UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Changes in root hair morphogenesis and nodulation of Phaseolus vulgaris L. in the presence of bacillus… Srinivasan, Murali 1996

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

Item Metadata

Download

Media
831-ubc_1996-148424.pdf [ 9.11MB ]
Metadata
JSON: 831-1.0087865.json
JSON-LD: 831-1.0087865-ld.json
RDF/XML (Pretty): 831-1.0087865-rdf.xml
RDF/JSON: 831-1.0087865-rdf.json
Turtle: 831-1.0087865-turtle.txt
N-Triples: 831-1.0087865-rdf-ntriples.txt
Original Record: 831-1.0087865-source.json
Full Text
831-1.0087865-fulltext.txt
Citation
831-1.0087865.ris

Full Text

CHANGES IN ROOT HAIR MORPHOGENESIS AND NODULATION OF Phaseolus vulgaris L. IN THE PRESENCE OF BACILLUS AND RHIZOBIUM by MURALI SRINIVASAN B.Sc, Kerala Agricultural University, 1986 M.Sc, Kerala Agricultural University, 1990 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 September 1996 © Murali Srinivasan, 1996 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 The University of British Columbia Vancouver, Canada Date ^ 7 fcpTtmkr I Ills DE-6 (2/88) ii ABSTRACT Twenty two Bacillus spp. isolates from the Phaseolus vulgaris rhizosphere were identified using Biolog™, GC-FAME and 23S rDNA analysis. Some Bacillus isolates produced copious amounts of indoleacetic acid in vitro, in culture medium supplemented with L-tryptophan. Spontaneous cc-methyl tryptophan resistant mutants of Bacillus megaterium S49 exhibited altered IAA production and excreted tryptophan into the culture medium. Coinoculation of Rhizobium etli TAL 182 with some Bacillus spp. under gnotobiotic conditions promoted nodulation of P. vulgaris. In contrast, decreased nodulation was observed in response to low and high lAA-producing Bacillus mutants, suggesting that the nodulation response in bean was not dependent on the in vitro level of IAA production by the Bacillus coinoculant. Coinoculation of Bacillus spp. and Rhizobium spp. resulted in a unique and significant synergistic enhancement of lateral root formation and root hair initiation which was independent of plant host. These observations provided direct evidence for a beneficial interaction on root morphology which may account for enhanced nodulation by creating additional infection sites for the Rhizobium. The presence of live Rhizobium and Bacillus was necessary for the root hair proliferation response. Coinoculation of Rhizobium TAL 182 with cell extracts of Bacillus S49 failed to induce root hair proliferation or enhance nodulation of P. vulgaris. A biologically active Rhizobium TAL 182 was essential for the observed response, indicating that root hair proliferation and enhanced nodulation were direct consequences of the bacterial species interaction. Nodulation and symbiotic specificity of Rhizobium TAL 182 were altered by the presence of Bacillus S49. Split-root experiments showed that coinoculation suppressed Ill host-controlled regulation of nodulation. Bacillus S49 facilitated heterologous nodulation of Rhizobium TAL 182 on P. acutifolius, indicating that a host plant response was also involved in the bacterial interaction. Exogenous aminoethoxy vinyl glycine, a negative regulator of ethylene biosynthesis, inhibited root hair formation in P. vulgaris. These inhibitory effects were reversed by the presence of both Bacillus S49 and Rhizobium TAL 182. The interaction between bacterial species on root hair induction and nodulation patterns in P. vulgaris is discussed. iv TABLE OF CONTENTS Page Abstract ii List of Tables vi List of Figures viii Appendix x Abbreviations xi Acknowledgments xii Foreword xiii Chapter 1 General Introduction 1 Chapter 2 Influence of IAA producing Bacillus isolates on the nodulation of Phaseolus vulgaris by Rhizobium etli 10 2.1. Introduction 11 2.2. Materials and Methods 12 2.3. Results 18 2.4. Discussion 39 Chapter 3 Synergistic interaction between Rhizobium etli and Bacillus megaterium enhances root hair formation in Phaseolus spp. and in Arabidopsis thaliana 44 3.1. Introduction 45 3.2. Materials and Methods 46 3.3. Results 49 3.4. Discussion 58 Chapter 4 Synergistic interaction between Rhizobium and Bacillus alters root hair morphogenesis and enhances nodulation in Phaseolus vulgaris 61 4.1. Introduction 62 4.2. Materials and Methods 64 4.3. Results 69 4.4. Discussion 83 Chapter 5 Nodulation of Phaseolus vulgaris by Rhizobium etli is enhanced by the presence of Bacillus spp 91 5.1. Introduction 92 5.2. Materials and Methods 94 5.3. Results 100 5.4. Discussion 114 Chapter 6 Altered root hair formation in Phaseolus vulgaris in the presence of Rhizobium, Bacillus and the ethylene biosynthesis inhibitor amino ethoxy vinyl glycine 119 V Page 6.1. Introduction 1 1 8 6.2. Materials and Methods 1 2 0 6.3. Results 1 2 3 6.4. Discussion ^ 2 8 Chapter 7 Summary 1 3 2 7.1. Proposal of a working hypothesis for future research 135 Bibliography 1 4 2 Appendix 1 5 6 vi LIST OF TABLES Page Table 2.1. Identification and origin of Bacillus isolates used in this study 22 Table 2.2. Indoleacetic acid (IAA) production by Bacillus isolates and Rhizobium etli 23 Table 2.3. Growth of Bacillus megaten'um S49 in the presence of 2.5mM a-methyl tryptophan 30 Table 2.4. Influence of Bacillus isolates on nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) of P. vulgaris by Rhizobium etli (TAL 182) in the greenhouse 33 Table 2.5. Influence of Bacillus isolates on nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) of P. vulgaris by Rhizobium etli (TAL 182) in the greenhouse 34 Table 2.6. Influence of Bacillus isolates on nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) of P. vulgaris by Rhizobium etli (TAL 182) in the greenhouse 35 Table 2.7. Effect of Bacillus isolates on nodulation of Phaseolus vulgaris cv. 'Contender1 inoculated with Rhizobium TAL 182 36 Table 2.8. Influence of Bacillus isolates on nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) of P. vulgaris by Rhizobium etli (TAL 182) under gnotobiotic conditions in the growth chamber 37 Table 2.9. Root fresh weight and root dry weight of P. vulgaris due to coinoculation of R. etli TAL 182 with B. megaten'um S49 or a-methyl tryptophan resistant mutants of S49 under gnotobiotic conditions in the growth chamber 40 Table 3.1. Lateral root production on five day old seedlings of Phaseolus vulgaris treated with indoleacetic acid (IAA), or inoculated with Bacillus megaterium (S49 or S49-1D) and/or Rhizobium etli (TAL 182) 57 Table 4.1. Effect of Rhizobium and Bacillus inoculation on lateral root formation in Phaseolus vulgaris grown in Leonard jars under gnotobiotic conditions 80 VII Page Table 4.2. Influence of Rhizobium and Bacillus inoculations on lateral root formation in Phaseolus vulgaris grown in test tubes under axenic conditions 81 Table 4.3. Influence of cell-free extracts of Rhizobium and Bacillus on lateral root formation in Phaseolus vulgaris grown in test tubes under axenic conditions 82 Table 5.1. Nodulation suppression induced by R. etli TAL 182 with and without B. megaterium S49 as co-inoculant on P. vulgaris 'Contender" 103 Table 5.2. Effect of bacterial inoculation on nodule number (NON), nodule fresh weight (NFW), and nitrogenase activity (ARA) in Phaseolus acutifolius under gnotobiotic conditions in Leonard jars 106 Table 5.3. Nodulation of P. vulgaris by R. etli TAL 182 in the presence of cell preparations of B. megaterium S49 110 Table 5.4. Nodulation of P. acutifolius 24 days after inoculation with R. etli TAL 182 alone or with B. megaterium S49 111 Appendix I. Preliminary results on the detection of cytokinins in the cell free super-natant of some Bacillus isolates 157 Appendix II. Mean number of lateral roots on Phaseolus vulgaris three and five days after the germinants were inoculated with Rhizobium etli TAL 182 and Bacillus polymyxa L6 160 LIST OF FIGURES VIII Page Figure 2.1. Phylogenetic dendrogram showing the diversity of IAA producing Bacillus isolates 19 Figure 2.2. The avena coleoptile straight growth assay for auxin-like activity in Bacillus isolates extracts 25 Figure 2.3. HPLC separation of (A) authentic IAA standard at 5 ug/ml, (B) IAA fraction in the methanol extract of B. megaterium S49, (C) TLC eluted fraction Rf 0.4-0.6 of B. megaterium S49 and, (D) co-chromatography of the TLC eluted band of B. megaterium S49 (R, 0.4-0.6) with an authentic IAA eluted from TLC at R, 0.4-0.6 . . . 26 Figure 2.4. Mass spectrometric span of (A) 100 picograms of an authentic indole acetic acid standard and, (B) thin layer chromatograph eluted band corresponding to Rf 0.4-0.6 of Bacillus polymyxa L6 27 Figure 2.5. (A) Excretion of L-tryptophan and, (B) production of IAA by wild-type B. megaterium S49 and a-methyl tryptophan resistant mutants of S49 (S491D and S49-1W) 31 Figure 2.6. Effect of coinoculation of a-methyl tryptophan resistant mutants of Bacillus megaterium S49 with Rhizobium etli TAL 182 on nodule number (NON), nodule fresh weight (NFW), and nitrogenase activity (ARA) in P. vulgaris'Couleudef 38 Figure 3.1. Root hair production by P. vulgaris in response to bacterial coinoculation 50 Figure 3.2. Root hair production by P. acutifolius in response to bacterial coinoculation 53 Figure 3.3. Root hair production by Arabidopsis thaliana in response to bacterial coinoculation 55 Figure 4.1. Schematic diagram of the cup assay procedure used to facilitate physical separation of bacterial species on plant roots 67 Figure 4.2. Root hair production by P. vulgaris in response to dual inoculation of Rhizobium etli (TAL 182) with different Bacillus spp 70 Figure 4.3. Root hair production by P. vulgaris in response to inoculation with cell extracts of Rhizobium etli TAL 182 and B. megaterium S49 73 ix Page Figure 4.4. Root hair production by P. vulgaris in response to inoculation with Tn5 mutant (MLC 35) of Rhizobium etli TAL 182 and B. megaten'um S49 76 Figure 4.5. Analysis of bacterial populations in cup assay experiments 78 Figure 4.6. Effect of dual inoculation of Rhizobium etli TAL 182 and Bacillus megaten'um S49 on (A) nitrogenase activity (B) nodule number and, (C) nodule fresh weight of P. vulgaris grown under gnotobiotic conditions in the growth chamber 84 Figure 5.1. Root hair formation in P. vulgaris in response to bacterial inoculation 101 Figure 5.2. Rate of new nodule formation on Phaseolus vulgaris 'Contender1 in growth pouches when inoculated (A) with R. etli TAL 182 alone and, (B) R. etli TAL 182 with B. megaterium S49 105 Figure 5.3. Effect of timing of inoculation of Rhizobium etli TAL 182 and B. megaterium S49 on (A) nodule number, (B) nodule fresh weight and, (C) nitrogenase activity of P. vulgaris •Contender1 grown under gnotobiotic conditions in the growth chamber 107 Figure 5.4. Detection of heterologous nodulation of P. acutifolius by R. etli TAL 182 in the presence of B. megaterium S49 by using fluorescent antibodies specific to Rhizobium TAL 182 112 Figure 5.5. Presence of R. etli TAL 182 inside the surface-sterilized nodules of P. acutifolius as determined by multiplex PCR 115 Figure 6.1. Root hair formation in P. vulgaris in response to inoculation with Rhizobium, Bacillus and treatment with 1-aminocyclopropane-1-carboxylic acid (ACC) 124 Figure 6.1. Root hair formation in P. vulgaris in response to inoculation with Rhizobium, Bacillus and treatment with aminoethoxy vinyl glycine 126 Appendix I. Effect of dual inoculation of R. etli TAL 182 and B. polymyxa L6 on (A) nodule number (B) nodule fresh weight and, (C) nitrogenase activity of P. vulgaris grown under gnotobiotic conditions in the growth chamber 158 Appendix V. Efect of timing of inoculation of R. etli TAL 182 and B. polymyxa 6 on (A) nodule number, (B) nodule fresh weight and, (C) nitrogenase activity of P. vulgaris 'Contender1 grown under gnotobiotic conditions in the growth chamber 162 X APPENDIX Page Appendix IV. Calculation of percent suppression in split-root experiments 161 LIST OF ABBREVIATIONS ACC 1 -aminocyclopropane-1 -carboxylic acid ACE autoclaved cell preparations ARA acetylene reduction assay AVG aminoethoxy vinyl glycine CCM combined carbon medium CFE cell-free extract CFU colony forming units Da daltons GC-FAME gas chromatography-fatty acid methyl ester GC-MS gas chromatography-mass spectrometry HPLC high performance liquid chromatography IAA indoleacetic acid 2iP 2-isopentyl adenine MW molecular weight kD kilo daltons MHB mycorrhization helper bacteria NaOAc sodium acetate NFW nodule fresh weight NifTAL nitrogen fixation in tropical agricultural legumes NON nodule number NPR nodulation promoting rhizobacteria PCR polymerase chain reaction PGPR plant-growth promoting rhizobacteria SEM scanning electron microscopy TIBA triiodo benzoic acid t Z trans zeatin t ZR trans zeatin riboside WCE Whole (French pressed) cell extract xii Acknowledgments This thesis is a testimony to the support of my family and friends throughout my academic career. I owe the greatest thanks to my wife Elisa Ulrika Stabell for her compassion, kindness and support. I am at loss for words to define the many sacrifices she has made to enable me to complete my degree. My love and gratitude for her, in return, is boundless. Many thanks to my family for keeping their faith in my abilities. Special thanks are due to my uncle and aunt for unflinching support and also to my in-laws, Eva and Charles for their support and wonderful Christmas holidays. The most endearing part of doing my Ph.D. has been the people I have become acquainted with during these years. Many thanks to my dearest friend, Christof Rampitsch for his help, support, wonderful friendship and for many pints of beer and e-mail correspondence that we have had over the past five years. His humour and wisdom has helped me to maintain my sanity during these years. Brian Holl is one of those people who has not only been a good friend, but also a great mentor. I owe a great deal to Brian Holl without whom this thesis would never have seen the light of the day. He has given me the unique opportunity to think and work independently on my research at the same time providing guidance and direction without condescension. He has shown me the very human face of science and it has been wonderful five years working with him. He has installed in me a sense of confidence to become a good scientist in the years to come. I am indeed very lucky to have wonderful people as my committee members. Many thanks to Chris Chanway for his critical comments, helpful suggestions, encouragement and goodwill during the course of this study. A big thanks to Jim Kronstad for his gracious support, help, encouragement, critical comments and for introducing me to the field of molecular biology. I am very fortunate to have worked closely with Robert Copeman both as a graduate student and as a teaching assistant. His compassion, constant encouragement, and support has played a vital role in the completion of my thesis. I also thank Brian Ellis for all his help and goodwill. It has been a revelation and great pleasure to have collaborated with Daniel Petersen during the past two years. Many thanks to him for his help with PCR, SEM, critical analysis of my experiments and for his friendship. Many thanks to Hugues Massicotte for his interesting ideas and suggestions, for sharing his vast knowledge and also for his friendship. I thank Dr. George Eaton, Soenarto, Dean Spanner and Shibairo for demystifying statistical analysis for me, Elaine Humphrey for her help with SEM and Mesfin Tesfaye for many hours of friendly debating and discussions. I thank Pat Warne, Shirley Warne, Thomas Vogt, Kevin Spollen, Bjom Orvar, Jacques Grandmaison, and Tony Millar for being great comrades and for having played an important role in keeping me physically fit as well as making sure of, the demise of my liver! Many thanks to Debbie Wheeler for being a great friend, a good office mate, and for helping me with her artistic talents; to Arthur Yee, Monica Lam, Suprayogi, Kate Sircom, Daryl Louie, Linda Tackaberry, Susan Petersen, Darlene O'Neill, Nancy Furness, Julia Rogers, Anne Winkelmann and Grant McKegney for being wonderful friends; to Stefanie Butland for her friendship, interesting intellectual conversations, encouragement and critical reading of manuscripts; to Jennifer for the swimming lessons; to Pam Scales, Len Ward, Ashley Herath, Leroy Scrubb, Bev Busch, Robyne Allen and Derek White for all their help and well wishes; to my friend Veronica Anthony for introducing me to Canada and to the beautiful city of Vancouver. Foreword Portions of this thesis have been submitted for publication as follows: (i) Portions of Chapter 2 to Canadian Journal of Microbiology as "Influence of IAA producing Bacillus isolates on the nodulation of Phaseolus vulgaris by Rhizobium etli under gnotobiotic conditions" by Srinivasan, M., Petersen D.J., and Holl, F.B. This manuscript has been accepted for publication in Canadian Journal of Microbiology. (ii) Portions of Chapter 3 to Planta titled " A unique rhizosphere bacteria synergism promoting root hair initiation" by Srinivasan, M., Petersen D.J., and Holl, F.B. (iii) Portions of Chapter 5 to Canadian Journal of Microbiology titled " Nodulation of Phaseolus vulgaris by Rhizobium etli is enhanced by the presence of Bacillus" by Srinivasan, M., Petersen D.J., and Holl, F.B. This manuscript has been accepted for publication in Canadian Journal of Microbiology. (iv) Portions of Chapter 6 of this thesis have been accepted for publication in Microbiological Research as " Altered root hair morphogenesis in Phaseolus vulgaris in response to bacterial coinoculation and the presence of aminoethoxy vinyl glycine" by Srinivasan, M., Petersen D.J., and Holl, F.B. The senior author initiated the investigations, carried out the majority of the laboratory work and wrote the manuscripts for journal submission. The second author was responsible for 23S rDNA sequencing and PCR analysis and helped in the analysis of the scanning electron microscopy experiments. All work was performed with the financial assistance and research supervision of Dr. Brian Holl. (v) In addition, a portion of the material presented in Chapter 4 of this thesis has been accepted for publication in FEMS Microbiological Letters titled "Bacillus polymyxa stimulates increased Rhizobium etli populations and nodulation when co-resident in the rhizosphere of Phaseolus vulgaris" by Petersen, D.J., Srinivasan, M., and Chanway, C P The senior author initiated the investigation, carried out the laboratory work (bacterial population estimation and PCR analysis) and wrote the manuscript for journal submission. The second author conducted the plant assays described in the paper. This work was performed under the supervision of Dr. CP . Chanway. (Murali Srinivasan) Petersen) (Dr. C. P. Chanway) (Dr. F.B. Holl) 1 Chapter 1 General Introduction Basic research is what I am doing when I do not know what I am doing Wernher von Braun (1912-1977) 2 The term rhizosphere refers to the unique soil environment which is under the influence of plant roots (Hiltner 1904, cited in Curl and Truelove 1986). Development of sustainable agricultural systems necessary to counteract the decline in quality of soils, demands that continuing advances in biological productivity be achieved in a sustainable manner (Schippers et al. 1995). Exploiting the biological potential of microbial species in the rhizosphere may offer an important target for sustained agricultural productivity. Bacteria are the most abundant of soil microorganisms by virtue of their ability to utilize a wide range of carbon and nitrogen sources and to grow rapidly (Glick 1995). While non-rhizosphere field soil may contain up to 100 million bacteria per gram dry weight of soil, root associated soil (rhizosphere) has been shown to contain over a billion bacteria expressed on a comparable basis (Curl and Truelove 1986). Since the majority of soil nutrients used by the plant pass through the rhizosphere, this region has significant potential for biological activity which may affect plant growth. 1.1. Mechanisms of plant-growth promotion by rhizobacteria Plant-growth promoting rhizobacteria (PGPR) are free-living bacteria that are present on or near the roots of plants and are involved in a beneficial association with plants (Kloepper et al. 1988; 1989). PGPR have been isolated from plant parts and from rhizosphere soil, and selected for their ability to improve plant growth, either directly or indirectly (reviewed by Glick 1995). Bacteria belonging to the genera Azospirillum, Bacillus, Pseudomonas, Streptomyces, and Serratia have been identified as PGPR (Kloepper et al. 1989). Indirect effects of PGPR on plant growth include the suppression of deleterious rhizobacteria (DRB) or pathogenic microorganisms (De Freitas and Germida 1991), and the production of iron sequestering compounds (siderophores) (Bakker et al. 1993) or antibiotics (Keel et al. 1992; Thomashow and 3 Weiler 1995). The direct promotion of plant growth by PGPR usually involves either providing the plant with a factor that is produced by the bacterium or by facilitating the uptake of nutrients/substances from the environment (Glick 1995). Direct PGPR effects on plant growth have been described in a variety of studies: (i) Azospirillum spp. enhanced plant growth by providing fixed atmospheric nitrogen (Volpin and Kapulnik 1994); (ii) Azospirillum spp. and Pseudomonas spp. enhanced mineral uptake of the plant (Murty and Ladha 1988; De Freitas and Germida 1992; Germida and De Freitas 1994) ; (iii) Pseudomonas spp. and Bacillus spp. solubilized minerals such as phosphorus so that they are readily available for plant growth (Kloepper et al. 1988); (iv) Pseudomonas spp. produced low molecular weight substances and enzymes leading to changes in plant growth and development (Lambert and Joos 1989; Glick et al. 1994b, 1995) ; (v) Pseudomonas spp. altered plant growth by changing endogenous phytohormonal levels (Glick et al. 1994b; Xie at al. 1996) and; (vi) phytohoromone production by Azospirillum, Bacillus and Pseudomonas enhanced various stages of plant growth (Tien et al. 1979; Kapulnik et al. 1985a; Holl et al. 1988; Chanway et al. 1991; Barbieri and Galli 1993; Xie et al. 1996). It is clear that PGPR are of great interest for their potential to enhance plant growth and to increase productivity in an ecologically sustainable manner. 1. 2. Are mixtures of compatible strains are more effective than single bacterial species in promoting plant growth and/or nodulation ? The root systems of all higher plants are associated with a distinct community of beneficial and non-beneficial microorganisms capable of synthesizing phytohormones (Frankenberger and Arshad 1995). Most work with PGPR has been performed with single strains and under laboratory conditions. However, it is highly unlikely that a 4 single free-living bacterial strain is ever omnipotent in the rhizosphere and thus singularly responsible for enhanced plant growth (Lambert and Joos 1989). In this context, Garbaye (1994) reviewed recent work which indicates the ability of some bacterial species to enhance mycorrhizal root tip formation and subsequently plant (seedling) growth. These bacteria are referred to as mycorrhization helper bacteria (MHB). The ability of rhizobacteria to promote nodulation of legumes by Rhizobium was recognized by many workers and the term nodulation promoting rhizobacteria (NPR) was coined by Kloepper et al. (1988) to describe such microorganisms. Nodulation promoting rhizobacteria belonging to the genera Azospirillum (Schmidt et al. 1988; Yahalom et al. 1990), Pseudomonas (Grimes and Mount 1984; Bolton et al. 1990; De Freitas et al. 1993), Streptomyces (Li and Alexander 1990) and Bacillus (Halverson and Handelsman'1991) have been reported. The beneficial influence of NPR on nodulation of legumes by Rhizobium has been variously attributed to their ability to produce phytohormones (Schmidt et al. 1988), toxins (Knight and Langston-Unkefer 1988), or antibiotics (Li and Alexander 1990), as well as by other unidentified mechanisms (Halverson and Handelsman 1991). Such observations suggest that particular microorganisms have considerable potential to alter the activity of specific rhizosphere microflora such as Rhizobium (Schroth and Becker 1990). Some of the most intimate beneficial interactions between plants and microbes, and between different microbes take place on the surface of the root (Lynch 1988). A variety of complex microbial interactions might be anticipated across the diversity of environments created in the plant rhizosphere. It is unlikely that any one rhizobacterium would be predominant and effective in all environments and hence mixtures of compatible strains might be more significant than a single bacterial species in promoting 5 plant growth. 1. 3. Role of phytohormones in the nodulation of legumes 1.3.1. Role of auxins The role of phytohormones such as auxins and cytokinins in plant organogenesis is well established. Auxins may also be important for maintaining functional root nodules, as effective (nitrogen-fixing) nodules have a higher IAA content than ineffective (nitrogen non-fixing) nodules (Badenoch-Jones et al. 1983). The importance of auxin in nodule organogenesis was confirmed when external application of triiodo benzoic acid (TIBA), an auxin transport inhibitor, induced the formation of pseudonodules on a non-nodulating line of alfalfa (Hirsch et al. 1989). External application of IAA mimicked the effect of Azospirillum on nodulation in Medicago sativa (Schmidt et al. 1988), further implicating auxin in nodule organogenesis. Many Rhizobium spp. are also known to produce IAA in vitro (Ernstsen et al. 1987; Atzorn et al. 1988). While the above studies clearly suggest that bacterially derived IAA could be involved in nodule formation, there is conflicting evidence whether such involvement is essential for nodulation (Wang et al. 1982; Atzorn et al 1988). 1.3. 2. Role of cytokinins Nodule initiation is a property often associated with plant hormones (Hirsch et al. 1989). Cytokinins are more effective than auxins in nodule initiation since they are primarily responsible for stimulating mitosis in plant tissues (Relic et al. 1993). Root nodules of both legumes and non-legumes generally contain elevated amounts of cytokinins which may be either of microbial origin or a response to microbial stimulation of host tissue synthesis (Puppo and Rigaud 1978). Cytokinin of rhizobial origin is involved in nodulation and early nodulin (nodule protein) induction in alfalfa (Long and 6 Cooper 1988). Rhizobium spp. are capable of synthesizing cytokinins in vitro (Morris 1986; Sturtevant and Taller 1989), while many rhizobacteria produce cytokinin-like compounds in vitro (Frankenberger and Arshad 1995) with mixed cultures showing higher production than single cultures (Cacciari et al. 1989). Nevertheless, unequivocal evidence for a direct role for microbially derived cytokinins in nodule formation or organogenesis in legumes has not yet been reported (Dehio and de Bruijn 1992). 1. 3. 3. Role of ethylene Root hairs are tubular-shaped, tip-growing structures that arise from root epidermal cells (Ridge 1995). They serve as sites for rhizobial infection in legumes (Bauer 1981), in addition to enhancing water and nutrient uptake by roots (Clarkson 1985). Root hair formation and differentiation in Arabidopsis are dependent upon the ethylene/auxin ratios in the plant (Masucci and Schiefelbein 1994). Ethylene has also been shown to be a positive regulator of root hair formation in Arabidopsis (Tanimoto et al. 1995). Exogenous application of ethylene has been shown to suppress nodule formation (Lee and LaRue 1992), although this effect can be reversed by exogenous application of the ethylene biosynthesis inhibitor aminoethoxy vinyl glycine (AVG) (Zaat et al. 1989). Many rhizosphere microorganisms including Rhizobium spp. (Billington et al. 1979), Pseudomonas spp. and Bacillus spp. (Mansouri and Bunch 1989) are capable of synthesizing ethylene in vitro, in the presence of the precursor methionine (MET). Inoculation of canola with Pseudomonas putida strain GR12-2 resulted in root elongation, which was attributed to changes in endogenous ethylene levels in the plant (Glick et al. 1994a,b; Xie et al. 1996). Ethylene of microbial origin has been directly implicated in altered growth of etiolated pea seedlings (Arshad and Frankenberger 1988). The ability of microbially derived ethylene to influence nodulation on legumes 7 directly remains unproven. Nevertheless, ethylene-related changes in response to microbial infection leading to enhanced or suppressed nodulation have been observed in Vicia sativa (van Brussel et al. 1986; Zaat et al. 1989). 1. 4. Influence of rhizobacteria on the nodulation of legumes by Rhizobium Genetic and biochemical mechanisms involved in Rhizobium-\egume symbiosis have been well characterized. Studies on the influence of rhizobacteria on nodulation of legumes by rhizobia have primarily focused on the Azospirillum-Rhizobium interaction. In the legume-RA?/zo6/t7/77 system, Azospirillum application promoted plant growth that was generally followed by increased nodule formation (Yahalom et al. 1990; 1991). However, many reports indicate that under gnotobiotic conditions, simultaneous application of Azospirillum and Rhizobium did not always result in a promotion of nodulation, and under some circumstances, even inhibited the ability of Rhizobium to nodulate its host (Plazinski and Rolfe 1985; Volpin and Kapulnik 1994). Enhanced nodulation responses under gnotobiotic conditions were however observed with the application of Azospirillum prior to, or following, inoculation with Rhizobium (Plazinski and Rolfe 1985; Volpin and Kapulnik 1994). In contrast, a positive influence of Azospirillum on nodulation of legumes by Rhizobium in field and greenhouse studies has been observed in response to the simultaneous application of Azospirillum and Rhizobium, indicative of the synergistic associations one might anticipate in natural ecosystems (Okon et al. 1988). These results suggest that simultaneous cell-to-cell interaction(s) between the two bacterial species was not essential for the plant nodulation response, and that excreted bacterial products or soluble factors may be involved in the process. Therefore, it is reasonable to speculate that Azospirillum may exert its influence through 8 the host plant by causing morphological changes in root growth (for example, by inducing the plant to produce more root hairs), and not by a direct interaction with Rhizobium. Evidence implicating phytohormones produced by Azospirillum in enhanced nodulation of legumes by Rhizobium under gnotobiotic conditions has, to date, been circumstantial rather than conclusive. 1.4.2. Role of Bacillus spp. in promoting nodulation of legumes by Rhizobium Bacillus spp. are widely distributed in temperate soils (Priest and Grigorova 1990) and have been implicated in plant growth promoting activity (Holl et al. 1988; Chanway et al. 1991; Shishido et al. 1996). Bacillus cereus and Bacillus subtilis have shown nodulation promotion of legumes with Rhizobium in both field experiments and under gnotobiotic conditions (Halverson and Handelsman 1991; Turner and Backman 1991) . None of the above studies explain adequately the mechanism(s) of nodulation promotion by Bacillus spp. There is little evidence implicating a plant response that could be singularly attributed to the interaction between Bacillus and Rhizobium resulting in enhanced nodulation. Nodule formation is primarily controlled by the plant (Caetano-Anolles and Gresshoff 1991). Hence, any changes to plant root growth or morphogenesis in response to the presence of both rhizosphere bacteria may influence nodulation of the plant by Rhizobium. While substantial progress has been made in elucidating the mechanism(s) of the \egume-Rhizobium interaction, information is scarce on rhizobacterial interactions and their potential to influence the \egume-Rhizobium symbiosis. 1.5. Hypothesis and objectives of this study Bacillus spp. are attractive as plant-growth promoting and nodulation promoting agents because they occur naturally in the rhizosphere, because they may be 9 endophytic in plant root tissues, and because the cells can be produced easily in large quantities and stored for long periods of time without loss of viability (Hwang et al. 1996). The general objective of this study was to evaluate the influence of Bacillus spp. in promoting nodulation of the legume, Phaseolus vulgaris (L.) by its homologous microsymbiont, Rhizobium etli. The study is predicated on the hypothesis that rhizosphere interactions between microorganisms may have significant potential to influence plant growth and plant-microbe interactions such as legume nodulation by Rhizobium. Specific objectives include: (i) to determine whether the ability of Bacillus spp. to promote nodulation of Phaseolus vulgaris by Rhizobium etli is dependent on phytohormone production (especially indoleacetic acid) in vitro, (ii) to investigate the effect of the simultaneous presence of Bacillus spp. and Rhizobium on lateral root and root hair development in P. vulgaris, (iii) to determine if viable, interacting Bacillus and Rhizobium are a prerequisite for invoking a plant response in terms of root hair morphogenesis and lateral root development, (iv) to determine the effect of Bacillus spp. on nodulation pattern and nodulation specificity of Rhizobium etli and, (v) to investigate whether the interaction between Rhizobium etli and Bacillus spp. is independent of ethylene-related changes within P. vulgaris with regard to root hair morphogenesis. 10 Chapter 2 Influence of IAA producing Bacillus isolates on the nodulation of Phaseolus vulgaris by Rhizobium etli Every method is imperfect Charles-Jean-Henri-Nicolle (1932) 11 2.1. Introduction The direct effects of PGPR have been most commonly attributed to the production of phytohormones, notably auxins (Brown 1972; Tien et al. 1979). The role of phytohormones in organogenesis in plants has been well established. Numerous rhizosphere-inhabiting, diazotrophic bacteria belonging to the genera Azospirillum, Azotdbacter, Pseudomonas and Bacillus produce copious amounts of phytohormones (Brown 1972; Tien et al. 1979; Holl et al. 1988). The capability of such bacteria to produce physiologically active amounts of growth hormones, particularly IAA, has been correlated to their growth-promoting ability (Barbieri et al. 1986; Loper and Schroth 1986). Rhizobacteria of the genera Azospirillum, Bacillus, Pseudomonas, and Streptomyces also promote nodulation in many legume species (Iruthayathas et al. 1983; Grimes and Mount 1984; Li and Alexander 1990). The ability of Azospirillum to produce phytohormones has been hypothesized to play a major role in promoting nodulation in legumes (Schmidt ef al. 1988; Yahalom et al. 1990). Auxins may also be important for maintaining functional root nodules, since effective nodules have a higher IAA content than ineffective nodules (Badenoch-Jones et al. 1983). Changes in the phytohormone balance in roots may also elicit nodule formation. External application of S the auxin transport inhibitor triiodobenzoic acid (TIBA) induced the formation of pseudo-nodules in a non-nodulating line of alfalfa, indicating phytohormone involvement independent of Nod factor activity (Hirsch et al. 1989). Schmidt et al. (1988) demonstrated that application of IAA mimicked the effect of Azospirillum on nodulation in Medicago sativa. Many workers have demonstrated the accumulation of substantial 12 amounts of auxin in nodules compared to the surrounding root tissues in both legumes and non-legumes (Dullart 1970; Dangar and Basu 1987). Many Rhizobium spp. are known to produce IAA in vitro (Badenoch-Jones et al. 1983; Ernstsen et al. 1987; Atzorn et al. 1988). While some studies clearly implicate bacterially derived IAA in nodule formation (Kaneshiro and Kwolek 1985; Hunter 1987), there is conflicting evidence to suggest that such involvement may not be essential for nodulation (Wang et al. 1982; Atzorn et al. 1988). Despite studies which suggest that IAA produced by bacteria can influence the biological response between plant and bacterial symbionts, the direct effect of phytohormone producing rhizobacteria on legume nodulation remains unproven. The objective of this study was to determine whether the nodulation promoting ability of Bacillus spp. is dependent on their ability to produce IAA in vitro. 2. 2. Materials and Methods 2. 2.1. Isolation and characterization of strains Bacillus spp. isolates were obtained from the rhizosphere of Phaseolus vulgaris at seven and fourteen days after planting from a field in which beans had been cultivated for the previous ten years. Isolation of bacteria from the root surface and from the inside of roots was done according to the procedure described by O'Neill et al. (1992). Isolates of the genus Bacillus were obtained by enrichment under anaerobic conditions as described by Holl et al. (1988) using serial dilution of root-soil extracts. A total of 120 bacterial colonies were isolated and characterized in this study. Carbon source utilization was assayed using Biolog™ multiwell analysis. Classification was accomplished using the Biolog™ database supplemented by the inclusion of numerous 13 other Bacilli previously isolated (Massicotte and Chanway, 1994). Species identification was further refined using fatty-acid analysis (GC-FAME) at Auburn University according to Kloepper et al. (1992). Both Biolog™ and GC-FAME analysis were in agreement at the genus level. Where differences in species designation occurred, the GC-FAME classification was used. The comparative database for the GC-FAME analysis includes a substantial range of known rhizosphere bacteria. Phylogenetic analyses were conducted using 23S rDNA described by Petersen et al. (1995). Rhizobium etli TAL 182 was obtained from NifTAL, Paia, Hawaii. 2. 2 .2. Bacterial growth media Bacillus spp. were routinely maintained on combined carbon medium agar slants (Rennie 1981). Rhizobium etli TAL 182 was maintained on yeast mannitol agar slants (Bohlool and Schmidt, 1970). Broth cultures were grown to early stationary phase on combined carbon medium for Bacillus spp. and in yeast mannitol broth for Rhizobium etli TAL 182. The cells were harvested by centrifugation, washed twice with sterile phosphate buffer (0.1M, pH 6.8-7.0) and resuspended in the same buffer to a desired cell density before inoculation. 2. 2. 3. Isolation of indoleacetic acid (IAA) Screening of isolates for IAA production IAA production by Bacillus spp. isolates was assessed by the rapid in situ assay described by Brie et al. (1991), using 2% 0.5M FeCI3 in 35% perchloric acid as the reagent for color development. Bacterial colonies that formed distinct red halos were further analyzed for IAA production using thin layer chromatography (TLC), high performance liquid chromatography (HPLC), the avena coleoptile bioassay and gas 14 chromatography-mass spectrometry (GC-MS). Extraction of indoleacetic acid Indoleacetic acid (IAA) in Bacillus cultures grown in the medium described by Holl et al. (1988) was extracted using ethyl acetate described by Tien et al. (1979). Rhizobium etli TAL 182 was grown in a medium described by Ernstsen et al. (1987). Cultures were incubated on a rotary shaker in the dark at 30° C until early stationary phase in 500 ml flasks containing 200 ml of medium with or without supplemental L-tryptophan (100 ug/L). The cultures were harvested at early stationary phase (72 h) and centrifuged at 2000 x g for 30 minutes. The supernatant was supplemented with dithiothreitol (0.05g) to prevent oxidation of indole compounds and filtered (0.45 um Millipore) to remove cell debris. The supernatant volume was reduced 75% by rotary evaporation at 37° C, adjusted to pH 2.7 with 1N HCL, and partitioned three times with an equal volume of ethyl acetate (99.9%; HPLC grade) as described previously (Tien et al. 1979). The ethyl acetate fractions were pooled and reduced to dryness by rotary evaporation at 37° C. The dry residue was dissolved in one ml of absolute methanol. The methanol extracts thus obtained were stored in the dark at -20°C until further use. Extraction of IAA was performed three times for each isolate using identical experimental conditions. Thin layer chromatography . Detection of IAA using thin layer chromatography was accomplished by applying 10 ul of the methanol extracts to a 0.25 mm thick silica gel plate (Merck, Germany). The plates were developed with a freshly prepared solvent mixture of chloroform : methanol: water (84 :14:1). Detection of indole compounds on thin layer plates was 15 done according to Ehmann (1977). IAA was identified at Rf 0.53 as verified by an authentic IAA standard (Sigma Chemical Co, St. Louis, MO). Bioassay The avena coleoptile straight growth assay (Nitsch and Nitsch 1956) was used for the detection of biological activity resulting from auxin-like substances in the methanol extract. Aliquots (250 ul) of the methanol extracts were reduced to dryness under vacuum and the residues were dissolved in 0.1 M phosphate buffer, pH 6.4 containing 1% sucrose (w/v), and the bioassay was conducted as described by Holl et al. (1988). Specific fractions representing 10 Revalues were eluted from untreated portions of the thin layer chromatograms and tested for auxin-like activity by the same assay. High performance liquid chromatography Methanol extracts (20 ul) were injected onto a 250mm x 4.6mm (ID.), C:18 Nucleosil column (Metachem) using a Gilson 714 liquid chromatograph equipped with a differential UV detector absorbing at 280nm. The indole compounds were eluted using an isocratic gradient of 2% acetic acid in acetonitrile as solvent B and 2% acetic acid in water (HPLC grade) as solvent A at a flow rate of 1ml/min. Each extract was analyzed at least three times by HPLC and the retention times for peaks were compared to those of an authentic standard for IAA. Indoleacetic acid peaks were also confirmed by co-chromatography with an authentic standard. Quantitation of IAA in the samples was done by comparison of peak areas. The IAA band corresponding to R, 0.53 was removed from the TLC plate and eluted for three hours at 4° C in the dark in 2 ml of absolute methanol; the eluate was filtered and evaporated under vacuum and the 16 residue redissolved in 0.5 ml of absolute methanol. Aliquots (30 ul) of this extract was analyzed by HPLC as described above to confirm the presence of IAA. Gas chromatography-mass spectrometry The methanol extract from the TLC plate was analyzed by GC -MS using 2 ul of trimethylsilyl derivatized samples. Analysis was performed with a Fisons Trio 2000 mass spectrometer, using an ID DB-5 column at 50° C, programmed at 8°/min. GC - MS analysis of an authentic IAA standard was done in the same manner. 2. 2. 4. Isolation of spontaneous a-methyl tryptophan resistant mutants Isolation and characterization of a-methyl tryptophan resistant mutants was done as described by Smidt and Kosuge (1978). B. megaterium S49 at 10s cfu ml"1 was grown on Spizizien minimal-glucose agar (Spizizien 1958) with either a centrally located 6mm filter paper disk impregnated with 100 ul of 5mM filter-sterilized a-methyl tryptophan or a 6mm well in the agar filled with 100 ul of 5 mM a-methyl tryptophan. Single a-methyl tryptophan resistant colonies growing within zones of inhibition on these plates were streaked on to the same medium and resultant single colonies were streaked out three times for purification. The resistant mutants thus obtained were maintained in the same medium containing 2.5 mM a-methyl tryptophan. IAA production by resistant mutants was determined by TLC as described above, and by the colorimetric method of Gordon and Weber (1951), using 20 ml cultures grown in Spizizien minimal-glucose medium containing 2.5mM a-methyl tryptophan. For plant inoculation experiments the resistant mutant isolates were grown as described above to the required cell density. 17 Assay for tryptophan excretion Strains were assayed for excretion of tryptophan as described by Hoch et al. (1971), using 0.05% acid hydrolyzed casein and/or 50 ug per ml of phenylalanine in the medium where appropriate. Samples were removed periodically (until early stationary phase) and centrifuged at 2000 x g for 10 minutes. The supernatants were examined fluorometrically using a Shimazdu RF 540 spectrofluorometer at an excitation wavelength of 280 nm and an emission wavelength of 350 nm and the amount of tryptophan excreted was determined from a standard curve obtained using L-tryptophan (Sigma Chemical Co, St. Louis, MO). 2. 2. 5. Plant material and growth Seeds of Phaseolus vulgaris L. 'Contender' were cultivated in growth chambers using modified Leonard jar assemblies (Blauenfeldt er al. 1994), containing 1:1 (v/v) of sterile planting medium of industrial sand and Turface™. Seeds were surface sterilized by soaking in 95% ethanol for 30 seconds and then in 20% Chlorox (1.5% sodium hypochlorite) for 15 minutes and rinsed seven times with sterile distilled water. Surface sterilized seeds were germinated aseptically on moist filter paper in sterile petri plates. Two day old pregerminated seeds with uniform radicle length were inoculated with early stationary phase bacterial suspensions of Rhizobium etli TAL 182, Bacillus spp. and di-methyl tryptophan resistant mutants of B. megaterium S49 at a cell density of approximately 2 x 108 cfu/ml. Plants were grown under a controlled environment with a 16 h photoperiod and 22°C/18° C day night temperatures, and 350 uE m2 s"1 illumination and were harvested at 24 to 26 days after planting. The experiment was laid out as a completely randomized design with seven replicates for each treatment. 18 Acetylene reduction assay (ARA) The ARA system as described by Dilworth (1966) was used to measure nitrogenase activity. After one hour incubation, the gas mixture was analyzed by flame -ionization gas chromatography using a Hewlett Packard 5830A gas chromatograph (Holl et al. 1988). Total soluble protein in nodules was estimated according to Bradford (1976). The difference -spectral method (Appleby and Bergersen 1980) was used to measure the leghemoglobin content in the root nodules. Statistical analysis was performed using standard analysis of variance (ANOVA) techniques (SAS Institute Inc., Cary, NC). 2.3. Results Initial classification of strains isolated from the rhizosphere of P. vulgaris using Biolog™ and 23S rDNA analysis demonstrated the presence of several Bacillus spp. both in the rhizosphere (soil adhering to the root surface) and inside root tissues. Endophytic isolates were identified predominantly as Bacillus megaterium whereas those from outside of the roots included various Bacillus spp. - B. megaterium, B. brevis, B. pumilus, B. licheniformis and B. azotoformans. Direct sequencing of an approximately 850 bp band amplified from the 23S rDNA using PCR primers BAC11X and BAC19 (Petersen et al. 1995) yielded genotypic differences which were not distinguishable using metabolic fingerprinting or GC-FAME analyses. DNA sequence alignments were constructed and compared to other sequences included in the Ribosomal Database Project (Larsen et al. 1993) to yield the dendrogram shown in Figure 2.1. Bacillus spp. S89 and B17 were not included in the 23S rDNA analysis; sequence data for these isolates was unavailable at the time of original analysis, and 19 Figure 2.1. Phylogenetic dendogram showing the diversity of IAA producing Bacillus isolates. The dendogram from 23S rDNA nucleotide sequences was constructed by Phylogenetic Analysis Using Parsimony (PAUP 3.0). The numbers at the nodes are boot strap values for the nodes (based on 1000 bootstrap resamplings). Rhizobium etli TAL 182 was used to outgroup the tree. B. megaterium (S49) B. megaterium (A15) B. polymyxa (L6) B. polymyxa (L5) B. polymyxa (PW2) 8. brevis (S72) B. mycoides (T10) 6. megaterium (Led 02) 8. pumilus (Bet 10a) 5. pumilus (Bet 11a) 6. azotoformans (S08) R ef// (TAL 182) 21 several other examples of B. megaterium were already included in the study. Bacterial identification using GC-FAME was consistent with the phylogenetic analysis. The origin of different Bacillus isolates used in this study is shown in Table 2.1. The thirteen isolates in this table include nine obtained in the current study and an additional four isolates from other plant and ecosystem environments. The amounts of indoleacetic acid (IAA) produced by different Bacillus isolates in the presence and absence of supplemental L-tryptophan (100 ug/L) is shown in Table 2.2. Of the 22 Bacillus isolates obtained from the P. vulgaris rhizosphere, forty-five percent were IAA producers, with significant differences noted in the in vitro production by different bacterial strains. Addition of L-tryptophan to the medium resulted in production of significantly higher amounts of IAA in seven of the nine Bacillus isolates tested. Two strains (S72 and T10) which produced low amounts of IAA in unsupplemented culture medium showed no appreciable response in IAA production following addition of L-tryptophan. In contrast to the Bacillus spp., no IAA could be detected in extracts of TAL 182 (Rhizobium etli) in the unsupplemented culture medium, while the addition of exogenous tryptophan (100 ug/L) resulted in low detectable amounts of IAA (0.35 ug/ml). The detection limit of IAA using the HPLC method described was approximately 0.1-0.2 ug/ml. Quantification of IAA was carried out over several months and some variation in detection sensitivity might be anticipated due to HPLC column wear and ageing UV detector. Thin layer chromatography of the methanol extracts revealed the presence of several indole compounds. IAA was detected at an R, value of 0.53 which corresponded to the R, of an authentic IAA standard. 22 Table 2.1. Identification and origin of Bacillus isolates used in this study Isolate GC-FAME ID Isolate origin Source A15 B. megaterium P. vulgaris root interior This study S49 B. megaterium P. vulgaris root interior " Led 02 B. megaterium P. vulgaris root interior " S89 B. megaterium P. vulgaris root interior B17 B. megaterium P. vulgaris rhizosphere S72 B. brevis P. vulgaris rhizosphere " Bet 10a B. pumilus P. vulgaris rhizosphere Bet 11a B. pumilus P. vulgaris rhizosphere " S08 - P. vulgaris rhizosphere L5 B. polymyxa Mixture of L. perenne and Chanway et al. (1991) T. repens root and soil L6 B. polymyxa Mixture of L. perenne and Holl et ai. (1989) T. repens root and soil PW2 B. polymyxa Lodgepole pine root Shishido et al. (1995) interior T10 B. mycoides Pseudotsuga menziesii Massicotte and mycorrhizosphere Chanway (1994) - = not determined. This isolate was identified as B. azotoformans by the Biolog™ carbon utilization profile. 23 Table 2.2. Indoleacetic acid (IAA) production by Bacillus isolates and Rhizobium etli IAA production (ug/ml)** Isolate (+trp)t (SD) (-trp) (SD) JAU82(Rhizobium etlif 0.35 (0.05) nd nd S49(S. megaterium) 4.80 (0.33) 1.08 (.005) * A15(B. megaterium) 4.48 (0.86) 0.89 (.001) * B17(B. megaterium) 4.88 (0.28) 1.05 (.010) * S89(B. megaterium) 1.80 (0.25) 0.40 (0.07) * Led 02(B. megaterium) 2.70 (0.30) 0.70 (0.09) * S72(B.brevis) 0.40 (0.04) 0.30 (.010) Bet10a(B. pumilus) 3.86 (0.25) 0.45 (.010) * L6(B. polymyxa) 3.04 (0.37) 0.84 (.001) * L5(B. polymyxa) 2.59 (0.61) 0.41 (.008) * PW2(B. polymyxa) 2.39 (0.15) 1.03 (.016) * T10(B. mycoides) 0.58 (0.09) 0.54 (.010) ** Mean of six replicates in two separate experiments using HPLC analysis as outlined in materials and methods, nd = not detected. * Grown in a medium with supplemental L-tryptophan at 100 ug/L. * Grown in medium as described by Ernstsen et al. (1987). * IAA production was significantly higher (p < 0.05) in the presence of supplemental L-tryptophan as determined by t-test. 24 The presence of IAA in the methanol extracts of isolates S49 and L6 was further supported by the avena coleoptile straight growth assays, showing significant auxin-like activity compared to the control (Figure 2.2). TLC eluted fractions corresponding to Rf 0.4-0.6 (corresponding to the mobility of authentic IAA standard) of B. megaterium S49 and B. polymyxa L6 showed biologically responsive auxin-like activity, with a 155% increase in coleoptile growth. The authentic IAA standard on TLC eluted at R, 0.4-0.6 also gave a significant increase in coleoptile growth compared to the control (0.1M phosphate buffer with 1% sucrose). Methanol extracts and the TLC fractions corresponding to R, 0.4-0.6 were analysed by HPLC and GC-mass spectrometry. In the HPLC analysis, peak activity corresponded to the elution time for the IAA standard (9.9 minutes) (Figure 2.3 A, B). The TLC band at Rf 0.4-0.6 also gave a prominent HPLC peak corresponding to the IAA standard (Figure 2.3C). Co-chromatography of the unknown samples from TLC with a TLC-derived authentic IAA standard produced a single enhanced peak in HPLC at the same retention time (9.9 min) as the standard alone (Figure 2.3D). Further confirmation that the TLC eluate (R, 0.4-0.6) contained IAA was obtained from the UV absorption spectrum which corresponded to that of an authentic IAA (data not shown). In addition to S49, IAA in the culture filtrate extracts derived from other Bacillus isolates was also identified by HPLC and TLC as described above. The GC-mass spectral analysis of the TLC eluted band corresponding to Rf 0.4-0.6 for the B. polymyxa isolate L6, yielded two prominent ions (m/z 130 and m/z 189) in the spectrum of indole acetic acid and the mass spectrum of these two ions was very similar to that of indoleacetic acid (Figure 2.4). At the onset of this work, isolate L6 was 25 Figure 2.2. The avena coleoptile straight growth assay for auxin-like activity in Bacillus isolates extracts. Treatments were as follows: (1) Control, (2) IAA at 1ug / ml, (3) IAA at 5 ng / ml, (4) IAA at 10 ug / ml, (5) S49 cell free extract, (6) L6 cell free extract, (7) S49 Rf 0-0.2, (8) S49 Rr0.2-0.4, (9) S49 Rf 0.4-0.6, (10) S49 Rr0.6-0.8, (11) S49 R,0.8-1.0, (12) L6 Rf 0.4-0.6. Each value represents the mean of ten replicates. Means with the same letter are not significantly different at p < 0.05. 26 A B 50-40-30-20-10-. 0 20 mVFS 50 40 30 20 10 0 20 mVFS 10 12 10 12 50 40 30 20 10 0 2 20 mVFS D 75 60 45 30 15 0 2 20 mVFS 10 12 10 12 Retention time (min) Figure 2.3. HPLC separation of (A) authentic IAA standard at 5 ug / ml, (B) IAA fraction in the methanol extract of B. megaterium S49, (C) TLC eluted fraction of Rf 0.4-0.6 of B. megaterium S49 and, (D) co-chromatography of the TLC eluted band of B. megaterium S49 (Rr 0.4-0.6) with authentic IAA eluted from TLC at Rr 0.4-0.6. 27 Figure 2.4. Mass spectrometric span of (A) 100 picograms of an authentic indoleacetic acid standard and (B) thin layer chromatograph eluted band corresponding to R, 0.4-0.6 of B. polymyxa L6. 100 130 to 00 29 the most characterized growth promoting strain was selected for GC-MS analysis. The subsequent analogous biochemical analyses for IAA production by B. megaterium S49 were consistent with the earlier data derived from B. polymyxa L6. Cost constraints precluded repeating the GC-MS analysis on the B. megaterium S49 extracts. The a-methyl tryptophan-resistant mutants of B. megaterium S49 were obtained using 5mM a-methyl tryptophan on Spizizien glucose-minimal agar plates (Spizizien 1958) as described previously (Smidt and Kosuge 1978). The growth rates of the a -methyl tryptophan resistant mutants derived from B. megaterium S49 are given in Table 2.3. Manifestation of a-methyl tryptophan inhibition is observed through an extension of the lag phase in the growth curve, while logarithmic growth remains unaltered (Smidt and Kosuge 1978). A similar growth response was observed for Bacillus S49-1D and S49-1W in this study. It is also evident from Table 2.3 that wild-type B. megaterium S49 was more susceptible to a-methyl tryptophan inhibition than the resistant mutants (S49-1D and S49-1W). Addition of 0.05% hydrolyzed casein and/or 50 ug/ml phenylalanine to the growing medium increased the growth rates of the resistant mutants. The a -methyl tryptophan resistant mutants of Bacillus S49 excreted tryptophan, which could be measured spectrofluorometrically, into the growing medium. Isolate S49-1D excreted higher amounts of tryptophan (2.40 ug/ml) at early stationary phase than isolate S49 -1W (1.51 ug/ml) . The reverse was true for IAA production by these two resistant mutants (S49-1W > S49-1D) (Figure 2.5). Having established the lAA-producing ability of our Bacillus isolates, we subsequently investigated the influence of coinoculation of these isolates with Rhizobium etli TAL 182 on nodulation of P. vulgaris. Initially nine IAA producing Bacillus isolates 30 Table 2.3. Growth of Bacillus megaterium S49 in the presence of 2.5mM a-methyl tryptophan Isolate Extension of lag phase (h)* Minimal-glucose medium with 2.5mM a-MT a b c S49 18 18 S49-1D 12 09 06 S49-1W 18 12 12 * Difference between duration of lag phase in cultures simultaneously inoculated into Spizizien minimal-glucose medium and Spizizien minimal-glucose medium with 2.5mM a-methyl tryptophan. a = Growth in Spizizen minimal-glucose medium with 2.5mM a-methyl tryptophan. b = 2.5mM a-methyl tryptophan with 50 ug / ml of phenylalanine, c = 2.5mM a-methyl tryptophan with 0.05% acid hydrolyzed casein. - = not determined. 31 6 12 24 36 48 60 Hours of growth Figure 2.5. (A) Excretion of L-tryptophan and, (B) production of IAA by wild-type B. megaterium S49 and a-methyl tryptophan resistant mutants of S49 (S49-1D and S49-1W). Wild-type S49 was grown in Spizizien minimal medium with 100 ug /ml of L-tryptophan. The mutant isolates were grown in Spizizien minimal medium with 2.5mM a-methyl tryptophan , 0.05% acid hydrolyzed casein and 50 ug /ml of phenylalanine. Each value represents the mean of six individual samples in two separate experiments. Bars represent the standard error of the mean. 32 from the P. vulgaris rhizosphere and isolate L6 from a perennial pasture (Holl et al. 1988) were tested for their ability to promote nodulation in P. vulgaris in the greenhouse. Five Bacillus isolates promoted nodulation under these non-sterile conditions (Table 2.4-2.6). Three of these five isolates (S49, S72 and A15) also promoted nodulation under gnotobiotic conditions in growth chambers, suggesting a direct influence of these isolates on the nodulation of P. vulgaris by TAL 182 (Table 2.7-2.8). Some of the control plants that received one ml of sterile yeast mannitol medium as inoculum developed a few pseudonodules, devoid of any Rhizobium. The formation of spontaneous nodules or nodulation in the absence of Rhizobium has been reported (Caetano-Anolles et al. 1990; Blauenfeldt et al. 1994). The positive effects of coinoculation of such Bacillus isolates on nodulation in P. vulgaris 'Contender' were readily discernible. Nodule number (NON), nodule fresh weight (NFW), nitrogenase activity (ARA), leghemoglobin content, and total soluble proteins (TSP) were significantly increased by coinoculation compared to the TAL 182 single inoculation (Table 2.7). iAA-producing Bacillus isolated from the rhizosphere of other plant species were tested for their ability to promote nodulation of P. vulgaris by Rhizobium TAL 182. Amongst the Bacillus isolates tested, B. polymyxa L6 produced the greatest enhancement of nodule number while B. megaterium S49 produced the highest nodule fresh weights and nitrogenase activity, and B. brevis S72 resulted in the highest accumulation of total soluble proteins in the nodules (Table 2.7). Coinoculation of the a-methyl tryptophan resistant mutants of B. megaterium strain S49 (S49-1D and S49-1W) with Rhizobium TAL 182 inhibited nodulation in P. vulgaris. Significant reductions in nodule number (NON), nodule fresh weight (NFW), 33 Table 2.4. Influence of Bacillus isolates on nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) of P. vulgaris by Rhizobium etli (TAL 182) in the greenhouse1 Treatment NON (± SE) NFW ARA (n moles C 2 H 4 (g/plant ± SE) h"1 plant1 ± SE) Control- 6.5 ± 2.46 d 0.05 ± 0.02 c 0.11 ± 0.05 c Rh\zobium TAL 182 43 ± 3.21 c 0.19 ±0.02 6 4.92 ± 0.24 b TAL 182 + L6 66 ± 7.16 a 0.28 ± 0.04 a 6.54 ± 0.53 a (B. polymyxa) TAL 182 + S49 62 ± 6.54 a 0.26 ± 0.03 a 6.02 ± 0.36 a (S. megaten'um) TAL182 + B17 46 ± 4.95 b,c 0.17 ± 0.02 b 4.68 ± 0.27 b (B. megaterium) TAL 182 + Led 02 60 ± 5.51 a,b 0.24 ± 0.03 a,b 6.05 ± 0.35 a (B. megaten'um) T Surface-sterilized P. vulgaris seeds were grown in four-inch pots with 1:1 (v/v) of steam pasteurized soil and Turface® as growth medium. Plants were inoculated at three days after planting with one ml of 2 x 108 cfu / ml of bacterial suspensions. A booster inoculation was given five days after the first inoculation. Control plants received one ml of sterile one-tenth strength yeast mannitol broth as inoculum. Plants were harvested at 21 days after planting. The experiment was laid out as a randomized block design with seven replicates for each treatment. • Nodules formed on control plants were smaller than those observed on plants that received Rhizobium inoculum. Generally, control plants were chlorotic in appearance. Values represent the means of seven replicates. Means were separated using analysis of variance (ANOVA). Means followed by the same letter in a column are not significantly different at p < 0.05, as determined by Duncan's multiple range test. 34 Table 2.5. Influence of Bacillus isolates on nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) of P. vulgaris by Rhizobium etli (TAL 182) in the greenhouse1 Treatment NON (± SE) NFW ARA (n moles C 2 H 4 (g/plant ± SE) h_1 plant"1 ± SE) Control 5.9 ± 1.45 c 0.03 ± 0.01c 0.10 ± 0.05 c Rhizobium TAL 182 44 ±4.65 6 0.17 ± 0.02 c 4.66 ± 0.40 b TAL182+A15 68 ± 8.64 a 0.25 ± 0.03 a 6.68 ± 0.64 a (B. megaterium) TAL182 + T10 37 ± 3.40 b 0.14 ± 0.02 b 3.95 ± 0.16 b (B. mycoides) TAL 182 + Bet 10a 38 ± 3.80 b 0.14 ± 0.02 b 3.85 ± 0.23 b (B. pumilus) 1 Plant growth conditions, bacterial inoculation, experimental conditions, and statistical analysis are outlined in Table 2.4. Means were separated using analysis of variance (ANOVA). Means followed by the same letter in a column do not differ significantly at p < 0.05, as determined by Duncan's multiple range test. 35 Table 2.6. Influence of Bacillus isolates on nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) of P. vulgaris by Rhizobium etli (TAL 182) in the greenhouse1 Treatment NON (± SE) NFW ARA(n moles C 2 H 4 (g/plant ± SE) h"1 plant1 ± SE) Control 6.2 ± 2.39 c 0.03 ± 0.01 c 0.11± 0.04 d Rhizobium TAL 182 47 ± 5.56 b 0.16 ± 0.02 6 4.80 ± 0.35 b TAL182+S08 35 ±3.80 6 0.13 ± 0.01 b 3.85 ± 0.30 b,c (B. azotoformans) TAL 182 + S72 71 ± 7.32 a 0.36 ± 0.06 a 7.54 ± 0.43 a (S. brevis) TAL182+S89 37 ± 2.90 b 0.13 ± 0.01 b 3.70 ± 0.21 c (B. megaterium) r Plant growth conditions, bacterial inoculation, experimental conditions, and statistical analysis are outlined in Table 2.4. Means were separated using analysis of variance (ANOVA). Means followed by the same letter in a column are not significantly different at p < 0.05, as determined by Duncan's multiple range test. Table 4. Effect of Bacillus isolates on nodulation of Phaseolus vulgaris cv. 'Contender* inoculated with TAL 1821 Treatment strain # of nodules±SE NFW 2±SE (9) ARA 3±SE nmol C2H4/h/plant Leghemoglobin ug/g FW nodules TSP 4 ug/g FW nodules Control5 3.5 ±1.8d* 0.02 ±0.01d* - - -TAL 182 74.0 ±5.7° 0.6 ±0.1° 7.5 ±0.7C 137.4 ±11.1° 867.6 ±82.4C L6+TAL 182 138.4 ±6.0a 1.0 ±0.1" 14.3 ±0.6a 324.2 ±50.5ab 1885.7 ±50.0bd L5+TAL 182 128.6 ±6.8ab 0.9 ±0.1a 13.6 ±1.4a 302.2 ±46.3a 2190.5 ±97.0ab S49+TAL 182 132.4 ±4.6a b 1.1 ±0.1 a b 16.1 ±1.2ab 230.8 ±28.9a 1979.1 ±126.3bd S72+TAL 182 121.9 ±3 . r b 0.9 ±0.1a 13.3 ±0.4a 210.8 ±20.0a 2452.1 ±156.4a A15+TAL 182 101.1 ±6.2b 0.8 ±0.1" 13.2 ±1.2" 141.0 ±16.8C 1781.6 ±49.5" 1 Plants were grown under gnotobiotic conditions in Leonard jars containing a sterilized industrial sand-Turface™ mixture as described in the text. Values represent means of seven replicates of each treatment taken at 24 DAP. * Pseudonodules Means were separated using analysis of variance and means followed by the same letter in a column do not differ significantly at p < 0.05 as determined by Duncan's multiple range test. 2 NFW=Nodule fresh weight. 3 ARA=Acetylene reduction assay expressed as n moles of ethylene per hour per plant. 4 TSP=Total soluble proteins 5 Sterile one tenth strength yeast extract mannitol broth used as an inoculant. 37 Table 2.8. Influence of Bacillus isolates on nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) of P. vulgaris by Rhizobium etli (TAL 182) under gnotobiotic conditions in the growth chamber1 Treatment NON (± SE) NFW ARA (n moles C 2 H 4 h"1 (g/plant ± SE) plant"1 ± SE) Control 1.30 ± 0.56 c 0.004 ± 0.002c 0.00 ± 0.00 c Rhizobium TAL 182 89.80 ± 5.97a 0.470 ± 0.029a 9.80 ± 0.82 a TAL182+Led 02 78.10 ± 4.85 b,a 0.373 ± 0.035 b 7.23 ± 0.46 b (B. megaterium) TAL182+T10 67.00 ± 5.52 b 0.356 ± 0.030 b 6.48 ± 0.48c (B. mycoides) TAL182+B17 78.14 ± 4.62 b,a 0.392 ± 0.026 b 7.46 ± 0.53 b (B. megaterium) f Plant growth conditions, bacterial inoculation, experimental conditions, and statistical analysis are described in Chapter two, section 2.2.5. Means followed by the same letter in a column are not significantly different at p < 0.05, as determined by Duncan's multiple range test. 38 con 182 S49/182 S49 1D/182 S49 1W/182 Figure 2.6. Effect of coinoculation of a-methyl tryptophan resistant mutants of Bacillus megaterium S49 with Rhizobium etli TAL 182 in nodule number (NON), nodule fresh weight (NFW), and nitrogenase activity (ARA) in P. vulgaris 'Contender*. Each value represents the mean of six replicates. Means were separated by Duncan's multiple range test using analysis of variance. Means followed by the same letter are not significantly different at p < 0.05. This experiment was repeated once with similar results. 39 and nitrogenase activity (ARA) also resulted from coinoculation (Fig 2.6). Root growth was also inhibited as a result of coinoculation of TAL 182 with the a-methyl tryptophan resistant mutants of Bacillus S49 (Table 2.9). 2. 4. Discussion Of 120 isolates characterized from the rhizosphere of P. vulgaris, twenty two belonged to the genus Bacillus as identified by Biolog™ and GC-FAME analysis. The phylogenetic relationships of this multiplicity of Bacillus spp. associated with plant roots were demonstrated by 23S rDNA analysis. This diversity may reflect the dependence of rhizosphere microorganisms upon specific root exudates for their carbon sources. Mavingui et al. (1992) reported that B. polymyxa strains isolated from the rhizoplane of wheat (Triticum aestivum L.) were capable of metabolizing sorbitol whereas rhizosphere and non-rhizosphere isolates did not show such capability. Such diversity may also represent a closer specific association between the plant and Bacillus spp. which provides some benefit to the plant by producing phytohormones (Holl et al. 1988), suppressing deleterious rhizobacteria and/or by an unknown mechanism(s) (Chanway et al. 1991; Halverson and Handelsman 1991; Turner and Backman 1991). Forty five percent of the 22 Bacillus isolates produced IAA as determined by the rapid in situ assay, although quantities of IAA produced by individual strains varied markedly. This variation was not phylogenetically-determined since 23S rDNA sequence analysis did not cluster the IAA producers. However, the complexity of the rhizosphere environment and the intricacies of plant host/bacterial interactions likely alter the metabolism of both host and symbiont. Hence, the quantity of phytohormone produced at the root surface may be expected to vary from those measured in this study; such 40 Table 2.9. Root fresh weight and root dry weight of P. vulgaris due to coinoculation of Rhizobium etli TAL 182 with Bacillus megaterium S49 or a-methyl tryptophan resistant mutants of S49, under gnotobiotic conditions in the growth chamber1 Treatment Root fresh weight Root dry weight" (g/plant ± SE) (g/plant ± SE) Control* 3.16 ± 0.12c 0.32 ± 0.04c Rhizobium TAL 182 3.91 ± 0.116 0.41 ± 0.056 TAL 182 + S49 5.27 ± 0.19a 0.60 ± 0.07a TAL 182 + S49-1D 1.34 ± 0.12d 0.15 ± 0.04d TAL 182 + S49-1W 1.13±0.09d 0.11 ± 0.03d * Plant growth conditions, bacterial inoculation, experimental conditions, and statistical analysis are described in Chapter two, section 2.2.5. Means were separated using analysis of variance (ANOVA). Means followed by the same letter in a column are not significantly different at p < 0.05, as determined by Duncan's multiple range test. * Control plants received one ml of sterile phosphate buffer (pH 6.8) as inoculum. ** Roots were dried at 70°C for 72 hours, to a constant dry weight. 41 differences may account for the different plant responses observed. Previous investigations of the beneficial effects of coinoculation of Rhizobium spp. and Azospirillum spp. on nodulation attributed these effects to bacterial-derived phytohormones (Yahalom et al. 1990). Similarly, Bacillus isolates observed in this study that enhanced nodulation in P. vulgaris also produced significant amounts of IAA, in vitro. Further analysis of the morphogenic potential of indole compounds produced from Bacillus cultures showed a variable relationship between in vitro IAA production, plant growth stimulation and/or nodulation enhancement. For example, isolates such as B. brevis S72 produced only small amounts of IAA, but markedly enhanced nodulation over Rhizobium inoculation alone. Such observations suggest that the enhanced nodulation observed required the presence of both bacterial species and may not be solely dependent on in vitro IAA production by the Bacillus spp. This response supports the view that combinations of bacteria are more effective in promoting nodulation (Grimes and Mount 1984; Caetano-Anolles and Bauer 1988). Support for the view that Bacillus isolates influenced nodulation on P. vulgaris by R.etli TAL 182 was provided by the observation that the a-methyl tryptophan mutants S49-1D and S49-1W inhibited nodulation by Rhizobium TAL 182, and that host plants showed poor root growth. These changes in root growth and nodulation may have occurred as a consequence of changes in the phytohormonal balance of the rhizosphere since IAA production by the a-methyl tryptophan mutants was appreciably different from that of the wild-type. Furthermore, other metabolic changes were likely associated with a-methyl tryptophan resistance; different carbon source utilization patterns (Biolog™), the presence of a prolonged lag phase, and excretion of tryptophan reflected such 42 changes. These differences may also have contributed to altered symbiotic efficiency of Rhizobium (TAL 182). Nevertheless, the ability of the a-methyl tryptophan mutants to affect nodulation of P. vulgaris adversely under gnotobiotic conditions leads to the speculation that Bacillus spp. may have a direct effect on either the Rhizobium or the plant. Similar results have been reported by Hunter (1987) with 5-methyl tryptophan resistant mutants of Bradyrhizobium japonicum on soybeans. In addition to enhancement of nodulation, coinoculation of Bacillus isolates with Rhizobium TAL 182 increased nitrogenase activity and total soluble nodule protein content of the nodules. Higher nitrogenase activity may be attributed to the increase in leghemoglobin content in the root nodules. Interactions between plants and microorganisms are often associated with the induction in the plant of a complex set of biochemical responses and morphogenetic changes (Young et al. 1991; Kamoun et al. 1993). It is highly likely that genes encoding nodule proteins can be regulated by Rhizobium (Nap and Bisseling 1990). The increased leghemoglobin and total soluble protein content observed in this study may be due to unique physiological conditions imposed during nodule development by the presence of both bacterial species on the plant roots. The ability of Rhizobium to alter the regulation of an already functional gene(s) and impose physiological changes during nodule organogenesis has been reported previously (Bennett et al. 1989; Nap and Bisseling 1990). This study implicates naturally occurring phytohormone- producing Bacillus spp. in promoting nodulation in P. vulgaris by Rhizobium etli TAL 182. Nodule organogenesis clearly involves phytohormones (Hirsch and Fang 1994), but it is not yet clear whether elevated phytohormonal levels in nodules are the result of a plant 43 response to bacterial infection or a direct function of bacterial activity (Dullart 1970). Realizing the importance of cytokinins in nodule organogenesis, the ability of the Bacillus isolates to produce cytokinins in vitro was evaluated. Some of the Bacillus isolates tested produced trans zeatin, isopentyl adenine (2iP) and trans zeatin riboside. Further detailed analysis of cytokinin production by Bacillus spp. was not carried out in this study for the following reasons: (i) inconsistent results obtained in the present study and, (ii) technical difficulties encountered in identifying and quantifying cytokinins. Preliminary results on the detection of cytokinins in the cell cultures of Bacillus isolates are given for information in Appendix I. A possible explanation for increased nodulation in response to coinoculation in this study is the creation of additional infection sites for Rhizobium. In other studies, increased nodulation observed due to coinoculation of Azospirillum with Rhizobium has been attributed to the creation of more infection sites on the roots for Rhizobium by Azospirillum (Iruthayathas et al. 1983; Plazinski and Rolfe 1985). Further investigation of the physiological basis of the plant response to Bacillus inoculation should enhance the basic understanding of the plant-microbe interaction. The ability of some Bacillus species in this study to promote nodulation under gnotobiotic conditions irrespective of in vitro IAA production suggests a broader role for these microorganisms in directing the development of plant tissues in natural ecosystems. The primary objective of the work described in the next chapter was the evaluation of the response of root development in Phaseolus vulgaris to the presence of Bacillus and Rhizobium. 44 Chapter 3 Synergistic interaction between Rhizobium etli and Bacillus megaterium enhances root hair formation in Phaseolus spp. and in Arabidopsis thaliana I have the result, but I do not yet know how to get it Karl Friedrich Gauss (1777-1855) 45 3.1. Introduction In the previous chapter, phytohormone-producing Bacillus spp. were shown to be capable of nodulation enhancement of P. vulgaris inoculated with Rhizobium etli TAL 182. Those observations are consistent with the involvement of phytohormones in the promotion of plant growth and the influence of bacteria on root growth (Tien et al. 1979). Both lateral root development (Barbieri et al. 1986) and root hair production (Barbieri and Galli 1993) have been shown to be enhanced by bacterially-derived indoleacetic acid (IAA) (Yahalom et al. 1991). Changes in root length and density of root hairs due to microbial activity have also been reported (Morgenstem and Okon 1987; De Freitas etal. 1993). Root development and root hair morphogenesis are significant factors in the development of a healthy plant. In particular, root hairs play an important role in water and nutrient uptake (Clarkson, 1985) and, in both legumes (Bauer 1981) and non-legumes (Newcomb and Wood 1987), have been implicated in the symbiotic infection process. Abnormal root and root hair morphogenesis have been described in various studies of Arabidopsis auxin-resistant mutants (Mirza et al. 1984; Estelle and Somerville 1987). None of the reported root hair mutations affected the spacing or distribution of root hairs, however, or the performance of the plants under optimal water and nutrient conditions (Schiefelbein and Somerville 1990). Deformation of legume root hairs in response to Rhizobium spp. inoculation has been correlated with the production of IAA by these microbes (Nutman 1956; Fahraeus and Ljunggren 1968). Phytohormone effects have also been implicated in the increased nodule numbers observed on legume hosts in response to combined inoculation with Rhizobium and Azospirillum (Sarig et al. 1986; Yahalom et al. 1990). 46 While phytohormones clearly participate in such developmental responses, other work suggests that IAA is not the sole, or primary, factor responsible (Wang et al. 1982; Plazinski and Rolfe 1985; Atzorn et al. 1988). Since auxins are most likely involved in both root development and nodulation, the main objective of this study was to determine the effect of IAA -producing and nodulation-promoting B. megaterium S49 and Rhizobium etli TAL 182, on lateral root and root hair development in the common bean (Phaseolus vulgaris), and on root hair development in P. acutifolius and Arabidopsis thaliana. 3. 2. Materials and methods 3. 2.1. Bacterial strains and growth media The isolation and characterization of Bacillus megaterium S49 and a-methyl tryptophan resistant mutants of S49 (S49-1D) have been described previously (Chapter 2; sections 2.2.1; 2.2.4.). B. megaterium S49 was routinely maintained on combined carbon medium (Rennie, 1981) and Rh\zobium TAL 182 was maintained on yeast mannitol agar slants (Bohlool and Schmidt, 1970). Growth conditions for a-methyl tryptophan mutants of S49 have been described previously (chapter 2; section 2.2.4.). Broth cultures were grown to early stationary phase in combined carbon medium for Bacillus spp. and in yeast mannitol broth for Rhizobium spp. The cells were harvested by centrifugation, washed twice with sterile 0.1M phosphate buffer (pH 6.8-7.0) and resuspended in the same buffer to the desired cell density before inoculation. 3. 2. 2. Plant material and growth Seeds of Phaseolus vulgaris cv. 'Contender1 were surface sterilized in 30% (v/v) hydrogen peroxide for 15 minutes. Sterile 2-day old pregerminated seedlings with 47 uniform radicle length (approximately 1 cm) were immersed for 15 minutes in 0.1 M 6 -1 phosphate buffer (pH 6.8-7.0) either alone (control) or containing 10 cfu ml of the appropriate bacterial inoculant. Seeds treated with IAA were incubated for 15 minutes in 5 or 30 ug m l 1 IAA in 0.1 M phosphate buffer (pH 6.8-7.0). Seedlings were cultured axenically by transfer to a 20 x 150 mm test tube containing 10 ml of one-tenth strength plant growth medium (Broughton and Dilworth, 1971), in which the seed was supported above the medium by a 200 ul plastic pipette tip. Medium for treatments which included IAA were supplemented with appropriate concentrations of IAA (5 or 30 ug ml1). Tubes o 2 - 1 were incubated at 25 C with a 16h photoperiod at 200 uE m s illumination. Seeds of P. acutifolius were obtained from NifTAL, Paia, Hawaii. The seeds were sterilized, pregerminated, treated and grown as described above. A random sample of lateral roots from approximately 10 plants/treatment was used for scanning electron microscopy. The base and tip (approximately 0.8 cm each) of the lateral roots were excised and discarded. The remaining central portion of the root was used for the scanning electron microscopy analysis. Seeds of Arabidopsis thaliana derived from the Columbia wild type were obtained from the Arabidopsis Stock Centre (Columbus, OH). Seeds were surface sterilized and germinated on an agarose nutrient medium as described (Schiefelbein and Somerville, 1990). Roots of two day old seedlings were treated with 50 ul of buffer solution or bacterial suspension (106cfu ml1) in a sterile plastic petri dish for 15 minutes. After inoculation, seedling roots were washed three times in sterile distilled water and transferred to petri plates containing the nutrient agar medium of Schiefelbein and Somerville (1990). Washing was necessary to prevent overgrowth of the culture medium by the bacteria. The plates were sealed with Parafilm® and incubated vertically 48 2 -1 under continuous fluorescent illumination (approximately 100 u E m s ) at room temperature. At five days, the terminal 0.5-1.0 cm of radicle was excised and prepared for scanning electron microscopy as described below. Approximately 20 plants were scored for each treatment in each of two experiments. 3. 2. 3. Scanning electron microscopy (SEM) The root segments were fixed overnight in 2.5% glutaraldehyde in 0.05 M sodium cacodylate buffer (pH 7.2). Samples were washed three times (15 minutes each) with 0.05 M sodium cacodylate buffer, and post-fixed in 1% osmium tetroxide in 0.1 M sodium cacodylate buffer (pH 7.2) for two hours. The samples were rinsed once with de-ionized water, dehydrated through a 30% to 100% series of ethanol rinses, and critical point dried from 100% ethanol in a Balzers CPD 020 system using liquid carbon dioxide as the exchange medium/ After mounting on double-sided sticky aluminum stubs, specimens were coated with gold-palladium (approximately 25nm) in a Nanotech Semprep II Sputter coater, and viewed with a Cambridge 250T scanning electron microscope. All samples were viewed and the typical observed response was scored (e.g. no proliferation, slight proliferation, massive proliferation), and recorded photographically. 3. 2. 4. Estimation of lateral root numbers For the estimation of lateral root numbers, seeds of P. vulgaris were sterilized, pregerminated, treated and grown as described above. Germinated seedlings were harvested at five days after seeding and lateral roots counted. Statistical comparisons were carried out by the SAS computer program (SAS Institute, Inc., Cary, NC) using Fisher's LSD. 49 3.-3. Results 3. 3. 1. Root hair production by P. vulgaris in response to various treatments The root hair production responses in P. vulgaris to the various treatments are shown in Figure 3.1 (A-H). Single inoculation with either Rhizobium TAL 182 (Figure 3.1.C) or Bacillus S49 (Figure 3.1E) produced an increase in root hair production compared to control (Figure 3.1 A). Dual inoculation of these two isolates resulted in greatly enhanced production of root hairs (Figure 3.1F). Application of IAA alone at 5 ug ml 1 (Figure 3.1B) did not show any evidence of increased numbers of root hairs. Application of IAA with Rhizobium TAL 182 resulted in the elongation of root hairs (Figure 3.1D), which may reflect its involvement in root hair morphogenesis. Further evidence for the de-linking of IAA effects and bacterial interaction is found in the response to TAL 182 + S49-1D coinoculation (Figure 3.1.G). Although the Bacillus mutant S49-1D produces significantly lower IAA levels in culture (Chapter 2; section 2.3), it nevertheless does increase root hair production compared to the control (Figure 3.1.A). The results described were obtained from five separate experiments conducted over a period of several months. The presence or absence of prolific root hair initiation in response to the treatments was consistent across experiments. Of the 30 lateral roots examined from 15 plants receiving the dual inoculation, 10% showed proliferation equivalent to single inoculation by S49 alone, while 90% showed massive proliferation. 3. 3. 2. Compatible \egume-Rhizobium combination is not required for root hair proliferation The P. acutifolius used in these experiments is a line of bean which is normally nodulated by Bradyrhizobium spp. and is resistant to nodulation by Rhizobium TAL 182 50 Figure 3.1. Root hair production by P. vulgaris in response to bacterial coinoculation. (A) 0.1M phosphate buffer (pH 7.0) (control) (B) IAA (5 ug ml"1) (C) Rhizobium etli (TAL 182) (D) IAA (5 ug ml"1) and Rhizobium etli (TAL 182) (E) Bacillus megaterium (S49) (F) Rhizobium (TAL 182) and Bacillus (S49) (G) Rhizobium (TAL 182) and Bacillus (S49-1D) (H) Rhizobium (TAL 182) and Bacillus (S49-1D) and IAA (5 ug ml"1) 52 (Somasegaran ef al. 1991). The absence of any root hair stimulation comparable to dual inoculation by inoculation with Rhizobium TAL 182 alone (Figure 3.2B) is consistent with the known host nodulation specificity and the previous observations with P. vulgaris. Single inoculation with Bacillus S49 enhanced root hair production (Figure 3.2C) compared to the control (Figure 3.2A), while dual inoculation with Rhizobium TAL 182 and Bacillus S49 produced a more extensive proliferation of root hairs on the lateral root surface (Figure 3.2D). 3. 3. 3. Arabidopsis shows increased root hair production due to coinoculation Since the root hair proliferation response did not require a compatible legume-Rhizobium combination, it was prudent to investigate whether a similar reaction could be observed following dual inoculation of an unrelated, non-legume genus. The root hair response of Arabidopsis to dual inoculation (Figure 3.3C) mimicked the extensive proliferation observed in the Phaseolus tests. Single inoculation of Arabidopsis yielded little evidence of changes in root hair morphogenesis (Figure 3.3B). The plant response was clearly not related to any natural association of Arabidopsis and the challenge bacteria. The results indicate that root hair proliferation required the participation of the bacterial partners, but was independent of the plant host. 3. 3. 4. Bacterial inoculation and root development To determine the impact of bacterial inoculation and associated phytohormone production on root development, lateral roots were counted on five day old seedlings of P. vulgaris treated with various combinations of bacteria and IAA (Table 3.1). Lateral root production was clearly stimulated by exogenous IAA (5 ug ml1) and by inoculation with Rhizobium TAL 182 and/or Bacillus S49, whereas no effect was observed for treatments which included higher IAA levels (30 ug ml1) or the a-methyl tryptophan 53 Figure 3. 2. Root hair production by P. acutifolius in response to bacterial coinoculation. (A) 0.1M phosphate buffer (pH 7.0) (control) (B) Rhizobium etli (TAL 182) (C) Bacillus megaterium (S49) (D) Rhizobium (TAL 182) and Bacillus (S49) 54 40OHM 20KV i 400HM 20KV 55 Figure 3. 3. Root hair production by Arabidopsis thaliana in response to bacterial coinoculation. (A) 0.1M phosphate buffer (pH 7.0) (control) (B) Rhizobium etli (TAL 182) (C) Rhizobium etli (TAL 182) and Bacillus megaterium (S49) Single inoculation of Arabidopsis thaliana with B. megaterium S49 gave similar results as treatment (B). 57 Table 3.1. Lateral root production on five day old seedlings of Phaseolus vulgaris treated with indoleacetic acid (IAA), or inoculated with Bacillus megaterium (S49 or S49-1D) and/or Rhizobium etli (TAL 182) Treatment (Number of lateral roots per sample*) Control 15.2 a IAA (5 ug ml'1) 42.0 d IAA (30 ug mr1) 24.8 a,b,c TAL 182 30.0 a,b,c S49 40.5 d S49-1D 23.6 a,b,c TAL 182 + S49 30.2 b,c,d TAL 182 + S49-1D 18.1 a,o TAL 182 + S49 + IAA (5 ug ml1) 32.4 c,d TAL 182 + S49-1D + IAA (5 ug ml"1) 38.5 c,d 1 Values shown are the means of ten replicates in two separate experiments. Means were separated using analysis of variance (ANOVA). Means followed by the same letter are not significantly different (p < 0.05; LSD = 11.9). mutant S49-1D. Notably, the absence of lateral root enhancement by the combination -1 of S49-1D was reversed by the addition of supplemental IAA (5 ug ml ). 3. 4. Discussion Bacillus spp. are widely distributed in temperate soils. Phytohormone production by Bacillus has been shown to play a role in plant growth promotion (Holl et al. 1988). Bacillus megaterium S49 and Rhizobium etli TAL 182 were used to study the interaction of a plant growth-promoting rhizobacterium and the legume microsymbiont on lateral root and root hair development. Bacillus S49 produces IAA in culture and responds to the addition of supplemental tryptophan to the culture medium, whereas Rhizobium TAL 182 produces barely detectable amounts of IAA under similar growth conditions (Chapter 2; section 2.3). An a-methyl tryptophan resistant mutant of Bacillus S49 (S49-1D) has also been isolated; the mutant produces lower levels of IAA in culture and its growth cycle exhibits an extended lag phase compared to the parent strain (Chapter 2; section 2.3). The importance of root hairs as infection sites for Rhizobium led to the investigation of the impact of dual inoculation on the production of root hairs in Phaseolus. Combined inoculations of Rhizobium and Bacillus gave a clear indication of a synergistic enhancement of root hair initiation. This enhancement was not mimicked by either supplemental IAA alone or by coinoculation with a low IAA producing Bacillus mutant. The inability of a low-IAA-producing Bacillus mutant to promote lateral root formation is consistent with the findings of Barbieri and Galli (1993), who reported similar effects on wheat root growth when inoculated with an analogous Azospirillum mutant, (SpM 7918). IAA supplementation of Rhizobium TAL 182 inoculated plants increased root hair elongation, but did not produce additional proliferation. Failure of 59 such treatments to produce a stimulatory response comparable to dual inoculation, as well as slightly increased root hair production using the Bacillus mutant S49-1D, indicates that the IAA effect on root hair elongation could be delineated from the bacterial interaction which produced prolific initiation of root hair development. In Azospirillum-plant interactions, changes in root morphology and root hair formation are attributed to the production of auxin by the bacteria. The effects of auxin are dependent on the concentration of Azospirillum on the plant roots (Harari et al. 1988). In clover, application of the synthetic auxin 2,4-D results in root hair elongation (Ridge 1995). The stimulatory,effect of dual inoculation was further demonstrated using a non-compatible legume host (P. acutifolius) for Rhizobium etli which also showed prolific root hair initiation in response to coinoculation with Rhizobium TAL 182 and Bacillus S49. These observations clearly indicate that the synergistic response is not linked to the host specificity of the Rhizobium isolate. Given the positive response using a non-compatible host, the effect of inoculation on an unrelated plant genus which had no prior exposure to either of the bacterial isolates was examined. Not surprisingly, single species inoculation of Arabidopsis thaliana yielded little evidence of changes in root hair morphogenesis. In contrast plants receiving dual inoculation exhibited obvious proliferation of root hairs, confirming the view that the observed response was likely a consequence of interaction between the bacterial partners, rather than a host- directed event. While considerable attention has been directed at the plant-microbe interaction, the observations in this study provide clear evidence of an bacterial "dialogue" which results in a significant change in plant root morphogenesis. Root hair proliferation in response to a single rhizosphere bacterial species has been reported previously (Umali-60 Garcia et al. 1980; Kapulnik 1985a,b; De Freitas et al. 1993; Pacovsky, 1990). The present work describes the potential for interspecific bacterial interactions in the plant rhizosphere, and is the first report of a unique response in root hair morphogenesis induced by the interaction of two bacterial species. This interaction and its effect on root hair development are likely to have significant implications for the infection potential and nodulation of legumes, and for water and nutrient relations of plant species in natural ecosystems. 61 Chapter 4 Synergistic interaction between Rhizobium and Bacillus alters root morphogenesis and enhances nodulation in Phaseolus vulgaris The in vitro axiom: "Under the most rigorously controlled conditions of pressure, temperature, volume, humidity and other variables, the organism will do as it damn well pleases". Brian Ellis (1990) 62 4.1. Introduction Root hairs are epidermal tubular outgrowths in the root system of higher plants (Ridge 1995; Schiefelbein and Somerville 1990). In legumes, they serve as attachment and infection sites for nitrogen-fixing Rhizobium (Bauer 1981, Wood and Newcomb 1989). Factors which adversely affect root hair formation in legumes also reduce nodulation by Rhizobium (Tu 1981). Many aspects of root hair development are under the genetic control of the plant (Cardus 1979; Mirza et al. 1984; Hochmuth et al. 1985), but non-specific factors that alter root hair growth include calcium ions (Weisenseel et al. 1975; Picton and Steer 1983) and water stress (Schnall and Quatrano 1992). Plant hormones such as auxins and ethylene (Cormak 1962; Mirza et al. 1984; Tretyn et al. 1991; Masucci and Schiefelbein 1994) also affect root hair growth. Many of the root hair morphogenetic changes induced by the non-specific and hormonal factors can be mimicked by nitrogen-fixing bacteria on compatible legume hosts (Callaham and Torrey 1981; Wood and Newcomb 1989). Plant growth promoting rhizobacteria (PGPR) (Kloepper et al. 1989) have also been reported to enhance nodulation of legumes by the Rhizobium!Bradyrhizobium microsymbiont (Grimes and Mount 1984; Halverson and Handelsman 1991). Indoleacetic acid (an auxin) has been associated with adventitious root initiation (Blakesley et al. 1991) and lateral root development (Wightman et al. 1980). Experimental evidence has implicated bacterially-mediated phytohormone production as a causal element in plant growth promotion (Tien et al. 1979; Holl et al. 1988) by altering the hormone balance within the affected plant (Glick 1995). Rhizosphere microbial production of indoleacetic acid (IAA) has also been implicated in lateral root 63 development (Barbieri et al. 1986; Dubeikovsky et al. 1993), root hair production (Barbieri and Galli 1993), root elongation (Kapulnik et al. 1985b), root hair branching and differentiation of root hairs (Patriquin et al. 1983; Jain and Patriquin 1984). However there is conflicting evidence regarding the primacy of bacterially derived IAA in causing these developmental changes (Atzorn et al. 1988; Harari et al. 1988). Various reports implicate IAA in root hair deformation and curling (Farhaeus and Ljunggren 1968) and ascribe to it a critical role in nodule development (Hirsch et al. 1989). Although the influence of IAA produced by root-associated bacteria in promoting nodulation in legumes has been documented (Schmidt et al. 1988; Yahalom et al. 1990), other studies have indicated that IAA may not be involved in root hair deformation (Yao and Vincent 1976) and may not be essential for induction of nodule formation by Rhizobium (Wang et al. 1982). In chapter 2, it was shown that coinoculation of Phaseolus vulgaris with Rhizobium etli TAL 182 and Bacillus megaterium S49 induced massive proliferation of root hairs. This enhancement of root hair initiation was independent of the plant genotype or addition of exogenous IAA. Thus, a precise role for bacterial-derived IAA in root hair morphogenesis and nodule formation remains undefined. The effect of dual inoculation using different Bacillus species, cell preparations and mutants of both Rhizobium and Bacillus on lateral root production and root hair proliferation in P. vulgaris has been examined. The results of this analysis clearly indicate a lateral root formation response to IAA, and that viable bacteria are likely required to induce root hair proliferation. Since root hairs are the infection sites for Rhizobium, the consequences of a unique interaction between Bacillus and Rhizobium 64 on nodulation of P. vulgaris by Rhizobium etli TAL 182 were also investigated. 4. 2. Materials and Methods 4. 2. 1. Bacterial strains and growth media The origin, isolation and characterization of Bacillus S49 and Rhizobium TAL 182 have been described (Chapter 2). Two additional strains were used; Bacillus megaterium (Led 02) isolated from the Phaseolus vulgaris rhizosphere which shows high in vitro production of IAA (2.70 ug ml'1) and B. brevis (S72) a low in vitro IAA producer (0.40ug ml'1), as determined by HPLC analysis (Chapter 2). Rhizobium etli (MLC 35) a Tn5 mutant of Rhizobium etli TAL 182 which shows defective motility and nodulation (George and Robert 1991) was obtained from NifTAL, Paia, Hawaii. Bacillus spp. were routinely maintained on combined carbon medium (CCM) agar slants (Rennie 1981). Rhizobium TAL 182 was maintained on yeast mannitol agar slants (Bohlool and Schmidt, 1970). MLC 35 was maintained on TY plates (Beringer 1974) with nalidixic acid at 20 ug ml'1 and kanamycin at 50 ug ml'1 as described by George and Robert (1991). Broth cultures were grown to early stationary phase in combined carbon medium for Bacillus spp. and in yeast mannitol broth (YMB), with antibiotics, when appropriate, for Rhizobium spp. The cells were harvested by centrifugation, washed twice with sterile 0.1M phosphate buffer (pH 6.8) and resuspended in the same buffer to the appropriate cell density before inoculation. 4. 2. 2. Preparation of bacterial extracts Rhizobium TAL 182 and Bacillus S49 were grown in YMB and CCM, respectively, to early stationary phase. Cell-free extracts (CFE) were prepared by harvesting 50 ml cultures by centrifugation at 10000 x g for 30 minutes in a Sorvall® 65 G13 rotor. The cell free supernatant was passed through a 0.22 urn syringe filter and stored at -20 °C prior to use. Whole cell extracts of both bacterial species were obtained by harvesting eariy stationary phase cultures as described above, and rupturing cells by passage through a French press (AMINCO, 5000 psi). The cell debris thus obtained is referred to as whole cell extracts (WCE). Autoclaved preparations (ACE) were obtained by autoclaving early stationary phase cultures for 15 minutes at 121°C and 15 psi. The absence of live bacteria in all the above preparations was tested by streaking a loopful of each preparation on to yeast mannitol agar plates for Rhizobium and nutrient agar (Difco) plates for Bacillus spp. Both the WCE and ACE samples were stored at -20 ° C prior to use. 4. 2. 3. Scanning electron microscopy (SEM) Seeds of P. vulgaris 'Contender' were surface sterilized in 30% (v/v) hydrogen peroxide and rinsed seven times with sterile distilled water. Two-day old pregerminated seeds with uniform radicle length (approximately 1cm) were immersed for 15 minutes in 0.1M phosphate buffer (pH 6.8) either in the absence of bacteria (control) or containing 106 cfu ml"1 of the appropriate inoculant. For assessing the impact of whole cell extracts and/or cell-free preparations on root hair formation, pregerminated seeds were immersed for 15 minutes in the respective preparations, after dilution with sterile phosphate buffer to the desired cell equivalent. Media for treatments which included WCE, ACE or CFE were supplemented with one ml of the respective preparations. Seedlings were cultured axenically by transfer to a 20 x 150 mm test tube containing 10 ml of 1/10 strength nitrogen-free plant growth medium (Broughton and Dilworth 1971), in which the seed was supported above the medium by a 200 pi plastic pipette tip. 66 Tubes were incubated at 25° C with a 16h photoperiod at 200 uE m2 s'1 illumination. Lateral roots in each treatment were counted prior to preparation for SEM. A random sample of lateral roots from at least 10 or more plants from each treatment was selected for SEM. In general, lateral roots of uniform length were chosen for SEM. The base and tip (approximately 0.8 cm each) of the lateral roots were excised and discarded. The remaining central portion of the root was fixed in 2.5% glutaraldehyde and prepared for SEM analysis as described in Chapter 3. An in vitro "cup assay" was developed to allow bacterial growth on the roots, while facilitating physical separation of the two bacterial species. A 30, 000 D MW cut off membrane was used to physically separate the two bacterial species, while allowing extracellular products to diffuse between the inoculants and the developing root system. This experiment helped to determine if the physical presence of both Rhizobium and Bacillus on the plant roots was necessary to induce root hair proliferation. Seedlings were grown as described above in 20 x 200 mm test tubes with 20 ml of 1/10 strength nitrogen-free plant growth medium. The seeds were supported above the medium by a 1000 ul sterile plastic pipette tip fused to a polypropylene cup containing a polysulfone filter of 30,000 D molecular weight cut off (Nalgene). Physical separation of bacterial strains was accomplished by adding 400 ul of either the Rhizobium or the Bacillus to the cup (106 cfu ml"1) (Figure 4.1). Root samples were prepared for SEM analysis as described in Chapter 3. 4. 2. 4. Assessment of lateral root numbers in test tubes To assess lateral root numbers, plant material and growth conditions were as described above. Seedlings were harvested at five days of growth in test tubes and the number of lateral roots counted. Statistical comparisons were made with the SAS 67 Figure. 4.1. Schematic diagram of the cup assay procedure used to facilitate physical separation of bacterial species on plant roots. (A) Initial set up at day one and (B) at five days of growth. 68 computer program (SAS Institute Inc., Cary, NC) using analysis of variance (ANOVA) and Duncan's multiple range test. 4. 2. 5. Assessment of lateral root numbers in Leonard jars Seeds of P. vulgaris 'Contender' were surface sterilized, pregerminated and inoculated with bacterial suspensions as described above. Seeds were planted in modified Leonard jar assemblies (Blauenfeldt et al. 1994) containing sterile industrial sand-Turface ® (1:1 v/v) of plant growth medium. Seedlings were grown under gnotobiotic conditions in the growth chamber with a 16h photoperiod at 350 uE m2 s 1 illumination. Seedlings were harvested at four and seven days after planting and the lateral roots counted. Statistical comparisons were made as described above. 4. 2. 6. Early nodulation of P. vulgaris under gnotobiotic conditions in the growth chamber This experiment was carried out to determine the effectiveness of nodules formed at early stages of plant growth in response to coinoculation of Rhizobium with Bacillus spp. Plant growth medium and experimental conditions in growth chambers were as described above. Plants were harvested at 8, 10, 12, and 14 days after inoculation and nodule numbers and fresh weights were determined. Acetylene reduction assays (Dilworth 1966), were determined by flame-ionization gas chromatography as described by Holl et al. (1988). 4. 2. 7. PCR determination of bacterial growth and physical presence PCR was used to determine the efficacy of the "cup assay" experiments which allowed bacterial growth on plant roots while maintaining physical separation from either Rhizobium etli TAL 182 or Bacillus spp. in the cup. DNA of cells in one microlitre of 69 culture (plant growth medium vs cup) was amplified by multiplex PCR using 23S rDNA primers (BAC 11X, POL and TAL 182 primers), as described by Petersen et al. (1995). Amplification products of Bacillus and Rhizobium differ greatly in size (540 vs 650 bp respectively), and were separated by agarose gel electrophoresis, examined by staining with ethidium bromide and visualized under UV light. 4. 3. Results 4. 3. 1. Different Bacillus isolates induce different root hair morphogenetic responses when coinoculated with Rhizobium etli TAL 182 P. vulgaris plants grew well in liquid medium in test tubes, forming uniform normal root systems comparable to those plants grown on solid medium or in growth pouches. B. megaterium S49 induced root hair proliferation on P.vulgaris when co-inoculated with R.etli TAL 182, (Figure 4.2B) confirming the earlier observations reported in Chapter 3. The root hair induction responses due to inoculation with bacterial species were not influenced by lateral root size (diameter). Root hair proliferation responses for each treatment were judged by visual inspection by two independent observers and the figures shown represent the typical response for each treatment which consisted of 20 randomly chosen lateral roots from 10 different plants. Coinoculation of Rhizobium TAL 182 with either Bacillus spp. Led 02 or S72 did not enhance root hair formation on P. vulgaris, compared to coinoculation with B. megaterium S49 (Figure 4.2 B-D). This result also suggests that root hair proliferation due to coinoculation may not be solely dependent upon the ability of the Bacillus partner to produce IAA in vitro. Coinoculation of the higher IAA producing B. megaterium Led 02 with TAL 182 resulted in longer root hairs compared to the control and dual inoculation with the low IAA producer, B. brevis S72. Figure. 4.2. Root hair production by P. vulgaris in response to dual inoculation of Rhizobium etli TAL 182 with different Bacillus spp. A. Control (sterile 0.1M phosphate buffer [pH 6.8]) B. Rhizobium TAL 182 and Bacillus S49 C. Rhizobium TAL 182 with Bacillus Led 02 D. Rhizobium TAL 182 with Bacillus S72 71 400HM 20KV 72 Hence root hair proliferation in response to coinoculation can be discriminated from the effects of IAA produced by the Bacillus spp. These results are consistent with the previously observed effects on root hair morphogenesis using a-methyl tryptophan mutants of Bacillus S49 (Chapter 3). 4. 3. 2. Physical presence of both Bacillus and Rhizobium is necessary to induce root hair proliferation Separation of bacterial species using a 30,000 D MW cut off membrane prevented root hair proliferation, regardless of which bacteria was applied to the root system, indicating that both bacterial partners need to be physically present on the roots (Figure 4.3 H). The presence of either Bacillus or Rhizobium alone on the roots did not induce root hair proliferation (Figure 4.3 D-G), compared to dual inoculation of Rhizobium TAL 182 and Bacillus S49 (Figure 4.2 B). 4. 3. 3. Root hair proliferation is not induced in the absence of live bacterial species Culture tests confirmed the absence of viable bacterial cells in WCE, ACE and CFE preparations. Inoculation of P. vulgaris with either filter-sterilized CFE or WCE of Bacillus S49 with live Rhizobium TAL 182 failed to elicit root hair proliferation (Figure 4.3 B-C) compared to dual inoculation of live bacterial species (Figure 4.2 B and 4.4 A). Inoculation of WCE of both Bacillus S49 and Rhizobium TAL 182 failed to show a root hair response on 85% of lateral roots examined (Figure 4.3 B) compared to dual inoculation with live bacterial species (Figure 4.2 B and 4.4 A) whereas 15% of lateral roots screened showed proliferation comparable to dual inoculation with viable cultures. Figure. 4.3. Root hair production by P. vulgaris in response to inoculation with cell extracts of Rhizobium etli TAL 182 and B. megaterium S49 A. Control (sterile 0.1M phosphate buffer [pH 6.8]) B. Rhizobium TAL 182 WCE and Bacillus S49 WCE C. Rhizobium TAL 182 CFE and Bacillus S49 CFE D. Rhizobium TAL 182 WCE and Bacillus S49 (live cells) E. Rhizobium TAL 182 (live cells) and Bacillus S49 WCE F. Rhizobium TAL 182 CFE and Bacillus S49 (live cells) G. Rhizobium TAL 182 (live cells) and Bacillus S49 CFE H. Rhizobium TAL 182 (cup) and Bacillus S49 (live cells on plant roots) 74 75 Inoculation using ACE of S49 and TAL 182 gave similar results to the filter sterilized CFE (not shown). These results suggest that live (metabolically active) interacting bacterial partners may be necessary for inducing root hair proliferation in P. vulgaris. 4. 3. 4. Functional Bacillus and Rhizobium are necessary to induce root hair proliferation MLC 35 is a Tn5 mutant of Rhizobium etli capable of root hair deformation and nodule initiation, but which forms empty, non-fixing pseudonodules (George and Robert 1991). Coinoculation of Bacillus S49 with Rhizobium MLC 35 did not induce root hair proliferation (Figure 4.4 C), compared to dual inoculation of Bacillus S49 with wild-type Rhizobium TAL 182 (Figure 4.4 A). This observation is consistent with the non-proliferative response observed with coinoculation of an a-methyl tryptophan resistant mutant of Bacillus S49 (S49-1D) and Rhizobium TAL 182 (Figure 4.4 B). These results suggest that a fully functional (wild-type) Bacillus and Rhizobium may be necessary to induce root hair proliferation, even though the mechanism(s) of the interaction is unknown. 4. 3. 5. PCR verification of membrane separation of bacteria The PCR was used to determine the efficacy of the cup assay to separate the bacteria from the root system, while simultaneously monitoring their viability during the course of the experiment. One microlitre samples from the root medium and the "cup" were analyzed for the presence of bacteria of either genera using multiplex PCR using species-specific primers. No contamination of the root medium or the cup contents was observed (Figure 4.5), confirming the ability of the 30 kD Nalgene membrane to function as a selective filter, allowing passage only of lower molecular weight diffusible Figure 4.4. Root hair production by P. vulgaris in response to inoculation with Tn5 mutant (MLC 35) of Rhizobium etli TAL 182 and B. megaterium S49 A. Rhizobium TAL 182 and Bacillus S49 B. Rhizobium TAL 182 and Bacillus S49-1D C. Rhizobium etli MLC 35 and Bacillus S49 77 1 2 3 4 5 6 7 Figure. 4.5. Analysis of bacterial populations in cup assay experiments. Amplified fragments obtained by multiplex PCR using Bacillus S49 and Rhizobium TAL 182 specific primers, were separated on a 1.5% agarose gel. Amplified fragments obtained from pure cultures Rhizobium TAL 182 and Bacillus S49 were used as controls in lane 1 and 3, respectively. Samples removed from treatments with Rhizobium TAL 182 on roots and Bacillus S49 in the cup were analysed and run in lane 4 (root medium sample) and lane 5 (cup sample). Samples removed from treatments with Bacillus S49 on roots and Rhizobium TAL 182 in the cup were analysed and run in lane 6 (root medium sample) and lane 7 (cup sample). 100 bp ladder (BRL) was run in lane 2, and approximate sizes (in basepairs) of amplified fragments are indicated at left. 79 molecules. 4. 3. 6. Effect of coinoculation on lateral root growth The effect of Bacillus and Rhizobium inoculation on lateral root formation in P. vulgaris grown under gnotobiotic conditions in the growth chamber is summarized in Table 4.1. At four days of growth, single inoculation of B. megaterium S49 resulted in a significantly higher lateral root production in P. vulgaris compared to uninoculated control and single inoculation with Rhizobium etli TAL 182. At seven days after planting, single inoculation with B. megaterium S49 resulted in a significantly higher lateral root formation compared to other treatments. When the bacteria were physically separated from one another, the presence of Bacillus S49 on the plant roots consistently induced .significantly higher lateral root numbers (52.5 per plant), compared to the presence of Rhizobium TAL 182 (34.5 per plant) (Table 4.2). A similar effect on lateral root formation results was also observed when B. polymyxa L6 was present on the plant roots. Treatment combinations which included live Bacillus S49 cells on P. vulgaris roots consistently induced the formation of higher number of lateral roots compared to treatment combinations containing live Rhizobium TAL 182 cells (Table 4.3). The presence of live Bacillus S49 alone on the plant roots resulted in the formation of a significantly higher number of lateral roots compared to other treatments. Treatments which included live Rhizobium TAL 182 and cell preparations of Bacillus S49 resulted in a significantly lower number of lateral roots compared to treatments which had live Bacillus S49 with Rhizobium TAL 182 cell preparations. 4. 3. 7. Effect of coinoculation on nodulation and nitrogenase activity To determine if early nodulation in coinoculated plants is accompanied by 80 Table 4.1. Effect of Rhizobium and Bacillus inoculation on lateral root formation* in Phaseolus vulgaris grown in Leonard jars under gnotobiotic conditions Treatment Lateral root numbers Day 4 Day 7 Control* 14a 16a Rhizobium etli (TAL 182) 15.4a 236 B. megaterium (S49) 22.86 29.6c TAL 182 + S49 coinoculation 18a,6 236 • Control plants were inoculated with one ml of sterile 0.1M phosphate buffer (pH 6.8). f Lateral roots were counted at four and seven days after planting. Values represent the means of five replicates. Means were separated using analysis of variance (ANOVA). Means followed by the same letter in a column are not significantly different at p < 0.05, as determined by Duncan's multiple range test. This experiment was repeated once with similar results. 81 Table 4.2. Influence of Rhizobium and Bacillus inoculations on lateral root formation* in Phaseolus vulgaris grown in test tubes under axenic conditions Treatment Lateral root numbers Control** 19.25a Rhizobium etli (TAL 182) 34.506 S49* with Rhizobium etli TAL 182 on the roots 32.506 Bacillus S49 with TAL 182* on the roots 52.50c Bacillus L6 with TAL 182* on the roots 51.00c * Lateral roots were counted at five days of growth. ** Control plants received sterile 0.1M phosphate buffer (pH 6.8) as inoculum. Values represent the means of five replicates. Means were separated using analysis of variance (ANOVA). Means followed by the same letter are not significantly different at p < 0.05 as determined by Duncan's Multiple Range test. This experiment was repeated twice with similar results. * Physically separated from the plant root by a membrane (MW cutoff 30,000 Da). 82 Table 4.3. Influence of cell-free extracts of Rhizobium and Bacillus on lateral root formationt in Phaseolus vulgaris grown in test tubes under axenic conditions Treatment Lateral root numbers Rhizobium etli TAL 182 35.756.C Bacillus S49 48.38a S49 + TAL 182 ACE 44.37a,/> S49 + TAL 182 WCE 45.87a S49 + TAL 182 CFE 44.75a,6 TAL 182 + S49 ACE 31.86c,d TAL 182 + S49 WCE 30.57c,d TAL 182 + S49 CFE 26.28d f Lateral roots were counted at five days of growth. Means were separated using analysis of variance (ANOVA). Means of eight replicates followed by the same letter are not significantly different at p < 0.05 as determined by Duncan's multiple range test. This experiment was repeated once with similar results. Abbreviations used in the table: ACE = autoclaved cell extracts; WCE = French pressed whole cell extracts; CFE = Filter sterile cell free supernatant. 83 increased nitrogen fixation activity, plants were grown in Leonard jars under gnotobiotic conditions and harvested at 8, 10, 12 and 14 days after inoculation. The plants were evaluated for nodule number, nodule fresh weight, and nitrogenase activity. Control plants that received sterile phosphate buffer as inoculum did not form any nodules and were chlorotic. Plants that received Rhizobium TAL 182 and Bacillus S49 coinoculation showed significantly higher values for all the above parameters ( Figure 4.6 A-C). 4. 4. Discussion Previous studies have implicated bacterially derived IAA in root hair morphogenesis (Tien etal. 1979; Morgenstern and Okon, 1987; Barbieri and Galli, 1993). However, the experiments in this study indicated root hair proliferation in P. vulgaris was most likely a function of the synergistic interaction between a Rhizobium and a Bacillus and may not be dependent upon in vitro IAA production by Bacillus. To ascertain the mechanism(s) involved, the influence of a high IAA producing Bacillus megaterium Led 02 and a low IAA producing Bacillus brevis S72 coinoculated with Rhizobium TAL 182 on root hair formation in P. vulgaris was examined. Elongation of root hairs in response to coinoculation of Bacillus Led 02, an IAA producer, with Rhizobium TAL 182 is consistent with earlier observations, using IAA alone or in combination with B. megaterium S49. Changes to the ultrastructure of tomato root hairs in response to application of auxin have been reported (Ayling and Pring 1995). The absence of root hair proliferation in response to IAA or to single inoculation by lAA-producing Bacillus suggests that the root hair proliferation response is a direct consequence of a Rhizobium-Bacillus interaction. Inoculation with lAA-producing Azospirillum brasilense induced changes in root hair density on pearl millet (Tien et al. 1979) and on wheat (Barbieri and Galli 84 Figure. 4.6. Effect of dual inoculation of Rhizobium etli TAL 182 and Bacillus megaterium S49 on (A) nitrogenase activity (B) nodule number and (C) nodule fresh weight of P. vulgaris grown under gnotobiotic conditions in the growth chamber. Control plants that received sterile 0.1M phosphate buffer (pH 6.8) did not form any nodules. Each value represents the means of six replicates. Means of values for each parameter tested were significantly different at p < 0.05,as determined by Duncan's multiple range test. Bars represent the standard error of the mean. 86 1993). However, the effect of microbially derived IAA on root hair morphogenesis still remains unresolved as other studies indicate that IAA production may not be the sole factor in causing changes to root morphology and root hair morphogenesis (Harari et al. 1988; Yahalom et al. 1991). The root hair proliferation observed in P. vulgaris is a consequence of viable interacting bacterial species and thus differs from Azospirillum-plant interaction(s) where only a single bacterium is involved in inducing changes to root hair morphogenesis. The experiments in this study further confirm that the simultaneous physical presence of Rhizobium and Bacillus is necessary to induce root hair proliferation, since physical separation of either bacterial species precluded root hair proliferation in P. vulgaris. It is unlikely, therefore, that a low molecular weight substance is involved in the interaction, unless it is produced close to the site of action (root surface). Furthermore, viability of the bacterial inoculum was essential for the root hair proliferation response. These observations are consistent with the findings of Barbieri and Galli (1993) on the Azospirillum-wheai interaction. The most variable response was obtained when whole cell extracts of Rhizobium and Bacillus were applied to plant roots. Approximately 15% of such plants showed proliferation comparable to dual inoculation with live bacteria, thereby indicating that the mere presence of cell debris of the two bacterial species on plant roots may to a limited extent, trigger root hair proliferation. Such a phenomenon may be compared to the formation of shepherd's crooks in legumes, where Rhizobium needs to be physically present on the roots to cause such a response (Relic et al. 1993; van Rhijn and Vanderleyden 1995). Nevertheless, the results of this study are consistent with the view that soluble or extracelluar microbial 87 factors are not likely responsible for root hair proliferation in P. vulgaris. Functional Bacillus and Rhizobium may also be prerequisites for root hair proliferation. Coinoculation of B. megaterium S49 with Rhizobium etli Tn5 mutant MLC 35 failed to induce root hair proliferation in P. vulgaris. These data are consistent with the earlier observations that an a-methyl tryptophan resistant mutant of B. megaterium (S49-1D), also failed to affect root hair morphogenesis in dual inoculation experiments. However, changes to the IAA production potential of this mutant were also accompanied by other metabolic changes such as excretion of tryptophan into growing medium (Chapter 2), and hence this result should be viewed with caution. Nevertheless, it is interesting to note that changes to the growth pattern and/or metabolic phenotype of either Rhizobium or Bacillus resulted in a marked decrease in the synergistic induction of root hair proliferation response. This difference was also reflected in the decreased nodulation of P. vulgaris in response to dual inoculation of Rhizobium TAL 182 and 6. megaterium mutant isolate, S49-1D (Chapter 2). The physical presence of B. megaterium S49 on the roots resulted in the formation of more lateral roots than on control or co-inoculated plants. Such a response may be the consequence of Bacillus-derived IAA on plant root growth. At appropriate concentrations, IAA induces formation of lateral and adventitious roots (Scott 1972; Thimann 1972; Blakesley et al. 1991; King et al. 1995). Many authors have provided circumstantial evidence that bacterial production of plant hormones resulted in a stimulation of root branching or lateral root growth (Harari et al. 1988; Chanway et al. 1991). Lateral root development is known to result in significant disruption of existing root tissues and in the development of potential entry points for microorganisms 88 (Mahaffee et al. 1994). Bacillus S49 was originally isolated from the root interior of P. vulgaris 'Contender', and it is tempting to speculate that lateral root formation may help facilitate colonization of root internal tissues by this microorganism. Preliminary results of another study in this laboratory (Petersen et al. personal communication) have indicated a drastic decline in Bacillus S49 populations on the P. vulgaris root surface, five to seven days after inoculation, as determined by dilution-plating and PCR analyses. This decrease in rhizoplane population coincided with an increase in the population of Bacillus S49 in the root interior. Further experimentation is required to relate internal root colonization of Bacillus S49 to lateral root development in P. vulgaris. Lateral root enhancement in response to dual inoculation may be attributed to the temporal and spatial changes in phytohormonal levels on the root surface in response to changing bacterial populations, which in turn may regulate aspects of plant root development. Such a response is similar to observations of lateral root formation and nodulation observed in Medicago polymorpha inoculated with an Azospirillum-Rhizobium combination (Yahalom et al. 1991). Root hairs are infection sites for Rhizobium in legumes (Bauer 1981) and changes to root hair formation or density are likely to affect nodulation. Diaz et al. (1995) observed that R. leguminosarum bv. trifolii consistently induced formation of more nodules on psMransformed white clover hairy roots than on wild-type clover; transformed plants also formed nodules much earlier than the wild type. In P. vulgaris, most nodules are formed on lateral roots (George et al. 1992) and the increased nodulation observed in this study due to coinoculation may be due to the availability of greater numbers of infection sites for the Rhizobium microsymbiont, as a result of 89 increased root hairs on lateral roots. In these experiments, coinoculated plants formed significantly higher numbers of effective nodules at early stages of plant growth. Increased nodulation in P. vulgaris due to dual inoculation suggests that normal regulation of nodule numbers has been at least partially overcome in the presence of Bacillus S49. Such a response is consistent with previously reported studies on nodulation of alfalfa by R. meliloti in the presence of Pseudomonas spp. (Knight and Langston-Unkefer 1988). These results are also consistent with the view that combinations of bacteria were more effective than single strain inoculation in promoting plant growth and/or nodulation (Grimes and Mount 1984; Park et al. 1988). These data demonstrate a unique interaction between two rhizosphere bacteria in promoting lateral root development and root hair proliferation in P. vulgaris, which ultimately resulted in enhanced nodulation. A similar interaction between Rhizobium TAL 182 and nodulation promoting Bacillus polymyxa L6 in enhancing root hair induction and lateral root formation on P. vulgaris has also been observed. The results of coinoculation of B. polymyxa L6 and Rhizobium TAL 182 on lateral root formation and nodule formation in P. vulgaris are shown for information in Appendix ll-lll. The results of this study also indicate that the root hair proliferation response may require a "dialogue" between two viable rhizosphere bacteria. Based on the currently available information, it is tempting to suggest a model which may account for the altered root hair morphogenesis in P. vulgaris in response to coinoculation. The presence of Bacillus spp. on the plant root might induce the plant to release compounds leading to the induction or derepression of Rhizobium genes that are involved in synthesis of a factor causing root hair proliferation. It is also possible that a Rhizobium-90 derived factor may be modified by Bacillus or vice-versa, and that such a modified factor is involved in root hair proliferation. Evaluation of such models has been hindered by the lack of availability of suitable mutants of Rhizobium etli TAL 182 and/or Bacillus megaterium S49. In future, development of appropriate Rhizobium and Bacillus mutants may assist in the determination of the genetic and biochemical basis of this synergistic interaction. 91 Chapter 5 Nodulation of Phaseolus vulgaris by Rhizobium etli is enhanced by the presence of Bacillus From beans to bacteria to the scavenger-hunter bond between larval crabs and jelly fish, symbiotic relationships show that peaceful existence is part of the very foundation of nature. Robert Masello (1986) 92 5.1. Introduction Root nodule formation is a complex developmental process involving sequential exchange of chemical signals between the bacterial microsymbiont and the host plant (Fisher and Long 1992; Fellay et al. 1995). Both plant and bacterial signals have been identified and shown to play a major role in the specificity of the Rhizobium-\egume interaction (Relic et al. 1994; Diaz et al. 1995; Fellay et al. 1995). None of the cascade of steps that leads to nodulation is completely specific for a given host-microsymbiont combination (Fellay et al. 1995). Shantaram (1986) has observed ineffective (non-fixing) nodulation of soybean by a non-specific microsymbiont, Rhizobium phaseoli. Similar anomalous (heterologous) nodulation resulting in the formation of ineffective nodules has also been reported by other workers (Bal et al. 1982; Shantaram and Wong 1982). Autoregulation is one of several host factors controlling nodulation (Pierce and Bauer 1983). Jamming of nodulation (Jan), where pre-treatment of host roots with pure Nod-factor abolishes nodulation by subsequent, symbiotic bacteria (van Brussel et al. 1993), and competitive nodulation blocking (Cnb), where pre-treatment of plants with an incompatible Rhizobium blocks subsequent nodulation by compatible Rhizobium (Dowling et al. 1988) have been identified as regulatory phenomena in many legume hosts. Mellor and Collinge (1995) consider these processes to be extensions of autoregulation of nodulation. It is evident from the above studies that successful nodulation of the legume host is predicated on the genetic compatibility of the host plant and the rhizobial microsymbiont. Specific microorganisms have considerable potential to alter the composition and activity of rhizosphere microflora such as Rhizobium (Schroth and Becker 1990). Nodulation-promoting bacteria (Kloepper et al. 1988) belonging to the Azospirillum 93 (Schmidt ef al. 1988), Pseudomonas (Bolton et al. 1990), Streptomyces (Li and Alexander 1990) and Bacillus (Halverson and Handelsman 1991) genera have been reported. The beneficial effects of these bacteria have been variously attributed to their ability to produce various compounds including phytohormones (Schmidt et al. 1988), toxins (Knight and Langston-Unkefer 1988), or antibiotics (Li and Alexander 1990), to suppress deleterious rhizobacteria (Turner and Backman 1991) or through some other unidentified mechanism (Halverson and Handelsman 1991). Coinoculation of wild-type strains or nod gene mutants of Rhizobium meliloti along with an Agrobacterium cured of the Ti plasmid, enhanced nodulation of alfalfa by a wild type Rhizobium meliloti (Caetano-Anolles and Bauer 1988). None of these studies has adequately explained the mechanism of heterologous nodulation promotion or the influence of associated bacteria on the nodulation of legumes by rhizobia. While substantial progress has been made in elucidating the mechanisms of RhizobiumAegume interaction, there is little information available on the direct influence of rhizosphere bacteria on nodulation of legumes by Rhizobium. It has been shown in previous chapters that coinoculation of Rhizobium etli TAL 182 with Bacillus spp. induced root hair proliferation in Phaseolus vulgaris and enhanced nodulation of P. vulgaris by Rhizobium. Since root hairs are the infection sites for Rhizobium, it was important to determine whether proliferation of root hairs in response to coinoculation would affect the nodulation capability of Rhizobium etli. The objective of this study was to demonstrate the specificity of the synergistic interaction between Rhizobium and Bacillus in enhancing root hair formation in P. vulgaris . The influence of Bacillus spp. on the nodulation capability of Rhizobium etli in overcoming host mediated autoregulation, and in inducing heterologous nodulation on Phaseolus acutifolius was 94 also examined in this study. 5. 2. Materials and Methods 5.2.1. Bacterial strains and media • Bacillus spp. were originally isolated from the rhizosphere of Phaseolus vulgaris as described previously (Chapter 2). Rhizobium etli TAL 182 and Bradyrhizobium spp. TAL 644 were obtained from NifTAL, Paia, Hawaii. Species identity was determined by analysis of carbon-source utilization (BIOLOG™). The distinct genera Bradyrhizobium and Rhizobium were further confirmed by sequencing a 320 bp region of the 23S rDNA (Petersen et al. 1995). Rhizobium etli TAL 182 and Bradyrhizobium spp.TAL 644 share only 68.7% nucleotide sequence identity within this region, and oligonucleotides were produced which distinguish each strain. Bacillus spp. were routinely maintained on combined carbon medium agar slants (Rennie 1981). Stock cultures of Rhizobium etli TAL 182 and Bradyrhizobium spp. TAL 644 were maintained on yeast mannitol agar slants (Bohlool and Schmidt 1970). Broth cultures were grown to early stationary phase in combined carbon medium for Bacillus spp. and in yeast mannitol broth for both Rhizobium TAL 182 and Bradyrhizobium TAL 644. The cells were harvested by centrifugation, washed twice with 0.1M sterile phosphate buffer (pH 6.8) and resuspended in the same buffer to a specified cell density before inoculation. 5.2.2. Preparation of bacterial extracts Cell free extracts (CFE) of Bacillus megaterium S49 were prepared by harvesting 50-ml cultures by centrifugation at 10000 x g for 30 minutes using a Sorvall ® GSA-13 rotor. The supernatant was filtered (0.22 pm) to produce CFE. Autoclaved cell preparations (ACE) were obtained by autoclaving early stationary phase cultures for 15 minutes at 121°C and 15 psi. The absence of live bacteria in the above preparations 95 was confirmed by streaking a loopful of each preparation on nutrient agar (Difco) plates. 5.2.3. Plant material and growth Seeds of P. vulgaris 'Contender' were obtained from Pacific Northwest Seed Co., Vernon, BC. Seeds of P. acutifolius were obtained from NifTAL, Paia, Hawaii. Seeds were surface sterilized by soaking in 30% (v/v) H 2 0 2 for 15 minutes and rinsed seven times with sterile distilled water and were germinated aseptically on moist filter paper in sterile petri plates. Two-day old pre-germinated seeds with uniform radicle lengths were inoculated by incubation for 15 minutes in 20 ml early stationary phase bacterial suspensions of Rhizobium ef//TAL182, Bradyrhizobium TAL 644 and B. megaterium S49 at a cell density of 2 x 108 cfu/ml. The experiment was established in a completely randomized design with six replicates per treatment and was repeated once with comparable results to those presented here. 5.2.4. Scanning electron microscopy (SEM) Plant growth conditions and scanning electron microscopic analysis for studying root hair formation in P. vulgaris in response to dual inoculation of Bacillus S49 and Bradyrhizobium TAL 644 were as described previously (Chapters 2 and 3). 5.2.5. Delayed inoculation experiment A time course analysis of inoculation effect was conducted using modified Leonard jar assemblies (Blauenfeldt et al. 1994) containing sterile planting medium of industrial sand and Turface™ (1:1 v/v). Cells of Bacillus megaterium S49 were inoculated on to two-day old pregerminated seedlings and plants grown in growth chambers with a 16 h photoperiod at 350 uE m2 s'1 illumination and 22°C/18°C day night temperatures. Rhizobium etli TAL 182 inoculation was delayed by four, seven or ten days after Bacillus spp. inoculation. One ml of Rhizobium etli TAL 182 bacterial 96 suspension at 2 x 108 cfu/ml was applied directly to the lower stem of each plant at the point of emergence from soil. Coinoculated plants received simultaneous inoculation of both bacterial species (2 x 108 cfu/ml) at time zero. Control plants were inoculated with one ml of sterile 0.1M phosphate buffer (pH 6.8) at each inoculation period. One-tenth strength nitrogen-free sterile nutrient medium (Broughton and Dilworth 1971) was used to water the plants. Plants were harvested at 14 days after Rhizobium inoculation and the nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (acetylene reduction activity) were determined. The acetylene reduction assay described by Dilworth (1966) was used to measure nitrogenase activity. After one hour of incubation, the gas mixture was analyzed by flame-ionization gas chromatography using a Hewlett Packard 5830A gas chromatograph (Holl et al. 1988). Analysis of variance of the data was performed using the SAS computer program (SAS Inc., Cary, NJ). 5.2.6. Determination of speed of nodulation in growth pouches To compare the speed of nodulation by Rhizobium TAL 182 in the presence and absence of Bacillus S49, seeds of P. vulgaris were surface sterilized and germinated as described above. Two-day old pre-germinated seeds were transferred to sterile plastic growth pouches (Mega International, MN) containing 15 ml of one tenth strength nitrogen-free sterile nutrient medium (Broughton and Dilworth 1971). The treatments were arranged in a completely randomized design with ten replicates. Two days after transfer, Rhizobium etli TAL 182, with and without B. megaterium S49, was inoculated individually on each of ten replicate plants at 108 cfu/ml. One ml of each bacterial suspension containing approximately 108 cfu was used as inoculum. Ten-ml of nitrogen-free nutrient solution was maintained in the growth pouches for the duration of the 97 experiment. The location of newly emerged nodules was marked on the growth pouch as soon as they were visible; new nodules were counted on alternate days up to 15 days after inoculation. 5.2.7. Nodulation of P. vulgaris using Bacillus spp. extracts The growth pouches and experimental conditions were identical to those described above. For treatments involving autoclaved cell preparations (ACE) or cell-free extracts (CFE) extracts, the pre-germinated seeds were incubated in the respective preparations for 15 minutes after treatment with live Rhizobium TAL 182. Plants were grown in the growth chambers with a 16 h photoperiod at 350 uE m2 s"1 illumination and 22°C/18°C day night temperatures. 5.2.8. Split-root experiment in growth pouches Modified growth pouches for the split-root experiment were prepared as described by George et al. (1992). A ten-ml volume of sterile nitrogen-free nutrient medium was maintained on each side of the growth pouch. Seeds of P. vulgaris were surface sterilized and germinated as described above. The split-root systems were generated by cutting off approximately 75mm of the root tip of two day old radicles and transferring a single plant to each modified growth pouch. Five days after transfer, the primary side of the split-root was inoculated with a one ml suspension (108cfu/ml) of Rhizobium etli TAL 182. The opposite (secondary) side received Rhizobium TAL 182 alone or in combination with Bacillus S49 (one ml suspension containing 108 cfu) after a given delay period. For each delay period, the positive control consisted of no inoculation on the primary side combined with a delayed inoculation on the secondary side. Plants that received sterile phosphate buffer at both inoculation events were used as controls and grown under the same conditions described above. Plants were 98 harvested 24 days after the primary inoculation and the nodules counted. Percentage suppression of nodulation was calculated based on the formula of George et al. (1992), which is given in Appendix IV for information. The data was analyzed using the analysis of variance in the SYSTAT PC computer package (SYSTAT Inc., Evanston) after arcsine transformation of percent values. 5.2.9. Nodulation studies on P. acutifolius Bacterial inoculation of P. acutifolius using B. megaterium S49 in combination with either Rhizobium etli TAL 182 or Bradyrhizobium TAL 644 was performed on two-day old pre-germinated seeds as described above. Plant growth conditions were as described for the delayed inoculation experiment in Leonard jars. Nodules were harvested from 24 day old plants and stored at -70 0 C . 5.2.10. Detection of R. etli TAL 182 in P. acutifolius nodules by fluorescent antibody staining Fluorescein isothiocyanate conjugated (FITC) polyclonal antibodies specific to Rhizobium etli TAL 182 were obtained from NifTAL, Paia, Hawaii. Nodules were surface sterilized by sequential immersion in 95% ethanol for two minutes, 1.2% sodium hypochlorite (three minutes) and five rinses with sterile deionized water. Nodules were halved with a sterile scalpel and incubated for three hours in microcentrifuge tubes containing 200 pi of sterile deionized water at 30° C, to release Rhizobium. The tubes were centrifuged briefly (2000 x g for 15 seconds) to remove nodule debris; 20 pi of the supernatant was applied to each well of a clean toxoplasmosis microscopic slide (Bellco Glass Inc., NJ) and dried on a slide warmer at 50° C. The samples were fixed by immersion in acetone for 10 minutes and air-dried. The fluorescent antibodies were diluted 1:100 in one ml of 10mM phosphate buffered saline (pH 7.2) containing 2% 99 blotto (10g skim milk powder, 0.2g NaN3 in 100 ml de- ionized water), and 20 ul of antibody solution was used for each reaction. The slides were incubated in a moist chamber at 37° C for one hour, rinsed with water once, and dried in a slide warmer as described above. A small drop of mounting fluid was placed on each well and covered with a cover slip. The edges of the cover slip were anchored with nail polish and the samples were observed under epifluorescence using a Zeiss ™ axiophot microscope (100x objective; 10x occular) with a fluorescein filter-set. The excitation wavelength used was 400-450nm with an LP 520 barrier filter. 5.2.11. Polymerase chain reaction (PCR) Nodules were surface sterilized as described above with the addition of a subsequent one minute rinse in 0.3M NaOAc, pH 4.2, followed by three rinses with sterile deionized water to remove extraneous DNA molecules. When the final water rinse was decanted from the sterilized nodules, a small volume of rinse water remained in the microcentrifuge tube with the nodules. A 50 pi aliquot of this liquid was removed for use as a negative control. Nodules were then crushed in 50 pi sterile deionized water to release bacteria. Whole-cell bacterial DNA amplification by multiplex PCR was conducted as described by Petersen et al. (1995) using the conserved forward primer BAC11X in combination with both Rhizobium etli TAL 182-specific and B. megaterium S49-specific reverse primers. The presence or absence of amplified product, as well as the fragment size (540 bp for Rhizobium TAL 182), was determined by electrophoretic analysis using 1.5% agarose gels stained with ethidium bromide and viewed under UV light. 100 5.3. Results 5.3.1. Root hair formation in P. vulgaris in response to dual inoculation with Bacillus S49 and Bradyrhizobium TAL 644 Scanning electron microscopy revealed that inoculation of P. vulgaris with Bradyrhizobium TAL 644 alone did not enhance root hair formation (Figure 5.1B), compared to the uninoculated control (Figure 5.1 A). Coinoculation of Bradyrhizobium TAL 644 with Bacillus S49 did not result in enhanced root hair formation (Figure 5.1C), compared to P. vulgaris plants that received dual inoculation of Rhizobium TAL 182 and Bacillus S49 (Figure 5.1D). Nevertheless, the presence of Bradyrhizobium TAL 644 and Bacillus S49 resulted in the formation of more root hairs in P. vulgaris compared to the uninoculated control (Figure 5.1 A) and plants that received a single inoculation of Bradyrhizobium TAL 644 (Figure 5.1 B). 5.3.2. Nodulation suppression induced by Rhizobium TAL 182 is overcome by coinoculation with Bacillus S49 Inoculation of Rhizobium TAL 182 on the primary side of the split-root resulted in suppression of nodulation on the secondary side of the split-root after a given delay period (Table 5.1). However, when coinoculated with Bacillus S49, Rhizobium TAL 182 partially overcame this suppression; significant effects on nodulation were observed for both the four and seven day delay periods, compared to suppression with single inoculation of Rhizobium TAL 182 at time zero. These results suggest that coinoculation of Rhizobium TAL 182 and Bacillus S49 enhances the nodulation potential of TAL 182 by partially overcoming host-mediated suppression of nodulation. This effect was time-dependent, since a ten day delay in secondary inoculation produced no effect of coinoculation on nodulation suppression. 101 Figure 5.1. Root hair formation in P. vulgaris in response to bacterial inoculation. (A) Uninoculated control (sterile 0.1M phosphate buffer, pH 6.8) (FJ) Bradyrhizobium TAL 644 (C) Bradyrhizobium TAL 644 with Bacillus S49 (D) Rhizobium TAL 182 with Bacillus S49. 102 103 Table 5.1. Nodulation suppression induced by Rhizobium etli TAL 182 with and without B. megaterium S49 as co-inoculant on P. vulgaris 'Contender' Treatment Delay in secondary Percent suppression inoculation (days) TAL 182 1 7.79 b 4 21.71c 7 51.36 d 10 92.08 e TAL 182 + S49 1 0.62 a 4 7.90 b 7 30.03 c 10 90.59 e Primary side of the split-root was inoculated with Rhizobium TAL 182 at day zero, while the secondary side was inoculated with Rhizobium TAL 182 or Rhizobium TAL 182 with Bacillus S49 following a given delay period. Values represent the means of five replicates. Means were separated using analysis of variance (ANOVA). Means followed by the same letter are not significantly different at p < 0.05, as determined by Fisher's LSD. Statistical analysis was carried out on arcsine transformed percentage values. 104 5.3.3. Rate of nodulation of P. vulgaris by Rhizobium TAL 182 is enhanced in the presence of Bacillus S49 The ability of a given Rhizobium strain to form nodules rapidly on a host plant can affect the outcome of competition (de Oliveira and Graham 1990). The ability of B. megaterium S49 to enhance the nodulation rate by Rhizobium etli TAL 182 was tested by observing the time course of nodule formation on P. vulgaris. Coinoculation of Rhizobium TAL 182 and Bacillus S49 resulted in early nodulation of P. vulgaris, compared to single inoculation with TAL 182 (Figure 5.2). These co-inoculated plants showed peak nodulation 11 days after inoculation, whereas singly inoculated plants showed peak nodulation 13 days after inoculation. In addition to earlier nodule formation, co-inoculated plants also formed more nodules during the first two weeks of plant growth (47 per plant) compared to plants that received a single inoculation of Rhizobium TAL 182 (34 per plant) (Figure 5.2). Coinoculation of Bradyrhizobium TAL 644 with B. megaterium S49 did not significantly enhance nodulation of P. acutifolius by TAL 644 (Table 5.2). 5.3.4. Effect of timing of inoculation on nodulation of P.vulgaris by Rhizobium TAL 182 The relative timing of inoculation of Bacillus S49 and Rhizobium TAL 182 had significant effects on the nodulation capability of TAL 182. When both bacterial species were simultaneously inoculated onto P. vulgaris, the nodulation of P. vulgaris by Rhizobium TAL 182 was significantly enhanced compared to single inoculation with Rhizobium TAL 182, as evidenced by higher values for nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) per plant (Figure 5.3). An initial 105 A 50 T 45 c S° 40 Q. w 35 a g 30 3 •D O c o o 2 25 20 15 10 TAL 182 B 50 T J 4 5 a. 40 | 35 jj 30 •D g 20 5 15 c 10 25 o o z 0 5 --S49 + TAL 182 + 7 9 11 13 Days after inoculation 14 Figure 5. 2. Rate of new nodule formation of R. etli TAL 182 on Phaseolus vulgaris 'Contender' in growth pouches when inoculated (A) alone and, (B) with B. megaterium S49. Each value represent the means of nodule numbers from ten replicate plants. 106 Table 5.2. Effect of bacterial inoculation on nodule number (NON), nodule fresh weight (NFW) and nitrogenase activity (ARA) in Phaseolus acutifolius under gnotobiotic conditions in Leonard jars Treatment NON (±SE) NFW ARA (n moles C 2 H 4 h"1 (g/plant ±SE) plant"1 ±SE) Control 0.00 a 0.000 a 0.000 a Bradyrhizobium 44.33 ± 4.55 b 0.34 ± 0.03 b 5.59 ± 0.29 b TAL 644 Bradyrhizobium 41.00 ± 2.99 b 0.31 ± 0.02 b 5.50 ± 0.24 b TAL 644 + Bacillus S49 Bradyrhizobium 39.80 ± 3.24 b 0.31 ± 0.02 b 5.49 ± 0.23 b TAL 644 + Bacillus L6 Values represent the means of six replicates. Means were separated using ANOVA and means followed by the same letter in a column do not differ significantly at p < 0.05 as determined by Duncan's multiple range test. 107 Figure 5.3. Effect of timing of inoculation of Rhizobium etli TAL 182 and B. megaterium S49 on (A) nodule number, (B) nodule fresh weight and (c) nitrogenase activity of P. vulgaris 'Contender' grown under gnotobiotic conditions in the growth chamber. Control plants that received 0.1M sterile phosphate buffer did not form any nodules. Each value represent the means of six replicates. Means of values for each parameter tested were significantly different at p < 0.05, as determined by Duncan's multiple range test. Bars represent the standard error of the mean. 109 inoculation with Bacillus S49 followed by Rhizobium inoculation after a given delay period (4, 7,or 10 days) significantly reduced nodulation of P. vulgaris by TAL 182 relative to that observed in the absence of Bacillus S49. A similar delay in Bacillus spp. inoculation nullified the coinoculation nodulation enhancement of P. vulgaris by Rhizobium TAL 182 (data not shown). These results suggest that direct interaction between the two bacterial species may be necessary for enhanced nodulation of P. vulgaris by Rhizobium TAL 182. 5.3.5. Effect of non-viable cell preparations of S49 on the nodulation of P. vulgaris by Rhizobium TAL 182 Autoclaved cell preparations (ACE) and cell-free preparations (CFE) of Bacillus S49 were tested for viability before they were mixed with Rhizobium TAL 182 for dual inoculation. Neither inhibition nor stimulation of nodulation of P. vulgaris by Rhizobium TAL 182 was observed in the absence of live Bacillus S49 cells, compared to single inoculation with Rhizobium TAL 182 (Table 5.3). 5.3.6. Heterologous nodulation of P. acutifolius by TAL 182 in the presence of Bacillus S49 The nodulation response of P. acutifolius to inoculation with Rhizobium TAL 182 alone, or in combination with Bacillus S49 was examined and is presented in Table 5.4. Inoculation of P. acutifolius with Rhizobium TAL 182 alone resulted in the formation of a few very small white swellings (pseudonodules) on the roots. Such structures showed no evidence of Rhizobium TAL 182 internally as revealed by fluorescent antibodies specific to Rhizobium TAL 182 (Figure 5.4 D). The specificity of the fluorescent antibody was confirmed by using a pure culture of Rhizobium TAL 182 (10"6 110 Table 5.3. Nodulation of P. vulgaris by Rhizobium etli TAL 182 in the presence of cell preparations of B. megaterium S49 Treatment No of nodules/plant (means ± SE) Nodule fresh weight (g/plant ± SE) Control TAL 182 TAL 182 + S49 TAL 182 +S49 (CFE) TAL 182 + S49 (ACE) 2.66 ± 1.14 a 46.66 1 3.79 b 69.50 ± 3.37 c 46.30 ± 3.14 b 46.50 ± 2.90 b 0.001 ± 0.0007a 0.28 ± 0.04 b 0.40 ± 0.02 c 0.31± 0.02 b 0.27 ± 0.03b Abbreviations: CFE= Cell free supernatant; ACE = Autoclaved cell preparation Values represent the means of six replicates. Means followed by the same letter in a column are not significantly different at p < 0.05 as determined by Duncan's multiple range test. dilution) as a positive control. Fluorescent antibody specific to Rhizobium TAL 182 did not cross react with pure cultures of Bradyrhizobium TAL 644 or with the same microsymbiont within the nodules of P. acutifolius. Rhizobium TAL 182 could not be detected inside such pseudonodules even when an undiluted nodule eluate was used for antibody detection. However, coinoculation of Rhizobium TAL 182 and Bacillus S49 resulted in the formation of a few nodules that were slightly pink and contained Rhizobium TAL 182 as detected by fluorescent antibodies in an undiluted nodule eluate (Figure 5.4 C). Dilutions greater than 10'2 of the nodule eluate did not show the presence of Rhizobium TAL 182. Direct smearing of the nodules onto microscopic slides and subsequent detection with fluorescent antibody was precluded by the Table 5.4. Nodulation of P. acutifolius 24 days after inoculation with R. etli TAL 182 alone or with B. megaterium S49 Treatment No of No of nodules No of nodules Percent nodules bearing plants analyzed * bearing TAL 182 TAL 182 Control 10 0 0 0 TAL 182 14 11t 0 0 TAL182 + S49 14 23t 4¥ 17.4 * Total number of nodules observed on fourteen plants in two separate experiments, t Small white nodule-like outgrowths (pseudonodules) ¥ Nodules pink in colour, slightly bigger compared to the rest 112 Figure 5.4. Detection of heterologous nodulation of P. acutifolius by Rhizobium ef//'TAL 182 in the presence of Bacillus megaterium S49 by using fluorescent antibodies specific to Rhizobium TAL 182. (A) pure culture of Rhizobium TAL 182, (B) Rhizobium TAL 182 in the nodule of its homologous host, P. vulgaris, (C) Rhizobium TAL 182 in the nodule of P. acutifolius and, (D) the nodule smear from singly inoculated plants. Arrow indicates the presence of Rhizobium TAL 182. 1 1 3 114 presence of heavy background fluorescence. The ineffectiveness of nodules formed by Rhizobium TAL 182 on P. acutifolius was apparent by the chlorotic appearance of the plants and stunted growth. Acetylene reduction assays of plants showed no evidence of nitrogenase activity. Only three of the 15 (20%) coinoculated plants formed such nodules and the frequency of heterologous nodulation was low (17.4%) (Table 5.4). Thepresence of Rhizobium TAL 182 inside the ineffective nodules of coinoculated plants was further confirmed by PCR amplification of surface sterilized nodules using Rhizobium TAL 182 specific primers, which showed an approximately 540 bp band corresponding to TAL 182 (Figure 5.5). PCR analysis of the pseudonodules of P. acutifolius plants that received single inoculation of Rhizobium TAL 182 did not show the presence of this microsymbiont inside these nodules (data not shown). Simultaneous PCR analysis with a second primer specific for B. megaterium showed no evidence of S49 within these nodules. 5.4. Discussion This study has shown that the presence of viable B. megaterium S49 could significantly influence the Rhizobium-iegume symbiotic interaction. The presence of both bacterial species is apparently necessary to enhance nodulation of P. vulgaris by Rhizobium etli TAL 182, as delayed inoculation of Rhizobium or Bacillus or coinoculation of Rhizobium TAL 182 with non-viable cell preparations of Bacillus S49 failed to show enhancement of nodulation comparable to dual inoculation. It has been shown in previous chapters that coinoculation of Bacillus S49 with Rhizobium TAL 182 enhances root hair formation in P. vulgaris and on P. acutifolius. Since root hairs are infection sites for rhizobia (Bauer, 1981), it is possible that the nodulation enhancement observed for P. vulgaris -in response to coinoculation may be attributed to increased numbers of 11 1 2 3 4 5 600 —> Figure 5.5. Presence of R. etli TAL 182 inside the surface-sterilized nodules of P. acutifolius as determined by multiplex PCR. Amplification products from PCR using Rhizobium TAL 182-specific and Bacillus S49-specific primers as described in text (section 5.2.11.) were separated on 1.5% agarose gels. Lane (1) Surface-sterilized nodule exterior, Lane (2) surface-sterilized nodule interior, of dual inoculated plants. B. megaterium S49 pure culture in Lane (4) and R. etli TAL 182 pure culture in Lane (5) were used as controls. Bands corresponding to R. etli TAL 182 (540 bp) were clearly distinguished from 650 bp B. megaterium S49 fragment as shown by the bright 600 bp fragment of the 100 bp ladder (Lane 3), and indicated by an arrow at left. The lower band denotes unincorporated primers and nucleotides. 116 infection sites available for Rhizobium TAL 182. Under the gnotobiotic conditions used in these experiments, enhanced nodulation may reflect an increased susceptibility of P. vulgaris root hairs to infection by Rhizobium TAL 182, triggered by the presence of B. megaterium S49. Coinoculation of Rhizobium etli TAL 182 with Bacillus polymyxa L6, a demonstrated plant growth promoter (Holl et al. 1988), produced similar nodulation responses on P. vulgaris (Appendix V). The presence of a few pseudonodules on control plants may be attributed to spontaneous nodulation by P. vulgaris in the absence of Rhizobium; such a phenomenon has been previously observed in clovers (Blauenfeldt etal. 1994). It has been shown by others that coinoculation of symbiotically deficient mutants of R. meliloti and heterologous members of the Rhizobiaceae contributed to enhanced nodulation by wild-type R. meliloti on alfalfa (Caetano-Anolles and Bauer 1988). Similar enhancement of nodulation of legumes due to coinoculation of bacterial mixtures has been reported previously (Rolfe and Gresshoff 1980; Grimes and Mount 1984). The failure of cell-free preparations of B. megaterium S49 to stimulate nodulation of P. vulgaris by Rhizobium etli TAL 182 suggests that simple triggering of nodulation responses may not be solely due to extracellular substances produced by the Bacillus. These data reinforce the hypothesis that viable bacteria are required to participate in nodulation enhancement, as well as the possible involvement of a stimulatory substance(s) produced by Bacillus which may be subject to regulation by Rhizobium or vice-versa. Such an hypothesis has been proposed by Caetano-Anolles and Bauer (1988) to account for the beneficial association of Agrobacterium and R. meliloti in enhancing alfalfa nodulation. Bacillus S49 may also prime the plant to facilitate late rhizobial infection, similar to the results observed for Azospirillum-Rhizobium interactions 117 on clovers (Plazinski and Rolfe 1985). The inability of Rhizobium TAL 182 to form effective nodules after delayed inoculation may be a consequence of decreased susceptibility to infection of the root hairs of P. vulgaris . Susceptibility of host root cells to infection by Rhizobium is both transient and developmental^ regulated (Bhuvaneswari etal. 1981). Autoregulation by the plant host ultimately restricts nodulation (Pierce and Bauer 1983). Although Rhizobium TAL 182 shows nodulation suppression in split-root experiments (George et al. 1992), coinoculation with Bacillus S49 was able to partially offset that host-mediated suppression. It is tempting to speculate that coinoculation may induce the production of nod off compounds (compounds that suppress the expression of common nod-genes in Rhizobium) in the plant (Peters and Long 1988; Grandmaison and Ibrahim 1994; Mellor and Collinge 1995), to enable late infections on the secondary side of the split-root. Such mobilization of nod off compounds in split-root systems has been hypothesized by Caetano-Anolles and Gresshoff (1990). In P. vulgaris, delayed suppression of nodulation may be attributed to the requirement for proliferation of bacteria inside the nodules; Rhizobium infection per se does not trigger the suppression response observed in other legumes (George and Robert 1991). Partial reversal of suppression in this study may also reflect higher numbers of infection sites (root hairs) facilitated by the presence of Bacillus S49. Coinoculation of Rhizobium TAL 182 and Bacillus S49 also induced heterologous nodulation of TAL 182 on P. acutifolius. Although P. vulgaris and P. acutifolius belong to the same genus, they are in different cross-inoculation groups (Allen and Allen 1981; Somesegaran et al. 1991). It was not possible to detect or re-isolate Rhizobium TAL 182 from the few pseudonodules observed on singly inoculated plants using PCR and 118 dilution-plating techniques. Heterologous nodulation of P. acutifolius by Rhizobium TAL 182 coinoculated with Bacillus S49 implicates the Bacillus more directly in the nodulation process. The enhancement of root hair production in P. vulgaris in response to coinoculation may have influenced both the infection process and the potential for heterologous nodulation. Similar heterologous nodulation on hairy roots of clover plants by a non-specific symbiont, R. leguminosarum bv. viciae, has been reported (Diaz et al. 1995). It is interesting to note that coinoculation of Bacillus S49 (or B. polymyxa L6) with Bradyrhizobium TAL 644 neither induced root hair proliferation on P. vulgaris nor enhanced nodulation by Bradyrhizobium TAL 644 on its host, P. acutifolius. These results suggest that the Rhizobium genotype might be more decisive than Bacillus in expressing the synergistic effect. Thus, reduced root hair proliferation in P. vulgaris in response to dual inoculation with Bradyrhizobium TAL 644 and Bacillus S49 reflects a degree of specificity in the synergistic interaction of Rhizobium and Bacillus on root hair formation and nodulation enhancement. Specific interaction between Rhizobium genotypes and Azospirillum in enhancing nodulation of legumes has been reported (Iruthayathas et al. 1983). Since the legume host can control and optimize events that lead to the formation of nodules (Caetano-Anolles and Gresshoff 1991), the lack of enhanced root hair formation and nodulation in P. acutifolius in response to coinoculation may be attributed to the host plant regulation of such events. Further work using well-characterized Bacillus and Rhizobium mutants may help elucidate the mechanism(s) of such an interaction and its consequences on plant nodulation and productivity. 119 Chapter 6 Altered root hair formation in Phaseolus vulgaris in the presence of Rhizobium, Bacillus and the ethylene biosynthesis inhibitor aminoethoxy vinyl glycine If I have seen a little further it is by standing on the shoulder of giants. Isaac Newton (1642-1727) 120 6.1. Introduction Ethylene has been implicated in a number of processes that affect plant growth and development (reviewed by Kieber and Ecker 1993). External application of the ethylene precursor, 1-aminocyclopropane-1-carboxylic acid (ACC) directly enhances ethylene production in plant tissues (Adams and Yang 1979; Liirssen et al. 1979; Tanimoto et al. 1995). Both ethylene and auxin are involved in root hair formation in Arabidopsis thaliana (Masucci and Schiefelbein 1994) and induction of root hair formation and inhibition of longitudinal root growth have been attributed to the effects of ethylene on plant roots (Goodlass and Smith 1979; Feldman 1984; Zaat et al. 1989). Exogenously applied ethylene promoted root hair formation in pea, faba bean and lupine (Abeles et al. 1992). External application of aminoethoxy vinyl glycine (AVG), which is an inhibitor of ethylene biosynthesis in plants, resulted in complete loss of root hair formation in Arabidopsis (Tanimoto et al. 1995), suggesting that ethylene may be a positive regulator of root hair formation (Dolan et al. 1994; Tanimoto et al. 1995). Among the diverse group of soil microorganisms capable of producing ethylene (Primrose 1979), members of both Rhizobium and Bacillus genera are capable of producing ethylene in vitro, when grown in the presence of the precursor, methionine (Billington et al. 1979; Mansouri and Bunch 1989). Ethylene of microbial origin is capable of influencing plant growth either directly (Arshad and Frankenberger 1988) or indirectly (Glick et al. 1995). Root elongation in canola in response to Pseudomonas putida inoculation was postulated to involve deamination of plant-derived ACC, leading to lower concentrations of ethylene in plant tissues (Glick et al. 1994a, b). Nod-factors produced by Rhizobium may also induce localized production of ethylene in root tissues leading to the aberrant thick and short root (Tsr) phenotype observed in Vicia sativa 121 subspp. nigra (van Spronsen et al. 1995). Such observations suggest that microbial activity in the vicinity of plant roots may elicit ethylene-mediated changes to root morphology. It has been shown in previous chapters of this thesis that co-inoculation of Rhizobium etli TAL 182 and Bacillus megaterium S49 induced proliferation of root hairs in P. vulgaris and that the plant response was dependent upon two viable interacting bacterial species. To investigate the possible role of endogenous ethylene in the proliferation response, root hair formation in P. vulgaris was examined in the presence of Bacillus, Rhizobium, ACC (an ethylene precursor) and AVG (an ethylene biosynthesis inhibitor). The main objective of this study was to determine the effect of AVG on the Bacillus-Rhizobium induced root hair proliferation in P. vulgaris. 6. 2. Materials and Methods 6. 2. 1. Bacterial strains and growth media The isolation and characterization of Bacillus megaterium S49 has been described previously (Chapter 2). The media and growth conditions of both Bacillus megaterium S49 and Rhizobium etli TAL 182 have been described in chapter two. Cells were harvested by centrifugation, washed twice with sterile 0.1M phosphate buffer (pH 6.8) and resuspended in the same buffer to a specified cell density before inoculation. 6. 2. 2. Plant material and growth conditions Seeds of P. vulgaris 'Contender' were surface sterilized in 30% hydrogen peroxide for 15 minutes, followed by seven rinses with sterile distilled water. The seeds were germinated aseptically on moist filter paper in petri plates. Stock solutions of aminoethoxy vinyl glycine (AVG) (2.5 mM), and 1-aminocyclopropane-1-carboxylic acid (ACC)(25 mM) (Sigma, St. Louis, MO), were prepared in distilled water, filter-sterilized 122 and stored at -20° C. Sterile two-day old pregerminated bean seedlings with uniform radicle length (approximately 1cm) were immersed for 15 minutes in either sterile 0.1M phosphate buffer (pH 6.8) (control) or buffer containing 106cfu ml"1 of bacterial inoculant. Seeds treated with AVG or ACC were incubated for 15 minutes in sterile phosphate buffer (pH 6.8) containing AVG (50 uM or 100 uM) or 100 uM ACC. Seedlings were cultured axenically in a 20 x 150 mm test tube containing 10 ml one-tenth strength nitrogen-free plant growth medium (Broughton and Dilworth 1971), in which the seed was supported above the medium by a 200 pi plastic pipette tip. Medium for treatments which included AVG or ACC were supplemented with 50 pM and/or 100 pM AVG or 100 pM ACC. Tubes were incubated at 25° C with a 16h photoperiod at 200 pE m2 s"1 illumination. A random sample of 20 lateral roots from approximately ten plants per treatment was used for scanning electron microscopy (SEM). The base and tip (approximately 0.8 cm each) of the selected lateral roots were excised and discarded. The remaining central portion of the root was used for SEM analysis. 6. 2. 3. Morphological characterization of root hair formation using scanning electron microscopy Root segments from each treatment were fixed overnight in 2.5% glutaraldehyde in 0.05M sodium cacodylate buffer (pH 7.2). Samples were washed three times (15 minutes each) with 0.05M sodium cacodylate buffer (pH 7.2) and post-fixed in 1% osmium tetroxide in 0.1 M sodium cacodylate buffer (pH 7.2) for two hours. The samples were rinsed once in de-ionized water, dehydrated through a 30% to 100% series of ethanol rinses, and critical point dried from 100% ethanol in a Balzers CPD 020 system using liquid carbon dioxide as the exchange medium. After mounting on double-sided sticky aluminum tabs, specimens were coated with gold-palladium (approximately 25 123 nm) in a Nano-tech II sputter coater, and viewed with a Cambridge 250T SEM. All samples were viewed by two independent skilled observers and typical observed response scored (e.g. no proliferation, slight proliferation, massive proliferation) and recorded photographically. 6. 3. Results 6. 3. 1. ACC induces root hair formation in P. vulgaris Enhancement of root hair formation in response to dual inoculation with Rhizobium and Bacillus are shown in Figure 6.1 A, B as previously reported in Chapter three. The concentrations of AVG and ACC used in this study were comparable to those reported by Tanimoto et al. (1995) in Arabidopsis thaliana. No aberrant root phenotypes were observed in response to treatment with ACC, AVG and/or either bacterial species. Root hair formation was enhanced by the addition of 100 uM ACC (Figure 6.1C), and single inoculation with either Rhizobium TAL 182 (Figure 6.1D) or Bacillus S49 (Figure 6.1 E). The addition of 100 uM ACC to dual inoculated plants showed no obvious synergistic enhancement of root hair formation (compare Figure 6.1F and 6.1B). 6. 3. 2. AVG inhibits root hair formation in P. vulgaris The presence of 50 uM AVG in the growth medium completely inhibited root hair formation in uninoculated control plants (Figure 6.2 A), and severely suppressed root hair formation in plants inoculated with either Rhizobium TAL 182 (Figure 6.2 B) or Bacillus S49 (Figure 6.2 C). The suppression of root hair formation by 50 uM AVG was largely reversed in dual inoculated plants (Figure 6.2 D). However, higher concentrations of AVG (100 uM) inhibited root hair formation even in dual inoculated 124 Figure 6.1. Root hair formation in P. vulgaris in response to inoculation with Rhizobium , Bacillus and treatment with 1-aminocyclopropane-1-carboxylic acid (ACC). (A) Uninoculated control (0.1M sterile phosphate buffer, pH 6.8) (B) Rhizobium etli TAL 182 and Bacillus megaterium S49 dual inoculated control (C) Uninoculated control with 100 uM ACC (D) Rhizobium TAL 182 and 100 uM ACC (E) Bacillus S49 and 100 uM ACC (F) Rhizobium TAL 182 and Bacillus S49 supplemented with 100 uM ACC 1 2 5 126 Figure 6. 2. Root hair formation in P. vulgaris in response to inoculation with Rhizobium, Bacillus and treatment with aminoethoxy vinyl glycine (AVG) (A) Uninoculated control with 50 uM AVG (B) Rhizobium etli TAL 182 and 50 uM AVG (C) Bacillus megaterium S49 and 50 uM AVG (D) Rhizobium TAL 182 and Bacillus S49 supplemented with 50 uM AVG (E) Rhizobium TAL 182 and Bacillus S49 supplemented with 100 uM AVG 127 4 O 0 H M 2 C I K V 128 plants (Figure 6.2 E); root hair morphogenesis in these plants was similar to that observed for B. megaterium S49 single inoculated plants supplemented with 50uM AVG (Figure 6.2 C). Plants treated with both AVG and ACC showed variable root hair formation responses; hence the results of this treatment are not presented here. Approximately 60% of treated plants showed little root hair formation, similar to uninoculated control with 50 uM AVG (Figure 6.2 A), while the remaining plants showed root hair formation similar to the uninoculated control with 100 uM ACC (Figure 6.1 C). 6. 4. Discussion It was shown in Chapter 3 that a synergistic interaction between Rhizobium etli TAL 182 and Bacillus megaterium S49 enhanced root hair formation in P. vulgaris and this enhancement was independent of the plant host. This study further demonstrates that enhancement of root hair formation likely overcomes the inhibitory effects of ethylene biosynthesis inhibitor aminoethoxy vinyl glycine (AVG), in the presence of the two interacting bacterial species. The inhibitory effects of AVG on root hair formation were clearly evident in uninoculated control plants and in those which received a single inoculation of either Rhizobium TAL 182 or Bacillus S49. The reversal of AVG inhibition by dual inoculation of Rhizobium TAL 182 and Bacillus S49 indicated the ability of P. vulgaris roots in the presence of the two bacteria, to overcome the inhibitory effects of AVG at 50 uM concentration. This observation was surprising since AVG is considered to strongly inhibit ethylene biosynthesis and hence root hair formation (Tanimoto et al. 1995). This study is consistent with the data presented in (Chapter 3 and Chapter 4) on root hair proliferation in P. vulgaris in response to coinoculation with the two bacteria. At higher concentrations of AVG (100 uM), root hair formation in P. vulgaris was 129 inhibited even in the presence of both bacterial species, which is consistent with the hypothesis that AVG is a negative regulator of root hair formation (Dolan et al. 1994; Tanimoto et al. 1995). Based on the root hair formation response in the presence of AVG, the effect of the addition of the ethylene precursor ACC to the plant growth medium on proliferation of root hairs in P. vulgaris was investigated. The presence of ACC in the growing medium of plants that received a single inoculation of either Rhizobium TAL 182 or Bacillus S49 resulted in a significantly enhanced root hair formation compared to the uninoculated bacterial control. This enhanced root hair formation may be due to increased ethylene biosynthesis in plant tissues, in the presence of exogenous ACC. Thus, the results of this study implicate ethylene as a positive regulator of root hair formation in P. vulgaris and are consistent with previously reported studies (Abeles et al. 1992; Dolan et al. 1994; Tanimoto et al. 1995). However, the presence of ACC in the growth medium of plants that received dual inoculation of Rhizobium TAL 182 and Bacillus S49 showed no obvious enhancement of root hair formation, compared to singly inoculated plants supplemented with ACC. Ethylene production in plants can also be induced by a variety of biological and physical stimuli (Adams and Yang 1981; Yang and Hoffmann 1984). The thick and short root (Tsr) phenotype in Vicia sativa is attributed to excess ethylene production, induced by Rhizobium nod factors (van Spronsen et al. 1995). It is likely that the presence of both Rhizobium and Bacillus on plant roots may induce an increase in ethylene biosynthesis in the root epidermal cells, leading to root hair differentiation. Ethylene and auxin are involved in root hair formation in Arabidopsis (Masucci and Schiefelbein 1994). B. megaterium S49 produces IAA in vitro (Chapter 2) and it is tempting to speculate that nod factors produced by Rhizobium TAL 182 may act 130 synergistically with IAA produced by Bacillus S49 to cause local changes in ethylene concentrations in root epidermal cells. The root hair proliferation observed in this study may also reflect systemic cell changes in response to the interacting bacteria, similar to the mechanism described for microtubule reorientation in cortical cells in Rhizobium-vetch interaction (Zaat et al. 1989; van Spronsen et al. 1995). Nevertheless, a role for ethylene in causing root hair proliferation in the presence of AVG/ACC and interacting Rhizobium and Bacillus cannot be ruled out, since it is not clear to what extent AVG could reduce or ACC induce endogenous ethylene production in P. vulgaris. The variable root hair formation observed on plants treated with both AVG and ACC may be due to differential uptake of these compounds by the plant roots. Ethylene of microbial origin could directly affect plant growth (Arshad and Frankenberger 1988). Under the experimental conditions of this study, both Rhizobium TAL 182 and Bacillus S49 did not produce detectable amounts of ethylene in vitro, even in the presence of its precursor L-methionine in the growing medium. Hence it is unlikely that ethylene production by these microorganisms could be a factor in the root hair proliferation responses observed. Nevertheless, the possibility of ethylene production by Rhizobium and/or Bacillus at the site of action (root surface) cannot be ruled out. The root hair proliferation observed in P. vulgaris in response to dual inoculation with Rhizobium and Bacillus may be analogous to the formation of shepherd's crooks as described for the Rhizobium-\egume interaction (Relic et al. 1993; van Rhijn and Vanderleyden 1995); both responses appear to require the physical contact between the plant and bacteria. Dual bacterial inoculation apparently disrupts the type of epidermal cell patterning described for root hair cells and non-root hair cells (Dolan et al. 1994). 131 The ACC/AVG response in combination with inoculation, suggest that the bacterial-plant interaction is mediated, at least in part, through ethylene. The root hair response reflects the unique potential for interaction between plants and their respective rhizosphere microbial populations in influencing the growth and physiology of the host. This is the first study to show unequivocal direct influence of interacting bacterial species on root hair formation in a leguminous plant, which has subsequently been shown to significantly influence nodulation of P. vulgaris by Rhizobium etli TAL 182. 132 Chapter 7 Summary / do not mind if you think slowly. But I do object when you publish more quickly than you think. Wolfgang Pauli (1900-1958) 133 The \egume-Rhizobium symbiosis is considered to be the most efficient system for biological nitrogen fixation. Roots of Phaseolus vulgaris (common bean) are nodulated by Rhizobium etli and Rhizobium tropici. In contrast to other legumes, the common bean is considered a poor nitrogen fixer, which may be a function of several genetic and environmental factors. Nodules formed in P. vulgaris are of the determinate type and plants have a bushy type habitat with a short growth cycle. Hence nitrogen fixation may not contribute as significantly to the nitrogen nutrition of the plant as in other nitrogen fixing legumes. The free-living, nitrogen-fixing bacteria of the genus Bacillus live in close association with plants, and may promote plant growth under appropriate conditions. The importance of Bacillus as plant-growth promoting rhizobacteria has received much attention in the last decade mainly from experiments under greenhouse and growth chamber conditions, and in some field inoculation experiments. Positive effects of Bacillus inoculation are mainly attributed to improved root development (i.e., increased root dry weight, root branching and root surface area), promotion of seedling emergence and general plant growth. Available evidence indicates that production of plant-growth promoting substances by Bacillus is at least partly responsible for these observed effects. The ability of Bacillus to enhance plant growth by suppression of deleterious rhizobacteria (DRB) and plant pathogenic fungi has been the focus of many researchers in the past few years. In this context, it is worth mentioning that suppression of deleterious microflora by Bacillus was achieved only in the presence of associated bacteria on plant roots. Thus, the growth promoting ability of Bacillus by suppressing deleterious rhizosphere microflora may be dependent upon its interaction with other root-associated microflora. 134 The effect of coinoculation of legumes with Rhizobium and Bacillus has not been extensively studied, unlike the Azospirillum-Rhizobium associations. Nevertheless, positive effects due to combined inoculation of Rhizobium and Bacillus in enhancing nodulation of legumes, mainly on soybeans, has been observed. Increases in dry-matter production and nitrogen content, increased number of nodules and, a general improvement in root development were attributed to the effects of coinoculation. However, none of the studies explain adequately the mechanism(s) of nodulation promotion due to coinoculation. A specific plant response to the presence of Bacillus alone or in combination with Rhizobium, has not yet been implicated, to explain enhanced nodulation. As in Azospirillum-Rhizobium associations, increases in root growth and enhanced nodulation in coinoculation experiments have been attributed to the effect of plant growth substances produced by Bacillus. Such evidence has to date, been circumstantial, rather than conclusive. In the present study, attempts were made to evaluate the effects of combined inoculation of Rhizobium etli TAL 182 and Bacillus on nodulation of P. vulgaris under controlled conditions, and to investigate possible mechanism(s) involved in this interaction. The results of the study are summarized as follows: (i) Some of the Bacillus isolates produced IAA in vitro, both in the presence and absence of L-tryptophan in the growing medium. The amount of IAA produced in a culture medium supplemented with 100 ug ml"1 of L-tryptophan was significantly higher than the amounts produced in unsupplemented medium. Bacillus isolates that enhanced nodulation of P. vulgaris by Rhizobium TAL 182 also produce IAA in vitro. A variable relationship was observed between in vitro IAA production by Bacillus isolates and their ability to promote nodulation on P. vulgaris by Rhizobium TAL 135 182. Enhanced nodulation responses likely require the presence of both bacterial species and may not be solely dependent on in vitro IAA production by Bacillus isolates. Spontaneous a-methyl tryptophan resistant mutants of B. megaterium strain S49 showed altered IAA production compared to the wild-type and excreted tryptophan into the growing medium. Both low and high IAA producing mutants decreased nodulation of P. vulgaris by Rhizobium TAL 182, suggesting that Bacillus isolates may have a direct effect on either the Rhizobium or the plant. Some of the Bacillus isolates also produced cytokinin-like compounds in vitro. A detailed study of cytokinin production by Bacillus isolates was precluded by inconsistency and lack of reproducibility in the HPLC analysis and hence was not pursued further. (ii) Combined inoculation of Rhizobium TAL 182 and B. megaterium S49 resulted in a synergistic enhancement of root hair initiation. This enhancement was not mimicked by either supplemental IAA or by coinoculation of Rhizobium TAL 182 with a low IAA producing B. megaterium mutant, S49-1D. The stimulatory effect of dual inoculation was a primary consequence of an interaction between Rhizobium TAL 182 and Bacillus S49, rather than a host directed event. The lack of host determination was illustrated by root hair proliferation induced on P. acutifolius and Arabidopsis thaliana in response to dual inoculation of Rhizobium TAL 182 and Bacillus S49. (iii) Simultaneous physical presence of viable Rhizobium and Bacillus was necessary to induce root hair proliferation. Physical separation of either bacterial species precluded root hair proliferation in P. vulgaris, which suggests that a low molecular weight diffusible substance is not involved in the interaction leading to root hair proliferation. P. vulgaris did not show altered root hair formation when inoculated with 136 non-viable cell preparations of both Rhizobium and Bacillus. The root hair proliferation responses observed in response to dual inoculation of Rhizobium with Bacillus isolates were not dependent on the ability of the Bacillus isolates to produce IAA in vitro. Fully functional Rhizobium and Bacillus may also be prerequisites for the synergistic interaction leading to root hair proliferation. Coinoculation of Bradyrhizobium spp. TAL 644 with Bacillus S49 did not enhance root hair formation in P. vulgaris, suggesting a degree of specificity for the microbial partners in the interaction. (iv) Presence of viable Bacillus significantly influenced Rhizobium TAL 182- P. vulgaris symbiosis by inducing formation of effective nodules at early stages of plant growth. Coinoculation of Bacillus S49 with Rhizobium TAL 182 partly offset the host-mediated suppression of nodulation, indicative of a plant response to the interacting bacterial species. Enhancement of nodulation was not observed if Rhizobium inoculation was delayed, even when Bacillus was present on the roots. Coinoculation of Rhizobium TAL 182 with non-viable cell preparations of B. megaten'um S49 also failed to promote nodulation of P. vulgaris by Rhizobium TAL 182. Heterologous nodulation of P. acutifolius by Rhizobium TAL 182 in the presence of Bacillus S49 implicates Bacillus directly in the nodulation potential of Rhizobium. Nodulation promotion appears to be Rhizobium specific, as coinoculation of Bacillus with Bradyrhizobium TAL 644 failed to enhance nodulation of P. acutifolius by its homologous microsymbiont, Bradyrhizobium TAL 644. (v) The presence of aminoethoxy vinyl glycine (AVG) in the growth medium resulted in a complete loss of root hair formation in P. vulgaris. Plants that received a single inoculation of either Rhizobium TAL 182 or Bacillus S49 in the presence of AVG 137 also did not show enhancement of root hair formation. The inhibitory effect of AVG on root hair formation was nullified by the dual presence of both Rhizobium TAL 182 and Bacillus S49 on P. vulgaris roots. The ACC/AVG response in combination with bacterial inoculation, suggest that the bacterial-plant interaction is mediated, at least in part, through ethylene and dual inoculation apparently disrupts the epidermal cell patterning described for root hair cells and non-root hair cells. The results of this study show that Bacillus enhances nodulation of P. vulgaris by Rhizobium TAL 182. This enhancement could be explained, at least in part, by the synergistic enhancement of root hair formation leading to increased infection sites for Rhizobium. This study provides clear evidence of a synergistic bacterial interaction which results in a significant change in plant root morphogenesis. The results reflect the potential for interspecific bacterial interactions in the plant rhizosphere, and this is the first study to demonstrate a plant response in root hair morphogenesis induced by the interaction of two bacterial species. This interaction, and its effect on root hair morphogenesis, has also been shown to have significant implications for the infection potential and nodulation of P. vulgaris by Rhizobium TAL 182 under gnotobiotic conditions. 138 7.1. Proposal of a working hypothesis for future research Much progress has been achieved in recent years in understanding the nature of nodulation, the regulation of its physiological processes and in identifying the chemical nature of Rhizobium signals that trigger nodulation. Little is understood of the mechanism(s) by which associated rhizobacteria promote nodulation of legumes by Rhizobium. Based on the results of this study, it is tempting to present an hypothetical model on which future research may be carried out to determine the basis of enhanced nodulation in response to coinoculation of Rhizobium and Bacillus. The proposals are based on the model presented in the diagram. The letters for each heading refers to the relevant portion of the schematic diagram. This model is purely intended to stimulate experimentation and, by any standards, is not comprehensive. A. Legume p\ant-Rhizobium interaction This interaction has been well characterized and hence will not be considered further here. B. Legume plant-Bac/7/us interaction The following may be appropriate for future studies: (I) To determine whether seed or root exudates of P. vulgaris contain a compound(s) that might be used exclusively by Bacillus for spermosphere and/or rhizosphere colonization. The endophytic habitat of some Bacillus spp. may also reflect their ability to use a particular substrate as a carbon source. (ii) To determine whether a plant-derived compound might be modified by Bacillus to enhance Rhizobium populations in the rhizosphere, or alter symbiotic activity. The influence of such a modified compound(s) on root growth might also be relevant. (iii) To determine whether Bacillus enhances or alters the secretion of nod gene 139 Legume host (P. vulgaris) Rhizobium ^ Partial offset of auto regulation V Root hair proliferation ^ Bacillus Heterologous nodulation Nodulation enhancement i Enhanced nitrogen-fixation Enhanced plant-growth Simplified scheme outlining events leading to nodulation enhancement and alteration of nodulation potential of Rhizobium in the presence of Bacillus. 140 inducing compounds from seeds/roots of P. vulgaris, that lead to altered Rhizobium activity. C. Rhizobium-Bacillus interaction The following investigations may provide evidence for the mechanism(s) involved in the synergistic interaction. (i) To determine whether Bacillus directly alters or enhances the expression of nod genes in Rhizobium. (ii) To determine whether a factor or compound produced by Bacillus is transferred to or is communicated to the plant by Rhizobium. (iii) To determine whether a Rhizobium gene product or signal is modified by Bacillus or vice-versa as a trigger to root hair proliferation. (iv) To examine the interaction between the nod factor produced by Rhizobium and IAA produced by the Bacillus on root hair proliferation. This interaction can further be characterized by use of Rhizobium mutants deficient in Nod factor production and Bacillus mutants deficient in IAA production. D. Rhizobium-Bacillus interaction and root hair proliferation (i) To determine whether the presence of interacting Rhizobium and Bacillus induces altered gene expression in root epidermal cells involved in root hair differentiation. Quantitative and qualitative variation of known root hair development (RHD) gene products may contribute to elucidating the mechanism(s) involved in root hair proliferation in response to coinoculation. E. Alteration of regulation of nodulation due to interacting Rhizobium and Bacillus (i) "Nod off compounds are primarily the products of plant secondary metabolism (Mellor and Collinge 1995). Analysis of the quantitative and qualitative variations of 141 "nod off' compounds in root tissues in response to dual inoculation may contribute to our understanding of the mechanism(s) influencing autoregulation in root nodulation. F. Heterologous nodulation of Rhizobium in the presence of Bacillus (i) To examine whether there is a variation of oligosaccharides (nod factor binding ligands) in root hairs induced by the presence of Bacillus. For example, oligosaccharide metabolism is involved in the production of specific nod factor binding ligands. Variation in carbohydrate profiles in root hairs in response to the presence of Bacillus may influence legume-microsymbiont recognition. Interference with the process (by the presence of Bacillus) may lead to heterologous nodulation. 142 Bibliography Abeles, F.B., Morgan, P.W., and Saltveit Jr, M.E. (1992). Ethylene in plant biology. Academic Press, San Diego. Adams, D.O., and Yang, S.F. (1979). Ethylene biosynthesis: identification of 1-aminocyclopropane-1-carboxylic acid as an intermediate in the conversion of methionine to ethylene. Proc. Natl. Acad. Sci. USA. 76: 170-174. Adams, D.O., and Yang, S.F. (1981). Ethylene the gaseous plant hormone: mechanism and regulation of biosynthesis. Trends. Biochem. Sci. 6: 648-652. Allen, O.N., and Allen, E.K. (1981). The leguminosae, a source book of characteristics, uses and nodulation. University of Wisconsin Press, Madison. Andersen, R.A., and Kemp, T.R. (1979). Reversed-phase high-performance liquid chromatography of several plant cell division factors (cytokinins) and their cis and trans isomers. J. Chromatogr. 172: 509-512. Appleby, C.A., and Bergersen, F.J. (1980). Preparation and experimental use of leghemoglobin. In Methods of evaluating biological nitrogen fixation. Edited by Bergersen, F.J. John Wiley and Sons Ltd, Chichester, pp 315-335. Arshad, M., and Frankenberger Jr, W.T. (1988). Influence of ethylene produced by soil microorganisms on etiolated pea seedlings. Appl. Environ. Microbiol. 54: 2728-2732. Atzorn, R.A., Crozier, A., Wheeler, C.T., and Sandberg, G. (1988). Production of gibberellins and indole-3-acetic acid by Rhizobium phaseoli in relation to nodulation of Phaseolus vulgaris roots. Planta. 175: 532-538. Ayling, S.M., and Pring, R.J. (1995). lndole-3-acetic acid induces rapid changes in the ultrastructure of tomato root hairs, (abstract No 360) Plant Physiol. 109: S-78. Badenoch-Jones, J., Rolfe, B.G., and Letham, D.S. (1983). Phytohormones, Rhizobium mutants and nodulation in legumes. Plant Physiol. 73: 347-352. Bakker, P.A.H.M., Raaymakers, J.M., and Schippers, B. (1993). Role of iron in the suppression of bacterial plant pathogens by fluorescent pseudomonads. In Iron chelation in plants and microorganisms. Edited by Barton, L.L., and Hemming, B.C. Academic Press, San Diego, pp 269-281. Bal, A.K., Shantaram, S., and Wong, P.P. (1982). Nodulation of pole bean (Phaseolus vulgaris L.) by Rhizobium species of two cross-inoculation groups. Appl. Environ. Microbiol. 44: 965-971. Barbieri, P., and Galli, E. (1993). Effect of wheat root development of inoculation with 143 an Azospirillum braziliense mutant with an altered indole-3-acetic acid production. Res. Microbiol. 144: 69-75. Barbieri, P., Zanelli, T., Galli, E., and Zanetti, G. (1986). Wheat inoculation with Azospirillum brazilians Sp6 and some mutants altered in nitrogen fixation and indole-3-acetic acid production. FEMS. Microbiol. Lett. 36: 87-90. Bauer, W.D. (1981). Infection of legumes by rhizobia. Ann. Rev. Plant Physiol. 32: 407-449. Bennett, M.J., Lightfoot, D.A., and Cullimore, J.V. (1989). cDNA sequence and differential expression of the gene encoding the glutamine synthetase polypeptide of Phaseolus vulgaris L. Plant. Moi. Biol. 12: 553-565. Beringer, J.E. (1974). R factor transfer in Rhizobium leguminosarum. J. Gen. Microbiol. 84: 188-198. Bhuvaneswari, T.V., Bhagwat A.A., and Bauer, W.D. (1981). Transient susceptibility of root cells in four common legumes to nodulation by rhizobia. Plant Physiol. 68: 1114-1149. Billington, D.C, Golding, B.T., and Primrose, S.B. (1979). Biosynthesis of ethylene from methionine. Biochem. J. 182: 827-836. Blakesley, D., Weston, G.D., and Hall, J.F. (1991). The role of endogenous auxin in root initiation. 1. Evidence from studies on auxin application, and analysis of endogenous levels. Plant Growth Regul. 10: 341-353. Blauenfeldt, J., Joshi, P.A., Gresshoff, P.M., and Caetano-Anolles, G. (1994). Nodulation of white clover (Trifolium repens) in the absence of Rhizobium. Protoplasma. 179: 106-110. Bohlool, B.B., and Schmidt, E.L. (1970). Immunofluorescent detection oi Rhizobium japonicum in soils. Soil. Sci. 110: 229-236. Bolton Jr, H., Elliott, L.F., Turco, R.F., and Kennedy, A.C. (1990). Rhizoplane colonization of pea seedlings by Rhizobium leguminosarum and a deleterious root colonizing Pseudomonas sp. and effects on plant growth. Plant Soil. 123: 121-124. Bradford, M.M. (1976). A rapid and sensitive method for the quantification of microgram quantities of protein using the principle of protein dye binding. Anal. Biochem. 72: 244-254. Brie, J.M., Richard, B.M., and Silverstone, S.E. (1991). Rapid in situ assay for indoleacetic acid production by bacteria immobilized on nitrocellulose membrane. Appl. Environ. Microbiol. 57: 331-340. 144 Broughton, W.J., and Dilworth, M.J. (1971). Control of leghaemoglobin synthesis in snake beans. Biochem. J. 125: 1075-1080. Brown, M.E. (1972). Plant growth substances produced by microorganisms of soil and rhizosphere. J. Appl. Bact. 35: 443-451. Cacciari, I., Lippi, D., Pietrosanti, T., and Pietrosanti, W. (1989). Phytohormone-like substances produced by single and mixed diazotrophic cultures of Azospirillum and Azotobacter. Plant Soil. 115: 655-657. Caetano-Anolles, G., and Bauer, W.D. (1988). Enhanced nodule initiation on alfalfa by wild-type Rhizobium meliloti co-inoculated with nod gene mutants and other bacteria. Planta. 174: 385-395. Caetano-Anolles, G., and Gresshoff, P.M. (1990). Early induction of feedback regulatory responses governing nodulation in soybean. Plant Science. 71: 69-81. Caetano-Anolles, G., and Gresshoff, P.M. (1991). Plant genetic control of nodulation. Ann. Rev. Microbiol. 45: 345-382. Caetano-Anolles, G., Joshi, P.A., and Gresshoff, P.M. (1990). Spontaneous nodules induce feedback suppression of nodulation in alfalfa. Planta. 183: 77-82. Callaham, D.A., and Torrey, J.G. (1981) The structural basis for infection of root hairs of Trifolium repens by Rhizobium. Can. J. Bot. 59: 1647-1664. Cardus, J.R. (1979). Selection for root hair length in white clover, Trifolium repens. Euphytica. 28: 489-494. Carroll, W.R. (1934). A study of Rhizobium species in relation to nodule formation on the roots of Florida legumes.l. Soil Sci. 37: 117-135. Chanway, CP. , Radley, R.A., and Holl, F.B. (1991). Inoculation of conifer seed with plant growth promoting Bacillus strains causes increased seedling emergence and biomass. Soil. Biol. Biochem. 23: 575-580. Clarkson, D.T. (1985). Factors affecting mineral nutrient acquisition by plants. Ann. Rev. Plant Physiol. 36: 77-115. Cormak, R.G.H. (1962). Development of root hairs in angiosperms. Bot. Rev. 28: 446 -464. Curl, E.A., and Truelove, B. (1986). The Rhizosphere, Advanced series in Agricultural Sciences. Vol 15. Springer-Verlag, New York. Dangar, T.K., and Basu, P.S. (1987). Studies on plant growth substances: IAA metabolism and nitrogenase activity in root nodules of Phaseolus aureus Roxb. mungo. 145 Biol. Plant. 29: 350-354. De Freitas, J. R., and Germida, J.J. (1990). A root tissue culture system to study wheat-rhizobacteria interactions. Appl. Microbiol. Biotechnol. 33: 589-595. De Freitas, J. R., and Germida, J.J. (1991). Pseudomonas cepacia and Pseudomonas putida as winter wheat inoculants for Rhizoctonia solani. Can. J. Microbiol. 37: 780-784. De Freitas, J. R., and Germida, J.J. (1992). Growth promotion of winter wheat by fluorescent pseudomonads under growth chamber conditions. Soil. Biol. Biochem. 24: 1127-1135. De Freitas, J. R., Gupta, V.V.S.R., and Germida, J.J. (1993). Influence of Pseudomonas syringae R25 and P. putida R105 on the growth and N 2 fixation (acetylene reduction activity) of pea (Pisum sativum L.) and field bean (Phaseolus vulgaris L). Biol. Fert. Soils. 16: 215-220. Dehio, C , and de Bruijn, F.J. (1992). The early nodulin gene SrEnod2 from Sesbania rostrata is inducible by cytokinin. Plant J. 2: 117-128. de Oliveira L.A., and Graham, P.H. (1990). Speed of nodulation and competitive ability among strains of Rhizobium leguminosarum bv. phaseoli. Arch. Microbiol. 153: 311-315. Diaz, C.L., Spaink, H.P., Wijffelman, C.A., and Kijne, J.W. (1995). Genomic requirements of Rhizobium for nodulation of white clover hairy roots transformed with the pea lectin gene. Moi. Plant. Microbe Interact. 8: 348-356. Dilworth, M.J. (1966). Acetylene reduction by nitrogen fixing preparations from Clostridium pasteurianum. Biochem. Biophys. Acta. 127: 285-294. Dolan, L., Duckett, CM., Grierson, C, Linstead, P., Schneider, K., Lawson, E., Dean, C , and Roberts, K. (1994). Clonal relationships and cell patterning in the root epidermis of Arabidopsis. Development. 120: 2465-2474. Dubeikovsky, A.N., Mordhukova, E.A., Kochetkov, VV., Polikarpova, F.Y., and Boronin, A.M. (1993). Growth promotion of black currant softwood cuttings by recombinant strain Pseudomonas fluorescens BSP 53a synthesizing an increased amount of indole-3-acetic acid. Soil. Biol. Biochem. 25: 1277-1281. Dowling, D.N., Stanley, J., Broughton, W.J. (1988). Competitive nodulation blocking (Cnb) is determined by nod-genes in Rhizobium leguminosarum. In Nitrogen fixation: hundred years after. Edited by Bothe, H., de Bruijn, F.J., and Newton. W.E. Gustav Fisher, Stuttgart, pp 443-448. Dullart, J. (1970). The auxin content of root nodules and roots of Alnus glutinosa (L.) Vill. J. Exp. Bot. 21: 975-984. 146 Ehmann, A. (1977). The Van Urk - Salkowski reagent-a senstive and specific chromogenic reagent for silica gel thin layer chromatographic detection and identification of indole derivatives. J. Chromatogr. 132: 267-276. Ernstsen, A., Sandberg, G., Crozier, A., and Wheeler, C.T. (1987). Endogenous indoles and the biosynthesis and metabolism of indole-3-acetic acid in cultures of Rhizobium phaseoli. Planta. 171: 422-428. Estelle, M.A., and Somerville, C. (1987). Auxin-resistant mutants of Arabidopsis thaliana with an altered morphology. Mol. Gen. Genet. 206: 200-206. Fahraeus, G., and Ljunggren, J. (1968). Pre-infection phases of the legume symbiosis. In The ecology of soil bacteria. Edited by Gray, T.R.G., and Parkinson, D. Liverpool University Press, UK. pp 928-936. Feldman, L.J. (1984). Regulation of root development. Ann. Rev. Plant Physiol. 35: 223-242. Fellay, R., Rochepeau, P., Relic, B., and Broughton, W.J. (1995). Signals to and emanating from Rhizobium largely control symbiotic specificity. In Pathogenesis and host specificity in plant diseases: histopathological, biochemical, genetic and molecular basis, Vol 1. Edited by Singh, U.S., and Singh, R.P. Pergamon/Elsevier Science Ltd, pp 199-220. Fisher, R.F., and Long, S.R. (1992). Rhizobium-p\ar\i signal exchange. Nature (London). 357: 655-660. Frankenberger Jr, W.T., and Arshad, M. (1995). Phytohormones in soils. Microbial production and function. Marcel Dekker Inc, New York. Garbaye, J. (1994). Helper bacteria: a new dimension to the mycorrhizal symbiosis. New Phytologist. 128: 197-210. George, M.L.C., and Robert, F.M. (1991). Autoregulatory response of Phaseolus vulgaris L. to symbiotic mutants of Rhizobium leguminosarum bv. phaseoli. Appl. Environ. Microbiol. 57: 2687-2692. George, M.L.C., Robert, F.M., and Bohlool, B.B. (1992). Nodulation suppression by Rhizobium leguminosarum bv. phaseoli in bean split-root systems. Symbiosis. 12: 95-105. Germida, J.J., and De Freitas, J.R. (1994). Growth promotion of cabbage, lettuce, and onion by fluorescent pseudomonads under growth chamber conditions. In Improving plant productivity with rhizosphere bacteria. Edited by Ryder, M.H., Stephens, P.M., and Bowen, G.D. Proceedings of the third international workshop on plant-growth promoting rhizobacteria. Adelaide, Australia, pp 37-29. 147 Glick, B.R. (1995). The enhancement of plant growth by free living bacteria. Can. J. Microbiol. 41: 109-117. Glick, B.R., Jacobson, C.B., Schwarze, M.M.K., and Pasternak, J.J. (1994a). Does the enzyme 1-aminocyclopropane-1-carboxylate deaminase play a role in plant growth-promotion by Pseudomonas putida GR 12-2? In Improving plant productivity with rhizosphere bacteria Edited by Ryder, M.H., Stephens, P.M., Bowen, G.D. Commonwealth Scientific and Industrial Research Organization, Adelaide, Australia, pp 150-152. Glick, B.R., Jacobson, C.B., Schwarze, M.M.K., and Pasternak, J.J. (1994b). 1-Aminocyclopropane-1-carboxylic acid deaminase mutants of the plant-growth promoting rhizobacteria Pseudomonas putida GR 12-2 do not stimulate canola root elongation. Can. J. Microbiol. 40: 911-915. Glick, B.R., Karaturovic, D.M., and Newell, P.C. (1995). A novel procedure for rapid isolation of plant growth promoting pseudomonads. Can. J. Microbiol. 41: 533-536. Goodlass, G., and Smith, K.A. (1979). Effects of ethylene on root elongation of pea (Pisum sativum L.) and white clover (Trifolium repens L). Plant Soil. 51: 387-395. Gordon, S.A., and Weber, R.P. (1951). Colorimetric estimation of indoleacetic acid. Plant Physiol. 26: 192-195. Grandmaison, J., and Ibrahim, R. (1994). Ultrastructural localization of a diprenylated isoflavone in Rhizobium lupini-Lupinus albus symbiotic association. J. Exp. Bot. 46. 231-237. Grimes, H.D., and Mount, M.S. (1984). Influence of Pseudomonas putida on nodulation of Phaseolus vulgaris. Soil. Biol. Biochem. 16: 27-30. Harari, A., Kigel, J., and Okon, Y. (1988). Involvement of IAA in the interaction between Azopsirillum brazilians and Panicum miliaceum roots. Plant Soil. 110: 275-282. Halverson, L.J., and Handelsman, J. (1991). Enhancement of soybean nodulation by Bacillus cereus UW85 in the field and in a growth chamber. Appl. Environ. Microbiol. 57: 2767-2770. 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 in Vicia sativa. Plant Physiol. 105: 787-797. Hirsch, A.M., Bhuvaneswari, T.V., Torrey, J.G., and Bisseling, T. (1989). Early nodulin genes are influenced in alfalfa outgrowths elicited by auxin transport inhibitors. Proc. Natl. Acad. Sci. USA. 86: 1244-1248. 148 Hirsch, A.M., and Fang, Y. (1994). Plant hormones and nodulation: what is the connection? Plant. Mol. Biol. 26: 5-9. Hoch, S.O., Roth, C.W., Crawford, I.P., and Nester, E.W. (1971). Control of tryptophan biosynthesis by the methyl tryptophan resistant gene in Bacillus subtilis. J. Bacteriol. 105: 38-45. Hochmuth, G.J., Gabelman, W.H., and Gerloff, G.C. (1985). A gene affecting tomato root morphology. Hort. Sci. 20: 1099-1011. Holl, F.B., Chanway, CP., Turkington, R., and Radley, R.A. (1988). Response of crested wheat grass {Agropyron cristatum L), Perennial ryegrass (Lolium perenne) and white clover (Trifolium repens L.) to inoculation with Bacillus polymyxa. Soil. Biol. Biochem. 20: 19-24. Horgan, R., and Kramers, M.R. (1979). High-performance liquid chromatography of cytokinins. J. Chromatogr. 173: 263-270. Hunter, W.J. (1987). Influence of 5-methyl tryptophan resistant Bradyrhizobium japonicum on soybean root nodule indole-3-acetic acid content. Appl. Environ. Microbiol. 53: 1051-1055. Hwang, S.F., Chang, K.F., Howard, R.J., Deneka, B.A., and Turnbull, G.D. (1996). Decrease in incidence of Pythium damping-off of field pea by seed treatment with Bacillus spp. and metalaxyl. Z. Pflkrankh. PflSchutz. 103: 31-41. Iruthayathas, E.E., Gunasekaran, S., and Vlassak, K. (1983). Effect of combined inoculation of Azospirillum and Rhizobium on nodulation and N 2 - fixation of winged bean and soybean. Sci. Hort. 20: 231-240. Jain, D.K., and Patriquin, D.G. (1984) Root hair deformation, bacterial attachment and plant growth in wheai-Azospirillum associations. Appl. Environ. Microbiol. 48: 1208-1213. Kamoun, S., Young, M., Glasiock, C.B., and Tyler, B.M. (1993). Extracellular protein elicitors from Phytophthora: host specificity and induction of resistance to bacterial and fungal pathogens. Mol. Plant. Microbe. Interact. 6: 15-25. Kaneshiro, T., and Kwolek, W.F. (1985). Stimulated nodulation of soybeans by Rhizobium japonicum mutant (B-14075) that catabolizes the conversion of tryptophan to indol3yl-acetic acid. Plant Sci. 42: 231-240. Kapulnik, Y., Gafny, R., and Okon, Y. (1985a). Effect of Azospirillum spp. inoculation on root development and NO3" uptake in wheat (Triticum aestivum cv.Miriam) in a hydroponic system. Can. J. Bot. 63: 627-631. Kapulnik, Y., Okon, Y., and Henis, Y. (1985b). Changes in root morphology of wheat 149 caused by Azospirillum inoculation. Can. J. Microbiol. 31: 881-887. Keel, C , Schnider, U., Maurhofer, M., Voisard, C , Laville, J., Burger, U., Wirthner, P., Haas, D., and Defago, G. (1992). Suppression of root diseases by Pseudomonas fluorescens CHAO: importance of the bacterial secondary metabolite 2,4-diacetylphloroglucinol. Moi. Plant. Microbe Interact. 5: 4-13. Kieber, J.J., and Ecker, J.R. (1993). Ethylene gas: it is not just for ripening any more! Trends. Genet. 9: 356-362. Kim, S.U., Chatfield, M.J., and Armstrong, D.J. (1985). Trace enrichment of cytokinins using neutral polystyrene resins. J. Chromatogr. 321: 175-184. King, J.J., Stimart, D.P., Fisher, R.H., and Bleecker, A.B. (1995). A mutation altering auxin homeostasis and plant morphology in Arabidopsis. Plant Cell. 7: 2023-2037. Kloepper, J.W., Hume, D.T., Scher, F.M., Singleton, C , Tipping, B., Laliberte, M., Frauley, K., Kutchaw, T., Simonson, C , Lifshitz, R., Zaleska, I., and Lee, L. (1988). Plant growth promoting bacteria on canola (rapeseed). Plant Disease. 72: 42-46. Kloepper, J.W., Lifshitz, R., and Zablotowicz, R.M. (1989). Free living bacterial inocula for enhancing crop productivity. Trends. Biotechnol. 7: 39-43. Kloepper, J.W., Mclnroy, J.A., and Bowen, K.L. (1992). Comparative identification by fatty acid analysis of soil, rhizosphere, and geocarposphere bacteria of peanut {Arachis hypogaea L.). Plant Soil. 139: 85-90. Knight, T.J., and Langston-Unkefer, P.J. (1988). Enhancement of symbiotic dinitrogen fixation by a toxin releasing plant pathogen. Science. 241: 951-954. Lambert, B., and Joos, H. (1989). Fundamental aspects of rhizobacterial plant growth promotion research. Trends Biotechnol. 7: 215-219. Larsen, N., Olsen, G.J., Maidak, B.L., McCaughey, M.J., Overbeek, R., Macke, T.J., Marsh, T.L., and Woese, C.R. (1993). The ribosomal data base project. Nucl. Acids. Res. 21: 3021-3023. Lee, H.J., and La Rue, Th.A. (1992). Exogenous ethylene inhibits nodulation of Pisum sativum L. cv Sparkle. Plant Physiol. 100: 1759-1763. Letham, D.S. (1971). Regulators of cell division in plant tissues. XII. A cytokinin bioassay using excised radish cotyledons. Physiol. Plant. 25: 397-401. Li D-M., and Alexander, M. (1990). Factors affecting co-inoculation with antibiotic producing bacteria to enhance rhizobial colonization and nodulation. Plant Soil. 129: 195-201. 150 Long, S.R., and Cooper, J. (1988). Overview of symbiosis. In Molecular genetics of plant microbe interactions. Edited by Palacois, R., and Verma, D.P.S. The American Phytopathological Society Press, St. Paul, pp 163-178. Loper, J.E., and Schroth, M.N. (1986). Influence of bacterial sources of indole-3-acetic acid on root elongation of sugar beet. Phytopathology. 76: 386-389. LUrssen, K., Naumann, K., and Schroder, R. (1979). 1-Aminocyclopropane-1-carboxylic acid- an intermediate of ethylene biosynthesis in higher plants. Z. Pflanzenphysiol. 92: 285-294. Lynch, J. (1988). Microbes are rooting for better crops. New Scientist. 28: 45-49. Mahaffee, W.F., Kloepper, J.W., van Vuurde, J.W.L., and van den Brink. M. (1994). Endophytic colonization of Phaseolus vulgaris by Pseudomonas fluorescens strain 89B-27 and Enterobacter asburiae strain JM 22. In Improving plant productivity with rhizosphere bacteria Edited by Ryder, M.H., Stephens, P.M., Bowen, G.D. Commonwealth Scientific and Industrial Research Organization, Adelaide, Australia, pp 180 Mansouri, S., and Bunch, A.W. (1989). Bacterial ethylene synthesis from 2-oxo-4-thiobutyric acid and from methionine. J. Gen. Microbiol. 135: 2819-2827. Massicotte, H.B., and Chanway, CP . (1994). Patterns of bacterial colonization of the ectomycorrhizae of conifers. In Proceedings of the 30th Canadian Botanical Association/Association Botanique Canadienne Meeting, June 26-30, Calgary, pp 34. Masucci, J.D., and Schiefelbein, J.W. (1994). The rhd6 mutation of Arabidopsis thaliana alters root-hair initiation through auxin-and ethylene-associated process. Plant Physiol. 106: 1335-1346. Mavingui, P., Laguerre, G., Berge, O., and Heulin, T. (1992). Genotypic and phenotypic variability of Bacillus polymyxa in soil and in the rhizosphere of wheat. Appl. Environ. Microbiol. 58: 1894-1903. Mellor, R.B., and Collinge, D.B. (1995). A simple model based on known plant defence reactions is sufficient to explain most aspects of nodulation. J. Exp. Bot. 46: 1-18. Mirza, J.I., Olsen, G.M., Iversen, T.H., and Maher, E.P. (1984). The growth and gravitropic responses of wild-type and auxin resistant mutants of Arabidopsis thaliana. Physiol. Plant. 60: 516-522. Morgenstern, E., and Okon, Y. (1987). The effect of Azospirillum brazilians and auxin on root morphology in seedlings of Sorghum bicolorx Sorghum sudanense. Arid. Soil. Res. Rehab. 1: 115-127. 151 Morris, R.O. (1986). Genes specifying auxin and cytokinin biosynthesis in phytopathogens. Ann. Rev. Plant Physiol. 87: 509-538. Murty, M.G., and Ladha, J.K. (1988). Influence of Azospirillum inoculation on the mineral uptake and growth of rice under hydroponic conditions. Plant Soil. 108: 281-285. Nap, J.N., and Bisseling, T. (1990). The roots of nodulins. Physiol Plant. 79: 407-414. Newcomb, W., and Wood, S.M. (1987). Morphogenesis and fine structure of Frankia (Actinomycetales). The microsymbiont of nitrogen-fixing actinorrhizal root nodules. Int. Rev. Cytol.109: 1-88. Nitsch, J.P., and Nitsch, C. (1956). Studies on the growth of coleoptile and first internode sections. A new sensitive straight-growth test for auxins. Plant Physiol. 31: 94-111. Nutman, P.S. (1956). The influence of the legume in root-nodule symbiosis. A comparative study of host determinants and function. Biol. Rev. Cambridge. Philos. 31: 109-151. Okon, Y., Fallik, E., Sarig, S., Yahalom, E., and Tal S. (1988). Plant growth promoting effects of Azospirillum. In Nitrogen-fixation. One hundred years after. Edited by Bothe E., de Bruijn, F.J., and Newton, W.E. Gustav Fisher, Stuttgart, pp 741-746. O'Neill, G.A., Chanway, CP., Axelrood, P.E., Radley, R.A., and Holl, F.B. (1992). An assessment of spruce growth response specificity after inoculation with coexistent rhizosphere bacteria. Can. J. Bot. 70: 2347-2353. Pacovsky, R.S. (1990). Development and growth effects in the Sorghum-Azosp/r///urn association. J. Appl. Bact. 68: 555-563. Park, C.S., Paulitz, T.C, and Baker, R. (1988). Biocontrol of fusarium wilt of cucumber resulting from interactions between Pseudomonas putida and non-pathogenic isolates of Fusarium oxysporum. Phytopathology. 78: 190-194. Patriquin, D.G., Dobereiner, J., and Jain, D.K. (1983). Sites and processes of association between diazotrophs and grasses. Can. J. Microbiol. 29: 900-915. Peters, N.K., and Long, S.R. (1988). Alfalfa root nodules and compounds which promote or inhibit induction of R. meliloti nodulation genes. Plant Physiol. 88: 99-112. Petersen, D.J., Holl, F.B., and Chanway, CP. (1995). Use of species- and strain-specific PCR primers for identification of conifer-root associated Bacillus spp. FEMS Microbiol. Lett. 133: 71-76. Picton, J.M., and Steer, M.W. (1983). Evidence for the role of Ca 2 + ions in tip extension 152 in pollen tubes. Protoplasma. 115: 11-17. Pierce, M., and Bauer, W.D. (1983). A rapid autoregulatory response governing nodulation in soybean. Plant Physiol. 73: 286-290. Plazinski, J., and Rolfe, B.G. (1985). Influence of Azospirillum strains on nodulation of clovers by Rhizobium strains. Appl. Environ. Microbiol. 49: 984-989. Priest, F.G., and Grigorova, R. (1990). Methods for studying the ecology of endospore forming bacteria. Meth. Microbiol. 22: 565-591. Primrose, S.B. (1979). A review, ethylene and agriculture: the role of microbes. J. Appl. Bacteriol. 46: 1-25. Puppo, A., and Rigaud, J . (1978). Cytokinins and morphological aspects of French-bean roots in the presence of Rhizobium. Physiol Plant. 42: 202-206. Relic, B., Perret, X., Estrada-Garcia, M.T., Kopcinska, J., Golinowski, W., Krishnan, H.B., Pueppke, S.G., and Broughton, W.J. (1994). Nod factors of Rhizobium are a key to the legume door. Moi. Microbiol. 13: 171-178. Relic, B., Talmont, F., Kopcinska, J., Golinowski, W., Prome, J-C, and Broughton, W.J. (1993). Biological activity of Rhizobium sp. NGR 234 nod-factors on Macroptilium atropurpureum. Moi. Plant. Microbe Interact. 6: 764-774. Rennie, R.J. (1981). A single medium for the isolation of acetylene - reducing (dinitrogen - fixing) bacteria from soils. Can. J. Microbiol. 27: 8-14. Ridge, R.W. (1995). Recent developments in the cell and molecular biology of root hairs. J. Plant. Res. 108: 399-405. Rolfe, B.G., and Gresshoff, P.M. (1980). Rhizobium trifolii mutant interactions during the establishment of nodulation in white clover. Aust. J. Biol. Sci. 33: 491-504. Sarig, S., Kapulnik, Y., and Okon, Y. (1986). Effect of Azospirillum inoculation on nitrogen fixation and growth of several winter legumes. Plant Soil. 90: 335-342. Schiefelbein, J.W., and Somerville, C. (1990). Genetic control of root hair development in Arabidopsis thaliana. Plant Cell. 2: 235-243. Schippers, B., Scheffer, R.J., Lugtenberg, B.J.J., and Weisbeek, P.J. (1995). Biocoating of seeds with plant growth-promoting rhizobacteria to improve plant establishment. Outlook on Agric. 24: 179-185. Schmidt, W., Martin, P., Omay, S.H., and Bangerth, F. (1988). Influence of Azospirillum on nodulation of legumes. In Azospirillum IV. genetics, physiology and 153 ecology. Edited by Klingmuller, G.K. Springer-Verlag, Heidelberg, pp 92-100. Schnall, J., and Quatrano, R. (1992). Abscisic acid elicits the water stress response in root hairs of Arabidopsis thaliana. Plant Physiol. 100: 216-218. Schroth, M.N., and Ole Becker, J. (1990). Concepts of ecological and physiological activities of rhizobacteria related to biological control and plant growth promotion. In Biological control of soil-borne plant pathogens. Edited by Hornby, D. CAB International, pp 389-414. Scott, T.K. (1972). Auxin and roots. Ann. Rev. Plant Physiol. 23: 235-258. Segovia, L., Young, P.W., and Martinez-Romero, E. (1993). Reclassification of American Rhizobium leguminosarum biovar phaseoli type 1 strains as Rhizobium etli sp. nov. Int. J. Syst. Bact. 43: 374-377. Shantaram, S. (1986). Anomalous nodulation of soybean by a promiscuous strain of Rhizobium phaseoli. Cytobios. 46: 53-64. Shantaram, S., and Wong, P.P. (1982). Recognition of leguminous hosts by a promiscuous Rhizobium strain. Appl. Environ. Microbiol. 43: 677-685. Shishido, M., Massicotte, H.B., and Chanway, CP. (1996). Effect of plant growth promoting Bacillus strains on pine and spruce seedling growth and mycorrhizal infection. Annals. Bot. in press. Smidt, M., and Kosuge, T. (1978). The role of indole-3-acetic acid accumulation by alpha methyl tryptophan resistant mutants of Pseudomonas savastanoi in gall formation on oleanders. Physiol. Plant Path, 13: 203-214. Somasegaran, P., and Bohlool, B.B. (1990). Single-strain versus multistrain inoculation: effect of soil mineral N availability on rhizobial strain effectiveness and competition for nodulation on chick-pea, soybean and dry bean. Appl. Environ. Microbiol. 56: 3298-3303. Somasegaran, P., Hoben, H.J., and Lewinson, L. (1991). Symbiotic interactions of Phaseolus acutifolius and P. acutifolius x P. vulgaris hybrid progeny in symbiosis with Bradyrhizobium spp and Rhizobium leguminosarum bv. phaseoli. Can. J. Microbiol. 37: 497-503. Spizizien, J. (1958). Transformation of biochemically deficient strains of Bacillus subtilis by deoxyribonucleate. Proc. Natl. Acad. Sci. USA. 44: 1072-1078. Sturtevant, D.B., and Taller, B.J. (1989). Cytokinin production by Bradyrhizobium japonicum. Plant Physiol. 89: 1247-1252. 154 Tanimoto, M., Roberts, K., and Dolan, L. (1995). Ethylene is a positive regulator of root hair development in Arabidopsis thaliana. Plant J. 8: 943-948. Thimann, K.V. (1972). The natural plant hormones in plant physiology. In The Hormones. Edited by Steward, F.C. Academic Press, New York, pp 3-145. Thomashow, L.S., and Weiler, D.M. (1995). Current topics in the use of introduced bacteria for biological control: mechanisms and antifungal metabolites. In Plant-Microbe Interactions, Vol 1. Edited by Stacey, G., and Keen, N. Chapman and Hall, New York, pp 187-225. Tien, T.M., Gaskins, M.H., and Hubbell, D.H. (1979). Plant growth substances produced by Azospirillum brazilians and their effect on the growth of pearl millet {Pennisetum americanum L). Appl. Environ. Microbiol. 37: 1016-1024. Tretyn, A., Wagner, G., and Felle, H. (1991). Signal transduction in Sinapis alba root hairs; auxins as external messengers. J. Plant Physiol. 139: 187-193. Tu, J.C. (1981). Effect of salinity on Rhizobium-rooi hair interaction, nodulation and growth of soybean. Can. J. Plant. Sci. 61: 231-239. Turner, J.T., and Backman, P.A. (1991). Factors relating to peanut yield increases following Bacillus subtilis seed treatment. Plant Disease. 75: 347-353. Umali-Garcia, M., Hubbell, D.H., Gaskins, M.H., and Dazzo, F.B. (1980). Association of Azospirillum with grass roots. Appl. Environ. Microbiol. 39: 219-226. van Brussel, A.A.N., Tak, T., Wijfjes, A.H.M., Spaink, H.P., Smit, G., Diaz, C.L., and Kijne, J.W. (1993). Jamming of nodulation of Vicia sativa ssp. nigra by nod factors of Rhizobium leguminosarum bv. viciae. In New Horizons in Nitrogen Fixation. Edited by, Palacios, R., Mora, J., and Newton, W.E. Kluwier Academic Publisher, Dordrecht, pp 261. van Brussel, A.A.N., Zaat, S.A.J., Canter Cramers, H.C.J., Wijffelman, C.A., Pees, E., Tak, T., and Lugtenberg, B.J.J. (1986). Role of plant root exudate and sym plasmid-localized nodulation genes in the synthesis by Rhizobium leguminosarum of Tsr factor, which causes thick and short roots on common vetch. J. Bacteriol. 165: 517-522. van Rhijn, P., and Vanderleyden, J. (1995). The Rhizobium-P\ar\\ symbiosis. Microbiol. Rev. 59: 124-142. van Spronsen, P.C, van Brussel, A.A.N., and Kijne, J.W. (1995). Nod factors produced by Rhizobium leguminosarum biovar viciae induce ethylene related changes in root cortical cells of Vicia sativa spp. nigra. Eur. J. Cell. Biol. 68: 463-469. Volpin, H., and Kapulnik, Y. (1994). Interaction of Azospirillum with beneficial soil 155 microorganisms. In Azospirillum/plant interactions. Edited by Okon, Y. CRC Press, Boca Raton, pp 111-118. Wang, T.E., Wood, E.A., and Brewin, N.J. (1982). Growth regulators, Rhizobium and nodulation in peas; indole-3-acetic acid from the culture medium of nodulating and non-nodulating strains of Rhizobium leguminosarum. Planta. 155: 345-349. Weisenseel, M.H., Nuccitelli, R., Jaffe, L.F. (1975). Large electrical currents traverse growing pollen tubes. J. Cell. Biol. 66: 556-567. Wightman, F., Schneider, E.A., and Thimann, K.V. (1980). Hormonal factors controlling the initiation and development of lateral roots. II. Effects of exogenous growth factors on lateral root formation in pea roots. Physiol. Plant. 49: 304-314. Wood, S., and Newcomb, W. (1989). Nodule morphogenesis: the early infection of alfalfa Medicago sativa root hairs by Rhizobium meliloti. Can. J. Bot. 67: 3108-3122. Xie, H., Pasternak, J.J., and Glick, B.R. (1996). Isolation and characterization of mutants of the plant growth -promoting rhizobacterium Pseudomonas putida GR12-2 that overproduce indoleacetic acid. Curr. Microbiol. 32: 67-71. Yahalom, E., Dovrat, A., Okon, Y., and Czosnek, H. (1991). Effect of inoculation with Azospirillum brazilians strain Cd and Rhizobium on the root morphology of burr medic (Medicago polymorpha L). Israel. J. Bot. 40: 155-164. Yahalom, E., Okon, Y., and Dovrat, A. (1990). Possible mode of action of Azospirillum brazilians strain Cd on the root morphology and nodule formation in burr medic (Medicago polymorpha). Can. J. Microbiol. 36: 10-14. Yang, S.F., and Hoffmann, N.E. (1984). Ethylene biosynthesis and its regulation in higher plants. Ann. Rev. Plant Physiol. 35: 155-189. Yao. P.Y., and Vincent, J.M. (1976). Factors responsible for the curling and branching of clover root hairs by Rhizobium. Plant Soil. 45: 1-16. Young, R.S., Pharis, R.P., Reid, D., Reddy, M.S., Lifshitz, R., and Brown, G. (1991). PGPR: Is there a relationship between plant growth regulators and the stimulation of plant growth or biological activity ? In The second international workshop on plant growth - promoting rhizobacteria. Edited by Keel, C , Koller, B., and De'faso. G. Bull. Srop. Interiaken, Switzerland, October 14-October 19. pp 102-103. Zaat, S.A.J., van Brussel, A.AN., Lugtenberg, B.J.J., and Kijne, J.W. (1989). The ethylene inhibitor aminoethoxyvinyl glycine restores normal nodulation by Rhizobium leguminosarum biovar. viciae on Vicia sativa subsp. nigra by suppressing the 'thick and short roots' phenotype. Planta. 177: 141-150. 156 Appendix 157 Appendix I. Preliminary results on the detection of cytokinins in the cell free supernatant of some Bacillus isolates Strain Cytokinin production t-Zeatina t-Zeatin riboside" 2iPa B. megaterium S49 + + + B. megaterium A15 + - + B. megaterium B17 - nd + B. polymyxa L6 + + + B. polymyxa PW2 + + + B. brevis S72 + + + B. mycoides T10 - - + B. pumilus (Bet 10a) + nd + a As detected by TLC, Sephadex LH-20 chromatography, HPLC and radish cotyledon expansion bioassay. b As detected by Phytodetek® immunoassay. nd= not determined +/- signs indicate the presence or absence of the compound. Trace enrichment of cytokinin-like compounds in the cell free supernatant of Bacillus isolates was accomplished using Amberlite XAD-2 chromatography, as described by (Kim et al. 1985). Separation of cytokinin-like-compounds after trace enrichment was accomplished using Sephadex LH-20 chromatography (Sturtevant and Taller 1989). Fractions obtained after separation by Sephadex chromatography were analysed by HPLC, to identify trans zeatin and isopentyl adenine (2iP), as desribed by either Horgan and Kramers (1979) or Andersen and Kemp (1979). Sephadex LH-20 fractions were also used for radish cotyledon expansion bioassay (Letham 1971), to ascertain for cytokinin-like activity. Identification of trans zeatin riboside was accomplished using the Phytodetek ® immunoassay after reverse-phase chromatography of cell free supernatant using C18 Sep-pak column (Sturtevant and Taller 1989). 158 Appendix II. Effect of dual inoculation of Rhizobium etli (TAL 182) and Bacillus polymyxa (L6) on (A) nodule number, (B) nodule fresh weight, and (c) nitrogenase activity of P. vulgaris 'Contender' grown under gnotobiotic conditions in the growth chamber. Control plants that received sterile 0.1M phosphate buffer did not form any nodules. Each value represents the means of six replicates. Means of values for each parameter tested were significantly different at p < 0.05, as determined by Duncan's multiple range test. Bars represent the standard error of the mean. 160 Appendix III. Mean number of lateral roots on Phaseolus vulgaris1 three and five days after the germinants were inoculated with Rhizobium etli TAL 182 and Bacillus polymyxa L6 Treatment Number of lateral roots Day 3 Day 5 Uninoculated control* 4.75 18.25 Rhizobium TAL 182 2.50 25.0 Rhizobium TAL 182 and 7.75** 27.5 Bacillus L6 f Plants were grown in modified Leonard jar assemblies as described in Chapter 2. * Sterile 0.1M phosphate buffer (pH 6.8) * * Significantly different from Rhizobium treatment, p < 0.001 (t-test). 161 Appendix IV. Calculation of percent suppression in split-root experiments (adapted from George et al. 1992). One side of the split-root system (primary side or E side) was inoculated at time zero, and the second side (D side) with increasing delay. At time zero, the uninoculated control consisted of early inoculation on both sides of the split-root with sterile 0.1M phosphate buffer (pH 6.8). For each delay period, the control consisted of a delayed inoculation only on the secondary side of the split-root (O/D control). In this experiment R. etli TAL 182 was the primary inoculant (E) and R. etli TAL 182 alone or in combination with B. megaterium S49 was the secondary inoculant (D). Inhibition of nodulation on the secondary side of the split-root was expressed as percent suppression, using the formula: % suppression = 100 - % nodulation The percent nodulation was derived by expressing the extent of secondary nodulation (D side of E/D treatments) as percentage of the nodulating potential of that side of the split-root at the time of the delayed inoculation (D side of O/D control) at each delay period. By quantifying suppression this way, the analysis was standardized with the comparison of nodules at the same developmental stage. 162 Appendix V. Effect of timing of inoculation of R. etli (TAL 182) and B. polymyxa (L6) on (A) nodule number, (B) nodule fresh weight, and (c) nitrogenase activity of P. vulgaris 'Contender' grown under gnotobiotic conditions in the growth chamber. Control plants that received sterile 0.1M phosphate buffer did not form any nodules. Each value represents the means of six replicates. Means of values for each parameter tested were significantly different at p < 0.05, as determined by Duncan's multiple range test. Bars represent the standard error of the mean. 0 4 7 10 Delay in Rhizobium inoculation (days) 

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.831.1-0087865/manifest

Comment

Related Items