 11 12 13 14 1. ATCC9930 99.9 92.6 92.6 92.6 92.6 92.9 93.5 92.9 91.8 90.9 91.0 79.3 79.6 2. Rm1021 92.7 92.7 92.7 92.7 93.0 93.6 93.0 92.0 91.1 91.1 79.4 79.7 3. TAL910 100 100 100 97.2 96.3 97.2 97.8 92.3 91.3 80.0 80.3 4. TAL909 100 100 97.2 96.3 97.2 97.8 92.3 91.3 80.0 80.3 5. TAL798 100 97.2 96.3 97.2 97.8 92.3 91.3 80.0 80.3 6. ICMP2173 97.2 96.3 97.2 97.8 92.3 91.3 80.0 80.3 7. BCSC826 99.3 100 95.7 93.5 92.3 80.3 80.6 8. BCRC02 92.9 94.8 94.4 93.2 80.5 80.8 9. ANU843 95.7 93.5 92.3 80.3 80.6 10. TAL182 90.7 89.6 79.1 79.3 11. A. vitis (a) 98.8 80.8 80.9 12. A. vitis (b) 81.2 81.4 13. B.japonicum (a) 99.3 14. 6. japonicum (b) i 50 0 . 000 P a t r i s t i c d i s t a n c e 0 . 0435 0 . 087 100 100 I — B.japonicum (a) B.japonicum (b) A. v i t i s (a) — A. vitis (b) 100 82 99 100 100 R.meliloti(ATCC9930) R.meliloti(Rml021) 100 ICMP2173 - TAL7 98 - TAL90 9 TAL910 R. etli (TAL182) 92 100 |-i?.2. bv t r i f o l i i (BCSC826) •R.l. bv t r i f o l i i (BCRC02) — R.l. bv t r i f o l i i (ANU843) Figure 2.8. Phylogenetic tree from the aligned sequences of a 23S rDNA fragment from Rhizobium species and related bacteria. The tree was constructed from a Jukes-Cantor distance matrix by the neighbor-joining method (Saitou and Nei 1987). The numbers at the nodes are the bootstrap values for the nodes (based on 1000 bootstrap samplings). The patristic distance (Nei and Li 1979) provides a horizontal scale that may be used to calculate the distance between species in the tree; the shorter the distance between selected pairs the closer the relationship. Previously published sequences included in the analysis are listed in Table 2.1b. 51 2.4. Discussion Rhizobium strains that formed effective nodules on T. repens, T. pratense, T. hybridum and T. fragiferum failed to produce effective symbiotic associations with T. semipilosum and vice versa. These results are consistent with earlier observations of separate effectiveness groups within the Rhizobium-Trifolium species interactions (Burton 1980; Vincent 1974). The results from the Biolog™ system also suggested that strains effective on T. semipilosum may have a broader overall metabolic profile compared to Rhizobium strains effective on T. repens, T. pratense, T. hybridum and T. fragiferum. Although cluster analysis of the Biolog™ data based on carbohydrate or amino acid substrates alone did not produce identical clusters, phenotypic clusters based on carbohydrate substrates reflected clear-cut strain differentiation along the host sub-group categories of the strains included in this study. The pattern of substrate utilization has been used previously for numerical taxonomic purposes, and the Biolog™ system was developed for that purpose (Bochner 1989). It is not yet known whether the failure of a given strain to produce oxidation values on a given substrate under these experimental conditions denotes an absolute inability for substrate utilization by that strain. Nevertheless, the Biolog™ test results appear to correlate with the different genetic backgrounds of Rhizobium isolates that are effective on different clover hosts. More direct genetic analysis of Rhizobium strains from different effectiveness groups of the Rhizobium-Trifolium interaction appears to be 52 promising. The comparison of 16S rDNA sequences indicated that all the isolates effective with T.semipilosum, T. repens, T. pratense, T. hybridum and T. fragiferum were identical in nucleotide sequence to that of the Type strain R. leguminosarum bv trifolii (ATCC 14480) effective on T. repens, T. pratense, T. hybridum and T. fragiferum. The results of this study substantiate the view that PCR-based DNA banding patterns and 23S rDNA sequencing offer alternative tools to discriminate closely related root-nodule forming bacteria. A relationship between host specificity and genetic divergence of clover strains from different effectiveness categories has been revealed by PCR-based DNA banding patterns and 23S rDNA sequence analysis, illustrating the ability of both techniques to assess genetic diversity and to differentiate strains. The utility of PCR-DNA banding pattern analysis using arbitrary or ERIC primers to determine genetic relationships among nodule-forming organisms has been demonstrated previously with many strains including R. leguminosarum bv trifolii (Harrison et al. 1992; Strain et al. 1994), R. leguminosarum bv. viciae (Strain et al. 1994), R. meliloti (de Bruijn 1992), and Bradyrhizobium (Judd et al. 1993). However, the variability of RAPD markers between different laboratories has limited its acceptance (Abed et al. 1995; Laguerre et al. 1996). Consistent with that view, the results of this study using both primer techniques were not fully coincident with each other, failing to identify a conserved lineage for the R. meliloti Type strain (ATCC 9930), which was included as a control. However, both PCR-DNA banding patterns defined polymorphic differences that could be 53 used to group the clover Rhizobium included in this study. The dendrogram derived from ERIC-PCR analysis showed that strains from T. semipilosum formed a tight cluster that was distinct from Rhizobium strains that are effective on other Trifolium species. Results from this study confirm an earlier observation, based upon DNA homology, that clover rhizobia appear to be highly heterogeneous (Jarvis et al. 1980). Furthermore, the array of amplification products from clover rhizobia using ERIC primers is consistent with the earlier observation by Versalovic et al. (1991) that ERIC sequences are ubiquitous in gram-negative bacteria. The utility of 16S and 23S ribosomal regions for genetic analysis is gaining wider acceptance due to their combination of both highly conserved and highly variable regions (Lane 1991; Ludwig and Schleifer 1994). However, the 16S rDNA nucleotide sequences among strains effective on T. semipilosum, T. repens and T. pratense were identical to each other. Although the 16S rDNA region was used previously to examine Rhizobium phylogeny (Eardly et al. 1992; Segovia et al. 1993; Young et al. 1991), previous results indicated little sequence divergence among Rhizobium species and almost none within species (Eardly et al. 1992; Jarvis et al. 1992). Phylogenetic relationships from the 16S rDNA analysis substantiated the discriminating power of and the potential utility of the 23S rDNA segment as an alternative to traditional Rhizobium species classification. In this study it was possible to characterize and distinguish Rhizobium, Agrobacterium and Bradyrhizobium strains by sequence differences within a hypervariable region 54 of the 23S rDNA. The neighbor-joining method provides a topological relationship among the sequences without taking account of the evolutionary rate (Saito and Nei 1987), whereas UPGMA has a clock. The phylogenetic trees developed with these methods produced similar tree topologies, indicating that the phylogenetic relationship determined by the distance matrix methods were reliable, as suggested by Nei 1987, and the Rhizobium phylogeny constructed by using the Jukes-Cantor distances from the nucleotide sequence analysis of 23S rDNA region was relevant. Furthermore, the resulting phylogenetic relationships for the Rhizobium isolates, based on their 23S rDNA sequence analysis, are consistent with previously described taxonomic relationships for these species (Jordan 1984), reflecting the differentiation of the microsymbionts of alfalfa (ATCC 9930 & Rm1021) from those of clover (ANU843, BCSC826, BCRC02, TAL910, TAL909, TAL798 & ICMP2173) and bean (R. etli strain TAL182). The separation of rhizobia isolated from T. semipilosum from strains effective on T. repens, T. pratense, T. fragiferum and T. hybridum using 23S rDNA sequence analysis is noteworthy as it is consistent with the pattern of symbiotic effectiveness in the cross-inoculation data. Because of a low level (45%) of DNA relatedness between strains from T. semipilosum and two reference strains of R. leguminosarum bv trifolii, Jarvis et al. (1980) suggested that strains from the tropical clover, 'Safari', were less closely related to the standard inoculant strains for T. repens than are most strains from temperate clovers. In contrast, an average of 75% DNA 55 homology was observed between 20 strains from temperate clovers and a reference strain of R. leguminosarum bv trifolii. Based on the partial 23S rDNA sequence analysis, the three taxa represented in this study, R. leguminosarum bv. trifolii, R. meliloti and R. etli, were clearly more closely related to each other than to the strains of B. japonicum. The potential utility of the 23S rDNA sequencing as an alternative tool for characterization of strains was also shown by the two Bradyrhizobium strains which showed a distinct lineage from both Rhizobium and Agrobacterium clusters. The relatively low sequence similarity between Rhizobium and B. japonicum strains is consistent with a separate evolutionary origin, or early divergence of the two genera (Sprent 1994). The primer set, BAC11 and BAC19, used in this study for 23S rDNA sequencing has been shown previously to amplify DNA from a number of root-associated Bacillus species as well as Azorhizobium strains (Petersen et al. 1995), suggesting that BAC11 and BAC19 primer sequences may be conserved among a wide range of bacterial species and are therefore useful for PCR amplification and sequencing of the 23S rDNA region from many Rhizobium species. No appreciable differences were obtained in the size of the PCR product from Rhizobium as viewed in gel electrophoresis. However, there were large alignment gaps within domains III and IV when Rhizobium strains and the two previously published sequences for Bradyrhizobium were compared to the analogous E. coli sequence. This difference suggests that the amplified product 56 from the Rhizobium strains included in this study may have undergone deletion(s) in the sequenced region, a phenomenon which appears to be common within the members of Proteobacteria (Ludwig and Schleifer 1994). Several features of the sequenced 23S rDNA region are variable and some appear to be species- and/or strain-specific sequences. Diagnostic probes designed for these regions are likely to be useful to analyze nitrogen-fixing populations, or to track and recover released inoculum strains. 57 CHAPTER THREE Differentiation of clover Rhizobium groups by PCR amplification of 23S rDNA areas 3.1. Introduction Host species-Rhizobium strain interactions have been reported in clovers (Chapter 2; Mytton 1975) and it has been suggested that herbage productivity from clover-based pastures may be improved by selecting highly compatible partnerships (Mytton 1975). In soils with indigenous Rhizobium populations, however, inter-strain competition often makes it difficult for a new inoculant strain to establish in a high proportion of nodules (Crush 1987). Detection methods are required to monitor inoculant strains and to study their fate in the environment. Relatively simple, reliable and rapid procedures are desirable for testing the persistence and saprophytic competence of inoculant strains in soil. The most common means of detecting specific bacteria has been to use a selective medium for recovery of spontaneous antibiotic-resistant derivatives of the parent strain. This method has some advantages and is still often used in conjunction with newer methods. There is now a wide variety of additional methods to track microbes in the environment (Drahos et al. 1986; Sessitsch et al. 1996; Wilson et al. 1995). The potential use of introduced marker genes, such as the E. coli gusA (Sessitsch et al. 1996; Wilson et al. 1995), lacZ (Drahos et al. 1986) or celB (Sessitsch et al. 1996), in monitoring genetically engineered Rhizobium in the environment has been described (Wilson et al. 1995). Ludwig (1994) also demonstrated the application of rRNA probes using dot blots for specific detection of non-engineered Bradyrhizobium strains. The comparisons of partial 23S rDNA nucleotide sequences from 59 Rhizobium described in Chapter 2 revealed divergent regions that may represent species- and/or strain-specific sequences. The application of PCR using oligonucleotide primers directed to those group-specific unique rDNA sequences would seem an attractive option to develop a rapid and reliable system to differentiate Rhizobium. The objective of this work was to develop a PCR assay which would facilitate group-specific differentiation of Rhizobium capable of nodulating different clover species. 3.2. Materials and Methods 3.2.1. Bacterial Strains Bacterial isolates used for designing and evaluating the utility of the group-specific primers were included in Table 2.1a. Rhizobium isolates were maintained on YMA slants stored at 4°C (Vincent 1970). 3.2.2. DNA preparation Bacterial isolates were grown for 48 hours on YMB for Rhizobium (Vincent, 1970) or combined carbon medium for root-associated bacterial species (Rennie 1981) at 30°C on an orbital shaker. Genomic DNA was extracted using the technique described by Jagadish and Szalay (1985). The resulting concentration and purity of DNA was assessed by a Genequant spectrophotometer (Pharmacia, LKB). 3.2.3. Direct amplification from Rhizobium cell cultures Cultures were grown overnight in YMB (Vincent 1970) to a concentration of approximately 108 cfu/ml. One ml of each culture was centrifuged at 4°C and the 60 resulting pellet was resuspended in sterile distilled water. Approximately 106 cells in a 6 //L volume was added to each amplification reaction. 3.2.4. Preparation of nodule extracts for PCR amplification T. semipilosum and T. repens plants were inoculated with their respective compatible Rhizobium strains TAL910 and ANU843. The plants were grown under gnotobiotic conditions in Magenta™ jars in a growth chamber as described in Chapter 2. Nodules were collected six weeks after inoculation and stored at -80°C. Nodules were crushed individually in 50 //I sterile distilled water in a 1.5 ml_ microcentrifuge tube and centrifuged. An aliquot (6 //L) of the supernatant was used for each PCR amplification reaction. 3.2.5. PCR Amplification PCR amplifications, with either 25 //L (extracted DNA) or 30 //L (direct amplification from cell cultures) reaction mixtures, were carried out using an initial denaturation at 94°C for 1-min, 29 cycles of 1-min denaturation at 94°C, 30-sec annealing at (50-52°C) and 1-min extension at 72°C and final extension for 5 min at 72°C. The initial denaturation period during PCR involving samples from cultures and nodule samples was extended to 5-min at 94°C to accommodate cellular breakdown. Samples containing template DNA from strains TAL910 and BCSC826, representative strains of the two clover Rhizobium groups, and a sample lacking template DNA, were included in every run as controls. The amplification products were electrophoresed in a 1.5% agarose gel that contained ethidium bromide in 1x Tris EDTA buffer and photographed under UV light. 61 3.3. Results and Discussion 3.3.1. Design of group specific oligonucleotide primers The partial nucleotide sequences of the 23S rDNA region were determined for 10 Rhizobium strains using BAC11 and BAC19 primers (Chapter 2). The sequenced region was contained within helices 45 to 69 (helix numbering according to E. coli, Gutel and Fox 1988) and included both highly conserved and highly variable divergent sequences. A secondary structure model of the sequenced 23S rDNA region from Rhizobium was developed by comparison with a previously published model for E. coli (Gutell and Fox 1988), using the method of Zuker and Stiegler (1981) as provided with the PC/GENE® software (Figure 3.1). Where the aligned sequences varied, the integrity of the proposed helical regions for E. coli was maintained by considering base changes downstream. Five regions of the model (designated A-E in Figure 3.1) were found to vary substantially among Rhizobium species. Among these variable regions, A (helix 45 of E. coli numbering), D (helices 55-59 ) and E (helix 63) exhibited both species- and strain-specific differences (Figure 3.2). The presence of such a variation in region D was the basis for designing two group-specific reverse primers, SAC1 (5-CCTGAT-AATCCAATAAACCA) and WRC1 (5'-CCATCCAATCCAATAAGAAT-3'), which could be exploited to distinguish clover Rhizobium from two different effectiveness (Burton 1980; Vincent 1974) and genetic groups as was revealed in Chapter 2. A A A , U A A C a a a a' . ac , U A „ • A A Q C a II Q A A 0 c c A U a c A c c a A 0 c u « A A o 0 > . ° A U ° . U a a A c • a c . a c • a * . a A • U a' c a . c u A U u ° A / U C Q U Q O Q • • • • AUCC • a * c c A U u O o a U G A O C A O Q A A U U C Q U C C A A U A A u u u u Q . c c • a A • U c • a a c u u A U a a u u ' u a A a c c U U O O a c U A ° 0 U O O U C A C C A a A c A a ' _ * A a y a A c o a A U U Q C C A u a c c , ° A A U Q c c a' p. . QC c c u c a A A u A A a B U • A U c a c • a ' A A i a A • U u • a a . c " a o A Q A O A_ A A . C • C • a*-a a u c A A A * a * a a-A < 0 ° A U A a a a . c u a . A . u 0 ° °A * A a u IV A A 0 C U . A A 0 0 . C Q . C A A Q A C A A C a . c C • 0 a • c U • A U . A a • c c • 0 A _ Q. u & °u A * a u* A U.A U . A „ 0 . C c • a a u . a c C O c a a . u C O Q C Q . c a . c a a u A *U' c a u • A C A A A C A JA uuuau 0 u . a , c • a ' C O a . c C O . Q O U A C C A U A . A c a u • u • a . c - _ a . u . A •  A  A • U • a* O C Q 0 A ' U a A C A U C U A u Q u u a Q A A A g Aa ocoua C M U C C Q C A C A A A a u o ' a • a o c a A "A 5 a Q Figure 3.1. Potential secondary structure model of the 23S rDNA gene segment from a consensus sequence of ten Rhizobium strains (Figure 2.7 of Chapter 2). Positions which are variable among the strains are represented as (.). Regions that are variable among the sequenced Rhizobium species are indicated as A-E. The prediction was done by the method of Zuker and Stiegler (1981) as provided with the PC/GENE® software in comparison to the previously published model of Gutell and Fox (1988). The structure was drawn using the CARD software (Winnepenninckx et al. 1995). Region A 63 C A G A A 0 U • A G • C U • A a • c u • a A • u ATCC 9930 C • G U • A G • C C • G A • U U • G A • U TAL910.TAL909, TAL798 ICMP2173 BCSC826, ANU843, BCRC02 Region D U G U G U G U G U U u u u u C • G . _ U • A U • G C * G G • u u • A * • y G - C A • U U • A C • G C • G U • G U - G U • G U . G G • C G • C A U - G U - G G ' C c - G G • C BCSC826, BCRC02 TAL910, TAL909, ANU843 ATCC 9930 TAL798, ICMP2173 G • C G . C C • G C • G U- G G . C U . G Region E C A G A C A C A G A G A G - C C - G C - G G - C U - A U ' A L! U C U U U C • G U • G U Q • U " * G U - A U - G U-.A U - A u u A • u A - u * • y C - G C - G C - G C - G C - G C - G C - G C - G A - 0 A - U C - G U - G C - G C - G C - G C - G C - G C - G A • U A • U TAL910, TAL909, BCSC826, BCRC02 ANU843 ATCC 9930 TAL798, ICMP2173 Figure 3.2. Predicted secondary structures of the species- and strain-specific 23S rRNA sequences from ten strains of clover Rhizobium and one R. melliloti strain (ATCC9930). The sequences correspond to regions A, D, and E of Figure 3.1 and are analogous to helices 45, 58 and 63 of E. coli sequence (Gutell and Fox 1988). 64 While SAC1 primer sequences were specific for Rhizobium strains (TAL910, TAL909 & TAL798) effective on the tropical clover T. semipilosum, WRC1 primer sequences were designed for strains (ANU843, BCSC826 & BCRC02) effective on temperate clovers such as T. repens and T. pratense. The positions of the group-specific reverse primers along with the positions of universal primers, BAC11 and BAC19 are given in Figure 3.3. 3.3.2. Specificities of oligonucleotide primers Identification and specific detection of organisms by PCR would require a high specificity of the oligonucleotide primers for which they are designed. The specificities of the reverse primers, SAC1 or WRC1 , in combination with the forward primer, BAC11, were tested using eight clover strains from two strain effectiveness groups of the tropical and temperate clover species. Both primers, SAC1 and WRC1 , gave amplification products of the expected size (approx. 370 bp), and unequivocally identified the clover strain groups for which they were designed (Figures 3.4a, b), indicating that clover Rhizobium could be differentiated by using these primers. Further evidence that the PCR product contained the correct sequence was provided by restriction digestion with Taq\, a unique restriction site in the amplified region. Digestion of the PCR product with this restriction enzyme produced the expected fragments, one of approximately 80 bp and the other of 290 bp (Figure 3.5). 1125 1931 - h -Universal Amplification BACll - >>- - III BAC19 - -<< -Group-specific B A C H Amplification ->>--S A C I or WRCl III Figure 3.3. Schematic representation of the position of PCR primers along the 23S rDNA segment. Primers BAC11 and BAC19 are universal primers allowing amplification from many Rhizobium and root-associated bacterial species tested. Primers SAC1/BAC11 and WRC1/BAC11 produce group-specific amplification by PCR. E1125 and E1931 indicate the corresponding nucleotide positions in E. coli 23S rDNA (Brosuis et al. 1980). Empty spaces represent gaps that were found during sequence alignment (see Figure 2.7). 6 6 Figure 3.4. Group-spec i f i c detect ion of c lover Rhizobium by P C R using (A) S A C 1 / B A C 1 1 and (B) W R C 1 / B A C 1 1 primers from extracted D N A . 1 kb marker (M), T A L 9 1 0 (lane 1), T A L 9 0 9 (lane 2 ) ,TAL798 (lane 3), T A L 5 7 7 (lane 4), B C S C 8 2 6 ( l a n e 5), A N U 8 4 3 (lane 6), B C R C 0 2 ( lane 7), B C W C 0 1 (lane 8), and the no-template P C R negat ive control ( lane 9). Numbers at the left indicate band s ize in kb. M 1 2 3 4 0.5 0.08 Figure 3.5. Restr ict ion digest ion ana lys is of the 370-bp P C R product ampl i f ied by W R C 1 / B A C 1 1 primers from extracted D N A . Taq I restriction e n z y m e w a s used fol lowing the manufacturers instructions. 1 kb marker (M), und igested P C R product from B C S C 8 2 6 (lane 1), restriction e n z y m e d igested B C S C 8 2 6 ( lane 2), undigested P C R product from A N U 8 4 3 (lane 3), and restriction e n z y m e d igested A N U 8 4 3 D N A (lane 4). Numbers at the left indicate band s ize in kb. 67 3.3.3. Conservation of 20-bp rDNA sequence among Rhizobium species Since only a few strains representing a limited range of Rhizobium species have been sequenced, the degree of nucleotide sequence similarity within the 23S rDNA region among Rhizobium species in general, and R. leguminosarum bv trifolii isolates in particular, is unknown. High sequence similarity of the target DNA region chosen for primer annealing is a prerequisite for PCR to serve as an effective identification tool. To infer that there is some degree of conservation within the region used for PCR amplification, SAC1 or WRC1 primers in combination with the forward BAC11 primer were tested with extracted DNA from other Rhizobium and root-associated bacterial species, (i) R. leguminosarum bv trifolii strains A total of 28 clover strains isolated from eight different Trifolium species were tested. The results using extracted DNA showed that 14 of the clover strains tested gave amplification products of the expected size with the SAC1/BAC11 primer combination (Table 3.1). All of these strains were isolated from and/or fully effective with T. semipilosum only. These strains did not produce amplification products with the WRd /BAC 11 primers. The WRC1/BAC11 primer combination successfully generated the expected product with the remaining 14 of the clover Rhizobium strains tested; these strains were isolated from and/or fully effective with T. repens, T. pratense, T. hybridum, T. subterraneum, T. medium, T. ambiguum or T. burchellianum (Table 3.1). 'Table 3.1. PCR amplification reactions of Rhizobium and root-associated bacterial species using group-specific SAC1/BAC11 and WRC1/BAC11 primers. 68 P C R amplification with primers Strain BAC11/SAC1 BAC11/WRC1 BAC11/BAC19 R. leguminosarum bv trifolii TAL910 3 + - + TAL909 3 + - + TAL798 3 + - + TAL577 3 + - + ICMP1313 3 + - + ICMP2173 3 + - + ICMP4936 3 + - + E2154 3 + - + E2315 3 + - + E2228 3 + - + E2167 3 + - + E2260 3 + - + E2264 3 + - + E2169 3 + - + ANU843 b - + + BCWC01 b - + + ATCC14480 0 - + + BCRC02 C - + + BCSC826 d - + + E2270 6 - + + E2099 6 - + + USDA2213' - + + USDA2060 f - + + USDA2181 9 - + + USDA2 1 80 9 - + + USDA2 1 82 9 - + + USDA2043 h - + + R. etli TAL182 •+• - + USDA9032 (CFN42) + - + 69 Table 3.1. continued... P C R amplification with primers Strain BAC11/SAC1 BAC11/WRC1 BAC11/BAC19 R. leguminosarum bv viciae USDA2370 + + NRG457 + + NRG480 + + R.leguminosarum bv phaseoli USDA2671 (RCR3644) + + R. tropicii USDA9030 (CIAT899) - + USDA9039 (CFN299) - + R. meliloti ATCC9930 - + Rm1021 - + NRG34 - + NRG85 - + Root-associated bacteria* B. polymyxa - + B. polymyxa - + B. megaterum - + B. pumilus - + B. azotoformans - + B. brevis - + Clover-hosts and/or Rhizobium isolated from: a=T. semipilosum; b=T. repens; C=T. pratense; d=T. subterraneum; e=T. burchellianum; '=7. hybridum; 9= T. ambiguum; and h=T. medium * PCR analysis using BAC11/BAC19 primers on root-associated bacterial species was from Petersen et al. (1995). 70 (ii) . Sym-plasmid cured R. leguminosarum bv trifolii strain Strain ANU845 is a nonnodulating derivative of the parent strain ANU843, which is effective with T. repens and T. subterraneum (Rolfe et al. 1980). As with ANU843, the expected PCR product from ANU845 was obtained only with WRC1/BAC11 primer set (Table 3.1), but no amplification was observed with primers SAC1/BAC11. These observations were expected as strain ANU845 is a symbiotic plasmid cured derivative of the wild type strain ANU843. Specific detection of ANU845 has also shown the usefulness of the primers in this study in detecting nodulating and non-nodulating variants of Rhizobium in a group-specific manner. (iii) Other Rhizobium species A total of 12 additional reference strains representing five other Rhizobium species was also tested (Table 3.1). SAC1/BAC11 primers amplified the predicted PCR product using template DNA from two strains of R. etli (TAL182 and CFN42). This result was consistent with nucleotide sequencing of the 23S rDNA region from TAL182 which had revealed identical sequences at the 20-bp rDNA region used for annealing of primer SAC1 (Chapter 2). No other Rhizobium species produced amplification products with the SAC1/BAC11 primer set. In contrast, the W R C 1 / BAC11 primers successfully generated the predicted size product with extracted DNA from three strains of R. leguminosarum bv viciae and one R. leguminosarum bv phaseoli strain (Table 3.1). Four strains of R. meliloti and two strains of R. tropicii did not produce PCR amplification with any of 71 the reverse primers. However, these strains were amplified when the universal primers, BAC11 and BAC19, were used (Table 3.1). (iv) Root-associated bacterial species None of the group-specific primers yielded PCR products from the root-associated bacterial species tested in this study. Nevertheless, previous results have demonstrated a successful PCR amplification from these organisms using BAC11 and BAC19 primers (Petersen et al. 1995). 3.3.4. Detection of clover Rhizobium from intact cells Harrison et al. (1992) described the feasibility of amplifying DNA using PCR from intact Rhizobium cells. PCR using the group-specific primers developed in this study also produced the predicted size band using intact cells directly from liquid culture medium (Figures 3.6a, b). Similarly, positive amplification of the appropriate size band was evident in agarose gels when single colonies grown on solid medium (YMA) were used as template DNA (data not shown). These applications would eliminate the necessity for DNA extraction prior to analysis, and do not require probe labelling and hybridization steps. 3.3.5. Detection of clover Rhizobium from nodule samples Attempts to use samples from individual crushed nodules directly in a PCR assay failed to produce positive amplification (Figure 3.7). When PCR tubes containing nodule extracts were spiked with an equivalent amount of extracted DNA as the positive controls, amplification produced a band of the expected size, although the intensity of the bands was weaker than those of positive controls using extracted DNA alone (Figure 3.7). It was considered likely that the presence of host DNA 72 M 1 (A) 3 4 5 6 7 M 1 (B) 3 4 5 6 1.0 0.5 1.0 0.5 Figure 3.6. Group-specif ic detection of clover Rhizobium by P C R using (A) S A C 1 / B A C 1 1 , and(B) W R C 1 / B A C 1 1 primers from cell suspension directly from liquid culture. 1 kb marker (M), TAL910 (lane 1), TAL909 (lane 2), TAL798 (lane 3), A N U 8 4 3 (lane 4), B C R C 0 2 (lane 5), B C W C 0 1 (lane 6) and the no-template P C R negative control (lane 7). 73 M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Figure 3.7. P C R amplification of clover Rhizobium by using W R C 1 / B A C 1 1 primers and nodule tissue as a source of template DNA. 7. repens was inoculated with the R. I. bv trifolii strain A N U 8 4 3 and nodules were collected six weeks after inoculation. Individual nodules were crushed in Eppendorf tubes and the supernatant was used as a source of template DNA. 1 kb D N A marker (M), nodule extracts (lanes 1-10), nodule tissue + 100 ug D N A from A N U 8 4 3 (lane 11-12), extracted D N A from A N U 8 4 3 as positive contrl (lane 13), and the no-template P C R negative control (lane 14). 74 as well as the presence of other undefined substances in nodule tissue extracts may have decreased the sensitivity of the amplification reaction. The use of PCR with arbitrary primers, REP or ERIC primers has proven to be a fast, sensitive and reliable method for DNA fingerprinting and determine strain relationships with other rhizobial species (de Bruijn 1992; Harrison et al. 1992; Laguerre et al. 1996). The RAPD-PCR method has also been used to identify genomic targets for specific PCR analysis with other microbial species (Pooler et al. 1996). Results from this study also confirmed that both group-specific primers were useful in differentiating clover Rhizobium groups by PCR. Furthermore, the primers in this study provided additional information that different versions of the 20-bp primer region may be found in nodule-forming and root-associated bacterial species. Additional sequencing of 23S rDNA from other Bradyrhizobium and Rhizobium strains will clarify the potential utility of such unique sequences. The ability to amplify DNA for specific detection directly from cells in liquid culture and from solid medium provides an important step towards field monitoring of clover Rhizobium. Further study is needed to determine the limit of detection by PCR regarding the minimum number of Rhizobium cells needed for amplification. The possibility of using nodule extracts directly in PCR assays, while not successful in this study, would provide a valuable rapid detection method. 75 CHAPTER FOUR Nodulation characteristics and symbiotic effectiveness of T. semipilosum inoculated with compatible and incompatible microsymbionts 76 4.1. Introduction Results from the preceding chapters revealed that strains effective on the tropical perennial clover, T. semipilosum, can clearly be distinguished from those strains effective on other perennial Trifolium species including T. repens, T. pratense, T. hybridum and T. fragiferum. Furthermore, Rhizobium strains that are isolated from, and compatible with, the latter Trifolium species failed to produce effective symbiotic associations with T. semipilosum (Chapter 2). Plant responses from reciprocal inoculations were NodVFix", indicating that the sequence leading to a functional nodule had been interrupted. A quantitative and qualitative diagnosis of the symbiotic interaction involving T. semipilosum and Rhizobium is an effective prerequisite to detailed biochemical, molecular and genetical analyses. The nodulation characteristics of T. semipilosum with an incompatible R. leguminosarum bv trifolii strain (ANU843) were investigated in comparison to effective nodulation by a compatible strain (TAL910). A microscopic study of early nodule development was employed to compare the spatial pattern of infection and nodule occupancy by inoculant strains using GUS-marked Rhizobium strains and their respective parent strains. 4.2. Materials and Methods 4.2.1. Plant culture and inoculation using Leonard jar assemblies The procedures for seed-surface sterilization and germination were described in Chapter 2. Leonard jar assemblies (Vincent 1970) were constructed using autoclavable plastic Magenta™ jars. Two-day-old seedlings (4 seedlings/jar) were 77 planted aseptically into an autoclaved industrial sand and Turface® mix (50% each by volume) contained in a Magenta™ jar which was inserted into a second Magenta™ jar containing N-free medium and connected by a cloth wick. Control treatments received either sterile N-free nutrient solution as an uninoculated control, or sterile nutrient solution containing 70 mg L"1 nitrogen in the form of KN0 3 . One week after transplanting, seedlings were thinned to two per jar and each seedling of the inoculation treatments was inoculated with 1 ml of a Rhizobium strain suspension containing approximately 108 cfu/ml. Inoculants were prepared as described previously, and the experiments were carried out using growth conditions as described in Chapter 2. Treatments in each experiment were arranged in a completely randomized design. Data collection included, but was not limited to, nitrogenase activity using the acetylene reduction assay (ARA), plant dry matter (DM) yield, plant nitrogen (N) and carbon (C) content. 4.2.2. Acetylene Reduction Assay Nitrogenase activity was determined using the reduction of acetylene (C 2H 4) to ethylene (C 2H 2) (Somasegaran and Hoben 1994). Two clover plants were placed in 1-L volume plastic Nalgene bottle with a lid and 10% of the air in the bottle was replaced with C 2 H 4 gas (Holl et al. 1988). Bottles were maintained at 26°C with occasional shaking. Duplicate gas samples (1 ml_ each) were withdrawn after 2 h and were analyzed for C 2 H 2 production using a gas chromatograph (Hewlett Packard 530A) equipped with a flame ionization detector. Data are presented as 78 total nitrogenase activity (TNA) (//mole C 2 H 2 plant"1 h"1) or specific nitrogenase activity (SNA) (//mole C 2 H 2 g"1 whole plant DM). 4.2.3. Plant DM and Nitrogen Analysis The plants used for the ARA were partitioned into shoots and roots and oven dried for 48-72 hours and weighed. The shootroot ratio was calculated from the DM values. The N and C concentration (%) of shoots from five replicates (two plants in each replicate) were determined by dry combustion (CHN analyzer, Carlo Erba Model 1106, Department of Oceanography, UBC). The total N or C content in shoot tissue was calculated as the product of N or C concentration (%) and shoot DM. 4.2.4. Relative Effectiveness The relative effectiveness of inoculation treatments was determined by comparing the mean DM of shoots or roots of treated plants with reference plants receiving mineral N, or plants of uninoculated controls. The formula described by Ferreira and Marques (1992) was used to calculate relative effectiveness (RE) as follows: RE = [(DM rDMc)/(DMN-DMc)]*100 where, DM;= DM of treatment, DMC=DM of untreated control, and DMN=DM of mineral N-reference. 79 In addition, the percent DM increase over control treatments was calculated as follows: % increase over control= [(DM rDMc)/DMJ*100 where, DMj= DM of treatments, and DMC=DM of untreated controls. 4.2.5. Marking Rhizobium with the transposon GUS gene The E. coli strain S17-1 >l-p/r(pCAM111), carrying the mTn5givsA11 transposon with spectinomycin resistance (Wilson et al. 1995) was used as a donor strain in a biparental mating with R. /eguminosarum bv trifolii strains TAL910 or ANU843. Biparental mating and selection of transconjugants were conducted as described by Wilson et al. (1995). Donor and recipient cells were grown overnight, separately in YMB and mixed in Eppendorf tubes at a ratio of 5:1 E. coli.Rhizobium. The mating mixture was spread onto YMA plates and incubated at 30°C overnight. After mating, cells were suspended in sterile saline solution and plated on BD minimal media (Brown and Dilworth 1975) containing 0.2% mannitol and 75-100//g ml_"1 spectinomycin. Selected transconjugants were inoculated onto T. semipilosum and T. repens grown in NifTAL tubes in cross-inoculation assays. Their effectivity for nodulation and nitrogen fixation were compared with their respective parent strains using the ARA. 80 4.2.6. Symbiotic Effectiveness Study (a) Cross-inoculation experiments The symbiotic effectiveness of six wild-type Rhizobium strains effective on T. semipilosum or T. pratense was studied in cross-inoculation experiments using Leonard jar assemblies. Plants were inoculated 1 wk after transplanting and were grown for 5 wk after inoculation. (b) Further evaluation of T. pratense and T. repens The symbiotic effectiveness of seven wild type strains capable of effectively nodulating T. pratense and T. repens was evaluated in Leonard jar assemblies. Plants were inoculated 1 wk after transplanting and were grown for 5 wk after inoculation. Data collection included nitrogenase activity using the acetylene reduction assay (ARA), plant DM, and plant N and C content. This study was arranged in a completely randomized design and repeated three times with five replicates of each treatment. (c) Further evaluation of T. semipilosum The symbiotic effectiveness of four wild type strains (TAL910, TAL909, TAL798 and TAL577), capable of effectively nodulating T. semipilosum, was evaluated in Leonard jar assemblies. Plants were inoculated 1 wk after transplanting and were grown for 5 wk after inoculation. This study was repeated four times with 10 replicates of each treatment. 81 (d) Onset and development of nitrogen fixation in T. semipilosum nodules The objective of this experiment was to investigate the onset of nitrogen fixation and DM accumulation of T. semipilosum over time up to 35 days after inoculation. The set-up of this experiment was essentially the same as described for the previous experiment [4.2.6. (c)] except that seedlings were inoculated 2 d after transplanting and sequential measurements (harvests) were taken at 5, 6, 7, 9, 12, 21 and 35 d after inoculation. This study was repeated twice with five replicates of each treatment. 4.2.7. Microscopic Analysis Several experiments were conducted to study the spatial pattern of infection and nodule occupancy of inoculant strains using microscopic analysis following methylene blue and GUS staining as well as hematoxylin-stained nodule sections, (i) Methylene blue staining Experiments for methylene blue staining involved seedlings of T. semipilosum and 7". repens, grown on seedling-agar medium (Jensen 1962) as described in Chapter 2, and inoculated with wild-type strains TAL910 or ANU843. Uninoculated seedlings were included as controls. Plants were harvested every day for 10 d after inoculation and examined for root hair deformation and infection thread formation by light microscopy following fixation and methylene blue staining. Whole plants were fixed in 2.5% glutaraldehyde in 0.2M sodium cacodylate buffer (pH 7.2) for 15 min under vacuum and 15 min at atmospheric pressure 82 (Truchet et al. 1989), followed by rinsing three times for 15 min each with 0.2 M sodium cacodylate buffer (pH 7.2). Samples were dehydrated with an ethanol series (20 - 70% ethanol) and stored in 70% ethanol at room temperature until needed. They were rehydrated progressively with an ethanol series (70%-20% ethanol and finally with distilled water) and cleared in a 1.25% sodium hypochlorite solution in deionized water for 15 min under vacuum, and a further 15 min at atmospheric pressure (Truchet et al. 1989). Samples were stained for about 15 min with 0.005% methylene blue solution prepared in deionized water. Following staining, plants were rinsed in deionized water and root segments were cut and mounted on a microscope slide. Slides were examined by light microcopy for the presence of root hair deformation, root hair curling and infection thread formation. This experiment was repeated three times involving over 100 seedlings from each treatment in each experiment, (ii) GUS staining Experiments for GUS staining were conducted essentially as described above except that seedlings were inoculated with GUS-marked Rhizobium strains TAL910::gusA11 or ANU843::gusA11 and were grown under gnotobiotics conditions in either Petri plates or in Leonard jar assemblies. At each sampling time, whole plants were immersed in X-GlucA staining buffer consisting of 50 mM NaP0 4 (pH 7.0), 1 mM EDTA, 0.1% Sarkosyl, 0.1% Triton X-100, and 50//g X-Gluc (5-bromo-4-chloro-3-indolyl-/?-D-glucuronic acid) per mL (Wilson et al. 1995). To suppress endogenous GUS activity of any other 83 species of plant-associated bacteria adhering to the root surface, 100 //g chloramphenicol was added to the detection buffer (Wilson et al. 1995). Plants were stained under vacuum for 30 minutes at room temperature followed by incubation at 37°C overnight. After staining, whole plants were cleared with 1.25% sodium hypochlorite solution for 15 minutes, followed by rinsing with several changes of deionized water. Whole roots were observed by light and/or dissecting microscopy and photographs were taken using the Zeiss photomicroscope [Section 4.2.7. (Hi)], (iii) Examination of nodule sections Plants for this study were inoculated with wild-type strains TAL910 or ANU843 and were grown under gnotobiotics conditions in Leonard jar assemblies as described above. Plants were harvested at 16, 21 and 30 d after inoculation and nodules with approximately 2 mm long root segments attached were excised and fixed in Caruna solution (ethyl alcohol:acetic acid:chloroform 6:1:3) (Sass 1958) for 12 h and dehydrated: initially with 70% ethanol for 10 h and 100% ethanol for 3x 10 h followed by serial dehydration in mixtures of alcohol and xylene (70:30, 50:50, 30:70, and 0:100 ethanol:xylene) for 24 h each. Samples were immersed in xylene for 3x 24 h and embedded in a paraffin/wax mixture. Serial longitudinal sections were cut with a rotary microtome and stained with hematoxylin. Observations and photographs were made using a Zeiss universal microscope (Zeiss, W. Germany), equipped with a Type CS mechanical shutter, BEW1 microexposure meter on the basic unit and C-35 camera attachment. 84 4.2.8. Nodule Number and Distribution (a) Wild type strain inoculants The pattern of nodule number and distribution on T. semipilosum and T. repens was studied using wild-type strains compatible for the respective clover hosts grown in NifTAL tubes as described in Chapter 2. The numbers of nodules that appeared on the main root and lateral roots were determined separately at 30 d after inoculation. There were 20 replicates for each treatment. (b) GUS-marked strain inoculants The pattern of nodule number and distribution on T. semipilosum was studied using GUS-marked compatible and incompatible strains in Leonard jar assemblies. Following staining for GUS activity [Section 4.2.6(H)], the number of nodules and their locations were determined at 15-21 d after inoculation under a dissecting microscope. The experiment was repeated three times with at least 70 plants for each treatment in each experiment. 4.2.9. Data Analyses Analysis of variance was performed using the General Linear Model procedure of the SAS for Windows program (SAS Institute, 1989). The Bonferoni test was used for mean separation when the overall F-test was significant at P<0.05. When necessary, ANOVA was performed on data following log or arcsin transformation. 85 4.3. RESULTS 4.3.1. Cross-inoculation experiments The results of cross-inoculation experiments involving T. semipilosum, and T. pratense using selected wild-type Rhizobium strains are presented visually in Figure 4.1. The Rhizobium strains effective on the respective host plants were discriminated by the two clover species resulting in species-specific host-strain combinations, but the distinction was that of Fix+ versus Nod+. It was clear from Figure 4.1 that strains TAL910, TAL909, TAL798 and TAL577 produced vigorous plant growth only with T. semipilosum, while the other strains tested, ATCC 14480 and BCSC826, were highly effective only on T. pratense. Once the highly specific strain x Trifolium species interaction was apparent, further analyses were carried out on the host T. semipilosum separately from T. pratense and T. repens. 4.3.2. Symbiotic effectiveness of T. repens and T. pratense There were highly significant Rhizobium x Trifolium species interactions in shoot DM, root DM, total DM, shoot to root ratio, TNA and SNA (Figures 4.2 and 4.3). The effects of Rhizobium strains on plant shoot N and C concentrations were also significant (Table 4.1). There were no significant interaction effects. The relative effectiveness of Rhizobium strains also was highly significant (Figure 4.4). The interaction effect was not significant. 86 TAL910 TAL909 TAL798 TAL577 ATCC 14480 BCSC826 NILL KN0 3 Figure 4.1. Host-specificity and symbiotic effectiveness of Rhizobium strains. T. semipilosum (top row) and T. pratense (middle row) were cross-inoculated with Rhizobium strains (bottom row). Reference plants shown were uninoculated treatments (NILL), and the mineral N-treatment (KN03). Plants were grown in Leonard jar assemblies in growth chambers and harvested at 35 days after inoculation. Details of strains are listed in Table 2.1a. 87 (A) * 2 IT. pratense IT. repens (C) ~ 8 o o ro 6 Q. O) E °, 4 2 2 S T . pratense • T. repens (B) J 6 E 4 ° 2 IT. pratense IT. repens Strain (D) HT. pratense • T. repens Figure 4.2. Efffect of Rhizobium inoculation on shoot DM (A), root DM (B), whole plant DM (C), and shootroot ratio of T. repense and T. pratense grown in N-free medium in Leonard jar assemblies at 35 days after inoculation. Data are means of 5 replicates of 2 plants each. Bars indicate =+ SE mean. Strains: (A)=ATCC 14480, (B)=BCSC826, (G)=BCRC02, (H)=BCWC01, (l)=E2270, and (J)=E2093. c a o o E < z 2000 1500 1000 500 (A) 3T. pratense IT. repens Strain (B) •o O) X CM o o E < z • T. pratense • T. repens Strain Figure 4.3. Total Nitrogenase Activity (TNA) (A) and Specific Nitrogenase Activity (TNA) (B) of T. repens and T. pratense grown in N-free medium Leonard jar assemblies at 35 days after inoculation. Data are means of 5 replicates of 2 plants each. Bars indicate + SE mean. Strain designations listed in Figure 4.2. 89 Table 4.1. Shoot nitrogen and carbon composition of nodulated T. repens and T. pratense plants at 35 days after inoculation. P e r P l a n t s h o o t S t r a i n % N mg N % C C : N r a t i o A T C C 144 80 3 . 04 + 0 . 1 6 a 5 . 5 + 0 . 7 5 a 41 . 1 + 0 . 4 2 a 1 3 . 9 + 0 . 9 4 b BCSC826 3 . 08 + 0 . l l a 5 . 0 + 0 . 5 2 a 41 . 0 + 0 . 4 9 a 1 3 . 5 + 0 . 5 3 b BCRC02 2 . 99 + 0 . 0 6 a 4 . 8 + 0 . 5 1 a b 40 . 3 + 0 . 4 5 a b 1 3 . 6 + 0 . 3 3 b BCWC01 2 .2 0 + 0 . 1 5 b c 3 . 0 + 0 . 74 b 40 . 3 + 0 . 2 9 a b 1 8 . 9 + 1 . 2 3 a b E2270 2 . 68 + 0 . 0 1 a b 2 . 8 + 0 . 2 2 b c 40 . 8 + 0 . 4 6 a 1 5 . 4 + 0 . 4 9 b E2093 1 . 84 + 0 . 1 8 c 0 . 9 + 0 . 1 9 ° 39 . 1 + 0 . 5 3 b 2 3 . 8 + 3 . 1 3 3 There were no interactions between Trifoliium species and values for the strains are are averaged over the two plant species. Mean of 5 replicates of 2 plants each with + SEM. Means within a column followed by the same letter are not significantly different (P<0.05). Strain Figure 4.4. Relative effectiveness of Rhizobium strains.(A) shoot, (B) root, and (C) whole plant DM accumulation of inoculation treatments as percentage DM of the mineral N-treatment. Data are average of T. repens and T. pratense at 35 days after inoculation. Bars indicate + SE mean. Strain designations are listed in Figure 4.2. 91 The variables measured in this study were subjected to multivariate statistical analysis. The multivariate test statistic was highly significant (P<0.001) and hence a canonical variate analysis was performed using the SAS DISC procedure (SAS Institute, 1989). The first canonical variate which measured plant species differences accounted for 69.6% of the variation, while a further 12.3% of the variation was accounted by the second canonical variate which measured the degree of symbiotic effectiveness. When the Rhizobium x Trifolium species combinations were plotted against the first two canonical axes a definite separation was observed among the different Rhizobium x Trifolium combinations (Figure 4.5). 4.3.3. Symbiotic effectiveness of T. semipilosum Four Rhizobium strains capable of nodulating T. semipilosum plants were examined for their symbiotic effectiveness on cultivar 'Safari'. Shoot and root DM at 35 d after inoculation are shown in Table 4.2. The effectiveness of all of the inoculant strains was readily discernible in the shoot and root DM accumulation and shoot:root ratio values. Values from inoculated plants were significantly higher (P<0.05) than for uninoculated T. semipilosum plants (Table 4.2). Shoot DM was more responsive to inoculation than root DM (Figure 4.6). On average inoculated plants showed a 200% or more DM increase over uninoculated plants (Figure 4.6a); however, shoot DM of inoculated plants produced only 25% of the shoot DM of the mineral N-treatments (Figure 4.6b). There were no significant differences in relative effectiveness among the four 92 3 2 1 CM < 0 O -1 -2 -3 -4 T. pratense R*H -6 -4 PLANT CO LLI z HI > Ill SPECIES O H g m >-CO w* W*F o W*G T. > w repens W*H :0 W«J -2 0 CAN1 4 6 Figure 4.5. Graph of the canonical variate analysis showing Rhizobium x Trifolium species relationships on the first two canonical axis. Moving from the origin along the first axis to the right reprents T. repens, while moving from the origin to the left represents T. pratense. Moving from the origin towards the top of the plot represents highly effective symbiotic associations as plants showed higher plant DM yield and fixed more nitrogen as evidenced by ARA. Those combinations plotted to the bottom of the plot from the origin represented least effective symbiotic associations. Symbols represent R = T. pratense, W = T. repens, E = ATCC 14480, F = BCSC826, G = BCRC02 , H = BCWC01, I = E2270, and J = E2093. 93 Table 4.2. Dry matter and shoot:root ratio of T. sempilosum plants at 35 days after inoculation with Rhizobium strains infective on clover hosts. ma plant"1 Shoot :root Strain Shoot DM Root DM ratio TAL577 77+6.6a 18+1.83 4.7+0.43 TAL798 66+6.63 16+1.23 4.1++0.33 TAL909 70+7.63 15+1.63 4.9+0.63 TAL910 68+3.9a 15+0.93 4.5+0.33 C O N T R O L 19+1.5" 9+1.3" 2.5+0.4" Values are mean of 10 replicates of 2 plants each with + S E mean. Means followed by the same letter in each column are not significantly different (P<0.05). 94 400 300 c o o > ° 200 CD a> oo CO 100 Shoot DM Whole Plant D V f A B C D A B C D S t r a i n A B C D 35 30 25 to CO CD S 20 co 15 CO > • eg CD 1 0 S/ioof D M R o o f D M fk i i W/Jo/e P/anf D M rii A B C D O A B C D O A B C D O St ra in Figure 4.6. Relative effectiveness of four Rhizobium strains on 7. semipilosum grown in N-free medium: (a) DM accumulation as a percent of the uninoculated control, and (b) DM accumulation as a percent of the mineral N treatment. Strains: A = TAL577, B = TAL798, C = TAL909, D = TAL910, and O = Uninoculated control. Plants were harvested 35 days after inoculation. Data are means of 10 replicates of 2 plants each and bars indicate + S E mean. inoculant strains. Whole plant DM accumulation, TNA and SNA over a period of 35 d after inoculation are shown in Figures 4.7 - 4.9. There was no detectable nitrogen fixation activity (ethylene production) by nodulated T. semipilosum during the first 7 d after inoculation. Nitrogenase activity was first detected 9 d after inoculation. There were no significant differences among inoculant strains in whole plant DM and TNA recorded for the sampling periods up to 35 d after inoculation (Figures 4.7 & 4.8). A similar trend was observed for SNA except at 21 d after inoculation (Figure 4.9), at which time strains TAL910 and TAL909 showed significantly higher SNA values than strains TAL798 and TAL577. The concentrations of N and C in shoot tissue were determined 35 d after inoculation. There were no significant differences among treatments in percentage N or C (Table 4.3). Consistent with shoot DM production, total nitrogen and carbon accumulation by shoots of inoculated plants were significantly greater than uninoculated controls (Table 4.3). 4.3.4. Nodulation responses of T. semipilosum by strains TAL910 and ANU843 Nodules formed by strain TAL910 were large, red-pigmented, spherical and located predominantly on the upper primary root of T. semipilosum. Plants were green and healthy and nodule effectiveness was confirmed by the acetylene reduction assay. 96 Figure 4.7. Effect of Rhizobium inoculation on whole plant DM of T. semipilosum grown in N-free medium. Data are means of 5 replicates of 2 plants each and bars indicate + SE mean. 97 1 , 0 0 0 8 0 0 0 5 7 9 1 2 2 1 3 5 Days after inoculation Figure 4.8. Effect of Rhizobium inoculation on Total Nitrogenase Activity (TNA) of T. semipilosum grown in N-free medium. Data are means of 5 replicates of 2 plants each and bars indicate + SE mean. 98 14 Days after inoculation Figure 4.9. Effect of Rhizobium inoculation on Specific Nitrogenase Activity (SNA) of T. semipilosum grown in N-free medium. Data are means of 5 replicates of 2 plants each and bars indicate + SE mean. 99 Table 4.3. Shoot nitrogen and carbon composition of T. semipilosum plants at 35 days after inoculation with Rhizobium strains infective on clover hosts. Plant shoot strain % N mg N % C mg C C:N ratio TAL577 3.2+0.11a 2.5 a 40.3 a 30.9 a 12.6a TAL798 3.2+0.12a 2.1 a 39.8 a 26.2 a 12.5a TAL909 3.0+0.093 2.1 a 39.4 a 27.5 a 13.2a TAL910 3.1+0.053 2.1 a 40.0 a 27.0 a 12.9 a C O N T R O L 3.0+0.12a 0.6b 38.8 a 7.2b 13.1b Values are mean of 5 replicates of 2 plants each with + S E mean. Means followed by the same letter in each column are not significantly different (P<0.05). 100 In contrast, nodule-like structures that were formed by strain ANU843 were white, variable in size but generally small. Most nodule-like structures were associated with, or in close proximity to, a lateral root. The ARA failed to detect nitrogenase activity in T. semipilosum plants inoculated with either the parent or GUS-marked strain ANU843, over the course of 30 d after inoculation. Plants were severely chlorotic and often died even before uninoculated control plants. 4.3.5. Assessment of marked Rhizobium strains for nodulation ability Transconjugants were selected for growth and development of blue colonies on YMA plates containing spectinomycin and XGIuc. Twelve colonies from each marked Rhizobium strain were tested for their ability to nodulate effectively T. repens and T. semipilosum. Two colonies of each marked strain that showed the best plant performance after four weeks of growth in NifTAL tubes were selected for further analysis. The effectivity of these colonies was also compared with that of their respective parent strains by using plants grown in Leonard jar assemblies in a controlled environment chamber. Strain effectiveness was measured by scoring plant appearance and by the ARA. There were no significant differences in plant appearance between those inoculated with marked strains and their respective parent strains (Figure 4.10). Twenty one d after inoculation, there were comparable rates of acetylene reduction between clover hosts nodulated by marked strains (TAL910::gusA11 or ANU843::gusA11) and their respective parent strains (data not shown), indicating that the interaction was effective. 101 Figure 4.10. Host-specificity and symbiotic performance of Rhizobium strains marked with mTn5gfusA11 and their parent strains (ANU843 and TAL910) . Plants were grown in Leonard jar assemblies in growth chambers and harvested 21 days after inoculation. 102 Incompatible interactions using marked strains and T. semipilosum or T. repens hosts produced stunted and chlorotic plants (For example see Figure 4.10) and as expected no ARA was detected. Marked strains demonstrated comparable effectiveness with their respective parent strains. To demonstrate expression of the gusA gene in nodules formed by marked strains, plants were incubated with the GUS detection buffer. Nodules occupied by the GUS-marked strains, including very young nodules, were detected by the GUS assay (Figure 4.11), confirming nodule occupancy by the inoculant strain and expression of the GUS + trait. Plant roots from uninoculated controls or effectively nodulated by wild-type strains did not show a response to staining with the GUS detection buffer. This difference in expression was found to be useful in monitoring and comparing the spatial pattern of infection and occupancy of nodules in inoculation studies as illustrated in Figure 4.12. 4.3.6. Root hair deformation and infection thread formation Root hair deformation and infection thread formation on T. semipilosum plants was examined by light microscopy following methylene blue staining or histochemical staining for GUS activity. Plants receiving sterile N-free nutrient solution were included as controls and over 100 plants from each treatment were used. Strain TAL910 or its derivative, TAL910::gusA11, were able to cause root hair deformation including curling, branching and twisting of root hairs. Strain ANU843 or its derivative, ANU843::gusA11, also caused root hair branching and twisting. Twisted root hairs were also seen rarely in uninoculated control plants. 103 Figure 4.11. T. repens inoculated with GUS marked Rhizobium strain ANU843::gusA11 as seen under a dissecting microscope at 30 days after inoculation in the G U S substrate X-GlucA. Nodules induced by the marked strain are identified by the development of a blue color. Note staining and detection of even very young nodules (arrow). Plant roots from uninoculated controls or plants nodulated by wild-type strains did not show a blue color. Bar = 1 mm. 104 Figure 4.12. Early steps of host microsymbiont interactions can be followed by chemical staining of B-glucuronidase using GUS-marked strains. Blue coloration identifies inoculant strains expressing gusA gene within infected root hair cells. The root system of T. repens inoculated with ANU843::gusA11 was viewed by light microscopy following staining for GUS:(a) infection thread (Arrow head); (b) round shaped nodule primordium (P) colonized by marked rhizobia ; (c) emerging nodule; and (d) N-fixing nodule. Note the localization of the blue color predominantly around the infection zone of the mature nodule. Bar = 0.1 mm. 105 Moderately curled root hairs and infection thread-like structures were observed in T. semipilosum plants inoculated with the compatible strains TAL910 and TAL910::gusA11, despite the fact that nodule formation from this association is localized primarily around the upper region of the root where few root hairs are located. Infection thread-like structures were also detected on T. repens root hairs after four d of inoculation with their respective compatible rhizobial strain ANU843::gusA11 (Figure 4.12) and its parent strain ANU843. In contrast, similar structures were not detected on any T. semipilosum plant inoculated with the incompatible rhizobial strains ANU843::gusA11 or its parent strain ANU843. Following GUS staining, swellings could be discriminated as pre-emergent nodules rather than initiating lateral roots. Even in areas where root nodules were beginning to develop on T. semipilosum following inoculation with ANU843::gusA11, infection threads were not found within any of the root hairs examined. However, the ability of incompatible rhizobia to colonize nodule initiation sites and enter the root system at the point of lateral root emergence was reflected in the histochemical analysis (Figure 4.13). The development of blue color following GUS staining was indicative of nodule occupancy by inoculant incompatible strains on those nodules induced at or around the junction between lateral roots and the tap root (Figure 4.13, c-d). 106 Figure 4.13. T. semipilosum inoculated with ANU843::gftvsA11 and viewed by light microscopy after histochemical staining of B-glucuronidase activity of whole root system. Many nodules during the incompatible interactions were formed at the junction between lateral roots (LR) and tap root (TR): (a) rhizobia are seen colonizing lateral root emergence sites (arrows); (b) emerging nodule; (c & d) nodules at 25 days after inoculation. Note development of blue color indicating occupancy of nodule-like structure by inoculant strain. Also compare the absence of distinct zone of staining in nodules seen here compared to those induced by a compatible interaction in Figure 4.12. Bar = 0.1 mm. 107 4.3.7. Nodule number and distribution The number of nodules that were induced on the main root and lateral roots of T. repens and T. semipilosum were determined separately at 30 d after inoculation. Trifolium semipilosum inoculated with a compatible rhizobia formed on average 3.3 nodules plant"1 (Figure 4.14); the majority of the nodules were on the upper region of the primary root. In contrast, T. repens nodulated by two different compatible strains formed on average, 12 or 16 nodules/plant. About 65% of the total nodules formed on T. repens were associated with lateral roots. In three separate experiments, T. semipilosum inoculated with the marked strains TAL910::gusA11 or ANU843::gusA11 was grown in Leonard jar assemblies for 15-21 d after inoculation. Following staining for GUS, the number of nodules and their locations were recorded using at least 70 plants for each treatment in each experiment. Since similar trends were found in all the experiments, the results of only one of the experiments are reported. Mean values of 3.7 nodules/plant were observed in compatible interactions, compared to 8.2 nodules/plant for incompatible interactions; the incompatible Rhizobium strain induced significantly greater numbers of nodules on T. semipilosum (Figure 4.15). In the compatible combination, nodules were located on the primary root (59.8%), lateral roots (5.0%)and at the junction between the primary and lateral roots (35.3%). For incompatible interactions nodule distribution was 20.1%, 18.3% and 61.6% respectively for the analogous positions. More than 95% of all nodules formed during a compatible interaction showed a positive GUS reaction, whereas 108 20 15 CO CD O O 10 CD - O 0 rofa/ A/O.. . of nodules ori lateral roots On tap roots S x A W x R W x F S x A W x R W x F S x A W x R W x F Trifolium species x Rhizobium strain Figure 4.14. Nodulation characteristics of T. semipilosum (S) and T. repens (W) inoculated with Rhizobium strains: A = TAL910, R = BCSC826, and F = ANU843. Plants were grown in N-free seedling agar medium (Jensen 1962) in NifTAL tubes (Somasoearn and Hoben 1994) and were harvested 35 days after inoculation. Data are means of 20 plants and bars indicate + SE mean. o 3 T3 O C 10 8 Whole plant at the junction on tap roots on lateral root n IT I "1~ imi • # of total nodules « # of GUS+ nodules 910 843 910 843 910 843 910 843 GUS-marked strain Figure 4.15. Nodule number and distribution of T. semipilosum inoculated with a compatible (TAL901::gusA11) and incompatible (ANU843::gusA11) Rhizobium in N-free medium in Magenta jars. Plants were harvested 15 days after inoculation and stained for GUS. Data are means of 70 plants and bars indicate + SE mean. 110 only 33% of all nodules from incompatible interactions were similarly stained; 70% of these reactive nodules were located at the junction between the primary root and lateral roots (Figure 4.15). 4.3.8. Nodule anatomy Longitudinal sections of T. semipilosum root nodules induced by the compatible rhizobial strain, TAL910, stained with hematoxylin were used to evaluate the anatomy of a normal nodule. Essential features of these nodules included an apical meristem and a zone of central nodule tissue which appeared to be colonized by rhizobia (Figure 4.16). Cells of the nodule meristem were uniform in shape and contained a centrally-located nucleus. Mean nodule dimensions were 1.3 mm long and 0.7 mm wide at day 16 after inoculation; nodules continued to elongate to approximately 2.6 mm long and 1.1 mm wide at 30 d after inoculation. In contrast, longitudinal sections of nodule-like structures induced by the incompatible rhizobial strain, ANU843, revealed that the nodules formed by ANU843 lacked a well-defined, clear sequence of developmental zones arising from the nodule meristem (Figures 4.17 & 4.18). Their meristematic activity was limited and the central area resembled normal nodules seen in Figure 4.16. These nodule-like structures were smaller than wild-type induced nodules ranging in size from 0.9 mm long and 0.6 mm wide at 16 d after inoculation, and 0.9 mm long and 0.7 mm wide at 30 d after inoculation. 111 Figure 4.16. Light micrographs of longitudinal section nodules induced by wild-type strain TAL910 on T. semipilosum plants: Nodules were sectioned at 16 days after inoculation (a & b), and 21 days after inoculation (c). Abbreviations are : M = nodule meristem; IZ = infection zone; C = Cortex; U = uninfected tisssue; and V = vascular tissue. Bar = 0.1 mm. 112 Figure 4.17. Light micrographs of longitudinal section nodules induced by the incompatible strain A N U 8 4 3 on T. semipilosum plants: Nodules were sectioned at 16 days after inoculation (a & b), and 21 days after inoculation (c). V = vascular tissue. B a r = 0 . 1 m m . 1 Figure 4.18. Light micrographs of longitudinal section nodules induced by the incompatible strain A N U 8 4 3 on T. semipilosum plants: Nodules were sectioned at 21 days after inoculation (a & b), and 30 days after inoculation (c & d) Bar = 0.1 mm. 114 Comparison of longitudinal sections from nodules induced by the incompatible strain on the different parts of the root system showed heterogeneous anatomy (Figure 4.19). Although the general tissue organization of nodules formed on both tap roots and lateral roots by strain ANU843 was similar to that found in normal nodules, considerable differences were evident in the extent of nodule meristem development, and in the enlargement of nodule cells distal to the nodule meristem. 4.4. Discussion It was shown in Chapter 2 that Rhizobium strains effective on temperate clover species such as T. repens and T. pratense failed to produce effective symbiotic associations with the tropical perennial clover, T. semipilosum, and vice versa. The study using Leonard jar assemblies in this chapter further demonstrated that symbiotic effectiveness of Rhizobium strains effective on T. semipilosum and T. pratense was species-specific. Furthermore, strains effective on T. pratense and T. repens differed in their degree of effectiveness, reflecting the presence of highly significant Rhizobium x Trifolium species interactions, even between host plants which had responded similarly in earlier cross-inoculation experiments. The data presented in this chapter show that Rhizobium strains that are effective on T. semipilosum caused similar plant responses in shoot and root DM production, and shoot N accumulation in N-free growth conditions. 115 Figure 4.19. Light micrographs of longitudinal section nodules induced by incompatible strain ANU843 on T. semipilosum plants: Nodules formed on tap roots (a), lateral roots (b), and junction between tap and lateral roots (c) were picked at 16 days after inoculation. See Figure 4.16. for abbreviations. Bar = 0.1 mm. 116 The four evaluated strains were equal in relative effectiveness, and total and specific nitrogenase activity. Under the experimental conditions in this study, effectively nodulated plants had a lag period of at least eight d before the onset of nitrogen fixation. This lag period may have accounted for the observed significant DM differences between inoculated and N-treatments; plants receiving inorganic nitrogen were grown in N-amended medium from the first day of the experiment. T. semipilosum plants inoculated with wild-type effective strains showed nodule formation predominantly near the upper region of the tap root compared to T. repens plants where nodules were induced primarily on lateral roots. This difference in the pattern of nodulation sites between the two Trifolium species suggests that nodule distribution on the root system may be controlled by the host. This observation is consistent with nodulation distribution differences which have been described at the genus level (Somasegaran and Hoben 1994). In this chapter, the anatomy of T. semipilosum root nodules formed in response to infection by the wild-type R. leguminosarum bv trifolii strain TAL910 was found to be analogous to those reported for other indeterminate nodules. In contrast, R. leguminosarum bv trifolii strain ANU843, effective on T. repens (Rolfe et al. 1980), formed root nodules on T. semipilosum that were ineffective (Fix" phenotype). This anomalous nodulation was characterized by the formation of numerous nodules, which were white, and smaller than those formed with the compatible microsymbiont; Dart (1977) observed that ineffective rhizobial strains often form more nodules than effective strains. The increased nodule numbers 117 were located at sites on the lateral roots and at the junctions of lateral and tap roots. Histochemical analysis clearly showed that strain ANU843 was able to initiate the formation of nodule-like primordia, and/or enter the root system predominantly near lateral root emergence sites. The GUS assay confirmed nodule occupancy by the ineffective inoculant strain; approximately 70% of all stained nodule-like structures were localized at the junctions of lateral and tap roots. However, the failure to detect infection thread formation and the location of ineffective nodules suggested that infection by the incompatible strain occurred primarily through the epidermis at the point of lateral root emergence. This route of infection has previously been described for root nodulation of other plant systems including peanut (Arachis hypogea) (Bergersen 1982; Sprent 1979), but is unusual in the genus Trifolium. Previous reports suggested that lateral root emergence sites result in disruption of existing host tissues and represent potential zones of root exudation (Hale et al. 1978). Root exudates are known to enhance growth rates of bacteria (D'Acry-Lameta and Jay 1987; Hartwig et al. 1991), act as chemoattractants (Aguilar et al. 1988; Armitage et al. 1988; Caetano-Anolles et al. 1988), and as transcriptional signals in the communication between host plants and rhizobia (Djordjevic et al. 1987b; Maxwell et al. 1989). There is no direct evidence in this study which helps explain strain ANU843's ability to enter the root system of T. semipilosum at the point of emerging lateral roots. However, it is appealing to 118 speculate that the route of entry by the incompatible rhizobia which occurred at the point of emerging lateral roots may have been a result of trophic chemotaxis. The incompatible R. leguminosarum bv trifolii strain ANU843 was able to cause root hair branching in T. semipilosum. Although root hair branching is considered to be a moderately specific host response (Vincent 1980), it is likely that the root hair branching of T. semipilosum in this study may have resulted from the action of Nod factors produced by the incompatible rhizobial strain. Such examples in the literature include root hair deformation of Vicia sativa induced by externally applied Nod factors from incompatible strains such as B. japonicum (Carlson et al. 1993) and R. loti (Lopez-Lara et al. 1995). It has been known for some time that initiation of the nodule meristem and switching on of nodule primordia by a potential host is triggered at a distance by diffusible signals from rhizobia (Truchet et al. 1980); external applications of the appropriate concentrations of Nod factors can stimulate cortical cell division and nodule organogenesis (Lerouge et al. 1990; Nap and Bisseling 1990; Spaink et al. 1991; Spaink 1992). It is not clear whether the observation of a somewhat similar nodular tissue organization by nodules formed by the incompatible strain with that of normal nodules was related to the production of Nod factors by strain ANU843 during an incompatible interaction with T. semipilosum. In spite of these earlier observations, the stage at which the sequence leading to functional nodule formation in T. semipilosum during interactions with Rhizobium effective on white clover may have been interrupted remains to be 119 seen. Based on observations from this study, a model can be formulated in which the observed nodulation phenotypes of T. semipilosum with compatible (TAL910) and incompatible (ANU843) strains can be explained. R. leguminosarum bv trifolii strain ANU843 appears to have produced Nod factors that may have caused T. semipilosum to initiate nodule development, but were not able to overcome a physiological regulation exerted by the host plant. Further study of Rhizobium-T. semipilosum should focus on the molecular details of the interaction emphasizing the genetic, molecular and biochemical basis of host specific nodulation as it relates to nod gene induction and Nod factor amounts and structure. CHAPTER FIVE Summary 121 The genus Trifolium is comprised of at least 240 legume species which are diverse in morphology, habitat and ecology (Williams 1987; Zohary and Heller 1984). Like many other legumes, Trifolium species have the capacity to interact symbiotically with the soil bacterium Rhizobium to convert atmospheric nitrogen to ammonia. This symbiotic association appears to be highly specific: no single Rhizobium strain is able to nodulate all Trifolium species (Burton 1980; Vincent 1974). The two most widely cultivated temperate perennial species, T. repens and T. pratense, show similar responses to Rhizobium inoculation, although one particular Rhizobium strain does not necessarily interact better than another over a wide range of cultivars (Mytton 1975). In contrast, strains that are effective on temperate Trifolium appear to be incompatible with tropical Trifolium species (Burton 1980; Vincent 1974). Burton (1980) and Vincent (1974) proposed about 12 clover host groups based on their cross-inoculation responses using Rhizobium strains. Despite such observations, all Rhizobium strains that nodulate Trifolium species are grouped in a single species classification as R. leguminosarum bv trifolii (Jordan 1984). Furthermore, most of our understanding of nodule development in Trifolium has been derived almost exclusively from temperate or Mediterranean species such as T. repens, T. pratense and T. subterraneum (Weinman et al. 1991), which may not represent the spectrum of Rhizobium-Trifolium interactions in nature. There are many agriculturally-important host plants in tropical and sub-tropical systems which have not been investigated in great detail, including T. 122 semipilosum. Trifolium semipilosum (Kenya white clover) is a perennial clover native to Sub-Saharan Africa, which has a growth habit similar to T. repens, but is reputed to be more tolerant of drought (MacKay 1973) and acidic soils (Jones and Jones 1982). The objectives of this study were to examine Rhizobium from perennial Trifolium species for their host range and symbiotic effectiveness and to determine the phylogenetic relationship of Rhizobium isolated from T. semipilosum with rhizobia that are effective on temperate Trifolium species. The results of this study are summarized as follows: (i) Among the clover hosts, degrees of specificity were reflected in nodulation responses ranging from no nodulation to completely effective nodulation. The results from the cross-inoculation experiments clearly indicated that specificity in the clover symbiosis was not limited to early nodulation, but might also influence later stages of nodulation and/or establishment of effective N-fixation. For example, Rhizobium strains that formed effective nodules on T. repens, T. pratense, T. hybridum and T. fragiferum produced NodVFix" phenotypes on T. semipilosum and vice versa, indicating that the sequence of events leading to a functional nodule had been interrupted. (ii) Results from substrate utilization analysis using the Biolog™ system suggested that Rhizobium strains effective on T. semipilosum may have a broader metabolic profile than strains effective on T. repens, T. pratense, 123 T. hybridum and T. fragiferum. Cluster analysis based on carbohydrate substrates produced clear-cut strain differentiation among these clover host groups. It is not known whether failure of a given strain to metabolize a given substrate under the experimental conditions in this study denotes an absolute inability for substrate utilization by that strain. Nevertheless, the Biolog™ test results correlate with the clover host grouping observed in the cross-inoculation experiments. These results together may reflect the different genetic backgrounds of Rhizobium isolates that are effective on different clover hosts. RAPD-PCR, ERIC-PCR, and PCR-based nucleotide sequence analysis of 16S and 23S rDNA regions were used for genetic analysis of strains. A considerable level of genetic diversity was found using RAPD- and ERIC-PCR. Rhizobium strains effective on T. semipilosum formed a tight cluster that was distinct from strains effective on temperate Trifolium species. Comparative nucleotide sequence analysis of 23S rDNA regions clustered Rhizobium strains effective on T. semipilosum, T. repens, T. pratense, T. hybridum and T. fragiferum into two distinct groups. This grouping is noteworthy as it is consistent with the pattern of symbiotic effectiveness observed in cross-inoculation experiments. These data suggested that Rhizobium strains effective on T. semipilosum show a distinct genetic structure compared to strains effective on other clover hosts. 124 (iv) The results of this study substantiate the view that PCR-based DNA banding patterns and 23S rDNA sequencing offer effective tools to discriminate closely related root-nodule forming bacteria. Furthermore, unique features identified by secondary structure analysis of the sequenced 23S rDNA region were consistent with the hypothesis that 23S rDNA could be used to design species- or strain-specific Rhizobium probes. Two 20-bp primers were constructed that allowed group-specific detection and differentiation of clover Rhizobium by PCR. The target for DNA amplification was a 370-bp fragment of the sequenced 23S rDNA region; analysis of other Rhizobium, as well as other root-associated bacterial species, revealed that widespread variability occurred within the 20-bp segment of the primer annealing region. (v) Four Rhizobium strains effective on T. semipilosum produced similar plant responses for shoot and root DM, shoot:root ratio, shoot N and C accumulation, relative effectiveness, total and specific nitrogenase activity (ARA). (vi) Effective Rhizobium strains on T. semipilosum induced deformed root hairs (curling, branching and twisting) and infection threads, suggesting a similar route of infection as that for T. repens. Nodules induced by effective strains on T. semipilosum were localized near the upper region of the tap 125 root where fewer root hairs are located compared to T. repens which had nodules distributed largely on lateral roots. Rhizobium strain ANU843, effective on T. repens, caused anomalous nodulation on T. semipilosum producing Fix" phenotypes. The anomalous nodulation was characterized by the formation of significantly greater numbers of nodules, which were white and variable in size but generally smaller than normal. Increased nodule numbers were attributed to the induction of nodule-like structures at the lateral roots and at the junction of lateral roots and tap roots. Strain ANU843 also caused root hair branching and twisting on T. semipilosum, but infection threads were not detected on any of the plants examined. Microsymbionts for T. semipilosum and T. repens were transformed with a constitutively expressed gusA gene. These marked strains provided a visual assay for rhizobial infection. It was observed that 33% of nodules from the incompatible interaction showed a positive GUS reaction. Using a histochemical assay, strain ANU843 was shown to enter the root system of T. semipilosum mainly at the epidermal sites of emerging lateral roots, and 70% of all reactive nodules from an incompatible interaction were located at the junction between the primary root and lateral roots. 126 (ix) The anatomy of nodules induced by effective strains on T. semipilosum was analogous to those reported for other indeterminate nodules. The anatomy of some of the nodule-like structures during the incompatible interaction showed a similar nodular tissue organization to wild-type induced nodules, but with a restricted nodule meristem and apparently no differentiated, active bacteroid zone. Results from this study suggest that the symbiotic interaction between Rhizobium strains and T. semipilosum may be unique; T. semipilosum and T. repens strains do not cross inoculate, consistent with the view that Rhiozobium strains effective on these host plants do not belong to the same cross-inoculation group. The potential for differentiating and detecting Rhizobium groups effective on these hosts by means of specific DNA sequences provided an indirect measure of the heterogenity of the Trifolium-Rhizobium cross-inoculation group. Nevertheless, future research using molecular and biochemical approaches will be necessary to provide more direct evidence for molecular events in the T. semipilosum-Rhizobium interaction. Such studies should target the chemical structure and biological activity of nod factors from Rhizobium strains effective on T. semipilosum. 127 References 128 Abed, Y., Davin, A., Charrel, R. N., Bollet, C , and De Micco, P. 1995. Variation of RAPD-fingerprint patterns using different DNA-extraction methods with gram-positive bacteria. World J. Microbiol. Biotechnol. 11: 238-239. Aguilar, J. M. M., Ashby, A. M., Richards, A. J. M., Loake, G. J., Watson, M. D., Shaw, C. H. 1988. Chemotaxis of Rhizobium leguminosarum biovar phaseoli towards flavonoid inducers of the symbiotic nodulation genes. J. Gen. Microbiol. 134: 2741-2746. Ardourel, M., Demont, N., Debelle, F., Mai I let, F., de Billy, F., Prome, J. C , Denarie, and Trucht, G. 1994. Rhizobium meliloti lipooligosaccharide nodulation factors: different structural requirements for bacterial entry into target root hair cells and induction of plant symbiotic development responses. Plant Cell 6: 1357-1374. Armitage, J. P., Gallagher, A., and Johnston, A. W. B. 1988. Comparison of the chemotactic behaviour of Rhizobium leguminosarum with and without the nodulation plasmid. Mol. Microbiol. 2:743-748. Armstrong, J., Gibbs, A., Peakall, R., and Weiller, G. 1994. The RAPDistance package, ftp://life.anu.edu.au/pub/ Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. 1989. Current protocols in molecular biology. John Wiley and Sons, New York. Bendor, G. L. and Rolfe, B. G. 1985. A rapid plant assay for Parsponia-Rhizobium symbiosis. Plant Sci. 38:135-140. Bergersen, F. J. 1978. Physiology of legume symbiosis. In Interactions between non-pathogenic soil microorganisms and plants. Edited by Y.R. Dommergues and S.V. Krupa. Elsevier Scientific Publishing Company, Amsterdam, pp. 305-333. 129 Bergersen, F. J. 1982. Root nodules of legumes: structure and functions. Research studies press, John Wiley and Sons Ltd., England. 127 p. Bhuvaneswari, T. V., and Solheim, B. 1985. Root hair deformation in the white clover/Rhizobium trifolii symbiosis. Physiol. Plant. 63:25-34. Bochner, B. 1989. Breath prints at the microbial level. Am. Soc. Microbiol. News 55:536-539 Brewin, N. J. 1991. Development of the legume root nodule. Annu. Rev. Cell Biol. 7:191-226. Briel, B. T., and Triplett, E. W. 1996. The 16S rRNA sequence from R. leguminosarum bv trifolii T24. J . Bacteriol. 178:4150-4156. Brockwell, J., and Katznelson, J. 1976. Symbiotic characteristics of Rhizobium trifolii from Israel in association with 10 species of Trifolium. Aust. J. Agric. Res. 27:799-810. Brosius, J., Dull, T. J., and Noller, H. F. 1980. Complete nucleotide sequence of a 23S ribosomal RNA gene from Escherichia coli. Proc. Natl. Acad. Sci. U.S.A. 77:201-204. Brown, C. M., and Dilworth, M. J. 1975. Ammonia assimilation by Rhizobium cultures and bacteroids. J . Gen. Microbiol. 122:61-67. Burton, G. C. 1980. New developments in inoculating legumes. In recent advances in biological nitrogen fixation. Edited by N.S. Subba Rao. Edward Arlond. Great Britain, pp. 380-405 Caetano-Anolles, G., Crist-Estes, D. K., Baner, W. D. 1988. Chemotaxis of Rhizobium meliloti to the plant flavone luteolin requires functional nod genes. J. Bacteriol. 170:3164-3169. 130 Carlson, R. W., Sanjuan, J., Bhat, R., Glushka, J., Spaink, H. P., Wijfjes, A. H. M., van Brussel, A. A. N., Stokkermans, T. J. W., Peters, N. K. 1993. The structures and biological activities of the lipo-oligosaccharide nodulation signals produced by type I and type II strains of Bradyrhizobium japonicum. J . Biol. Chem. 268:18372-18381. Chen, W. X., Li., G. S. , Qi, Y. L., Wang, E. T., Yuan, H. L. and Li, J. L. 1991. Rhizobium huakii sp. nov. isolated from the root nodules of Astragalus sinicus. Int. J . Syst. Bacteriol. 41:275-280. Crush, J. R. 1987. Nitrogen fixation. In White clover. Edited by M.J. Baler and W.M. Williams. C.A.B. Abstracts, U.K. pp. 185-20. Dakora, F. D., Joseph, C. M., Phillips, D. A. 1993. Common bean root exudates contain elevated levels of daidzein and coumestrol in response to Rhizobium inoculation. Mol. Plant-Microbe Interact. 6: 665-668. D'Arcy-Lameta, A. A., and Jay, M. 1987. Study of soybean and lentil root exudates III. Influence of soybean isoflavonoids on growth of rhizobia and some rhizospheric microorganisms. Plant Soil 101:267-272. Dart, P. 1977. Infection and development of leguminous nodules. In A treatise on dinitrogen fixation. Edited by R.W. F. Hardy and W.S. Silver. John Wiley & Sons, Inc., New York. pp. 367-472 Debelle, F., Rosenberg, C , Vasse, J., Mai I let, F., Martinex, E., Denarie, J., Truchet, G. 1986. Assignment of symbiotic developmental phenotypes to common and specific nodulation (nod) genetic loci of Rhizobium meliloti. J. Bacteriol. 168:1075-1086. de Bruijn, F. J. 1992. Use of repetitive (Repetitive Extragenic Palindromic and Enterobacterial Repetitive Intergeneric Consensus) sequences and the polymerase chain reaction to fingerprint the genomes of Rhizobium meliloti isolates and other soil bacteria. Appl. Environ. Microbiol. 58: 2180-2187. 131 Denarie, J. and Cullimore, J. 1993. Lipo-oligosaccharide nodulation factors: a minireview of new class of signalling molecules mediating recognition and morphogenesis. Cell 74: 951-954. Denarie, J., Debelle, F., and Rosenberg, C. 1992. Signalling and host range variation in nodulation. Ann. Rev. Microbiol. 46: 497-531. Dharmatilake, A. J. and Bauer, W. D. 1992. Chemotaxis of Rhizobium meliloti towards nodulation gene inducing compounds from alfalfa. Appl. Environ. Microbiol. 58: 1153-1158. Djordjevic, M. A. Gabriel, D. W., and Rolfe, B. G. 1987a. Rhizobium-ihe refined parasites of legumes. Ann. Rev. Phytopathol. 25:145-168. Djordjevic, M. A., Redmond, J. W., Batley, M., Rolfe, B. G. 1987b. Clovers secret specific phenolic compounds which either stimulate or repress nod gene expression in Rhizobium trifolii. EMBO J. 6:1173-1179. Djordjevic, M. A., Schofield, P. R., and Rolfe, B. G. 1985. Tn5 mutagenesis of Rhizobium trifolii host specific nodulation genes result in mutants with altered host range ability. Mol. Gen. Genet. 200:463-471. Drahos, D. J., Hemming, B. C , and McPherson, S. 1986. Tracking recombinant organisms in the environment: /?-galactosidase as a selectable, non-antibiotic marker for fluorescent Pseudomonads. Bio/Technology, 4:439-443. Dreyfus, B., Garcia, J. L, and Giliis, M. 1988. Characterization of Azorhizobium caulinodans gen. nov., sp. nov., a stem-nodulating nitrogen-fixing bacterium isolated from Sesbania rostrata. Int. J . Syst. Bacteriol. 38:89-98. Drouin, P., Prevost, D., and Antoun, H. 1994. (Agriculture and Agri-food Canada, Ste-Foy, Que.) Unpublished data. 132 Dudley, M. E., Jacobs, T. W., and Long, S.R. 1987. Microscopic studies of cell division induced in alfalfa roots by Rhizobium meliloti. Planta 171:289-301. Eardly, B. D., Hannaway, D. B., Bottemley, P. J. 1985. Characterization of rhizobia from ineffective alfalfa nodules: ability to nodulate bean plants Phaseolus vulgaris L. Appl. Environ. Microbiol 50:1422-1427. Eardly, B. D., Materon, L. A., Smith, N. H., Johnson, D. A., Rumbaugh, M. D., and Selander, R. K. 1990. Genetic structure of natural populations of the nitrogen-fixing bacterium Rhizobium meliloti. Appl. Environ. Microbiol. 56:187-194. Eardly, B. D., Young, J. P. W., and Selander, R. K. 1992. Phylogenetic position of Rhizobium sp. Strain Or191, a symbiont of both Medicago sativa and Phaseolus vulgaris, based on partial sequence of the 16S rRNA and n/'/H genes. Appl. Environ. Microbiol. 58: 1809-1815. Felsenstein, J. 1995. PHYLIP (Phylogeney Inference Package), version 3.57c, Department of Genetics, University of Washington, Seattle, U.S.A. Ferreira, E. M., and Marques, J. F. 1992. Selection of Portuguese Rhizobium leguminosarum bv trifolii strains for production of legume inoculants. I. Screening for effectiveness in laboratory conditions. Plant Soil 147:151-158. Fisher, R. F. and Long, S. R. 1992. Rhizobium-p\ant signal exchange. Nature 357: 655-660. Flores, M., Gonzalez, V., Pardo, M. A., Leija, A., Martinez, E., Romero, D., Pinero, D., Davila G., and Palacios, R. 1988. Genomic instability in Rhizobium phaseoli. J. Bacteriol. 170:1191-1196. Gibson, A.H. 1987. Evaluation of nitrogen fixation by legumes in the greenhouse and growth chamber. In Symbiotic nitrogen fixation technology. Edited by G. H. Elkan. Marcel Dekker, Inc., U.S.A. pp. 321-369. 133 Goethals, K., Mergaert, P., Geelen, D., Desomer, J., van-Montagu, M, Holster, M. 1993. Signalling in interactions between plants and bacteria. In Plant signals in interactions with other organisms. Edited by J . Schultz & I. Raskin. Am. Soc. Plant Physiol, pp. 153-163. Graham, P. H., Sadowsky, M. J., Keyser, H. H., Barnet, Y. M., Bradley, R. S., Cooper, J. E., de Ley, D. J., Jarvis, B. D. W., Roslycky, E. B., Strijdom, B. W., and Young, J. P. W. 1991. Proposed minimal standards for the description of new genera and species of root-and stem-modulating bacteria. Int. J. Syst. Bacteriol. 41: 582-587. Gutell, R. R., and Fox, G. E. 1988. A compilation of large subunit RNA sequences presented in a structural format. Nucleic Acids Res. 16(suppl.); r175-r269. Hale, M. G., L. D. Moore, L. D. and Griffin, G. J. 1978. Root exudates and exudation. In Interactions between non-pathogenic soil microorganisms and plants. Edited by Y.R. Dommergues and S.V. Krupa. Elsevier Scientific Publishing Company, Amsterdam, p. 163-203. Harrison, S. P., Mytton, L. R., Skot, L., Dye, M., and Cresswell, A. 1992. Characterization of Rhizobium isolates by amplification of DNA polymorphism using random primers. Can. J. Microbiol. 38: 1009-1015. Hartwig, U. A., Joseph, C. M., Phillips, D. A. 1991. Flavonoids released naturally from alfalfa seeds enhance growth rate of Rhizobium meliloti. Plant Physiol. 95: 797-803. Hartwig, U. A., Maxwell, C. A., Joseph, C. M., Phillips, D. A. 1990. Chrysoeriol and luteolin released from alfalfa seeds induce nod gene in Rhizobium meliloti. Plant Physiol. 92: 116-122. Heidstra, R., Geurts, R., Franssen, H., Spaink, H. P., van Kammen, A., and Bisseling, T. 1994. Root hair deformation activity of nodulation factors and their fate on Vicia sativa. Plant Physiol. 105:787-797. 134 Heller, W., and Forkman, G. 1988. Biosynthesis, pp. 399-425. In The flavonoids: advances in research since 1980. Edited by J.B. Harborne. Chapman and Hall Ltd., London. Hernandez-Lucas, I., Segovia, L, Martinez-Romero, E., Pueppke, S. G. 1995. Phylogenetic relationships and host range or Rhizobium spp. that nodulate Phaseolus vulgaris L. Appl. Environ. Microbiol. 61:2775-2779. Hirsch, A. M. 1992. Developmental biology of legume nodulation. New Phytol. 122: 211-237. Holl, F. B., Chanway, C. P., Turkington, R., and Radley, R. A. 1988. Response of crested wheat grass (Agropyron cristatum L.), perennial rye grass (Lolium perenne) and white clover (Trifolium repens L.) to inoculation with Bacillus polymyxa. Soil Biol. Biochem. 20:19-24. Horvath, B., Bachem, C. W., Schell, J., and Kondorosi, A. 1987. Host-specific regulation of nodulation genes in Rhizobium is mediated by a plant signal interacting with the nodD gene product. EMBO J. 6:841-848. Huang, S. S., Djordjevic, M. A., and Rolfe, B. G. 1995. Microscopic analysis of the effect of Rhizobium leguminosarum biovar trifolii host specific nodulation genes in the infection of white clovers. Protoplasma 172:180-190. Huber, I. and Selenska-Pobell, S. 1994. Characterization of Rhizobium galegae by REP-PCR, PFGE and 16S rRNA sequencing. In Proceedings of the 14 th North American Conference on symbiotic nitrogen fixation. Eds. P.H.Graham, M.J. Sadowsky and C P . Vance. Kluwer Academic Publishers, Dordrecht, p. 153-158. Hungaria, M., Joseph, C. M., and Phillips, D. A. 1991a. Anthocanidins and flavonols, major nod gene inducers from seeds of a black-seeded common bean (Phaseolus vulgaris L.). Plant Physiol. 97 : 751-758. 135 Hungaria, M., Joseph, CM., and Phillips, D. A. 1991b. Rhizobium nod gene inducers exuded naturally from roots of common bean (Phaseolus vulgaris L) . Plant Physiol. 97 : 759-764. Innes, R. W., Kuempel, P. L., Plazinski, J., Canter-Cremers, H., and Rolfe, B. G. 1985. Plant factors induce expression of nodulation and host-range gene in Rhizobium trifolii. Mol. Gen. Genet. 201: 426-432. Jagadish, M., and Szalay, A. A. 1985. Directed transposon Tn5 mutagenesis and complementation in slow growing broad-host range Cowpea Rhizobium. Mol. Gen. Genet. 196: 290-300. Jarvis, B. D. W., Downer, H. L., and Young, J. P. W. 1992. Phylogeny of fast-growing soybean-nodulating rhizobia supports synonymy of Sinorhizobium & Rhizobium and assignment to Rhizobium fredii. Int. J. Syst. Bacteriol. 42:93-96. Jarvis, B. D. W., Dick, A. G., and Greewood, R. M. 1980. Deoxyribonucleic acid homology among strains of Rhizobium trifolii and related species. Int. J. Syst. Bacteriol. 30: 42-52. Jarvis, B.D.W., Pankhurst C. E. and Patel, J. J. 1982. Rhizobium loti, a new species of legume root nodule bacteria. Int. J. Syst. Bacteriol. 32:378-380. Jensen, H. L. 1962. Proc. Linn. Soc. N.S.W. 67:205 (Cited bv Gibson 1987). Jones, R. J. and Jones, R. M. 1982. Observation on the persistence and potential for beef production of pastures based on Trifolium semipilosum and Leucaena leucocephala in subtropical and coastal Queensland. Trop. Grassl. 16:24-29. Johnston, A. W. B., and Beringer J. E. 1975. Identification of the Rhizobium strains in pea root nodules using genetic markers. J. Gen. Microbiol. 87:343-350. 136 Jordan, D. C. 1984. Family III. Rhizobiaceae Conn 1938. In Bergey's Manual of Systematic Bacteriology, Vol I. Edited by N.R.Kreig and J.G.Holt. Williams and Wilkins Co., Baltimore, p.234-254. Judd, A. K., Schneider, M., Sadowsky, M. J. and De Bruijn, F. J. 1993. Use of repetitive sequences and the polymerase chain reaction technique to classify genetically related Bradyrhizobium japonicum serocluster 123 strains. Appl. Environ. Microbiol. 59: 1702-1708. Jukes, T. H., and Cantor, C. R. 1969. Evolution of protein molecules. In Mammalian protein metabolism. Edited by H.N. Murro. Academic Press, Inc., New York. p. 21-132. Kapp, D., Niehans, K., Quandt, J., Mtiller, P., and Puhler 1990. Cooperative action of R. meliloti nodulation and infection mutants during the process of forming mixed infected alfalfa nodules. Plant Cell 2:139-151. Kundig, C , Beck, C , Hennecke, H., and Gottfert, M. 1995. A single rRNA gene region in B. japonicum. J . Bacteriol. 177:5151-5154. Laguerre, G., Fernandez, M. P., Edel, V, Normand, P., and Amarger, N. 1993. Genomic heterogeneity among French Rhizobium strains isolated from Phaseolus vulgaris. Int. J . Sys. Bacteriol. 43:761-767. Laguerre, G., Allard, M. R., Revoy, F. and Amarger, N. 1994. Rapid identification of Rhizobia by restriction fragment length polymorphism analysis of PCR-amplified 16S rRNA genes. Appl. Environ. Microbiol. 60:56-63. Laguerre, G., Mavingui, P., Allard, M. R., Charnay, M. P., Louvrier, P., Mazurier, S. I., Rigottier-Gois, L., and Amarger, N. 1996. Typing of rhizobia by PCR DNA fingerprinting and PCR-restriction fragment length polymorphism analysis of chromosomal and symbiotic gene regions:application to Rhizobium leguminosarum and its different biovars. Appl. Environ. Microbiol. 62: 2029-2036. 137 Lane, D. J. 1991. 16S/23S rRNA sequences. In Nucleic acid techniques in bacterial systematics. Edited by Stackebrandt and M. Goodfellow. John Wiley and Sons, New York. pp.115-147. Leroughe, P., Roche, P., Faucher, C , Mai I let, F., Truchet, G., Prome, C , and Denarie, J. 1990. Symbiotic host-specificity of Rhizobium meliloti is determined by a sulphated and acylated glucosamine oligosaccharide. Nature 344:781-784. Libbenga, K. R., and Harkes, P. A. A. 1973. Initial proliferation of cortical cells in the formation of root nodules in Pisum sativum. Planta 114:17-28. Lindstrom, K. 1989. Rhizobium galegae, a new species of legume root nodule bacteria. Int. J . Syst. Bacteriol. 39: 365-367. Long, S. R. 1985. Genetics of Rhizobium nodulation. In Plant-microbe interactions. Vol. 1. Edited by T. Kosuge and E.W. Nester. Macmillan, New York. p. 265-306. Long, S. R. 1989. Rhizobium genetics. Annu. Rev. Genet. 23:483-506. Lopez-Lara, I. M., van den Berg, J. D. J., Thomas-Oates, J. E., Glushka, J., Lugtenberg, B. J. J., and Spaink, H. P. 1995. Structural identification of the lipo-chitin oligosaccharide nodulation signals of Rhizobium loti. Mol. Microbiol. 15:627-638. Lucas, I. H., Segovia, L., Martinez-Romero, E., and Pueppke, S. G. 1995. Phylogenetic relationships and host range of Rhizobium spp. that nodulate Phaseolus vulgaris L. Appl. Environ. Microbiol. 61: 2725-2779. Ludwig, W., Dorn, S., Springer, N., Kirchhof, G., and Schleifer, K. H. 1994. PCR-based preparation of 23S rRNA-targeted group-specific polynucleotide probes. Appl. Environ. Microbiol. 60: 3236-3244. Ludwig, W., and Schleifer, K. H. 1994. Bacterial phytogeny based on 16S & 23S rRNA sequence analysis. FEMS Microbiol. Review 15: 155-173. 138 MacKay, J. E. 1973. Register of Australian herbage plant cultivars. 1. Clover. Trifolium semipilosum Fres. Var. glabrescens Gillet (Kenya white clover) cv. Safari. J . Aust. Inst. Agric. Sci. 39:274-276. Maniatis, T., Fritsch, E. F., and Sambrook, J. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor, New York. Martinez-Romero, E. 1994. Recent developments in Rhizobium taxonomy. Plant Soil 161: 11-20. Martinez-Romero, E., Segovia, L., Mercante, F. M., Franco, A. A., Graham, P., and Pardo, M. A. 1991. Rhizobium tropici: a novel species nodulating Phaseolus vulgaris L. beans and Lucaena sp. trees. Int. J . Syst. Bacteriol. 41: 417-426. Maxwell, C. A., Hartwig, U. A., Joseph, C. M., and Phillips, D. A. 1989. A chalcone and two related flavonoids released from alfalfa roots induce nod genes of Rhizobium meliloti. Plant physiol. 91: 842-847. Mclver, J., Djordjevic, M. A., Weinman, J. J., Bender, G. L., and Rolfe, B. G. 1989. Extension of host range of Rhizobium leguminosarum bv trifolii caused by point mutation in nodD that result in alterations in regulatory function and recognition of inducer molecules. Mol. Plant-microbe Interact. 2:97-106. Mulligan, J. T., and Long, S. R. 1985. Induction of Rhizobium meliloti nodC expression by plant exudates requires nodD. Procc Natl Acad. Sci. U.S. 82 : 6609-6613. Mytton, L. R. 1975. Plant genotype x Rhizobium strain interaction in white clover. Ann. Appl. Biol. 80:103-107. Mytton, L. R., and Livesey, C. J. 1983. Specific and general effectiveness of Rhizobium trifolii populations from different agricultural locations. Plant Soil 73:299-305. 139 Nap, J. P., and Bisseling, T. 1990. Developmental biology of a plant-prokaryote symbiosis: The legume root nodule. Science 250:948-954. Nei, M., and Li, W. H. 1979. mathematical model for studying genetic variation in terms of restriction endonucleases. Proc. Natl. Acad. Sci. U.S.A. 76:5269-5273. Nei, M. 1987. Molecular evolutionary genetics, Columbia University Press, New York. Newcomb, W. 1981. Nodule morphogenesis and differentiation. Int. Rev. Cytol. Suppl. 13:246-298. Otten, L., and de Ruff ray, P. 1996. (Institute of Plant Molecular Biology of the CNRS, Strasbourg, France) Unpublished data. Otten, L., de Ruffray, P., de Lajudie, P., and Michot, B. 1996. Sequence characterization of a ribosomal RNA operon from Agrobacterium vitis. Mol. Gen. Genet. 251:99-107. Peters, N. K., Frost, J. W., and Long, S. R. 1986. A plant flavone, luteolin, induces expression of R. meliloti nodulation genes. Science 233: 977-980. Peters, N. K., and Long, S. R. 1988. Alfalfa root exudates and compounds which promote or inhibit induction of Rhizobium meliloti nodulation genes. Plant Physiol. 88:396-400. Petersen, D. J., Shishido, M., Holl, F. B., and Chanway, C. P. 1995. Use of species-and strain-specific PCR primers for identification of conifer root-associated Bacillus spp. FEMS Microbiol. Lett. 133: 71-76. Phillips, D. A., Dakora, F. D., Sande, E., Joseph, C. M., and Zon, J. 1994. Synthesis, release and transmission of alfalfa signals to rhizobial symbionts. Plant Soil 161 : 69-80. 140 Pooler, M. R., Ritchie, D. F., and Hartung, J. S. 1996. Genetic relationships among strains of Xanthomonas fragariae based on random amplified polymorphic DNA PCR, repetitive extragenic palindromic PCR, and enterobacterial repetitive intergenic consensus PCR data and generation of multiplexed PCR primers useful for the identification of this phytopathogen. Appl. Environ. Microbiol. 62:3132-3127. Qian, L., and Wilkinson, M. 1993. DNA fragment purification: removal of agarose ten minutes after electrophoresis. Biotechniques 10: 736-738. Recourt, K., Schripsema, J., Kijne, J. W., van Brussel, A. A. N., and Lugtenberg, B. J. J. 1991. Inoculation of Vicia sativa subsp. nigra roots with Rhizobium leguminosarum bv. viciae results in release of nod gene activating flavonones and chalcones. Plant Mol. Biol. 16: 841-852. Redmond, J. W., Batley, M., Djordjevic, M. A., Innes, R. W., Kumpel, P. L., and Rolfe, B. G. 1986. Flavones induce expression of nodulation genes in Rhizobium. Nature 323:632-635. Rennie, R. J. 1981. A simple medium for the isolation of acetylene-reducing (dinitrogen- fixing) bacteria from soils. Can. J . Microbiol. 27:8-14. Robinson, A. 1969. Host selection for effective Rhizobium trifolii by red clover and subterranean clover. Aust. J . Agric. Res. 20:1053-1060. Roche, P., Debelle, F., Maillet, F., Lerouge, P., Faucher, C , Truchet, G., Denarie, J., and Prome, J.C. 1991. Molecular basis of symbiotic host specificity in Rhizobium meliloti: nodH and nodPQ genes encode the sulfation of lipo-oligosaccharide signals. Cell 67:1131-1143. Rolfe, B. G., Batley, M., Redmond, J. W., Richardson, A.E ., Simpson, R. J., Basasm, B J . , Sargent, L. L, Weinman, J. J., Djordjevic, M. A., and Dazzo, F. B. 1988. Phenolic compounds secreted by legumes. In Nitrogen fixation hundred years after. Edited by Bothe, de Bruijn and Newton. New York. p. 405-408. 141 Rolfe, B. G., Gresshoff, P. M., and Shine, J. 1980. Rapid screening for symbiotic mutants of Rhizobium and white clover. Plant Sci. Lett. 19:277-284. Rossen, L., Sherman, C. A., Johnston, W. B., and Downie, J. A. 1985. The nodD gene of Rhizobium leguminosarum is autoregulated and in the presence of plant exudate induces the nodA, B, C genes. EMBO J. 4:3369-3373. Saitou, N., and Nei, M. 1987. The neighbour-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4: 406-425. SAS Institute 1989. Software release for Windows version 6.08, SAS Institute Inc., Cary, N.C., U.S.A. Sass, J. E. 1958. Botanical microtechnique. The Iowa State College Press, Ames, Iowa. p. 721. Schmidt, P. E., Parniske, M., and Werner, D. 1991. Production of the phytoalexin glyceollin I by soybean roots in response to symbiotic and pathogenic infection. Bota Acta 105: 18-25. Schmidt, P. E., Broughton, W. J., and Werner, D. 1994. Nod factors of Bradyrhizobium japonicum and Rhizobium sp. NGR234 induce flavonoid accumulation in soybean root exudate. Mol. plant-microbe inter. 7: 384-390. Scholia, M. H., and Elkan G. H., 1984. Rhizobium fredii sp. nov., a fast-growing species that effectively nodulates soybeans. Int. J. Syst. Bacteriol. 34: 484-486. Segovia, L, Young, P. W., and Martinez-Romero, E. 1993. Reclassification of American Rhizobium leguminosarum biovar phaseoli type I strains as Rhizobium etli sp. nov. Int. J. Syst. Bacteriol. 43:374-377. Selenska-Pobell, S. 1994. How to monitor released bacteria. Plant Soil, 166: 187-191. 142 Sessitsch, A., Wilson, K. J., Akkermans, A. D. L., and de Vos, W. M. 1996. Simultaneous detection of different Rhizobium strains marked with either the Escherichia coli gusA gene or the Pyrococcus furiosus ce/B gene. Appl. Environ. Microbiol. 62:4191-4194. Shishido, M., Loeb, B. M., and Chanway, C. P. 1995. External and internal root colonization of lodgepole pine seedlings by two growth-promoting Bacillus strains originated from different root microsites. Can. J. Microbiol. 41:707-713. Somasegaran, P., and Hoben, H. J. 1994. Handbook for Rhizobia. Springer Laboratory, Springer-Verlag, New York. Spaink, H.P. 1992. Rhizobial lipo-oligosacchardes: answers and questions. Plant Mol. Biol. 20:977-987. Spaink, H. P., Sheeley, D. M., van Brussel, A. A. N., Glushka, J., York, W. S., Tak, T., Geiger, O., Kennedy, E. P., Reinhold, V. N., and Lugterberg, B. J. J. 1991. A novel highly unsaturated fatty acid moiety of lipo-oligosaccharide signals determines host specificity of Rhizobium. Nature 354:125-130. Spaink, H. P., Wijffeleman, C. A., Pees, E., Okker, R. J. H., and Lugtenberg, B. J. J. 1987. Rhizobium nodulation gene nodD as a determinant of host specificity. Nature 328:337-340. Sprent, J.I. 1979. The biology of nitrogen-fixing organisms. McGraw-Hill Book Company, U.K. p. 196. Sprent, J. I. 1994. Evolution and diversity in the \egume-Rhizobium symbiosis:chaos theory? Plant Soil 161:1-10. Springer, N., Ludwig, W., and Hardarson, G. 1993. A 23S rRNA targeted specific hybridization probe for B. japonicum. Syst. Appl. Microbiol. 16:468-470. 143 Srinivasan, M. 1996. Changes in root hair morphogenesis and nodulation of Phaseolus vulgaris L. in the presence of Bacillus and Rhizobium. Ph.D. thesis, University of British Columbia, Canada. Strain, S. R., Leung, K., Whittam, T., de Bruijn, F., and Bottomley, P. J. 1994. Genetic structure of Rhizobium leguminosarum biovar trifolii and viciae populations in two Oregon soils under different plant communities. Appl. Environ. Microbiol. 60: 2772-2778. SYSTAT Inc. 1992. Sotware release for Windows version 5.0., Evanston, IL, U.S.A. Truchet, G., Camut, S., de Billy, F., Odorico, R., and Vasse, J. 1989. The Rhizobium-\egume symbiosis: two methods to discriminate between nodules and other root-derived structures. Protoplasma 149: 82-88. Truchet, G., Michel, M., and Denarie, J. 1980. Sequential analysis of organogenesis of lucerne (Medicago sativa) root nodules using symbiotically-defective mutants of Rhizobium meliloti. Differentiation 16: 163-172. Truchet, G., Roche, P., Lerouge, P., Vasse, J., Camut, S., de Billy, F., Prome, and Denerie, J. 1991. Sulphated lipo-oligosaccharide signals of the symbiotic prokaryote Rhizobium meliloti elicit root nodule organogenesis on the host plant Medicago sativa. Nature 351:670-673. Turgeon, B. G., and Bauer, W. D. 1982. Early events in the infection of the soybean by Rhizobium japonicum. Time course cytology of the initial infection process. Can. J. Bot. 60:152-161: Vahjen, W., Munch, V. J., and Tebbe, C. C. 1995. Carbon source utilization of soil extracted microorganisms as a tool to detect the effects of soil supplemented with genetically engineered and non-engineered Corynebacterium glutamicum and a recombinant peptide at the community level. FEMS Microbiol. Ecol. 18:317-328. 144 van Berkum, P., Desta Beyene, and Eardly, B. D. 1996. Phylogenetic relationships among Rhizobium species nodulating the common bean (Phaseolus vulgaris L) . Int. J . Syst. Bacteriol. 46:240-244. Vasse, J., and Truchet, G. 1984. The Rhizobium-legume symbiosis: observation of root infection by bright-field microscopy after staining with methylene blue. Planta 161:487-489. Versalovic, J., Koeuth, T. and Lupski, J. R. 1991. Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genome. Nucleic Acids Res. 19: 6823-6831. Vickery, M. L. and Vickery, B. 1981. Secondary Plant Metabolism. MacMillan Press, London. Vincent, J. M. 1970. A manual for the practical study of root-nodule bacteria. IBP handbook No. 15. Blackwell Scientific Publications, Oxford. Vincent, J. M. 1974. Root-nodule symbiosis with Rhizobium. In The biology of nitrogen fixation. Edited by Quispel. North-Holland Publishing company, Amsterdam, p. 265-341. Vincent, J. M. 1980. Factors controlling the \egume-Rhizobium symbiosis. In Nitrogen fixation, vol. II, Edited by W.E. Newton and W.H. Orme-Johnson, University Park Press., Baltimore, pp. 103-129. Weinman, J. J., Djordjevic, M. A., Howies, P. A., Anoli, T., Henderson, W. L., Mclver, J., Oakes, M., Creaser, E. H., and Rolfe, B. G. 1991. The use of the genus Trifolium for the study of plant-microbe interactions. In Advances in molecular genetics of plant microbe interactions, volume I. Edited by H. Hennecke and D.P.S. Verma. Kluwer Academic Publishers, Dordrecht, Netherlands, pp. 168-173. Willems, A., and Collins, M. D. 1993. Phylogenetic analysis of Rhizobium and Agrobacterium based 16S rRNA gene sequences. Int. J . Syst. Bacteriol. 43:305-313. 145 Williams, W. H. 1987. White clover taxonomy and biosynthesis.p.323-342. In white clover. Edited by M.J. Baker and W.M. Williams.C.A.B. Abstracts, U.K. Wilson, K. J., Sessitsch, A., Corbo, J. C , Giller, K. E., Akkermans, A. D. L., and Jefferson, R. A. 1995. ^ -Glucuronidase (GUS) transposons for ecological and genetic studies of rhizobia and other Gram-negative bacteria. Microbiol. 141:1691-1705. Winnepennickx, B., Van de Peer, Yves, Backeljau, T., and De Wachter, R. 1995. CARD: a drawing tool for RNA secondary structure models. Biotechniques 18: 1060-1063. Yao, P. Y., and Vincent, J. M. 1976. Host specificity in the root hair "curling factor" of Rhizobium spp. Aust. J . Biol. Sci. 22:413-423. Young, J. P. W., Downer, H. L., and Eardly, B. D. 1991. Phylogeny of the phototrophic Rhizobium strain BTAM by polymerase chain reaction-based sequencing of a 16S rRNA gene segment. J. Bacteriol. 173: 2271-2277. Young, J. P. W. and Johnston, A. W. B. 1989. The evolution of specificity in the \egume-Rhizobium symbiosis. Trends Ecol. Evol. 4: 341-349. Zak, J. C , Willig, M. R., Moorhead, D. L., and Wildman, H. G. 1994. Functional diversity of microbial communities: a quantitative approach. Soil Biol. Biochem. 26:1101-1108. Zohary, M., and Heller, D. 1984. The genus Trifolium. The Israel Academy of Sciences and Humanities, Jerusalem. Zuker, M., and Stiegler, P. 1981. Optimal computer folding of large RNA sequences using thermodynamics and auxiliary information. Nucleic Acid Res. 9: 133-148. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088128/manifest

Comment

Related Items