Symbiotic specificity and nodulation in the southern African legume clade Lotononis s. l. and description of novel rhizobial species within the

Alphaproteobacterial genus Microvirga

by

Julie Kaye Ardley

This thesis is presented for the degree of

Doctor of Philosophy

of

Murdoch University

2011

DECLARATION

I declare that this thesis is my own account of my research and contains as its

main content work which has not previously been submitted for a degree at any

tertiary education institution.

Julie Kaye Ardley

ii

TABLE OF CONTENTS

Publications arising from this thesis………………………………………………….ix

Acknowledgements……………………………………………………………………………x

Abstract…………………………………………………………………………………………..xii

Chapter 1: Introduction and literature review……………………………………1 1.1 The legume-rhizobia symbiosis….……………….…….……………………………………..………..3 1.2 The legume-rhizobia symbiosis in Australian agriculture.………………………….……….4 1.2.1 Background…………………………………………………………………………………….…….…..4 1.3 Legumes.……….……………………………………………………………………………………..……….….6 1.3.1 Evolution of legumes…………………………………………….……….…………….…….…….6 1.3.2 Legume taxonomy and phylogeny………………………………………………….…………7 1.3.2.1 Phylogeny of the Papilionoideae…………………………….………….…………..….8 1.4 Rhizobia…..…………………………………………………………………….………………………………..11 1.4.1 History of rhizobial classification……………………………………………………...……..11 1.5 Bacterial systematics….……………………….…………….………………………………….….……..13 1.5.1 Genotypic classification methods………………………………………………..…….……15 1.5.2 Phylogenetic classification methods…………………..…………………..………….…..15 1.5.3 Phenotypic classification methods………………………………………..…..…………...18 1.5.4 The effect of lateral gene transfer on bacterial phylogeny…………………...…19 1.6 General recommendations for classifying and describing bacteria……….………….21 1.7 Rhizobial symbiotic genes………………………………………………………..………………………23 1.7.1 The role of nodulation genes and Nod factors…………………………………………24 1.8 Rhizobial infection and nodule formation in legumes………………………………………25 1.8.1 Infection pathways in legumes……………………………..…………………………….…..26

iii 1.8.2 Nodule types………………………………………………………..……………………….………..29 1.8.2.1 Formation and structure of nodules infected via infection threads………………………………………………………………………………………….…….……….30 1.8.2.2 Formation and structure of intercellularly infected nodules …………….31 1.9 Specificity and effectiveness in the legume-rhizobia symbiosis………………………..32 1.10 The Lotononis s. l. clade………………………………………………………………………………...34 1.10.1 The genus Listia…………………………………………………………………………………....36 1.10.2 The genus Leobordea…………………………………………………………………….………37 1.10.3 The genus Lotononis s. str……………………………………………………………………..38 1.11. The Lotononis s. l. -rhizobia symbiosis…………………………………………………………..39 1.12 Summary of current knowledge of the symbiotic relationships between Lotononis s. l. and associated rhizobia……………………………………………………………………44 1.13 Aims of this thesis…………………………………………………………………………………….……46

Chapter 2: Determining symbiotic specificity within Lotononis s.l. ….47 2.1 Introduction…………………………………………………………………………………………………….49 2.1.1 Background……………………………………………………………………………………………..49 2.1.2 Experimental approach……………………………………………………………………………50 2.2 Materials and methods…………………………………………………………………………………...53 2.2.1 Rhizobial strains………………………………………………………………………………………53 2.2.2 Host plants………………………………………………………………………………………………53 2.2.3 Glasshouse experimental design……………………………………………………………..55 2.2.4 General glasshouse procedures……………………………………………………………….58 2.2.4.1 Preparation of plant material……………………………………………………………59 2.2.4.2 Preparation of inoculum……………………………………………………………………60 2.2.4.3 Harvesting…………………………………………………………………………………………61 2.2.5 Statistics………………………………………………………………………………………………….61 2.2.6 Molecular fingerprinting………………………………………………………………………….61 2.2.7 Amplification and sequencing of 16S rRNA genes……………………………………63 2.2.8 Amplification and sequencing of nodA genes………………………………………….66 2.3 Results……………………………………………………………………………………………………………..68 iv 2.3.1 Selection of inoculant strains………………………………………………………………..…68 2.3.1.1 Selection of a representative Listia angolensis strain………………...……..68 2.3.1.2 Authentication of non-pigmented Lotononis s. l. strains……………………69 2.3.2 Symbiotic interactions of Lotononis s. l. species with phylogenetically diverse Lotononis s. l. rhizobia………………………………………………………………………….71 2.3.3 Symbiotic specificity within Listia species………………………………………………..75 2.3.4 Nodule structure in Lotononis s. l. species……………………………………………….75 2.3.5 Amplification and sequencing of 16S rRNA genes……………………………………77 2.3.6 Amplification and sequencing of nodA…………………………………………………….84 2.3.6.1 nodA sequences of WSM2598 and WSM2667…………………………………..84 2.3.6.2 nodA sequences of the remaining Lotononis s. l. rhizobia…………………85 2.4 Discussion………………………………………………………………………………………………………..90 2.4.1 Nodule structure in Lotononis s. l. species……………………………………………….90 2.4.2. Diversity of Lotononis s. l. rhizobia………………………………………………………….90 2.4.2 1 The Methylobacterium lineage………………………………………………………….91 2.4.2.2 The Microvirga lineage……………………………………………………………………..93 2.4.2.3 The Bradyrhizobium, Ensifer and Mesorhizobium lineages…………….….94 2.4.3 Symbiotic relationships within Lotononis s. l. hosts………………………………...96 2.4.4 Phylogeny of nodA genes…………………………………………………………………………98 2.4.5 Environmental factors that may be effectors of diversification in rhizobia associated with Lotononis s. l. species……………………………………………………………103 2.4.6 Putative evolution of symbiotic patterns within Lotononis s. l. …………….107

Chapter 3 Nodule structure in Lotononis s. l. species and infection and nodule formation in Listia angolensis and Listia bainesii…………….….111 3.1 Introduction…………………………………………………………………………………………………..113 3.1.1 Modes of infection and nodulation in legumes……………………………………..113 3.2 Materials and methods………………………………………………………………………………….115 3.2.1 Legume hosts, bacterial inoculant strains and growth conditions………….115 3.2.2 Harvesting……………………………………………………………………………………………..117 3.2.3 Nodule initial staining and sectioning, nodule sectioning and light microscopy…………………………………………………………………………………………………....118

v 3.3 Results……………………………………………………………………………………………………………119 3.3.1. Localisation of infection site…………………………………………………………………119 3.3.2 Infection and nodule organogenesis in Listia angolensis and Listia bainesii…………………………………………………………………………………………………………..121 3.3.3 Localisation of the infection site in Listia angolensis and Listia bainesii plants inoculated at 85 days old…………………………………………………………………….124 3.3.4 Internal structure of Lotononis s. l. nodules…………………………………………..125 3.4 Discussion………………………………………………………………………………………………………128

Chapter 4 Characterisation and description of novel Listia angolensis and Lupinus texensis rhizobia……………………………………..…………………133 4.1 Introduction…………………………………………………………………………………………………..135 4.2 Materials and methods………………………………………………………………………………….137 4.2.1 Bacterial strains and culture conditions…………………………………………………137 4.2.2 DNA amplification and sequencing………………………………………………………..137 4.2.2.1 Amplification and sequencing of the 16S rRNA gene………………………137 4.2.2.2 Amplification and sequencing of the nodA gene…………………………….139 4.2.3 Phenotypic characterisation………………………………………………………………….139 4.2.3.1 Morphology…………………………………………………………………………………….139 4.2.3.2 Extraction of pigments……………………………………………………………………140 4.2.3.3 Temperature range…………………………………………………………………………141 4.2.3.4 Optimal growth temperature………………………………………………………….141 4.2.3.5 Growth curves and determination of mean generation time………….141 4.2.3.6 Determination of sodium chloride and pH tolerance………………………141 4.2.3.7 Anaerobic growth……………………………………………………………………………143 4.2.3.8 Antibiotic sensitivity……………………………………………………………………….143 4.2.3.9 Substrate utilisation………………………………………………………………………..144 4.2.3.9.1 Utilisation of sole carbon sources in Biolog GN2 plates…………….144 4.2.3.9.2 Utilisation of substrates in Biolog Phenotype Microarray plates……………………………………………………………………………………………………..144 4.2.3.9.3 Growth on sole carbon substrates……………………………………………145 4.2.3.10 Biochemical characterisation………………………………………………………..147 vi 4.2.4 Host range…………………………………………………………………………………………….148 4.3 Results……………………………………………………………………………………………………………149 4.3.1 Amplification and sequencing of the 16S rRNA gene……………………………..149 4.3.2 Amplification and sequencing of nodA…………………………………………………..151 4.3.3 Phenotypic characterisation………………………………………………………………….153 4.3.3.1 Morphology…………………………………………………………………………………….153 4.3.3.2 Characterisation of pigments………………………………………………………….155 4.3.3.3 Growth characteristics……………………………………………………………………156 4.3.3.4 Antibiotic sensitivity……………………………………………………………………….156 4.3.3.5 Substrate utilisation………………………………………………………………………..158 4.3.3.5.1 Utilisation of sole carbon sources in Biolog GN2 plates…………….158 4.3.3.5.2 Utilisation of substrates in Biolog Phenotype Microarray plates………………………………………………………………………………………………………162 4.3.3.5.3 Growth in sole carbon substrate broths……………………………………168 4.3.3.6 Biochemical characterisation………………………………………………………….169 4.3.4 Host range…………………………………………………………………………………………….171 4.4 Discussion………………………………………………………………………………………………………173 4.4.1 Genotypic characterisation of novel Microvirga species………………………..173 4.4.2 G+C content, DNA:DNA hybridisation and cellular fatty acid analysis…….176 4.4.3 Phenotypic characterisation of novel Microvirga species………………………176 4.4.3.1 Substrate utilisation………………………………………………………………………..181 4.4.3.1.1 Critique of the use of Biolog plates in determining substrate utilisation………………………………………………………………………………………………..181 4.4.3.1.2 Substrate utilisation in Listia angolensis and Lupinus texensis strains 4.4.4 Biochemical characteristics of Listia angolensis and Lupinus texensis strains…………………………………………………………………………………………………………….187 4.4.5 Host range………………………………………………………………………………………….…187 4.5 Summary, emended description of Microvirga and description of Microvirga lotononidis sp. nov., Microvirga zambiensis sp. nov. and Microvirga lupini sp. nov. ……………………………………………………………………………………………………………………………188

vii 4.5.1 Emended description of Microvirga (Kanso & Patel, 2003 emend. Zhang et al, 2009, emend. Weon et al, 2010)……………………………………………………………….189 4.5.2 Description of Microvirga lotononidis sp. nov. ………………………………………189 4.5.3 Description of Microvirga zambiensis sp. nov. ………………………………………190 4.5.4 Description of Microvirga lupini sp. nov. ………………………………………………191

Chapter 5 General discussion………………………………………………………...193 5.1 Symbiotic relationships within Lotononis s. l. ………………………………………………..195 5.1.1 Habitat as a driver of rhizobial diversity………………………………………………..199 5.1.2 Lotononis s. l. rhizobia have diverse nodA sequences……………………………201 5.1.3 Specificity and effectiveness within Lotononis s. l. ……………………………….203 5.1.4 Models of legume-rhizobia symbiosis……………………………………………………204 5.2 Infection and nodule organogenesis in Listia spp. …………………………………………207 5.3 Characterisation of novel rhizobial species of Microvirga………………………………209

Bibliography…………………………………………………………………………………..213

Appendix……………………………………………………………………………………….249

viii PUBLICATIONS ARISING FROM THIS THESIS

Ardley, J. K., Garau, G., Yates, R. J., Parker, M. A., O’Hara, G. W., Reeve, W.

G., De Meyer, S., Walker R., Dilworth, M. J., Ratnayake, S., Willems, A.,

Watkin, E., and Howieson, J. G. 2011. Variations on a theme: novel rhizobial strains of Burkholderia, Methylobacterium and Microvirga demonstrate the diversity of root nodule bacteria/legume symbioses. Advances in Legume Systematics (in press).

Ardley, J. K., Parker, M. A., De Meyer, S., Trengove, R. D., O'Hara, G. W.,

Reeve, W. G., Yates, R. J., Dilworth, M. J., Willems, A. & Howieson, J. G.

Microvirga lupini sp. nov., Microvirga lotononidis sp. nov., and Microvirga zambiensis sp. nov. are Alphaproteobacterial root nodule bacteria that specifically nodulate and fix nitrogen with geographically and taxonomically separate legume hosts. International Journal of Systematic and Evolutionary Microbiology. Published online ahead of print December 23, 2011, doi: 10.1099/ijs.0.035097-0.

ix ACKNOWLEDGEMENTS

I would like to take this opportunity to thank all the many people who have helped or supported me during this PhD.

Firstly, I would like to express my sincerest appreciation and gratitude to my four supervisors Professor John Howieson, Dr Graham O’Hara, Dr Wayne Reeve and Dr Ron Yates. You have all given generously of your time and knowledge in your various areas of expertise and have provided invaluable guidance and support. It has been a privilege and a pleasure to work with you.

I would also like to thank Professor Mike Dilworth for contributing so much to this project, whether in the form of scientific expertise on all things rhizobial, or in the provision of insightful editorial comments during the preparation of this thesis.

Thanks also to Dr Ravi Tiwari and Dr Lambert Bräu for always being prepared to lend an ear and to Dr Vanessa Melino for being a kind and enthusiastic colleague, and generator of great discussions. To Regina Carr, many thanks for being a great friend and for providing valuable advice and assistance during the many glasshouse trials. Many thanks also to Gordon Thomson for his expert and beautiful histological preparations and generous advice on microscopy and to Dominie Wright for kindly giving advice on the preparation of Biolog plates.

I wish to thank Dr Catherine Masson-Boivin, Dr Xavier Perret and Dr Bharat

Patel for kindly providing the strains ORS 2060, NGR234 and FaiI4, respectively. I also wish to thank Dr Matt Parker, Dr Anne Willems and Dr Sofie De Meyer for the

x time, hard work and expertise they have supplied to the collaborative research into novel rhizobial strains of Microvirga.

I have been lucky enough, during the course of these studies, to meet many wonderful researchers in the field of nitrogen fixation, none more so than Professor

Janet Sprent, who has been a most generous source of encouragement, ideas and inspiration.

I also count myself very fortunate to have been able to work with all the past and present members of the Centre for Rhizobium Studies, who have made the CRS such an enthusiastic and supportive research environment. Especial thanks to Sharon,

Rebecca, Macarena, Felipe, Liza, Rob, Jason, Will, Yvette and Tina, who have shared the journey and provided friendship, support, advice, good humour, good conversation and, on occasion, great times together.

Finally, to my family, many thanks for all your love and patience.

xi ABSTRACT

Lotononis s. l. is a legume clade within the Crotalarieae tribe, with a centre of origin in South Africa. After taxonomic revision, the three genera Listia, Leobordea and Lotononis s. str. are now recognised. The N2-fixing symbiosis between Listia bainesii and pigmented Methylobacterium rhizobia is known to be highly specific, while a recent study has shown that Listia angolensis is effectively nodulated by a novel lineage of root nodule bacteria. The symbiotic relationships of Lotononis s. l. species outside the Listia genus have not yet been examined. The work presented in this thesis sought to determine the identity of rhizobia isolated from Listia,

Leobordea and Lotononis s. str. hosts, to examine the phylogeny of their nodA genes and to quantify the nodulation and N2 fixation capabilities of Lotononis s. l.- associated rhizobia on eight taxonomically diverse Lotononis s. l. hosts.

Additionally, this research sought to examine the processes of infection and nodule initiation in L. angolensis and L. bainesii and to validly name and characterise the novel L. angolensis rhizobia.

Amplification and sequencing of nearly full length fragments of the 16S rRNA gene showed that the rhizobia isolated from nodules of Lotononis s. l. species were phylogenetically diverse. Strains isolated from Leobordea and Lotononis s. str. hosts were most closely related to Bradyrhizobium spp., Ensifer meliloti,

Mesorhizobium tianshanense and Methylobacterium nodulans. The Listia angolensis microsymbionts, together with closely related Lupinus texensis rhizobia, were identified as novel species of the Alphaproteobacterial genus Microvirga. The

xii phylogeny of the nodA genes correlated more with the rhizobial 16S rRNA genes than with the taxonomy of the host plants.

The nodulation and effectiveness trials confirmed the symbiotic specificity of the genus Listia. L. bainesii nodulated only with the representative pigmented

Methylobacterium strain WSM2598. As measured by plant shoot dry weight, this symbiosis was highly effective. L. angolensis was effectively nodulated only by

Microvirga rhizobia, but formed ineffective nodules with the pigmented

Methylobacterium and M. nodulans strains. WSM2598 was effective only on Listia species (other than L. angolensis). In contrast, the Microvirga strain WSM3557 was partially effective on some Leobordea and Lotononis s. str. species. No clear pattern of symbiotic association was seen in the Leobordea and Lotononis s. str. hosts.

Nodulation in these species was usually promiscuous and often ineffective for N2 fixation.

The nodules formed on Listia species were lupinoid, whereas the nodules of

Leobordea and Lotononis s. str. species were indeterminate. Micrographs of all sectioned nodules, whether lupinoid or indeterminate, showed a mass of central, uniformly infected tissue, with no uninfected interstitial cells. Rhizobial infection and nodulation in L. angolensis and L. bainesii did not appear to involve root hair curling or the development of infection threads. Nodule organogenesis followed a process similar to that observed in Lupinus species, with nodule primordia developing in the outer cortex.

xiii The polyphasic characterisation of the Microvirga rhizobia associated with

Listia angolensis and Lupinus texensis resulted in the description of three new rhizobial species: Microvirga lotononidis, M. zambiensis and M. lupini. Microvirga species possess several phenotypic properties that are unusual in rhizobia, including the ability to grow at relatively elevated temperatures and the presence of pigmentation in most strains. The rhizobial Microvirga strains WSM3557T and Lut6T have been included in the Genomic Encyclopedia for Bacteria and Archaea Root

Nodule Bacteria (GEBA-RNB) sequencing project (Kyrpides & Reeve, collaborative

CSI project (http://genome.jgi-psf.org/programs/bacteria-archaea/GEBA-RNB.jsf) and should provide new insights into the evolution of and genomic architecture required for rhizobial symbionts.

xiv Chapter 1

CHAPTER 1

Introduction and literature review

1 Chapter 1

2 Chapter 1

1.1 The legume-rhizobia symbiosis

Biological nitrogen fixation (BNF), in which bacteria reduce atmospheric nitrogen (N2) to the biologically usable form of ammonia, is a highly important source of nitrogen (N) in both natural and agricultural systems, providing about 65% of the biosphere’s N (Lodwig et al., 2003). In terms of the quantity of N2 fixed and in its importance to agricultural systems, the symbiosis between root nodule bacteria

(RNB, collectively known as rhizobia) and legumes is the most ecologically and economically important form of BNF. It is estimated to contribute approximately 50

- 70 Tg (Tg = 1012 g) of fixed N to agricultural systems per year (Herridge et al.,

2008). The rate of N2 fixation by nodulated legumes varies from 57 to 600 kilograms per hectare per year compared with 0.1 to 0.5 kg ha-1 year-1 obtained from free-living

Azotobacter and Clostridium (Evans & Barber, 1977). Unlike the N2-fixing associations of epiphyte species of Gluconacetobacter, Azospirillum,

Herbaspirillum, and Azoarcus with various grasses, most of the N2 fixed in nodules is transferred to and assimilated by the plant for its growth (Hirsch et al., 2001).

Legumes are important components of agriculture as both pastures and pulses, not only for their input of fixed N but also in their contribution to the agronomy of crop production (Graham & Vance, 2003). Grain legumes rank as the second major type of field crop after cereals and constitute an important component of people’s diet

(Broughton et al., 2003; Peoples et al., 1995).

3 Chapter 1

1.2 The legume-rhizobia symbiosis in Australian agriculture

1.2.1 Background

Southern Australian agricultural systems are characterised by a Mediterranean climate, with predominantly winter rainfall varying from 250 – 1000 mm annually

(Nichols et al., 2007). Soils are often acid, have a low clay content and low organic matter, and tend to be inherently infertile (Howieson et al., 2000; Nichols et al.,

2007). Introduced legumes are established components of these agricultural systems

(Howieson et al., 2000). Traditionally, pasture legumes have been Mediterranean annuals such as medics and subterranean clover (Loi et al., 2005), but recent challenges to the sustainability of these practices have emerged. Howieson et al.

(2000) have identified several threats to current practice, such as the emergence of new pests and diseases, less reliable winter rainfall and, in particular, the development of dryland salinity.

Dryland salinity occurs when an increase in groundwater recharge raises the water table, bringing highly saline water to the surface (Hatton et al., 2003). In these agricultural systems, it is attributable to the replacement of native deep-rooted perennial vegetation with shallow-rooted annual crops and pastures (Dear et al.,

2003). In Western Australia, approximately 3 million hectares are estimated to be affected and more than 6 million hectares are potentially at risk (George et al., 1997).

For this reason, greater use of perennials in these farming systems has been advocated, as a means of preventing the development of dryland salinity (Cocks,

2003; Dear et al., 2003). Lucerne (Medicago sativa) is a perennial pasture legume that has been shown to lower ground water levels (Cocks, 2003), but its use is constrained by its poor adaptation to low rainfall, acid soils and waterlogging

4 Chapter 1

(Cocks, 2001; Dear et al., 2003). Researchers have therefore been engaged in an examination of the global flora for climatically and edaphically adapted perennial legumes that are potential alternatives to lucerne (Howieson et al., 2008; Li et al.,

2008; Nichols et al., 2007). These legumes include several species of Listia

(previously Lotononis (Boatwright et al., 2011)) that have shown potential as alternatives to lucerne in areas of high acidity, low rainfall or waterlogging (Dear &

Ewing, 2008; Yates et al., 2007). As part of this development of exotic pasture legumes, an interspecific crossbreeding program has been undertaken by the Centre for Rhizobium Studies to obtain Listia hybrids with non-shattering pods and larger seed size (Anon., 2006; Howieson, 2006).

The introduction of exotic legume pasture plants also requires a concomitant evaluation of their rhizobial partners, as N2 fixation will be compromised if inoculants are poorly adapted to the target environment, poorly competitive against resident rhizobial strains or sub-optimal for fixation effectiveness (Howieson &

Ballard, 2004; Sessitsch et al., 2002). The rhizobiology of Listia is interesting, as species in this genus are nodulated only by strains of pigmented Methylobacterium or, in the case of Listia angolensis, by light-pink-pigmented strains that are likely to belong to a new genus of rhizobia (Yates et al., 2007). Before proceeding with a description of the research into symbiotic relationships within this group of legumes, however, it is appropriate to review the current state of knowledge of the legume- rhizobia symbiosis.

5 Chapter 1

1.3 Legumes

1.3.1 Evolution of legumes

Legumes belong in the Rosid 1 clade of the angiosperms. The ten families of flowering plants that are able to form endosymbiotic associations with nodulating

N2-fixing bacteria are placed within this clade, according to a phylogeny based on the chloroplast rbcL gene (Soltis et al., 1995). This suggests that, rather than N2- fixing symbioses evolving in disparate families, the underlying genetic architecture necessary for nodulation and fixation is confined to one lineage of closely related angiosperm taxa (Soltis & Soltis, 2000). Parasponia, a member of the Ulmaceae

(and included in the Rosid 1 clade) is the only non-legume known to form a N2- fixing symbiosis with rhizobia (Sprent, 2007).

The Leguminosae are the third largest flowering plant family, comprising over

700 genera and some 20,000 species (Lewis et al., 2005). They are found throughout temperate and tropical regions and are particularly diverse in seasonally dry tropical forests and xeric climate temperate shrublands. Several species are aquatic (Lewis et al., 2005). Their morphology is equally diverse, ranging from large trees to perennials, climbing vines and annual herbs (Doyle & Luckow, 2003).

The first legumes are believed to have evolved some 60 million years ago

(mya) in the Paleocene epoch (Sprent, 2007). It has been hypothesised that the earliest legumes first appeared in the boreotropical forests of North America (Doyle

& Luckow, 2003), but it is now thought more likely that they arose in a succulent biome in seasonally dry tropical forest to the north of the Tethys seaway (Figure 1.1)

(Schrire et al., 2005) Their nitrogen-demanding metabolism is thought to be an

6 Chapter 1 adaptation to climatically variable arid or semi-arid habitats (Wojciechowski et al.,

2004). Rapid diversification of the Leguminosae then followed, such that by the middle Eocene (50 mya) most of the major lineages had arisen and are present in the fossil record (Cronk et al., 2006; Lavin et al., 2005).

Tethys seaway

Figure 1.1. Map of the globe 60 million years ago, showing the position of the continents and the Tethys seaway. Adapted from Scotese (2004).

1.3.2 Legume taxonomy and phylogeny

Using the terminology of Lewis et al. (2005), the Leguminosae are divided into three subfamilies, the Caesalpinioideae, the Mimosoideae and the Papilionoideae.

These three subfamilies are further divided into groups of genera called tribes. The

Papilionoideae are the largest subfamily, containing 28 tribes, 476 genera and around

14,000 species. The Mimosoideae have 4 tribes, 77 genera and 3,000 species and there are currently 4 tribes, 171 genera and 2,250 species in the Caesalpinioideae.

(Doyle & Luckow, 2003; Lewis et al., 2005). Both the Papilionoideae and the

7 Chapter 1

Mimosoideae are monophyletic. Caesalpinioids, in contrast, appear to be a paraphyletic grouping of diverse lineages (Lewis et al., 2005).

Recent molecular phylogeny based on analyses of plastid gene sequences shows the caesalpinioids forming a basal clade within which the other subfamilies are nested (Lavin et al., 2005; Wojciechowski et al., 2004). The Mimosoideae appear to have diverged relatively recently (40 mya) and are a sister group to a caesalpinioid lineage. In contrast, the papilionoids constitute an early branching from the major caesalpinioid clades. This accords with the fossil evidence; the oldest known legume fossils, dating from around 56 mya, are of leaves, fruits and flowers that are similar to extant genera in the papilionoid genistoid clade (Lavin et al., 2005;

Wojciechowski et al., 2004).

1.3.2.1 Phylogeny of the Papilionoideae

Molecular analyses have resolved a number of distinct groups in the

Papilionoideae and can be used, together with fossil evidence, to estimate when particular legume “crown” (diversification) clades first appeared (Lavin et al., 2005;

Wojciechowski et al., 2004). The phylogenetic analysis of Wojciechowski et al.

(2004) supports seven major sub-clades that are informally recognised as the

Cladrastis clade, genistoid sensu lato, dalbergioid sensu lato, mirbelioid, millettioid, and robinioid clades, and the inverted-repeat-lacking clade (IRLC) (Figure 1.2.)

(Cronk et al., 2006). This phylogeny is in agreement with that of McMahon &

Sanderson (2006), who have additionally recognised the Amorpheae clade (between the genistoids and the dalbergioids) and the Indigofereae clade (between the milletioids and the robinioids). The Cladrastis clade is basal to the other sub-clades.

8 Chapter 1

Figure 1.2. The major monophyletic clades of the papilionoid legumes. Figure taken from Cronk et al. (2006).

Cladrastis species have not been observed to nodulate and at least one species appears to lack this capacity (Foster et al., 1998). The genistoid clade dates from

56.4 mya. It includes the agriculturally important genus Lupinus (Sprent, 2007) and the large, mainly African tribe Crotalarieae (Boatwright et al., 2008). The Lotononis sensu lato clade, comprising the genera Listia, Leobordea and Lotononis sensu stricto, is included within tribe Crotalarieae (Boatwright et al., 2011). The dalbergioid clade is also an early diverging lineage, dating from 55.3 mya, and

9 Chapter 1 includes the stem-nodulated Aeschynomene species and economically important genera such as Dalbergia, Arachis and the forage legume Stylosanthes (Lavin et al.,

2001). The mirbelioid group consists mostly of endemic Australian tribes and originated 48 mya (Lavin et al., 2005; Sprent, 2007; Wojciechowski et al., 2004).

The milletioid crown dates from 45.2 mya and is divided into two sub-clades, the

Milletieae and the Phaseoleae. The latter includes the agriculturally important species

Phaseolus vulgaris and Glycine max (Sprent, 2007; Wojciechowski et al., 2004).

The two remaining clades are both included in the Hologalegina group, which comprises the majority of temperate, herbaceous legumes, including the well-known peas, chickpeas, clovers and lucerne and the model legumes Medicago truncatula and Lotus japonicus (Lavin et al., 2005; Wojciechowski et al., 2004). The first clade in this group is the robinioids, consisting of the Sesbanieae, Loteae and Robineae tribes and dated at 48 mya. The second clade is the comparatively recent (39 mya) inverted repeat loss clade (IRLC), named after the loss of one copy of the 25- kilobase chloroplast inverted repeat and containing members of the tribes Cicereae,

Trifolieae, Hedysareae and Fabeae (Sprent, 2007; Wojciechowski et al., 2000).

The existence of rhizobia predates that of legumes by several hundred million years (Turner & Young, 2000). In contrast to the monophyletic Leguminosae, rhizobia are a polyphyletic group and genera capable of nodulating hosts are found in both the Alpha- and Betaproteobacteria (Sawada et al., 2003). The following is a brief examination of rhizobial systematics.

10 Chapter 1

1.4 Rhizobia

Rhizobia are defined as Gram-negative, saprophytic soil or water micro- organisms that can form N2-fixing symbioses with legumes or Parasponia by eliciting nodules on the roots or stems of their hosts (Masson-Boivin et al., 2009;

Sprent, 2007). Within the nodule, the free-living form of the microsymbiont differentiates into bacteroids that convert atmospheric N2 to ammonia (Jones et al.,

2007).

1.4.1 History of rhizobial classification

Prior to 1982, all root-nodule bacteria were included in the genus Rhizobium and were classified according to the legumes they were able to nodulate (Jordan,

1982; Sawada et al., 2003). Six species of Rhizobium were recognised: R. leguminosarum, R. trifolii, R. phaseoli, R. meliloti, R. japonicum (nodulating Glycine max) and R. lupini (Fred et al., 1932). These early studies understandably focussed on temperate, agriculturally important legume hosts and thus were not representative of the diversity of the Leguminosae, or their microsymbionts (Willems, 2006). The idea of classifying rhizobia according to their host range and grouping legumes according to their microsymbionts (the “cross-inoculation concept”) proved to be flawed, however, as some rhizobial strains are capable of nodulating a wide range of legumes (Pueppke & Broughton, 1999; Young & Haukka, 1996). Similarly, various legumes (notably those in tribe Phaseoleae) are nodulated by a broad spectrum of rhizobia (Perret et al., 2000). That there was some correlation between rhizobial host range and physiological properties resulted in the proposal that slow-growing strains be placed in a new genus, Bradyrhizobium, to distinguish them from the fast-growing

Rhizobium (Jordan, 1982; Young & Haukka, 1996). At this time, the classification of

11 Chapter 1 rhizobia, and bacteria in general, was still based on phenotypic properties such as cell morphology and physiological and biochemical markers (Schleifer, 2009). The advent of molecular techniques, in particular gene sequencing, revolutionised bacterial taxonomy by permitting the development of phylogenies that reflect bacterial genomic relatedness (Woese, 1987). This, along with studies of novel isolates from indigenous legumes in diverse biogeographical areas, has demonstrated that rhizobia are a larger and more diverse group that originally supposed (Lindström et al., 2010; Sawada et al., 2003; Willems, 2006).

Phylogenies based on a comparative analysis of the 16S rRNA gene sequence show that rhizobia are polyphyletic, being distributed in both the Alpha- and

Betaproteobacteria and intermingled with taxa that do not contain legume microsymbionts (Figure 1.3) (Masson-Boivin et al., 2009; Sawada et al., 2003).

Currently, 12 rhizobial genera and over 70 species have been described (Weir, 2011)

(http://www.rhizobia.co.nz/taxonomy/rhizobia.html). The majority of the currently described rhizobial species belong to the genera Rhizobium, Bradyrhizobium,

Mesorhizobium and Ensifer (including former Sinorhizobium), but novel

Alphaproteobacterial microsymbiont strains of Devosia (Rivas et al., 2002),

Methylobacterium (Sy et al., 2001), Ochrobactrum (Trujillo et al., 2005) and

Shinella (Lin et al., 2008) have also recently been characterised, along with rhizobial

Betaproteobacteria belonging to Burkholderia and Cupriavidus (Amadou et al.,

2008; Bontemps et al., 2010; Chen et al.,2007, 2008; Garau et al., 2009).

12 Chapter 1

Figure 1.3. Unrooted phylogenetic tree of 16S rRNA gene sequences from selected Alpha-, Beta- and Gammaproteobacteria. Genera in bold font contain rhizobia. Figure taken from Masson-Boivin et al. (2009).

Potential new rhizobial taxa are expected to conform to the proposed minimal standards for the description of new genera and species outlined by Graham et al.

(1991) and to the current recommendations for the characterisation of prokaryotes

(Stackebrandt et al., 2002; Wayne et al., 1987). To understand what this entails, it is worth examining the elements and methodologies upon which modern bacterial systematics – the taxonomy and phylogeny of prokaryotes – are now based.

1.5 Bacterial systematics

A classification system for living organisms should group taxa with similar genotypes and phenotypes together. Classification is traditionally based on the

“species concept” (Cohan, 2002; Rosselló-Móra & Amann, 2001). This is less well defined for prokaryotes than for eukaryotes, which are delineated far more by morphology and reproductive isolation. A further complicating factor in assigning

13 Chapter 1 bacteria to discrete, hierarchical taxonomies is the phenomenon of horizontal (also known as lateral) gene transfer (HGT or LGT), the process by which genetic material is transmitted between individual organisms. It contrasts with the vertical transmission of genes by direct descent from one generation to the next. HGT is a common and widespread occurrence in prokaryotes and has played an integral role in their evolution, diversification and speciation (Ochman et al., 2000).

Bacterial systematics now relies on a polyphasic approach that integrates genotypic data (derived from DNA and RNA present in the cell), phenotypic data

(such as chemotaxonomic markers) and phylogenetic information (for example sequences of highly conserved genes) (Schleifer, 2009; Vandamme et al., 1996). The species is considered to be the basic unit of taxonomy and is defined as a group of strains, including the type strain, that share 70% or greater DNA-DNA relatedness and with 5ºC or less ∆Tm (difference in the DNA-DNA hybrid melting points)

(Stackebrandt & Goebel, 1994; Wayne et al., 1987). Bacterial classification is expected to reflect the degree of relatedness of different microorganisms. The standard method of determining this phylogenetic relatedness is by sequencing the

16S or 23S rRNA gene (Woese, 1987). As a practicality, strains with <97% 16S rRNA gene sequence similarity to any known taxa are considered to belong to a different species, as species having 70% or greater DNA similarity usually have more than 97% sequence identity (Gevers et al., 2005; Stackebrandt & Goebel, 1994;

Vandamme et al., 1996). A phylogenetically based taxonomy must also show phenotypic consistency (Vandamme et al., 1996; Wayne et al., 1987). The following is a summary of polyphasic taxonomy methods as reviewed by Vandamme et al.

14 Chapter 1

(1996), as well as a critique of the limitations of some of the methods used in this approach.

1.5.1 Genotypic classification methods

Two genotypic methods are widely used in describing bacterial taxa. The first of these is the determination of the DNA base ratio, that is, determination of the mole percent guanosine plus cytosine. Organisms that differ by more than 10 mol% do not belong in the same genus and species should be less than 5 mol% different

(Schleifer, 2009; Wayne et al., 1987). Similar DNA base ratios, however, do not necessarily imply phylogenetic relatedness (Rosselló-Móra & Amann, 2001). The second method is the percent DNA-DNA hybridisation, which is an indirect measure of the sequence similarity between two entire genomes. DNA-DNA hybridisation is used to delineate species; as stated above, strains are considered to belong to the same species if they share 70% or greater DNA-DNA relatedness (Stackebrandt &

Goebel, 1994). One problem with DNA-DNA hybridisation is that it can sometimes be difficult to compare results, as these are dependent on the stringency of the hybridisation conditions (Vandamme et al., 1996). In addition, incremental databases cannot be developed for this method (Schleifer, 2009).

1.5.2 Phylogenetic classification methods

The development of molecular protocols for sequencing the small subunit ribosomal RNA (SSU rRNA) genes allowed the construction of valid bacterial phylogenies (Woese, 1987). The rRNA genes are considered the best means of studying phylogenetic relationships, as they are present in all bacteria, are functionally constant and are composed of highly conserved as well as more variable domains (Schleifer, 2009; Vandamme et al., 1996). The extensive publication of 16S

15 Chapter 1 rRNA gene sequences and the availability of the sequence data in databases such as the National Centre for Biotechnology Information (NCBI) GenBank allow rapid comparison of new isolates with named species and the subsequent construction of phylogenetic trees.

There are several problems though with using the rRNA genes as strict determinants of bacterial phylogenetic placement. Some bacterial taxa can contain more than one copy of the 16S rRNA gene and these alleles may have a relatively high sequence divergence (Acinas et al., 2004; Amann et al., 2000; Rainey et al.,

1996). The sequences of different ribosomal RNA operon (rrn) genes may also provide divergent phylogenetic information. The topology of a phylogenetic tree generated from diverse rhizobial 16S rRNA gene sequences was found to be significantly different from that of the corresponding tree assembled with 23S rRNA gene sequences (van Berkum et al., 2003). In addition, rrn operons are prone to homologous recombination, resulting in mosaic structures and consequent difficulties in accurately determining phylogenies. Evidence of mosaicism in 16S rRNA gene sequences has been found in species of Bradyrhizobium, Ensifer, Mesorhizobium and

Rhizobium (van Berkum et al., 2003; Vinuesa et al., 2005). Finally, comparative 16S rRNA gene sequence analysis can also lack resolving power at and below the species level, due to the extent of gene conservation (Schleifer, 2009). Bradyrhizobia, for example, share a high level of 16S rDNA sequence similarity and it is therefore difficult to evaluate the interrelationships of strains (Willems, 2006). For all these reasons, the rRNA genes should therefore not be used as sole determinants of bacterial taxonomies.

16 Chapter 1

The difficulties with resolving rRNA gene-based phylogenies led to the recommendation that sequences of protein-encoding “housekeeping” genes (required for the maintenance of basic cellular function) be included in the phylogenetic analysis of a taxon (Stackebrandt et al., 2002). These genes have a higher degree of sequence divergence than the rRNA genes, allowing greater resolution of taxonomies at the species or sub-species level (Martens et al., 2008; Palys et al., 2000). Ideally, several sets of genes that are widely distributed among taxa, in single copy and from diverse chromosomal loci should be chosen (Gevers et al., 2005; Stackebrandt et al.,

2002). Housekeeping genes such as atpD, dnaK, glnA, gyrB, GSI and GSII, recA and rpoB have been widely used in phylogenetic studies of rhizobia and other bacteria

(Adékambi & Drancourt, 2004; Gaunt et al., 2001; Martens et al., 2008; Stepkowski et al., 2003; Turner & Young, 2000; Vinuesa et al., 2005). As with the 16S rRNA gene, the availability of extensive sequence data for housekeeping genes facilitates the phylogenetic analysis of a given group of isolates.

The intraspecific variation between bacterial strains can be identified by amplification of highly conserved repetitive intergenic DNA sequences. These are short (usually <200bp), non-coding DNA sequences found in extragenic locations and widely distributed in prokaryotic genomes (Lupski & Weinstock, 1992). PCR amplification of these sequences generates products which, when separated on an agarose gel, give a characteristic banding pattern that can be used as a genomic fingerprint to identify bacteria at the species, sub-species or strain level (Versalovic et al., 1991). There are several families of these sequences, two of which are the enterobacterial repetitive intergenic consensus (ERIC) sequences and the BOX sequences (Lupski & Weinstock, 1992; Martin et al., 1992). These have been used to

17 Chapter 1 distinguish rhizobial strains in a number of studies (Alberton et al., 2006; de Bruijn,

1992; Ormeño-Orrillo et al., 2006).

1.5.3 Phenotypic classification methods

Phenotypic description requires the analysis of morphological, physiological and biochemical features and is a requirement of a valid bacterial species definition

(Rosselló-Móra & Amann, 2001; Wayne et al., 1987). Perhaps as importantly, the determination of bacterial phenotypes not only allows for the classification of bacteria, but also provides important information about the roles played by a species in the natural environment and its adaptation to a particular environmental niche

(Bochner, 2009; Fenchel & Finlay, 2006).

Classical phenotypic tests constitute the basis for the formal description of bacterial taxa (Vandamme et al., 1996). These include the determination of bacterial morphology (cellular shape, presence of endospores, flagella, Gram staining, inclusion bodies) and colonial characteristics. The physiological and biochemical analyses provide data on growth at different temperatures, pH values, salt concentrations or aerobic/anaerobic conditions; in the presence of various antibiotics; on various compounds, including sole carbon substrates; and data on the presence or activity of various enzymes (Rosselló-Móra & Amann, 2001; Vandamme et al.,

1996).

Commercial, miniaturised, phenotypic fingerprinting systems, such as Biolog and API, can provide rapid and reproducible results under standardised conditions

(Rosselló-Móra & Amann, 2001; Vandamme et al., 1996). The Biolog system

(http://www.biolog.com/phenoMicro.html) is based on microplates that can assay up

18 Chapter 1 to nearly 2000 culture traits and in so doing are able to provide an analysis of the physiology of the cell (Bochner, 2003; Bochner, 2009). The technology uses cell respiration to reduce a tetrazolium dye, forming a strong colour that can be read by a computerised microplate reader (Bochner, 1989). The bioMérieux

(http://www.biomerieux.com/servlet/srt/bio/portail/home) API strips typically contain 20 miniature biochemical tests. Both these methods can be used in conjunction with a database to identify bacteria, particularly in clinical diagnostics.

Chemotaxonomy – the collection and use of data on the chemical composition of cell constituents in order to classify the bacteria - is a more recent addition to phenotypic methods. It makes use of differences in the distribution of particular chemicals, notably amino acids, proteins, lipids and sugars, amongst specific bacterial taxa (Rosselló-Móra & Amann, 2001). Two standard methods used are cellular fatty acid composition and whole-cell protein analysis. Cellular fatty acids are the major constituents of lipids in cell membranes and lipopolysaccharides and their variability in chain length, double-bond position and substituent groups has been useful in characterising bacterial taxa (Vandamme et al., 1996). Whole cell protein analysis compares the protein patterns obtained from sodium dodecyl sulfate- polyacrylamide gel electrophoresis. It is a reliable method for comparing and grouping large numbers of closely related strains (Vandamme et al., 1996).

1.5.4 The effect of horizontal gene transfer on bacterial phylogeny

The effect of HGT on bacterial evolution and the ecological and pathogenic character of bacterial species is considerable. Bacterial species are open to gene transfer from many other species, even those that are distantly related (Cohan, 2002) and it is estimated that 5%–15% of the genes in a typical bacterial genome have been

19 Chapter 1 acquired from other species (Ochman et al., 2000). The impact of such transfer is that molecular phylogenies calculated for different molecules from the same set of species are only rarely completely congruent (Gogarten et al., 2002). Perhaps more importantly, HGT of entire operons has been detected (Omelchenko et al., 2003), thus providing a mechanism for the potential gain of new metabolic capabilities or to confer bacterial antibiotic resistance, pathogenicity or photosynthetic or symbiotic ability (Barcellos et al., 2007; Boucher et al., 2003; Ochman et al., 2000; Ochman &

Moran, 2001; Sullivan et al., 1995).

This phenomenon of dynamic prokaryotic genomes has led some researchers to suggest that a network model, rather than a tree, is a more accurate depiction of bacterial phylogeny (Kloesges et al., 2011; Koonin & Wolf, 2008; Kunin et al.,

2005). Others have argued, however, that carefully selected gene sequences can still provide valid phylogenies of bacterial lineages (Daubin et al., 2003). This view is based on the concept that bacterial genomes consist of three distinct pools of genes: the “core” genome, the “character or lifestyle” genes and the “accessory” genes

(Schleifer, 2009). The core genome consists of a group of essential genes that are common to all genomes of a phylogenetically coherent group of bacteria. It preferentially contains informational or housekeeping genes that are stable and less prone to HGT and thus are suitable candidates for phylogenetic analysis. The second gene pool contains genes, such as those that code for specific metabolic properties, which allow the bacteria to survive in a particular environment. The accessory genes are non-essential, less conserved and often strain specific (Schleifer, 2009) and help to determine the specific ecological properties of an organism (Fraser et al., 2009).

They are notably more A+T-rich, more prone to HGT and more usually found on

20 Chapter 1 mobilisable genetic elements such as plasmids and chromosomal islands (Daubin et al., 2003; Young et al., 2006).

The frequency of HGT is positively correlated not only with accessory genes, but also with genome size, with physical proximity (i.e. among species sharing the same habitat) and with the degree of relatedness of the bacterial species (Kloesges et al., 2011). In the latter case, phylogenetic distance imposes barriers to HGT due to the restricted host ranges of transmissible agents such as bacteriophages and conjugative plasmids, and to differences in the apparatuses of transcription and translation (Lawrence & Hendrickson, 2003).

Within the framework outlined above, how should the available methodologies be used to validly describe and classify bacterial (and in particular, rhizobial) species?

1.6 General recommendations for classifying and describing bacteria

Although there is no official classification for prokaryotes, the classification system represented by Bergey’s Manual of Systematic Bacteriology

(http://www.bergeys.org/pubinfo.html) is widely accepted and is the standard reference work on bacterial classification. Bacterial nomenclature, on the other hand, is governed by the Bacteriological Code (Lapage et al., 1992). A list of current prokaryotic names with standing is maintained by J. P. Euzéby at http://www.bacterio.cict.fr/. Guidance on the use of current methodologies in the taxonomic characterisation of prokaryote strains is provided in a series of reports and

21 Chapter 1 notes published in the International Journal of Systematic and Evolutionary

Microbiology (Stackebrandt et al., 2002; Tindall et al., 2010; Wayne et al., 1987).

The general recommendations for the description of new genera and species of rhizobia can be found at http://edzna.ccg.unam.mx/rhizobial-taxonomy/node/12 and in Graham et al. (1991). The recommendations are summarised as follows:

• New taxon descriptions should be based on a minimum of at least three

distinct strains, as revealed by molecular markers and, where possible, several

populations from different ecological settings should be sampled.

• A range of different molecular markers suitable to uncover and analyse

genetic diversity at different phylogenetic/taxonomic depths should be used.

Generating full-length 16S rDNA sequences for a few carefully selected

strains, along with the partial sequencing of two protein-coding core loci (e.g.

recA and rpoB) and at least one symbiotic locus (e.g. nifH, nodA or nodC) is

recommended.

• Phenotypic tests should include host range as well as pH and temperature

growth-range, salt tolerance, growth on different carbon and nitrogen sources,

antibiotic resistance profiling and fatty acid methyl-ester analysis. Phenotypes

and chemotaxonomic markers that are relevant as ecologically adaptive traits

should also be included.

• DNA-DNA hybridization should be determined for three distinct strains of a

new potential taxon in order to get an estimate of the standard deviations of

homology values (genome heterogeneity) within the new taxon.

22 Chapter 1

1.7 Rhizobial symbiotic genes

As stated above, rhizobia are polyphyletic and intermingled with non- symbiotic prokaryotes. Symbiotic ability appears to be conferred by a group of approximately 400 genes that form part of the accessory gene pool and enable nodulation and N2 fixation with a particular legume (Broughton & Perret, 1999;

Young et al., 2006). These symbiotic genes are clustered together in potentially transferable genomic elements such as plasmids or megaplasmids in Rhizobium or

Sinorhizobium species, or genomic islands in Azorhizobium, Bradyrhizobium and some Mesorhizobium species (Freiberg et al., 1997; González et al., 2003; Kaneko et al., 2002; Lee et al., 2008; Sullivan & Ronson, 1998; Young et al., 2006).

Regions encoding the rhizobial symbiotic genes are marked by the presence of insertion sequence elements, transposases, phage-related integrases and related genes that putatively provide mechanisms for HGT (MacLean et al., 2007). HGT of symbiotic loci between both closely and distantly related rhizobial taxa has been demonstrated in several studies (Andam et al., 2007; Barcellos et al., 2007;

Cummings et al., 2009; Johnston et al., 1978; Nandasena et al., 2007a; Sullivan &

Ronson, 1998). What is the extent of HGT of symbiotic loci between distantly related rhizobia? As well as direct evidence that this occurs (Barcellos et al., 2007),

HGT across different genera can be inferred where phylogenetically diverse rhizobia nodulating the same legume host possess the same or similar nodulation genes, or where the phylogeny of the nodulation genes reflects the host plant taxonomy rather than the rhizobial chromosomal lineage (Chen et al., 2003; Haukka et al., 1998;

Laguerre et al., 2001; Lu et al., 2009; Suominen et al., 2001). On the other hand, there is evidence that HGT is more prevalent in closely related rhizobia and occurs

23 Chapter 1 within, but not between, different genera (Wernegreen & Riley, 1999). Phylogenetic studies of the symbiotic loci of diverse rhizobia support the idea that the bacterial chromosomal background is an important determinant in nodulation gene transfer between rhizobial strains (Haukka et al., 1998; Moulin et al., 2004).

1.7.1 The role of nodulation genes and Nod factors

The rhizobial nodulation (nod, nol and noe) genes are structural and regulatory genes that confer the ability to infect and nodulate the legume host. They form part of a molecular dialogue between the host and the rhizobia, in which plant-derived flavonoids induce expression of the nod genes and the subsequent synthesis of lipo- chito-oligosaccharide (LCO) Nod factors (NFs) (Perret et al., 2000). The NFs in turn induce plant genes that control and coordinate the separate processes of bacterial infection and nodule organogenesis (D’Haeze & Holsters, 2002).

The LCO backbone is composed of usually four or five β1-4-linked N-acetyl glucosamine residues that carry an N-linked C16 or C18 acyl chain on the terminal non-reducing sugar (Gough & Cullimore, 2011). It is encoded by the canonical nodABC genes that have to date been found in all rhizobia (excepting some bradyrhizobial strains that form stem nodules on Aeschynomene species (Miché et al., 2010)), whereas other nodulation loci encode substituent groups that “decorate” the LCO core (Perret et al., 2000) (Figure 1.4). The variations in LCO structure

(number of glucosamine residues, length and degree of saturation of acyl tail and addition of substituent groups) are characteristic of a rhizobial strain or species

(D’Haeze & Holsters, 2002; Dénarié et al., 1996) and although NF structure alone cannot be used to predict host range, there is a correlation between the type of NF produced and the rhizobial host range (Dénarié et al., 1996; Perret et al., 2000).

24 Chapter 1

Rhizobia synthesise a suite of NFs, which can consist of from two to 60 individual

NF structures, according to the rhizobial strain or environmental factors (D’Haeze &

Holsters, 2002; Morón et al., 2005). The quantity and variety of the NF structures is also a determinant of rhizobial host range: the exceptionally broad host range strain

NGR234 secretes high concentrations of a large family of NF structures, in which a wide palette of substituent groups variously decorate the LCO core (Schmeisser et al., 2009).

A

Figure 1.4. General structure of Nod factors produced by rhizobia. The substituent groups that decorate the lipo-chito-oligosaccharide core and the genes that code for the oligosaccharide moiety, the acyl chain and the substituent groups are indicated. Figure taken from Perret et al. (2000).

Studies of the model legumes Lotus japonicus and Medicago truncatula show that it is the specific recognition of NFs by LysM receptor-like kinases in the epidermis of the root hair that triggers the complex signalling pathway controlling bacterial infection and nodule morphogenesis (Kouchi et al., 2010; Madsen et al.,

2010; Oldroyd & Downie, 2008).

1.8 Rhizobial infection and nodule formation in legumes

A key aspect of the legume-rhizobia symbiosis is the organogenesis of a specialised structure, the nodule, which develops on the roots or stems of the host plant. Within the cells of the nodule the differentiated bacteria (bacteroids) are housed in plant membrane-bound compartments (symbiosomes), where they reduce

25 Chapter 1

N2 to ammonia (Oldroyd & Downie, 2008). Nodule shape and structure are determined by the host plant and have taxonomic value (Corby, 1981; Lavin et al.,

2001; Sprent & James, 2007). The legume host also controls the infection pathway.

Sprent (2007; 2008b) has identified several types of nodule and two modes of infection that are characteristic of the different legume groups in which they are found.

1.8.1 Infection pathways in legumes

Within the Papilionoideae, rhizobial infection may proceed intracellularly via root hair curling, with the bacteria confined to an infection thread (IT) before their release into nodule primordium cells, or may be directly through the epidermis without IT involvement in nodulation (Sprent & James, 2007). The majority of legume taxa, including species in the tribe Phaseoleae and in the hologalegoid clade, are usually infected via a root hair pathway (Sprent, 2009) and this well-studied mode of bacterial entry has been the subject of several reviews. Briefly, infection begins with the root hair curling around and enclosing attached rhizobia. Within the curl, the plant cell wall is locally hydrolysed, the plasma membrane invaginates and new cell wall material is deposited to form the IT, which then grows through the epidermal cell and into the root cortex, while the bacteria within the thread divide and proliferate. Cytoplasmic bridges (also known as pre-infection threads) form in the cortical cells to guide the IT towards the developing nodule primordium. Once it reaches the nodule primordium, the IT ramifies, followed by the release of bacteria into the primordial nodule cells. The rhizobia, now surrounded by a peribacteroid membrane (symbiosomes), differentiate into their symbiotic forms (bacteroids) and begin to fix N2 (Brewin, 2004; Gage, 2004; Maunoury et al., 2008).

26 Chapter 1

The alternative intercellular mode of infection, where rhizobia enter the host plant directly through the epidermis and are not confined to infection threads, is found in at least 25% of extant legumes and is characteristic of the more basal genistoid and dalbergioid clades (Sprent, 2008a). This type of infection (commonly referred to as crack entry) has been observed in Aeschynomene fluminensis (Loureiro et al., 1995), Arachis hypogaea (Boogerd & van Rossum, 1997), Cytisus (previously

Chamaecytisus) proliferus (Vega-Hernández et al., 2001) Genista tinctora (Kalita et al., 2006) Lupinus species (González-Sama et al., 2004; Tang et al., 1992) and

Stylosanthes species (Chandler et al., 1982). Intercellular infection occurs by penetration of the middle lamella, often at the junction between a root hair base and an adjoining epidermal cell (González-Sama et al., 2004; Uheda et al., 2001), and in dalbergioid legumes is associated with the emergence of lateral or adventitious roots

(Lavin et al., 2001). Bacteria then spread through the cortex in an intercellular matrix. In some legumes, this invasion appears to induce local cell death, creating an intercellular space filled with rhizobia (Boogerd & van Rossum, 1997; Vega-

Hernández et al., 2001). The rhizobia eventually penetrate the altered cell wall of nodule primordium cells. Mitotic division of the newly infected cells and symmetrical distribution of the symbiosomes (similar to host cell organelles), gives rise to nodules with characteristic uniformly infected central tissue (Boogerd & van

Rossum, 1997; Fedorova et al., 2007; Sprent & James, 2007).

Studies of both IT (Lotus japonicus and Medicago truncatula) and non-IT

(Lupinus albus) legumes show that the infected nodule cells undergo cycles of endoreduplication, leading to enlarged polyploid cells (González-Sama et al., 2006;

Maunoury et al., 2008). In some legumes, differentiation of the bacteroids also

27 Chapter 1 involves endoreduplication and an increase in cell size. This has been studied most notably in the Inverted Repeat Lacking Clade legumes, where bacteroid differentiation is terminal and bacteroids are unable to resume growth after release from a nodule; conversely bacteroids within phaseoloid nodules are not swollen or polyploid and remain viable (Maunoury et al., 2008; Mergaert et al., 2006).

Although the type of bacteroid differentiation is dependent upon the host plant, swollen bacteroids have been found in legume species belonging to five out of the six major papilionoid subclades and this host-imposed control is considered by Oono et al. (2010) to have evolved independently within multiple legume clades.

It has been postulated that intercellular infection is likely to be the ancestral mode of rhizobial entry and it remains a default pathway, as crack entry can occur in host plants that are usually infected by root hair curling (Sprent & James, 2007;

Sprent, 2008a). An example is the semi-aquatic robinioid legume Sesbania rostrata.

Rhizobial infection in this plant is via root hair curling, but in flooded conditions

(where waterlogged roots are mainly hairless) bacteria enter via epidermal fissures at the sites of adventitious root emergence. Subsequent IT formation then guides the rhizobia towards the nodule primordium (Capoen et al., 2010). Similar crack entry followed by the development of ITs has been observed in aquatic mimosoid

Neptunia species and in flooded Lotus uliginosus plants (James et al., 1992; James &

Sprent, 1999; Subba-Rao et al., 1995). Lonchocarpus muehlbergianus (tribe

Milletieae), which does not produce root hairs, also appears to be nodulated via an epidermal infection followed by the formation of ITs (Cordeiro et al., 1996).

Strikingly, the Lotus japonicus ROOT HAIRLESS mutant can be effectively nodulated via crack entry through the cortical surface of the nodule primordium and

28 Chapter 1 subsequent development of ITs (Karas et al., 2005; Madsen et al., 2010). The symbiosis between rhizobia and the non-legume Parasponia, where crack entry is followed by the retention of rhizobia in thin-walled ITs called fixation threads

(Becking, 1992), further supports the theory of ancestral intercellular infection.

1.8.2 Nodule types

Nodules are formed when rhizobia induce the dedifferentiation of the root cortical cells, activating their cell cycle and establishing a cluster of meristematic cells that give rise to the nodule primordium (Maunoury et al., 2008). Legume nodules have traditionally been divided into two types: determinate and indeterminate, according to whether the nodule possesses a persistent meristem

(Gibson et al., 2008). Sprent (2007), however, has identified several major types of nodule (Figure 1.5), based on nodule structure, meristem persistence and the presence or absence of ITs and this terminology will be used in this review. Notably, in legumes infected via ITs, the active N2-fixing nodule tissue contains a mix of infected and uninfected cells, whereas in intercellularly infected legumes the central tissue of the N2-fixing nodule is uniformly infected (Sprent & James, 2007; Sprent,

2009).

29 Chapter 1

Figure 1.5. Main features of legume nodules. (a-b) Nodules in which infection threads carry bacteria to individual cells derived from the meristem and the central tissue contains infected and uninfected cells. (a) Indeterminate form as found in the Caesalpinioideae, Mimosoideae and Papilionoideae subfamilies, where infection threads carry bacteria to individual cells derived from the meristem. There is a gradient of developmental zones and the nodule may be branched or unbranched. (b–e) Nodules are found only in the Papilionoideae. (b) Determinate (desmodioid) nodule, rounded in shape, with no age gradient in the infected tissue. (c-e) Nodules in which infection threads are not associated with nodulation and the central uniformly infected tissue is formed by division of a few initially infected cells. (c) Determinate dalbergioid nodule, associated with a lateral root. (d) Indeterminate nodule with an age gradient, found in many genistoid legumes. It may be unbranched or branched. (e) The lupinoid nodule, which is a variant of (d) where the meristems grow around the subtending root. Found in some genistoid legumes such as Lupinus and Listia. IC = infected cells; UC = uninfected cells; M = meristem; C = cortex; R = subtending root; LR = lateral root; arrow indicates a gradation in age of infected cells, youngest towards arrowhead. Figure taken from Sprent (2007).

1.8.2.1 Formation and structure of nodules infected via infection threads

Both determinate and indeterminate nodules are found in legumes that are infected via ITs and (usually) root hair curling. The model legumes Lotus japonicus and Medicago truncatula, respectively, provide well-studied examples of each nodule type (Maunoury et al., 2008). Indeterminate nodules occur in the majority of

30 Chapter 1 legumes. They originate from cell division in the inner cortex and retain a meristem, giving them a characteristic cylindrical or lobed shape (Gualtieri & Bisseling, 2000).

The nodule cells are divided into a gradient of developmental zones, with a persistent apical meristem (zone I), an infection zone (zone II), a fixation zone (zone III) and, in mature nodules, a senescent zone (zone IV) that is established proximal to zone III

(Guinel, 2009; Maunoury et al., 2008). Determinate (or desmodioid) nodules are found in the phaseoloid clade and some members of tribe Loteae (Sprent, 2009).

They develop from cells in the hypodermal region or outer cortex and the meristem ceases to divide at an early stage (Guinel, 2009; Szczyglowski et al., 1998). Because of this, the nodules are spherical in shape and the cells are all at a similar developmental stage (Gualtieri & Bisseling, 2000).

1.8.2.2 Formation and structure of intercellularly infected nodules

Determinate and indeterminate nodules are also found in the genistoid and dalbergioid legumes, in which infection threads are absent. The distinctive aeschynomenoid nodule, with determinate growth, a central mass of uniformly infected tissue and associated with a lateral or adventitious root, is synapomorphic for the dalbergioid clade (Lavin et al., 2001). In studied dalbergioid legumes, the meristematic zone arises in the cortex some distance from the infection site. The rhizobia penetrate to the cortex via an intercellular matrix (Arachis) or by the progressive collapse of invaded cells (Aeschynomene and Stylosanthes); infected meristematic cells then divide repeatedly to form the nodule (Alazard & Duhoux,

1990; Boogerd & van Rossum, 1997; Chandler et al., 1982).

In contrast, genistoid nodules have a persistent meristem and are thus indeterminate. The lupinoid nodule found in species of Lupinus and Listia is a

31 Chapter 1 variation on this, with lateral meristems forming a “collar” nodule that encircles the subtending root (Guinel, 2009; Sprent, 2009). Nodule primordia of the genistoid legume Cytisus proliferus develop in the inner cortex, and rhizobial dissemination through the cortex is accompanied by the collapse of host cells (Vega-Hernández et al., 2001). In contrast, the nodule primordium of Lupinus species originates from a single infected hypodermal or outer cortical host cell (González-Sama et al., 2004;

Tang et al., 1992).

It has been speculated that legume infection via a crack entry pathway is less specific than that of root hair infection (Boogerd & van Rossum, 1997; Sprent,

2007). Specificity is not always associated with a root hair curl infection, however, as demonstrated by promiscuous plants such as Phaseolus vulgaris (Laeremans &

Vanderleyden, 1998; Martínez-Romero, 2003). Legume specificity and a narrow rhizobial host range do seem to be features of the symbiosis between temperate legumes in the hologalegoid tribes Vicieae and Trifolieae and their cognate rhizobia

(Sprent, 2007). As specificity and effectiveness appear to be linked (Sprent, 2007) and are critical aspects in the assessment and development of legumes and inoculants in agricultural systems (Howieson et al., 2008; Sessitsch et al., 2002), it is important that these terms are defined and explained.

1.9 Specificity and effectiveness in the legume-rhizobia symbiosis

Both legume hosts and rhizobial microsymbionts vary in their symbiotic ability. Legumes in tribe Phaseoleae are known to be promiscuous; that is, able to nodulate with a broad spectrum of rhizobia (Perret et al., 2000). Conversely,

32 Chapter 1

Biserrula pelecinus is known only to nodulate with strains of Mesorhizobium

(Nandasena et al., 2007b; Nandasena et al., 2009). Some legume cultivars (of clover and soybean, for example) are more promiscuous than others, with important implications for agriculture (Drew & Ballard, 2010; Graham, 2008). Similarly, rhizobia such as the strains NGR234 and USDA257 have exceptionally broad host ranges (Pueppke & Broughton, 1999), while the narrow host range Azorhizobium caulinodans is compatible only with Sesbania species (Perret et al., 2000).

A fundamental aspect of the legume-rhizobia relationship is the effectiveness of the symbiosis, i.e. the amount of N2 fixed by the rhizobia and made available to the plant. A wide variation in the effectiveness of symbiotic interactions can exist

(Sprent, 2007; Thrall et al., 2000). Based on this variation, Howieson et al. (2005) have defined four categories of symbiotic interaction:

1. no symbiotic interaction, i.e. plants do not nodulate

2. an ineffective or parasitic interaction, where nodules form but there is

no N2 fixation

3. a partially effective symbiosis, where fixation produces 20–75% of

the biomass achieved by a nitrogen-fed control

4. an effective symbiosis, where nodulated plants produce > 75% of the

biomass achieved by a nitrogen-fed control

A greater understanding of the factors that govern specificity and effectiveness in legume-rhizobia symbioses will be required if agricultural systems are to provide the sustainable increases in productivity needed to cope with an increasing world

33 Chapter 1 population, higher nitrogen fertilizer prices and other pressures (Howieson et al.,

2008). This is illustrated by the approach used to develop exotic legumes for southern Australian agricultural systems, in response to environmental and economic factors (Howieson et al., 2000; Howieson et al., 2008). Several species in the

Lotononis s. l. clade were included in the list of exotic legumes targeted for evaluation in this approach, in particular several species in the genus Listia.

1.10 The Lotononis s. l. clade

The Lotononis s. l. clade is grouped within tribe Crotalarieae (and thus is in the genistoid clade), has a centre of origin in South Africa and consists of some 150 species, divided into 15 sections (Figure 1.6) (van Wyk, 1991). The taxonomy has recently been revised and the three distinct clades within Lotononis s. l. are now recognised at the generic level as Listia, Leobordea and Lotononis s. str. (Boatwright et al., 2011). A brief description of genera, sections and species relevant to this thesis is provided below (Figure 1.7), and is based on the synopsis of van Wyk (1991) and the revisions to the taxonomy given in Boatwright et al. (2011).

34 Chapter 1

uchlora ynclistus otononis Cleistogama Polylobium Listia Monocarpa E Krebsia L Lipozygis Leptis Leobordea S Aulacinthus Buchenroedera Oxydium Digitata

2

1 3

Figure 1.6. Cladogram for Lotononis s. l., showing the relationships of the different sections. The diagram is based on the cladogram featured in van Wyk (1991). The three numbered clades are now recognised at the generic level as (1) Listia, (2) Leobordea and (3) Lotononis s. str., according to the taxonomy of Boatwright et al. (2011). The monospecific Euchlora section is now strongly supported as being more closely related to Crotalaria and Bolusia than to Lotononis s. l. (Boatwright et al. (2008; 2011)). Sections relevant to this thesis are in blue font.

35 Chapter 1

Figure 1.7. Species of Lotononis s. l.: (a) Listia angolensis, (b) Listia bainesii (c) Leobordea longiflora, (d) Leobordea platycarpa, (e) Leobordea stipulosa, (f) Leobordea bolusii, (g) Leobordea polycephala, (h) Lotononis crumanina, (i) Lotononis delicata, (j) Lotononis falcata, (k) Lotononis laxa, (l) Lotononis pungens

1.10.1 The genus Listia

Listia consists of seven species of herbaceous perennials: L. angolensis, L. bainesii, L. heterophylla (previously Lotononis listii), L. marlothii, L. minima, L.

36 Chapter 1 solitudinis and L. subulata. Species in this section are of particular interest as pasture plants as they are perennial, lack the poisonous metabolites found in some other

Lotononis s. l. sections (van Wyk & Verdoorn, 1990) and are able to produce adventitious, almost stoloniferous roots on the lower branches. These adventitious roots are a property unique to Listia, and are thought to be associated with the seasonally wet habitats, such as ditches, riverbanks and dambos (shallow grassy wetlands) where these species are found. Nodulation has been studied in all species except the very rare L. minima, and the nodules found to be of the lupinoid form

(Yates et al., 2007; R. Yates unpublished data). L. angolensis has a tropical distribution, occurring in the uplands encircling the Zaire basin. The L. heterophylla distribution extends from South Africa into southern Central Africa. L. bainesii is native to Botswana, Mozambique (south), Namibia and South Africa, while the remaining species are endemic to South Africa.

1.10.2 The genus Leobordea

Leobordea comprises the sections Digitata, Lipozygis, Leptis, Leobordea and

Synclistus. The six species in the Digitata section all have woody perennial bases and are mostly found in a narrow distribution in the dry mountainous region of the north- western Cape Province. Leobordea longiflora (syn. Lotononis speciosa) has the longest flowers in the genus. Lipozygis section species are perennial suffrutescent pyrophytic herbs, restricted to summer rainfall grassland areas of the eastern parts of southern Africa. Leobordea foliosa occurs at high altitudes. Species in the Leptis section are perennial suffrutescent herbs, or shrublets, or annuals; with a disjunct distribution in central and southern Africa and in the Mediterranean region.

Leobordea calycina is endemic to southern Africa and widely distributed in the eastern and central interior. Leobordea mollis is found only in the western Cape.

37 Chapter 1

Species in the Leobordea section are all annuals that occur in the dry parts of southern Africa. The distribution of Leobordea platycarpa extends through eastern tropical Africa and Mauritania to Pakistan and the Cape Verde Islands, which makes it the most widespread species of the genus. Leobordea stipulosa is found in the

Transvaal, Zimbabwe and the southern border of Zambia. The group of species in the

Synclistus section differs from all other Lotononis s. l. sections in having heads of sessile flowers. They are described as annuals, but in the case of Leobordea bolusii and Leobordea polycephala are more likely to be short-lived herbaceous perennials, as the plants can be grown from cuttings (see Chapter 2, Materials and Methods).

The section is endemic to the Cape Province in South Africa.

1.10.3 The genus Lotononis s. str.

Lotononis s. str. includes the sections Oxydium, Monocarpa, Cleistogama,

Polylobium, Lotononis, Aulacinthus, Krebsia and Buchenroedera. The genus is chemically distinct from Leobordea and Listia in that its members are cyanogenic

(except for L. sect. Cleistogama) and accumulate macrocyclic pyrrolizidine alkaloids. The monotypic section Euchlora was formerly placed within this group, but recent studies have found Euchlora to be more closely related to Crotalaria and

Bolusia rather than Lotononis s. l., and to merit generic status (Boatwright et al.,

2008; Boatwright et al., 2011). Euchlora hirsuta (formerly Lotononis hirsuta), the sole species, is a geophytic herb with a north-western distribution in the Cape

Province. Although the GRIN database (http://www.ars-grin.gov/~sbmljw/cgi- bin/taxnodul.pl) lists E. hirsuta as nodulated, field observations of this species have so far found no evidence of nodulation (J. Howieson & R. Yates, pers. comm.).

Species in the Oxydium section are herbaceous annuals or perennials. Lotononis crumanina is a perennial that occurs on limestone or lime-rich soils in the central

38 Chapter 1 parts of southern Africa. Lotononis delicata is an annual; Lotononis falcata is a common, widespread annual in southern Africa and Lotononis laxa is the most common and widely distributed perennial species in Lotononis s. l., with a range extending along the east coast of southern Africa to Ethiopia. The two species in the

Cleistogama section are very common, short-lived perennials that can be found in most habitats. Their distribution extends into the north-eastern and eastern Cape, with the range of Lotononis pungens also extending into the Orange Free state.

1.11. The Lotononis s. l. -rhizobia symbiosis

Renewed interest in the potential of Lotononis s. l. species as perennial pasture plants in southern Australian agricultural systems has prompted recent research into the rhizobia that nodulate these legumes (Yates et al., 2007). Although the molecular studies aimed at identifying and characterising these rhizobia and their host specificity genes have just begun, previous research on Lotononis s. l. rhizobia was conducted in Australia and Africa some fifty years ago, as part of the development of

Listia bainesii for tropical and sub-tropical pastures (Bryan, 1961; Sandmann, 1970).

Data from this era, while unable to give a precise identification of Lotononis s. l. rhizobia, provide records of rhizobial phenotypes and cross-inoculation experiments.

Sandmann’s work in Zimbabwe showed that Lotononis s. l. species were nodulated by phenotypically diverse rhizobia and that the pigmented rhizobia isolated from Listia bainesii and Listia heterophylla formed a separate cross- inoculation group to isolates from Listia angolensis. Similar studies were conducted by the CSIRO in Queensland, Australia in the 1950s and 60s on various Lotononis s. l. species and cultivars (Eagles & Date, 1999) (Table 1.1).

39 Chapter 1

Table 1.1. The CB Collection accession list of Lotononis s. l. hosts, rhizobia isolated from these hosts and effectiveness of the rhizobia on Lotononis s. l. species (Eagles & Date, 1999). Data on ineffectiveness are not given in the cited reference. Original Host Strain Effective on Listia angolensis CB1297 L. angolensis Listia angolensis CB1298 L. angolensis Listia angolensis CB1299 L. angolensis Listia angolensis CB1321 L. angolensis Listia angolensis CB1322 L. angolensis Listia angolensis CB1323† L. angolensis, Leobordea mucronata Listia angolensis CB2406 L. angolensis Listia angolensis CB2645 L. angolensis Listia bainesii CB360 L. bainesii, Lisita heterophylla Listia bainesii CB376‡ L. bainesii, Lotononis laxa, Lotononis leptoloba Listia bainesii CB730 L. bainesii, L. heterophylla Listia bainesii CB1775 L. bainesii, L. heterophylla Listia bainesii CB1776 L. bainesii Listia bainesii CB1777 L. bainesii Listia bainesii CB1778 L. bainesii Listia bainesii CB1779 L. bainesii Listia bainesii CB2343 L. bainesii, L. heterophylla Listia bainesii CB2344 L. bainesii Listia heterophylla CB768 L. bainesii, L. heterophylla Leobordea calycina* CB2676 L. calycina Leobordea mucronata CB2695 L. mucronata Leobordea mucronata CB2705 L. mucronata Leobordea platycarpa CB2648 Lotononis laxa Leobordea stipulosa CB1394 L. stipulosa *Referred to as Lotononis orthorrhiza in the cited reference. †Published performance information lists this strain as effective on some accessions of Listia angolensis and most Leobordea mucronata and Lotononis laxa (Eagles & Date, 1999). ‡ Published performance information lists this strain as effective on Listia bainesii, Leobordea mucronata, Lotononis laxa and Lotononis leptoloba (Eagles & Date, 1999).

The CB Collection data on the host range and effectiveness of rhizobia isolated from and/or inoculated onto species of Listia (L. angolensis, L. bainesii and L. heterophylla), Leobordea (L. calycina, L. mucronata, L. platycarpa and L. stipulosa) and Lotononis s. l. (L. laxa and L. leptoloba) indicate that:

40 Chapter 1

1) the rhizobia were phenotypically diverse;

2) L. angolensis, L. bainesii and L. heterophylla were not nodulated by

isolates from other Lotononis s. l. species;

3) L. angolensis isolates and the isolates from L. bainesii and L.

heterophylla formed separate cross-inoculation groups.

Most of the literature on Lotononis s. l. rhizobia is based on studies conducted on Listia bainesii, which is recognised as a model of acute symbiotic specificity

(Pueppke & Broughton, 1999). Its microsymbionts were first formally described by

Norris (1958), who showed that L. bainesii seedlings inoculated with a wide range of rhizobia would only nodulate with their specific red-pigmented isolate, designated strain CB360. This rhizobial strain was also highly specific. In glasshouse trials of 31 legume species from 21 genera, it was able to nodulate only species in the tribes

Aeschynomeneae and Crotalarieae and was effective only on L. bainesii, although other Lotononis s. l. species were not tested (Norris, 1958). Strain CB376, also isolated from L. bainesii, has similarly been described as having an extremely narrow host range (Broughton et al., 1986). In Australia, CB360 was used as the commercial inoculant strain for L. bainesii from 1958 – 1963, when it was superseded by CB376

(Bullard et al., 2005). The colony colour, serological and symbiotic properties of both these strains remained stable over a 5 – 12 year period following their introduction in Queensland (Diatloff, 1977). These and other field trials have indicated that pigmented L. bainesii strains are able to persist in acidic, sandy, infertile soils (Diatloff, 1977; Yates et al., 2007).

41 Chapter 1

The pigmented L. bainesii rhizobial strains have since been identified as a species of Methylobacterium (Jaftha et al., 2002; Yates et al., 2007). The only other rhizobial Methylobacterium species described to date is the non-pigmented M. nodulans, which specifically nodulates species of Senegalese Crotalaria (Sy et al.,

2001). Methylobacteria are typically pink-pigmented facultative methylotrophs that are characterised by their ability to utilise methanol and other C1 compounds, as well as a variety of multicarbon substrates (Green, 1992; Lidstrom, 2006). The pigments of strain CB360 have been identified as carotenoids (Godfrey, 1972). Similarly, pigments in strain CB376 were found to be carotenoids (diapolycopenedioic acid glucosyl ester and diapolycopenedioic acid diglucosyl ester), with structures identical to those from Methylobacterium rhodinum (formerly Pseudomonas rhodos) (Britton et al., 2004; Kleinig & Broughton, 1982). Interestingly, bacterial pellets of CB360 extracted from nodules were not pigmented, indicating that carotenoid synthesis is suppressed in this environment (Godfrey, 1972). Carotenoid pigments are believed to play a role in protecting bacteria from ultraviolet (UV) radiation (Jacobs et al.,

2005). Law (1979) has noted, however, that while pigmented L. bainesii rhizobia are highly resistant to UV irradiation, non-pigmented mutants are also resistant to such treatment. In addition to carotenoid pigments, bacteriochlorophyll a has been found in the L. bainesii rhizobial strains 4-46 (incorrectly referred to as a synonym of

CB376; D. Fleischman, pers. comm.), 4-144 and xct14 (Fleischman & Kramer, 1998;

Giraud & Fleischman, 2004; Jaftha et al., 2002), which is consistent with the weakly phototrophic nature of other species of Methylobacterium (Garrity et al., 2005;

Hiraishi & Shimada, 2001). The sequenced genome of strain 4-46, along with other

Methylobacterium strains, is available at the USA Joint Genome Institute

(http://genome.jgi-psf.org/programs/bacteria-archaea/index.jsf).

42 Chapter 1

Strains of pigmented methylobacteria have now been isolated from, and are effective on, all studied Listia species, with the exception of L. angolensis, which forms ineffective nodules with these rhizobia (Yates et al., 2007; R. Yates, unpublished data). Surprisingly, a study of the methylobacterial isolates from L. bainesii, L. heterophylla and L. solitudinis showed that they are, uniquely for

Methylobacterium spp., unable to grow on methanol as a sole carbon source (Ardley et al., 2009). It has since been suggested that long term maintenance of C1 metabolism in methylobacteria requires relatively frequent use of C1 compounds to prevent the rapid loss of this trait (Lee et al., 2009).

Another unexpected finding was that the Listia angolensis microsymbionts, based on their 16S rRNA gene sequences, belong to a novel genus of rhizobia (Yates et al., 2007). Strains WSM3674 and WSM3686, isolated from L. angolensis, were subsequently shown to be closely related to rhizobia that specifically nodulate

Lupinus texensis plants endemic to Texas, USA (Andam & Parker, 2007). The 16S rRNA-based phylogenetic tree constructed by Andam & Parker (2007) indicated that the Listia angolensis and Lupinus texensis strains were most closely related to

Microvirga flocculans (previously Balneimonas flocculans) (Weon et al., 2010), a species described from a strain isolated from a Japanese hot spring (Takeda et al.,

2004). Currently, four other Microvirga species have been named and characterised:

M. subterranea (Kanso & Patel, 2003), M. guangxiensis (Zhang et al., 2009), M. aerophila and M. aerilata (Weon et al., 2010), isolated from Australian geothermal waters, Chinese rice field soil and Korean atmospheric samples (two strains), respectively.

43 Chapter 1

Published data on the nodulation record or symbiotic specificity of other species of Lotononis s. l. remain scant. Studies on isolates obtained in 2002 from nodules of South African Lotononis s. l. hosts established that the rhizobia isolated from species outside the genus Listia were non-pigmented, genotypically and phenotypically diverse and unable to nodulate either Listia bainesii or Listia heterophylla (Ardley, 2005; R. Yates, unpublished data).

1.12 Summary of current knowledge of the symbiotic relationships between Lotononis s. l. and associated rhizobia

A brief summary of previous studies of the Lotononis s. l. -rhizobia symbiosis indicates the following:

• Lotononis s. l. species are nodulated by diverse rhizobia, although this has not

been formally reported in the literature.

• The symbiosis between Listia bainesii and strains of pigmented

methylobacteria is both effective and highly specific. This effectiveness and

specificity extends to other studied Listia species, with the exception of L.

angolensis.

• L. angolensis can be ineffectively nodulated by pigmented methylobacteria,

but forms effective nodules only with bacteria that belong to a group of novel

and undescribed rhizobia.

There has as yet been no detailed examination of the level of specificity in

the L. angolensis symbiosis or in Lotononis s. l. species outside the genus Listia.

A first step would be to quantify the nodulation and N2 fixation capabilities of

44 Chapter 1

Lotononis s. l. isolates on host species from a range of Lotononis s. l. sections. To

that purpose, the availability of seeds collected from a number of South African

Lotononis s. l. species in 2007, together with a collection of phylogenetically

diverse Lotononis s. l. isolates, allowed the design of glasshouse trials to compare

the symbiotic interactions of different taxonomic groups of Lotononis s. l. hosts

and their cognate rhizobia.

Secondly, given the important role Nod factors play in determining specificity

(Section 1.7), does the phylogeny of the nod genes of Lotononis s. l. rhizobia provide a clue to the basis of specificity in this genus? Comparison of the phylogeny of the rhizobial chromosomal background with that of the nod genes can also help elucidate the evolution and biogeography of the symbiosis between native populations of

Lotononis s. l. host plants and their associated rhizobia.

Another aspect of symbiosis in Listia bainesii that warrants study is the mechanism of infection and nodule formation. This is of interest, given that:

1. Studies of other legume hosts in the genistoid clade show a crack or

epidermal, rather than a root hair curl, intercellular infection process

with no, or transient, development of infection threads (Section 1.8)

2. If infection in L. bainesii occurs by epidermal entry without IT

formation, it presents an example of unusual symbiotic specificity in a

legume host with a non-root-hair/IT-mediated infection process.

Finally, according to their 16S rRNA gene sequences, the Listia angolensis isolates belong to a group of novel and as yet uncharacterised rhizobia (Yates et al.,

45 Chapter 1

2007). This presented an opportunity to make a contribution to the literature on rhizobia by naming and describing these new symbiotic bacteria. Importantly, the availability of L. angolensis isolates from the WSM Culture Collection, together with a group of phylogenetically related Lupinus texensis isolates (Andam & Parker,

2007), allowed for a polyphasic description of this novel group of root nodule bacteria, based on a number of strains from different hosts and widely separate geographical areas. Professor Matt Parker, whose laboratory at the State University of New York (SUNY) was responsible for the work on the L. texensis isolates, was approached to be a part of this collaboration. Professor Anne Willems and her student Sofie De Meyer at the University of Ghent were included in the collaboration for their expertise in DNA: DNA hybridization, determination of G + C content and cellular fatty acid analysis, with the aim of publishing a valid name and description of the genus and species.

1.13 Aims of this thesis

Accordingly, the aims of this thesis are to:

1. Assess and compare the symbiotic specificity of Lotononis s. l. species and

their diverse rhizobial isolates in glasshouse trials, and determine the

phylogeny of the isolates’ chromosomal and symbiotic backgrounds.

2. Investigate the process of infection and nodule initiation and the nodule

morphology of Listia angolensis and Listia bainesii.

3. Perform a range of genotypic and phenotypic studies as part of the

requirements for validly naming and describing the novel Listia angolensis

and Lupinus texensis isolates.

46 Chapter 2

CHAPTER 2

Determining symbiotic specificity within Lotononis s. l.

47 Chapter 2

48 Chapter 2

2.1 Introduction

2.1.1 Background

Between 2002 and 2007, the Centre for Rhizobium Studies undertook collections of nodule isolates and seeds from a range of Lotononis s. l. species growing in diverse sites in South Africa. This collection of legume germplasm and rhizobia, together with Listia angolensis strains sourced from the CB Strain

Collection, allowed an examination of the nodulation and nitrogen fixation capabilities of rhizobia associated with Lotononis s. l. across a range of taxonomically different Lotononis s. l. hosts. Together with a molecular analysis of the rhizobial core and symbiosis genes, this study sought to answer the following questions:

1. What is the extent of nodulation and the nodule morphology found

within Lotononis s. l. species?

2. Is the symbiotic specificity of the genus Listia maintained?

3. How does the symbiotic specificity of L. angolensis compare with other

Listia species?

4. Are there patterns of rhizobial specificity in the other Lotononis s. l.

sections?

5. Given that this is a study of native legumes and microsymbionts

obtained from their centre of diversity, does this study shed light on the

mechanisms by which the evolution and diversification of legume

plants affects their relationship with their preferred microsymbionts?

49 Chapter 2

6. Do the diverse Lotononis s. l. rhizobia also possess diverse symbiotic

genes and is there a relationship between host and symbiotic gene

phylogeny?

2.1.2 Experimental approach

The work was divided into four phases:

1. Selection of appropriate inoculant strains and Lotononis s. l. hosts.

2. Design of a glasshouse trial that would determine the nodulation and

nitrogen-fixing capabilities of Lotononis s. l. species inoculated with

diverse Lotononis s. l. rhizobial isolates.

3. Additional glasshouse trials to examine further the degree of symbiotic

specificity within the genus Listia.

4. Sequencing of rhizobial genes.

The Lotononis s. l. isolates available for this study consisted of three groups of strains:

1. Sixty seven pink-pigmented methylobacterial isolates obtained from

nodules of South African Listia bainesii, Listia heterophylla and Listia

solitudinis and authenticated as effective nodulators of L. bainesii

(Ardley, (2005); R. Yates, unpublished data).

2. Seven non-pigmented genetically and phenotypically diverse isolates

from South African Leobordea and Lotononis s. str. species, i.e.

obtained from Lotononis s. l. species outside the genus Listia. Partial

16S rRNA sequences identified the isolates as species of

Bradyrhizobium, Ensifer (syn. Sinorhizobium), Mesorhizobium or

Methylobacterium (Ardley, 2005). They were unable to nodulate Listia

50 Chapter 2

bainesii or Listia heterophylla, but were authenticated as RNB after

nodulation tests on promiscuous host plants (Ardley, 2005).

3. Seven Listia angolensis strains (six that formed pale pink colonies and

one non-pigmented strain). These were derivatives of the original CB

strains, obtained by inoculating glasshouse grown L. angolensis plants

(accession No. 8639, ARC—LBD Animal Production Institute,

Pretoria, South Africa) with each strain and reisolating from the

nodules according to the methods of Vincent (1970) (Yates et al.,

2007). Two of these strains, WSM3674 and WSM3686, have been

identified as belonging to a novel genus of rhizobia (Yates et al., 2007).

Inoculant strains were selected from these three groups on the basis of primary host species, ability to nodulate at least one species of Lotononis s. l. and phenotypic and genotypic diversity. Additionally, where there were a number of strains from the same host (as in the Listia angolensis strains) or the same cross-inoculation group (as in the pigmented methylobacteria) a representative strain was chosen on the basis of its effectiveness on its original host plant.

Previous studies on the pink-pigmented methylobacteria indicated that strain

WSM2598 was highly effective on both Listia bainesii and Listia heterophylla and was able to ineffectively nodulate Listia angolensis (Yates et al., 2007). As data on the effectiveness of the L. angolensis strains was available for only two strains

(WSM3674 and WSM3686 (Yates et al., 2007)), a glasshouse trial was set up to determine the relative effectiveness of all L. angolensis strains on this host plant. The seven non-pigmented strains isolated from other Lotononis s. l. species had been

51 Chapter 2 authenticated as rhizobia, but in most cases it was not possible to authenticate them on their original hosts, due to a lack of available seed or to an inability to identify the host plant to species level. It was therefore important to confirm that these strains were able to nodulate Lotononis s. l. hosts. This was undertaken in a subsequent glasshouse trial that evaluated the ability of the strains to nodulate a taxonomically diverse range of Lotononis s. l. species.

The Lotononis s. l. hosts were selected similarly, according to their taxonomic group and the availability of germplasm. Where sufficient quantities of germplasm existed, Lotononis s. l. species were included in the large-scale nodulation and nitrogen fixation glasshouse experiment. Other Lotononis s. l. species were used to confirm that the non-pigmented Leobordea and Lotononis s. str. isolates were able to nodulate Lotononis s. l. hosts. Symbiotic specificity within the genus Listia was further examined by assessing the ability of a range of rhizobia, including type strains, to nodulate L. angolensis, L. bainesii and L. heterophylla.

The rhizobial strains inoculated onto the Lotononis s. l. hosts were identified by amplification and sequencing of a fragment of the 16S rRNA gene. The phylogeny of the symbiotic genes was assessed by amplifying and sequencing a fragment of the nodA gene. This gene was chosen as it is one of the genes that is required for synthesis of the Nod factor and is an important determinant of host range, due to the role of the Nod factor acyl tail in host specificity (Dénarié et al., 1996; Haukka et al.,

1998; Roche et al., 1996) (Section 1.6.1).

52 Chapter 2

2.2 Materials and Methods

2.2.1 Rhizobial strains

The Lotononis s. l. isolates used in this study were obtained from the Western

Australian Soil Microbiology (WSM) collection, housed at the Centre for Rhizobium

Studies, Murdoch University. The strains and the details of their collection sites are listed in Table 2.1. A map of southern Africa, showing site locations, is given in

Figure 2.1. Other rhizobial strains used in this chapter are listed in Table 2.2. All strains were stored at -80ºC as 12% (v/v) glycerol/media stocks. Strains were routinely subcultured on modified ½ lupin agar (½ LA) (Yates et al., 2007), or on ½

LA with succinate replacing glucose and mannitol as a carbon source, or on TY agar

(Beringer, 1974). Broth cultures were grown in TY or ½ LA. Plate and broth cultures were grown at 28ºC and shaking cultures were incubated on a gyratory shaker at 200 rpm.

2.2.2 Host plants

The Lotononis s. l. species used in the glasshouse experiments are shown in

Table 2.3, along with details of their taxonomic grouping. Seeds of L. angolensis, L. bainesii and L. heterophylla were obtained from the Department of Agriculture

Western Australia and cultivar names or line numbers for these species are also given in Table 2.3. Seed for all other Lotononis s. l. species was collected between 2002 and 2007 from wild plants growing at various sites in South Africa.

53 Chapter 2

Table 2.1. List of Lotononis s. l. isolates used in this chapter. Strain Affiliation* Host Lotononis Geographical Site Site Latitude Longitude GH Experiment Reference (Synonym) s. l. section source (Collector) No. pH 1 2 3 4 Pink-pigmented methylobacteria WSM2598 Methylobacterium Listia Listia South Africa 35 6.0 S29 01 540 E29 36 491 * * Yates et al. (2007) bainesii (Yates, Real & Law) Non-pigmented Lotononis s. l. strains WSM2596 Bradyrhizobium Leobordea Lipozygis South Africa 34 5.5 S29 04 414 E29 36 491 * * Ardley (2005) foliosa (Yates, Real & Law) WSM2624 Mesorhizobium Lotononis ND South Africa 20 7.0 S32 14 082 E22 55 081 * * Ardley (2005) s. l. sp. (Yates, Real & Law) WSM2632 Bradyrhizobium Lotononis ND South Africa 41 5.5 S28 38 902 E31 55 793 * * Ardley (2005) s. l. sp. (Yates, Real & Law) WSM2653 Ensifer Lotononis ND South Africa 22 8.0 S32 21 109 E22 36 127 * * Ardley (2005) s. l. sp. (Yates, Real & Law) WSM2667 Methylobacterium Leobordea Leptis South Africa 53 6.0 S25 35 432 E28 22 668 * * Ardley (2005) calycina (Yates, Real & Law) WSM2783 Bradyrhizobium Lotononis ND South Africa 49 6.5 S27 12 979 E31 10 635 * * Ardley (2005) s. l. sp. (Yates, Real & Law) WSM3040 Ensifer Lotononis Oxydium South Africa 14 6.5 S32 11 075 E23 55 631 * * Ardley (2005) s. str. laxa (Yates, Real & Law) Listia angolensis strains WSM3557 Microvirga† Listia Listia Chipata ‡, Zambia ND S 13.65 E 32.63 * * * Eagles & Date (1999) (CB1322) angolensis (Verboom) WSM3673 Listia Listia Chipata, Zambia ND S 13.65 E 32.63 * Eagles & Date (1999) (CB1321) angolensis (Verboom) WSM3674 Microvirga† Listia Listia Chipata, Zambia ND S 13.65 E 32.63 * Eagles & Date (1999) (CB1323) angolensis (Verboom) Yates et al. (2007) WSM3675 Listia Listia Mbale, Zambia ND S 08.51 E 31.20 * Eagles & Date (1999) (CB2406) angolensis (Verboom) WSM3686 Microvirga† Listia Listia Chipata, Zambia ND S 13.65 E 32.63 * Eagles & Date (1999) (CB1297) angolensis (Verboom) Yates et al. (2007) WSM3692 Listia Listia Chipata, Zambia ND S 13.65 E 32.63 * Eagles & Date (1999) (CB1299) angolensis (Verboom) WSM3693 Microvirga† Listia Listia Chipata, Zambia ND S 13.65 E 32.63 * Eagles & Date (1999) (CB1298) angolensis (Verboom) * As determined by Ardley (2005). † Determined from this study (see also Chapter 4). ‡ Formerly known as Fort Jameson. ND = not determined.

54 Chapter 2

L. angolensis strains

RNB strains

WSM2667

WSM2632

WSM2783

WSM2624 WSM2596

WSM2653 WSM3040 WSM2598

Figure 2.1. Geographical locations of southern African sites from which Lotononis s. l. strains were collected. Map image taken from: http://www.eecrg.uib.no/projects/AGS_BotanyExp/Drakensberg/Pictures/Biomes.jpg

2.2.3 Glasshouse experimental design

Four glasshouse experiments were conducted in this study:

Experiment 1 was an evaluation of the symbiotic abilities of the seven Listia angolensis strains on their L. angolensis host, with the aim of selecting the most effective strain for inclusion in Experiment 3. Each treatment consisted of three replicate pots, initially sown with six plants and thinned to four plants when the seedlings were three weeks old. Uninoculated nitrogen-free and supplied nitrogen

(N+) controls were included.

55 Chapter 2

Table 2.2. List of rhizobial strains used to inoculate Listia angolensis, Listia bainesii and Listia heterophylla (Experiment 4 in this chapter).

Strain Identification Original Geographical Reference/source host origin

ORS 571T Azorhizobium Sesbania Senegal Dreyfus et al. caulinodans rostrata (1988) USDA 76T Bradyrhizobium Glycine max USA Kuykendall et al. elkanii (1992) USDA 6T Bradyrhizobium Glycine max USA Jordan (1982) japonicum 2281T Bradyrhizobium Glycine max China Xu et al. (1995) liaongense WSM3937 Burkholderia sp. Rhynchosia South Africa Garau et al. (2009) ferulifolia WSM4187 Burkholderia sp. Lebeckia South Africa J. Howieson ambigua (Unpublished data) TTR 38T Ensifer arboris Prosopis Sudan Nick et al. (1999) chilensis NGR234 Ensifer fredii Lablab New Guinea Trinick (1980) purpureus WSM419 Ensifer medicae Medicago Sardinia Howieson & Ewing murex (1986) Sm1021 Ensifer meliloti Medicago Australia* Meade et al. (1982) sativa ORS 609T Ensifer saheli Sesbania Senegal De Lajudie et al. cannabina (1994) ORS 1009T Ensifer terangae Acacia laeta Senegal De Lajudie et al. (1994) ORS 2060T Methylobacterium Crotalaria Senegal Sy et al. (2001) nodulans podocarpa Lut6 Microvirga Lupinus USA Andam & Parker lupini † texensis (2007) ORS 992T Rhizobium Neptunia Senegal De Lajudie et al. (Allorhizobium) natans (1998b) undicola Control strains WSM2598 Methylobacterium Listia South Africa Yates et al. (2007) sp. bainesii WSM3557 Microvirga Listia Zambia This study lotononidis † angolensis T = Type strain * StreptomycinR derivative of the introduced wild-type; see Terpolilli (2009) † Determined from this study (see also Chapter 4)

56 Chapter 2

Table 2.3. List of host plants in this chapter and experiments in which they are used. Accession numbers or cultivars, where relevant, are given in brackets. Genus Species Glasshouse Experiment (Lotononis s. l. section)† (Accession No. /cultivar) 1 2 3 4 Listia Listia angolensis (8363) * * * Listia bainesii (Miles) * *

Listia heterophylla * (2004 CRSL69) Leobordea Leobordea longiflora * (Digitata) Leobordea Leobordea mollis * (Leptis) Leobordea Leobordea platycarpa * (Leobordea) Leobordea stipulosa * Leobordea Leobordea bolusii * (Synclistus) Leobordea polycephala * Lotononis s. str. Lotononis crumanina * (Oxydium) Lotononis delicata * Lotononis falcata * Lotononis laxa * Lotononis s. str. Lotononis pungens * (Cleistogama) Euchlora Euchlora hirsuta * †Based on the taxonomy of van Wyk (1991) and Boatwright et al. (2011).

Experiment 2 authenticated the seven strains isolated from Leobordea and

Lotononis s. str. species as able to nodulate Lotononis s. l. species hosts, as a prerequisite for their inclusion in Experiment 3. Because of the scarcity of germplasm, only one to four plants of each species were inoculated separately with each strain. Plants were assessed for nodulation only, which was deemed to be effective if the plants were green and the nodules pink in colour; an uninoculated nitrogen-free control was included.

Experiment 3 investigated the symbiotic interactions of strains of diverse

Lotononis s. l. rhizobia inoculated separately onto eight Lotononis s. l. species from a range of taxonomic groups. As in Experiment 1, nodulation and nitrogen-fixing

57 Chapter 2 capabilities were determined using three replicate pots per treatment. Each pot was sown with up to six germinated seeds (depending on seed availability) and thinned to four seedlings three weeks after planting, or was planted with four cuttings.

Uninoculated nitrogen-free and supplied nitrogen (N+) controls were included in the treatments. The exception to this experimental set-up was Leobordea stipulosa, where insufficient seedlings germinated to allow statistically valid replicates.

Experiment 4 was used to evaluate symbiotic specificity within the genus

Listia, especially in regard to the specificity of Listia angolensis as compared with other species in this section. Seedlings of L. angolensis, Listia bainesii and Listia heterophylla were grown in closed screw-topped polycarbonate vials (500 ml) and inoculated separately with the rhizobia shown in Table 2.2. Treatments were duplicated and an uninoculated control was used. Treatments were repeated, using a pot system rather than vials, for strains that formed nodules or nodule-like structures on the host.

2.2.4 General glasshouse procedures

All plants, in both pots and vials, were grown in the axenic sand culture system described in Howieson et al. (1995) and in Yates et al. (2004). Briefly, free-draining pots were lined with absorbent paper, filled with a 3:2 mix of yellow sand and washed river sand, moistened, and then sterilised by steam treatment or autoclaving.

Each pot was flushed twice with hot, sterile, deionised (DI) water to remove inorganic nitrogen. Sterile polyvinyl chloride tubes (25 mm diameter) with lids were inserted into the sand mix for supply of water and nutrients. Post-inoculation, the soil surface was covered with sterile alkathene beads (Figure 2.2a). The closed vial

58 Chapter 2

system consisted of mixed sand medium (400g) and DI H2O (50 ml) and was sterilised by autoclaving (Figure 2.2b).

Figure 2.2. Axenic sand culture system used in glasshouse experiments: a) pots and b) screw-topped vials. Scale bar = 4 cm.

All experiments were conducted in a naturally lit, controlled temperature

(maximum 24°C) glasshouse. The pots were watered as required with sterile DI water. Sterile nitrogen-free nutrient solution (20 ml) (Howieson et al., 1995) was supplied weekly to each pot. All vials had a single dose of nutrient solution (20 ml) added at the time of planting. Nitrogen-fed (N+) controls received 5 ml of 0.1 M

KNO3 per pot weekly.

2.2.4.1 Preparation of plant material

Seeds were lightly scarified, then surface sterilised by immersion in ethanol

(70% (v/v); 60 s) transferred to hypochlorite (4% (w/v); two min), then rinsed in six changes of sterile DI water. The seeds were germinated in the dark at room temperature on water agar (1.0%) plates and aseptically sown into the pots or vials when the radicles were 1 – 3 mm in length. For two Leobordea species, L. bolusii

59 Chapter 2 and L. polycephala, plant germplasm was obtained by taking tip cuttings from pot- grown plants maintained in the glasshouse. The cuttings (approximately 6 cm long) were stripped of their lower leaves and placed in DI water, then transferred to hypochlorite (1% (w/v); one min) and rinsed twice in sterile DI water. The cut end was then dipped in striking powder (active ingredient 4-indol-3-yl butyric acid

(0.3%)) and aseptically planted. Pots with cuttings were supplied as required with sterile DI water for the first three weeks, and then given a single dose (20 ml) of

-1 nutrient solution containing 0.25 g l KNO3. The rooted cuttings were subsequently supplied with sterile DI water and weekly doses of sterile nitrogen-free nutrient solution (20 ml). To maintain sterility, all pots were covered with plastic wrap until inoculation. Vials were sown with four germinated seeds per vial.

2.2.4.2 Preparation of inoculum

For Experiments 1, 2 and 3, inoculant strains were aseptically washed off agar plates with approximately 60 ml sterile sucrose solution (1% (w/v)) and resuspended in a sterile 100 ml screw-topped container. Plants were inoculated with 1 ml of this solution delivered by a sterile syringe to each seedling or cutting. Inocula for

Experiment 4 were prepared by resuspending five loopfuls of plate culture in 10 ml of sucrose solution. Inocula contained approximately 3.0 x 107 – 1.0 x 109 live cells per ml, determined from colony counts of media plates spread with a series of inocula dilutions. Pots and vials with germinated seeds were inoculated after the seedlings had emerged. This was from seven to ten days after sowing, due to the variable growth rates of the different Lotononis s. l. species. Cuttings were inoculated six weeks after planting. To avoid cross-contamination during inoculation, all plants except for Euchlora hirsuta (Experiment 2) were grouped by rhizobial strain. As only one E. hirsuta seed germinated, this seedling was inoculated with 1 ml of a mix

60 Chapter 2 of all rhizobial strains, prepared by resuspending two loopfuls of each plate culture into sucrose solution (10 ml).

2.2.4.3 Harvesting

Plants were harvested eight weeks post-inoculation for Experiment 1 and ten weeks post-inoculation for Experiments 2 and 3. Plants in Experiment 4 were harvested six weeks post-inoculation. For quantification of nitrogen fixation, the aboveground biomass was excised and dried at 60°C, then weighed. Nodules were assessed for colour, morphology, number and distribution on the root system. Where nodulation was scored, a scale of 1-10 was applied, using the system given in Fettell et al. (1997). Selected nodules were excised and fixed overnight at 4°C in 3% (v/v) glutaraldehyde in 25 mM phosphate buffer (pH 7.0) in preparation for nodule sectioning. Fixed material was washed in phosphate buffer, then dehydrated in a series of acetone solutions and infiltrated with Spurr’s resin for sectioning. Sections were stained with 1% (w/v) methylene blue and 1% (w/v) azur II and examined under an Olympus BX51 photomicroscope.

2.2.5 Statistics

General analyses of variance using a 5% least significant difference (LSD) were calculated on the data sets using GenStat 12® (Release 12.1, Lawes

Agricultural Trust, Rothamsted Experimental Station).

2.2.6 Molecular fingerprinting

Rhizobia were re-isolated from nodules and identified, where required, by PCR fingerprinting with ERIC primers (Versalovic et al., 1991) (Table 2.4).

61 Chapter 2

Table 2.4. Primers used in this study.

Primer Sequence Reference

ERIC

ERIC 1R 5'- ATGTAAGCTCCTGGGGATTCAC -3' Versalovic et al. (1991) ERIC 2F 5'- AAGTAAGTGACTGGGGTGAGCG -3' Versalovic et al. (1991) 16S rRNA

a FGPS6 5'- GGAGAGTTAGATCTTGGCTCAG -3' Normand et al. (1992) FGPS1509 5'- AAGGAGGGGATCCAGCCGCA -3' Normand et al. (1992) 420F 5’- GATGAAGGCCTTAGGGTTGT -3’ Yanagi & Yamasato (1993) 800F 5'- GTAGTCCACGCCGTAAACGA -3' Yanagi & Yamasato (1993) 1100F 5'- AAGTCCCGCAACGAGCGCAA -3' Yanagi & Yamasato (1993) 520R 5'- GCGGCTGCTGGCACGAAGTT -3' Yanagi & Yamasato (1993) 920R 5'- CCCCGTCAATTCCTTTGAGT -3' Yanagi & Yamasato (1993) 1190R 5'- GACGTCATCCCCACCTTCCT -3' Yanagi & Yamasato (1993) 16S-1924r 5′- GGCACGAAGTTAGCCGGGGC -3′ Sy et al. (2001) 16S-1080r 5′- GGGACTTAACCCAACATCT -3′ Sy et al. (2001) nodA

b M4-46 nodD 89-109f 5'- ATTGCGGGCTGGCTTAGGTTG -3' This study b M4-46 nodB 29-48r2 5'- CGCGCCAATGACACAGAACG -3' This study nodA-1 5’- TGCRGTGGAARNTRNNCTGGGAAA -3’ Haukka et al. (1998) nodA-2 5'- GGNCCGTCRTCRAAWGTCARGTA -3' Haukka et al. (1998) a Symbols: A, C, G, T – standard nucleotides; R – A, G; W – A, T; N – all. b The position of the primer in the corresponding sequence of the Methylobacterium sp. 4-46 target gene.

The ERIC primers were also used in PCR fingerprinting to confirm that the

Listia angolensis isolates were separate strains, rather than clones. DNA template was prepared from whole cells, using fresh plate culture resuspended in 0.89% (w/v)

NaCl to an OD600 of 6.0. Each PCR reaction was set up according to Table 2.5 and each set of reactions included a negative control, in which DNA template was replaced with an equal volume of sterile PCR-grade water. The thermal cycling conditions are shown in Table 2.6. The PCR was performed on an iCycler (Biorad).

Amplification products were visualized by agarose gel electrophoresis, using a 1.5%

(w/v) agarose in TAE gel, which contained 0.01% (v/v) SYBR® Safe DNA gel stain

62 Chapter 2

10,000X (Invitrogen), submerged in TAE running buffer (40 mM Tris acetate, 1 mM

EDTA, pH 8.0). A 5.0 µL aliquot of 6X loading dye (Promega) was added to each sample prior to electrophoresis and a 1kb DNA ladder (Promega) was used as a marker. The gels were run at 80V for 3.5 h, then visualised under UV light using the

BIORAD Gel Doc 2000 system.

Table 2.5. Reaction components for ERIC PCR amplification.

Component Volume (µl)

5 X Gitschier buffer 5.0

Bovine serum albumin (BSA) (20 mg/ml) 0.4 Dimethyl sulfoxide (DMSO) 2.5

Sterile PCR-grade H2O (Fisher Biotech) 12.35 dNTPs (25 mM) 1.25 ERIC 1R (50 µM) 1.0 ERIC 2F (50 µM) 1.0 Taq DNA polymerase (Invitrogen) 0.2

DNA template (whole cells, OD600nm = 6.0) 1.0 25 0

Table 2.6. Thermal cycling conditions for ERIC PCR amplification.

Temperature (ºC) Time No. of cycles

94 7 min 1 94 30 s 50 1 min 35 65 8 min 65 16 min 1 14 ∞ 1

2.2.7 Amplification and sequencing of 16S rRNA genes

Nearly full length PCR amplification of the 16S rRNA gene was performed using the universal eubacterial primers FGPS6 and FGPS1509 (Nesme et al., 1995;

Normand et al., 1992) (Table 2.4). DNA template was prepared as for ERIC PCRs, 63 Chapter 2

but with an OD600nm of 2.0. The optimised PCR reaction and thermal cycling conditions are shown in Table 2.7 and Table 2.8 respectively. The PCR amplicons were purified directly, using a QIAquickTM PCR Purification Kit (Qiagen), or with a

QIAquickTM Gel Extraction Kit (Qiagen) following electrophoresis in a 1% (w/v) agarose gel and excision of the amplified gene products.

Table 2.7. Reaction components for PCR amplification of the 16S rRNA gene.

Component Volume (µl)

5x PCR Polymerisation Buffer (Fisher Biotech) 5.0

MgCl2 (15 mM) 2.5

Sterile PCR-grade H2O (Fisher Biotech) 15.3 Forward primer FGPS6 (50 µM) 0.5 Reverse primer FGPS1509 (50 µM) 0.5 Taq DNA polymerase (Invitrogen) 0.2

DNA template (whole cells, OD600nm = 2.0) 1.0 25 0

Table 2.8. Thermal cycling conditions for 16S rDNA PCR amplification.

Temperature (ºC) Time No. of cycles

94 5 min 1 94 30 s 55 30 s 35 70 1 min 72 7 min 1 14 ∞ 1

Sequencing PCRs were performed on purified 16S rDNA in a 96-well microplate, using FGPS6 and FGPS1509 and internal primers designed by Yanagi &

Yamasato (1993) and Sy et al. (2001) (Table 2.4) and the BigDye Terminator 3.1 mix (Applied Biosystems). The sequencing PCR reaction and the thermal cycling parameters are shown in Table 2.9 and Table 2.10, respectively. Sequencing

64 Chapter 2 reactions were purified by an ethanol/EDTA and sodium acetate precipitation, as recommended by Applied Biosystems. Sequence reads were obtained from the ABI model 377A automated sequencer (Applied Biosystems). Sequences were manually edited and aligned using Genetool Lite (version 1.0; Double Twist Inc., Oakland,

CA, USA). Searches for sequences with high sequence similarity to the sample 16S rDNA were conducted using BLASTN (Altschul et al., 1990) against sequences deposited in the National Centre for Biotechnology Information GenBank database.

A phylogenetic tree was constructed with MEGA version 4.0 (Tamura et al., 2007), using the neighbour-joining method (Saitou & Nei, 1987) and the Maximum

Composite Likelihood model, and bootstrapped with 1000 replicates. For comparison, trees were also constructed using the minimum evolution (ME) and maximum parsimony (MP) methods.

Table 2.9. Reaction components for PCR sequencing of 16S rDNA.

Component Volume (µl)

Purified DNA template (10-40 ng) 4 BigDye Terminator 3.1 mix 4

Sterile PCR-grade H2O (Fisher Biotech) 1 Primer (10 µM) 1 10

Table 2.10. Thermal cycling conditions for 16S rDNA sequencing PCR.

Temperature (ºC) Time No. of cycles

96 2 min 1 96 10 s 50 5 s 25 60 4 min 14 ∞ 1

65 Chapter 2

2.2.8 Amplification and sequencing of nodA genes

Amplification of the nodA gene of Lotononis s. l. rhizobia was initially attempted by designing primers based on conserved regions of the nodD and nodB genes in the Listia bainesii microsymbiont Methylobacterium sp. 4-46. The sequenced genome of this strain is available on the USA Joint Genome Institute’s

Microbial Genomics Program database (http://genome.jgi-psf.org/programs/bacteria- archaea/index.jsf). The Methylobacterium sp. 4-46 nodD and nodB gene sequences were aligned and compared with those of other rhizobia that displayed high sequence identity to the target genes, based on searches conducted using BLASTN (Altschul et al., 1990) against sequences deposited in GenBank (Appendix I).

Four primers were designed from these alignments. Combinations of primers were tested: the primer set, PCR reaction components and cycling conditions that gave optimum results are shown in Table 2.4, Table 2.11 and Table 2.12 respectively. Forward primer M4-46 nodD 89-109f starts from the 5’ end of nodD

(codon 4271507 in the Methylobacterium sp. 4-46 genome sequence) and reverse primer M4-46 nodB 29-48r2 ends in the 5’ end of nodB (codon 4270473). They amplify a product of approximately 1000 bp. As these primers yielded appropriate products with only two of the rhizobial strains, the nodA primers developed by

Haukka et al. (1998) were also trialled (Table 2.4). The optimised cycling conditions for these primers were modified slightly from the original conditions cited by

Haukka et al. (1998) and are shown in Table 2.13. Purification and sequencing of the nodA PCR amplicons was as described for the 16S rRNA gene, but with the use of the requisite nodA primers.

66 Chapter 2

Table 2.11. Reaction components for PCR amplification of nodA.

Component Volume (µl)

5x PCR Polymerisation Buffer (Fisher Biotech) 5

Mg Cl2 (10 mM) 2.5

Sterile PCR-grade H2O (Fisher Biotech) 15.3 Forward primer (50 µM) 0.5 Reverse primer (50 µM) 0.5 Taq DNA polymerase (Invitrogen) 0.2

DNA template (whole cells, OD600nm = 2.0) 1.0 25 0

Table 2.12. Thermal cycling conditions for nodA PCR amplification, using primers based on the Methylobacterium sp. 4-46 sequenced genome.

Temperature (ºC) Time No. of cycles

94 4 min 1 94 30 s 52 30 s 35 70 30 s 72 5 min 1 14 ∞ 1

Table 2.13. Thermal cycling conditions for nodA PCR amplification, using primers developed by Haukka et al. (1998).

Temperature (ºC) Time No. of cycles

94 4 min 1 94 45 s 55 45 s 35 68 2 min 70 5 min 1 14 ∞ 1

BLASTN and BLASTX (Altschul et al., 1990) were used to conduct searches for sequences with high sequence similarity to the sample nodA sequences against sequences deposited in the National Centre for Biotechnology Information GenBank database. A protein-coding phylogenetic tree was constructed with MEGA version

67 Chapter 2

4.0 (Tamura et al., 2007), using the neighbour-joining method (Saitou & Nei, 1987) and the Maximum Composite Likelihood model, and bootstrapped with 1000 replicates. Trees were also constructed using the minimum evolution and maximum parsimony methods.

2.3 Results

2.3.1 Selection of inoculant strains

Glasshouse trials were used to select the most symbiotically effective Listia angolensis strain and to confirm that the strains isolated from Leobordea and

Lotononis s. str. species were able to nodulate Lotononis s. l. species.

2.3.1.1 Selection of a representative Listia angolensis strain

The symbiotic effectiveness of the seven L. angolensis strains on their original host was assessed through visual observation of plant appearance and nodulation, and by measurement of nodule numbers and the dry weight of plant shoots (Figure

2.3). Plants inoculated with all strains except WSM3692 were green and had increased biomass compared with the uninoculated N-free control. Plants treated with WSM3692 were small and pale, with biomass similar to the uninoculated control. All plants, other than the controls, were nodulated. WSM3692 induced a large number of nodules, but plants received a low nodulation score, as nodules were very small and white, with most occurring on the lateral roots. Nodules induced by the other strains were pink, larger in size and distributed more on the hypocotyl and taproot. On the basis of these results, WSM3692 was considered to be ineffective for nitrogen fixation on this accession of L. angolensis. Effectiveness, as measured by the dry weight of shoots, varied in the other strains, with some being only partially effective. WSM3557 was chosen as the most symbiotically competent strain, as it

68 Chapter 2 produced both the greatest biomass and the highest nodulation score on the L. angolensis host. A fingerprinting PCR, using ERIC primers, confirmed that the N2- fixing L. angolensis rhizobia were separate strains, rather than clones (Appendix II).

10 0.08 9 0.07 a 8 0.06 7

0.05 6

5 0.04

4 0.03

3 b b 0.02 (g/plant) weight Shoot dry 2 b Mean nodules/nodule score per plant per score nodules/nodule Mean c 0.01 1 c c c c 0 0 WSM3557 WSM3673 WSM3674 WSM3675 WSM3686 WSM3692 WSM3693 Uninoc N+ Inoculation treatments Figure 2.3. Symbiotic ability of Listia angolensis strains on L. angolensis, assessed by nodule number ( ), nodulation score ( ) and dry weight of shoots ( ) of plants harvested after eight weeks growth. For shoot dry weight, treatments which share a letter are not significantly different according to Fisher’s LSD test (P<0.05).

2.3.1.2 Authentication of non-pigmented Lotononis s. l. strains

The Leobordea and Lotononis s. str. rhizobia were assessed for their ability to nodulate a range of Lotononis s. l. hosts (Table 2.14). All strains were authenticated as able to nodulate at least one species of Lotononis s. l. and were therefore included in the study of the symbiotic interactions of diverse strains of Lotononis s. l. rhizobia with Lotononis s. l. species from a range of taxonomic groups (Experiment 3).

Interestingly, the Lotononis laxa isolate WSM3040 was unable to nodulate this host.

All host species in this experiment, with the exception of E. hirsuta, formed nodules with at least one inoculant strain.

69 Chapter 2

Table 2.14. Ability of non-pigmented strains isolated from Leobordea and Lotononis s. str. species to nodulate a range of Lotononis s. l. and Euchlora hosts. Nodulation is in bold type. Isolate Original host Nodulation of Lotononis s. l. (section) host

Leobordea Leobordea Leobordea Lotononis Lotononis Euchlora mollis polycephala pungens delicata laxa hirsuta† (Leptis) (Synclistus) (Cleistogama) (Oxydium) (Oxydium) (Euchlora) Bradyrhizobium WSM2596 Leobordea foliosa N- N+F+ N- N+F- N- N- (Lipozygis) WSM2632 Lotononis s. l. sp. ND N+F+ N+F- N+F- ND N- WSM2783 Lotononis s. l. sp. N- N+F+ N+F- N+F+ N- N- Ensifer WSM2653 Lotononis s. l. sp. N+F- N- N+F+ N- N+F- N- WSM3040 Lotononis laxa N+F- N- N+F+ N- N- N- (Oxydium) Mesorhizobium WSM2624 Lotononis s. l. sp. N+F- N- N- N- N- N- Methylobacterium WSM2667 Leobordea calycina N+F+ N- N- N+F- N- N- (Leptis) N+F+ = nodulation and nitrogen fixation; N+F- = nodulation, but no nitrogen fixation; N- = no nodulation, ND = not determined. † The E. hirsuta seedling was inoculated with a mix of all the RNB in this Table, plus WSM2598 (isolated from and effective on Listia bainesii) and WSM3557 (isolated from and effective on Listia angolensis).

70 Chapter 2

Although the nodules of the Lotononis s. l. species varied in size and shape, all were indeterminate in structure (Figure 2.4). Molecular fingerprinting, using ERIC primers, confirmed nodule occupancy by the inoculant strain (Appendix II).

Figure 2.4. Nodulation in Lotononis s. l. spp.. Scale bar = 5 mm. a) ineffective nodules on Leobordea mollis inoculated with WSM2624; b) effective nodules on Leobordea polycephala inoculated with WSM2632; c) effective nodules on Lotononis delicata inoculated with WSM2783; d) ineffective nodules on Lotononis laxa inoculated with WSM2653; e) effective nodules on Leobordea pungens inoculated with WSM2653.

2.3.2 Symbiotic interactions of Lotononis s. l. species with phylogenetically diverse Lotononis s. l. rhizobia

The symbiotic relationships between Lotononis s. l. species and the rhizobia that nodulate them are summarised in Table 2.15. Figures for the nodulation and N2- fixation of rhizobial strains on the individual Lotononis s. l. species are given in

Appendix III (Figures 1 - 8). A notable feature of these symbioses was the extreme symbiotic specificity exhibited by Listia bainesii (Figure 2, Appendix III). This host species was nodulated only by the pigmented Methylobacterium strain WSM2598.

The symbiosis was highly effective, producing plant biomass equivalent to that

71 Chapter 2 obtained for the N+ control (Figure 2, Appendix III). Listia angolensis was less specific, as host plants consistently, or occasionally, formed ineffective nodules with the methylobacterial strains WSM2598 and WSM2667, respectively. Effective nodulation, however, was observed only with the novel rhizobial strain WSM3557

(Table 2.15).

The nodulation ability and effectiveness of Lotononis s.l.-associated rhizobia on Leobordea and Lotononis s. str. species is shown in Appendix III (Figures 3 – 8) and Table 2.15. In some species, the response of plants to the N+ treatment has been much greater than their response to rhizobial inoculation, and in these cases a rescaled figure that omits the N+ treatment has also been included, to reveal partial rhizobial effectiveness. Statistical analysis has not been performed for the Leobordea stipulosa results, as apart from the WSM2598 and WSM3557 treatments, too few plants were available to provide valid data. No uninoculated control data are included for this species, as the single plant that was available for this treatment died.

Many of the Leobordea and Lotononis s. str. species could be considered promiscuous, as they were nodulated by a range of Lotononis s. l. rhizobia. There did not appear to be any obvious taxonomically based pattern of symbiotic specificity in these species. Leobordea platycarpa, for example, was comparatively specific, being nodulated by only three strains (WSM2653 and WSM3040 (Ensifer spp.) and

WSM3557), yet Leobordea stipulosa, in the same taxonomic section as L. platycarpa

(Leobordea), was nodulated by all inoculants except the two Ensifer strains.

Nodulation was seldom effective.

72 Chapter 2

Table 2.15. Summary of nodulation and fixation ability of Lotononis s. l. species inoculated with rhizobia isolated from Lotononis s. l. species. Strain

Lotononis s. l. Lotononis s. l. Section* sp. Bradyrhizobium Ensifer Mesorhizobium Methylobacterium Microvirga WSM WSM WSM WSM WSM WSM WSM WSM WSM 2596 2632 2783 2653 3040 2624 2598 2667 3557 Listia Listia. angolensis N- N- N- N- N- N- N+F- N+/-F- N+F+

Listia bainesii N- N- N- N- N- N- N+F+ N- N-

Digitata Leobordea N+/-F- N+/-F- N+F- N+/-F- N- N+/-F- N+F- N+F- N+F- longiflora

Leobordea Leobordea N- N- N- N+F+ N+F- N- N- N- N+F+ platycarpa

Leobordea N+F- N+F- N+F- N- N- N+F- N+F- N+F- N+F- stipulosa

Synclistus Leobordea bolusii N+/-F+/- N+F+ N+F+ N- N+/-F- N+/-F- N+F- N+/-F- N+/-F+/-

Oxydium Lotononis N+/-F+/- N+/-F- N+/-F- N+F+ N+F+ N+/-F- N- N- N+F+/- crumanina

Lotononis falcata N+F- N+F- N+/-F- N- N+/-F- N- N- N- N+F- * according to van Wyk (1991)

73 Chapter 2

Nodulation of the Leobordea and Lotononis s. str. species was seldom effective. This was a not unexpected finding, given that none of these species was an original host of the inoculant strains used in this experiment. Moreover, the effectiveness was in most cases only partial as plants were green, but biomass was much smaller than for the N+ control. There was also variation in the response of individual plants both to inorganic nitrogen and to inoculation, as evidenced by the large error bars seen for some treatments. This is illustrated in the case of WSM2653 on Leobordea platycarpa, where the plant response ranged from ineffective to partially effective and effective nodulation (Figure 2.5).

Figure 2.5. Inoculant response of Leobordea platycarpa plants to WSM2653 (10 weeks post-inoculation): a) effective nodulation; b) ineffective nodulation; c) partially effective nodulation

Similarly, WSM2596 on Leobordea bolusii and WSM3557 on L. platycarpa were effective or partially effective on some plants but failed to nodulate others. All inoculant strains were able to nodulate at least two Lotononis s. l. species (Table

2.15). The strain with the broadest host range was the Listia angolensis isolate

74 Chapter 2

WSM3557, which was able to nodulate all Lotononis s. l. species except for Listia bainesii and was partially effective on Leobordea bolusii, Leobordea platycarpa and

Lotononis crumanina. Nodule occupancy was confirmed in most cases, as inoculant strains were successfully reisolated from harvested nodules. Reisolates were not obtained from nodules of Leobordea stipulosa inoculated with WSM2596,

WSM2624, WSM2632 and WSM2783 or Lotononis falcata inoculated with

WSM2632 or WSM3557. Additionally, reisolates were not obtained from nodules of

Leobordea longiflora and Lotononis crumanina inoculated with WSM2624.

2.3.3 Symbiotic specificity within Listia species

The symbiotic specificity of L. angolensis, L. bainesii and L. heterophylla was evaluated by inoculating seedlings of these species with rhizobia of a diversity of genera and species (Table 2.2). L. bainesii and L. heterophylla were nodulated only by the pink-pigmented methylobacterial strain WSM2598, and nodulation was always effective. L. angolensis was slightly less specific, being nodulated effectively only by WSM3557, but forming ineffective nodules consistently with WSM2598 and occasionally with Methylobacterium nodulans strain ORS 2060.

2.3.4 Nodule morphology in Lotononis s. l. species

Nodule morphology clearly differentiated Listia species from the other

Lotononis s. l. taxa. Both Listia angolensis and Listia bainesii formed lupinoid nodules, primarily on the hypocotyl and taproot (Figure 2.6 a and b), whereas nodules of Leobordea and Lotononis s. str. species were indeterminate and distributed throughout the root system (Figure 2.6 c – h).

75 Chapter 2

Figure 2. 6. Nodule morphology in Lotononis s. l. spp. (inoculant strain in brackets). Scale bar is 10 mm, except for 2.15g and inserts. a) Listia angolensis (WSM3557); b) Listia bainesii (WSM2598); c) Leobordea longiflora (WSM2598); d) Leobordea bolusii (WSM2783) e) Lotononis crumanina (WSM2596); f) Lotononis falcata (WSM3557) (scale bar in insert = 2 mm; g) Leobordea platycarpa (WSM2653) (scale bar = 5 mm; scale bar in insert = 3 mm); h) Leobordea stipulosa (WSM2632) (scale bar in insert = 2 mm).

76 Chapter 2

2.3.5 Amplification and sequencing of 16S rRNA genes

The 16S rRNA gene of the pink-pigmented Methylobacterium strain

WSM2598 had previously been sequenced (Yates et al., 2007). Nearly full-length portions of the 16S rRNA gene were amplified and sequenced for the remaining eight Lotononis s. l. strains.

BLASTN sequence identity searches confirmed that the Lotononis s. l. strains were phylogenetically diverse. Strains WSM2596, WSM2632 and WSM2783 were most closely related to Bradyrhizobium, WSM2653 and WSM3040 were identified as Ensifer strains, WSM2624 was related to Mesorhizobium and WSM2667 was most closely related to Methylobacterium nodulans. The Listia angolensis strain

WSM3557 was not related to any currently described rhizobial genus, but instead grouped with the other L. angolensis strains WSM3674 and WSM3686, and with root nodule bacteria isolated from Lupinus texensis (Andam & Parker, 2007). The highest sequence identity to a validly named microbial species was to Microvirga

(formerly Balneimonas) flocculans (Takeda et al., 2004; Weon et al., 2010).

Analysis of a 1381 bp segment of the 16S rRNA gene from the bradyrhizobial strains showed that WSM2632 and WSM2783 had the same sequence and shared

99.1% sequence identity with WSM2596 (Table 2.16). The sequence identity of

WSM2596 with the type strains of Bradyrhizobium elkanii and Bradyrhizobium japonicum was 99.4% and 98.8%, respectively, while WSM2632 and WSM2783 had

98.7% and 98.0% sequence identity with these type strains (Table 2.16).

77 Chapter 2

Table 2.16. Pairwise percent identity and number of nucleotide mismatches (in brackets) for a 1381 bp fragment of the 16S rRNA gene of Lotononis s. l. Bradyrhizobium strains and type strains of Bradyrhizobium elkanii (USDA 76T) and Bradyrhizobium japonicum (USDA 6T). WSM2596 WSM2632 WSM2783 USDA 76T USDA 6T 99.1 99.1 99.4 98.8 WSM2596 (15) (15) (11) (20) 99.1 100 98.7 98.0 WSM2632 (15) (0) (19) (36) 99.1 100 98.7 98.0 WSM2783 (15) (0) (19) (36) T 99.4 98.7 98.7 98.0 USDA 76 (11) (19) (19) (36) T 98.8 98.0 98.0 98.0 USDA 6 (20) (36) (36) (36)

The current phylogeny of Bradyrhizobium divides this genus into two groups.

The first group includes B. japonicum, B. canariense, B. liaoningense and B. yuanmingense strains and the second comprises strains related to B. elkanii, B. pachyrhizi and B. jicamae (Menna et al., 2009). The NJ phylogenetic tree grouped

WSM2596, WSM2632 and WSM2783, with high bootstrap support, in this second clade (Figure 2. 7). Phylogenetic trees constructed using MP or ME methods gave similar topologies. The closest relatives to WSM2596 on the basis of 16S rRNA sequence identity were a group of strains (ApE4.8, LcCT6, aeky10 and jws91-2) isolated from diverse desmodioid and phaseoloid wild legumes from the USA and

Japan (Parker & Kennedy, 2006; Qian et al., 2003). In contrast, WSM2632 and

WSM2783 were grouped, in a well-supported clade, with rhizobia isolated from diverse tropical or sub-tropical wild hosts (strains Ai1a-2, Ai4.2 and Cp5-3, from

Andira inermis (Dalbergieae) and Centrosema pubescens (Phaseoleae) in Costa Rica and Barro Colorado Island, Panama (Parker, 2003; 2004; 2008); and strain ARRI 218

(genospecies Y) from Indigofera linifolia, growing in Kakadu National Park, in northern Australia (Lafay & Burdon, 2007).

78 Chapter 2

B. sp. ApE4.8 (DQ354611) Apios B. sp. aeky10 (AF417051) Amphicarpaea 65 B. sp. jws91-2 (AF514700) Amphicarpaea B. sp. LcCT6 (DQ354610) Lespedeza

WSM2596 Lotononis s. l. 53 B. sp. onl92.6 (DQ354612) Amphicarpaea B. sp. SEMIA 621 (FJ390908) Kummerowia

56 B. sp. ORS204 (AF514796) Pterocarpus 89 B. sp. R5 (EU364711) Vigna 96 B. elkanii USDA 76T (U35000) Glycine 87 B. pachyrhizi PAC48T (AY624135) Pachyrhizus

B. jicamae PAC68T (AY624134) Pachyrhizus 85 B. sp. RSB6 (FJ514039) Retama B. sp. ARRI 218 (AJ785291) Indigofera B. sp. Cp5-3 (AY187548) Centrosema 96 B. sp. Ai4.2 (EF569650) Andira 66 WSM2632 Lotononis s. l. 93 WSM2783 Lotononis s. l.

B. sp. Ai1a-2 (AF514703) Andira

77 B. sp. RC2d (EU364716) Vigna B. iriomotense EK05T (AB300992) Entada B. canariense BTA-1T (AJ558025) Chamaecytisus 99 B. japonicum USDA 6T (U69638) Glycine

60 B. sp. WSM2793 (EU481826) Rhynchosia B. liaoningense 2281T (AF208513) Glycine

B. sp. GC1d (EU364703) Vigna

65 B. sp. R10 (EU364714) Vigna 90 B. yuanmingense B071T (AF193818) Lespedeza A. caulinodans ORS 571T (D11342)

0.01

Figure 2.7. NJ phylogenetic tree, bootstrapped with 1000 replicates, showing the relationships of Bradyrhizobium strains associated with Lotononis s. l. (in bold type) with related strains and type strains of Bradyrhizobium, based on aligned sequences of the 16S rRNA genes. GenBank accession numbers are in brackets. Also shown is the original host. Geographical origin is indicated if the strain is African ( ) or South African ( ). Scale bar, 1% sequence divergence (one substitution per 100 nucleotides). A., Azorhizobium; B., Bradyrhizobium. T = type strain.

79 Chapter 2

The fast-growing strains WSM2653 and WSM3040 had identical (1383 bp fragment) 16S rRNA gene sequences. The NJ phylogenetic tree grouped these strains with rhizobia closely related to Ensifer meliloti (Figure 2. 8). They shared 100% sequence identity with Ensifer strains isolated from Lotus species and Phaseolus vulgaris from the Canary Islands (León-Barrios et al., 2009; Zurdo-Piñeiro et al.,

2009), Prosopis alba in northern Spain (Iglesias et al., 2007) and Canadian

Medicago sativa (Bromfield et al., 2010) (strains Lma-x, Lse-2, GVPV12, RPA13 and T15, respectively (Figure 2.8)). Strains from Tunisian Argyrolobium uniflorum

(STM4038) and Genista saharae (STM4028) (Mahdhi et al., 2007; 2008) were also closely related. WSM2653 and WSM3040 shared sequence identities of 99.7% (5 bp mismatches), 99.7% (4 bp mismatches), 99.7% (5 bp mismatches), and 98.5% (21 bp mismatches) with the type strains of E. meliloti, Ensifer medicae, Ensifer numidicus and Ensifer garamanticus, respectively. The WSM2653 and WSM3040 16S rRNA gene sequences contained the specific primer sequence that allows differentiation of

E. meliloti from E. medicae (Garau et al., 2005).

WSM2624 was most closely related to strains of Mesorhizobium tianshanense

(Figure 2.8). A 1452 bp fragment of the 16S rRNA gene had sequence identity of

99.8% (2 bp mismatches) with the M. tianshanense type strain A-1BST, isolated from diverse native legumes in arid north-western China (Chen et al., 1995; Jarvis et al.,

1997), and 99.6% (6 bp mismatches) with RCAN03, an effective nodulant of chickpeas (Cicer arietinum) in northern Spain (Rivas et al., 2007). WSM2624 was also closely related to the strains LBER1 and Ala-3 that effectively nodulate the endemic Canary Islands legumes Lotus berthelotii (Lorite et al., 2010) and Anagyris latifolia (Donate-Correa et al., 2007), respectively.

80 Chapter 2

E. sp. STM 4028 (EF100524) Genista E. meliloti T15 (EU928877) Medicago WSM3040 Lotononis s. l. E. meliloti RPA13 (EF201802) Prosopis 59 E. sp. STM 4O38 (EF152477) Argyrolobium E. meliloti Lma-x (FM178847) Lotus E. meliloti Lse-2 (FM178848) Lotus 61 E. sp. GVPV12 (FJ462794) Phaseolus WSM2653 Lotononis s. l. E. meliloti WSM1022 (EU885227) Medicago 62 E. meliloti WSM1701 (AF533684) Lotus E. meliloti IAM 12611T (D14509) Medicago T 79 E. medicae A 321 (L39882) Medicago E. numidicus ORS 1407T (AY500254) Argyrolobium 60 E. arboris HAMBI 1552T (Z78204) Prosopis E. garamanticus ORS 1400T (AY500255) Argyrolobium 100 100 E. sp. STM 4015 (EF100514) Genista E. kummerowiae CCBAU 71714T (AY034028) Kummerowia E. fredii LMG 6217T (X67231)Glycine 67 E. saheli LMG 7837T (X68390) Sesbania 52 E. kostiensis LMG 19227T (AM181748) Acacia E. terangae LMG 7834T (X68388) Acacia E. adhaerens LMG 20216T (AM181733) 99 E. adhaerens LMG 9954 (AM181735) Leucaena M. alhagi CCNWXJ12-2T (EU169578) Alhagi T 99 M. loti ATCC 33669 (D14514) Lotus T 97 M. ciceri UPM-Ca7 (U07934) Cicer M. albiziae CCBAU 61158T (DQ100066) Albizia 82 98 M. amorphae SEMIA 806 (FJ025125) Lotus M. amorphae ACCC 19665T (AF041442) Amorpha T 45 58 M. huakuii IFO 15243 (D13431) Astragalus M. sp. AC39e1 (GQ847892) Acacia 77 M. plurifarium LMG 11892T (Y14158) Acacia 68 T M. mediterraneum UPM-Ca36 (AM181745) Cicer 98 M. sp. LBER1 (FN563430) Lotus 72 M. sp. Ala-3 (AM491621) Anagyris

63 M. tianshanense A-1BST (AF041447) Glycyrrhiza WSM2624 97 Lotononis s. l. 69 M. tianshanense RCAN03 (AY195846) Cicer A. caulinodans ORS 571T (D11342) Sesbania

0.01 Figure 2.8. NJ phylogenetic tree, bootstrapped with 1000 replicates, showing the relationships of Ensifer and Mesorhizobium strains associated with Lotononis s. l. (in bold type) with related strains and type strains of Ensifer and Mesorhizobium, based on aligned sequences of the 16S rRNA genes. GenBank accession numbers are in brackets. Also shown is the genus of the original host (E. adhaerens LMG 20216T was isolated from soil). Scale bar, 1% sequence divergence (one substitution per 100 nucleotides). A., Azorhizobium; E., Ensifer; M., Mesorhizobium. T = type strain.

81 Chapter 2

Analysis of the nearly full-length (1416 bp) 16S rRNA gene sequence of

WSM2667 placed this strain in the genus Methylobacterium (Figure 2.9). The closest relationship was to the Methylobacterium nodulans type strain ORS 2060T (Sy et al.,

2001), with 99.9% (1 bp mismatch) sequence identity. WSM2667 was also closely related to Methylobacterium isbiliense DSM 17168T (Gallego et al., 2005; Kato et al., 2008) and to strains of the pigmented methylobacteria isolated from Listia species, such as Methylobacterium sp. 4-46 (Renier et al., 2008), WSM2598,

WSM2799 and WSM3032 (Yates et al., 2007), with sequence identities of 98.3%,

98.0%, 98.0%, 98.0% and 97.5%, respectively.

Strain WSM3557 was identified, with high bootstrap support, as being in the genus Microvirga and family Methylobacteriaceae (Figure 2.9). Its closest characterised phylogenetic relative was Microvirga flocculans TFBT (Takeda et al.,

2004; Weon et al., 2010). As well as grouping with the Listia angolensis strains

WSM3674 and WSM3686 (Yates et al., 2007) and Lupinus texensis strains Lut5 and

Lut6 (Andam & Parker, 2007), WSM3557 had high sequence identity with several other root nodule isolates. These included the authenticated strains AC72a, isolated from nodules of Phaseolus vulgaris growing in Ethiopia (Wolde-Meskel et al., 2005) and ARRI 185 (genospecies AL), from northern Australian Indigofera linifolia

(Lafay & Burdon, 2007). The 16S rRNA gene sequences of several unpublished nodule isolates also clustered within this clade. Strain SWF66521 was isolated from a Sesbania species in Yunnan Province, China, while strain TP1 was obtained from

Tephrosia purpurea growing in the Thar Desert, India (Figure 2.9). Further analysis of the taxonomy and phylogeny of WSM3557 and the other novel rhizobial isolates from Listia angolensis and Lupinus texensis is covered in Chapter 4.

82 Chapter 2

Mtb. sp. WSM2598 58 (DQ838527) Li. bainesii 53 Mtb. sp. WSM2799 (DQ848137) Li. bainesii 100 Mtb. sp. 4-46 (GI170738367) Li. bainesii Mtb. sp. WSM3032 99 (DQ848137) Li. solitudinis Mtb. isbiliense DSM 17168T (AB302929) Mtb. nodulans ORS 2060T (GI220920054) C. podocarpa 86 55 100 WSM2667 Le. calycina Mtb. variabile GR3T (NR_042348) 95 98 Mtb. aquaticum GR16T ((NR_025631) T 80 Mtb. organophilum ATCC 27886 (NR_041027) Mtb. oryzae CBMB20T (NR_043104) T 75 Mtb. populi BJ001 (NR_029082) Mtb. extorquens IAM 12631T (AB175632) 100 62 Mtb. zatmanii DSM 5688T (NR_041031) DSM14364T (FR733708) Mi. subterranea Mi. sp. ARRI 185 (AJ785294) I. linifolia

100 83 Mi. sp. AC72a (AY776209) P. vulgaris 100 Mi. sp. SWF66521 (GQ922064) Sesbania sp. 68 71 Mi. sp. TP1 (HM042680) T. purpurea Mi. flocculans TFBT (NR_036762) 100 Mi. sp. Lut5 (EF191407) Lu. texensis Mi. sp. Lut6 (EF191407) Lu. texensis Mi. sp. WSM3557 (HM362432) Li. angolensis Mi. sp. WSM3686 (DQ838529) Li. angolensis 100 79 Mi. sp. WSM3674 (DQ838528) Li. angolensis B. japonicum USDA 6T (U69638) G. max

0.01

Figure 2.9. NJ model phylogenetic tree, bootstrapped with 1000 replicates, showing the relationships of Methylobacterium and Microvirga strains isolated from Listia species (in bold type) with related strains and type strains of Methylobacterium and Microvirga, based on aligned sequences of the 16S rRNA genes. GenBank accession numbers are in brackets. The host species of rhizobial strains are given. Bootstrap values are indicated on branches only when higher than 50%. Scale bar, 1% sequence divergence (one substitution per 100 nucleotides). B., Bradyrhizobium; Mi., Microvirga; Mtb., Methylobacterium. C., Crotalaria; G., Glycine; I., Indigofera; Le., Leobordea; Li., Listia; Lu., Lupinus; P., Phaseolus; T., Tephrosia. T = type strain.

83 Chapter 2

2.3.6 Amplification and sequencing of nodA

2.3.6.1 nodA sequences of WSM2598 and WSM2667

PCR amplification using the primers M4-46 nodD 89-109f and M4-46 nodB

29-48r2 yielded products of approximately 1000 bp for the two methylobacterial strains WSM2598 and WSM2667. All other strains failed to yield an amplification product with these primers. Sequences of 937 bp and 965 bp were obtained for

WSM2598 and WSM2667, respectively, and contained a portion of nodD, the NodD nod box binding region, the complete nodA gene and a portion of nodB. Both strains had nodA genes of 645 bp length, with the proteins deduced from the nodA gene sequences containing 214 amino acids. The sequence identity between WSM2598 and WSM2667 was less than 80%. WSM2598 had highest sequence identity with the

Listia bainesii symbiont strain Methylobacterium sp. 4-46 (99.4%) while the

WSM2667 nodA sequence was most closely related to that of Methylobacterium nodulans ORS 2060 (99.5%) (Table 2.17). The nod box region of WSM2598 had the same sequence as that of Methylobacterium sp. 4-46. Similarly, the WSM2667 and

ORS 2060 sequences shared 100% sequence identity in this region (Figure 2.10).

Table 2.17. Pairwise percent identity and number of nucleotide mismatches (in brackets) for the 645 bp nodA gene of the Lotononis s. l. methylobacterial strains and the reference strains Methylobacterium sp. 4-46 and Methylobacterium nodulans ORS 2060T. T WSM2598 4-46 WSM2667 ORS 2060 99.4 WSM2598 79.4 79.1 (4) 99.4 Mtb sp. 4-46 79.7 79.4 (4) 99.5 WSM2667 79.4 79.7 (3) T 99.5 ORS 2060 79.1 79.4 (3)

84 Chapter 2

Meth 4-46 GATCGAGGCCCTTGAAGCGCATCTACCCTCCTATCCACCACCCAGATGCG WSM2598 GATCGAGGCCCTTGAAGCGCATCTACCCTCCTATCCACCACCCAGATGCG ORS 2060 TATCCATCGTTTGGATATGCTGCATCAAAACAATCGATTTTACAAATGTC WSM2667 TATCCATCGTTTGGATATGCTGCATCAAAACAATCGATTTTACAAATGTC pRL1JI TATCCATTCCATAGATGATTGCCATCCAAACAATCAATTTTACCAATCTT

Figure 2.10. Comparison of nod box sequences in the promoter regions of nodA genes from methylobacterial strains WSM2598, WSM2667 and Methylobacterium sp. 4-46 (isolated from Lotononis s.l. hosts), and from Methylobacterium nodulans ORS 2060 (associated with Crotalaria species). A-T-C-N9-G-A-T repeats (or related motifs) are in blue type. The sequence from the symbiotic plasmid pRL1JI of Rhizobium leguminosarum bv. viciae (Feng et al., 2003) is included for comparison.

2.3.6.2 nodA sequences of the remaining Lotononis s. l. rhizobia

A partial sequence of the nodA and nodB genes was obtained for WSM2632,

WSM2653, WSM2783, WSM3040 and WSM3557 using the primers nodA-1 and nodA-2. No amplification products could be obtained for WSM2596 or WSM2624.

The 638 bp sequences of the bradyrhizobial strains WSM2632 and WSM2783 were identical, while a 631 bp fragment of nodA from the E. meliloti strains WSM2653 and WSM3040 had 100% sequence identity. The 636 bp sequence of the Microvirga strain WSM3557 had sequence identity of 84.9% or less with WSM2632,

WSM2653, WSM2783 and WSM3040. Rhizobial strains identified by BLASTN as having nodA sequences that were closely related to WSM3557 had sequence identities of 83% or less. Alignment of a 565 bp nodA fragment showed that the

Bradyrhizobium, Ensifer, Methylobacterium nodulans, pigmented Methylobacterium and Microvirga strains all possessed distinct nodA sequences, with 84.9% or less sequence identity between each of these groups (Table 2.18).

85 Chapter 2

Table 2.18. Pairwise percent identity for a 565 bp fragment of the nodA gene of Lotononis s.l. strains.

WSM2598 WSM2667 WSM2632 WSM2783 WSM2653 WSM3040 WSM3557

WSM2598 80.4 71.5 71.5 66.2 66.2 70.1

WSM2667 80.4 83.5 83.5 72.6 72.6 81.0

WSM2632 71.5 83.5 100 79.5 79.5 84.9

WSM2783 71.5 83.5 100 79.5 100 84.9

WSM2653 66.2 72.6 79.5 79.5 100 75.6

WSM3040 66.2 72.6 79.5 79.5 100 75.6

WSM3557 70.1 81.0 84.9 84.9 75.6 75.6

The complete or nearly complete nodA sequences of the Lotononis s. l. strains were analysed, along with those of rhizobial strains identified as having highest

BLASTN or BLASTX sequence similarity, and several reference strains. The resulting NJ phylogenetic tree is shown in Figure 2.11. Although the clades within the tree were clearly defined and mostly well supported by high bootstrap values, the higher branches were not well supported, as evidenced by low bootstrap values. The phylogenetic relationships between the nodA genes of the Lotononis s. l. strains and those of other rhizobia therefore could not be determined with confidence. Trees generated using the ME and MP methods yielded similar results.

The Lotononis s. l. strains in the NJ nodA tree formed several polyphyletic clusters that intermingled with other taxonomically diverse rhizobial strains. It is interesting to note that all strains, except for the methylobacteria WSM2598 and

WSM2667, shared a three base pair deletion with reference Ensifer, Mesorhizobium,

Microvirga and Rhizobium strains that was not present in the nodA sequences of

Bradyrhizobium, Burkholderia and Methylobacterium rhizobia (Figure 2.12).

86 Chapter 2

WSM2653 and WSM3040 grouped with a cluster of African Ensifer strains that nodulate Acacia and Prosopis (Ba et al., 2002; Nick et al., 1999), with sequence identities ranging from 76% to 78%. The Lotononis s. l. bradyrhizobia WSM2632 and WSM2783 formed a separate lineage from Bradyrhizobium strains isolated from

Australian native legumes (WSM1735 and WSM1790) and from the clade of remaining bradyrhizobia, which included a number of strains isolated from native

African legumes. WSM3557 was a sister group to the Lotononis s. l. bradyrhizobia.

The Methylobacterium strains WSM2598 and WSM2667 were included in a well- supported clade that contained the other methylobacterial rhizobia strains ORS 2060 and 4-46. There was a large divergence, however, between the Methylobacterium nodulans strains (ORS 2060 and WSM2667) and the strains isolated from Listia bainesii (4-46 and WSM2598).

87 Chapter 2

57 E. sp. ORS1085 (AJ302677) A. tortilis Senegal 100 E. sp. ORS1230 (AJ302681) A. tortilis Senegal 99 E. arboris HAMBI 1396 (EF457955) P. chilensis Kenya 90 E. terangae ORS1009T (Z95237) A. laeta Senegal E. meliloti WSM2653 Lotononis s. l. South Africa E. meliloti WSM3040 Lo. laxa 63 100 South Africa 99 M. plurifarium CFN ESH18 (EU291996) Ac. angustissima Mexico M. plurifarium ORS1255 (AJ302680) A. tortilis Senegal 84 Mi. sp. Lut6 (EF191414) Lu. texensis USA R. tropici CFN299 (X98514 ) Ph. vulgaris Brazil 80 E. meliloti ORS1044 (AJ300232) A. tortilis Senegal 90 E. sp. BR827 (Z95232) Leu. leucocephala Brazil Br. sp. WSM2632 Lotononis s. l. South Africa 100 Br. sp. WSM2783 Lotononis s. l. South Africa Mi. sp. WSM3557 (HQ435534) Li. angolensis Zambia M. sp. CCNWAX24-1 (EU722471) Alh. sparsifolia China M. sp. CCNWGS0011 (GQ871220) T. lanceolata China 67 73 M. sp. ICMP 14330 (DQ100401) So. microphylla NZ 53 M. albiziae CCBAU 61158T (GQ167236) Alb. kalkora China Bur. sp. WSM3937 (EU219867) Rh. ferulifolia South Africa T 100 Bur. tuberum STM678 (AJ302321) Cyclopia South Africa 100 Bur. tuberum DUS833 (EF566976) Cyclopia South Africa

100 Mtb. sp. WSM2598 Li. bainesii South Africa Mtb. sp. 4-46 (CP000943) Li. bainesii South Africa 99 Mtb. nodulans WSM2667 Le. calycina South Africa 100 Mtb. nodulans ORS 2060T (CP001349) Cr. podocarpa Senegal 62 Br. sp. WSM1735 (AJ890293) Rh. minima NW Aust 100 Br. sp. WSM1790 (AJ890286) I. colueta NW Aust Br. japonicum ORS1816 (AJ430722) Cr. hyssopifolia Senegal Br. sp. F3b (AM117517) Les. cuneata Japan 100 Br. elkanii USDA 76T (AM117554) G. max USA 68 82 Br. sp. USDA3259 (AM117558) Ph. lunatus USA 90 100 Br. sp. ORS1812 (AJ430723) Ab. stictosperma Senegal 97 Br. sp. ORS170 (AJ300258) F. albida Senegal 53 Br. sp. CP283 (AJ920038) Parasponia New Guinea 58 Rhizobial sp. STM251 (AJ300259) Ch. sanguinea Guinea E. fredii NGR234 (U00090) L. purpureus New Guinea R. leguminosarum WSM1325 (NC_012850) Trifolium sp. Greece 99 E. medicae WSM419 (NC_009636) M. murex Sardinia A. caul. ORS 571T (NC_009937) S. rostrata Senegal

0.05 Figure 2.11. NJ nodA phylogenetic tree, bootstrapped with 1000 replicates, showing the relationships of the Lotononis s.l.-associated rhizobia (in bold type) and other rhizobia. GenBank accession numbers are in brackets. Also shown is the original host and the geographical location of the strain. Scale bar, 5% sequence divergence (five substitutions per 100 nucleotides). A., Azorhizobium; Br., Bradyrhizobium; Bur., Burkholderia; E., Ensifer; M., Mesorhizobium; Mtb., Methylobacterium; Mi., Microvirga; R., Rhizobium. A., Acacia;Ab., Abrus; Ac., Acaciella; Alb., Albizia; Alh.,

Alhagi; Ch., Chidlowia; Cr., Crotalaria; F., Faidherbia; I., Indigofera; L., Lablab; Le., Leobordea; Les., Lespedeza; Leu., Leucaena; Li., Listia; Lo., Lotononis s. str;

Lu., Lupinus; M., Medicago; P., Prosopis; Ph., Phaseolus; Rh., Rhynchosia; S., T Sesbania; So., Sophora; T., Thermopsis. = type strain.

88 Chapter 2

A. caulinodans ORS 571 TGCGGATCCATTTTCAGCGCTT---CTGTCGCAA E. arboris HAMBI1396 TGCAAAGACACGTGGAAAGATT---CGGCCGACA E. fredii NGR234 TGTACAAGCTTGTGGGCAGACT---CTGCCGAAA E. medicae WSM419 TGCGGAACCACGTTGAGAGATA---TTGCCAGAA E. meliloti WSM2653 TGCAGAAGCACATTACCAGATT---CGTCCGACA E. meliloti WSM3040 TGCAGAAGCACATTACCAGATT---CGTCCGACA Ensifer sp. BR827 TGCGCAAACATATTGAGAGGTT---CGCCCGGTA Ensifer sp. ORS 1044 TGCGCAAGCATGTAGAGCGGTT---CGCCCGATA Ensifer sp. ORS 1085 TGCAAAGGCACGTGGAAAGATT---CGGCCGACA Ensifer sp. ORS 1230 TGCAAAGGCACGTGGAAAGGTT---CGGCCGACA E. terangae ORS 1009 TGCAGAAGCACGTCGAAAGATT---CGGTCGGCA Mes. albizieae CCBAU61158 TGCGGAACCATGTTGAGAGGTT---CTGCAGAGA Mes. plurifarium ORS 1255 TGCATAAACACGTTGCAAGATT---GGGGCGAGG Mes. plurifarium CFN ESH18 TGCGCAAACACGTTGCAAGATT---CGAGCGACA Mes. sp. CCNWAX24-1 TGCGGAACCATCTTGAGAGGTT---TGGCAGACA Mes. sp. CCNWGS0010 TGCGGAACCATGTTGAGAGGTT---CTGCCGAGG Mes. sp. CCNWSX426 TGCGGAACCATATGGAGAGGTT---CTGCAGAGA Mes. sp. ICMP 14330 TGCGAAACCATATTGTAAGATT---CTTCAGAGA Microvirga WSM3557 TGCAGAAGCATGTTGCAAGGCT---GGGCCGGCA Microvirga Lut6 TGCAGAAACATGTTGAGAGGTT---CGGCCGATA R. tropici CFN299 TGCAAAAGCATATCGAAAGGTT---CGGCCGTTA R. leguminosarum WSM1325 TGCGGAACCACGTGGAGAGATT---CTGCCGAGA

Br. sp. WSM2632 TGCAGAAGCATGTTGAAAGGGT---CGCCCGGCA Br. sp. WSM2783 TGCAGAAGCATGTTGAAAGGGT---CGCCCGGCA

Br. elkanii USDA 76 TTGAGAAGCATCTGACCAGACTGGTCGGAAGGCG Br. sp. CP 283 TTGAGAGGCATCTCACCAGGCTGGTCGGAAGGCA Br. sp. f3b TTGAGAAGCATCTTACCAGGCTGGTCGGCCGGCA Br. sp. ORS 1812 TTGAGAAGCATCTCACCAGGCTGGTCGGAAGGCA Br. sp. ORS 1816 TCGAGAAACATATGACGCGACTGGTCCAAAGGCA Br. sp. ORS 170 TTGAGAAGCATCTCACCAGGCTAGTCGGAAGGCA Br. sp. USDA 3259 TTGAGAAGCATCTCACCAGGCTGGTCGGAAGGCA Br. sp. WSM1735 TGAAGGGCCATCTTACAAGGTTGCTGACCCGCCG Br. sp. WSM1790 TGAAGGGCCATCTTACAAGGTTGCTGACCCGCCG Burk. tuberum STM678 TGCGGCAACATATTGCAAGGCTGCTCGGCCGACC Burk. tuberum DUS833 TGCGGCAACATATTGCAAGGCTGCTCGGCCGACC Burk. sp. WSM3937 TGCGGCAACATATTGCAAGGCTGCTCGGCCGACA Mtb. sp. 4-46 TTCAGAAACACCTCACCAGGCTGCTCGGTAAGGC Mtb. sp. WSM2598 TTCAGAAACACCTCACCAGGCTGCTCGGTAAGGC Mtb. nodulans ORS 2060 TGCAGAAACATCTTACAAGGCTGCTCGGTAAGGC Mtb. nodulans WSM2667 TGCAGAAACATCTTACAAGGCTGCTCGGTAAGGC

Figure 2.12. Alignment of a portion of the nodA sequences of the Lotononis s. l. rhizobia (in bold type) and reference strains featured in Figure 2.11. The aligned sequences begin from position 388 of the complete Azorhizobium caulinodans ORS 571 nodA sequence. The Lotononis s. l. bradyrhizobial strains WSM2632 and WSM2783 (highlighted) share a three base pair deletion with reference Ensifer, Mesorhizobium, Microvirga and Rhizobium strains that is not present in the nodA sequences of Burkholderia, Methylobacterium and other Bradyrhizobium rhizobia. A. = Azorhizobium; Br = Bradyrhizobium; Burk. = Burkholderia; E. = Ensifer; Mes. = Mesorhizobium; Mtb = Methylobacterium; R = Rhizobium.

89 Chapter 2

2.4 Discussion

This study reports on the type and extent of nodulation within Lotononis s. l. and the diversity of root nodule bacteria associated with this genus. Additionally, it describes the possible interactions between the symbiotic and the biogeographical relationships between these host plants and their associated rhizobia.

2.4.1 Nodule morphology in Lotononis s. l. species

Previous authors (Corby, 1981; Lavin et al., 2001; Sprent & James, 2007) have noted the value of nodule morphology and structure as a taxonomic tool, and this study shows that nodule type can be used as a taxonomic marker to distinguish Listia species from other genera in the Lotononis s. l. clade. Listia angolensis, Listia bainesii and Listia heterophylla form lupinoid nodules. Recent studies have confirmed that all examined species in the genus Listia have this type of nodule (R.

Yates, unpublished data). Lupinoid nodulation appears to be a property unique to

Listia, whereas Leobordea and Lotononis s. str. species all have indeterminate nodules. All tested host species formed nodules, with the exception of the geophytic

Euchlora hirsuta. The inability of this host to nodulate with any of the rhizobial strains used in this study supports previous observations that record a lack of nodulation in field studies of this species (J. Howieson, R. Yates, unpublished data), although the GRIN Rhizobial Nodulation Data (http://www.ars- grin.gov/~sbmljw/cgi-bin/taxnodul.pl) lists it as nodulated.

2.4.2. Diversity of Lotononis s. l. rhizobia

This study has revealed that wild Listia, Leobordea and Lotononis s. str. species are nodulated by phylogenetically diverse rhizobia. The rhizobia that nodulate Lotononis s. l. species belong to five different genera in the

90 Chapter 2

Alphaproteobacteria, one of which (Microvirga) has not previously been described as containing rhizobial species. Other studies have similarly reported on wild legumes, such as species of Lotus, Dalbergia, Ononis, Pterocarpus and Sophora, that are nodulated by natural populations of genotypically diverse rhizobia (Lorite et al.,

2010; Rasolomampianina et al., 2005; Rincón et al., 2008; Sylla et al., 2002; Zhao et al., 2010). What is unusual about the Lotononis s. l. symbiotic associations, however, is firstly, the very wide taxonomic diversity of the rhizobia and secondly, the varying levels of specificity exhibited by the different host species, ranging from the highly specific Listia bainesii to the more promiscuous Leobordea bolusii.

Some degree of microsymbiont diversity is to be expected, given that South

Africa is the centre of diversity for Lotononis s. l. and rhizobial centres of diversity are thought to coincide with those of their legume hosts (Andronov et al., 2003; Lie et al., 1987). The extent of this taxonomic diversity prompts an examination of the biogeography and symbiotic relations of the rhizobial lineages that are associated with Lotononis s. l. and of the climatic, edaphic and host plant factors that may have contributed to the development of microsymbiont diversity in this group of legumes.

2.4.2 1 The Methylobacterium lineage

Methylobacterium is a widespread genus that contains many plant-associated and endophytic species (Madhaiyan et al., 2007, 2009b; Sy et al., 2005; Van Aken et al., 2004). Two rhizobial lineages of Methylobacterium have so far been identified: the pigmented Listia strains (Jaftha et al., 2002; Yates et al., 2007) and the non- pigmented M. nodulans, which specifically nodulates species of Senegalese

Crotalaria (Jourand et al., 2004; Sy et al., 2001). Nodulating Methylobacterium strains appear to have both a limited geographical distribution and host range, with

91 Chapter 2 an affinity to African crotalarioid host legumes, although a recent study has described M. nodulans strains isolated from Indian Crotalaria juncea and Sesbania aculeata (Madhaiyan et al., 2009a). This geographical distribution reflects the predominantly African distribution of the Crotalarieae tribe, although the large genus

Crotalaria is pantropical (van Wyk, 2005).

The isolation of WSM2667 from Leobordea calycina (and its ability to nodulate other Lotononis s. l. species) extends the host range of M. nodulans from a narrow group of Crotalaria species to other genera in the Crotalarieae tribe.

Although M. nodulans and the pigmented Listia methylobacteria have distinct and highly specific host ranges, WSM2598 and WSM2667 were able to nodulate the same subset of Lotononis s. l. species, with the notable exception of Listia bainesii.

The methylobacterial strains WSM2598, WSM2667 and ORS 2060 were also the only rhizobia able to nodulate Listia angolensis other than its cognate Microvirga microsymbionts, although the Methylobacterium-induced nodules were always ineffective.

The results of this study support the finding that Listia methylobacteria have a narrow host range. WSM2598 was effective only on Listia species (other than L. angolensis) and ineffectively nodulated or failed to nodulate other Lotononis s. l. hosts. Pigmented methylobacteria have been isolated, over a wide geographical range, from all studied Listia species except L. angolensis (Yates et al., 2007; R.

Yates, unpublished data). The individual microsymbiont strains vary in their level of effectiveness, but none has been found to be ineffective on these hosts (R. Yates,

92 Chapter 2 unpublished data) and these methylobacteria would appear to constitute a symbiotically stable population of effective and highly specific microsymbionts.

2.4.2.2 The Microvirga lineage

This study is the first report of Microvirga strains that are capable of nodulation and N2 fixation with legumes. Although the small number of isolates obtained from L. angolensis precludes definitive conclusions, the data presented here indicate that rhizobial Microvirga species preferentially nodulate this legume host.

The symbiotic relationship between Microvirga strains and L. angolensis is not as tight as that of the other Listia species and their associated methylobacteria.

Symbiotic effectiveness appears to be more variable: in this study, some Microvirga strains were only partially effective on L. angolensis, and Eagles & Date (1999) have noted that WSM3674 (= CB1323; see Chapter 4) is effective on only some accessions of L. angolensis. Rhizobial Microvirga strains appear to have a greater host range and geographical distribution than the Methylobacterium rhizobia.

WSM3557 had the broadest host range of all strains in this study, nodulating all

Lotononis s. l. hosts except L. bainesii in Experiment 3, and being partially effective on three of these species in addition to effectively nodulating L. angolensis.

Although rhizobial species of Microvirga have not yet been formally described, published reports indicate that they have a broad tropical or sub-tropical distribution and nodulate taxonomically diverse legume hosts: Lupinus texensis in Texas, USA,

Phaseolus vulgaris in Ethiopia and northern Australian Indigofera linifolia (Andam

& Parker, 2007; Lafay & Burdon, 2007; Wolde-Meskel et al., 2005).

93 Chapter 2

2.4.2.3 The Bradyrhizobium, Ensifer and Mesorhizobium lineages

It is not possible to gain a complete picture of the symbiotic relationships between Lotononis s. l. species and their cognate rhizobia, due to the small number of isolates obtained from South African Leobordea and Lotononis s. str. species. It is illustrative of the genotypic diversity of these isolates though, that the seven collected strains belong to four different genera (Bradyrhizobium, Ensifer,

Mesorhizobium and Methylobacterium) and two separate lineages of bradyrhizobia.

Their identification has been based solely on sequences of the 16S rRNA gene, however, which is not adequate for delineating rhizobial species (van Berkum et al.,

2003). Although sequencing additional loci, such as housekeeping genes, would be required for a more accurate determination of taxonomic status, the existing data do suggest interesting phylogenetic and biogeographical relationships in these lineages.

WSM2653 and WSM3040 are strains of Ensifer meliloti, which is classically considered to be a specific microsymbiont of the genera Medicago, Melilotus and

Trigonella (Fred et al., 1932). Recent studies, however, have identified strains of E. meliloti that effectively nodulate Cicer arietinum in Tunisia (Ben Romdhane et al.,

2007), Phaseolus vulgaris in Tunisia and the Canary Islands (Mnasri et al., 2007;

Zurdo-Piñeiro et al., 2009) and endemic Lotus species in the Canary Islands (León-

Barrios et al., 2009). E. meliloti strains have also been isolated from diverse wild legumes growing in arid regions of northern Africa. Host plants include Acacia tortilis and Prosopis farcta (tribe Mimoseae) (Ba et al., 2002; Fterich et al., 2011);

Argyrolobium uniflorum, Genista saharae and Retama raetam (Genisteae);

Hippocrepis bicontorta and Lotus spp. (Loteae); Hedysarum carnosum (Hedysareae);

94 Chapter 2

Ononis natrix (Trifolieae) (Mnasri et al., 2009); and Colutea arborescens (Galegeae)

(Ourarhi et al., 2011).

The predominant Mesorhizobium species that has so far been associated with

African legumes is M. plurifarium, which has been described as nodulating species of the mimosoid genera Acacia, Leucaena and Prosopis and the caesalpinioid genus

Chaemaecrista (Ba et al., 2002; de Lajudie et al., 1998a). Recent studies have also reported novel genospecies of Mesorhizobium isolated from Ethiopian Acacia abyssinica, A. tortilis (Mimoseae) and Sesbania sesban (Sesbanieae) (Degefu et al.,

2011); Anagyris latifolia (Thermopsideae) and Lotus spp. (Loteae) in the Canary

Islands (Donate-Correa et al., 2007; Lorite et al., 2010) and endemic South African

Lessertia spp (Galegeae) (Howieson et al., 2008; M. Gerding Gonzalez, unpublished thesis). Mesorhizobium tianshanense, the species with highest sequence identity to

WSM2624, has not previously been reported as associated with African legumes. M. tianshanense was thought to have a limited geographical distribution. Han et al.

(2010) considered this species to be endemic to Xinjiang region, China, although strains of M. tianshanense have been isolated from nodules of Cicer arietinum

(Cicereae) in Spain and Portugal (Laranjo et al., 2004; Rivas et al., 2007). It appears to be adapted to diverse legume hosts, as the host range of the originally described species encompasses legumes in tribe Phaseoleae and in the genistoid and galegoid clades (Chen et al., 1995).

The Lotononis s. l. bradyrhizobia form two distinct lineages that do not correspond to any of the diverse bradyrhizobial strains that have previously been isolated from native South African or African legume hosts (Figure 2. 7) (Boulila et

95 Chapter 2 al., 2009; Garau et al., 2009; Steenkamp et al., 2008; Sylla et al., 2002). Their biogeography is interesting, as the rhizobial strains most closely related to the

Lotononis s. l. bradyrhizobia were isolated from host plants that are phylogenetically and geographically distant from Lotononis s. l. species (Parker & Kennedy, 2006;

Parker, 2008; Qian et al., 2003). Bradyrhizobia have been reported to nodulate various African Crotalaria species (Moulin et al., 2004; Sy et al., 2001), but in general there is a paucity of data available on the microsymbionts of native African crotalarioid legumes. The phylogenetic and biogeographical relationships of bradyrhizobial microsymbionts of this group of legumes may become clearer with further sampling. The strain most closely related to the Lotononis s. l. bradyrhizobia that had both geographical and host plant affinity was RSB6, which nodulates

Algerian Retama (tribe Genisteae) species (Boulila et al., 2009).

What factors are possible drivers of the phylogenetic diversity seen in the

Lotononis s. l.-associated rhizobia? As host plants have been shown to play a key role in shaping rhizobial diversity (Han et al., 2010) and plants are more likely than rhizobia to exercise partner choice (Simms & Taylor, 2002), an examination of the patterns of symbiotic association between the different Lotononis s. l. taxonomic groups and their associated rhizobia may provide an explanation for this microsymbiont diversity.

2.4.3 Symbiotic relationships within Lotononis s. l. hosts

The Lotononis s. l. species can be divided into two groups, in terms of their symbiotic relationships. Listia species have high symbiotic specificity and are associated with unusual rhizobia. For Listia species other than L. angolensis, specificity is acute, as these legumes nodulate only with their associated pigmented

96 Chapter 2 methylobacteria. Nitrogen fixation is linked with specificity: the Listia methylobacteria are effective or highly effective on all Listia species except L. angolensis (Yates et al., 2007; R. Yates, unpublished data). In comparison, in the L. angolensis-Microvirga symbiosis, there has been some diminution of specificity in the host plant, a broadening of the host range of the rhizobial partner and greater variability of nitrogen fixation effectiveness.

In contrast to the Listia species, there appears to be no relationship between host plant taxonomy and rhizobial chromosomal background in the symbiotic associations seen in Leobordea and Lotononis s. str. species. Most of these host species could be considered promiscuous, as they were able to nodulate with diverse rhizobial strains. Similarly, the Bradyrhizobium, Ensifer and Methylobacterium strains isolated from Leobordea and Lotononis s. str. species had broad host ranges across this group of legumes. The Mesorhizobium strain WSM2624, while being the least infective, also had no obvious pattern of host nodulation. Similar high promiscuity has been noted in other studies of native populations of rhizobia and legumes (Han et al., 2010; Lafay & Burdon, 1998; Weir, 2006; Zhao et al., 2010).

The effectiveness of the symbioses was generally poor and can be explained by several factors. Firstly, none of the Leobordea and Lotononis s. str. species, with the exception of Lotononis laxa, is a homologous host of the inoculant strains used in this study. Rhizobia that are not the usual partners of these host species would therefore not be expected to be as symbiotically compatible. Secondly, rhizobial strains isolated from a population of a wild legume host can vary in their symbiotic effectiveness (Odee et al., 2002; Singleton & Tavares, 1986). Unlike WSM2598 and

97 Chapter 2

WSM3557, the remaining rhizobia in this study have not been screened for nitrogen fixation effectiveness and may be suboptimal strains. Thirdly, the host plants were usually obtained from a collection of wild seeds. Individual plants within a population would be expected to have considerable genetic variability, which has been linked with variations in symbiotic compatibility with rhizobial partners (Abi-

Ghanem et al., 2011; Drew & Ballard, 2010; Wilkinson et al., 1996). This could also explain the differences in the responses of individual plants to inoculation, and the ineffective nodulation of WSM3040 on its cognate L. laxa host, which has previously been noted for L. laxa isolates on various L. laxa accessions (The CB

Collection Database, unpublished data). This putative lack of broad symbiotic compatibility across a population of the host species is in contrast to the broad-scale effectiveness of methylobacterial strains across the Listia cross-inoculation group.

The rhizobial host range is usually a product of the nodulation genes that are present in the genome (Perret et al., 2000). The phylogeny of the rhizobial nod genotype has been found to correlate with that of the host plant, most notably in studies of temperate legumes (Mergaert et al., 1997; Suominen et al., 2001).

Conversely, studies of other legume-rhizobia associations have found no clear association between the host plant and the nod gene systematics (Lafay et al., 2006;

Moulin et al., 2004). Does an examination of the nodA phylogeny of Lotononis s. l. rhizobia provide insights into the evolution of symbioses in this group of legumes?

2.4.4 Phylogeny of nodA genes

The nodA sequences of the Lotononis s. l. rhizobia are polyphyletic, forming four separate clades in which the nodA lineages are associated with the rhizobial genomic species, although the low bootstrap values mean that the topology of the

98 Chapter 2 tree is not well supported. The nodA sequences of the Lotononis s. l. rhizobia are intermingled with nodA of rhizobia isolated from host plants that are not closely related to Lotononis s. l. (Figure 2.11). This indicates firstly, that the symbiotic genes of Lotononis s. l. rhizobia were derived from different sources or have evolved divergently and secondly, that those Leobordea and Lotononis s. str. hosts which are promiscuous nodulators are not stringently selecting the rhizobial symbiotic genotype.

The nodA sequences of the methylobacterial strains WSM2598 and WSM2667 are related to nodA of Methylobacterium sp. 4-46 (isolated from L. bainesii) and M. nodulans ORS 2060 (which specifically nodulates Crotalaria podocarpa, Crotalaria glaucoides and Crotalaria perrottetti (Renier et al., 2008; Sy et al., 2001)). Both

WSM2667 and ORS 2060 are unable to nodulate L. bainesii and L. heterophylla, although they do form occasional ineffective nodules on L. angolensis (this study).

The nodA and nod box sequences of the Methylobacterium rhizobia (4-46,

WSM2598, ORS 2060 and WSM2667) suggest that while there is a close phylogenetic relationship between the nodulation genes of M. nodulans and the

Listia methylobacteria, the branch depth indicates their divergence occurred some time ago. It has also been suggested that the nodulation genes of the

Methylobacterium rhizobia have been acquired by horizontal gene transfer (HGT) from Bradyrhizobium strains (Moulin et al., 2004). Their considerable divergence again suggests an ancient origin for this putative HGT event. In this study, the nodA sequences of the Methylobacterium strains appear to be more closely related to the main Bradyrhizobium branch than to the other Lotononis s. l. rhizobia.

99 Chapter 2

The ORS 2060 nod genes are organised in a single operon that encodes NodA,

NodB, NodC, NodI, NodJ and NodH proteins and is almost identical to the nod operon found in the sequenced genome of Methylobacterium sp. 4-46 (Renier et al.,

2008). It can be hypothesised from this that Listia and Crotalaria hosts that are nodulated by methylobacteria have similar Nod factor requirements. Crotalaria species are postulated to belong to two inoculation groups: most are nodulated by bradyrhizobia that may produce fucosylated Nod factors, while C. podocarpa, C. glaucoides and C. perrottetti are specifically nodulated by M. nodulans, which produces sulfated Nod factors (Renier et al., 2008) The Methylobacterium nodA sequences are in a sister clade to those of several Burkholderia tuberum strains that nodulate South African fynbos legumes (Elliott et al., 2007; Garau et al., 2009).

Interestingly, Burkholderia tuberum STM678 (previously named as Bradyrhizobium aspalati; see Elliott et al. (2007)) does not produce sulfated Nod factors (Boone et al., 1999).

The nodA sequences of the Bradyrhizobium, Ensifer and Microvirga strains do not group with rhizobia that nodulate other crotalarioid legumes. This may, however, be more a reflection of the current lack of data on rhizobia associated with this group of host plants. The nodA sequences of the E. meliloti strains WSM2653 and

WSM3040 are clearly grouped with diverse Ensifer and Mesorhizobium strains that form a nodulation group for Acacia, Prosopis and Leucaena (Ba et al., 2002). The nodA lineages of the bradyrhizobial isolates WSM2632 and WSM2783 are grouped apart from the Ensifer strains. Interestingly, they are also well separated from the clade containing all other Bradyrhizobium sequences, including ORS 1816, isolated from Crotalaria hyssopifolia in Senegal (Moulin et al., 2004). A previous study

100 Chapter 2 found that the nodA phylogeny of African bradyrhizobia placed them in the large, pan-tropical Clade III (Steenkamp et al., 2008). Moulin et al. (2004) considered bradyrhizobia to form a monophyletic branch within the nodA tree and their deduced

NodA proteins contain 209–211 amino acids (with the exception of the photosynthetic stem-nodulating Bradyrhizobium strains, which encode 197 amino- acid proteins). The published Methylobacterium strains 4-46 and ORS 2060, and

WSM2598 and WSM2667 from this study, along with Burkholderia tuberum strain

STM678 also contain deduced NodA proteins of over 200 residues. In contrast,

NodA in other rhizobial genera is usually composed of 195–198 amino acids. The greater length of the bradyrhizobial NodA is the result of a 12–13 amino acid segment at the N-terminal end (Moulin et al., 2004). The primers used in this study did not amplify this section of nodA, but it would be interesting to obtain the complete nodA sequences for WSM2632 and WSM2783, to determine whether, as the phylogenetic tree suggests, they are more closely related to Ensifer and

Mesorhizobium nodA, rather than other bradyrhizobial strains.

The position of the Microvirga strain WSM3557 in the nodA phylogenetic tree is intriguing. It forms a sister clade to the bradyrhizobial isolates WSM2632 and

WSM2783 and is separate from the only other nodulating Microvirga nodA sequence described to date, that of Lut6, which specifically nodulates Lupinus texensis, a species endemic to Texas, USA (Andam & Parker, 2007). Notably, WSM3557 is also separate from the nodA lineages of the Methylobacterium strains that specifically nodulate the remaining Listia species. Microvirga has not previously been described as a genus containing rhizobial species. It is presumed that the

101 Chapter 2 symbiotic genes that confer the ability to nodulate Listia angolensis have been acquired from another rhizobial lineage via HGT, but due to the low bootstrap values for the higher branches of the nodA tree, the source of the donor and the phylogeny of these nod genes are unclear.

Although the nodA gene is a host range determinant (Debellé et al., 1996;

Lortet et al., 1996), the congruence between nodA phylogeny and the taxonomy of the legume host is stronger in some associations than others. Host plants in the

Galegeae, Trifolieae and Vicieae tribes (the galegoid clade) are nodulated by diverse rhizobia with similar nodA sequences; which is related to the requirement of these legumes for Nod factors that are N-acylated with unsaturated fatty acids (Debellé et al., 2001; Suominen et al., 2001). Several studies of other legume species that are nodulated by phylogenetically diverse rhizobia have also found the nod genes to be highly similar, regardless of rhizobial chromosomal background (Ba et al., 2002;

Haukka et al., 1998; Laguerre et al., 2001; Lu et al., 2009). Conversely, Han et al.

(2010) and Zhao et al. (2010), in studies of rhizobia isolated from wild legumes in

China, concluded that the relations between the microsymbiont chromosomal background, nod genes and host plant were promiscuous and the nodA or nodC lineages were clearly associated with the rhizobial genomic background. Canary

Island Lotus species also have Ensifer meliloti and Mesorhizobium microsymbionts in which distinct nodC lineages are related to the different chromosomal genotypes, although in this case the E. meliloti strains have a restricted host range (Lorite et al.,

2010).

102 Chapter 2

Symbiotic relationships between the Lotononis s. l. legumes and their associated rhizobia appear to follow the latter model, where nodulation gene lineages are related to the microsymbiont chromosomal background. It is not possible, from the data on nodA presented in this study, to trace the lines of descent of the symbiotic genes that allow nodulation of Lotononis s. l. hosts, or say whether they came from a single, or multiple, ancestors. There appears to be no evidence for a “Lotononis s. l.” or a “Listia” nodulation genotype as such, although the very high nodA sequence identity seen within, rather than between, the Lotononis s. l.-associated

Bradyrhizobium, Ensifer, M. nodulans and pigmented Methylobacterium strains argues for selection pressure on nod genes and possible HGT within those chromosomal backgrounds. The collection and study of more rhizobial isolates from

Leobordia and Lotononis s. l. species and the acquisition of complete nodA sequences for all Lotononis s. l. rhizobia could provide more robust phylogenies of symbiotic loci and a greater understanding of the evolution of symbiotic relationships within Lotononis s. l. species.

If there is no evidence for selection pressure on Lotononis s. l.-associated rhizobial for a particular nodulation genotype, what other factors could account for the range of microsymbiont diversity and specificity seen in this group of host plants?

2.4.5 Environmental factors that may be effectors of diversification in rhizobia associated with Lotononis s. l. species

Rhizobial diversity in Lotononis s. l. can also be looked at in the context of the distribution and evolution of this clade of legume hosts within the Crotalarieae tribe.

Crotalarieae are monophyletic and sub-endemic to Africa (Boatwright et al., 2008),

103 Chapter 2 with initiation of the crown clade of Crotalarieae dated to approximately 46.3 mya

(Edwards & Hawkins, 2007). Lotononis s. l. has been included in the Crotalarieae

Cape floral clade, which is defined as a clade that is considered to originate in the remarkably species-rich Cape Floristic Region (CFR) (Edwards & Hawkins, 2007).

The species density of Lotononis s. l. in southern Africa is also notably high (van

Wyk, 1991). In looking at the radiations and subsequent speciation of the Cape clades, Cowling et al. (2009) have advanced the theory that topo-edaphic heterogeneity of the CFR, along with relative climatic stability, are factors that contribute to the exceptional diversity seen in this region. Another consequence of this edaphic heterogeneity may be that legume species are faced with selecting symbiotic partners of diverse genotypes, due to variable rhizobial saprophytic competence across different eco-regions.

Rhizobia are not obligate symbionts (Young & Johnston, 1989). Each generation of legume hosts is required to select its microsymbiont partner from a pool of free-living soil rhizobia (Martínez-Romero, 2009). Rhizobial saprophytic competence is thus important in determining which strains are available to be selected by the plant host. Environmental selection implies that distinctive microbial assemblages are maintained in different contemporary environments (Martiny et al.,

2006). In the case of the bacterial soil community, its composition is controlled primarily by edaphic factors, in particular soil pH (Fierer & Jackson, 2006).

Rhizobial populations in the soil are also known to be affected by pH, in addition to other factors such as soil fertility and clay content, temperature, rainfall and the host plant distribution (Dilworth et al., 2001; Hirsch, 1996; Howieson & Ballard, 2004).

Populations of Bradyrhizobium have been shown to increase as soil pH declines

104 Chapter 2

(Slattery et al. 2004) and Rhizobium tropici and Mesorhizobium loti are noted acid tolerant species; conversely E. meliloti (and other Ensifer species) are sensitive to acid stress and more tolerant of alkaline conditions (Graham & Parker, 1964;

Graham, 2008; Herridge, 2008; Zahran, 1999). The Listia methylobacteria have been found to be well adapted to acid, infertile soils (Diatloff, 1977; Yates et al., 2007).

Studies on diverse legumes have found that the microsymbiont genotype correlates with eco-regions and edaphic factors (Bala & Giller, 2006; Diouf et al., 2007; Garau et al., 2005; Han et al., 2009; Lu et al., 2009). In the present study, the number of rhizobial strains is too small to attempt a correlation between soil pH and microsymbiont genotype, but it is interesting to note that the soil pH of the sites where the bradyrhizobial isolates were collected was on average lower than that of the Ensifer collection sites (Table 2.1).

Similarly, the seasonally waterlogged habitat favoured by Listia species may be a factor in their utilisation of Methylobacterium and Microvirga strains as microsymbionts. Non-rhizobial species of these genera are known to inhabit aquatic environments. Pink-pigmented methylobacteria are frequently found in fresh water habitats (Green, 1992) and M. isbiliense, which is closely related to the rhizobial methylobacteria (Figure 2.9), was described from strains isolated from Spanish drinking water (Gallego et al., 2005). L. angolensis has a seasonally waterlogged habitat but a more tropical distribution than other Listia species; thus it is interesting to note that three of the five validly described non-rhizobial Microvirga species are from thermal waters or (presumably) waterlogged rice field soil (Kanso & Patel,

2003; Takeda et al., 2004; Weon et al., 2010; Zhang et al., 2009).

105 Chapter 2

The nodulation of the hypocotyl that is seen in Listia species is also unusual in legumes. Other accounts of hypocotyl nodulation appear to be restricted to subspecies of Arachis hypogaea (Nambiar et al., 1982) and to semi-aquatic species of Aeschynomene, which also form stem nodules (Boivin et al., 1997; Loureiro et al.,

1995). These legumes form aeschynomenoid nodules, however, with bacterial infection proceeding via epidermal cracks created by emerging lateral roots, rather than the lupinoid type seen in Listia species (Boogerd & van Rossum, 1997; Sprent,

2009). Arachis hypogaea and Aeschynomene species are nodulated by strains of bradyrhizobia, some of which, in the case of the Aeschynomene microsymbionts, may be pigmented, photosynthetic bacteria and/or lack canonical nodABC genes

(Miché et al., 2010; Van Rossum et al., 1995; Yang et al., 2005). Pigmentation is rare in RNB: the photosynthetic bradyrhizobia and the Methylobacterium and

Microvirga strains that nodulate Listia species are the only pigmented rhizobia to be described so far. The pigments found in the photosynthetic bradyrhizobia and in

Methylobacterium and Microvirga species have been identified as carotenoids

(Giraud et al., 2004; Jaftha et al., 2002; Kanso & Patel, 2003). In the photosynthetic bradyrhizobia, carotenoids protect against oxidative stress and function as part of the photosynthetic system, which may facilitate bacterial infection and contribute to symbiotic effectiveness (Giraud et al., 2000; Giraud & Fleischman, 2004).

Carotenoids are also prevalent in organisms inhabiting the phyllosphere and aquatic ecosystems, due to their photoprotective effect against ultraviolet radiation (Jacobs et al., 2005; Laurion et al., 2002). The presence of carotenoids in pigmented rhizobia may confer a selective advantage for bacterial survival on stems and hypocotyls, and in aquatic habitats, as compared with non-pigmented, soil-dwelling rhizobia. It should be noted, though, that other aquatic legumes such as Neptunia natans and

106 Chapter 2

Sesbania rostrata are nodulated by the non-pigmented rhizobia Devosia neptuniae and Azorhizobium caulinodans, respectively (Dreyfus et al., 1988; Rivas et al.,

2002).

In summary then, the Lotononis s. l. clade is nodulated by a remarkable diversity of rhizobia, with symbiotic associations that range from promiscuous through to highly specific. Given the above, is there a hypothesis that can explain the evolution of the symbiotic patterns seen in this group of legumes and their associated microsymbionts?

2.4.6 Putative evolution of symbiotic patterns within Lotononis s. l.

Edaphic heterogeneity driven by geomorphic change in the early and late

Miocene (23.8–5.3 mya), and simultaneous climatic deterioration, are hypothesised to have triggered the radiation of the Cape floral lineages (including the Lotononis s. l. clade) and their subsequent speciation (Cowling et al., 2009; Edwards & Hawkins,

2007). The edaphic heterogeneity of this landscape would presumably also promote the genotypic diversity of rhizobia available as potential symbiotic partners of these legume species, due to the varying saprophytic competencies of different rhizobial genera and strains (Graham, 2008).

It is proposed that in response to this, two specificity groupings have arisen in

Lotononis s. l. species. In the first group, Leobordea and Lotononis s. str. species are more or less promiscuous and able to nodulate, with varying degrees of effectiveness, with soil rhizobia that have a diversity of both chromosomal backgrounds and symbiotic genes. In the second group, the adaptation of Listia spp. to waterlogged habitats has consequently required the selection of microsymbionts that are more specialised inhabitants of these environments. The comparative rarity

107 Chapter 2 of rhizobial species of Microvirga and Methylobacterium suggests that their capacity for nodulation has been acquired through horizontal transfer of symbiotic genes from other rhizobial genera.

The mechanisms of HGT are favoured in closely related microorganisms

(Ochman et al., 2000). This concept accords with the results of Wernegreen & Riley

(1999), who have proposed that HGT in rhizobia is restricted across major chromosomal subdivisions, but supported among congeneric strains. A recent review of rhizobial symbiovars also supports this concept, as the biovars were nearly always found in species belonging to the same rhizobial genus (Rogel et al., 2011).

Nandasena et al. (2007) and Sullivan & Ronson (1998) have demonstrated that in mesorhizobia, HGT between closely related strains can result in the rapid evolution of symbiotic bacteria. In the Lotononis s. l. rhizobia, the 100% sequence identity of the bradyrhizobial nodA genes and likewise the 100% sequence identity of nodA in the Ensifer strains lend weight to the theory of HGT occurring among closely related strains.

It is presumed therefore that HGT between distantly related bacteria is a rarer event than that between closely related strains. That HGT between unrelated strains does occur in rhizobia has been documented in a study of Brazilian soybean isolates, where symbiotic genes from Bradyrhizobium japonicum inoculant strains were transferred to a presumably more saprophytically competent indigenous Ensifer fredii strain (Barcellos et al., 2007). Newly acquired genes require integration into the existing cellular regulatory circuits (Masson-Boivin et al., 2009). Moulin et al.

(2004) have suggested that the low frequency of nodulation gene transfer between

108 Chapter 2

Bradyrhizobium and distant rhizobial genera implies that symbiotic genes from the former require the correct chromosomal background to function. Interestingly, a study of 165 sequenced microbial genomes shows that B. japonicum USDA110 is an important “hub” for HGT, as it has one of the highest numbers of HGT partners

(Kunin et al., 2005).

Thus, in the proposed model of the evolution of symbiotic patterns in

Lotononis s. l., the radiation and subsequent speciation of ancestral plants has led to transference of symbiotic genes from the originally associated rhizobia to diverse, more saprophytically competent strains. A selective advantage is conferred upon rhizobial strains that successfully integrate the symbiotic genes into their chromosomal background and are able to nodulate the host plants. The symbiotic genes may then be rapidly disseminated via HGT to closely related rhizobial populations. Over time, the symbiotic genes evolve within each background, leading to allelic variation.

Species within Leobordea and Lotononis s. str. appear to have maintained the symbiotic promiscuity that Perret et al. (2000) consider to be the ancestral state. In contrast, the adaptation of Listia species to waterlogged habitats appears to involve the selection of microsymbionts that are more saprophytically competent in this environment. The waterlogged habitat may restrict the number of other rhizobial species that are available for HGT, in addition to the cellular mechanisms that restrict gene flow in genetically divergent organisms (Papke & Ward, 2004), thus contributing to symbiotic isolation for both the microsymbiont and the legume host.

In the Methylobacterium-Listia symbiosis, the extreme symbiotic specificity that has

109 Chapter 2 developed may be due to coevolution of the plant hosts and the rhizobia. Aguilar et al. (2004) have previously suggested that coevolution occurs in the centres of host genetic diversity for the symbiosis between Phaseolus vulgaris and Rhizobium etli.

The symbiosis of the more tropically adapted L. angolensis with rhizobial strains of

Microvirga appears to be a case of symbiont replacement. Such replacement has also been documented in the Phaseoleae tribe, in which species of Rhizobium have replaced Bradyrhizobium strains as the preferred microsymbionts in two Phaseolus species (Martínez-Romero, 2009).

Symbiotic specificity, and the mechanisms that govern rhizobial infection and nodule initiation have been studied extensively in a small number of legume hosts

(Goormachtig et al., 2004; Madsen et al., 2010; Oldroyd & Downie, 2008), but not in Listia species. A first step might be to investigate the methods by which the

Methylobacterium and Microvirga rhizobia infect and nodulate their hosts. To this end, the processes of infection and nodule initiation were studied in L. angolensis and L. bainesii, and are detailed in Chapter 3.

110 Chapter 3

CHAPTER 3

Nodule structure in Lotononis s. l. species and infection and nodule formation in Listia angolensis and Listia bainesii

111 Chapter 3

112 Chapter 3

3.1 Introduction

3.1.1 Modes of infection and nodulation in legumes

The route by which rhizobial infection proceeds and the shape and structure of the resulting nodule are under the control of the host plant and have taxonomic value (Corby, 1981; Lavin et al., 2001; Sprent & James, 2007). Two modes of infection and several nodule types, characteristic of the different legume groups in which they are found, have been described (Guinel, 2009; Sprent, 2007) (see also

Section 1. 7).

Within the Papilionoideae, rhizobial infection may proceed intracellularly via root hair curling, with the bacteria confined to an infection thread (IT) before their release into nodule primordium cells, or may be directly through the epidermis with no ITs involved in nodulation (Sprent & James, 2007). The former is characteristic of phaseoloid and hologalegoid legumes, while the latter is found in the more basal genistoid and dalbergioid clades. The nodules of root-hair-infected legumes contain central tissue that has a mixture of infected and uninfected cells. These nodules may be either indeterminate, with a persistent meristem and a gradient of developmental zones; or determinate (desmodioid), with a short-lived meristem and all cells at a similar developmental stage (Sprent & James, 2007). The genistoid and dalbergioid nodules, in which infection threads are absent, are characterised by a central mass of uniformly infected tissue. Dalbergioid nodules are determinate and associated with a lateral or adventitious root, while genistoid legumes have indeterminate nodules

(Lavin et al., 2001; Sprent & James, 2007). A variant of the indeterminate form is the lupinoid nodule, in which lateral meristems encircle the subtending root. In the genistoid tribe Crotalarieae, examples of indeterminate and lupinoid nodules are

113 Chapter 3 found in Crotalaria species and Listia species, respectively (Renier et al., 2011;

Yates et al., 2007).

Root hair infection is considered to be a feature of the more phylogenetically advanced legumes, in which recognition of Nod factors (NFs) by LysM receptor-like kinases in the epidermis of the root hair triggers a signalling pathway that controls infection and nodule formation (Madsen et al., 2010; Oldroyd & Downie, 2008).

This mode of infection is believed to provide the legume with a tighter and more selective control of bacterial passage through the epidermis (Madsen et al., 2010). In contrast, it has been speculated that the promiscuous nodulation seen in Arachis and

Stylosanthes may be attributed to their crack entry mode of infection (Boogerd & van

Rossum, 1997).

The rhizobial infection pathway in Listia species has not been determined, but the internal structure of the lupinoid nodules found in studied plants is that of a typical genistoid legume, with a central mass of uniformly infected tissue (Yates et al., 2007). Rhizobial infection in other genistoid legumes occurs via epidermal entry, rather than by root hair curling (González-Sama et al., 2004; Tang et al., 1992;

Vega-Hernández et al., 2001). If the mode of infection in Listia species does indeed occur via epidermal entry, then the Listia- Methylobacterium symbiosis represents an interesting example of symbiotic specificity in legumes with a non root-hair- mediated infection process.

Little is known of the molecular mechanisms that govern specificity and signalling pathways in non-IT legumes. An examination of the molecular dialogue

114 Chapter 3 and signal pathways that underlie rhizobial infection and nodule morphogenesis in the Listia –Methylobacterium and Listia -Microvirga symbioses is beyond the scope of this study. An investigation of the infection route and the development of nodule primordia would, however, be a useful first step in determining what mechanisms may govern specificity in these symbioses. It is also worthwhile to study the internal structure of the indeterminate nodules found in Lotononis s.l. species outside the genus Listia, as this may provide further evidence as to whether nodule structure is a useful taxonomic marker for the genistoid clade. For these purposes, light microscopy was used to observe the processes of infection and nodulation in Listia angolensis and Listia bainesii and to examine nodule sections from Lotononis s. l. species featured in Chapter 2.

3.2 Materials and Methods

3.2.1 Legume hosts, bacterial inoculant strains and growth conditions

Internal nodule structure was examined for selected nodules of the Listia,

Leobordea and Lotononis s. str. host species featured in Chapter 2. The species examined and their inoculant strains are listed in Table 3.1. The preparation of plant material, inocula and growth conditions is described in Chapter 2.

115 Chapter 3

Table 3.1. List of host plants and rhizobial strains featured in this chapter. Lotononis s. l. Lotononis s. l. Inoculant strain Nodule genus and section* species phenotype

Listia L. angolensis Methylobacterium sp. WSM2598 Fix- Microvirga sp. WSM3557 Fix+ L. bainesii Methylobacterium sp. WSM2598 Fix+ Leobordea L. mollis Methylobacterium nodulans Fix+ (Leptis) WSM2667 Leobordea L. platycarpa Ensifer meliloti WSM2653 Fix+ (Leobordea) Leobordea L. bolusii Bradyrhizobium sp. WSM2632 Fix+ (Synclistus) Lotononis s. str. L. crumanina Ensifer meliloti WSM2653 Fix+ (Oxydium) Lotononis s. str. L. pungens Ensifer meliloti WSM2653 Fix+ (Cleistogama) *Based on the taxonomy of van Wyk (1991) and Boatwright et al. (2011).

To study infection and nodulation, L. angolensis (Accession number

2004CRSL69) and L. bainesii (cv. Miles) were inoculated with their respective microsymbiont strains WSM3557 and WSM2598. Plants were grown in a naturally lit, controlled temperature (maximum 24°C) glasshouse, in either pots (using an axenic sand culture system (Howieson et al., 1995; Yates et al., 2007)) or in gnotobiotic growth pouches (CYG Seed Germination Pouch, Mega International,

West St. Paul, MN, USA). For the sand culture system, the preparation of pots was as previously detailed in the Materials and Methods for Chapter 2.

The methods for growing seedlings in growth pouches were based on protocols described in Journet et al. (2001). Growth pouches were prepared by adding 10 ml of 1x Fahräeus solution (modified from Vincent (1970) and containing

0.5 mM MgSO4, 0.7 mM KH2PO4, 0.8 mM Na2HPO4, 50 µM Fe-EDTA and 0.1 µg each of MnSO4, CuSO4, ZnSO4, H3BO3 and Na2MoO4) to each pouch. The pouches were wrapped in alfoil (to keep plant roots in the dark) and sterilised by autoclaving

116 Chapter 3

(121ºC, 20 min). After autoclaving and just prior to seeds being sown, 100 µl of 100 mM CaCl2 was added to each pouch, to give a final concentration of 1 mM CaCl2.

Seeds were prepared according to the protocols detailed in Chapter 2. For pot-grown plants, the seeds were germinated in the dark at room temperature on water agar

(1.0% (w/v)) plates and aseptically sown into pots (six seeds per pot) when the radicles were 1 – 3 mm in length. Inocula were prepared according to the protocols detailed in Chapter 2 and applied after the seedlings had emerged (from five to ten days after sowing). Selected pots of L. angolensis and L. bainesii seedlings were not inoculated until the plants were 85 days old.

For growth pouches, shaking broth cultures (5 ml) were grown in TY medium

(28ºC, 200 rpm) to late log phase, then diluted with TY medium to an OD600nm of

0.5. Seeds were germinated as above. Three-day-old seedlings were then transferred to sterile Petri dishes and incubated with 5 ml of the diluted broth cultures for one h, to allow the bacteria to colonise the seedlings. The seedlings were then planted aseptically in the growth pouches, which were then placed inside surface sterilised boxes fitted with clear PVA plastic lids. Moistened paper towels were placed in the boxes to maintain humidity and prevent drying out of the growth pouches, after which the boxes were transferred to the glasshouse. At seven days after inoculation

(dai), a further 5 ml of sterile deionised water, 5 ml of 1x Fahräeus solution and 100

µl of 100 mM CaCl2 were added aseptically to each pouch.

3.2.2 Harvesting

Plants were harvested at regular intervals. At harvest, plants grown in growth pouches were examined for nodulation. The roots of pot-grown plants were washed

117 Chapter 3 and sprayed with water to remove all soil and organic matter and similarly examined for nodulation.

3.2.3 Nodule initial staining and sectioning, nodule sectioning and light microscopy

Sampled plants were stained to highlight root and nodule initials, following a protocol adapted from O’Hara et al. (1988). The plants were initially stored in a fixative solution of 7% (v/v) acetic acid in absolute ethanol. After storage, whole plants were cleared in a 10% (w/v) KOH solution for 4.5 h, followed by washing in running water to remove any discoloration. They were then acidified in a solution of

0.25 M HCl for five min, before being placed in a 0.1% (w/v) Brilliant Green (No.

C086, ProSci Tech, QLD, Australia) solution for 30 min, followed by destaining overnight in tap water.

Sections of fresh plants that included the hypocotyl and the upper portion of the main root were excised and fixed overnight at 4°C in 3% (v/v) glutaraldehyde in

25 mM phosphate buffer (pH 7.0) in preparation for sectioning. Fixed material was washed in three changes of phosphate buffer. The samples were dehydrated in a rotator using a series of acetone solutions (30%, 50%, 70%, 90% and 100%) at 41ºC, with two changes of each solution, each of 15 min duration. Dehydrated samples were infiltrated with Spurr’s resin mixed with acetone, using an increasing succession of resin concentrations (5%, 10%, 15%, 20%, 30%, 40%, 50%, 70% and

90%). The material was kept in each solution for a minimum of 2 h. Infiltrated material was transferred into 100% Spurr’s resin, left at room temperature for 1–2 h and then transferred into fresh 100% Spurr’s resin for 5–8 h at room temperature or left overnight at 41ºC. Finally, in order to obtain good polymerisation, the material

118 Chapter 3 was embedded in fresh Spurr’s resin at 60ºC for 24 h. For light microscopy, 1 µm sections were cut using a Reichert-Jung 2050 microtome with a glass knife. Sections were dried onto glass slides at 60ºC and stained with 1% (w/v) methylene blue and

1% (w/v) azur II in 1% (w/v) sodium tetraborate (Richardson et al., 1960) for 3–5 min at room temperature. Stained sections were rinsed in water then dried. Excised nodules from selected Lotononis s. l. host species (Chapter 2) were similarly prepared for sectioning and microscopy. Specimens were examined under an

Olympus BX51 photomicroscope and photographed with an Olympus DP70 camera.

3.3 Results

3.3.1. Localisation of infection site

Pouch-grown seedlings of L. angolensis and L. bainesii were examined daily under a stereomicroscope for nodule initials. They were also examined under a photomicroscope for nodule and lateral root initials, after first being cleared and stained with Brilliant Green. Seedlings grown in pots were examined under a photomicroscope at seven, ten and 16 days after inoculation (dai) for nodule and lateral root initials, after clearing and staining. In both species and in both growth systems, nodule initials could be clearly distinguished at 16-17 dai. On seedlings of both L. angolensis and L. bainesii, nodule initials occurred most frequently on the plant hypocotyl. This was usually just above the root-shoot transition zone, which was marked by large numbers of root hairs, whereas the hypocotyl was mostly devoid of these (Figure 3.1). Nodule initials were also found further down the taproot. The root hairs did not appear to be involved in the infection process, as they were not seen on the epidermal surface of most of the hypocotyl nodule initials, although two root hairs seem to have arisen from basal cells on the epidermal surface of the nodule initial in Figure 3.1 e. Curled or deformed root hairs were not observed. 119 Chapter 3

Figure 3.1. Light microscopy of cleared and stained Listia angolensis and Listia bainesii plants, showing lateral root and nodule initials. The root zone is marked by large numbers of root hairs, which can be seen just below the hypocotyl. a, b) L. angolensis 17 days after inoculation (dai). c, d) L. bainesii 16 dai. e) L. bainesii 16 dai, showing lateral root and nodule initial. f) L. bainesii 16 dai showing lateral roots and nodule initials. H, hypocotyl; LR, lateral root; LRI, lateral root initial; NI, nodule initial; RH, root hair; S, stele.

Lateral root initials could also be seen along the hypocotyl and taproot, but could be distinguished from nodule initials by their position adjacent to the plant stele (Figure 3.1 e). In contrast, nodule initials arose in the outer cortex. Nodule

120 Chapter 3 initials in L. angolensis and L. bainesii were not associated with lateral roots, as they did not form in the junctions of these roots and the main root, although some nodules could be found close to emerging lateral roots (Figure 3.1 f).

3.3.2 Infection and nodule organogenesis in Listia angolensis and Listia bainesii

Transverse sections of the shoot/root junction of L. angolensis and L. bainesii seedlings showed root hairs, although these were not curled or deformed and infection threads were not observed (Figures 3.2 and 3.3). In L. angolensis the nodule primordium developed in the outer cortical layer immediately beneath an enlarged epidermal cell (EE) (Figure 3.2 b). At six dai, there was a proliferation of cells in the outer cortex directly under the enlarged epidermal cell. Both anticlinal and periclinal cell division was observed, and cells in the infection zone had prominent nuclei and dense cytoplasm. There was no simultaneous division of pericycle cells; rather, cell division appeared to spread from the outer to the inner cortex. The orientation of the plane of division of inner cortex cells was more commonly anticlinal. A similar pattern of nodule development was observed in L. bainesii. At six dai, a cluster of newly divided cells with prominent nuclei had formed in the outer cortex (Figure 3.3 a and b). At ten dai, the central tissue of the nodule primordium was uniformly infected with bacteria that did not appear to have differentiated into bacteroids

(Figure 3.3 c, d and e). Infection pockets, caused by the collapse and death of cortical cells, were not observed in either L. angolensis or L. bainesii.

121 Chapter 3

NP C X S

E

RH

C

a 200 µm

NP C NP EE EN S EN

RH C

b 50 µm c 50 µm

Figure 3.2. Light micrographs of Listia angolensis hypocotyl sections (transverse to the primary root axis) showing nodule primordia development after inoculation with WSM3557. a) Whole section, six days after inoculation. b) Higher magnification of the same section. c) Section from the same plant (resin has not been etched). Cells in the nodule primordium have divided repeatedly and show cytoplasmic staining and prominent nuclei (arrow). Cell division can also be seen in the inner cortex (arrowheads). C, cortex; E, epidermis; EE, enlarged epidermal cell; EN, endodermis; NP, nodule primordium; RH, root hair; S, stele; X, xylem

122 Chapter 3

S E C EN S C NP NP

a 200 µm b 50 µm

S

NP NP

d c 50 µm 50 µm

R

e 50 µm

Figure 3.3. Light micrographs of Listia bainesii hypocotyl sections (transverse to the primary root axis) showing nodule primordia development after inoculation with WSM2598. a) Whole section, six days after inoculation (dai) and b) at higher magnification. Cells in the nodule primordium have divided repeatedly and show cytoplasmic staining and prominent nuclei (arrow). Cell division can be seen in the inner cortex (arrowhead). c) Nodule primordium, ten dai and d) section from the same nodule (resin has not been etched). Cells in the nodule primordium are uniformly infected with bacteria. Another locus of infection is visible in the bottom left hand corner. e) The same section at higher magnification, showing details of rhizobia (arrow) infecting nodule primordium cells. C, cortex; E, epidermis; EN, endodermis; NP, nodule primordium; R, rhizobia; S, stele

123 Chapter 3

3.3.3 Localisation of the infection site in Listia angolensis and Listia bainesii plants inoculated at 85 days old

Uninoculated 85-day-old L. angolensis and L. bainesii plants were not much larger than young seedlings, possibly due to the small size of Listia spp. seeds. This growth pattern is typical of uninoculated L. angolensis and L. bainesii plants, in which leaves grow, but the internodes of the stem remain very short, resulting in a growth form similar to that of a rosette plant. The nodulation pattern in the plants inoculated at 85 days old (Figure 3.4) did not appear to differ from that observed in seedlings inoculated at seven days old (Figure 2.15 a, b; Chapter 2). The location of the infection site did not change, as collar nodules developed on the hypocotyl and main root, and occasionally on lateral roots, just as in the seedlings inoculated at seven days old.

Figure 3.4. Nodulation pattern in Listia angolensis and Listia bainesii plants inoculated at 85 days old and harvested 33 days after inoculation. a) L. angolensis b) L. bainesii.

124 Chapter 3

3.3.4 Internal structure of Lotononis s. l. nodules

Nodules taken from ten-week old inoculated Listia, Leobordea and Lotononis s. str. host species (Chapter 2) were sectioned and examined by light microscopy to compare their internal structure. In all cases, effective nodules contained uniformly infected central tissue with no uninfected interstitial cells, whether from lupinoid nodules of L. angolensis and L. bainesii (Figure 3.5 a – f) or from the indeterminate nodules of Lotononis s. l. species outside the genus Listia (Figure 3.6 a – f). The degree of vacuolation of bacteroid-containing cells appeared to vary between species

(Figure 3.5 a and f; Figure 3.6 a and d). In some nodules (Figures 3.5 b and 3.6 f), amyloplasts could be seen in the cortical uninfected cells next to the nodule central tissue. L. angolensis inoculated with WSM2598 produced ineffective nodules in which the bacteria were not differentiated into nitrogen-fixing bacteroids (Figure 3.5 c – e). Interestingly, bacteroids of WSM2598 in nitrogen-fixing nodules of L. bainesii (Figure 3.7 a) appeared to be far more swollen than the WSM3557 bacteroids in effective L. angolensis nodules (Figure 3.7 b).

125 Chapter 3

Figure 3.5. Light microscopy of sections of ten-week-old Listia angolensis and Listia bainesii nodules. In both species, the central tissue (*) of effective nodules contains infected cells only. a, b) Nitrogen-fixing nodule of L. angolensis inoculated with WSM3557. Amyloplast structures (arrow) can be seen in the peripheral cortical uninfected cells. c, d, e) Non-fixing nodule of L. angolensis inoculated with WSM2598. f) Nitrogen-fixing nodule of L. bainesii inoculated with WSM2598. S, stele; VB, vascular bundle

126 Chapter 3

Figure 3.6. Light microscopy of sections of ten-week-old Leobordea spp. and Lotononis s. str. spp. nodules. In all nodules, the central tissue (*) of effective nodules contains infected cells only. a) Leobordea bolusii inoculated with WSM2632. b) Leobordea mollis inoculated with WSM2667. c) Leobordea platycarpa inoculated with WSM2653. d) Lotononis crumanina inoculated with WSM2653 e, f) Lotononis pungens inoculated with WSM2653. Structures resembling amyloplasts (arrow) can be seen in the peripheral cortical uninfected cells.

127 Chapter 3

Figure 3.7. Light microscopy of sections of ten-week-old nodules, showing bacteroids. a) Listia bainesii inoculated with WSM2598. b) Listia angolensis inoculated with WSM3557.

3.4 Discussion

Nodulation in L. angolensis and L. bainesii occurs mainly on the hypocotyl and taproot. Remarkably, the infection site appears to be independent of the age of the plant at the time of inoculation, as the same pattern of nodulation was observed for plants inoculated at 85 days old as those inoculated at seven days old. This is in contrast to other studies of both root-hair-curl- and epidermally-infected legumes.

128 Chapter 3

Rhizobial invasion in legumes infected via root hair curling is confined to a small zone of root hairs that have nearly finished growing (root hair zone II) and are susceptible to deformation and infection (Bhuvaneswari et al., 1980, 1981; Gage,

2004; Heidstra et al., 1994; Maunoury et al., 2008). In vetch, about 80% of zone II root hairs were deformed three hours after exposure to NodRlv factor (Heidstra et al.,

1994). Similarly, rhizobial invasion in the epidermally infected genus Lupinus appears to be localised within a transient zone. In Lupinus angustifolius, nodules appeared most frequently in the region between the smallest emergent root hairs and the root tip at the time of inoculation. Epidermal root cells aged 13 h or over appeared not to be infected (Tang et al., 1992). Rhizobia preferentially accumulate in the root hair portion of the main root of Lupinus albus seedlings, c. 10–15 mm from the root tip (González-Sama et al., 2004).

Although infection and nodule formation in dalbergioid and genistoid legumes does not proceed via root hair curling and infection thread development, root hair deformation and transient infection threads have been observed in some species belonging to these clades. Bradyrhizobium strain BTA-1 induced root hair deformation and curling, and the development of infection threads (that subsequently aborted), in the genistoid legume tagasaste (Cytisus proliferus; formerly

Chamaecytisus proliferus) (Vega-Hernández et al., 2001). Deformation and curling of root hairs has been observed in Arachis, Stylosanthes and Lupinus species

(Boogerd & van Rossum, 1997; Chandler et al., 1982; González-Sama et al., 2004;

Łotocka et al., 2000). Structures similar to short, wide infection threads have occasionally been found in L. angustifolius (Tang et al., 1992) and L. albus nodules

(James et al., 1997). Infection threads were not found during the nodulation process

129 Chapter 3 in the Listia species (this work) and similarly were not seen in Lupinus albus nodulation (González-Sama et al., 2004) or in the invasion zone of Crotalaria podocarpa nodules (Renier et al., 2011).

Root hair cells seem to be important in the initial infection process in some dalbergioid and genistoid legumes. Rhizobia invade the middle lamella between enlarged basal hair cells and adjacent root hair or epidermal cells in Arachis and

Lupinus species (Boogerd & van Rossum, 1997; González-Sama et al., 2004; Tang et al., 1992). In Arachis, rhizobial infection is associated with tufts of root hairs that arise in the axils of young lateral roots and non-nodulation is strongly correlated with an absence of these hairs. Successful infection is restricted to penetration sites where enlarged root hair basal cells are found (Boogerd & van Rossum, 1997; Uheda et al.,

2001). Conversely, in Aeschynomene afraspera stem-nodulation sites, root hairs are not observed in the epidermis of the lateral root primordium (Alazard & Duhoux,

1990). Further study is required to determine whether or not root hair cells are involved in rhizobial infection of the hypocotyl in Listia angolensis and Listia bainesii. It could be hypothesised that in these Listia species, the ability of the hypocotyl to remain potent for infection may be correlated with the putative lack of a role for root hair cells.

Nodule organogenesis in Listia angolensis and Listia bainesii appears to follow a process similar to that observed in Lupinus albus and Lupinus angustifolius

(González-Sama et al., 2004; Tang et al., 1992). In the Lupinus studies, rhizobia penetrate intercellularly at the junction between epidermal cells and subsequently invade a cortical cell immediately beneath the epidermis. The infected cell divides

130 Chapter 3 rapidly, as do the rhizobia within the cell, to produce a nodule with characteristic uniformly infected cells in the central infected zone (Fernández-Pascual et al., 2007).

The incidence of cell division spreads progressively from the infection focus towards the inner cortex, with cellular division in the deeper cortical layers occurring several days after the initial infection (González-Sama et al., 2004; Tang et al., 1992). The same pattern was seen for nodule morphogenesis in the Listia species. The proliferation of outer cortical cells formed the nodule primordium and cellular division subsequently spread to the inner cortex. In this, the lupinoid nodule primordium is more similar to that of the desmodioid nodule, in that desmodioid nodule initiation (seen for example in Lotus japonicus) is typically found in the first outer cortical layer, immediately beneath the primary infection site (Guinel, 2009;

Szczyglowski et al., 1998). Conversely, both aeschynomenoid and indeterminate hologalegoid nodules arise from divisions in the root inner cortex (Alazard &

Duhoux, 1990; Guinel, 2009; Voroshilova et al., 2009). As well, in the aquatic robinioid legume Sesbania rostrata (where infection can occur intercellularly at lateral root bases) and in the aeschynomenoid genera Aeschynomene and

Stylosanthes, the infection process is associated with the collapse and death of cortical plant cells (Alazard & Duhoux, 1990; Chandler et al., 1982; D'Haeze et al.,

2003). In contrast, cell death does not appear to occur during the nodulation process in Listia species.

In keeping with the morphology found in typically genistoid nodules (Sprent,

2009) nodule sections of all Lotononis s. l. host species, whether of lupinoid or indeterminate nodules, had a mass of central, uniformly infected tissue, with no uninfected interstitial cells. It is interesting to note that nitrogen-fixing bacteroids of

131 Chapter 3

WSM2598 within L. bainesii nodules should appear to be more swollen than those of

WSM3557 within L. angolensis nodules. It has been suggested that swollen bacteroids may confer advantages to the host, such as enhanced nitrogen fixation capability (Mergaert et al., 2006; Oono & Denison, 2010). The symbiosis between

WSM2598 and L. bainesii is highly effective (Yates et al., 2007), but further study would be required to determine if a relationship between bacteroid size and effectiveness exists in this system.

In summary, the mechanism of morphogenesis in lupinoid nodules, in which cell division is initiated in the outer cortex beneath the infection site, appears to be the same for both genistoid Lupinus and crotalarioid Listia species. The salient characteristics of infection and nodule organogenesis in Listia species – epidermal infection, the lack of infection threads and a central mass of uniformly infected tissue

- are consistent with the nodulation process observed in the dalbergioid and genistoid clades. The ability of the root and hypocotyl epidermal cells to remain potent for infection is an interesting feature of the infection process in Listia, and one that merits further study.

132 Chapter 4

CHAPTER 4

Characterisation and description of novel Listia angolensis

and Lupinus texensis rhizobia

133 Chapter 4

134 Chapter 4

4.1 Introduction

The light-pink-pigmented rhizobial strains WSM3674 and WSM3686, isolated from nodules of Zambian Listia angolensis, have previously been identified as belonging to a novel lineage of root nodule bacteria (Yates et al., 2007). The 16S rRNA gene sequences of these strains showed them to be closely related to rhizobia that specifically nodulate Lupinus texensis plants growing in Texas, USA (Andam &

Parker, 2007). According to Andam & Parker’s phylogenetic tree, the closest relative of the Listia angolensis and Lupinus texensis strains was Microvirga flocculans

TFBT (previously Balneimonas flocculans), a strain isolated from a Japanese hot spring (Takeda et al., 2004; Weon et al., 2010).

The results of the present study have shown that strain WSM3557, an effective isolate of Lupinus angolensis, is also closely related to WSM3674,

WSM3686, the Lupinus texensis strains and M. flocculans (Chapter 2). These strains all grouped, with high bootstrap support, within the genus Microvirga (Chapter 2,

Figure 2.9), a member of the Methylobacteriaceae family and a bacterial lineage in which no strain has previously been identified as a legume symbiont. Microvirga was first described by Kanso & Patel (2003) from a single strain (FaiI4T) isolated from a bore sample of geothermal waters from the Australian Great Artesian Basin and given the name Microvirga subterranea. FaiI4T is thus the type strain of the genus.

The remaining described species are Microvirga guangxiensis (strain 25BT), obtained from rice field soil in Guangxi Province, China (Zhang et al., 2009) and Microvirga aerophila (5420S-12T) and Microvirga aerilata (5420S-16T), which were isolated from atmospheric samples taken in Suwon region, Republic of Korea (Weon et al.,

2010).

135 Chapter 4

The availability of six authenticated N2-fixing Listia angolensis strains in the

WSM Culture Collection (Chapter 2), together with 28 isolates from Lupinus texensis, presented an opportunity to develop polyphasic descriptions of this novel group of root nodule bacteria. A collaboration to achieve this aim comprised:

a) Professor Matt Parker at the State University of New York (SUNY), responsible for most of the sequencing of rRNA, housekeeping, nod and nif genes and the subsequent construction of phylogenetic trees.

b) Professor Anne Willems and her student Ms Sofie De Meyer at the

University of Ghent, included in the collaboration for their expertise in DNA: DNA hybridization, determination of G + C content and cellular fatty acid analysis.

c) Phenotypic data and additional DNA sequences, obtained by Ms Julie

Ardley at the Centre for Rhizobium Studies and detailed in this chapter.

Together, the data provide a basis to validly name and describe these novel rhizobial species of Microvirga.

Phenotypic studies were conducted initially on all Listia angolensis strains.

More detailed phenotyping was subsequently performed on the putative type strains

WSM3557T and WSM3693T. WSM3557T was chosen as it was the most effective strain for nodulation and nitrogen fixation on L. angolensis (Chapter 2), while the phenotypic and molecular studies performed on WSM3693T identified it as being sufficiently different to warrant consideration as a separate species. Lut5 and Lut6T were chosen as representative of the Lupinus texensis isolates because of the body of phenotypic and molecular data already obtained for these strains (Andam & Parker,

136 Chapter 4

2007), and Lut6T, as a putative type strain, was subsequently used for additional phenotyping.

4.2 Materials and Methods

4.2.1 Bacterial strains and culture conditions

The strains used in this study are listed in Table 4.1. The proposed type strains Lut6T (= LMG26460T; = HAMBI3236T), WSM3557T (= LMG26455T, =

HAMBI3237T) and WSM3693T (= LMG26454T, = HAMBI3238T) have been deposited in the BCCM/LMG and HAMBI Culture Collections. All strains were stored at -80ºC as 12% (v/v) glycerol/media stocks. Strains were routinely subcultured on modified ½ lupin agar (½ LA) (Yates et al., 2007), on ½ LA with succinate replacing glucose and mannitol as a carbon source, or on TY agar

(Beringer, 1974). Broth cultures were grown in TY or ½ LA. Plate and broth culture was grown at 28ºC or 37ºC and shaking cultures were incubated on a gyratory shaker at 200 rpm.

4.2.2 DNA amplification and sequencing

4.2.2.1 Amplification and sequencing of the 16S rRNA gene

Amplification and sequencing of the WSM3557T 16S rRNA gene has been detailed in Chapter 2. Nearly full length PCR amplification and sequencing of the

16S rRNA gene from WSM3693T was performed using the universal eubacterial primers FGPS6 and FGPS1509 (Nesme et al., 1995; Normand et al., 1992) and internal primers designed by Yanagi & Yamasato (1993) and Sy et al. (2001), according to the protocols given in Chapter 2.

137 Chapter 4

Table 4.1. List of strains used in this chapter.

Strain Synonym Host Geographical source Reference or (Collector) source

WSM3557T CB1322 Listia Chipata (Fort Eagles & Date LMG26455T angolensis Jameson), Zambia (1999) (Verboom) WSM3673 CB1321 Listia Chipata (Fort Eagles & Date angolensis Jameson), Zambia (1999) (Verboom) WSM3674 CB1323 Listia Chipata (Fort Eagles & Date angolensis Jameson), Zambia (1999) (Verboom) Yates et al (2007) WSM3675 CB2406 Listia Mbale,Zambia Eagles & Date angolensis (Verboom) (1999) WSM3686 CB1297 Listia Chipata (Fort Eagles & Date angolensis Jameson), Zambia (1999) (Verboom) Yates et al (2007) WSM3693T CB1298 Listia Chipata (Fort Eagles & Date LMG26454T angolensis Jameson), Zambia (1999) (Verboom) Lut5 Lupinus Texas, USA Andam & Parker texensis (Parker) (2007) Lut6T LMG26460T Lupinus Texas, USA Andam & Parker texensis (Parker) (2007) Reference strains Strain Host Comments Reference or source

Bacillus subtilis Hydrolyses starch Murdoch collection Produces acetoin Escherichia coli DH5α Negative control for Bethesda Research antibiotic tests Laboratories (1986) Escherichia coli KS272 Facultative anaerobe Guzman et al. (1995) Mesorhizobium cicerae Biserrula SmR derivative of Nandasena et al. WSM4114 pelecinus WSM1497 (2007) Mesorhizobium sp SpR derivative of CJ4 Sullivan et al. (1998) WSM4116 Methylobacterium sp Listia Pigmented Yates et al. (2007) WSM2598 bainesii Pseudomonas Obligate aerobe Murdoch collection fluorescens Ensifer fredii NGR234 Lablab RifR Trinick (1980) purpureus Ensifer medicae Medicago AmpR, CmR, NalR W. Reeve WSM419 murex (unpublished data) Ensifer medicae Medicago Derivative of WSM419 G. Garau MUR2110 murex GmR, KmR, TcR (unpublished data)

138 Chapter 4

Searches for sequences with high sequence identity to the sample 16S rDNA were conducted using BLASTN (Altschul et al., 1990) against sequences deposited in the National Centre for Biotechnology Information GenBank database. Genetool

Lite was used to align and calculate the pairwise percent sequence identity of a 1396 bp fragment of the 16S rRNA gene from the Listia angolensis and Lupinus texensis strains and the Microvirga species type strains. A phylogenetic tree was constructed with MEGA version 4.0 (Tamura et al., 2007), using the neighbour-joining (NJ) method (Saitou & Nei, 1987) and the Maximum Composite Likelihood model, and bootstrapped with 1000 replicates.

4.2.2.2 Amplification and sequencing of the nodA gene

Amplification and sequencing of the WSM3557T nodA gene has been detailed in Chapter 2. Amplification and sequencing of the nodA gene from WSM3693T was performed using the nodA primers developed by Haukka et al. (1998) and according to the protocols given in Chapter 2.

4.2.3 Phenotypic characterisation

4.2.3.1 Morphology

Colony morphology was studied on ½ LA plates. Strains were also assessed for growth on nutrient agar. Gram staining was performed according to standard methods (Vincent, 1970). Motility of exponential phase broth cultures was observed using a light microscope and the hanging drop method. To try to induce motility in the Lut5 and Lut6T strains, they were also grown, using a method modified from

Bowra & Dilworth (1981), on JMM minimal media plates (O'Hara et al., 1989) containing 0.1 mM of succinate as a carbon source, 0.05 % (w/v) yeast extract, 0.1 mM EDTA and 0.3% agar. One drop of 0.3 mM MgSO4 solution was applied to the

139 Chapter 4 edge of the resulting two day old culture and the cells resuspended by gentle pipetting, then examined for motility as previously described. The motile isolate

WSM3557T served as a positive control.

For electron microscopy, resuspended cells were collected from overnight ½

LA slopes to which 100 µl of sterile deionised (DI) water had been added. Scanning electron microscope (SEM) samples were prepared from aliquots of two day old plate culture resuspended in sterile reverse osmosis deionised (RODI) water. Strains were examined for spore formation by light microscopy after staining of stationary phase broth and plate cultures with malachite green (Beveridge et al., 2007).

Stationary phase cultures were also heated to 70ºC for 10 min, and then reinoculated onto fresh media and observed for growth.

4.2.3.2 Extraction of pigments

Plate cultures of WSM3557T and Lut6T were incubated in a 28ºC growth cabinet fitted with a Sylvania T8 F15W/GRO lamp that emitted radiation in the blue and red spectrum. The cultures were grown for nine days in total, with light supplied for 15 h per day for the first six days and the cultures subsequently grown without light. Plate culture (0.1 ml volume) was resuspended in 0.89% (w/v) saline, and centrifuged at 10, 000 x g for one minute to collect the pellet. Pigments were extracted with aliquots (3 x 1 ml) of a 7:2 (v/v) acetone: methanol mix at 0ºC under dim light. Absorption spectra of the extracts were recorded from 400-850 nm with a

Shimadzu UVmini-1240 scanning spectrophotometer and compared with the absorption spectrum obtained for the pink-pigmented methylobacterial strain

WSM2598 (Table 1) grown and extracted under the same conditions.

140 Chapter 4

4.2.3.3 Temperature range

The growth of strains was assessed over a range of temperatures (10-50ºC, at intervals of 5ºC, with 1ºC intervals from 38-46ºC) by resuspending a loopful of fresh plate culture in ½ LA broth and streaking the resulting culture onto ½ LA plates.

Plates were examined for growth up to 15 days after streaking.

4.2.3.4 Optimal growth temperature

Cultures were grown in TY broths to mid-log phase, and then subcultured into duplicate 50 ml TY broths in 250 ml conical flasks to give an initial optical density at OD600nm of approximately 0.05. The flasks were incubated in a shaking water bath (200 rpm) pre-set to the desired temperature and growth was measured by reading the OD600nm at regular intervals.

4.2.3.5 Growth curves and determination of mean generation time

Cultures were incubated at the optimum temperature, with growth measured as per Section 4.2.3.4 and the results plotted on a log/linear graph. Mean generation time was determined from reading the graph during the exponential growth phase; i.e. when the logarithm of absorbance against time produced a straight line.

4.2.3.6 Determination of sodium chloride and pH tolerance

Inocula were prepared by resuspending a loopful of fresh plate culture in ½

LA broth and incubating the cell suspension at 28ºC for 24 h to obtain stationary phase cultures. Aliquots (1 ml) were standardised to an OD600nm of 1.0 by the addition of ½ LA broth and serial dilutions made. An aliquot of 10 µl of the 10-1, 10-

3, 10-5 and 10-6 (or 10-7) dilutions was plated onto ½ LA media containing 0.0, 0.01,

0.5, 1.0, 1.5, 2.0, 2.5 or 3.0% (w/v) NaCl as shown in Figure 4.1. Plates were incubated at 28ºC and examined for growth daily, up to 6 days.

141 Chapter 4

10-1

10-3 Strain 1 Strain 2

10-5

10-7

Figure 4.1. Layout of aliquots of culture on NaCl plate.

To determine the ability of the strains to grow over a range of pH values, TY plates were prepared in the following manner: Universal indicator (Vogel, 1962) (5 ml l-1) was added to the TY medium and the following buffers, at 20 mM working concentration, were added for the pH values indicated:

MES (2-[N-morpholino]-ethane-sulfonic acid; pKa = 5.96): pH 5.5, 6.0

HEPES (4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid; pKa = 7.31): pH

7.0, 8.0, 8.5

CHES (N-Cyclohexyl-2-aminoethanesulfonic acid; pKA = 9.07): pH 9.00, 9.5,

10.0

Homopipes (Homopiperazine-n,n'-bis-2- (ethanesulfonic acid); pKa = 4.55):

pH 4.0, 4.5 and 5.0

The pH was adjusted by adding HCl or NaOH to the media. Agar (15 g l-1

(w/v) was added prior to autoclaving. For media at pH 4.0, 4.5 and 5.0, agarose was used as the setting agent, as agar plates did not set firmly enough at these values.

Media were prepared by combining separately autoclaved solutions of agarose and

142 Chapter 4 medium, as autoclaving the medium with agarose was found to affect the final pH.

Aliquots of culture were plated and examined for growth as for NaCl plates.

4.2.3.7 Anaerobic growth

Loopfuls of four day old plate culture were resuspended in 0.89% (w/v) saline, standardised to an OD600nm of 1.0 and serial dilutions made. Aliquots (100 µl) of the 10-5 dilutions were spread on plates of modified Hugh & Leifson’s medium

(Hugh & Leifson, 1953) containing either glucose or pyruvate as a carbon source.

One set of plates was placed in an anaerobic jar (BBL GasPac 100 Non-vented system), then two anaerobic generators (bioMérieux GENbox anaer, Ref 96 125) and an anaerobic indicator strip (bioMérieux, Ref 96 118) were added and the jar sealed.

The other set of plates was grown aerobically. Plates were incubated at 28ºC and examined for growth after ten days. Escherichia coli strain KS272 and Pseudomonas fluorescens (Table 4.1) served as positive and negative controls respectively.

4.2.3.8 Antibiotic sensitivity

Intrinsic antibiotic resistance was assessed for nine antibiotics at the following concentrations: ampicillin (50 and 100 µg ml-1), chloramphenicol (10, 20 and 40 µg ml-1), gentamycin (10, 20 and 40 µg ml-1), kanamycin (50 and 100 µg ml-

1), nalidixic acid (50 and 100 µg ml-1), rifampicin (50 and 100 µg ml-1), spectinomycin (50 and 100 µg ml-1), streptomycin (50 and 100 µg ml-1) and tetracycline (10 and 20 µg ml-1). Aliquots of culture were pipetted onto ½ LA plates containing antibiotics and examined for growth as per Section 4.2.2.4. Growth was compared with that of cultures inoculated onto TY plates devoid of antibiotics, and with positive and negative bacterial controls (Table 4.1).

143 Chapter 4

4.2.3.9 Substrate utilisation

4.2.3.9.1 Utilisation of sole carbon sources in Biolog GN2 plates

GN2-MicroPlates (Biolog Inc, CA, USA) were used to assess the ability of the strains to utilise sole carbon sources. The Biolog GN2 microplate consists of 96 wells containing 95 carbon sources and one blank well and is based on the catabolic production of NADH, which reduces a tetrazolium dye and results in a colour change from clear to purple in the microplate well (Bochner, 1989). In general, the manufacturer’s instructions for preparation and reading of the samples were followed, according to the guidelines recommended by Biolog for Ensifer meliloti strains. R2A agar (Reasoner & Geldreich, 1985), was used in place of the standard

Biolog Universal Growth (BUG) agar, as the strains grew poorly on the latter.

Fresh plate culture was streaked onto R2A agar plates, incubated for 24 h at

37ºC then resuspended in 15 ml of Biolog GN/GP Inoculation Fluid (IF) to a concentration of 85% ± 2% transmittance, as measured on a spectrophotometer.

Aliquots (150 µl) of the suspension were added to each well in duplicate microplates and incubated in a sealed container at 37ºC. Moistened paper towels were placed in the container to minimise evaporative loss from the wells. Colour development was monitored and microplate readings taken at 40, 72, 96 and 137 h using a Biorad 680 microplate reader with 595nm filter. The mean of the duplicate raw absorbance value for each well was expressed as a positive, negative or partial reaction, based on the cutoff values given for the Biorad software.

4.2.3.9.2 Utilisation of substrates in Biolog Phenotype Microarray plates

Additional phenotyping tests were performed on the provisional type strains

WSM3557T and Lut6T. Catabolism of carbon, nitrogen, phosphorus and sulfur 144 Chapter 4 compounds was assessed using Biolog Phenotype Microarray (PM) microplates

(panels PM1- 4 respectively). The PM technology is a high-throughput system that allows global analysis of cellular phenotypes (Bochner et al., 2001) The tetrazolium dye-based methodology is similar to that of the GN2 plates, but is a more sensitive system and uses a much wider range of substrates. Testing was performed by Biolog

PM Services (http://www.biolog.com/PM_Services.html) using the following protocol: The strains were grown on R2A agar prior to being suspended in inoculation fluid containing vitamin B12, biotin and dye mix G, with pyruvate as a carbon source in the phosphorus and sulfur panels. Duplicate cell suspensions at a concentration of 40% ± 2% transmittance were pipetted at 100 µl per well into the 96 well Biolog PM panels. Incubation at 37ºC and recording of phenotypic data over 56 h was performed by an OmniLog instrument. Utilisation of the substrate was scored as being either positive or negative, or was given as raw absorbance values.

4.2.3.9.3 Growth on sole carbon substrates

Growth supplement requirements were determined using JMM broths with succinate (20 mM) as the sole carbon source and NH4Cl (10 mM) replacing glutamate as a nitrogen source. The media contained either no vitamins; the standard vitamins added to JMM (biotin, thiamine and pantothenic acid); a vitamin solution containing all B group vitamins, as described in Egli & Auling (2005) for growth of

Chelatococcus asaccharovorans (Appendix IV); the B group vitamin solution plus casamino acids (0.01% (w/v)); or yeast extract (0.05% (w/v)). Fresh plate culture was resuspended in JMM medium devoid of vitamins, and then added to duplicate 5 ml broths to a final OD600nm of 0.05. The broths were incubated at 28ºC with shaking and observed for growth over six days. The minimal amount of yeast extract required for growth was similarly determined by adding 0.05%, 0.01%, 0.005% or 0.001%

145 Chapter 4

(w/v) yeast extract to the media. Glassware used to grow cultures (McCartney bottles) was soaked in a 10% (v/v) hydrochloric acid solution for at least 24 h and rinsed twice in RODI water prior to use. Lids of McCartney bottles were wrapped with parafilm prior to incubation to prevent contamination.

Strains were examined for growth on L (+) arabinose, D (+) cellobiose, β-D- fructose, α-D-glucose, glycerol, D-mannitol, acetate, succinate (all at 20 mM concentration), benzoate, p-hydroxybenzoate (both 3 mM), glutamate (10 mM), methanol (0.5%, v/v) and ethanol (20 mM) as sole carbon sources in JMM medium devoid of galactose and arabinose and with NH4Cl (10 mM) as a nitrogen source.

Glutamate (10 mM) was also added to JMM devoid of NH4Cl to determine the ability of the strains to utilise glutamate as a nitrogen source. Stock solutions of the carbon substrates (adjusted to pH 7.0 where necessary) were filter sterilised (0.22 µm filter) and added to the autoclaved JMM medium prior to inoculation.

Inocula were prepared by growing fresh plate culture in 5 ml broths of JMM medium containing sodium pyruvate (10 mM) as a carbon source and NH4Cl (10 mM) as a nitrogen source and supplemented with yeast extract (0.1% (w/v)).

Cultures were grown for 50 h to stationary phase, then centrifuged (20 800 x g for 30 s), washed twice with 0.89% (w/v) saline, resuspended in JMM medium devoid of carbon source and added to duplicate 5 ml broths of JMM containing one of the carbon substrates to a final OD600nm of 0.05. Inoculated culture media were incubated for 14 days before a visual assessment was made. Two negative controls were used: an uninoculated control containing JMM medium and various carbon sources, and a control devoid of carbon substrate but containing bacterial inoculant. Growth on the

146 Chapter 4

carbon substrate was assessed as being no growth (OD600nm was the same as for the minus carbon substrate control), poor (0.1 < OD600nm < 0.2), moderate (0.2 < OD600nm

< 0.5) or abundant (OD600nm > 1.0).

4.2.3.10 Biochemical characterisation

API –20E (bioMérieux) test strips were used to determine acid production from sugars and utilisation of various substrates. Fresh plate culture was resuspended in sterile RODI water containing either vitamin solution (Egli & Auling, 2005) or yeast extract (0.005% (w/v)) for the Listia angolensis and Lupinus texensis strains, respectively, and used to inoculate the strips. Strips were incubated at 28ºC for 40 h and then read in accordance with the manufacturer’s instructions.

Oxidase activity was detected by applying fresh plate culture to filter paper impregnated with a solution of 1% (w/v) tetra-methyl p-phenylene diamine HCl and

0.1% (w/v) ascorbic acid. Catalase activity was determined on fresh plate culture using 3% (v/v) hydrogen peroxide solution. Determination of nitrate reduction was performed by inoculating stationary phase TY broth culture into shaking TY broths

-1 supplemented with KNO3 (1 g l ) and growing for 24 h at 28ºC. Nitrate reduction to nitrite was revealed by adding 1 ml of 0.8% (w/v) sulphanilic acid in 5N acetic acid to the culture, followed by the addition of 1 ml of 0.8% (w/v) 8-aminonaphthalene-2- sulphonic acid (Cleve’s acid). The development of a red colour indicated the presence of nitrite. Powdered zinc was added to solutions that remained colourless; a subsequent development of red colour revealed residual nitrate, whereas the further reduction of nitrite was demonstrated by the solution remaining colourless.

Determination of starch hydrolysis was performed on TY agar supplemented with

0.4% (w/v) soluble starch. Bacillus subtilis and Escherichia coli (Table 4.1) served

147 Chapter 4 as positive and negative controls respectively. Oxidative or fermentative catabolism was determined using Hugh & Leifson’s basal medium (Hugh & Leifson, 1953), with L-arabinose, α-D-glucose or pyruvate as a carbon source. Test tubes containing

5 ml of medium were stab-inoculated with fresh plate culture. Anaerobic conditions were created by overlaying the stab culture with sterile paraffin and cultures were examined for growth and colour change in the medium after incubation at 37ºC for

48 h. The Voges-Proskauer test for production of acetoin from glucose was performed using Smith’s medium (Leone & Hedrick, 1954) supplemented with 0.5 g l-1 yeast extract. Test tubes containing 10 ml of medium were inoculated with 100 µl of fresh plate culture resuspended in 0.89% saline to an OD600nm of 1.0. The cultures were grown at 28ºC for 42 h, then aliquots of culture (1 ml) were placed in a test tube, to which was added 15 drops of 5% (w/v) α-naphthol in 100% ethanol followed by 5 drops of 40% (w/v) KOH in RODI water. The samples were vortexed after the addition of each reagent, then allowed to stand and monitored for colour development. A red colour was scored as positive, pink was weakly positive and amber was negative. B. subtilis and E. coli (Table 4.1) served as positive and negative controls respectively.

4.2.4 Host range

The host range of WSM3557T, WSM3693T, Lut5 and Lut6T was determined on taxonomically diverse legume hosts (Table 4.2) in glasshouse trials, using either a closed vial or open pot system according to the methods given in Chapter 2, but with the pots being sterilised by autoclaving, rather than steam treatment. Duplicate pots or vials were used for each treatment, along with uninoculated controls.

Confirmation of nodule occupancy was determined by reisolating the inoculant strain

148 Chapter 4 and verifying its identity with fingerprinting PCR, using ERIC primers (Versalovic et al., 1991) according to the methods given in Chapter 2.

Table 4.2. Legume species used to determine the host range of novel rhizobial species of Microvirga.

Tribe Host Geographical origin

Acacieae Acacia saligna Temperate Australia (Mimosoideae) Crotalaria juncea India and tropical Asia Listia angolensis Tropical Africa Crotalarieae Listia bainesii South Africa Listia heterophylla South Africa Lupinus Genisteae Mediterranean angustifolius Indigofera South Africa Indigofereae frutescens Indigofera patens South Africa Loteae Lotus corniculatus Africa and Asia Macroptilium Pantropical atropurpureum Phaseoleae Phaseolus vulgaris Tropical and warm temperate Americas Vigna unguiculata Africa

4.3 Results

4.3.1 Amplification and sequencing of the 16S rRNA gene

A nearly full-length portion of the 16S rRNA gene was amplified and sequenced for WSM3557T (1457 nt; GenBank Accession No. HM362432) and

WSM3693T (1452 nt; GenBank Accession No. HM362433). BLASTN analysis of these sequences showed that WSM3557T and WSM3693T were members of the

Alphaproteobacteria and, along with WSM3674, WSM3686, Lut5 and Lut6T, were most closely related to species of Microvirga. The pairwise percent sequence identity of a 1396 bp fragment of the 16S rRNA gene for the Listia angolensis and Lupinus texensis strains and the Microvirga species type strains showed that all strains in this clade shared at least 96.1% sequence identity (Table 4.3).

149 Chapter 4

Table 4.3. Pairwise percent identity and number of base pair mismatches (in brackets) for 1396 bp fragment of 16S rRNA gene from Listia angolensis strains (WSM numbers), Lupinus texensis strains (Lut5 and Lut6T), Microvirga flocculans TFBT, Microvirga aerilata 5420S-16T, Microvirga aerophila 5420S-12T, Microvirga guangxiensis 25BTand Microvirga subterranea FaiI4T.

Lut5 Lut6T WSM3557T WSM3674 WSM3686 WSM3693T 5420S-16T 5420S-12T TFBT 25BT FaiI4T 98.3 98.2 98.2 98.0 97.9 96.9 98.0 97.1 97.5 Lut5 100 (24) (25) (25) (29) (30) (51) (29) (44) (35) T 98.3 98.2 98.2 98.0 97.9 96.9 98.0 97.1 97.5 Lut6 100 (24) (25) (25) (29) (30) (51) (29) (44) (35) T 98.3 98.3 99.9 99.9 96.9 97.0 96.1 97.6 97.1 96.8 WSM3557 (24) (24) (1) (1) (44) (42) (56) (34) (46) (45) 98.2 98.2 99.9 96.9 96.9 96.1 97.6 97.1 96.7 WSM3674 100 (25) (25) (1) (45) (43) (57) (35) (45) (46) 98.2 98.2 99.9 96.9 96.9 96.1 97.6 97.1 96.7 WSM3686 100 (25) (25) (1) (45) (43) (57) (35) (45) (46) T 98.0 98.0 96.9 96.9 96.9 98.8 96.4 98.1 96.5 97.1 WSM3693 (29) (29) (44) (45) (45) (18) (51) (28) (52) (42) T 97.9 97.9 97.0 96.9 96.9 98.8 96.7 98.1 97.4 97.4 5420S-16 (30) (30) (42) (43) (43) (18) (53) (26) (40) (37) T 96.7 96.7 96.1 96.1 96.1 96.4 96.7 96.4 97.2 97.9 5420S-12 (51) (51) (56) (57) (57) (51) (53) (56) (41) (30) T 98.0 98.0 97.6 97.6 97.6 98.1 98.1 96.4 96.4 97.0 TFB (29) (29) (34) (35) (35) (28) (26) (56) (55) (42) T 97.1 97.1 97.1 97.1 97.1 96.5 97.4 97.2 96.4 97.6 25B (44) (44) (46) (45) (45) (52) (40) (41) (55) (33) FaiI4T 97.5 97.5 96.8 96.7 96.7 97.1 97.4 97.9 97.0 97.6 (35) (35) (45) (46) (46) (42) (37) (30) (42) (33)

150 Chapter 4

The Listia angolensis and Lupinus texensis strains formed three distinct groups. In the first group, the sequences of WSM3674 and WSM3686 were identical and shared 99.9% sequence identity (1 bp difference) with WSM3557T. They shared

98.2-98.3% sequence identity (24-25 bp difference) with the Lut5 and Lut6T strains that formed the second group. The Lut5 and Lut6T sequences were identical. Both these groups were most closely related to M. flocculans, having 97.6-98.0% sequence identity with strain TFBT (29-35 bp difference). In contrast, the L. angolensis strain

WSM3693T was most closely related to M. aerilata strain 5420S-16T, with 98.8% sequence identity (18 bp difference) and sharing only 96.9% sequence identity (44-

45 bp difference) with the other L. angolensis strains.

The NJ phylogenetic tree obtained from a 1396 bp fragment of the 16S rDNA sequences (Figure 4.2) demonstrated that the Listia angolensis and Lupinus texensis strains, along with the Microvirga species type strains, formed a monophyletic group that was clearly separated from Methylobacterium, Bosea and Chelatococcus lineages and supported by high (100%) bootstrap values. Microvirga subterranea was a basal member of this clade.

4.3.2 Amplification and sequencing of nodA

The 635 bp sequence obtained for WSM3693T, containing most of the nodA gene and part of the 5’ region of nodB, was identical to the sequence obtained for

WSM3557T (detailed in Chapter 2). A Bayesian phylogenetic tree of the Microvirga nodA genes is shown in Figure 2, Appendix VIII (Matt Parker unpublished data).

151 Chapter 4

77 WSM3674 (DQ838528)(DQ838528) 100 WSM3686 (DQ838529) 87 WSM3557T (HM362432) Lut5 (EF191407) 99 100 Lut6T (EF191408) Microvirga flocculans TFBT (AB098515) WSM3693T 100 57 (HM362433) 93 Microvirga aerilata 5420S-16T (GQ421849) Microvirga guangxiensis 25BT (EU727176) 92 Microvirga aerophila 5420S-12T (GQ421848) 83 73 Microvirga subterranea Fail4T (FR733708 ) Methylobacterium nodulans ORS 2060T (AF220763)

90 Methylobacterium organophilum DSM760T (AB175639) 99 Methylobacterium extorquens IAM12631T (AB175632) 99 100 Methylobacterium populi BJ001T (AY251818) Chelatococcus asaccharovorans TE2T (AJ294349)

98 Chelatococcus daeguensis K106T (EF584507) Bosea minatitlanensis DSM13099T (AF273081) T 100 Bosea thiooxidans DSM9653 (AJ250796) 94 34614T (AF288300) Bosea eneae Bradyrhizobium japonicum USDA6T (U69638)

0.01 Figure 4.2. NJ phylogenetic tree based on a comparative analysis of 16S rRNA gene sequences, showing the relationships between novel symbiotic Microvirga strains (indicated in bold) and closely related species. Numbers at the nodes of the tree indicate bootstrap values (expressed as percentages of 1000 replications). GenBank accession numbers are given in parentheses. Bradyrhizobium japonicum USDA6T was used as an outgroup. Scale bar for branch lengths shows 0.01 substitutions per site.

152 Chapter 4

4.3.3 Phenotypic characterisation

4.3.3.1 Morphology

Colonies of five of the six Listia angolensis strains grown on ½ LA were light pink, convex and smooth (Figure 4.3a, b). Pigmentation developed after several days.

The sixth strain, WSM3693T, was cream coloured. Cultures grown with glucose and mannitol as carbon sources typically produced more exopolysaccharide than those grown with succinate, and developed characteristic banding patterns (Figure 4.3 c).

Colonies of the Lupinus texensis strains were a pale orange on ½ LA (Figure 4.4 d).

All strains grew to 0.5 – 1.5 mm in three days at 28°C and grew well on nutrient agar. All cultures stained Gram-negative. No spores were observed, nor did cultures grow after heat treatment at 70°C. Cells of the Listia angolensis strains, observed by the hanging drop method, were motile. The Lupinus texensis strains Lut5 and Lut6T were grown on a range of media, but motility was never observed. Electron micrographs of the L. angolensis strains and Lut5 and Lut6T showed rod shaped cells

(0.4 - 0.5 µm x 1.0-2.2 µm), surrounded by a prominent capsule (Figures 4.4 a and b and 4.5 a, b and c). The L. angolensis strains all had at least one polar flagellum.

Flagella were not observed on Lut5 or Lut6T.

153 Chapter 4

1 mm a b

c d

Figure 4.3. Colony morphology and pigmentation of Listia angolensis and Lupinus texensis rhizobial strains: a) Three day old colony of WSM3557T on ½ LA b) Seven day old culture of WSM3557T on ½ LA, with succinate replacing glucose and mannitol as the carbon source c) Seven day old culture of WSM3557T on ½ LA, showing characteristic banding patterns d) Aged culture of Lut6T on ½ LA

500 nm 500 nm a b

Figure 4.4. Transmission electron micrograph of a) WSM3693T and b) Lut6T.

154 Chapter 4

5 µm 1 µm 10 µm a b c

Figure 4.5. Scanning electron micrograph of a) WSM3557T; b) Lut6T; c) Lut6T, showing a clearly visible capsule, due to shrinkage of the cytoplasm.

4.3.3.2 Characterisation of pigments

The pigment extracted from the Listia angolensis isolate WSM3557T was light pink in colour. The main absorption peak at 495nm (absorbance, 0.071A) (Figure 4.6 a) appeared similar to that obtained for the pigmented Methylobacterium strain

WSM2598 (absorbance, 0.834A) (Figure 4.6 b). The Lut6T pigment extract was bright yellow and had an absorption peak at 458 nm (absorbance 0.474A) (Figure 4.6 c). No absorption peaks were seen at wavelengths between 700 – 800 nm for

WSM3557T and Lut6T, whereas the data for WSM2598 indicated a very slight absorption peak at around 770 nm.

Figure 4.6. Absorption spectra of 7:2 acetone:methanol extracts from rhizobial strains. a) WSM3557T; b) WSM2598; c) Lut6T.

155 Chapter 4

4.3.3.3 Growth characteristics

The pink-pigmented L. angolensis strains grew at a temperature range of

15°C to 44/45°C, while WSM3693T had a growth range of 15°C to 38°C. For Lut5 and Lut6T, the range was 10°C to 43°C. The optimal growth temperature for Lut6T was 39°C and the mean generation time (MGT) at this temperature was 1.8 h. The optimal temperature and MGT for WSM3557T were 41°C and 1.6 h and for

WSM3693T these were 35°C and 1.7 h (Figure 4.7 a-f).

All strains grew optimally at pH 7.0 – 8.5 and over a range of pH 5.5 – 9.5 (or pH 6.0 – 9.5 for WSM3693T). They were not salt tolerant; best growth was observed at 0 – 0.5% (w/v) NaCl and isolates did not grow at > 2.0% (w/v) NaCl (Appendix

V). Colonies growing at 1.0 and 1.5% (w/v) NaCl were noticeably smaller.

WSM3557T, WSM3693T, Lut5 and Lut6T were strictly aerobic: no growth was observed on Hugh & Leifson’s medium incubated for ten days in an anaerobic jar.

4.3.3.4 Antibiotic sensitivity

All the L. angolensis strains showed high resistance to gentamycin.

WSM3557T, WSM3674, WSM3675 and WSM3686 were partially resistant to kanamycin and spectinomycin. WSM3674, WSM3675 and WSM3686 were additionally partially resistant to chloramphenicol and WSM3674 and WSM3675 to ampicillin. Lut5 and Lut6T were partially resistant to ampicillin, chloramphenicol, gentamycin and streptomycin and were sensitive to kanamycin, nalidixic acid, rifampicin, spectinomycin and tetracycline.

156 Chapter 4

a

b

c

Figure 4.7. Log/linear growth curves at temperature optima to determine mean generation time: a) Lut6T at 39°C; b) WSM3557T at 41°C; c) WSM3693T at 35°C.

157 Chapter 4

4.3.3.5 Substrate utilisation

4.3.3.5.1 Utilisation of sole carbon sources in Biolog GN2 plates

The utilisation of carbon substrates by Listia angolensis and Lupinus texensis strains inoculated into Biolog GN2 microplates is shown in Table 4.4. The substrates have been categorised and divided into 11 groups, according to Garland & Mills

(1991). Readings from the 96-h samples were used, as colour development was poor until 40 h incubation. Dextrin (well A3) was scored as a negative, as the initial absorbance readings were slightly above the threshold value but did not increase and no colour development was apparent.

In general, all strains had similar substrate utilisation patterns. A total of 35 of the 95 carbon sources gave positive results. The carbon sources utilised spanned most of the 11 designated categories, with none of the polymer, alcohol, phosphorylated chemical or amine substrates being used. The range of substrates utilised within each category was, however, quite narrow. Only 9 of 28 carbohydrates and 7 of 24 carboxylic acids were metabolised. The Lupinus texensis strains utilised 6 of the possible 20 amino acid sources. The pink-pigmented Listia angolensis strains were less specific in their utilisation, with 12 amino acids giving a positive result for at least one strain. WSM3693 utilised five amino acids.

158 Chapter 4

Table 4.4. Ability of Listia angolensis strains (WSM numbers) and Lupinus texensis strains (Lut5, Lut6T) to metabolise sole carbon sources in Biolog GN2 microplates. Carbon sources are categorised according to Garland & Mills (1991). Data is from 96 hr duplicate reads. Scored mean values (+ve = green; Weak = yellow; -ve = red)

Strain Carbon source WSM WSM WSM WSM Lut5 Lut6T 3557T 3674 3686 3693T Carbohydrates N-Acetyl-D-galactosamine N-Acetyl-D-glucosamine Adonitol L-Arabinose D-Arabitol D-Cellobiose i-Erythritol D-Fructose L-Fucose D-Galactose Gentiobiose α-D-Glucose m-Inositol α-Lactose Lactulose Maltose D-Mannitol D-Mannose D-Melibiose β-Methylglucoside Psicose D-Raffinose L-Rhamnose D-Sorbitol Sucrose D-Trehalose Turanose Xylitol Esters Pyruvic acid methyl ester Succinic acid methyl ester Polymers α-Cyclodextrin Dextrin Glycogen Tween 40 Tween 80

159 Chapter 4

Table 4.4. (cont)

Strain Carbon source WSM WSM WSM WSM Lut5 Lut6T 3557T 3674 3686 3693T Carboxylic acids Acetic acid cis-Aconitic acid Citric acid Formic acid D-Galactonic acid lactone D-Galacturonic acid D-Gluconic acid D-Glucosaminic acid D-Glucuronic acid α-Hydroxybutyric acid β-Hydroxybutyric acid γ-Hydroxybutyric acid p-Hydroxyphenylacetic acid Itaconic acid α-Ketobutyric acid α-Ketoglutaric acid α-Ketovaleric acid D,L-Lactic acid Malonic acid Propionic acid Quinic acid D-Saccharic acid Sebacic acid Succinic acid Alcohols 2,3-Butanediol Glycerol Amides Succinamic acid Glucuronamide Alaninamide Phosphorylated chemicals D,L-α-Glycerol phosphate Glucose-1-phosphate Glucose-6-phosphate

160 Chapter 4

Table 4.4. (cont)

Strain Carbon source WSM WSM WSM WSM Lut5 Lut6T 3557T 3674 3686 3693T Amino acids D-Alanine L-Alanine L-Alanyl-glycine L-Asparagine L-Aspartic acid L-Glutamic acid Glycyl-L-aspartic acid Glycyl-L-glutamic acid L-Histidine Hydroxy-L-proline L-Leucine L-Ornithine L-Phenylalanine L-Proline L-Pyroglutamic acid D-Serine L-Serine L-Threonine D,L-Carnitine γ-Aminobutyric acid Aromatic chemicals Urocanic acid Inosine Uridine Thymidine Brominated chemicals Bromosuccinic acid Amines Phenylethylamine Putrescine 2-Aminoethanol

Substrates that gave a positive reading for at least one of the pink-pigmented

Listia angolensis strains included adonitol, L-arabinose, D-cellobiose, D-fructose, α-

D-glucose, inositol, D-mannose, D-melibiose, L-rhamnose, pyruvic acid methyl ester, succinic acid methyl ester, acetic acid, D-galacturonic acid, β-hydroxybutyric acid, γ-hydroxybutyric acid, α-ketoglutaric acid, D,L-lactic acid, succinic acid, succinamic acid, urocanic acid and bromosuccinic acid. The Lut strains utilised all these substrates with the exception of D-mannose, succinic acid methyl ester and urocanic acid, and were additionally able to metabolise saccharic acid and xylitol.

WSM3693 utilised p-hydroxyphenylacetic acid but was unable to metabolise D-

161 Chapter 4 mannose, succinic acid methyl ester, β-hydroxybutyric acid, succinamic acid methyl ester, urocanic acid and bromosuccinic acid. L-alanine, L-alanyl-glycine, L- asparagine, L-aspartic acid, L-glutamic acid, glycyl-L-aspartic acid, glycyl-L- glutamic acid, L-histidine, hydroxy-L-proline, L-ornithine, L-phenylalanine, L- proline, L-serine and γ-aminobutyric acid were used as sole carbon substrates by the pink-pigmented Listia angolensis strains. Lut5 and Lut6T utilised L-alanine, L- glutamic acid, L-histidine, hydroxy-L-proline, L-ornithine and L-proline. WSM3693 utilised L-aspartic acid, L-glutamic acid, L-histidine, hydroxy-L-proline and L- ornithine.

4.3.3.5.2 Utilisation of substrates in Biolog Phenotype Microarray plates

Biolog PM plates 1 to 4 were used to test the metabolic abilities of

WSM3557T and Lut6T on 379 substrates, including 190 carbon sources, 95 nitrogen sources, 59 phosphorus sources and 35 sulfur sources. The complete list of substrates is shown in Table 1 (Appendix VI). The absorbance values of substrates giving a positive phenotype are shown in Table 4.5. It is possible that some of the carbon source readings are false positives. Dye G reacts with some pentoses and other carbon sources, including D and L-arabinose, D-glucosamine, dihydroxyacetone, L- lyxose and D-xylose (Biolog PM Services, pers. comm.). Absorbance readings for the C5 carbohydrates were generally higher than for other substrates. Conversely, some of the carbon source readings may be false negatives. Succinic acid and L- glutamic acid, for example, gave negative results, but were utilised (in the form of succinate and glutamate) as sole carbon substrates for WSM3557T grown in minimal media. D-mannitol and acetate supported the growth of all tested strains in minimal media, but also gave negative results on the PM plates.

162 Chapter 4

Table 4.5. Biolog PM microplate readings for substrates that gave positive scores for inoculated WSM3557T or Lut6T. Brackets ( ) around the reading indicate a negative score.

Average Reading Plate Well Substrate Substrate Type WSM3557T Lut6T PM2 B05 D-Arabinose C source; C5 carbohydrate 120.5 110.5 PM1 A02 L-Arabinose C source; C5 carbohydrate 124.0 115.5 PM1 H06 L-Lyxose C source; C5 carbohydrate 145.0 117.5 PM1 C04 D-Ribose C source; C5 carbohydrate 150.0 108.0 PM2 B09 2-Deoxy-D-Ribose C source; C5 carbohydrate 36.0 43.0 PM1 B08 D-Xylose C source; C5 carbohydrate 135.5 107.5 PM1 E09 Adonitol (=Ribitol) C source; C5 carbohydrate (11.5) 36 PM1 C07 D-Fructose C source; C6 carbohydrate (12.0) 37.0 D-Fructose-6- PM1 E04 Phosphate C source; C6 carbohydrate 20.5 13.5 PM1 C09 α-D-Glucose C source; C6 carbohydrate 24.0 35.5 PM2 E05 D-Glucosamine C source; C6 carbohydrate 150.0 170.0 C source; C6 carbohydrate PM1 F03 m-Inositol (polyol of cyclohexane) (13.0) 31.5 PM1 A11 D-Mannose C source; C6 carbohydrate 23.5 39.5 PM1 H05 D-Psicose C source; C6 carbohydrate 34.5 64.0 PM1 C06 L-Rhamnose C source; C6 carbohydrate 38.5 65.0 PM2 D06 D-Tagatose C source; C6 carbohydrate 79.0 65.5 PM1 F11 D-Cellobiose C source; C12 carbohydrate (13.5) 26.0 PM2 C12 Palatinose C source; C12 carbohydrate 42.0 38.5 C source; C12 carbohydrate PM2 E05 Arbutin (Glucosylated hydroquinone) 26.5 38.0 PM1 B09 L-Lactic acid C source, C3 carboxylic acid (16.0) 35.0 PM2 F02 D-Malonic acid C source, C3 dicarboxylic acid (0.0) 44.0 PM1 H08 Pyruvic acid C source, C3 carboxylic acid 67.0 52.5 PM1 F06 Bromosuccinic acid C source, C4 dicarboxylic acid (0.5) 74.5 PM1 F05 Fumaric acid C source, C4 dicarboxylic acid (0.0) 79.5 PM1 C03 D,L-Malic acid C source, C4 dicarboxylic acid (3.0) 87.0 PM1 G11 D-Malic acid C source, C4 dicarboxylic acid (5.0) 58.0 PM1 G12 L-Malic acid C source, C4 dicarboxylic acid (17.0) 78.5 PM1 A05 Succinic acid C source, C4 dicarboxylic acid (19.0) 94.5 PM2 F10 Succinamic acid C source, C4 carboxylic acid (10.5) 30.5 PM1 D06 α-Ketoglutaric acid C source, C5 carboxylic acid (20.5) 28.0 PM1 H10 D-Galacturonic acid C source, C6 carboxylic acid 55.0 39.5 5-Keto-D-Gluconic PM2 E12 acid C source, C6 carboxylic acid 78.5 90.0 PM2 F05 D-Oxalomalic acid C source, C6 tricarboxylic acid 29.0 38.0 PM1 A04 D-Saccharic acid C source, C6 dicarboxylic acid (0.0) 44.5 Mono- C source, methyl ester of PM1 G09 Methylsuccinate carboxylic acid 78.5 12.5 PM2 H09 Dihydroxyacetone C source, C3 alcohol 123.5 133.5 PM1 H07 Glucuronamide C source, C6 amide 45.0 40.5 PM1 A07 L-Aspartic acid C.source, amino acid (5.0) 33.0 PM1 B12 L-Glutamic acid C.source, amino acid (23.0) 52.5 PM1 G03 L-Serine C.source, amino acid 45.0 0.0 PM1 C05 Tween 20 C source, fatty acid (18.5) 27.0 PM2 A12 Pectin C source, polymer 24.0 37.0

163 Chapter 4

Table 4.5. (cont.)

Average Reading Plate Well Chemical Mode of Action WSM3557T Lut6T PM3 C03 D-Alanine N.source, amino acid (0.0) 27.5 PM3 A07 L-Alanine N.source, amino acid (1.0) 37.0 PM3 A08 L-Arginine N.source, amino acid (0.0) 55.5 PM3 A09 L-Asparagine N.source, amino acid (19.5) 96.5 PM3 C05 D-Aspartic acid N.source, amino acid (0.0) 38.0 PM3 A10 L-Aspartic acid N.source, amino acid (23.5) 96.0 PM3 C10 L-Citrulline N.source, amino acid (0.0) 25.0 PM3 A11 L-Cysteine N.source, amino acid 46.0 58.0 PM3 A12 L-Glutamic acid N.source, amino acid 48.5 111.0 PM3 B01 L-Glutamine N.source, amino acid 22.0 103.0 PM3 B03 L-Histidine N.source, amino acid (11.0) 69.0 PM3 B06 L-Lysine N.source, amino acid (0.0) 38.5 PM3 C12 L-Ornithine N.source, amino acid (4.0) 56.5 PM3 B09 L-Proline N.source, amino acid (0.5) 91.0 PM3 B10 L-Serine N.source, amino acid (7.0) 34.5 PM3 B12 L-Tryptophan N.source, amino acid 32.5 51.0 N-Acetyl-D- PM3 E11 Glucosamine N.source, other (1.0) 46.5 PM3 G05 Allantoin N.source, other (22.5) 31.5 PM3 G04 Alloxan N.source, other 46.5 68.5 γ-Amino-N-Butyric PM3 G08 acid N.source, other (2.5) 72.0 D,L-α-Amino- PM3 G10 Caprylic acid N.source, other 28.5 24.5 d-Amino-N-Valeric PM3 G11 acid N.source, other (1.0) 48.0 PM3 E09 D-Galactosamine N.source, other 101.0 124.0 PM3 E08 D-Glucosamine N.source, other 101.0 133.5 PM3 E06 Glucuronamide N.source, other (4.5) 26.0 PM3 E10 D-Mannosamine N.source, other 136.0 157.0 PM3 G01 Xanthine N.source, other 19.5 59.0 PM3 H01 Ala-Asp N.source, peptide (18.5) 53.5 PM3 H02 Ala-Gln N.source, peptide 26.5 71.5 PM3 H03 Ala-Glu N.source, peptide (20.5) 75.5 PM3 H04 Ala-Gly N.source, peptide (2.0) 45.5 PM3 H05 Ala-His N.source, peptide (0.5) 52.0 PM3 H08 Gly-Asn N.source, peptide (8.0) 63.0 PM3 H09 Gly-Gln N.source, peptide (20.0) 78.0 PM3 H10 Gly-Glu N.source, peptide (18.5) 63.0 PM4 B01 Thiophosphate P source, inorganic 25.5 26.5 Cysteamine-S- PM4 D03 Phosphate P source, organic, amine (1.0) 22.0 O-Phospho-D- PM4 D05 Serine P source, organic, amino acid (0.0) 23.5 O-Phospho-L- PM4 D06 Serine P source, organic, amino acid (0.0) 32.5 O-Phospho-L- PM4 D07 Threonine P source, organic, amino acid (0.0) 25.0 O-Phospho-L- PM4 E02 Tyrosine P source, organic, amino acid (0.0) 22.0 PM4 D04 Phospho-L-Arginine P source, organic, amino acid (0.0) 39.5 D,L-a-Glycerol P source, organic, C3 PM4 B03 Phosphate carbohydrate (0.0) 42.5

164 Chapter 4

Table 4.5. (cont.)

Average Reading Plate Well Chemical Mode of Action WSM3557T Lut6T D-2-Phospho- P source, organic, C3 PM4 B06 Glyceric Acid carbohydrate (0.0) 23.0 2-Deoxy-D-Glucose P source, organic, C6 PM4 C05 6-Phosphate carbohydrate (0.0) 24.0 D-Glucosamine-6- P source, organic, C6 PM4 C06 Phosphate carbohydrate (6.0) 39.0 D-Glucose-1- P source, organic, C6 PM4 C03 Phosphate carbohydrate (0.0) 24.5 6-Phospho- P source, organic, C6 PM4 C07 Gluconic Acid carbohydrate (0.0) 25.5 Inositol PM4 E11 Hexaphosphate P source, organic, C6 polyol 24.0 30.0 Adenosine 3`- PM4 A09 Monophosphate P source, organic, purine (0.0) 20.5 Adenosine 5`- PM4 A10 Monophosphate P source, organic, purine (0.0) 29.0 Adenosine 2`,3`- Cyclic PM4 A11 Monophosphate P source, organic, purine (0.0) 26.0 Guanosine 2`- PM4 B08 Monophosphate P source, organic, purine (0.0) 27.0 Guanosine 3`- PM4 B09 Monophosphate P source, organic, purine (0.0) 26.0 Guanosine 5`- PM4 B10 Monophosphate P source, organic, purine (0.0) 27.0 Guanosine 2`,3`- Cyclic PM4 B11 Monophosphate P source, organic, purine (0.0) 21.0 Cytidine 2`- PM4 C08 Monophosphate P source, organic, pyrimidine (0.0) 31.0 Cytidine 3`- PM4 C09 Monophosphate P source, organic, pyrimidine (0.0) 37.0 Cytidine 5`- PM4 C10 Monophosphate P source, organic, pyrimidine (0.0) 31.0 Cytidine 2`,3`- Cyclic PM4 C11 Monophosphate P source, organic, pyrimidine (0.0) 34.5 Thymidine 3`- PM4 E09 Monophosphate P source, organic, pyrimidine (0.0) 38.0 Thymidine 5`- PM4 E10 Monophosphate P source, organic, pyrimidine (0.0) 37.0 Thymidine 3`,5`- Cyclic PM4 E12 Monophosphate P source, organic, pyrimidine (0.0) 25.5 Uridine 2`- PM4 D08 Monophosphate P source, organic, pyrimidine (0.0) 29.0 Uridine 3`- PM4 D09 Monophosphate P source, organic, pyrimidine (0.0) 41.5 Uridine 5`- PM4 D10 Monophosphate P source, organic, pyrimidine (0.0) 36.0 Uridine 2`,3`-Cyclic PM4 D11 Monophosphate P source, organic, pyrimidine (0.0) 38.5 PM4 F08 D-Cysteine S-source, amino acid (0.0) 32.0 PM4 H04 D,L-Lipoamide S-source, fatty acid (17.5) 25.5 165 Chapter 4

The results for the PM3 and PM4 plates (nitrogen, phosphorus and sulfur sources) may also not be a true indication of WSM3557T and Lut6T metabolic abilities. The lack of a positive reaction for the substrates ammonia (PM3 plate, well

A02) and phosphate (PM4 plate, well A02) for either strain is illustrative, as these substrates have been used to grow the strains in minimal media, and ammonia is supplied as the nitrogen source in PM plates 1, 2 and 4 (Biolog PM Services, pers. comm.). Colour production in plates PM1 and PM2 only requires carbon oxidation, whereas plates PM3 and PM4 behave differently due to the defined media and may not work for all organisms because of this (Biolog PM Services, pers. comm.).

WSM3557T utilised 24 of the carbon sources, 10 of the nitrogen sources and two phosphorus sources. None of the sulfur sources gave a positive reading. Lut6T utilised 34 of the carbon sources, 35 of the nitrogen sources, 32 phosphorus sources and two sulfur sources. A variety of substrates were used as carbon sources. For

WSM3557T, these included carbohydrates (D- and L-arabinose, α-D-glucose, L- lyxose, D-mannose, palatinose, D-psicose, L-rhamnose, D-ribose, D-tagatose, D- xylose, 2-deoxy-D-ribose, D-fructose-6-phosphate, D-glucosamine and arbutin); carboxylic acids (D-galacturonic acid, 5-keto-D-gluconic acid, mono- methylsuccinate, oxalomalic acid and pyruvic acid); alcohols (dihydroxyacetone); amides (glucuronamide); amino acids (L- serine) and polymers (pectin). Lut6T metabolised all the above except for D-fructose-6-phosphate, mono-methylsuccinate and L- serine and was additionally able to utilise D-cellobiose, D-fructose, adonitol and m-inositol (carbohydrates); bromosuccinic acid, fumaric acid, α-ketoglutaric acid, L-lactic acid, D- and L-malic acid, D-malonic acid, D-saccharic acid, succinic

166 Chapter 4 acid and succinamic acid (carboxylic acids); L-glutamic acid and L-aspartic acid

(amino acids) and Tween 20 (fatty acid).

WSM3557T was able to metabolise four of 33 amino acids (L-cysteine, L- glutamic acid, L-glutamine and L-tryptophan), one peptide (Ala-Gln) and five other substrates (alloxan, D,L-α-amino-caprylic acid, D-galactosamine, D-glucosamine and D-mannosamine) as nitrogen sources. Thiophosphate and inositol hexaphosphate

(phytic acid), were utilised as phosphorus sources. Lut6T utilised the same substrates as WSM3557T and additionally twelve other amino acids, seven other peptides, six other nitrogen sources, and thirty other phosphorus sources. D-cysteine and D,L- lipoamide were utilised as sulfur sources.

Most of the carbon substrates in the GN2 plates form a subset of those available in the PM carbon source plates (PM 1 and 2). The results obtained for

WSM3557T and Lut6T on the PM plates are not directly comparable with GN2 plate data because of differences in media formulation and redox dyes, and different inoculation concentrations and incubation times (40% transmittance and 56 h for the

PM plates versus 85% transmittance and 96 h for the GN2 plates). Substrate utilisation in the PM plates was consistent with some GN2 substrate results, but differed in several instances, notably with amino acid substrates. A comparison of results is included in Table 1, Appendix VI. Substrates inoculated with WSM3557T that tested positive on GN2 and negative on PM plates included D-cellobiose, D- fructose and m-inositol (carbohydrates); bromosuccinic acid, L-lactic acid, methyl pyruvate, succinamic acid and succinic acid (carboxylic acids) and L-asparagine, L- aspartic acid, L-glutamic acid, L-histidine, hydroxy L-proline, ornithine and L-

167 Chapter 4 proline (amino acids). Carbon substrates that tested negative in WSM3557T- inoculated GN2 plates but positive in PM plates included psicose, glucuronamide and L-serine. For Lut6T, substrates that were positive on GN2 and negative on PM plates were methyl pyruvate, L-alanine and L-proline. Substrates that were negative on GN2 and positive on PM plates included D-mannose, D-psicose, α-ketoglutaric acid, glucuronamide and L-aspartic acid.

4.3.3.5.3 Growth in sole carbon substrate broths

The Biolog system measures the cellular respiration of an organism, rather than its ability to actually grow on a given substrate. The strains WSM3557T, WSM3693T,

Lut5 and Lut6T were therefore grown in minimal media broths to confirm that growth was possible on selected carbon substrates that gave positive results for the

Biolog GN2 plates. Several other substrates not present in the GN2 plates were also tested. As Microvirga subterranea had been shown to have an obligate requirement for yeast extract in minimal media (Kanso & Patel, 2003), all Listia angolensis and

Lupinus texensis strains were first assessed for their supplementary requirements for growth in minimal medium.

Yeast extract was an absolute requirement for growth of Lut5 and Lut6T in minimal media. The Listia angolensis strains required either yeast extract or the vitamin mix described by Egli & Auling (2005) for growth of Chelatococcus asaccharovorans in minimal media. The minimal amount of yeast extract required was 0.005% (w/v); this amount supported growth in media with carbon substrates added, but not growth in media devoid of carbon substrate. In minimal broth media

WSM3557T grew on L-arabinose, D-cellobiose, D-fructose, α-D-glucose, glycerol,

D-mannitol, succinate, glutamate and acetate. WSM3693T grew on all the above and 168 Chapter 4 additionally was able to grow on p-hydroxybenzoate. Neither strain could grow on benzoate, ethanol or methanol. Lut5 and Lut6T grew on L-arabinose, D-cellobiose,

D-fructose, α-D-glucose, D-mannitol, succinate, glutamate, p-hydroxybenzoate and ethanol, but were unable to grow on benzoate, glycerol or methanol (Table 4.6).

Table 4.6. Growth of WSM3557T, WSM3693T, Lut5 and Lut6T on sole carbon substrates in minimal media broths, measured 21 days after inoculation. Strain Type of Carbon Substrate WSM WSM substrate Lut5 Lut6T 3557T 3693T

L + Arabinose C5 carbohydrate xxx xxx xx xx (20 mM) D + Cellobiose C12 carbohydrate xxx xxx x x (20 mM) β D Fructose C6 carbohydrate xxx xxx x x (20 mM) D Glucose C6 carbohydrate xxx xxx xx xx (20 mM) D Mannitol C6 carbohydrate xxx xxx xx xx (20 mM) (polyol) Acetate C2 carboxylate xxx xxx xx xx (20 mM) Succinate C4 dicarboxylate xxx xxx xxx x (20 mM) Benzoate C7 aromatic 0 0 0 0 (3 mM) carboxylate p-hydroxybenzoate C7 aromatic 0 xx x x (3 mM) carboxylate Glutamate –NH Cl 4 Amino acid xxx xxx x xxx (10 mM) Glutamate +NH Cl 4 Amino acid xxx xxx xxx xxx (10 mM) Methanol C1 alcohol 0 0 0 0 (0.5% (v/v)) Ethanol C2 alcohol 0 0 xxx xx (20 mM) Glycerol C3 alcohol xxx xxx 0 0 (20 mM)

4.3.3.6 Biochemical characterisation

On API 20E strips, all strains were positive for urease and weakly positive for acetoin production from pyruvate. All except WSM3693T were weakly positive for tryptophan deaminase. β-Galactosidase, arginine dihydrolase, lysine decarboxylase,

169 Chapter 4 ornithine decarboxylase, indole and hydrogen sulphide were not produced by any strain. Citrate was not utilised and gelatin was not hydrolysed. Acid was produced from arabinose but not from glucose, mannitol, inositol, sorbitol, sucrose or amygdalin. Weak acid production was observed for WSM3673 and WSM3674 on rhamnose and for WSM3557T on melibiose. All strains were catalase positive and oxidase negative. Nitrate was reduced to nitrite by WSM3557T, WSM3674,

WSM3686 and WSM3693T, but not by WSM3673, Lut5 or Lut6T. Starch was not hydrolysed. Stab cultures of WSM3557T, WSM3693T, Lut5 and Lut6T in Hugh &

Leifson’s medium with arabinose, glucose or pyruvate as a carbon source grew in aerobic conditions and in test tubes overlaid with paraffin. Cultures grown in this medium acidified arabinose, but not glucose or pyruvate. Weak acetoin production was observed for WSM3557T, but not for WSM3693T, Lut5 or Lut6T, in Smith’s medium (Leone & Hedrick, 1954) containing glucose as a carbon source (Figure

4.8).

1 2 3 4 5 6

Figure 4.8. Production of acetoin from growth on glucose in Smith’s medium by 1) WSM3557T (weakly positive); 2) WSM3693T, 3) Lut5, 4) Lut6T, 5) Escherichia coli (all negative); 6) Bacillus subtillus (positive).

170 Chapter 4

4.3.4 Host range

The symbiotic ability of WSM3557T, WSM3693T, Lut5 and Lut6T was examined on a range of legumes from different phylogenetic clades (Table 4.7).

WSM3557T and WSM3693T only formed effective nodules on their original Listia angolensis host, but were able to induce occasional ineffective nodulation on several other legumes. WSM3693T was the more infective strain, forming nodules on Acacia saligna, Indigofera frutescens, Phaseolus vulgaris and Vigna unguiculata. Bacteria were reisolated from nodules of all these hosts, thus confirming nodule occupancy.

WSM3557T induced nodulation only on P. vulgaris and the nodules appeared devoid of bacteria, as the inoculant could not be reisolated from harvested nodules. Lut5 and

Lut6T did not induce nodulation on L. angolensis, Listia bainesii or Listia heterophylla. On the other legume hosts, they were able to nodulate only A. saligna and P. vulgaris, albeit ineffectively. Inoculant was reisolated from nodules only in the case of Lut5 on A. saligna. PCR amplification using ERIC primers confirmed the identity of the reisolated strains. Gel images of the fingerprinting patterns obtained from electrophoresis of the amplification products are shown in Appendix VII.

171 Chapter 4

Table 4.7. Nodulation (average number of nodules per plant out of eight* inoculated plants) and effectiveness of the Listia angolensis strains WSM3557T and WSM3693T and Lupinus texensis strains Lut5 and Lut6T on legume hosts harvested six weeks post inoculation.

Inoculant strain Tribe Host WSM3557T WSM3693T Lut5 Lut6T N+/- F- N+/- F- N+/- F- Acacieae Acacia (0.125) N- F- (0.875) (0.125) (Mimosoideae) saligna No Reisolated Reisolated reisolates Crotalaria N- F- N- F- N- F- N- F- juncea N+ F+ N+ F+ Listia (3.0) (3.12) N- F- N- F- Crotalarieae angolensis Reisolated Reisolated Listia bainesii N- F- N- F- N- F- N- F-

Listia N- F- N- F- N- F- N- F- heterophylla Lupinus Genisteae N- F- N- F- N- F- N- F- angustifolius N+/- F- Indigofera N- F- (0.8) ND ND frutescens Indigofereae Reisolated Indigofera N- F- N- F- ND ND patens Lotus Loteae N- F- N- F- N- F- N- F- corniculatus Macroptilium N- F- N- F- ND ND atropurpureum N+/- F- N+/- F- N+/- F- N+/- F- Phaseolus (4.5) (7.5) (6.75) (15.0) Phaseoleae vulgaris No No No Reisolated reisolates reisolates reisolates N+/- F- Vigna N- F- (3.0) N- F- N- F- unguiculata Reisolated *I. frutescens: four plants were inoculated with WSM3557T; five plants were inoculated with WSM3693T. I. patens: four plants were inoculated with WSM3557T; two plants were inoculated with WSM3693T.

172 Chapter 4

4.4 Discussion

4.4.1 Genotypic characterisation of novel Microvirga species

The 16S rRNA gene sequences of the Listia angolensis and Lupinus texensis strains show them to be most closely related to the type strains of Microvirga species. Based on the 95 % 16S rRNA gene sequence similarity that has been proposed as a ‘practicable border zone for genus definition’ (Ludwig et al., 1998), these Listia angolensis and Lupinus texensis rhizobia belong within the genus

Microvirga. It is notable that the lineages of the strains are not based on their geographical separation or on their symbiotic hosts, as WSM3557T, WSM3674 and

WSM3686 are more closely related to Lut5 and Lut6T than they are to WSM3693T.

Strains sharing less than 97% 16S rRNA gene sequence similarity are not considered to be members of the same species (Tindall et al., 2010), although

Stackebrandt & Ebers (2006) have suggested that this be changed to 98.7-99.0% shared sequence similarity. The pink-pigmented Listia angolensis strains

WSM3557T, WSM3674 and WSM3686 are members of the same species, on the basis of 99.9% sequence identity of their 16S rRNA gene sequences. Lut5 and Lut6T share 100% sequence identity and form a second species. WSM3693T is clearly separated from both the Lupinus texensis and the other Listia angolensis strains. As its closest relative is M. aerilata strain 5420S-16T, with 98.8% sequence identity, it merits consideration as a separate species.

As the 16S rRNA gene should not be used as a sole determinant of bacterial phylogenetic placement (Chapter 1, Section 1.5.2) portions of the housekeeping loci dnaK, gyrB, recA and rpoB were sequenced in five symbiotic Microvirga strains and 173 Chapter 4 in two non-symbiotic Microvirga species (M. flocculans TFBT and M. subterranea

DSM 14364T (= FaiI4T)) to further analyse the relationships in this group of bacteria.

A combined analysis of concatenated sequences was performed, using a phylogenetic tree inferred by MrBayes (Ronquist & Huelsenbeck, 2003), with nucleotide sites partitioned by codon position and a HKY substitution model. The program was run for a 250, 000 generation burn-in period and then results were sampled every 250 generations for an additional 250 000 generations. The topology of the phylogenetic tree obtained from this analysis supports that of the 16S rRNA gene tree and confirms that the Listia angolensis and Lupinus texensis strains belong in the genus

Microvirga (Matt Parker unpublished data, Figure 1, Appendix VIII).

The phylogeny of the rhizobial Microvirga symbiotic genes was also examined. A

Bayesian phylogenetic analysis of the nodA gene indicated that Microvirga nodA sequences were derived from two different sources (Matt Parker unpublished data,

Figure 2, Appendix VIII). The WSM3557T and WSM3693T nodA genes clustered in a strongly supported clade with reference strains in the genera Bradyrhizobium,

Burkholderia and Methylobacterium. The host plant species of the Bradyrhizobium strains (ORS 938, USDA 76T, USDA 110 and Lcamp8) are Rhynchosia minima,

Glycine max, Glycine max and Lupinus campestris, respectively (Stepkowski et al.,

2007). The Burkholderia strains (B. tuberum STM678T and Burkholderia sp.

WSM3930) are effective and specific nodulators of Cyclopia spp. (tribe Podalyrieae) and Rhynchosia ferulifolia (tribe Phaseoleae), respectively, which are legumes native to the South African fynbos vegetation (Elliott et al., 2007; Garau et al., 2009).

Methylobacterium nodulans ORS 2060T specifically nodulates Senegalese

174 Chapter 4

Crotalaria spp. (Sy et al., 2001), while Methylobacterium sp. 4-46 nodulates the highly specific Listia bainesii (Fleischman & Kramer, 1998) (see also Chapter 2).

The Lut6T nodA sequence was placed in an equally strongly supported clade with reference strains in the genera Rhizobium, Mesorhizobium and Ensifer (formerly

Sinorhizobium) isolated from diverse legume hosts (Acacia senegal, Acacia tortilis,

Acaciella angustissima, Phaseolus vulgaris, Prosopis chilensis) growing in Africa or the Americas (Matt Parker unpublished data, Figure 2, Appendix VIII). These results suggest that the rhizobial Microvirga strains from Listia angolensis and those from

Lupinus texensis acquired their nodA genes in separate horizontal gene transfer events and from different donor lineages.

In contrast to the nodA phylogeny, the Bayesian analysis of concatenated sequences for nifD and nifH clustered both the Listia angolensis and the Lupinus texensis rhizobia into a single well-supported group with affinities to Rhizobium etli

CFN42T (Matt Parker unpublished data, Figure 3, Appendix VIII). This clade, along with other Rhizobium, Mesorhizobium and Ensifer strains, was grouped separately from the nifDH sequences of Azorhizobium and Bradyrhizobium strains. Microvirga is not a close relative of Rhizobium for housekeeping gene loci (Matt Parker unpublished data, Figure 1, Appendix VIII). The close affinity of Microvirga nif genes to those of Rhizobium, Mesorhizobium and Ensifer therefore suggests that these genes were acquired through horizontal transfer.

175 Chapter 4

4.4.2 G+C content, DNA:DNA hybridisation and cellular fatty acid analysis

The DNA G+C content of strains Lut5, Lut6T, WSM3557T and WSM3693T ranges from 61.9-63.0%, which is consistent with values obtained for other

Microvirga spp. (Table 1, Appendix IX, Anne Willems and Sofie De Meyer, unpublished data). A DNA:DNA hybridization (DDH) value equal to or higher than

70% has been recommended as the threshold for the definition of members of a species (Tindall et al., 2010). DDH values confirmed that WSM3557T, WSM3693T,

Lut6T and M. flocculans LMG 25472T, with 19-34.5% hybridization, represented separate species; while Lut5 and Lut6T, with 97% DDH, belong to the same species

(Anne Willems and Sofie De Meyer, unpublished data, Table 1, Appendix IX).

The major cellular fatty acids were 18:1 w7c (52.58-53%) and 19:0 CYCLO w8c (17.25-17.65%) for WSM3557T and WSM3693T and 18:1 w7c (68.94-69.71%) and SF2 (15.41-16.06%) for Lut5 and Lut6T (Anne Willems and Sofie De Meyer, unpublished data, Table 2, Appendix IX). Cellular fatty acid composition was similar for all Microvirga spp. and is a feature that distinguishes this group of bacteria from the phylogenetically related Chelatococcus asaccharovorans LMG 25503T (Table 2,

Appendix IX and Weon et al. (2010)).

4.4.3 Phenotypic characterisation of novel Microvirga species

The phenotypic features that distinguish Lut6T, WSM3557T and WSM3693T from other Microvirga spp. are given in Table 4.8.

176 Chapter 4

Table 4.8. Differentiating phenotypic characteristics of the novel strains Lut6T, WSM3557T and WSM3693T and the type strains of closely related species of the genus Microvirga. Strains: 1, M. lupini sp. nov. Lut6T; 2, M. lotononidis sp. nov. WSM3557T; 3, M. zambiensis sp. nov. WSM3693T; 4, M. flocculans TFBT (Takeda et al., 2004); 5, M. subterranea FaiI4T (Kanso & Patel, 2003); 6, M. guangxiensis 25BT (Zhang et al., 2009); 7, M. aerophila 5420S-12T (Weon et al., 2010); 8, M. aerilata 5420S-16T (Weon et al., 2010). All strains are rod-shaped, strictly aerobic and positive for catalase but negative for arginine dihydrolase and indole production. (+ = positive, w = weak, - = negative, ND = not determined)

Characteristic 1 2 3 4 5 6 7 8 Thermal Isolation source Root nodule Root nodule Root nodule Hot spring Soil Air Air aquifer Light pink, Cream, White, Light pink, Light pink, Light pink, Light pink, Colony Pale orange mucilaginous mucilaginous rough smooth, smooth, smooth smooth Polar Flagella Non-motile Polar flagella Polar flagella Non-motile Non-motile Non-motile Non-motile flagella 0.4 - 0.5 x 0.4 - 0.5 x 0.4 - 0.5 x 0.5 - 0.7 x 1 x 1.5 - 0.6 – 0.8 x 0.8 – 1.1 x 1.2 – 1.5 x Cell size (µm) 1.0 - 2.2 1.0 - 2.2 1.0 - 2.2 1.5 - 3.5 4.0 1.3 – 2.1 1.6 – 4.2 1.6 – 3.3 Optimum temp (◦C) 39 41 35 40 - 45 41 37 ND ND Growth range (◦C) 10 - 43 15 - 44 15 – 38 20 – 45* 25 - 45 16 - 42 10 - 35 10 - 35 MGT 1.8 hrs 1.6 hrs 1.7 hrs ND 4.5 hrs 230 min ND ND Optimum pH 7.0 – 8.5 7.0 – 8.5 7.0 – 8.5 7.0 7.0 7.0 ND ND PH growth range 5.5 –9.5 5.5 –9.5 6.0 –9.5 ND 6 – 9* 5.0 – 9.5 7.0 – 10.0 7.0 – 10.0 Optimum NaCl % 0.0 – 0.5 0.0 – 1.0 0.0 – 0.5 ND 0 ND ND ND NaCl growth range (%) 0 – 1.5 0 – 2.0 0 – 2.0 0 – 1.5* 0 – 1% 0 – 2.0 0 – 2.0 0 – 3.0 Growth supplement Vitamins or Vitamins or Yeast Yeast extract No No ND ND required yeast extract yeast extract extract AztR Antibiotic sensitivity GmR GmR GmR ND VmR EryR ND ND KmR DNA G+C content (% mol) 61.9 62.9± 0.1 62.6 64 63.5 ± 0.5 64.3 62.2 61.5 Symbiotic nitrogen fixation Yes Yes Yes ND ND ND ND ND

177 Chapter 4

Table 4.8. (cont.)

Characteristic 1 2 3 4 5 6 7 8 Carbon sources utilised L-Arabinose + + + ND ND ND - - D-Cellobiose + + + ND - - - - D-Fructose + + + - - - ND ND α-D-Glucose + + + - - + - - Succinate + + + - - ND ND ND Ethanol + - - - - - ND ND Glycerol - + + - - - ND ND Mannitol + + + - ND + - - p-Hydroxybenzoate + - + - ND ND ND ND Hydrolysis of gelatin - - - +* + - - W Hydrolysis of starch - - - -* - - + + Acid production from - - - - W - ND ND α-D-Glucose Oxidase - - - + - + + + Urease + + + - - + - - Tryptophan deaminase W W - ND - ND ND ND Acetoin production W W W - - - ND ND Nitrate reduction - + + - + + - - Azt = aztreonam; Ery = erythromycin; Gm = gentamicin; Km = kanamycin; Vm = vancomycin * Data taken from Weon et al. (2010)

178 Chapter 4

The morphological and physiological characteristics of the Listia angolensis and Lupinus texensis strains are consistent with those recorded for Microvirga spp., but somewhat atypical when compared with other Alphaproteobacterial rhizobia.

Motility is a variable character in Microvirga spp.; M. flocculans and M. subterranea are motile, while M. aerilata, M. aerophila and M. guangxiensis are not (Kanso &

Patel, 2003; Takeda et al., 2004; Weon et al., 2010; Zhang et al., 2009). Motility and chemotaxis are important in the ecology of rhizobia, allowing the bacteria to colonise the comparatively nutrient rich rhizosphere and conferring advantages in competition and symbiotic efficiency (Caetano-Anollés et al., 1988; Tambalo et al., 2010). All the Listia angolensis strains were motile in both plate and broth culture. It was therefore surprising that under the same growth conditions Lut5 and Lut6T were non- motile. It is possible that motility and flagellar synthesis are inducible rather than constitutive properties in Lut5 and Lut6T and culture media do not provide the appropriate conditions for such induction. Non-motility of rhizobial strains in laboratory conditions has also previously been noted for type IIA strains of

Rhizobium tropici grown on soft agar (Martínez-Romero et al., 1991).

In keeping with other species of Microvirga, the Listia angolensis and

Lupinus texensis strains have comparatively high temperature optima of 35ºC

(WSM3693T), 39ºC (Lut6T) and 41ºC (WSM3557T) and are able to tolerate temperatures of 38-45ºC. The optimal temperature for most rhizobia is 25 – 30ºC

(Garrity et al., 2005), although this may be a reflection of the preponderance of species isolated from temperate biomes. Strains of Ensifer saheli, Ensifer terangae and Rhizobium tropici, isolated from tropical legumes, are able to grow at 40 – 44ºC

(de Lajudie et al., 1994), while Methylobacterium nodulans grows optimally at 30 –

179 Chapter 4

37ºC (Jourand et al., 2004). Photosynthetic bradyrhizobial strains isolated from nodules of the semi-aquatic legume Aeschynomene indica are also able to grow at

41ºC (van Berkum et al., 2006). The mean generation time of 1.6 – 1.8 h for the

Listia angolensis and Lupinus texensis strains is faster than that recorded for M. subterranea and M. guangxiensis and faster than many other Alphaproteobacterial rhizobia. Mean generation time for the fast-growing rhizobial genera Azorhizobium,

Rhizobium and Ensifer is between 1.5 – 6 h (Garrity et al., 2005).

Neither the Listia angolensis nor the Lupinus texensis strains were salt tolerant, or acid tolerant, and they grew best in neutral to slightly basic conditions.

Their NaCl tolerance and pH growth range are consistent with reports for other

Microvirga species. Strictly aerobic growth is also characteristic of Microvirga, but growth in stab cultures indicates that WSM3557T, WSM3693T and Lut6T (and by implication the other Listia angolensis and Lupinus texensis strains) are microaerobic. Intrinsic antibiotic resistance is a distinguishing phenotypic feature: the Listia angolensis and Lupinus texensis strains are resistant to gentamycin, while

M. subterranea strain FaiI4T was vancomycin-resistant and M. guangxiensis strain

25BT was resistant to aztreonam, erythromycin and streptomycin sulphate (Kanso &

Patel, 2003; Zhang et al., 2009).

Another feature that distinguishes these novel strains is their pigmentation, which has previously been noted in only two other rhizobial species: the pigmented methylobacteria associated with species of Listia (previously Lotononis (Boatwright et al., 2011)) and the photosynthetic bradyrhizobia isolated from Aeschynomene spp.

(Jaftha et al., 2002; Molouba et al., 1999; Norris, 1958; Yates et al., 2007). The

180 Chapter 4 pigments extracted from WSM3557T and Lut6T are probably carotenoids, based on the absorption spectra, which show peaks at 450nm (Lut6T) and 495nm

(WSM3557T) (Figure 4.6). This accords with results obtained for M. subterranea

(Kanso & Patel, 2003). Carotenoids typically absorb at between 400–500nm and individual carotenoids possess unique absorption spectra (Britton et al., 2004). The absorption spectrum (and thus the type of pigment) in Lut6T is clearly different from that found in WSM3557T. In addition to carotenoids, the photosynthetic bradyrhizobia and the pigmented methylobacteria also contain bacteriochlorophyll, although the quantity of bacteriochlorophyll in the Listia methylobacteria is small, relative to that found in the bradyrhizobia (Evans et al., 1990; Jaftha et al., 2002).

The absence of absorption peaks at 700–800nm indicates that WSM3557T and Lut6T do not contain bacteriochlorophyll, which is consistent with previous findings for M. subterranea (Kanso & Patel, 2003). The absence of pigmentation in WSM3693T suggests that this trait is not essential for either saprophytic or symbiotic competence in this novel group of rhizobia.

4.4.3.1 Substrate utilisation

4.4.3.1.1 Critique of the use of Biolog plates in determining substrate utilisation

The Biolog systems of phenotype analysis (the GN2 and PM microarrays) allow rapid metabolic testing of microorganisms and have a wide variety of applications (Bochner, 1989; Bochner, 2003). Biolog microarrays have been employed in several previous examinations of rhizobial metabolism. These include phenotypic characterisation of novel Rhizobium and Burkholderia species (Berge et al., 2009; Chen et al., 2007), examination of the metabolic diversity of rhizobia

181 Chapter 4 associated with particular legume hosts (Wolde-Meskel et al., 2004) and assessment of phenotypic variability in Ensifer meliloti populations (Biondi et al., 2009).

In this study, Biolog GN2 and PM plates were used to characterise the metabolic properties of the novel Listia angolensis and Lupinus texensis strains.

Although the results were generally consistent, some caution needs to be exercised in the interpretation, as a number of substrates have shown a potential to give false positive or false negative readings. These potential false readings fall into three categories:

1) Substrates where the results for the PM plate carbon substrates have

differed from those obtained using the GN2 plates, or from growth

studies in minimal media;

2) The surprisingly large differences in the substrate utilisation of

WSM3557T and Lut6T on the PM plates, particularly for nitrogen,

phosphorus and sulfur sources;

3) Substrates that have been reported in the literature to give potential

false positives.

The differences in the GN2 plates as compared with the PM plates may be partially explained by the data from the PM plates being read after 56 h, as compared with the 96-h read for the GN2 plates, with the extra incubation time allowing more readings to reach a threshold level in the latter. Substrates that were demonstrably false negatives for WSM3557T on the PM plates include cellobiose, fructose, acetate, succinate and glutamate, as these carbon sources supported growth of the strain in

182 Chapter 4 minimal media. Lut6T similarly grew on acetate, but returned a negative result for acetic acid in the PM plates.

The results for WSM3557T on the PM plates, as compared with those of

Lut6T, may have been affected by the differences in exopolysaccharide production in these strains. WSM3557T is noticeably more mucilaginous than Lut6T, and excessive mucopolysaccharide has been advanced as a possible reason for less than optimal signal to noise ratios in PM plate reads (Biolog PM Services, pers comm). This may account for the reduced number of positive results for WSM3557T. The negative result obtained with WSM3557T and Lut6T for sulfate as a substrate in the PM plates, or indeed obtained by WSM3557T for any sulfur source, is probably an artefact, given that sulfur is essential for bacterial protein synthesis, including nitrogenase

(Krusell et al., 2005).

Potential false positive reactions have previously been reported in PM plates for the substrates dihydroxyacetone (a triose), and the pentose sugars D-xylose, D- ribose, L-lyxose, and D- and L-arabinose (Line et al., 2010). The utilisation of L- arabinose by the Listia angolensis and Lupinus texensis strains has been confirmed by subsequent growth in minimal media on this substrate (this study).

4.4.3.1.2 Substrate utilisation in Listia angolensis and Lupinus texensis strains

The putative type strains WSM3557T, WSM3693T and Lut6T differ from other species of Microvirga on the basis of carbon substrate assimilation, in particular their growth on D-cellobiose and D-fructose. Their ability to grow on a variety of sole carbon sources markedly distinguishes them from M. subterranea

183 Chapter 4

FaiI4T and M. flocculans TFBT, which both utilise a comparatively narrow range of carbon substrates. This may be related to the more oligotrophic nature of the aquatic environments from which the latter were isolated. Of 37 carbon substrates tested, M. flocculans TFBT grew only on yeast extract, peptone and tryptone Takeda et al.,

2004). M. subterranea FaiI4T grew on yeast extract, tryptone, casein hydrolysate, xylose and acetate (Kanso & Patel, 2003). M. aerophila and M. aerilata (isolated from atmospheric samples) are also reported to be unable to assimilate any of the substrates present in the API 20NE and ID 32GN strips (Weon et al., 2010). The soil-borne M. guangxiensis 25BT utilises a slightly different set of carbon sources from the Listia angolensis and Lupinus texensis strains. This strain uses arabitol, (+)-

D-glucose, myo-inositol, (+)-maltose, (+)-D-mannitol, (+)-melezitose, (+)-melibiose, peptone, (+)-D-sorbitol and (+)-D-xylose, but not (+)-cellobiose, ethanol, (+)-D- fructose, glycerol or α-L-rhamnose (Zhang et al., 2009).

The carbon substrate utilisation patterns of the Listia angolensis and Lupinus texensis strains, in minimal media and on the GN2 and PM microplates, demonstrate evidence of their adaptation to a soil or rhizosphere environment. They are able to metabolise arabinose, glucose, mannose, rhamnose, xylose and galacturonic acid, which are present in legume and other plant root mucilages (Knee et al., 2001).

WSM3693T and the Lut isolates grow on p-hydroxybenzoate, one of the aromatic compounds obtained from the breakdown of lignin, an important plant-derived component of land-based biomass (Harwood & Parales, 1996). The utilisation of some substrates may also have symbiotic implications. Catabolism of inositol and rhamnose has been shown to be important to rhizobial competitiveness in some symbioses (Jiang et al., 2001; Oresnik et al., 1998). Cellobiose and pectin, along

184 Chapter 4 with arabinose, glucose, rhamnose, xylose and galacturonic acid, are components of plant cell walls (Delmer & Amor, 1995; Mort & Grover, 1988). Penetration of the plant cell wall occurs as part of the rhizobial infection process, and the ability to degrade pectin may play a role in this process. Symbiotic induction of a pectate lyase has been demonstrated for Rhizobium etli (Fauvart et al., 2009). Rhizobia also require the ability to utilise dicarboxylic acids, as the dicarboxylates fumaric, succinic and malic acid are primary carbon sources for bacteroids within nodules

(Lodwig & Poole, 2003). In contrast to M. flocculans TFBT and M. subterranea

FaiI4T, the Listia angolensis and Lupinus texensis strains are able to grow on succinate in minimal media.

In pea nodules, Rhizobium leguminosarum bacteroids become symbiotic auxotrophs and are reliant on the legume host for the supply of the branched chain amino acids leucine, isoleucine and valine (Prell et al., 2009). It is therefore interesting that these amino acids were not metabolised as either carbon or nitrogen sources in the PM and GN2 microplates. These are preliminary results that need to be confirmed, but may reflect differences in the functioning of the pea nodule, as compared with the structurally different nodules of Listia and Lupinus species

(Chapter 3). Utilisation of at least one of these branched chain amino acids as a carbon source has been demonstrated for strains of Mesorhizobium loti, Rhizobium leguminosarum, Rhizobium tropici, Ensifer fredii, Ensifer meliloti, Ensifer saheli and

Ensifer terangae, but not for Azorhizobium caulinodans (de Lajudie et al., 1994). On

Biolog PM microplates, E. meliloti 1021 utilises leucine, but not isoleucine and valine, as carbon sources, and all three amino acids as nitrogen sources (Biondi et al.,

2009).

185 Chapter 4

Metabolic versatility varies among different rhizobial taxa. A recent paper by

Biondi et al. (2009) used Biolog PM plates to analyse the metabolic capacity of E. meliloti strains and allows direct comparisons to be made between E. meliloti strain

Sm1021 and WSM3557T and Lut6T. Sm1021 utilised a far wider range of substrates

(84 of 190 carbon sources, 84 of 95 nitrogen sources, 55 of 59 phosphorus sources and 29 of 35 sulfur sources (Biondi et al., 2009)) than did either WSM3557T or

Lut6T (Section 4.3.2.5.2; Table 1, Appendix II). The Microvirga rhizobial strains described in this study thus appear to have a relatively narrow range of carbon substrate utilisation. Interestingly, the carbon substrate utilisation pattern of the

Microvirga rhizobia also differs from that of the pigmented Methylobacterium rhizobia associated with Listia spp., which on GN2 plates oxidise very few carbohydrates and all 24 carboxylic acid sources (Jaftha et al., 2002).

The requirement for supplementary growth factors varies in Microvirga species. Yeast extract is an absolute requirement for growth of M. subterranea

FaiI4T, Lut5 and Lut6T in minimal media, while the Listia angolensis strains require either yeast extract or the complex vitamin mix detailed in Egli and Auling (2005). In contrast, M. flocculans TFBT and M. guangxiensis 25BT require no supplements for growth in minimal media. Vitamin requirements vary amongst rhizobial strains; biotin, thiamine and calcium pantothenate are essential for the growth of many fast- growing species (Graham, 1963), but not required for some strains of slow-growing rhizobia (Stowers & Elkan, 1984). The Listia angolensis and Lupinus texensis strains appear to have the most stringent growth factor requirements of any rhizobial species. Interestingly, methylobacteria require no supplements for growth in minimal

186 Chapter 4 media (Green, 1992); this holds true also for the pigmented Listia methylobacteria (J.

Ardley, unpublished data).

4.4.4 Biochemical characteristics of Listia angolensis and Lupinus texensis strains

The results obtained for the Listia angolensis and Lupinus texensis strains from the API 20E strips and other biochemical tests are consistent with results obtained for other species of Microvirga (Table 4.8). All strains of bacteria in this clade gave negative results for indole production, glucose fermentation and arginine dihydrolase, but were positive for catalase. Hydrolysis of gelatin and starch, reduction of nitrate to nitrite, production of oxidase and urease and weak production of acetoin and tryptophan deaminase are distinguishing phenotypic features for the different species. The Listia angolensis and Lupinus texensis strains and M. guangxiensis 25BT did not produce acid from growth on D-glucose or D-mannitol

(Zhang et al., 2009; this study). M. subterranea FaiI4T is reported to show weak production of acid from D-glucose, although it is unable to grow on this substrate

(Kanso & Patel, 2003).

4.4.5 Host range

Previous reports have indicated that both the Listia angolensis and Lupinus texensis strains have a narrow host range (Andam & Parker, 2007; Yates et al.,

2007). The results in this study confirm this finding, as WSM3557T, WSM3693T,

Lut5 and Lut6T were able to elicit nodules on some other, usually promiscuous, hosts; but such nodulation was usually only occasional and always ineffective. The host range of the Lupinus texensis strains appears to be narrower than that of the

Listia angolensis strains. Lut6T is unable to nodulate other native North American

187 Chapter 4

Lupinus spp. (Andam & Parker, 2007), while WSM3557T, although unable to nodulate Listia bainesii and Listia heterophylla, is able to nodulate, and in some cases fix nitrogen, with other Lotononis s. l. species (Chapter 2). Furthermore,

WSM3557T and WSM3693T are able to form ineffective nodules on Lupinus texensis

(Matt Parker, unpublished data), whereas Lut5 and Lut6T are unable to nodulate

Listia angolensis. This is an interesting finding, considering the specificity of L. texensis. It also suggests that the molecular basis for specificity in these symbioses is not limited to Nod factors, given that the nodA phylogenetic tree (Figure 2, Appendix

VIII) places the nodA of Lut6T in a separate group to that of WSM3557T and

WSM3693T.

4.5 Summary, emended description of Microvirga and description of Microvirga lotononidis sp. nov., Microvirga zambiensis sp. nov. and Microvirga lupini sp. nov.

In summary, on the basis of 16S rRNA and concatenated housekeeping gene sequence identity, the Listia angolensis strains WSM3557T and WSM3693T and the

Lupinus texensis strains Lut5 and Lut6T belong to the genus Microvirga. The ability of these strains to nodulate and form N2-fixing symbioses with species of leguminous plants distinguishes them from previously reported species of Microvirga. Additional genotypic, phenotypic, and chemotaxonomic data support the classification of these strains as three new species within the genus Microvirga. The names M. lotononidis sp. nov., M. zambiensis sp. nov. and M. lupini sp. nov. are proposed, with the isolates

WSM3557T, WSM3693T and Lut6T representing the respective type strains.

188 Chapter 4

4.5.1 Emended description of Microvirga (Kanso & Patel, 2003 emend.

Zhang et al, 2009, emend. Weon et al, 2010)

The description remains as given by Kanso & Patel (2003), Zhang et al.

(2009) and Weon et al. (2010), with the following modifications. Some strains are capable of nodulation and symbiotic nitrogen fixation with legumes. The type species is Microvirga subterranea.

4.5.2 Description of Microvirga lotononidis sp. nov.

Microvirga lotononidis (lo.to.no'ni.dis. N.L. gen. n. lotononidis, of Lotononis, a taxon of leguminous plants, referring to the isolation source of the first strains, nodules of Listia angolensis, a species in the Lotononis s. l. clade.

Cells are strictly aerobic, asporogenous, Gram-negative rods (0.4-0.5 x 1.0-2.2

µm), motile with one or more polar flagella. Grows well on YMA, ½ lupin agar, TY agar and nutrient agar. On ½ LA after three days at 28 ºC, colonies are light pink, convex, smooth, mucilaginous and circular, with entire margins, 0.5-1.5 mm in diameter. Grows from 15-44/45 ºC; optimum temperature for the type strain is 41 ºC and mean generation time at this temperature is 1.6 h. Best growth is at pH 7.0-8.5

(range 5.5-9.5), and 0.0-1.0 % (w/v) NaCl (range 0-2.0 % (w/v)). Yeast extract or the vitamin mix detailed in Egli and Auling (2005) is an absolute requirement for growth in minimal media. The main cellular fatty acids are 18:1 ω7c and 19:0 cyclo ω8c.

Positive for catalase and urease and weakly positive for tryptophan deaminase and acetoin production. Oxidase, β-galactosidase, arginine dihydrolase, lysine decarboxylase, ornithine decarboxylase, indole and hydrogen sulphide production are negative, as is utilisation of citrate. Gelatin and starch are not hydrolysed. Nitrite is produced from nitrate. Acid is produced from growth on L-arabinose but not from

189 Chapter 4 growth on α-D-glucose or D-mannitol. Resistant to gentamicin and some strains are partially resistant to ampicillin, chloramphenicol, kanamycin and spectinomycin.

Sensitive to nalidixic acid, rifampicin, streptomycin and tetracycline. Assimilates L- arabinose, D-cellobiose, D-fructose, α-D-glucose, glycerol, D-mannitol, acetate, succinate and glutamate. The G + C content of the type strain is 62.8-63.0 %.

The type strain WSM3557T (= LMG26455T, = HAMBI3237T) and other strains were isolated from N2-fixing nodules of Listia angolensis originally collected in

Zambia.

4.5.3 Description of Microvirga zambiensis sp. nov.

Microvirga zambiensis (zam.bi.en'sis. N.L. fem. adj. zambiensis, of or belonging to Zambia, from where the type strain was isolated).

Cells are strictly aerobic, asporogenous, Gram-negative rods (0.4-0.5 x 1.0-

2.2 µm), motile with one or more polar flagella. Grows well on YMA, ½ lupin agar,

TY agar and nutrient agar. On ½ LA after three days at 28ºC, colonies are cream coloured, convex, smooth, mucilaginous and circular, with entire margins, 0.5-1.5 mm in diameter. Grows from 15-38 ºC; optimum temperature is 35 ºC and mean generation time at this temperature is 1.7 h. Best growth is at pH 7.0-8.5 (range 6.0-

9.5) and 0.0-0.5 % (w/v) NaCl (range 0-1.5 % (w/v)). Yeast extract or the vitamin mix detailed in Egli and Auling (2005) is an absolute requirement for growth in minimal media. The main cellular fatty acids are 18:1 ω7c and 19:0 cyclo ω8c.

Positive for catalase and urease and weakly positive for acetoin production. Oxidase,

β-galactosidase, arginine dihydrolase, lysine decarboxylase, ornithine decarboxylase, tryptophan deaminase, indole and hydrogen sulphide production are negative, as is

190 Chapter 4 utilisation of citrate. Gelatin and starch are not hydrolysed. Nitrite is produced from nitrate. Acid is produced from growth on L-arabinose but not from growth on α-D- glucose or D-mannitol. Resistant to gentamicin. Sensitive to ampicillin, chloramphenicol, kanamycin nalidixic acid, rifampicin, spectinomycin, streptomycin and tetracycline. Assimilates L-arabinose, D-cellobiose, D-fructose, α-D-glucose, glycerol, D-mannitol, acetate, succinate, p-hydroxybenzoate and glutamate. The G +

C content of the type strain is 62.6 %.

The type strain WSM3693T (= LMG26454T, = HAMBI3238T) was isolated from N2-fixing nodules of Listia angolensis originally collected in Zambia.

4.5.4 Description of Microvirga lupini sp. nov.

Microvirga lupini (lu.pi'ni. L. n. lupinus, a lupine and also a botanical generic name (Lupinus); L. gen. n. lupini, of Lupinus, isolated from Lupinus texensis.

Cells are strictly aerobic, asporogenous, Gram-negative non-motile rods (0.4-

0.5 x 1.0-2.2 µm). Grows well on YMA, ½ lupin agar, TY agar and nutrient agar. On

½ LA after three days at 28 ºC, colonies are pale orange, convex, smooth and circular, with entire margins, 0.5-1.5 mm in diameter. Grows from 10-43 ºC; optimum temperature is 39 ºC and mean generation time at this temperature is 1.8 h.

Best growth is at pH 7.0-8.5 (range 5.5-9.5) and 0.0-0.5 % (w/v) NaCl (range 0-1.5

% (w/v)). Yeast extract is an absolute requirement for growth in minimal media. The main cellular fatty acids are 18:1 ω7c and summed feature 2 (16:1 iso I / 14:0 3 OH / unknown 10.938). Positive for catalase and urease and weakly positive for tryptophan deaminase and acetoin production. Oxidase, β-galactosidase, arginine dihydrolase, lysine decarboxylase, ornithine decarboxylase, indole and hydrogen

191 Chapter 4 sulphide production are negative, as is utilisation of citrate. Gelatin and starch are not hydrolysed. Nitrite is not produced from nitrate. Acid is produced from growth on L- arabinose but not from growth on α-D-glucose or D-mannitol. Partially resistant to ampicillin, chloramphenicol, gentamicin and streptomycin and sensitive to kanamycin, nalidixic acid, rifampicin, spectinomycin and tetracycline. Assimilates

L-arabinose, D-cellobiose, D-fructose, α-D-glucose, D-mannitol, acetate, succinate, glutamate, ethanol and p-hydroxybenzoate. The G + C content of the type strain is

61.9 %.

The type strain Lut6T (= LMG26460T; = HAMBI3236T) and other strains were isolated from N2-fixing nodules of Lupinus texensis collected in Texas, USA.

192 Chapter 5

CHAPTER 5

General discussion

193 Chapter 5

194 Chapter 5

5.1 Symbiotic relationships within Lotononis s. l.

The nitrogen-fixing symbiosis between the southern African crotalarioid legume Listia bainesii (formerly Lotononis bainesii (Boatwright et al., 2011) and strains of pigmented Methylobacterium sp. has long been known to be a model of symbiotic specificity, both for the plant and the microsymbiont (Broughton et al.,

1986; Norris, 1958; Pueppke & Broughton, 1999). Until recently, however, this symbiosis had been examined in isolation and outside the context of the symbiotic relationships and biogeography of the Lotononis s. l. clade (comprising the genera

Leobordea, Listia and Lotononis s. str. (Boatwright et al., 2011)) in its southern

African centre of origin. Studies undertaken by the Centre for Rhizobium Studies

(Yates et al., 2007; R. Yates, unpublished data) partially addressed this issue by showing that all examined Listia species were effectively nodulated by pigmented

Methylobacterium sp. strains, except for the more tropically distributed Listia angolensis, which forms N2-fixing nodules with a novel lineage of rhizobia. No previous study, however, has examined the symbiotic relationships between the different Lotononis s. l. taxa and their cognate rhizobia. The work presented here sought to assess the nodulation and N2-fixation abilities of Lotononis s. l.- associated rhizobia on host species that were representative of the taxonomic diversity of the

Lotononis s. l. clade. To determine whether symbiotic specificity in the genus Listia was confined to L. bainesii, three Listia host species (L. angolensis, L. bainesii and

L. heterophylla) were also inoculated with type strains, or well-characterised strains, of diverse rhizobia. In addition, this work sought to analyse the phylogeny of the chromosomal and symbiotic genes of Lotononis s. l.-associated rhizobia, to investigate infection, nodule initiation and/or nodule morphology in Lotononis s. l.

195 Chapter 5 species and to validly name and characterise the L. angolensis rhizobia, identified in this study as novel species of Microvirga.

The results of these studies demonstrate three salient features, summarised in

Figure 5.1, of the symbioses between Lotononis s. l. species and their cognate rhizobia. Firstly, the rhizobia isolated from nodules of Lotononis s. l. species are remarkably diverse. Authenticated rhizobial strains able to nodulate Lotononis s. l. hosts have been identified as belonging to the genera Bradyrhizobium, Ensifer,

Mesorhizobium, Methylobacterium and Microvirga.

196 Chapter 5

Li. angolensis 1 Br. sp WSM2596 Li. bainesii Li. heterophylla* Br. sp WSM2632 Br. sp WSM2783 Le. longiflora

Le. foliosa† Mic. lotononidis WSM3557 2 Le. mollis‡ Le. calycina† Mtb. nodulans WSM2667

Mtb. sp. WSM2598 Le. platycarpa Le. stipulosa Mes. tianshanense WSM2624

Le. bolusii Le. polycephala ‡

Lo. crumanina Lo. delicata ‡ Lo. falcata 3 E. meliloti WSM2653 Lo. laxa ‡ E. meliloti WSM3040 Lo. pungens ‡

Figure 5.1. Symbiotic and phylogenetic relationships of Lotononis s. l. species and associated rhizobia. Rhizobial phylogeny is based on the 16S rRNA gene; Lotononis s. l. phylogeny is according to van Wyk (1991). Filled circles = effective nodulation; empty circles = ineffective nodulation; barred circles = the rhizobial strain was isolated from this host, but the host was not tested for nodulation with any rhizobia; no circle = no nodulation. Leobordea bolusii, for example, is effectively nodulated by the Bradyrhizobium and Microvirga strains, and forms ineffective nodules with the remaining inoculant strains. 1 = Listia; 2 = Leobordea; 3 = Lotononis s. str. Br. = Bradyrhizobium; E. = Ensifer; Mes. = Mesorhizobium; Mic. = Microvirga; Mtb. = Methylobacterium. * Listia heterophylla was only tested with WSM2598 and WSM3557. † These host species were not inoculated with any rhizobia. ‡ These host species were not inoculated with WSM2598 or WSM3557.

197 Chapter 5

Secondly, different specificity groups exist within the Lotononis s. l. clade.

Leobordea and Lotononis s. str. appear to be more or less promiscuous and able to nodulate (often ineffectively) with a range of Lotononis s. l. rhizobia. In contrast,

Listia species are more specific. This study has confirmed the extreme symbiotic specificity of L. bainesii, which does not nodulate with any of the Lotononis s. l. rhizobia, other than its cognate methylobacteria. Additionally, none of the diverse rhizobial strains (including Azorhizobium, Bradyrhizobium, Burkholderia, Ensifer and Rhizobium species and Methylobacterium nodulans) inoculated onto L. bainesii and L. heterophylla was able to elicit nodules on these host plants. From these and previous results (Norris, 1958; Yates et al., 2007; R. Yates, unpublished data) it appears likely that this pattern of extreme symbiotic specificity applies to all the

Listia species that are effectively nodulated by these methylobacteria. The single exception is L. angolensis, which can be nodulated by the Listia methylobacteria and by Methylobacterium nodulans strains ORS2060 and WSM2667 (but not by any other tested rhizobial strains), but forms effective symbioses only with the novel

Microvirga species.

Thirdly, the rhizobia associated with Lotononis s. l. species exhibit a wide range of variability in both host range and effectiveness. The extremely narrow host range of the Listia methylobacteria was confirmed. Methylobacterium strain

WSM2598 was highly effective on L. bainesii, but ineffective on L. angolensis and only able only to form (ineffective) nodules on the most promiscuous species of the remaining Lotononis s. l. hosts. Somewhat surprisingly, of all the rhizobial isolates, the Microvirga strain WSM3557 was effective over the widest range of Lotononis s. l. host species. The remaining rhizobial strains did not appear to be strongly

198 Chapter 5 associated with any taxonomic group of Lotononis s. l. and were generally poorly effective on these hosts.

As legume hosts radiate from their centre of origin and adapt to different climatic and edaphic conditions, what happens to their associations with their microsymbionts? The results obtained for the Lotononis s. l. clade raise interesting questions about this process and the development of symbiotic specificity that has occurred in the genus Listia.

5.1.1 Habitat as a driver of rhizobial diversity

It is not surprising that rhizobia isolated from Lotononis s. l. hosts should be taxonomically diverse. As South Africa is the centre of origin of these legumes, their microsymbionts would also be expected to have a wide range of genotypes, in accordance with gene centre theory (Andronov et al., 2003). Numerous previous studies have demonstrated the genotypic diversity of rhizobia isolated from wild legumes in a given geographical area (Han et al., 2009; Lorite et al., 2010; Odee et al., 2002; Rasolomampianina et al., 2005; Sylla et al., 2002; Zhao et al., 2010). In this regard, topo-edaphic heterogeneity, seen for example in the Cape Floristic

Region, may not only be a driver of diversification and speciation of the flora of this region (Cowling et al., 2009) but may also be a factor in the genotypic diversity of the rhizobia associated with the Lotononis s. l. clade. As rhizobia are not obligate, vertically transmitted symbionts (Young & Johnston, 1989), free-living root nodule bacteria require saprophytic competence in the host plant’s preferred habitat if they are to be available for selection by the host. There are characteristic differences in the ability of various species and genera of rhizobia to tolerate acid or alkaline conditions (Graham, 2008; Herridge, 2008; Zahran, 1999) and several studies on

199 Chapter 5 legumes nodulated by taxonomically diverse rhizobia have found a correlation between eco-regions and/or edaphic factors and the microsymbiont genotype (Bala &

Giller, 2006; Diouf et al., 2007; Garau et al., 2005; Han et al., 2009; Lu et al., 2009).

In the case of the genus Listia, the waterlogged habitat favoured by these plants (van Wyk, 1991) may be an important determining factor in their symbiotic association with strains of Methylobacterium and Microvirga, two bacterial genera not commonly known to contain rhizobial species. In addition, Listia species are distinguished by their adventitious roots, which arise from stem nodes, and by lupinoid nodules, which typically develop on the hypocotyl and tap root (Boatwright et al., 2011; Yates et al., 2007; this study). Methylobacterium and Microvirga rhizobia associated with Listia species may harbour traits that confer a competitive advantage for saprophytic competence in waterlogged environments, or for colonisation of the hypocotyl and rhizosphere in these conditions.

Members of the genus Methylobacterium have demonstrated saprophytic competence in aquatic habitats, being ubiquitous in water, as well as soil and air environments; numerous strains have been isolated from water samples (Gallego et al., 2005; Gallego et al., 2006; Green, 1992; Hiraishi et al., 1995). Methylobacteria are known to colonise plant surfaces (Madhaiyan et al., 2009b), although it is interesting that a recent study on the ability of epiphytic methylobacteria to colonise the Arabidopsis thaliana phyllosphere found that rhizobial Methylobacterium species are not competitive in this environment (Knief et al., 2010). Microvirga species have similarly been isolated from aquatic or potentially waterlogged environments (Kanso

& Patel, 2003; Takeda et al., 2004; Zhang et al., 2009). Additionally, the ability of

200 Chapter 5 all described Microvirga species (including the novel rhizobial Microvirga species characterised in this study) to tolerate relatively high temperatures correlates well with their isolation from thermal waters or subtropical regions (Kanso & Patel, 2003;

Takeda et al., 2004; Zhang et al., 2009) and with the isolation of rhizobial

Microvirga species from the tropically or sub-tropically distributed Listia angolensis and Lupinus texensis (Andam & Parker, 2007; van Wyk, 1991; Yates et al., 2007).

Strikingly, the legume nodule isolates AC72a and ARRI 185 (genospecies AL) that have now been identified as Microvirga strains (this study) have also come from tropically or sub-tropically distributed host plants (Lafay & Burdon, 2007; Wolde-

Meskel et al., 2005). Graham (2008) has recommended that new rhizobial organisms be examined for unusual traits that could influence their ecological performance.

Both the Methylobacterium and Microvirga microsymbionts are unusual and uncommon species of rhizobia. They merit further study to determine which genes and physiological factors contribute to saprophytic or competitive ability within the host plant’s environment.

Environmental pressures may drive the selection of saprophytically competent rhizobia, but do host plants provide a similar selective pressure for particular phylogenies of rhizobial nod genes? Evidence for this varies and can be examined in the context of the Lotononis s. l.-rhizobia symbioses.

5.1.2 Lotononis s. l. rhizobia have diverse nodA sequences

The nod genes encode Nod factors that are required by most rhizobia for infection and nodulation of the legume host and are important in determining host range and specificity (Dénarié et al., 1996; Perret et al., 2000). The nodA gene is a useful symbiotic marker as it encodes a product that is a host-specific determinant of

201 Chapter 5 the transfer of fatty acids in Nod factor biosynthesis (Dénarié et al., 1996; Haukka et al., 1998; Roche et al., 1996) and nodA phylogeny can give an indication of the Nod factor structure (Moulin et al., 2000; Moulin et al., 2004). Previous studies of legume-rhizobia symbiotic relationships have found a correlation between nod gene phylogeny and host range (Ba et al., 2002; Dobert et al., 1994; Haukka et al., 1998;

Laguerre et al., 2001; Lu et al., 2009); in other studies, however, this association is not as strong and nod gene phylogeny is more closely aligned with the rhizobial chromosomal background than with legume taxonomy (Han et al., 2010; Lorite et al., 2010; Zhao et al., 2010).

The nodA genes of rhizobia associated with the Lotononis s. l. clade are polyphyletic and their phylogenies appear to correlate more with those of the rhizobial 16S rRNA genes than with the taxonomy of the host plants, as the polyphyletic nodA lineages of the Lotononis s. l. rhizobia are associated with taxonomically diverse legumes (Figure 2.11). The high nodA sequence similarity seen in strains within genera (and within the pigmented Methylobacterium and

Methylobacterium nodulans species) is consistent with horizontal gene transfer

(HGT) and supports the view of Wernegreen & Riley (1999) that HGT occurs within, but not between, genetic subdivisions. That HGT does occur between rhizobial genera has been shown in studies by Andam et al. (2007), Barcellos et al.

(2007) and Cummings et al. (2009), which provide evidence for symbiotic gene transfer from Burkholderia to Cupriavidus strains, from a Bradyrhizobium strain to an indigenous Ensifer strain, and from an Ensifer strain to a novel rhizobial

Agrobacterium strain, respectively. Presumably, however, these are comparatively rare events due to the mechanistic barriers that constrain HGT between

202 Chapter 5 phylogenetically distant species (Lawrence & Hendrickson, 2003), whereas HGT between related rhizobia is more common. Physical proximity also appears to be important, as the main trends in gene transfer are observed among species from different taxa inhabiting the same habitat (Kloesges et al., 2011).

In this regard, it is significant that aquatic environments (which, in comparison to soil, are oligotrophic) support fewer bacterial taxa than soil, and saturated soils have both fewer taxa and a more uneven distribution than unsaturated soils (Horner-Devine et al., 2004; Zhou et al., 2002). The implications for the Listia- rhizobia symbioses are that their waterlogged habitat may reduce the diversity of the rhizobial pool available for plant selection and act as an additional barrier to the transfer of symbiotic genes across bacterial taxa.

5.1.3 Specificity and effectiveness within Lotononis s. l.

Previous studies of wild legumes nodulated by diverse rhizobia have noted the generally promiscuous nature of these relationships and the preference of endemic rhizobial populations of a certain geographical area for a wide spectrum of hosts (Han et al., 2010; Zahran, 2001). There appears to be a natural variation in the symbiotic infectivity and effectiveness of rhizobial isolates within wild host species

(Burdon et al., 1999; Odee et al., 2002; Thrall et al., 2000). Similarly, the variations in response to an inoculant strain (i.e. no nodulation, ineffective nodulation or effective nodulation) that were seen in individual plants within species of Leobordea and Lotononis s. str. (where seeds from wild plants, rather than from a particular accession were used) have been found in other studies (Wilkinson et al., 1996). This is likely to be related to seed provenance and variability in the host genotype, leading to differing symbiotic compatibilities with rhizobial partners.

203 Chapter 5

In contrast, all Listia species, other than L. angolensis, form effective N2- fixing associations with all tested strains of their cognate, pigmented methylobacteria

(Eagles & Date, 1999; Yates et al., 2007; R. Yates, unpublished data). It is remarkable that this effectiveness extends across a range of plant accessions and over

80 isolates collected across a broad geographical area in South Africa and Zimbabwe and is in direct contrast to the variability in effectiveness seen in individual plants within species of Leobordea and Lotononis s. str. Clearly, despite the variability in host genotypes (five Listia species are known to nodulate with pigmented methylobacteria (R. Yates, unpublished data); additionally, L. bainesii is highly allogamous (Real et al., 2004)), effective symbiosis with pigmented methylobacteria is strongly selected for in this group of legumes. The results confirm Sprent’s

(2008b) observations that specificity is associated with effectiveness, but show that this close mutual relationship is not confined to temperate areas, but also found in legumes from warmer latitudes.

Hosts are presumably under selection pressure to maximise symbiont effectiveness (Douglas, 1998). It has been experimentally shown that legumes are able to selectively favour the most beneficial RNB by imposing sanctions on non- fixing rhizobia (Kiers et al., 2008). In mutualisms, however, there is a potential for conflict of interest between the symbiont and the host (Herre et al., 1999). Factors that are suggested to align these interests are: vertical transmission of symbionts, genotypic uniformity of symbionts within individual hosts, spatial structure of populations leading to repeated interactions between would-be mutualists, and restricted options outside the relationship for both partners; conversely, horizontal transmission, multiple symbiont genotypes and varied options decrease symbiotic

204 Chapter 5 stability (Herre et al., 1999). The legume-rhizobia symbiosis falls into the latter category (Kiers et al., 2008). In the case of Listia species, however, the waterlogged habitat of the host plants may potentially reduce the diversity of rhizobial partners available for host selection. In such a case, the selection pressures may favour the co- evolution of the host and microsymbiont towards a more effective symbiosis and the development of specificity, as seen in the Listia-Methylobacterium symbiosis. Thrall et al. (2000) have noted that Acacia species with more limited distributions or tighter ecological requirements have a greater degree of specificity than widespread species.

An analysis of the rhizobial diversity present in habitats favoured by Listia species would confirm whether or not the host plants face a restricted choice of rhizobial partner. The reduced specificity seen in L. angolensis may be a function of the putative expansion of L. angolensis into more tropical habitats and symbiont replacement of Methylobacterium species by less symbiotically adapted Microvirga strains. Molecular clock analyses of the divergence of L. angolensis from other Listia species and of the genus Listia from the Lotononis s. l. clade could help answer these questions.

5.1.4 Models of legume-rhizobia symbiosis

From the data presented here and in previous studies, there appear to be several models of the relationships between legume hosts, rhizobial chromosomal backgrounds and nodulation genes:

1) Legumes are generally promiscuous and able to associate with a wide range of different rhizobial chromosomal backgrounds and nod gene lineages, as seen in species of Leobordea and Lotononis s. str. in this study and in Lathyrus,

Lotus and Sophora species (Han et al., 2010; Zhao et al., 2010)

205 Chapter 5

2) Legumes are associated with both a particular rhizobial chromosomal background and a nod gene lineage. This model is seen in species in the Vicieae tribe that are nodulated by Rhizobium leguminosarum bv vicieae strains containing a distinct nodD clade (Mutch & Young, 2004) and in European Genisteae bradyrhizobia, which all group within the nodA gene clade II (Kalita et al., 2006;

Moulin et al., 2004; Stepkowski et al., 2007).

3) Legumes associate with rhizobia of different chromosomal backgrounds, but harbouring similar nod genes. This infers horizontal transfer of the symbiotic genes. Rogel et al. (2011) have used the term “symbiovar” to distinguish symbiotically distinct subgroups within a single rhizobial species. Examples include the symbioses between Acacia tortilis and Phaseolus vulgaris and their diverse rhizobia (Ba et al., 2002; Laguerre et al., 2001).

4) Occasionally, a subset of legumes within a taxon is specifically nodulated by rhizobia that are both chromosomally distinct and harbour a different lineage of nod genes from the rhizobia that nodulate other hosts within that legume taxon. This has been observed in Listia (this study) and in species of Crotalaria (Renier et al.,

2008), Lupinus (Andam & Parker, 2007), Rhynchosia (Garau et al., 2009) and

Sesbania (Cummings et al., 2009).

This suggests that the different symbiotic models are functions of the weight assigned to selective pressures exerted by the environment and the host plant on rhizobia. Further research on the microsymbionts of wild legumes in their centres of diversity, especially in environments that exert a strong selective pressure on the rhizobia, could help to shed light on the evolution of these symbiotic relationships.

206 Chapter 5

5.2 Infection and nodule organogenesis in Listia spp.

The initial infection process in Listia species is remarkable, in that epidermal cells on the hypocotyl and root, unlike other studied legumes (Bhuvaneswari et al.,

1980; Tang et al., 1992), retain their potential for infection, regardless of age. This study also indicates that rhizobial infection and nodulation in Listia species, as in other genistoid legumes, does not involve root hair curling (RHC) and development of infection threads (ITs). It has confirmed the value of nodule morphology and structure as a marker for legume taxonomy, as lupinoid nodules are synapomorphic for the genus Listia. Nodules from Crotalarieae species are also confirmed to contain uniformly infected central tissue with no uninfected interstitial cells.

Root hair infection is considered to be an advanced feature that provides the legume with a tighter and more selective control of bacterial passage through the epidermis (Madsen et al., 2010), while epidermal entry has been linked with more promiscuous nodulation (Boogerd & van Rossum, 1997; Sprent, 2007). The effective and highly specific symbiosis of L. bainesii with its cognate methylobacterial rhizobia shows, however, that epidermal entry does not preclude specificity.

Most of the research on the molecular mechanisms that govern specificity and signalling pathways has concentrated on legumes that are infected via root hair curling, including the model plants Lotus japonicus and Medicago truncatula. The choice of these two species has been justified partly because they exhibit different developmental systems (determinate and indeterminate nodules, respectively)

(Maunoury et al., 2008). The mechanics of infection and nodule organogenesis in these model legumes are similar, however, whereas bacterial entry and nodule

207 Chapter 5 organogenesis in the epidermally infected dalbergioid and genistoid legumes are structurally different to the processes seen in IT legumes. There are several commonalities, but also several points of departure, in the nodulation signalling pathways of IT and non-IT legumes. The key symbiotic genes SymRK and CCaMK

(Oldroyd & Downie, 2008) have also been isolated and characterised in Lupinus species and Arachis hypogea, respectively (Mahé et al., 2011; Sinharoy & DasGupta,

2009). Nod factors are required for induction of meristematic activity in cortical cells, but not for rhizobial colonisation of the root cortex in the dalbergioid legume

Arachis (Ibáñez & Fabra, 2011); additionally, Nod-factor-independent infection and nodulation occurs in some Aeschynomene species nodulated by specific strains of

Bradyrhizobium (Giraud et al., 2007; Miché et al., 2010).

Deciphering the signalling pathways and the molecular basis of specificity in non-IT symbioses is important for several reasons. Firstly, it provides a greater understanding of the development and evolution of N2-fixing symbioses within legumes. Secondly, determining the means by which legume hosts exclude non- effective strains of rhizobia could arguably aid the maintenance of effective symbioses in agricultural systems, where competition for nodulation by ineffective strains is a constraint to N2 fixation (Howieson et al., 2008). Finally, a greater understanding of the nodulation pathway in legumes may offer the possibility of extending N2-fixing symbiosis to cereal and other non-legume crops. Charpentier &

Oldroyd (2010) have suggested that the comparatively primitive crack entry nodulation pathway is a more realistic target for transfer to cereals than the root hair infection process. In this case, knowledge of the molecular mechanisms that govern

208 Chapter 5 specificity, infection and nodule organogenesis in non-IT legumes will be a vital tool in the development of strategies to effect this transfer.

5.3 Characterisation of novel rhizobial species of Microvirga

An important outcome of this study has been the naming and description of three novel rhizobial species of Microvirga, a genus not previously known to contain strains capable of symbiotic nitrogen fixation with legumes, or to be associated with plants. This extends the range of bacterial chromosomal backgrounds that are able to support the processes of infection, nodulation and nitrogen fixation with a legume host. With the addition of the novel rhizobial species, there are now eight described species in the genus Microvirga. The rhizobial strains AC72a (Wolde-Meskel et al.,

2005) and ARRI 185 (Lafay & Burdon, 2007) also belong within this clade, as do several other as-yet-uncharacterised legume nodule isolates, based on the 16S rRNA gene sequences deposited in the GenBank database (accession numbers EU618028,

GQ922067, HM042680). This points to the existence of additional potentially novel rhizobial species within the genus Microvirga.

It is interesting to compare the rhizobial Microvirga with the described non- rhizobial species. The substrate utilisation patterns of the rhizobial Microvirga strains differ markedly from those of the aquatic Microvirga flocculans and Microvirga subterranea (Kanso & Patel, 2003; Takeda et al., 2004) and provide evidence of the adaptation of the rhizobial strains to the more nutrient-rich soil or rhizosphere environment. Rhizobial Microvirga utilise a wider range of carbon substrates, and in particular dicarboxylic acids, whereas M. flocculans and M. subterranea are unable to grow on these substrates. Utilisation of dicarboxylic acids, along with the acquisition of nodulation and nitrogen fixation genes, would appear to be a

209 Chapter 5 prerequisite for potentially rhizobial bacteria (Lodwig & Poole, 2003) and the limited metabolic capacity of some Microvirga strains appears to preclude them from attaining symbiotic capability.

As noted previously, described Microvirga species are able to tolerate relatively high temperatures. Strains related to M. subterranea and to Microvirga sp.

AC72a have been isolated from Colorado Plateau soil crusts and the rhizosphere of the endemic Thar Desert grass Lasiurus sindicus, respectively (Chowdhury et al.,

2009; Gundlapally & Garcia-Pichel, 2006); additionally, a Microvirga strain (TP1) has been isolated from nodules of Tephrosia purpurea growing in the Thar Desert

(H. S. Gehlot, unpublished data). These deserts are subject to aridity, high temperatures and high solar UV flux, indicating that Microvirga species may be well adapted to such environmental stresses. Rhizobial Microvirga strains may therefore have potential as inoculants in marginal agricultural areas.

Interesting questions can also be raised about the host range and biogeography of rhizobial Microvirga strains, especially as compared with the rhizobial Methylobacterium species. Nodulating Microvirga strains have a wider geographical distribution and host range, whereas rhizobial Methylobacterium strains appear to be confined to African crotalarioid legumes (Andam & Parker, 2007; Lafay

& Burdon, 2007; Sy et al., 2001; Wolde-Meskel et al., 2005; Yates et al., 2007). One of the outstanding questions about the evolution of rhizobia is what genetic predisposition is required for a potential bacterial recipient of nod and nif genes to evolve into a rhizobium (Masson-Boivin et al., 2009). To this could be added, what are the circumstances under which horizontal transfer of symbiotic genes and/or

210 Chapter 5 symbiont replacement is favoured? Willems (2006) has suggested that novel rhizobial species are likely to be found amongst genera that have at least some plant- associated species and are therefore probably more able to overcome plant defenses.

In this regard, Methylobacterium is a somewhat puzzling genus, as many strains are epiphytes or endophytes of diverse plant hosts and capable of colonising both the rhizosphere and phyllosphere (Andreote et al., 2009; Madhaiyan et al., 2009a;

2009b; Omer et al., 2004; Van Aken et al., 2004), yet the nodulating methylobacterial taxa so far described are associated almost exclusively with African crotalarioid legume hosts (Jaftha et al., 2002; Sy et al., 2001; Yates et al., 2007).

The genomes of eight Methylobacterium strains, including the Listia bainesii symbiont Methylobacterium sp. 4-46 (Giraud & Fleischman, 2004), M. nodulans

ORS 2060 and strains of four non-symbiotic species have now been sequenced and are available on the Integrated Microbial Genomes (IMG) database of the Joint

Genome Institute (JGI) (http://img.jgi.doe.gov/cgi-bin/w/main.cgi). The rhizobial

Microvirga strains Lut6 and WSM3557 have been entered in the JGI sequencing program and the sequenced genomes of these microbes should soon be available (W.

Reeve, pers. comm. (http://genome.jgi.doe.gov/genome-projects/pages/projects.jsf)).

These sequences will provide platforms for intra- and interspecies genomic comparisons in these genera and may shed further light on the traits required for saprophytic competence in a given environment and the evolution of rhizobial microsymbionts.

211 Chapter 5

212 Bibliography

Bibliography

213 Bibliography

Abi-Ghanem, R., Carpenter-Boggs, L. & Smith, J. (2011). Cultivar effects on nitrogen fixation in peas and lentils. Biology and Fertility of Soils, 47, 115-120.

Acinas, S. G., Marcelino, L. A., Klepac-Ceraj, V. & Polz, M. F. (2004). Divergence and redundancy of 16S rRNA sequences in genomes with multiple rrn operons. Journal of Bacteriology 186, 2629-2635.

Adékambi, T. & Drancourt, M. (2004). Dissection of phylogenetic relationships among 19 rapidly growing Mycobacterium species by 16S rRNA, hsp65, sodA, recA and rpoB gene sequencing. International Journal of Systematic and Evolutionary Microbiology 54, 2095-2105.

Aguilar, O. M., Riva, O. & Peltzer, E. (2004). Analysis of Rhizobium etli and of its symbiosis with wild Phaseolus vulgaris supports coevolution in centers of host diversification. Proceedings of the National Academy of Sciences of the United States of America 101, 13548-13553.

Alazard, D. & Duhoux, E. (1990). Development of stem nodules in a tropical forage legume, Aeschynomene afraspera. Journal of Experimental Botany 41, 1199- 1206.

Alberton, O., Kaschuk, G. & Hungria, M. (2006). Sampling effects on the assessment of genetic diversity of rhizobia associated with soybean and common bean. Soil Biology and Biochemistry 38, 1298-1307.

Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). Basic local alignment search tool. Journal of Molecular Biology 215, 403-410.

Amadou, C., Pascal, G., Mangenot, S., Glew, M., Bontemps, C., Capela, D., Carrere, S., Cruveiller, S., Dossat, C., Lajus, A. & other authors (2008). Genome sequence of the beta-rhizobium Cupriavidus taiwanensis and comparative genomics of rhizobia. Genome Research 18, 1472-1483.

Amann, G., Stetter, K. O., Llobet-Brossa, E., Amann, R. & Antón, J. (2000). Direct proof for the presence and expression of two 5% different 16S rRNA genes in individual cells of Haloarcula marismortui. Extremophiles 4, 373-376.

Andam, C. P., Mondo, S. J. & Parker, M. A. (2007). Monophyly of nodA and nifH genes across Texas and Costa Rican populations of Cupriavidus nodule symbionts. Applied and Environmental Microbiology 73, 4686–4690.

Andam, C. P. & Parker, M. A. (2007). Novel Alphaproteobacterial root nodule symbiont associated with Lupinus texensis. Applied and Environmental Microbiology 73, 5687-5691.

Andreote, F., Carneiro, R., Salles, J., Marcon, J., Labate, C., Azevedo, J. & Araújo, W. (2009). Culture-independent assessment of rhizobiales-related Alphaproteobacteria and the diversity of Methylobacterium in the rhizosphere and rhizoplane of transgenic eucalyptus. Microbial Ecology 57, 82-93.

214 Bibliography

Andronov, E. E., Terefework, Z., Roumiantseva, M. L., Dzyubenko, N. I., Onichtchouk, O. P., Kurchak, O. N., Dresler-Nurmi, A., Young, J. P. W., Simarov, B. V. & Lindström, K. (2003). Symbiotic and genetic diversity of Rhizobium galegae isolates collected from the Galega orientalis gene center in the Caucasus. Applied and Environmental Microbiology 69, 1067-1074.

Anon. (2006). Innovative Science Statement. Perth: Future Farm Industries CRC The University of Western Australia.

Ardley, J., O'Hara, G., Reeve, W., Yates, R., Dilworth, M., Tiwari, R. & Howieson, J. (2009). Root nodule bacteria isolated from South African Lotononis bainesii, L. listii and L. solitudinis are species of Methylobacterium that are unable to utilize methanol. Archives of Microbiology 191, 311-318.

Ardley, J. K. (2005). Characterisation of root nodule bacteria isolated from indigenous South African herbaceous legumes: Honours thesis, Murdoch University, Perth, Western Australia.

Ba, S., Willems, A., De Lajudie, P., Roche, P., Jeder, H., Quatrini, P., Neyra, M., Ferro, M., Promé, J. C., Gillis, M., Boivin-Masson, C. & Lorquin, J. (2002). Symbiotic and taxonomic diversity of rhizobia isolated from Acacia tortilis subsp raddiana in Africa. Systematic and Applied Microbiology 25, 130-145.

Bala, A. & Giller, K. E. (2006). Relationships between rhizobial diversity and host legume nodulation and nitrogen fixation in tropical ecosystems. Nutrient Cycling in Agroecosystems 76, 319-330.

Barcellos, F. G., Menna, P., da Silva Batista, J. S. & Hungria, M. (2007). Evidence of horizontal transfer of symbiotic genes from a Bradyrhizobium japonicum inoculant strain to indigenous diazotrophs Sinorhizobium (Ensifer) fredii and Bradyrhizobium elkanii in a Brazilian savannah soil. Applied and Environmental Microbiology 73, 2635-2643.

Becking, J.-H. (1992). The Rhizobium symbiosis of the nonlegume Parasponia. In Biological Nitrogen Fixation, pp. 497-559. Edited by G. S. Stacey, R. H. Burris & H. J. Evans. New York, London: Chapman & Hall.

Ben Romdhane, S., Tajini, F., Trabelsi, M., Aouani, M. E. & Mhamdi, R. (2007). Competition for nodule formation between introduced strains of Mesorhizobium ciceri and the native populations of rhizobia nodulating chickpea (Cicer arietinum) in Tunisia. World Journal of Microbiology & Biotechnology 23, 1195-1201.

Berge, O., Lodhi, A., Brandelet, G., Santaella, C., Roncato, M. A., Christen, R., Heulin, T. & Achouak, W. (2009). Rhizobium alamii sp nov., an exopolysaccharide-producing species isolated from legume and non-legume rhizospheres. International Journal of Systematic and Evolutionary Microbiology 59, 367-372.

Beringer, J. E. (1974). R factor transfer in Rhizobium leguminosarum. Journal of General Microbiology 84, 188-198.

215 Bibliography

Bethesda Research Laboratories (1986). BRL pUC host: E. coli DH5α competent cells. Focus 8 (2), 9.

Beveridge, T. J., Lawrence, J. R. & Murray, R. G. E. (2007). Sampling and staining for light microscopy. In Methods for General and Molecular Microbiology, pp. 19-33. Edited by C. A. Reddy, T. J. Beveridge, J. A. Breznak, T. M. Marzluf, T. M. Schmidt & L. R. Snyder. Washington, D.C.: American Society for Microbiology Press.

Bhuvaneswari, T. V., Turgeon, B. G. & Bauer, W. D. (1980). Early events in the infection of soybean (Glycine max L. Merr) by Rhizobium japonicum: I. Localization of infectible root cells. Plant Physiology 66, 1027-1031.

Bhuvaneswari, T. V., Bhagwat, A. A. & Bauer, W. D. (1981). Transient susceptibility of root cells in four common legumes to nodulation by rhizobia. Plant Physiology 68, 1144-1149.

Biondi, E. G., Tatti, E., Comparini, D., Giuntini, E., Mocali, S., Giovannetti, L., Bazzicalupo, M., Mengoni, A. & Viti, C. (2009). Metabolic capacity of Sinorhizobium (Ensifer) meliloti strains as determined by phenotype MicroArray analysis. Applied and Environmental Microbiology 75, 5396-5404.

Boatwright, J. S., le Roux, M. M., Wink, M., Morozova, T. & van Wyk, B. E. (2008). Phylogenetic relationships of tribe Crotalarieae (Fabaceae) inferred from DNA sequences and morphology. Systematic Botany 33, 752-761.

Boatwright, J. S., Wink, M. & van Wyk, B.-E. (2011). The generic concept of Lotononis (Crotalarieae, Fabaceae): Reinstatement of the genera Euchlora, Leobordea and Listia and the new genus Ezoloba. Taxon 60, 161-177.

Bochner, B. R. (1989). Sleuthing out bacterial identities Nature 339, 157-158.

Bochner, B. R., Gadzinski, P. & Panomitros, E. (2001). Phenotype microarrays for high-throughput phenotypic testing and assay of gene function. Genome Research 11, 1246-1255.

Bochner, B. R. (2003). New technologies to assess genotype-phenotype relationships. Nature Reviews Genetics 4, 309-314.

Bochner, B. R. (2009). Global phenotypic characterization of bacteria. FEMS Microbiology Reviews 33, 191-205.

Boivin, C., Ndoye, I., Molouba, F., de Lajudie, P., Dupuy, N. & Dreyfus, B. (1997). Stem nodulation in legumes : Diversity, mechanisms, and unusual characteristics. Critical Reviews in Plant Sciences 16, 1-30.

Bontemps, C., Elliott, G. N., Simon, M. F., dos Reis Júnior, F. B., Gross, E., Lawton, R. C., Neto, N. E., Loureiro, M. d. F., De Faria, S. M., Sprent, J. I., James, E. K. & Young, J. P. W. (2010). Burkholderia species are ancient symbionts of legumes. Molecular Ecology 19, 44-52.

216 Bibliography

Boogerd, F. C. & van Rossum, D. (1997). Nodulation of groundnut by Bradyrhizobium: a simple infection process by crack entry. FEMS Microbiology Reviews 21, 5-27.

Boone, C. M., Olsthoorn, M. M. A., Dakora, F. D., Spaink, H. P. & Thomas- Oates, J. E. (1999). Structural characterisation of lipo-chitin oligosaccharides isolated from Bradyrhizobium aspalati, microsymbionts of commercially important South African legumes. Carbohydrate Research 317, 155-163.

Boucher, Y., Douady, C. J., Papke, R. T., Walsh, D. A., Boudreau, M. E. R., Nesbø, C. L., Case, R. J. & Doolittle, W. F. (2003). Lateral gene transfer and the origins of prokaryotic groups. Annual Review of Genetics 37, 283-328.

Boulila, F., Depret, G., Boulila, A., Belhadi, D., Benallaoua, S. & Laguerre, G. (2009). Retama species growing in different ecological-climatic areas of northeastern Algeria have a narrow range of rhizobia that form a novel phylogenetic clade within the Bradyrhizobium genus. Systematic and Applied Microbiology 32, 245-255.

Bowra, B. J. & Dilworth, M. J. (1981). Motility and chemotaxis towards sugars in Rhizobium leguminosarum. Journal of General Microbiology 126, 231-235.

Brewin, N. J. (2004). Plant cell wall remodelling in the rhizobium-legume symbiosis. Critical Reviews in Plant Sciences 23, 293-316.

Britton, G., Liaaen-Jensen, S. & Pfander, H. (2004). Carotenoids handbook. Compiled by A.Z. Mercadante and E.S. Egeland. Basel; Boston; Berlin. Birkhäuser Verlag.

Bromfield, E. S. P., Tambong, J. T., Cloutier, S., Prevost, D., Laguerre, G., van Berkum, P., Thi, T. V. T., Assabgui, R. & Barran, L. R. (2010). Ensifer, Phyllobacterium and Rhizobium species occupy nodules of Medicago sativa (alfalfa) and Melilotus alba (sweet clover) grown at a Canadian site without a history of cultivation. Microbiology-Sgm 156, 505-520.

Broughton, W. J., Wong, C. H., Lewin, A., Samrey, U., Myint, H., Meyer, H., Dowling, D. N. & Simon, R. (1986). Identification of Rhizobium plasmid sequences involved in recognition of Psophocarpus, Vigna, and other legumes. The Journal of Cell Biology 102, 1173-1182.

Broughton, W. J. & Perret, X. (1999). Genealogy of legume-Rhizobium symbioses. Current Opinion in Plant Biology 2, 305-311.

Broughton, W. J., Hernández, G., Blair, M., Beebe, S., Gepts, P. & Vanderleyden, J. (2003). Beans (Phaseolus spp.) - model food legumes. Plant and Soil 252, 55-128.

Bryan, W. W. (1961). Lotononis bainesii Baker - a legume for subtropical pastures. Australian Journal of Experimental Agriculture and Animal Husbandry 1, 4-10.

217 Bibliography

Bullard, G. K., Roughley, R. J. & Pulsford, D. J. (2005). The legume inoculant industry and inoculant quality control in Australia: 1953-2003. Australian Journal of Experimental Agriculture 45, 127-140.

Burdon, J. J., Gibson, A. H., Searle, S. D., Woods, M. J. & Brockwell, J. (1999). Variation in the effectiveness of symbiotic associations between native rhizobia and temperate Australian Acacia: within-species interactions. Journal of Applied Ecology 36, 398-408.

Caetano-Anollés, G., Wall, L. G., De Micheli, A. T., Macchi, E. M., Bauer, W. D. & Favelukes, G. (1988). Role of motility and chemotaxis in efficiency of nodulation by Rhizobium meliloti. Plant Physiology 86, 1228-1235.

Capoen, W., Oldroyd, G., Goormachtig, S. & Holsters, M. (2010). Sesbania rostrata: a case study of natural variation in legume nodulation. New Phytologist 186, 340-345.

Chandler, M. R., Date, R. A. & Roughley, R. J. (1982). Infection and root-nodule development in Stylosanthes species by Rhizobium. Journal of Experimental Botany 33, 47-57.

Charpentier, M. & Oldroyd, G. (2010). How close are we to nitrogen-fixing cereals? Current Opinion in Plant Biology 13, 556-564.

Chen, W. X., Wang, E., Wang, S. Y., Li, Y. B., Chen, X. Q. & Li, Y. B. (1995). Characteristics of Rhizobium tianshanense sp. nov., a moderately and slowly growing root-nodule bacterium isolated from an arid saline environment in Xinjiang, People’s Republic of China. International Journal of Systematic Bacteriology 45, 153-159.

Chen, W.-M., Moulin, L., Bontemps, C., Vandamme, P., Béna, G. & Boivin- Masson, C. (2003). Legume symbiotic nitrogen fixation by β-Proteobacteria is widespread in nature. Journal of Bacteriology 185, 7266-7272.

Chen, W. M., de Faria, S. M., James, E. K., Elliott, G. N., Lin, K. Y., Chou, J. H., Sheu, S. Y., Cnockaert, M., Sprent, J. I. & Vandamme, P. (2007). Burkholderia nodosa sp nov., isolated from root nodules of the woody Brazilian legumes Mimosa bimucronata and Mimosa scabrella. International Journal of Systematic and Evolutionary Microbiology 57, 1055-1059.

Chen, W.-M., de Faria, S. M., Chou, J.-H., James, E. K., Elliott, G. N., Sprent, J. I., Bontemps, C., Young, J. P. W. & Vandamme, P. (2008). Burkholderia sabiae sp. nov., isolated from root nodules of Mimosa caesalpiniifolia. International Journal of Systematic and Evolutionary Microbiology 58, 2174-2179.

Chowdhury, S. P., Schmid, M., Hartmann, A. & Tripathi, A. K. (2009). Diversity of 16S-rRNA and nifH genes derived from rhizosphere soil and roots of an endemic drought tolerant grass, Lasiurus sindicus. European Journal of Soil Biology 45, 114-122.

218 Bibliography

Cocks, P. S. (2001). Ecology of herbaceous perennial legumes: a review of characteristics that may provide management options for the control of salinity and waterlogging in dryland cropping systems. Australian Journal of Agricultural Research 52, 137-151.

Cocks, P. S. (2003). Land-use change is the key to protecting biodiversity in salinising landscapes. Australian Journal of Botany 51, 627-635.

Cohan, F. M. (2002). What are bacterial species? Annual Review of Microbiology 56, 457-487.

Corby, H. D. L. (1981). The systematic value of leguminous root nodules. In Advances in legume systematics, Part 2, pp. 657-669. Edited by R. M. Polhill & P. H. Raven. Kew, U.K: Royal Botanic Gardens.

Cordeiro, L., Sprent, J. I. & McInroy, S. G. (1996). Some developmental and structural aspects of nodules of Lonchocarpus muehlbergianus. Naturalia, São Paulo 21, 9-21.

Cowling, R. M., Proches, S. & Partridge, T. C. (2009). Explaining the uniqueness of the Cape flora: Incorporating geomorphic evolution as a factor for explaining its diversification. Molecular Phylogenetics and Evolution 51, 64-74.

Cronk, Q., Ojeda, I. & Pennington, R. T. (2006). Legume comparative genomics: progress in phylogenetics and phylogenomics. Current Opinion in Plant Biology 9, 99-103.

Cummings, S. P., Gyaneshwar, P., Vinuesa, P., Farruggia, F. T., Andrews, M., Humphry, D., Elliott, G. N., Nelson, A., Orr, C., Pettitt, D. & other authors (2009). Nodulation of Sesbania species by Rhizobium (Agrobacterium) strain IRBG74 and other rhizobia. Environmental Microbiology 11, 2510-2525.

D’Haeze, W. & Holsters, M. (2002). Nod factor structures, responses, and perception during initiation of nodule development. Glycobiology 12, 79R-105R.

D'Haeze, W., De Rycke, R., Mathis, R., Goormachtig, S., Pagnotta, S., Verplancke, C., Capoen, W. & Holsters, M. (2003). Reactive oxygen species and ethylene play a positive role in lateral root base nodulation of a semiaquatic legume. Proceedings of the National Academy of Sciences of the United States of America 100, 11789-11794.

Daubin, V., Moran, N. A. & Ochman, H. (2003). Phylogenetics and the cohesion of bacterial genomes. Science 301, 829-832. de Bruijn, F. J. (1992). Use of repetitive (repetitive extragenic palindromic and enterobacterial repetitive intergeneric consensus) sequences and the polymerase chain reaction to fingerprint the genomes of Rhizobium meliloti isolates and other soil bacteria. Applied and Environmental Microbiology 58, 2180-2187.

219 Bibliography de Lajudie, P., Willems, A., Pot, B., Dewettinck, D., Maestrojuan, G., Neyra, M., Collins, M. D., Dreyfus, B., Kersters, K. & Gillis, M. (1994). Polyphasic taxonomy of rhizobia: emendation of the genus Sinorhizobium and description of Sinorhizobium meliloti comb. nov., Sinorhizobium saheli sp. nov., and Sinorhizobium teranga sp. nov. International Journal of Systematic Bacteriology 44, 715-733. de Lajudie, P., Willems, A., Nick, G., Moreira, F., Molouba, F., Hoste, B., Torck, U., Neyra, M., Collins, M. D., Lindström, K., Dreyfus, B. & Gillis, M. (1998a). Characterization of tropical tree rhizobia and description of Mesorhizobium plurifarium sp. nov. International Journal of Systematic Bacteriology 48, 369-382. de Lajudie, P., Laurent-Fulele, E., Willems, A., Torek, U., Coopman, R., Collins, M. D., Kersters, K., Dreyfus, B. & Gillis, M. (1998b). Allorhizobium undicola gen. nov., sp. nov., nitrogen-fixing bacteria that efficiently nodulate Neptunia natans in Senegal. International Journal of Systematic Bacteriology 48, 1277-1290.

Dear, B. S., Moore, G. A. & Hughes, S. J. (2003). Adaptation and potential contribution of temperate perennial legumes to the southern Australian wheatbelt: a review. Australian Journal of Experimental Agriculture 43, 1-18.

Dear, B. S. & Ewing, M. A. (2008). The search for new pasture plants to achieve more sustainable production systems in southern Australia. Australian Journal of Experimental Agriculture 48, 387-396.

Debellé, F., Plazanet, C., Roche, P., Pujol, C., Savagnac, A., Rosenberg, C., Promé, J.-C. & Dénarié, J. (1996). The NodA proteins of Rhizobium meliloti and Rhizobium tropici specify the N-acylation of Nod factors by different fatty acids. Molecular Microbiology 22, 303-314.

Debellé, F., Moulin, L., Mangin, B., Dénarié, J. & Boivin, C. (2001). nod genes and Nod signals and the evolution of the Rhizobium legume symbiosis. Acta Biochimica Polonica 48, 359-365.

Degefu, T., Wolde-Meskel, E. & Frostegård, Å. (2011). Multilocus sequence analyses reveal several unnamed Mesorhizobium genospecies nodulating Acacia species and Sesbania sesban trees in Southern regions of Ethiopia. Systematic and Applied Microbiology 34, 216-226.

Delmer, D. P. & Amor, Y. (1995). Cellulose biosynthesis. Plant Cell 7, 987-1000.

Dénarié, J., Debellé, F. & Promé, J.-C. (1996). Rhizobium lipo- chitooligosaccharide nodulation factors: Signaling molecules mediating recognition and morphogenesis. Annual Review of Biochemistry 65, 503-535.

Diatloff, A. (1977). Ecological studies of root-nodule bacteria introduced into field environments-6. Antigenic and symbiotic stability in Lotononis rhizobia over a 12- year period. Soil Biology and Biochemistry 9, 85-88.

220 Bibliography

Dilworth, M. J., Howieson, J. G., Reeve, W. G., Tiwari, R. P. & Glenn, A. R. (2001). Acid tolerance in legume root nodule bacteria and selecting for it. Australian Journal of Experimental Agriculture 41, 435-446.

Diouf, D., Samba-Mbaye, R., Lesueur, D., Ba, A., Dreyfus, B., de Lajudie, P. & Neyra, M. (2007). Genetic diversity of Acacia seyal Del. rhizobial populations indigenous to Senegalese soils in relation to salinity and pH of the sampling sites Microbial Ecology 54, 553-566.

Dobert, R. C., Breil, B. T. & Triplett, E. W. (1994). DNA sequence of the common nodulation genes of Bradyrhizobium elkanii and their phylogenetic relationship to those of other nodulating bacteria. Molecular Plant-Microbe Interactions 7, 564-572.

Donate-Correa, J., León-Barrios, M., Hernández, M., Pérez-Galdona, R. & del Arco-Aguilar, M. (2007). Different Mesorhizobium species sharing the same symbiotic genes nodulate the shrub legume Anagyris latifolia. Systematic and Applied Microbiology 30, 615-623.

Douglas, A. E. (1998). Host benefit and the evolution of specialization in symbiosis. Heredity 81, 599-603.

Doyle, J. J. & Luckow, M. A. (2003). The rest of the iceberg. Legume diversity and evolution in a phylogenetic context. Plant Physiology 131, 900-910.

Drew, E. A. & Ballard, R. A. (2010). Improving N2 fixation from the plant down: Compatibility of Trifolium subterraneum L. cultivars with soil rhizobia can influence symbiotic performance. Plant and Soil 327, 261-277.

Dreyfus, B., Garcia, J. L. & Gillis, M. (1988). Characterization of Azorhizobium caulinodans gen. nov., sp. nov., a stem-nodulating nitrogen-fixing bacterium isolated from Sesbania rostrata. International Journal of Systematic Bacteriology 38, 89-98.

Eagles, D. A. & Date, R. A. (1999). The CB Rhizobium/Bradyrhizobium Strain Collection. Genetic Resources Communication No. 30. St Lucia, Queensland, Australia: CSIRO Tropical Agriculture.

Edwards, D. & Hawkins, J. A. (2007). Are Cape floral clades the same age? Contemporaneous origins of two lineages in the genistoids s.l. (Fabaceae). Molecular Phylogenetics and Evolution 45, 952-970.

Egli, T. W. & Auling, G. (2005). Genus II. Chelatococcus In Bergey's Manual of Systematic Bacteriology pp. 433-437. Edited by D. J. Brenner, N. R. Krieg & J. T. Staley. New York: Springer.

Elliott, G. N., Chen, W. M., Bontemps, C., Chou, J. H., Young, J. P. W., Sprent, J. I. & James, E. K. (2007). Nodulation of Cyclopia spp. (Leguminosae, Papilionoideae) by Burkholderia tuberum. Annals of Botany 100, 1403-1411.

221 Bibliography

Evans, H. J. & Barber, L. E. (1977). Biological nitrogen fixation for food and fiber production. Science 197, 332-339.

Evans, W. R., Fleischman, D. E., Calvert, H. E., Pyati, P. V., Alter, G. M. & Rao, N. S. S. (1990). Bacteriochlorophyll and photosynthetic reaction centers in Rhizobium strain BTAi 1. Applied and Environmental Microbiology 56, 3445-3449.

Fauvart, M., Verstraeten, N., Dombrecht, B., Venmans, R., Beullens, S., Heusdens, C. & Michiels, J. (2009). Rhizobium etli HrpW is a pectin-degrading enzyme and differs from phytopathogenic homologues in enzymically crucial tryptophan and glycine residues. Microbiology-Sgm 155, 3045-3054.

Fedorova, E. E., de Felipe, M. R., Pueyo, J. J. & Lucas, M. M. (2007). Conformation of cytoskeletal elements during the division of infected Lupinus albus L. nodule cells. Journal of Experimental Botany 58, 2225-2236.

Fenchel, T. & Finlay, B. J. (2006). The diversity of microbes: resurgence of the phenotype. Philosophical Transactions of the Royal Society B: Biological Sciences 361, 1965-1973.

Feng, J., Li, Q., Hu, H. L., Chen, X. C. & Hong, G. F. (2003). Inactivation of the nod box distal half-site allows tetrameric NodD to activate nodA transcription in an inducer-independent manner. Nucleic Acids Research 31, 3143-3156.

Fernández-Pascual, M., Pueyo, J. J., de Felipe, M. R., Golvano, M. P. & Lucas, M. M. (2007). Singular features of the Bradyrhizobium–Lupinus symbiosis. Dynamic Soil, Dynamic Plant 1, 1–16.

Fettell, N. A., O'Connor, G. E., Carpenter, D. J., Evans, J., Bamforth, I., Oti- Boateng, C., Hebb, D. M. & Brockwell, J. (1997). Nodulation studies on legumes exotic to Australia: The influence of soil populations and inocula of Rhizobium leguminosarum bv. vicieae on nodulation and nitrogen fixation by field peas. Applied Soil Ecology 5, 197-210.

Fierer, N. & Jackson, R. B. (2006). The diversity and biogeography of soil bacterial communities. Proceedings of the National Academy of Sciences of the United States of America 103, 626-631.

Fleischman, D. & Kramer, D. (1998). Photosynthetic rhizobia. Biochimica Et Biophysica Acta-Bioenergetics 1364, 17-36.

Foster, C. M., Horner, H. T. & Graves, W. R. (1998). Nodulation response of woody Papilionoid species after inoculation with rhizobia and soil from Hawaii, Asia and North America. Plant and Soil 205, 103-111.

Fraser, C., Alm, E. J., Polz, M. F., Spratt, B. G. & Hanage, W. P. (2009). The bacterial species challenge: making sense of genetic and ecological diversity. Science 323, 741-746.

222 Bibliography

Fred, E. B., Baldwin, I. L. & McCoy, E. (1932). Root nodule bacteria and leguminous plants. University of Wisconsin Studies in Sciences: Number 5 Madison: University of Wisconsin Press.

Fterich, A., Mahdhi, M., Caviedes, M., Pajuelo, E., Rivas, R., Rodriguez- Llorente, I. & Mars, M. (2011). Characterization of root-nodulating bacteria associated to Prosopis farcta growing in the arid regions of Tunisia. Archives of Microbiology 193, 385-397

Freiberg, C., Fellay, R., Bairoch, A., Broughton, W. J., Rosenthal, A. & Perret, X. (1997). Molecular basis of symbiosis between Rhizobium and legumes. Nature 387, 394-401.

Gage, D. J. (2004). Infection and invasion of roots by symbiotic, nitrogen-fixing rhizobia during nodulation of temperate legumes. Microbiology and Molecular Biology Reviews 68, 280-300.

Gallego, V., García, M. T. & Ventosa, A. (2005). Methylobacterium isbiliense sp nov., isolated from the drinking water system of Sevilla, Spain. International Journal of Systematic and Evolutionary Microbiology 55, 2333-2337.

Gallego, V., García, M. T. & Ventosa, A. (2006). Methylobacterium adhaesivum sp. nov., a methylotrophic bacterium isolated from drinking water. International Journal of Systematic and Evolutionary Microbiology 56, 339-342.

Garau, G., Reeve, W. G., Bräu, L., Deiana, P., Yates, R. J., James, D., Tiwari, R., O'Hara, G. W. & Howieson, J. G. (2005). The symbiotic requirements of different Medicago spp. suggest the evolution of Sinorhizobium meliloti and S. medicae with hosts differentially adapted to soil pH. Plant and Soil 276, 263-277.

Garau, G., Yates, R. J., Deiana, P. & Howieson, J. G. (2009). Novel strains of nodulating Burkholderia have a role in nitrogen fixation with papilionoid herbaceous legumes adapted to acid, infertile soils. Soil Biology & Biochemistry 41, 125-134.

Garland, J. L. & Mills, A. L. (1991). Classification and characterization of heterotrophic microbial communities on the basis of patterns of community-level sole-carbon-source utilization. Applied and Environmental Microbiology 57, 2351- 2359.

Garrity, G. M., Brenner, D. J., Krieg, N. R. & Staley, J. T. (2005). Bergey's Manual Of Systematic Bacteriology, Volume 2: The Proteobacteria. New York: Springer.

Gaunt, M. W., Turner, S. L., Rigottier-Gois, L., Lloyd-Macgilp, S. A. & Young, J. P. (2001). Phylogenies of atpD and recA support the small subunit rRNA-based classification of rhizobia. International Journal of Systematic and Evolutionary Microbiology 51, 2037-2048.

George, R., McFarlane, D. & Nulsen, B. (1997). Salinity threatens the viability of agriculture and ecosystems in Western Australia. Hydrogeology Journal 5 6-21.

223 Bibliography

Gevers, D., Cohan, F. M., Lawrence, J. G., Spratt, B. G., Coenye, T., Feil, E. J., Stackebrandt, E., Van de Peer, Y., Vandamme, P., Thompson, F. L. & Swings, J. (2005). Re-evaluating prokaryotic species. Nature Reviews Microbiology 3, 733- 739.

Gibson, K. E., Kobayashi, H. & Walker, G. C. (2008). Molecular determinants of a symbiotic chronic infection. Annual Review of Genetics 42, 413-441.

Giraud, E. & Fleischman, D. (2004). Nitrogen-fixing symbiosis between photosynthetic bacteria and legumes. Photosynthesis Research 82, 115-130.

Giraud, E., Hannibal, L., Fardoux, L., Verméglio, A. & Dreyfus, B. (2000). Effect of Bradyrhizobium photosynthesis on stem nodulation of Aeschynomene sensitiva. Proceedings of the National Academy of Sciences of the United States of America 97, 14795-14800.

Giraud, E., Hannibal, L., Fardoux, J., Jaubert, M., Jourand, P., Dreyfus, B., Sturgis, J. N. & Vermeglio, A. (2004). Two distinct crt gene clusters for two different functional classes of carotenoid in Bradyrhizobium. Journal of Biological Chemistry 279, 15076-15083.

Giraud, E., Moulin, L., Vallenet, D., Barbe, V., Cytryn, E., Avarre, J. C., Jaubert, M., Simon, D., Cartieaux, F., Prin, Y. & other authors (2007). Legumes symbioses: Absence of nod genes in photosynthetic bradyrhizobia. Science 316, 1307-1312.

Godfrey, C. A. (1972). The carotenoid pigment and deoxyribonucleic acid base ratio of a rhizobium which nodulates Lotononis bainesii Baker. Journal of General Microbiology 72, 399-402.

Gogarten, J. P., Doolittle, W. F. & Lawrence, J. G. (2002). Prokaryotic evolution in light of gene transfer. Molecular Biology and Evolution 19, 2226-2238.

González, V., Bustos, P., Ramírez-Romero, M. A., Medrano-Soto, A., Salgado, H., Hernández-González, I., Hernández-Celis, J. C., Quintero, V., Moreno- Hagelsieb, G., Girard, L. & other authors (2003). The mosaic structure of the symbiotic plasmid of Rhizobium etli CFN42 and its relation to other symbiotic genome compartments. Genome Biology 4.

González-Sama, A., Lucas, M. M., de Felipe, M. R. & Pueyo, J. J. (2004). An unusual infection mechanism and nodule morphogenesis in white lupin (Lupinus albus). New Phytologist 163, 371-380.

González-Sama, A., de la Peña, T. C., Kevei, Z., Mergaert, P., Lucas, M. M., de Felipe, M. R., Kondorosi, E. & Pueyo, J. J. (2006). Nuclear DNA endoreduplication and expression of the mitotic inhibitor Ccs52 associated to determinate and lupinoid nodule organogenesis. Molecular Plant-Microbe Interactions 19, 173-180.

224 Bibliography

Goormachtig, S., Capoen, W., James, E. K. & Holsters, M. (2004). Switch from intracellular to intercellular invasion during water stress-tolerant legume nodulation. Proceedings of the National Academy of Sciences of the United States of America 101, 6303-6308.

Gough, C. & Cullimore, J. (2011). Lipo-chitooligosaccharide signalling in endosymbiotic plant-microbe interactions. Molecular Plant-Microbe Interactions 24, 867-878.

Graham, P. H. (1963). Vitamin requirements of root nodule bacteria. Journal of General Microbiology 30, 245-248.

Graham, P. H. & Parker, C. A. (1964). Diagnostic features in the characterisation of the root-nodule bacteria of legumes. Plant and Soil 20, 383-396.

Graham, P. H., Sadowsky, M. J., Keyser, H. H., Barnet, Y. M., Bradley, R. S., Cooper, J. E., De Ley, D. J., Jarvis, B. D. W., Roslycky, E. B., Strijdom, B. W. & Young, J. P. W. (1991). Proposed minimal standards for the description of new genera and species of root- and stem-nodulating bacteria. International Journal of Systematic Bacteriology 41, 582-587.

Graham, P. H. & Vance, C. P. (2003). Legumes: importance and constraints to greater use. Plant Physiology 131, 872-877.

Graham, P. H. (2008). Ecology of the root-nodule bacteria of legumes. In Nitrogen- fixing Leguminous Symbioses, pp. 23-58. Edited by M. J. Dilworth, E. K. James, J. I. Sprent & W. E. Newton. Dordrecht, The Netherlands: Springer.

Green, P. N. (1992). The genus Methylobacterium. In The Prokaryotes: a Handbook on the Biology of Bacteria: Ecophysiology, Isolation, Identification, Applications, pp. 2342-2349. Edited by A. Balows, H. G. Trüper, M. Dworkin, W. Harder & K.-H. Schliefer. New York: Springer-Verlag.

Gualtieri, G. & Bisseling, T. (2000). The evolution of nodulation. Plant Molecular Biology 42, 181-194.

Guinel, F. C. (2009). Getting around the legume nodule: I. The structure of the peripheral zone in four nodule types. Botany 87, 1117-1138.

Gundlapally, S. R. & Garcia-Pichel, F. (2006). The community and phylogenetic diversity of biological soil crusts in the Colorado Plateau studied by molecular fingerprinting and intensive cultivation. Microbial Ecology 52, 345-357.

Guzman, L. M., Belin, D., Carson, M. J. & Beckwith, J. (1995). Tight regulation, modulation, and high-level expression by vectors containing the arabinose pBAD promotor. Journal of Bacteriology 177, 4121-4130.

Han, L. L., Wang, E. T., Han, T. X., Liu, J., Sui, X. H., Chen, W. F. & Chen, W. X. (2009). Unique community structure and biogeography of soybean rhizobia in the saline-alkaline soils of Xinjiang, China. Plant and Soil 324, 291-305.

225 Bibliography

Han, T. X., Tian, C. F., Wang, E. T. & Chen, W. X. (2010). Associations among rhizobial chromosomal background, nod genes, and host plants based on the analysis of symbiosis of indigenous rhizobia and wild legumes native to Xinjiang. Microbial Ecology 59, 311-323.

Harwood, C. S. & Parales, R. E. (1996). The beta-ketoadipate pathway and the biology of self-identity. Annual Review of Microbiology 50, 553-590.

Hatton, T. J., Ruprecht, J. & George, R. J. (2003). Preclearing hydrology of the Western Australia wheatbelt: Target for the future? Plant and Soil 257, 341-356.

Haukka, K., Lindström, K. & Young, J. P. W. (1998). Three phylogenetic groups of nodA and nifH genes in Sinorhizobium and Mesorhizobium isolates from leguminous trees growing in Africa and Latin America. Applied and Environmental Microbiology 64, 419-426.

Heidstra, R., Geurts, R., Franssen, H., Spaink, H. P., van Kammen, A. & Bisseling, T. (1994). Root hair deformation activity of nodulation factors and their fate on Vicia sativa. Plant Physiology 105, 787-797.

Herre, E. A., Knowlton, N., Mueller, U. G. & Rehner, S. A. (1999). The evolution of mutualisms: exploring the paths between conflict and cooperation. Trends in Ecology & Evolution 14, 49-53.

Herridge, D. F. (2008). Inoculation technology for legumes. In Nitrogen-fixing Leguminous Symbioses, pp. 77-115. Edited by M. J. Dilworth, E. K. James, J. I. Sprent & W. E. Newton. Dordrecht, The Netherlands: Springer.

Herridge, D. F., Peoples, M. B. & Boddey, R. M. (2008). Global inputs of biological nitrogen fixation in agricultural systems. Plant and Soil 311, 1-18.

Hiraishi, A., Furuhata, K., Matsumoto, A., Koike, K. A., Fukuyama, M. & Tabuchi, K. (1995). Phenotypic and genetic diversity of chlorine-resistant Methylobacterium strains isolated from various environments. Applied and Environmental Microbiology 61, 2099-2107.

Hiraishi, A. & Shimada, K. (2001). Aerobic anoxygenic photosynthetic bacteria with zinc-bacteriochlorophyll. Journal of General and Applied Microbiology 47, 161-180.

Hirsch, A. M., Lum, M. R. & Downie, J. A. (2001). What makes the rhizobia- legume symbiosis so special? Plant Physiology 127, 1484-1492.

Hirsch, P. R. (1996). Population dynamics of indigenous and genetically modified rhizobia in the field. New Phytologist 133, 159-171.

Horner-Devine, M. C., Carney, K. M. & Bohannan, B. J. M. (2004). An ecological perspective on bacterial biodiversity. Proceedings of the Royal Society of London Series B-Biological Sciences 271, 113-122.

226 Bibliography

Howieson, J. (2006). CRS Annual Report, 2005/2006. Perth: Centre for Rhizobium Studies, Murdoch University.

Howieson, J. & Ballard, R. (2004). Optimising the legume symbiosis in stressful and competitive environments within southern Australia - some contemporary thoughts. Soil Biology & Biochemistry 36, 1261-1273.

Howieson, J. G. & Ewing, M. A. (1986). Acid tolerance in the Rhizobium meliloti - Medicago symbiosis. Australian Journal of Agricultural Research 37, 55-64.

Howieson, J. G., Loi, A. & Carr, S. J. (1995). Biserrula pelecinus L. - a legume pasture species with potential for acid, duplex soils which is nodulated by unique root-nodule bacteria. Australian Journal of Agricultural Research 46, 997-1009.

Howieson, J. G., O'Hara, G. W. & Carr, S. J. (2000). Changing roles for legumes in Mediterranean agriculture: developments from an Australian perspective. Field Crops Research 65, 107-122.

Howieson, J. G., Yates, R. J., O'Hara, G. W., Ryder, M. & Real, D. (2005). The interactions of Rhizobium leguminosarum biovar trifolii in nodulation of annual and perennial Trifolium spp. from diverse centres of origin. Australian Journal of Experimental Agriculture 45, 199-207.

Howieson, J. G., Yates, R. J., Foster, K. J., Real, D. & Besier, R. B. (2008). Prospects for the future use of legumes. In Nitrogen-fixing Leguminous Symbioses, pp. 363-393. Edited by M. J. Dilworth, E. K. James, J. I. Sprent & W. E. Newton. Dordrecht, The Netherlands: Springer.

Hugh, R. & Leifson, E. (1953). The taxonomic significance of fermentative versus oxidative metabolism of carbohydrates by various Gram negative bacteria. Journal of Bacteriology 66, 24-26.

Ibáñez, F. & Fabra, A. (2011). Rhizobial Nod factors are required for cortical cell division in the nodule morphogenetic programme of the Aeschynomeneae legume Arachis. Plant Biology, 13, 794-800.

Iglesias, O., Rivas, R., Garcia-Fraile, P., Abril, A., Mateos, P. F., Martínez- Molina, E. & Velazquez, E. (2007). Genetic characterization of fast-growing rhizobia able to nodulate Prosopis alba in North Spain. FEMS Microbiology Letters 277, 210-216.

Jacobs, J. L., Carroll, T. L. & Sundin, G. W. (2005). The role of pigmentation, ultraviolet radiation tolerance, and leaf colonization strategies in the epiphytic survival of phyllosphere bacteria. Microbial Ecology 49, 104-113.

Jaftha, J. B., Strijdom, B. W. & Steyn, P. L. (2002). Characterization of pigmented methylotrophic bacteria which nodulate Lotononis bainesii. Systematic and Applied Microbiology 25, 440-449.

227 Bibliography

James, E. K., Sprent, J. I., Sutherland, J. M., McInroy, S. G. & Minchin, F. R. (1992). The structure of nitrogen fixing root nodules on the aquatic mimosoid legume Neptunia plena. Annals of Botany 69, 173-180.

James, E. K., Minchin, F. R., Iannetta, P. P. M. & Sprent, J. I. (1997). Temporal relationships between nitrogenase and intercellular glycoprotein in developing white lupin nodules. Annals of Botany 79, 493-503.

James, E. K. & Sprent, J. I. (1999). Development of N2-fixing nodules on the wetland legume Lotus uliginosus exposed to conditions of flooding. New Phytologist 142, 219-231.

Jarvis, B. D. W., Van Berkum, P., Chen, W. X., Nour, S. M., Fernandez, M. P., Cleyet-Marel, J. C. & Gillis, M. (1997). Transfer of Rhizobium loti, Rhizobium huakuii, Rhizobium ciceri, Rhizobium mediterraneum, and Rhizobium tianshanense to Mesorhizobium gen. nov. International Journal of Systematic and Evolutionary Microbiology 47, 895-898.

Jiang, G. Q., Krishnan, A. H., Kim, Y. W., Wacek, T. J. & Krishnan, H. B. (2001). A functional myo-inositol dehydrogenase gene is required for efficient nitrogen fixation and competitiveness of Sinorhizobium fredii USDA191 to nodulate soybean (Glycine max [L.] Merr.). Journal of Bacteriology 183, 2595-2604.

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

Jones, K. M., Kobayashi, H., Davies, B. W., Taga, M. E. & Walker, G. C. (2007). How rhizobial symbionts invade plants: the Sinorhizobium-Medicago model. Nature Reviews Microbiology 5, 619-633.

Jordan, D. C. (1982). Transfer of Rhizobium japonicum Buchanan 1980 to Bradyrhizobium gen. nov., a genus of slow-growing, root nodule bacteria from leguminous plants. International Journal Of Systematic Bacteriology 32, 136-139.

Jourand, P., Giraud, E., Béna, G., Sy, A., Willems, A., Gillis, M., Dreyfus, B. & de Lajudie, P. (2004). Methylobacterium nodulans sp. nov., for a group of aerobic, facultatively methylotrophic, legume root-nodule-forming and nitrogen-fixing bacteria. International Journal of Systematic and Evolutionary Microbiology 54, 2269-2273.

Journet, E.-P., Barker, D., Harrison, H. J. & E., K. (2001). M. truncatula as biological material. In EMBO Practical Course on the New Plant Model System Medicago truncatula. http://www.isv.cnrs-gif.fr/embo01/manuels/pdf/embodebut.pdf

Kalita, M., Stepkowski, T., Łotocka, B. & Malek, W. (2006). Phylogeny of nodulation genes and symbiotic properties of Genista tinctoria bradyrhizobia. Archives of Microbiology 186, 87-97.

228 Bibliography

Kaneko, T., Nakamura, Y., Sato, S., Minamisawa, K., Uchiumi, T., Sasamoto, S., Watanabe, A., Idesawa, K., Iriguchi, M., Kawashima, K. & other authors (2002). Complete genomic sequence of nitrogen-fixing symbiotic bacterium Bradyrhizobium japonicum USDA110. DNA Research 9, 189-197.

Kanso, S. & Patel, B. K. C. (2003). Microvirga subterranea gen. nov., sp. nov., a moderate thermophile from a deep subsurface Australian thermal aquifer. International Journal of Systematic and Evolutionary Microbiology 53, 401-406.

Karas, B., Murray, J., Gorzelak, M., Smith, A., Sato, S., Tabata, S. & Szczyglowski, K. (2005). Invasion of Lotus japonicus root hairless 1 by Mesorhizobium loti involves the nodulation factor-dependent induction of root hairs. Plant Physiology 137, 1331-1344.

Kato, Y., Asahara, M., Goto, K., Kasai, H. & Yokota, A. (2008). Methylobacterium persicinum sp nov., Methylobacterium komagatae sp nov., Methylobacterium brachiatum sp nov., Methylobacterium tardum sp nov and Methylobacterium gregans sp nov., isolated from freshwater. International Journal of Systematic and Evolutionary Microbiology 58, 1134-1141.

Kiers, E. T., West, S. A. & Denison, R. F. (2008). Maintaining cooperation in the legume-rhizobia symbiosis: identifying selection pressures and mechanisms. In Nitrogen-fixing Leguminous Symbioses, pp. 59-76. Edited by M. J. Dilworth, W. E. Newton, E. K. James & J. I. Sprent. Dordrecht, The Netherlands: Springer.

Kleinig, H. & Broughton, W. J. (1982). Carotenoid pigments in a red strain of Rhizobium from Lotononis bainesii Baker. Archives of Microbiology 133, 164.

Kloesges, T., Popa, O., Martin, W. & Dagan, T. (2011). Networks of gene sharing among 329 Proteobacterial genomes reveal differences in lateral gene transfer frequency at different phylogenetic depths. Molecular Biology and Evolution 28, 1057-1074.

Knee, E. M., Gong, F. C., Gao, M. S., Teplitski, M., Jones, A. R., Foxworthy, A., Mort, A. J. & Bauer, W. D. (2001). Root mucilage from pea and its utilization by rhizosphere bacteria as a sole carbon source. Molecular Plant-Microbe Interactions 14, 775-784.

Knief, C., Frances, L. & Vorholt, J. A. (2010). Competitiveness of diverse Methylobacterium strains in the phyllosphere of Arabidopsis thaliana and identification of representative models, including M. extorquens PA1. Microbial Ecology 60, 440-452.

Koonin, E. V. & Wolf, Y. I. (2008). Genomics of bacteria and archaea: the emerging dynamic view of the prokaryotic world. Nucleic Acids Research 36, 6688- 6719.

Kouchi, H., Imaizumi-Anraku, H., Hayashi, M., Hakoyama, T., Nakagawa, T., Umehara, Y., Suganuma, N. & Kawaguchi, M. (2010). How many peas in a pod?

229 Bibliography

Legume genes responsible for mutualistic symbioses underground. Plant and Cell Physiology 51, 1381-1397.

Krusell, L., Krause, K., Ott, T., Desbrosses, G., Kramer, U., Sato, S., Nakamura, Y., Tabata, S., James, E. K., Sandal, N. & other authors (2005). The sulfate transporter SST1 is crucial for symbiotic nitrogen fixation in Lotus japonicus root nodules. Plant Cell 17, 1625-1636.

Kunin, V., Goldovsky, L., Darzentas, N. & Ouzounis, C. A. (2005). The net of life: Reconstructing the microbial phylogenetic network. Genome Research 15, 954- 959.

Kuykendall, L. D., Saxena, B., Devine, T. E. & Udell, S. E. (1992). Genetic diversity in Bradyrhizobium japonicum (Jordan, 1982) and proposal for Bradyrhizobium elkanii sp.nov. Canadian Journal of Microbiology 38, 501-505.

Laeremans, T. & Vanderleyden, J. (1998). Review: Infection and nodulation signalling in Rhizobium-Phaseolus vulgaris symbiosis. World Journal of Microbiology and Biotechnology 14, 787-808.

Lafay, B. & Burdon, J. J. (1998). Molecular diversity of rhizobia occurring on native shrubby legumes in southeastern Australia. Applied and Environmental Microbiology 64, 3989-3997.

Lafay, B., Bullier, E. & Burdon, J. J. (2006). Bradyrhizobia isolated from root nodules of Parasponia (Ulmaceae) do not constitute a separate coherent lineage. International Journal of Systematic and Evolutionary Microbiology 56, 1013-1018.

Lafay, B. & Burdon, J. J. (2007). Molecular diversity of legume root-nodule bacteria in Kakadu National Park, Northern Territory, Australia. PLoS One 2.

Laguerre, G., Nour, S. M., Macheret, V., Sanjuan, J., Drouin, P. & Amarger, N. (2001). Classification of rhizobia based on nodC and nifH gene analysis reveals a close phylogenetic relationship among Phaseolus vulgaris symbionts. Microbiology 147, 981-993.

Lapage, S. P., Sneath, P. H. A., Lessel, E. F., Skerman, V. B. D., Seeliger, H. P. R. & Clark, W. A. (1992). International Code of Nomenclature of Bacteria. Washington (DC): ASM Press.

Laranjo, M., Machado, J., Young, J. P. W. & Oliveira, S. (2004). High diversity of chickpea Mesorhizobium species isolated in a Portuguese agricultural region. FEMS Microbiology Ecology 48, 101-107.

Laurion, I., Lami, A. & Sommaruga, R. (2002). Distribution of mycosporine-like amino acids and photoprotective carotenoids among freshwater phytoplankton assemblages. Aquatic Microbial Ecology 26, 283-294.

230 Bibliography

Lavin, M., Pennington, R. T., Klitgaard, B. B., Sprent, J. I., de Lima, H. C. & Gasson, P. E. (2001). The dalbergioid legumes (Fabaceae): delimitation of a pantropical monophyletic clade. American Journal of Botany 88, 503-533.

Lavin, M., Herendeen, P. S. & Wojciechowski, M. F. (2005). Evolutionary rates analysis of Leguminosae implicates a rapid diversification of lineages during the Tertiary. Systematic Biology 54, 575-594.

Law, I. J. (1979). Resistance of Rhizobium specific for Lotononis bainesii to ultraviolet irradiation. Soil Biology & Biochemistry 11, 87-88.

Lawrence, J. G. & Hendrickson, H. (2003). Lateral gene transfer: when will adolescence end? Molecular Microbiology 50, 739-749.

Lee, K. B., De Backer, P., Aono, T., Liu, C. T., Suzuki, S., Suzuki, T., Kaneko, T., Yamada, M., Tabata, S., Kupfer, D. M. & other authors (2008). The genome of the versatile nitrogen fixer Azorhizobium caulinodans ORS571. BMC Genomics 9.

Lee, M. C., Chou, H. H. & Marx, C. J. (2009). Asymmetric, bimodal trade-offs during adaptation of Methylobacterium to distinct growth substrates. Evolution 63, 2816-2830.

León-Barrios, M., Lorite, M. J., Donate-Correa, J. & Sanjuán, J. (2009). Ensifer meliloti bv. lancerottense establishes nitrogen-fixing symbiosis with Lotus endemic to the Canary Islands and shows distinctive symbiotic genotypes and host range. Systematic and Applied Microbiology 32, 413-420.

Leone, F. A. & Hedrick, L. R. (1954). Sodium chloride as a replacement for phosphate in media for the bacterial production and determination of acetoin. Applied and Environmental Microbiology 2, 259-262.

Lewis, G., Schrire, B., Mackinder, B. & Lock, M. (2005). Legumes of the World. Richmond, Surrey: Royal Botanic Gardens, Kew.

Li, G. D., Lodge, G. M., Moore, G. A., Craig, A. D., Dear, B. S., Boschma, S. P., Albertsen, T. O., Miller, S. M., Harden, S., Hayes, R. C., Hughes, S. J., Snowball, R., Smith, A. B. & Cullis, B. C. (2008). Evaluation of perennial pasture legumes and herbs to identify species with high herbage production and persistence in mixed farming zones in southern Australia. Australian Journal of Experimental Agriculture 48, 449-466.

Lidstrom, M. E. (2006). Aerobic methylotrophic prokaryotes. In The Prokaryotes : a Handbook on the Biology of Bacteria : Ecophysiology and Biochemistry, pp. 618- 634. Edited by M. Dworkin, S. Falkow, E. Rosenberg, K. H. Schleifer & E. Stackebrandt. New York: Springer-Verlag.

Lie, T. A., Göktan, D., Engin, M., Pijnenborg, J. & Anlarsal, E. (1987). Co- evolution of the legume-Rhizobium association. Plant and Soil 100, 171-181.

231 Bibliography

Lin, D. X., Wang, E. T., Tang, H., Han, T. X., He, Y. R., Guan, S. H. & Chen, W. X. (2008). Shinella kummerowiae sp nov., a symbiotic bacterium isolated from root nodules of the herbal legume Kummerowia stipulacea. International Journal of Systematic and Evolutionary Microbiology 58, 1409-1413.

Line, J. E., Hiett, K. L., Guard-Bouldin, J. & Seal, B. S. (2010). Differential carbon source utilization by Campylobacter jejuni 11168 in response to growth temperature variation. Journal of Microbiological Methods 80, 198-202.

Lindström, K., Murwira, M., Willems, A. & Altier, N. (2010). The biodiversity of beneficial microbe-host mutualism: the case of rhizobia. Research in Microbiology 161, 453-463.

Lodwig, E. & Poole, P. (2003). Metabolism of Rhizobium bacteroids. Critical Reviews in Plant Sciences 22, 37-78.

Lodwig, E. M., Hosie, A. H. F., Bordès, A., Findlay, K., Allaway, D., Karunakaran, R., Downie, J. A. & Poole, P. S. (2003). Amino-acid cycling drives nitrogen fixation in the legume - Rhizobium symbiosis. Nature 422, 722-726.

Loi, A., Howieson, J. G., Nutt, B. J. & Carr, S. J. (2005). A second generation of annual pasture legumes and their potential for inclusion in Mediterranean-type farming systems. Australian Journal of Experimental Agriculture 45, 289-299.

Lorite, M. J., Donate-Correa, J., del Arco-Aguilar, M., Galdona, R. P., Sanjuán, J. & León-Barrios, M. (2010). Lotus endemic to the Canary Islands are nodulated by diverse and novel rhizobial species and symbiotypes. Systematic and Applied Microbiology 33, 282-290.

Lortet, G., Mear, N., Lorquin, J., Dreyfus, B., de Lajudie, P., Rosenberg, C. & Boivin, C. (1996). Nod factor thin-layer chromatography profiling as a tool to characterize symbiotic specificity of rhizobial strains: Application to Sinorhizobium saheli, S. teranga, and Rhizobium sp. strains isolated from Acacia and Sesbania. Molecular Plant-Microbe Interactions 9, 736-747.

Łotocka, B., Kopcinska, J., Gorecka, M. & Golinowski, W. (2000). Formation and abortion of root nodule primordia in Lupinus luteus L. Acta Biologica Cracoviensia Series Botanica 42, 87-102.

Loureiro, M. F., James, E. K., Sprent, J. I. & Franco, A. A. (1995). Stem and root nodules on the tropical wetland legume Aeschynomene fluminensis. New Phytologist 130, 531-544.

Lu, Y. L., Chen, W. F., Wang, E. T., Guan, S. H., Yan, X. R. & Chen, W. X. (2009). Genetic diversity and biogeography of rhizobia associated with Caragana species in three ecological regions of China. Systematic and Applied Microbiology 32, 351-361.

232 Bibliography

Ludwig, W., Strunk, O., Klugbauer, S., Klugbauer, N., Weizenegger, M., Neumaier, J., Bachleitner, M. & Schleifer, K. H. (1998). Bacterial phylogeny based on comparative sequence analysis. Electrophoresis 19, 554-568.

Lupski, J. R. & Weinstock, G. M. (1992). Short, interspersed repetitive DNA sequences in prokaryotic genomes. Journal of Bacteriology 174, 4525-4529.

MacLean, A. M., Finan, T. M. & Sadowsky, M. J. (2007). Genomes of the symbiotic nitrogen-fixing bacteria of legumes. Plant Physiology 144, 615-622.

Madhaiyan, M., Kim, B. Y., Poonguzhali, S., Kwon, S. W., Song, M. H., Ryu, J. H., Go, S. J., Koo, B. S. & Sa, T. M. (2007). Methylobacterium oryzae sp nov., an aerobic, pink-pigmented, facultatively methylotrophic, 1-aminocyclopropane-L- carboxylate deaminase-producing bacterium isolated from rice. International Journal of Systematic and Evolutionary Microbiology 57, 326-331.

Madhaiyan, M., Poonguzhali, S., Senthilkumar, M., Sundaram, S. P. & Sa, T. (2009a). Nodulation and plant-growth promotion by methylotrophic bacteria isolated from tropical legumes. Microbiological Research 164, 114-120.

Madhaiyan, M., Poonguzhali, S., Kwon, S.-W. & Sa, T.-M. (2009b). Methylobacterium phyllosphaerae sp. nov., a pink-pigmented, facultative methylotroph from the phyllosphere of rice. International Journal of Systematic and Evolutionary Microbiology 59, 22-27.

Madsen, L. H., Tirichine, L., Jurkiewicz, A., Sullivan, J. T., Heckmann, A. B., Bek, A. S., Ronson, C. W., James, E. K. & Stougaard, J. (2010). The molecular network governing nodule organogenesis and infection in the model legume Lotus japonicus. Nature Communications 1, 1-12.

Mahdhi, M., Nzoue, A., Gueye, F., Merabet, C., de Lajudie, P. & Mars, M. (2007). Phenotypic and genotypic diversity of Genista saharae microsymbionts from the infra-arid region of Tunisia. Letters in Applied Microbiology 45, 604-609.

Mahdhi, M., de Lajudie, P. & Mars, M. (2008). Phylogenetic and symbiotic characterization of rhizobial bacteria nodulating Argyrolobium uniflorum in Tunisian arid soils. Canadian Journal of Microbiology 54, 209-217.

Mahé, F., Markova, D., Pasquet, R., Misset, M.-T. & Aïnouche, A. (2011). Isolation, phylogeny and evolution of the SymRK gene in the legume genus Lupinus L. Molecular Phylogenetics and Evolution 60, 49-61.

Martens, M., Dawyndt, P., Coopman, R., Gillis, M., De Vos, P. & Willems, A. (2008). Advantages of multilocus sequence analysis for taxonomic studies: a case study using 10 housekeeping genes in the genus Ensifer (including former Sinorhizobium). International Journal of Systematic and Evolutionary Microbiology 58, 200-214.

Martin, B., Humbert, O., Camara, M., Guenzi, E., Walker, J., Mitchell, T., Andrew, P., Prudhomme, M., Alloing, G., Hakenbeck, R., Morrison, D. A.,

233 Bibliography

Boulnois, G. J. & Claverys, J.-P. (1992). A highly conserved repeated DNA element located in the chromosome of Streptococcus pneumoniae. Nucleic Acids Research 20, 3479-3483.

Martínez-Romero, E. (2003). Diversity of Rhizobium-Phaseolus vulgaris symbiosis: overview and perspectives. Plant and Soil 252, 11-23.

Martínez-Romero, E. (2009). Coevolution in Rhizobium-legume symbiosis? DNA and Cell Biology 28, 361-370.

Martínez-Romero, E., Segovia, L., Mercante, F. M., Franco, A. A., Graham, P. & Pardo, M. A. (1991). Rhizobium tropici, a novel species nodulating Phaseolus vulgaris L. beans and Leucaena sp. trees. International Journal of Systematic Bacteriology 41, 417-426.

Martiny, J. B. H., Bohannan, B. J. M., Brown, J. H., Colwell, R. K., Fuhrman, J. A., Green, J. L., Horner-Devine, M. C., Kane, M., Krumins, J. A., Kuske, C. R. & other authors (2006). Microbial biogeography: putting microorganisms on the map. Nature Reviews Microbiology 4, 102-112.

Masson-Boivin, C., Giraud, E., Perret, X. & Batut, J. (2009). Establishing nitrogen-fixing symbiosis with legumes: how many rhizobium recipes? Trends in Microbiology 17, 458-466.

Maunoury, N., Kondorosi, A., Kondorosi, E. & Mergaert, P. (2008). Cell biology of nodule infection and development. In Nitrogen-fixing Leguminous Symbioses, pp. 153-189. Edited by M. J. Dilworth, E. K. James, J. I. Sprent & W. E. Newton. Dordrecht, The Netherlands: Springer.

McMahon, M. M. & Sanderson, M. J. (2006). Phylogenetic supermatrix analysis of GenBank sequences from 2228 papilionoid legumes. Systematic Biology 55, 818- 836.

Meade, H. M., Long, S. R., Ruvkun, G. B., Brown, S. E. & Ausubel, F. M. (1982). Physical and genetic characterization of symbiotic and auxotrophic mutants of Rhizobium meliloti induced by transposon Tn5 mutagenesis. Journal of Bacteriology 149, 114-122.

Menna, P., Barcellos, F. G. & Hungria, M. (2009). Phylogeny and taxonomy of a diverse collection of Bradyrhizobium strains based on multilocus sequence analysis of the 16S rRNA gene, ITS region and glnII, recA, atpD and dnaK genes. International Journal of Systematic and Evolutionary Microbiology 59, 2934-2950.

Mergaert, P., Van Montagu, M. & Holsters, M. (1997). Molecular mechanisms of Nod factor diversity. Molecular Microbiology 25, 811-817.

Mergaert, P., Uchiumi, T., Alunni, B., Evanno, G., Cheron, A., Catrice, O., Mausset, A. E., Barloy-Hubler, F., Galibert, F., Kondorosi, A. & Kondorosi, E. (2006). Eukaryotic control on bacterial cell cycle and differentiation in the

234 Bibliography

Rhizobium-legume symbiosis. Proceedings of the National Academy of Sciences of the United States of America 103, 5230-5235.

Miché, L., Moulin, L., Chaintreuil, C., Contreras-Jimenez, J. L., Munive- Hernández, J. A., Villegas-Hernandez, M. D., Crozier, F. & Béna, G. (2010). Diversity analyses of Aeschynomene symbionts in Tropical Africa and Central America reveal that nod-independent stem nodulation is not restricted to photosynthetic bradyrhizobia. Environmental Microbiology 12, 2152-2164.

Mnasri, B., Mrabet, M., Laguerre, G., Aouani, M. E. & Mhamdi, R. (2007). Salt- tolerant rhizobia isolated from a Tunisian oasis that are highly effective for symbiotic N2-fixation with Phaseolus vulgaris constitute a novel biovar (bv. mediterranense) of Sinorhizobium meliloti. Archives of Microbiology 187, 79-85.

Mnasri, B., Badri, Y., Saidi, S., de Lajudie, P. & Mhamdi, R. (2009). Symbiotic diversity of Ensifer meliloti strains recovered from various legume species in Tunisia. Systematic and Applied Microbiology 32, 583-592.

Molouba, F., Lorquin, J., Willems, A., Hoste, B., Giraud, E., Dreyfus, B., Gillis, M., de Lajudie, P. & Masson-Boivin, C. (1999). Photosynthetic bradyrhizobia from Aeschynomene spp. are specific to stem-nodulated species and form a separate 16S ribosomal DNA restriction fragment length polymorphism group. Applied and Environmental Microbiology 65, 3084-3094.

Morón, B., Soria-Díaz, M. E., Ault, J., Verroios, G., Noreen, S., Rodríguez- Navarro, D. N., Gil-Serrano, A., Thomas-Oates, J., Megías, M. & Sousa, C. (2005). Low pH changes the profile of nodulation factors produced by Rhizobium tropici CIAT899. Chemistry & Biology 12, 1029-1040.

Mort, A. J. & Grover, P. B. (1988). Characterization of root hair cell walls as potential barriers to the infection of plants by rhizobia. Plant Physiology 86, 638- 641.

Moulin, L., Debellé, F., Giraud, E., Mangin, B., Dénarié, J. & Boivin-Masson, C. (2000). The nodA sequence of rhizobia gives clues on structural features of Nod factors. In Nitrogen Fixation: From Molecules to Crop Productivity, pp. 204. Edited by F. O. Pedrosa, M. Hungria, M. G. Yates & W. E. Newton. Dordrecht: Kluwer.

Moulin, L., Béna, G., Boivin-Masson, C. & Stepkowski, T. (2004). Phylogenetic analyses of symbiotic nodulation genes support vertical and lateral gene co-transfer within the Bradyrhizobium genus. Molecular Phylogenetics and Evolution 30, 720- 732.

Mutch, L. A. & Young, J. P. W. (2004). Diversity and specificity of Rhizobium leguminosarum biovar viciae on wild and cultivated legumes. Molecular Ecology 13, 2435-2444.

Nambiar, P. T. C., Dart, P. J., Srinivasa Rao, B. & Ramanatha Rao, V. (1982). Nodulation in the hypocotyl region of groundnut (Arachis hypogaea). Experimental Agriculture 18, 203-207.

235 Bibliography

Nandasena, K. G., O'Hara, G. W., Tiwari, R. P., Sezmiş, E. & Howieson, J. G. (2007a). In situ lateral transfer of symbiosis islands results in rapid evolution of diverse competitive strains of mesorhizobia suboptimal in symbiotic nitrogen fixation on the pasture legume Biserrula pelecinus L. Environmental Microbiology 9, 2496-2511.

Nandasena, K. G., O'Hara, G. W., Tiwari, R. P., Willlems, A. & Howieson, J. G. (2007b). Mesorhizobium ciceri biovar biserrulae, a novel biovar nodulating the pasture legume Biserrula pelecinus L. International Journal of Systematic and Evolutionary Microbiology 57, 1041-1045.

Nandasena, K. G., O'Hara, G. W., Tiwari, R. P., Willems, A. & Howieson, J. G. (2009). Mesorhizobium australicum sp. nov. and Mesorhizobium opportunistum sp. nov., isolated from Biserrula pelecinus L. in Australia. International Journal of Systematic and Evolutionary Microbiology 59, 2140-2147.

Nesme, X., Picard, C. & Simonet, P. (1995). Specific DNA sequences for detection of soil bacteria. In Nucleic Acids in the Environment, pp. 112-139. Edited by J. T. Trevors & J. D. Van Elsas. Berlin: Springer-Verlag.

Nichols, P. G. H., Loi, A., Nutt, B. J., Evans, P. M., Craig, A. D., Pengelly, B. C., Dear, B. S., Lloyd, D. L., Revell, C. K., Nair, R. M. & other authors (2007). New annual and short-lived perennial pasture legumes for Australian agriculture - 15 years of revolution. Field Crops Research 104, 10-23.

Nick, G., de Lajudie, P., Eardly, B. D., Suomalainen, S., Paulin, L., Zhang, X., Gillis, M. & Lindström, K. (1999). Sinorhizobium arboris sp. nov. and Sinorhizobium kostiense sp. nov., isolated from leguminous trees in Sudan and Kenya. International Journal of Systematic and Evolutionary Microbiology 49, 1359-1368.

Normand, P., Cournoyer, B., Simonet, P. & Nazaret, S. (1992). Analysis of a ribosomal RNA operon in the actinomycete Frankia. Gene 111, 119-124.

Norris, D. O. (1958). A red strain of Rhizobium from Lotononis bainesii Baker. Australian Journal of Agricultural Research 9, 629-632.

Ochman, H., Lawrence, J. G. & Groisman, E. A. (2000). Lateral gene transfer and the nature of bacterial innovation. Nature 405, 299-304.

Ochman, H. & Moran, N. A. (2001). Genes lost and genes found: Evolution of bacterial pathogenesis and symbiosis. Science 292, 1096-1098.

Odee, D. W., Haukka, K., McInroy, S. G., Sprent, J. I., Sutherland, J. M. & Young, J. P. W. (2002). Genetic and symbiotic characterization of rhizobia isolated from tree and herbaceous legumes grown in soils from ecologically diverse sites in Kenya. Soil Biology & Biochemistry 34, 801-811.

236 Bibliography

O'Hara, G. W., Dilworth, M. J., Boonkerd, N. & Parkpian, P. (1988). Iron- deficiency specifically limits nodule development in peanut inoculated with Bradyrhizobium sp. New Phytologist 108, 51-57.

O'Hara, G. W., Goss, T. J., Dilworth, M. J. & Glenn, A. R. (1989). Maintenance of intracellular pH and acid-tolerance in Rhizobium meliloti. Applied and Environmental Microbiology 55, 1870–1876.

Oldroyd, G. E. D. & Downie, J. A. (2008). Coordinating nodule morphogenesis with rhizobial infection in legumes. Annual Review of Plant Biology 59, 519-546. Omelchenko, M., Makarova, K., Wolf, Y., Rogozin, I. & Koonin, E. (2003). Evolution of mosaic operons by horizontal gene transfer and gene displacement in situ. Genome Biology 4, R55.

Omer, Z. S., Tombolini, R. & Gerhardson, B. (2004). Plant colonization by pink- pigmented facultative methylotrophic bacteria (PPFMs). FEMS Microbiology Ecology 47, 319-326.

Oono, R. & Denison, R. F. (2010). Comparing symbiotic efficiency between swollen versus nonswollen rhizobial bacteroids. Plant Physiology 154, 1541-1548.

Oono, R., Schmitt, I., Sprent, J. I. & Denison, R. F. (2010). Multiple evolutionary origins of legume traits leading to extreme rhizobial differentiation. New Phytologist 187, 508-520.

Oresnik, I. J., Pacarynuk, L. A., O'Brien, S. A. P., Yost, C. K. & Hynes, M. F. (1998). Plasmid-encoded catabolic genes in Rhizobium leguminosarum bv. trifolii: Evidence for a plant-inducible rhamnose locus involved in competition for nodulation. Molecular Plant-Microbe Interactions 11, 1175-1185.

Ormeño-Orrillo, E., Vinuesa, P., Zúñiga-Dávila, D. & Martínez-Romero, E. (2006). Molecular diversity of native bradyrhizobia isolated from Lima bean (Phaseolus lunatus L.) in Peru. Systematic and Applied Microbiology 29, 253-262.

Ourarhi, M., Abdelmoumen, H., Guerrouj, K., Benata, H., Muresu, R., Squartini, A. & Missbah El Idrissi, M. (2011). Colutea arborescens is nodulated by diverse rhizobia in Eastern Morocco. Archives of Microbiology 193, 115-124.

Palys, T., Berger, E., Mitrica, I., Nakamura, L. K. & Cohan, F. M. (2000). Protein-coding genes as molecular markers for ecologically distinct populations: the case of two Bacillus species. International Journal of Systematic and Evolutionary Microbiology 50, 1021-1028.

Papke, R. T. & Ward, D. M. (2004). The importance of physical isolation to microbial diversification. FEMS Microbiology Ecology 48, 293-303.

Parker, M. A. (2003). Genetic markers for analysing symbiotic relationships and lateral gene transfer in Neotropical bradyrhizobia. Molecular Ecology 12, 2447- 2455.

237 Bibliography

Parker, M. A. (2004). rRNA and dnaK relationships of Bradyrhizobium sp. nodule bacteria from four Papilionoid legume trees in Costa Rica. Systematic and Applied Microbiology 27, 334-342.

Parker, M. A. (2008). Symbiotic relationships of legumes and nodule bacteria on Barro Colorado Island, Panama: A review. Microbial Ecology 55, 662-672.

Parker, M. A. & Kennedy, D. A. (2006). Diversity and relationships of bradyrhizobia from legumes native to eastern North America. Canadian Journal of Microbiology 52, 1148-1157.

Peoples, M. B., Herridge, D. F. & Ladha, J. K. (1995). Biological nitrogen fixation: An efficient source of nitrogen for sustainable agricultural production? . Plant and Soil 174, 3-28.

Perret, X., Staehelin, C. & Broughton, W. J. (2000). Molecular basis of symbiotic promiscuity. Microbiology and Molecular Biology Reviews 64, 180-201.

Prell, J., White, J. P., Bourdes, A., Bunnewell, S., Bongaerts, R. J. & Poole, P. S. (2009). Legumes regulate Rhizobium bacteroid development and persistence by the supply of branched-chain amino acids. Proceedings of the National Academy of Sciences of the United States of America 106, 12477-12482.

Pueppke, S. G. & Broughton, W. J. (1999). Rhizobium sp. strain NGR234 and R. fredii USDA257 share exceptionally broad, nested host ranges. Molecular Plant- Microbe Interactions 12, 293-318.

Qian, J. H., Kwon, S. W. & Parker, M. A. (2003). rRNA and nifD phylogeny of Bradyrhizobium from sites across the Pacific Basin. FEMS Microbiology Letters 219, 159-165.

Rainey, F. A., Ward-Rainey, N. L., Janssen, P. H., Hippe, H. & Stackebrandt, E. (1996). Clostridium paradoxum DSM 7308T contains multiple 16S rRNA genes with heterogeneous intervening sequences. Microbiology 142, 2087-2095.

Rasolomampianina, R., Bailly, X., Fetiarison, R., Rabevohitra, R., Béna, G., Ramaroson, L., Raherimandimby, M., Moulin, L., De Lajudie, P., Dreyfus, B. & Avarre, J. C. (2005). Nitrogen-fixing nodules from rose wood legume trees (Dalbergia spp.) endemic to Madagascar host seven different genera belonging to α- and β-Proteobacteria. Molecular Ecology 14, 4135-4146.

Real, D., Dalla Rizza, M., Quesenberry, K. H. & Echenique, M. (2004). Reproductive and molecular evidence for allogamy in Lotononis bainesii Baker. Crop Science 44, 394-400.

Reasoner, D. J. & Geldreich, E. E. (1985). A new medium for the enumeration and subculture of bacteria from potable water. Applied and Environmental Microbiology 49, 1-7.

238 Bibliography

Renier, A., Jourand, P., Rapior, S., Poinsot, V., Sy, A., Dreyfus, B. & Moulin, L. (2008). Symbiotic properties of Methylobacterium nodulans ORS 2060(T): A classic process for an atypical symbiont. Soil Biology & Biochemistry 40, 1404-1412.

Renier, A., De Faria, S. M., Jourand, P., Giraud, E., Dreyfus, B., Rapior, S. & Prin, Y. (2011). Nodulation of Crotalaria podocarpa DC. by Methylobacterium nodulans displays very unusual features. Journal of Experimental Botany.

Richardson, K. C., Jarret, L. & Finke, E. H. (1960). Embedding in epoxy resins for ultrathin sectioning in electron microscopy. Stain Technology 35, 313-323.

Rincón, A., Arenal, F., González, I., Manrique, E., Lucas, M. M. & Pueyo, J. J. (2008). Diversity of rhizobial bacteria isolated from nodules of the gypsophyte Ononis tridentata L. growing in Spanish soils. Microbial Ecology 56, 223-233.

Rivas, R., Velázquez, E., Willems, A., Vizcaíno, N., Subba-Rao, N. S., Mateos, P. F., Gillis, M., Dazzo, F. B. & Martínez-Molina, E. (2002). A new species of Devosia that forms a unique nitrogen-fixing root-nodule symbiosis with the aquatic legume Neptunia natans (L.f.) Druce. Applied and Environmental Microbiology 68, 5217-5222.

Rivas, R., Laranjo, M., Mateos, P. F., Oliveira, S., Martínez-Molina, E. & Velázquez, E. (2007). Strains of Mesorhizobium amorphae and Mesorhizobium tianshanense, carrying symbiotic genes of common chickpea endosymbiotic species, constitute a novel biovar (ciceri) capable of nodulating Cicer arietinum. Letters in Applied Microbiology 44, 412-418.

Roche, P., Maillet, F., Plazanet, C., Debellé, F., Ferro, M., Truchet, G., Promé, J.-C. & Dénarié, J. (1996). The common nodABC genes of Rhizobium meliloti are host-range determinants. Proceedings of the National Academy of Sciences of the United States of America 93, 15305-15310.

Rogel, M. A., Ormeño-Orrillo, E. & Martínez-Romero, E. (2011). Symbiovars in rhizobia reflect bacterial adaptation to legumes. Systematic and Applied Microbiology 34, 96-104.

Ronquist, F. & Huelsenbeck, J. P. (2003). MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19, 1572-1574.

Rosselló-Móra, R. & Amann, R. (2001). The species concept for prokaryotes. FEMS Microbiology Reviews 25, 39-67.

Saitou, N. & Nei, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4, 406-425.

Sandmann, W. P. (1970). Results of formal greenhouse tests (1963-1969) of strains of Rhizobium and information about some other trials, pp. 1-56. Department of Research and Specialist Services, Marandellas, Zimbabwe: Zimbabwe Ministry of Agriculture.

239 Bibliography

Sawada, H., Kuykendall, L. D. & Young, J. M. (2003). Changing concepts in the systematics of bacterial nitrogen-fixing legume symbionts. Journal of General and Applied Microbiology 49, 155-179.

Schleifer, K. H. (2009). Classification of Bacteria and Archaea: Past, present and future. Systematic and Applied Microbiology 32, 533-542.

Schmeisser, C., Liesegang, H., Krysciak, D., Bakkou, N., Le Quéré, A., Wollherr, A., Heinemeyer, I., Morgenstern, B., Pommerening-Röser, A., Flores, M. & other authors (2009). Rhizobium sp. strain NGR234 possesses a remarkable number of secretion systems. Applied and Environmental Microbiology 75, 4035- 4045.

Schrire, B., Lewis, G. & Lavin, M. (2005). Biogeography of the Leguminosae. In Legumes of the World, pp. 21-54. Edited by G. Lewis, B. Schrire, B. Mackinder & M. Lock. Richmond, Surrey, UK: Royal Botanic Gardens, Kew.

Scotese, C. R. (2004). A continental drift flipbook. Journal of Geology 112, 729- 741.

Sessitsch, A., Howieson, J. G., Perret, X., Antoun, H. & Martínez-Romero, E. (2002). Advances in Rhizobium research. Critical Reviews in Plant Sciences 21, 323- 378.

Simms, E. L. & Taylor, D. L. (2002). Partner choice in nitrogen-fixation mutualisms of legumes and rhizobia. Integrative and Comparative Biology 42, 369- 380.

Singleton, P. W. & Tavares, J. W. (1986). Inoculation response of legumes in relation to the number and effectiveness of indigenous Rhizobium populations. Applied and Environmental Microbiology 51, 1013-1018.

Sinharoy, S. & DasGupta, M. (2009). RNA interference highlights the role of CCaMK in dissemination of endosymbionts in the Aeschynomeneae legume Arachis. Molecular Plant-Microbe Interactions 22, 1466-1475.

Slattery, J. F., Pearce, D. J. & Slattery, W. J. (2004). Effects of resident rhizobial communities and soil type on the effective nodulation of pulse legumes. Soil Biology & Biochemistry 36, 1339-1346.

Soltis, D. E., Soltis, P. S., Morgan, D. R., Swensen, S. M., Mullin, B. C., Dowd, J. M. & Martin, P. G. (1995). Chloroplast gene sequence data suggest a single origin of the predisposition for symbiotic nitrogen-fixation in angiosperms. Proceedings of the National Academy of Sciences of the United States of America 92, 2647-2651.

Soltis, E. D. & Soltis, P. S. (2000). Contributions of plant molecular systematics to studies of molecular evolution. Plant Molecular Biology 42, 45-75.

Sprent, J. I. (2007). Evolving ideas of legume evolution and diversity: a taxonomic perspective on the occurrence of nodulation. New Phytologist 174, 11-25.

240 Bibliography

Sprent, J. I. (2008a). 60Ma of legume nodulation. What's new? What's changing? Journal of Experimental Botany 59, 1081-1084.

Sprent, J. I. (2008b). Evolution and diversity of legume symbiosis. In Nitrogen- fixing Leguminous Symbioses, pp. 1-21. Edited by M. J. Dilworth, E. K. James, J. I. Sprent & W. E. Newton. Dordrecht, The Netherlands: Springer.

Sprent, J. I. (2009). Legume Nodulation: A Global Perspective. Oxford, U.K.: Wiley-Blackwell.

Sprent, J. I. & James, E. K. (2007). Legume evolution: Where do nodules and mycorrhizas fit in? Plant Physiology 144, 575-581.

Stackebrandt, E. & Goebel, B. M. (1994). Taxonomic note: a place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. International Journal of Systematic Bacteriology 44, 846-849.

Stackebrandt, E. & Ebers, J. (2006). Taxonomic parameters revisited: tarnished gold standards. Microbiology Today 33, 152-155.

Stackebrandt, E., Frederiksen, W., Garrity, G. M., Grimont, P. A. D., Kämpfer, P., Maiden, M. C. J., Nesme, X., Rosselló-Mora, R., Swings, J., Trüper, H. G., Vauterin, L., Ward, A. C. & Whitman, W. B. (2002). Report of the ad hoc committee for the re-evaluation of the species definition in bacteriology. International Journal of Systematic and Evolutionary Microbiology 52, 1043-1047.

Steenkamp, E. T., Stepkowski, T., Przymusiak, A., Botha, W. J. & Law, I. J. (2008). Cowpea and peanut in southern Africa are nodulated by diverse Bradyrhizobium strains harboring nodulation genes that belong to the large pantropical clade common in Africa. Molecular Phylogenetics and Evolution 48, 1131-1144.

Stepkowski, T., Czaplińska, M., Miedzinska, K. & Moulin, L. (2003). The variable part of the dnaK Gene as an alternative marker for phylogenetic studies of rhizobia and related Alpha Proteobacteria. Systematic and Applied Microbiology 26, 483-494.

Stepkowski, T., Hughes, C. E., Law, I. J., Markiewicz, L., Gurda, D., Chlebicka, A. & Moulin, L. (2007). Diversification of lupine Bradyrhizobium strains: Evidence from nodulation gene trees. Applied and Environmental Microbiology 73, 3254- 3264.

Stowers, M. D. & Elkan, G. H. (1984). Growth and nutritional characteristics of cowpea rhizobia. Plant and Soil 80, 191-200.

Subba-Rao, N. S., Mateos, P. F., Baker, D., Pankratz, H., Palma, J., Dazzo, F. B. & Sprent, J. I. (1995). The unique root-nodule symbiosis between Rhizobium and the aquatic legume Neptunia natans (L. f.) Druce. Planta 196, 311-320.

241 Bibliography

Sullivan, J. T., Patrick, H. N., Lowther, W. L., Scott, D. B. & Ronson, C. W. (1995). Nodulating strains of Rhizobium loti arise through chromosomal symbiotic gene transfer in the environment. Proceedings of the National Academy of Sciences of the United States of America 92, 8985-8989.

Sullivan, J. T. & Ronson, C. W. (1998). Evolution of rhizobia by acquisition of a 500-kb symbiosis island that integrates into a phe-tRNA gene. Proceedings of the National Academy of Sciences of the United States of America 95, 5145-5149.

Suominen, L., Roos, C., Lortet, G., Paulin, L. & Lindström, K. (2001). Identification and structure of the Rhizobium galegae common nodulation genes: evidence for horizontal gene transfer. Molecular Biology and Evolution 18, 907-916. Sy, A., Giraud, E., Jourand, P., Garcia, N., Willems, A., de Lajudie, P., Prin, Y., Neyra, M., Gillis, M., Boivin-Masson, C. & Dreyfus, B. (2001). Methylotrophic Methylobacterium bacteria nodulate and fix nitrogen in symbiosis with legumes. Journal of Bacteriology 183, 214-220.

Sy, A., Timmers, A. C. J., Knief, C. & Vorholt, J. A. (2005). Methylotrophic metabolism is advantageous for Methylobacterium extorquens during colonization of Medicago truncatula under competitive conditions. Applied and Environmental Microbiology 71, 7245-7252.

Sylla, S. N., Samba, R. T., Neyra, M., Ndoye, I., Giraud, E., Willems, A., de Lajudie, P. & Dreyfus, B. (2002). Phenotypic and genotypic diversity of rhizobia nodulating Pterocarpus erinaceus and P. lucens in Senegal. Systematic and Applied Microbiology 25, 572-583.

Szczyglowski, K., Shaw, R. S., Wopereis, J., Copeland, S., Hamburger, D., Kasiborski, B., Dazzo, F. B. & de Bruijn, F. J. (1998). Nodule organogenesis and symbiotic mutants of the model legume Lotus japonicus. Molecular Plant-Microbe Interactions 11, 684-697.

Takeda, M., Suzuki, I. & Koizumi, J. I. (2004). Balneomonas flocculans gen. nov., sp nov., a new cellulose-producing member of the α-2 subclass of Proteobacteria. Systematic and Applied Microbiology 27, 139-145.

Tambalo, D. D., Del Bel, K. L., Bustard, D. E., Greenwood, P. R., Steedman, A. E. & Hynes, M. F. (2010). Regulation of flagellar, motility and chemotaxis genes in Rhizobium leguminosarum by the VisN/R-Rem cascade. Microbiology 156, 1673- 1685.

Tamura, K., Dudley, J., Nei, M. & Kumar, S. (2007). MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution 24, 1596-1599.

Tang, C., Robson, A. D., Dilworth, M. J. & Kuo, J. (1992). Microscopic evidence on how iron deficiency limits nodule initiation in Lupinus angustifolius L. New Phytologist 121, 457-467.

242 Bibliography

Terpolilli, J. J. (2009). Why are the symbioses between some genotypes of Sinorhizobium and Medicago suboptimal for N2 fixation? PhD thesis, Murdoch University, Perth, Western Australia.

Thrall, P. H., Burdon, J. J. & Woods, M. J. (2000). Variation in the effectiveness of symbiotic associations between native rhizobia and temperate Australian legumes: interactions within and between genera. Journal of Applied Ecology 37, 52-65.

Tindall, B. J., Rosselló-Móra, R., Busse, H.-J., Ludwig, W. & Kämpfer, P. (2010). Notes on the characterization of prokaryote strains for taxonomic purposes. International Journal of Systematic and Evolutionary Microbiology 60, 249-266.

Trinick, M. J. (1980). Relationships amongst the fast-growing rhizobia of Lablab purpureus, Leucaena leucocephala, Mimosa spp., Acacia farnesiana and Sesbania grandiflora and their affinities with other rhizobial groups. Journal of Applied Bacteriology 49, 39–53.

Trujillo, M. E., Willems, A., Abril, A., Planchuelo, A.-M., Rivas, R., Ludeña, D., Mateos, P. F., Martínez-Molina, E. & Velázquez, E. (2005). Nodulation of Lupinus albus by strains of Ochrobactrum lupini sp. nov. Applied and Environmental Microbiology 71, 1318-1327.

Turner, S. L. & Young, J. P. W. (2000). The glutamine synthetases of rhizobia: phylogenetics and evolutionary implications. Molecular Biology and Evolution 17, 309-319.

Uheda, E., Daimon, H. & Yoshizako, F. (2001). Colonization and invasion of peanut (Arachis hypogaea L.) roots by gusA-marked Bradyrhizobium sp. Canadian Journal of Botany 79, 733-738.

Van Aken, B., Peres, C. M., Doty, S. L., Yoon, J. M. & Schnoor, J. L. (2004). Methylobacterium populi sp nov., a novel aerobic, pink-pigmented, facultatively methylotrophic, methane-utilizing bacterium isolated from poplar trees (Populus deltoides x nigra DN34). International Journal of Systematic and Evolutionary Microbiology 54, 1191-1196. van Berkum, P., Terefework, Z., Paulin, L., Suomalainen, S., Lindström, K. & Eardly, B. D. (2003). Discordant phylogenies within the rrn loci of rhizobia. Journal of Bacteriology 185, 2988-2998. van Berkum, P., Lelbold, J. M. & Eardly, B. D. (2006). Proposal for combining Bradyrhizobium spp. (Aeschynomene indica) with Blastobacter denitrificans and to transfer Blastobacter denitrificans (Hirsch and Muller, 1985) to the genus Bradyrhizobium as Bradyrhizobium denitrificans (comb. nov.). Systematic and Applied Microbiology 29, 207-215.

Van Rossum, D., Schuurmans, F. P., Gillis, M., Muyotcha, A., Van Verseveld, H. W., Stouthamer, A. H. & Boogerd, F. C. (1995). Genetic and phenetic analyses of Bradyrhizobium strains nodulating peanut (Arachis hypogaea L.) roots. Applied and Environmental Microbiology 61, 1599-1609.

243 Bibliography van Wyk, B.-E. (1991). A Synopsis of the Genus Lotononis (Fabaceae: Crotalarieae). Cape Town, South Africa: Rustica Press. van Wyk, B.-E. (2005). Tribe Crotalarieae. In Legumes of the World. Edited by G. P. Lewis, B. D. Schrire, B. Mackinder & M. Lock. Richmond, Surrey, UK: Royal Botanic Gardens, Kew. van Wyk, B.-E. & Verdoorn, G. H. (1990). Alkaloids as taxonomic characters in the tribe Crotalarieae (Fabaceae). Biochemical Systematics and Ecology 18, 503-516.

Vandamme, P., Pot, B., Gillis, M., DeVos, P., Kersters, K. & Swings, J. (1996). Polyphasic taxonomy, a consensus approach to bacterial systematics. Microbiological Reviews 60, 407–438.

Vega-Hernández, M. C., Pérez-Galdona, R., Dazzo, F. B., Jarabo-Lorenzo, A., Alfayate, M. C. & León-Barrios, M. (2001). Novel infection process in the indeterminate root nodule symbiosis between Chamaecytisus proliferus (tagasaste) and Bradyrhizobium sp. New Phytologist 150, 707-721.

Versalovic, J., Koeuth, T. & Lupski, R. (1991). Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Research 19, 6823-6831.

Vincent, J. M. (1970). A Manual for the Practical Study of the Root-Nodule Bacteria. Oxford, England: Blackwell Science Publications.

Vinuesa, P., Silva, C., Lorite, M. J., Izaguirre-Mayoral, M. L., Bedmar, E. J. & Martínez-Romero, E. (2005). Molecular systematics of rhizobia based on maximum likelihood and Bayesian phylogenies inferred from rrs, atpD, recA and nifH sequences, and their use in the classification of Sesbania microsymbionts from Venezuelan wetlands. Systematic and Applied Microbiology 28, 702-716.

Vogel, A. I. (1962). A Text-Book of Quantitative Inorganic Analysis. London: Longman Group.

Voroshilova, V. A., Demchenko, K. N., Brewin, N. J., Borisov, A. Y. & Tikhonovich, I. A. (2009). Initiation of a legume nodule with an indeterminate meristem involves proliferating host cells that harbour infection threads. New Phytologist 181, 913-923.

Wayne, L. G., Brenner, D. J., Colwell, R. R., Grimont, P. A. D., Kandler, O., Krichevsky, M. I., Moore, L. H., Moore, W. E. C., Murray, R. G. E., Stackebrandt, E., Starr, M. P. & Trüper, H. G. (1987). Report of the ad hoc committee on reconciliation of approaches to bacterial systematics. International Journal of Systematic Bacteriology 37, 463-464.

Weir, B. S. (2006). Systematics, specificity, and ecology of New Zealand rhizobia. PhD Thesis: School of Biological Sciences. The University of Auckland.

244 Bibliography

Weir, B. S. (2011). The current taxonomy of rhizobia. New Zealand rhizobia website. http://www.rhizobia.co.nz/taxonomy/rhizobia.html.

Weon, H.-Y., Kwon, S.-W., Son, J.-A., Jo, E.-H., Kim, S.-J., Kim, Y.-S., Kim, B.- Y. & Ka, J.-O. (2010). Description of Microvirga aerophila sp. nov. and Microvirga aerilata sp. nov., isolated from air, reclassification of Balneimonas flocculans Takeda et al. 2004 as Microvirga flocculans comb. nov. and emended description of the genus Microvirga. International Journal of Systematic and Evolutionary Microbiology 60, 2596-2600.

Wernegreen, J. J. & Riley, M. A. (1999). Comparison of the evolutionary dynamics of symbiotic and housekeeping loci: A case for the genetic coherence of rhizobial lineages. Molecular Biology and Evolution 16, 98-113.

Wilkinson, H. H., Spoerke, J. M. & Parker, M. A. (1996). Divergence in symbiotic compatibility in a legume-Bradyrhizobium mutualism. Evolution 50, 1470- 1477.

Willems, A. (2006). The taxonomy of rhizobia: an overview. Plant and Soil 287, 3- 14.

Woese, C. R. (1987). Bacterial evolution. Microbiology and Molecular Biology Reviews 51, 221-271.

Wojciechowski, M. F., Sanderson, M. J., Steele, K. P. & Liston, A. (2000). Molecular phylogeny of the “temperate herbaceous tribes” of papilionoid legumes: a supertree approach. . In Advances in Legume Systematics, part 9, pp. 277-298. Edited by P. S. Herendeen & A. Bruneau. Richmond, Surrey, UK: Royal Botanic Gardens, Kew.

Wojciechowski, M. F., Lavin, M. & Sanderson, M. J. (2004). A phylogeny of legumes (Leguminosae) based on analyses of the plastid matK gene resolves many well-supported subclades within the family. American Journal of Botany 91, 1846- 1862.

Wolde-Meskel, E., Terefework, Z., Lindström, K. & Frostegård, A. (2004). Rhizobia nodulating African Acacia spp. and Sesbania sesban trees in southern Ethiopian soils are metabolically and genomically diverse. Soil Biology & Biochemistry 36, 2013-2025.

Wolde-Meskel, E., Terefework, Z., Frostegård, A. & Lindström, K. (2005). Genetic diversity and phylogeny of rhizobia isolated from agroforestry legume species in southern Ethiopia. International Journal of Systematic and Evolutionary Microbiology 55, 1439-1452.

Xu, L. M., Ge, C., Cui, Z., Li, J. & Fan, H. (1995). Bradyrhizobium liaoningense sp. nov., isolated from the root nodules of soybeans. International Journal of Systematic Bacteriology 45, 706-711.

245 Bibliography

Yanagi, M. & Yamasato, K. (1993). Phylogenetic analysis of the family Rhizobiaceae and related bacteria by sequencing of 16S rRNA gene using PCR and DNA sequencer. FEMS Microbiology Letters 107, 115-120.

Yang, J. K., Xie, F. L., Zou, J., Zhou, Q. & Zhou, J. C. (2005). Polyphasic characteristics of bradyrhizobia isolated from nodules of peanut (Arachis hypogaea) in China. Soil Biology and Biochemistry 37, 141-153.

Yates, R. J., Howieson, J. G., Nandasena, K. G. & O'Hara, G. W. (2004). Root- nodule bacteria from indigenous legumes in the north-west of Western Australia and their interaction with exotic legumes. Soil Biology & Biochemistry 36, 1319-1329.

Yates, R. J., Howieson, J. G., Reeve, W. G., Nandasena, K. G., Law, I. J., Bräu, L., Ardley, J. K., Nistelberger, H. M., Real, D. & O'Hara, G. W. (2007). Lotononis angolensis forms nitrogen fixing, lupinoid nodules with phylogenetically unique, fast-growing, pink-pigmented bacteria, which do not nodulate L. bainesii or L. listii. Soil Biology & Biochemistry 39, 1680-1688.

Young, J. P. W. & Johnston, A. W. B. (1989). The evolution of specificity in the legume-Rhizobium symbiosis Trends in Ecology & Evolution 4, 341-349.

Young, J. P. W. & Haukka, K. E. (1996). Diversity and phylogeny of rhizobia. New Phytologist 133, 87-94.

Young, J. P. W., Crossman, L. C., Johnston, A. W. B., Thomson, N. R., Ghazoui, Z. F., Hull, K. H., Wexler, M., Curson, A. R. J., Todd, J. D., Poole, P. S. & other authors (2006). The genome of Rhizobium leguminosarum has recognizable core and accessory components. Genome Biology 7.

Zahran, H. H. (1999). Rhizobium-legume symbiosis and nitrogen fixation under severe conditions and in an arid climate. Microbiology and Molecular Biology Reviews 63, 968-989.

Zahran, H. H. (2001). Rhizobia from wild legumes: diversity, taxonomy, ecology, nitrogen fixation and biotechnology. Journal of Biotechnology 91, 143-153.

Zhang, J., Song, F., Xin, Y. H., Zhang, J. & Fang, C. (2009). Microvirga guangxiensis sp. nov., a novel Alphaproteobacterium from soil, and emended description of the genus Microvirga. International Journal of Systematic and Evolutionary Microbiology 59, 1997-2001.

Zhao, L. F., Deng, Z. S., Yang, W. Q., Cao, Y., Wang, E. T. & Wei, G. H. (2010). Diverse rhizobia associated with Sophora alopecuroides grown in different regions of Loess Plateau in China. Systematic and Applied Microbiology 33, 468-477.

Zhou, J., Xia, B., Treves, D. S., Wu, L. Y., Marsh, T. L., O'Neill, R. V., Palumbo, A. V. & Tiedje, J. M. (2002). Spatial and resource factors influencing high microbial diversity in soil. Applied and Environmental Microbiology 68, 326- 334.

246 Bibliography

Zurdo-Piñeiro, J. L., García-Fraile, P., Rivas, R., Peix, A., León-Barrios, M., Willems, A., Mateos, P. F., Martínez-Molina, E., Velàzquez, E. & van Berkum, P. (2009). Rhizobia from Lanzarote, the Canary Islands, that nodulate Phaseolus vulgaris have characteristics in common with Sinorhizobium meliloti isolates from mainland Spain. Applied and Environmental Microbiology 75, 2354-2359.

247 Bibliography

248 Appendix

Appendix

249 Appendix

250 Appendix

Appendix I

1 10 20 30 40 50 60 70 M4-46 ATGCGCTTCAAGGGCCTCGATCTCAATCTTCTCGTTGCGCTCGACGCTTTGATGAAAGAGAGAAATCTCA Mn ORS2060 ATGCGCTTCAAGGGCCTCGATCTCAATCTCCTCATCGCGCTCGACGCTTTGATGAAAGAGCGAAACCTCA Mes loti ATGCGCTTCAAAGGCCTTGATCTGAATCTTCTCGTCGTGCTCGACGCACTGCTGACCGAGCGGACCCTCA Brs NC92 ATGCGGTTCAAGGGCCTTGATCTAAATCTTCTTGTTGCGCTCGATGCTTTGATGACCGAACGCAATCTCA Brj USDA 191 ATGCGTTTTAAGGGCCTTGATCTCAATCTCCTCGTTGCGCTCGACGCACTGATGACCGAACGCAAACTCA Bre USDA 94 ATGCGGTTCAAGGGACTTGATCTAAACCTTCTCGTTGCGCTCGATGCTCTAATGACGGAGCGCAACCTCA Rlp ATGCGGTTCAAGGGCCTTGATCTGAATCTCCTCGTCGCACTCGACGCCCTGATGACCGAGCGTAATCTCA NGR234 ATGCGTTTTAAGGGCCTTGATCTCAATCTCCTCGTTGCGCTCGACGCACTGATGACCGAACGCAAACTCA

71 80 90 100 110 120 130 140 M4-46 CCCGCGCGGCAAGCAGTCTCAACCTAAGCCAGCCCGCAATGAGCGCGGCTGTCGCGCGTTTGCGC Mn ORS2060 CCCGCGCGGCATGCAGCATCAACCTGAGCCAACCCGCCATGAGCGCGGCCGTCGCGCGTTTGCG Mes loti CGGCGGCGGCAAGCAGCATCAACCTTAGTCAGCCGGCCATGAGCGCGGCCGTCGCCCGGCTGCGC Brs NC92 CGGCGGCGGCACGCAAAATACACTTGAGCCAGCCGGCCATGAGTGCCGCCGTTGCCCGGCTACGC Brj USDA 191 CGGCCGCTGCACGCAGCATCAACCTGAGCCAGCCGGCGATGAGCGCAGCCATCACCCGGCTTCGG Bre USDA 94 CAGCGGCGGCGCGCCAAATTAACCTGAGCCAGCCCGCCATGAGCGCTGCGATCGCGCGGCTGCGC Rlp CGGCGGCAGCGCGCAGCATCAACTTGAGCCAACCGGCCATGAGCGCGGCCGTCGGCAGGTTACGC NGR234 CGGCCGCTGCACGCAGCATCAACCTGAGCCAGCCGGCGATGAGCGCAGCCATCACCCGGCTTCGG GTTGGATTCGGTCGGGCGTTA Forward primer f1 GATTCGGTCGGGCGTTACTCG Forward primer f2

Figure 1. Alignments of nodD nucleotide sequences from rhizobial strains, showing forward primer binding sites. M4-46 = Methylobacterium sp. 4-46; Mn = Methylobacterium nodulans; Ml = Mesorhizobium loti; Brs = Bradyrhizobium sp.; Brj = Bradyrhizobium japonicum; Bre = Bradyrhizobium elkanii; Rlp = Rhizobium leguminosarum bv phaseoli.

251 Appendix

1 10 20 30 40 50 60 70 M4-46 GTGCTGGCGGAATACGGCGTGCCGGCAACGTTCTGTGTCATTGGCGCGTACGCGGCAAGGGAGCCGCTGC Mn ORS2060 GTGCTGGCTGAACACCGGGTGCCGGCGACGTTCTGCGTCATTGGCGCGTACGCAGCAAATGAGCCTGAAC Bus STM678 GTGCTGGCGCAACACCGGGTGCCAGCGACTTTCTGCGTCATCGGCGCGTACGCAGCGGACCAGCCGAAGC Brs LMTR 25 GTGCTCGCGGAACACCGAGTGCCCGCAACGTTCTTCGTCATCGGCGCGTACGCGGCGGACCAGCCGAAAC Brs LMTR 21 GTGCTCGCGGAACACCGAGTGCCCGCAACGTTCTTCGTCATCGGCGCGTACGCGGCGGACCAGCCGAAAC GCAAGACACAGTAACCGCGC Reverse primer r1

71 80 90 100 110 120 130 M4-46 TAATCCGGCGCATGATCTCAGATGGGCATGAGGTGGCAAACCACACCATGACTCACCCGGAT Mn ORS2060 TAATCCAAAGAATGATCGCAGAAGGGCACGAAGTCGCAAATCACACGATGACTCATCCGGAC Bus STM678 TGATCCGACGGATGATCGCAGAAGGACACGAGGTCGCAAACCACTCGATGACTCATCCGGAC Brs LMTR 25 TAATCCGACGAATGATCGCAGAAGGGCACGAGGTCGCAAACCACACGATGACGCATCCGGAC Brs LMTR 21 TAATCCGACGAATGATCGCAGAAGGGCACGAGGTCGCAAACCACACGATGACGCATCCGGAC GTCTACCCGTACTCCACCGTT Reverse primer r2

Figure 2. Alignments of nodB nucleotide sequences from rhizobial strains, showing reverse primer binding sites. M4-46 = Methylobacterium sp. 4-46; Mn = Methylobacterium nodulans; Bus = Burkholderia sp; Brs = Bradyrhizobium sp.

252 Appendix

Appendix II

1 2 3 4 5 6 1 2 10.000 bp

3000 bp

1000 bp

a b

Figure 1. Gel electrophoresis showing molecular fingerprinting patterns of N2-fixing Li. angolensis strains, generated with ERIC primers. a) Lane 1, 1 kb DNA marker; lane 2, WSM3557; lane 3, WSM3673; lane 4, WSM3674; lane 5, WSM3675; lane 6, WSM3686; lane 7. b) Lane 1, 1 kb DNA marker; lane 2, WSM3693. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 10,000 bp

3000 bp

1000 bp

Figure 2. Gel electrophoresis showing molecular fingerprinting patterns of isolates generated with ERIC primers. Lanes 1, 7, 11 and 15: 1 kb DNA marker; lanes 2 – 6, WSM2596 and reisolates from Lo. delicata and Le. polycephala; lanes 8 – 10, WSM2624 and reisolates from Le. mollis; lanes 12 – 14, WSM3040 and reisolates from Le. polycephala.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 10,000 bp

3000 bp

1000 bp

Figure 1b. Gel electrophoresis showing molecular fingerprinting patterns of isolates generated with ERIC primers. Lanes 1, 7, 13 and 19, 1 kb DNA marker; lanes 2 – 6, WSM2632 and reisolates from Lo. delicata and Le. polycephala; lanes 8 – 12, WSM2667 and reisolates from Lo. delicata and Le. mollis; lanes 14 – 18, WSM2783 and reisolates from Lo. delicata and Le. polycephala; lanes 20 – 25, WSM2653 and reisolates from Le. mollis and Le. polycephala; lane 26, no DNA template.

253 Appendix

Appendix III

L. angolensis

16 0.06

14 0.05 12 0.04 10

8 0.03

6 0.02 4 Mean nodules per plant

0.01 Shoot dry weight (g/plant) 2

0 0 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc N+ Inoculation treatments

Figure 1. Symbiotic ability of rhizobia associated with Lotononis s. l. on Listia angolensis, assessed by nodule number ( ) and dry weight of shoots ( ) of L. angolensis plants harvested after ten weeks growth.

L. bainesii

14 0.08

12 0.07

0.06 10 0.05 8 0.04 6 0.03 4 0.02 Mean nodules per plant

Shoot dry weight (g/plant) 2 0.01

0 0.00 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc N+ Inoculation treatment

Figure 2. Symbiotic ability of rhizobia associated with Lotononis s. l. on Listia bainesii, assessed by nodule number ( ) and dry weight of shoots ( ) of L. bainesii plants harvested after ten weeks growth.

254 Appendix

L. bolusii

30 0.08

0.07 25 0.06 20 0.05

15 0.04

0.03 10 0.02

Mean nodules5 per plant

0.01 Shoot dry weight (g/plant)

0 0.00 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc N+ Inoculation treatment

Figure 3. Symbiotic ability of rhizobia associated with Lotononis s. l. on

Leobordea bolusii, assessed by nodule number ( ) and dry weight of shoots ( ) of L. bolusii plants harvested after ten weeks growth.

L. longiflora

12 0.012

10 0.010

8 0.008

6 0.006

4 0.004

Mean nodules per plant 2 0.002 Shoot dry weight (g/plant)

0 0.000 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc N+ Inoculation treatment

Figure 4. Symbiotic ability of rhizobia associated with Lotononis s. l. on Leobordea longiflora, assessed by nodule number ( ) and dry weight of shoots ( ) of L. longiflora plants harvested after ten weeks growth.

255 Appendix

L. platycarpa

9 0.35

8 0.30 7 0.25 6

5 0.20

4 0.15 3 0.10

Mean nodules per Mean nodules plant 2 Shoot dry Shoot weight (g/plant) 0.05 1

0 0.00 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc N+ Inoculation treatment

Figure 5a. Symbiotic ability of rhizobia associated with Lotononis s. l. on

Leobordea platycarpa, assessed by nodule number ( ) and dry weight of L. platycarpa shoots ( ) of plants harvested after ten weeks growth.

L. platycarpa nodulated WSM3557 minus N+

9 0.035

8 0.030 7 0.025 6 0.020 5 4 0.015 3 0.010 2 Mean nodules per plant

0.005 Shoot dry weight ( g/plant) 1

0 0.000 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc Inoculation treatment

Figure 5b. Symbiotic ability of rhizobia associated with Lotononis s. l. on

Leobordea platycarpa, assessed by nodule number ( ) and dry weight of shoots ( ) of L. platycarpa plants harvested after ten weeks growth. The graph has

been rescaled by removing the N+ treatment to reveal the rhizobial inoculant response. Non-nodulated plants were removed from the WSM3557 treatment data.

256 Appendix

L. stipulosa

45 0.50 40 0.45 35 0.40 0.35 30 0.30 25 0.25 20 0.20 15 0.15 10 0.10 Mean nodules per plant 5 0.05 Shoot dry weight (g/plant) 0 0.00 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 N+ Inoculation treatment

Figure 6a. Symbiotic ability of rhizobia associated with Lotononis s. l. on Leobordea stipulosa, assessed by nodule number ( ) and dry weight of shoots ( ) of L. stipulosa plants harvested after ten weeks growth. No uninoculated control is included due to seedling death.

L. stipulosa minus N+

45 0.005 40 35 0.004

30 0.003 25

20 0.002 15

Mean nodules 10per plant 0.001 Shoot dry weight (g/plant) 5

0 0.000 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Inoculation treatment

Figure 6b. Symbiotic ability of rhizobia associated with Lotononis s. l. on Leobordea stipulosa, assessed by nodule number ( ) and dry weight of shoots ( ) of L. stipulosa plants harvested after ten weeks growth. No uninoculated control is included due to seedling death. The graph has been rescaled by removing the N+ treatment to reveal the rhizobial inoculant response.

257 Appendix

L. crumanina

6.0 0.50 0.45 5.0 0.40

0.35 4.0 0.30

3.0 0.25

0.20 2.0 0.15

Mean nodules per plant 0.10 1.0 Shoot dry weight (g/plant ) 0.05

0.0 0.00 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc N+ Inoculation treatment

Figure 7a. Symbiotic ability of rhizobia associated with Lotononis s. l. on Lotononis crumanina, assessed by nodule number ( ) and dry weight of shoots ( ) of L. crumanina plants harvested after ten weeks growth.

L. crumanina minus N+ control

6 0.016

0.014 5 0.012 4 0.010

3 0.008

0.006 2 0.004 Mean nodules Mean per nodules plant 1 dry Shoot weight (g/plant) 0.002

0 0.000 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc Inoculation treatment

Figure 7b. Symbiotic ability of rhizobia associated with Lotononis s. l. on Lotononis crumanina, assessed by nodule number ( ) and dry weight of shoots ( ) of L. crumanina plants harvested after ten weeks growth. The graph has been rescaled by removing the N+ treatment to reveal the rhizobial inoculant response.

258 Appendix

L. falcata

9 0.16

8 0.14 7 0.12 6 0.10 5 0.08 4 0.06 3 0.04

Mean nodules per Mean plant nodules 2 Shoot dry Shoot weight (g/plant) 1 0.02

0 0.00 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc N+ Inoculation treatment

Figure 8a. Symbiotic ability of rhizobia associated with Lotononis s. l. on Lotononis falcata, assessed by nodule number ( ) and dry weight of shoots ( ) of L. falcata plants harvested after ten weeks growth.

L. falcata minus N+

9 0.005 8 7 0.004

6 0.003 5

4 0.002 3

Mean nodules per Mean plant nodules 2

0.001 dry Shoot weight (g/plant) 1

0 0.000 WSM2596 WSM2598 WSM2624 WSM2632 WSM2653 WSM2667 WSM2783 WSM3040 WSM3557 Uninoc Inoculation treatment

Figure 8b. Symbiotic ability of rhizobia associated with Lotononis s. l. on Lotononis falcata, assessed by nodule number ( ) and dry weight of shoots ( ) of L. falcata plants harvested after ten weeks growth. The graph has been rescaled by removing the N+ treatment to reveal the rhizobial inoculant response.

259 Appendix

Appendix IV

Vitamin Stock (1000x) From Egli & Auling (2005)

Pyridoxine HCl (Vitamin B6) 100 mg l-1

Thiamine HCl (Vitamin B1) 50 mg l-1

Riboflavin (Vitamin B2) 50 mg l-1

Nicotinic acid (Niacin) (Vitamin B3) 50 mg l-1

D calcium pantothenate (Vitamin B5) 50 mg l-1 p aminobenzoic acid (Vitamin Bx) 50 mg l-1

Lipoic acid 50 mg l-1

Nicotinamide (Vitamin B3 amide) 50 mg l-1

Cobalamin (Vitamin B12) 50 mg l-1

Biotin (Vitamin B7) 20 mg l-1

Folic acid (Vitamin B9) 20 mg l-1

260 Appendix

Appendix V

Figure 1. Growth of novel rhizobial strains of Microvirga at varying concentrations of NaCl (0.01 – 2.0% (w/v)). Results are shown for Listia angolensis strains only.

261 Appendix

262 Appendix

Appendix VI Table 1. Substrate utilisation pattern of WSM3557T and Lut6T in Biolog PM plates compared with Sinorhizobium meliloti 1021 (Sm 1021; data from Biondi et al. (2009)) and compared with utilisation in GN2 plates and minimal media (MM). Green = +ve; yellow = weak +ve; red = -ve.

PM Well Tested Compound Type of compound S.m 1021 WSM3557 GN2 WSM3557 MM WSM3557 Lut6 GN2 Lut6 MM Lut6 PM1 Carbon A01 Sources-1 Negative Control PM1 Carbon A02 Sources-2 L-Arabinose C source; C5 carbohydrate PM1 Carbon A03 Sources-3 N-Acetyl-D-Glucosamine PM1 Carbon A04 Sources-4 D-Saccharic Acid C-Source, carboxylic acid PM1 Carbon A05 Sources-5 Succinic Acid C-Source, carboxylic acid PM1 Carbon A06 Sources-6 D-Galactose PM1 Carbon A07 Sources-7 L-Aspartic Acid C-Source, amino acid PM1 Carbon A08 Sources-8 L-Proline PM1 Carbon A09 Sources-9 D-Alanine PM1 Carbon A10 Sources-10 D-Trehalose PM1 Carbon A11 Sources-11 D-Mannose C source; C6 carbohydrate PM1 Carbon A12 Sources-12 Dulcitol PM1 Carbon B01 Sources-13 D-Serine PM1 Carbon B02 Sources-14 D-Sorbitol PM1 Carbon B03 Sources-15 Glycerol PM1 Carbon B04 Sources-16 L-Fucose PM1 Carbon B05 Sources-17 D-Glucuronic Acid PM1 Carbon B06 Sources-18 D-Gluconic Acid PM1 Carbon B07 Sources-19 D,L-α-Glycerol-Phosphate PM1 Carbon B08 Sources-20 D-Xylose C source; C5 carbohydrate PM1 Carbon B09 Sources-21 L-Lactic Acid C-Source, C3 carboxylic acid PM1 Carbon B10 Sources-22 Formic Acid PM1 Carbon B11 Sources-23 D-Mannitol C-source, C6 sugar alcohol PM1 Carbon B12 Sources-24 L-Glutamic Acid C-Source, amino acid

263 Table 1. (cont).

PM1 Carbon C01 Sources-25 D-Glucose-6-Phosphate PM1 Carbon C02 Sources-26 D-Galactonic Acid-γ-Lactone PM1 Carbon C03 Sources-27 D,L-Malic Acid C-Source, carboxylic acid PM1 Carbon C04 Sources-28 D-Ribose C source; C5 carbohydrate PM1 Carbon C05 Sources-29 Tween 20 C-Source, fatty acid PM1 Carbon C06 Sources-30 L-Rhamnose C source; C6 carbohydrate (methyl pentose) PM1 Carbon C07 Sources-31 D-Fructose C source; C6 carbohydrate PM1 Carbon C08 Sources-32 Acetic Acid PM1 Carbon C09 Sources-33α-D-Glucose C source; C6 carbohydrate PM1 Carbon C10 Sources-34 Maltose PM1 Carbon C11 Sources-35 D-Melibiose PM1 Carbon C12 Sources-36 Thymidine PM1 Carbon D01 Sources-37 L-Asparagine PM1 Carbon D02 Sources-38 D-Aspartic Acid PM1 Carbon D03 Sources-39 D-Glucosaminic Acid PM1 Carbon D04 Sources-40 1,2-Propanediol PM1 Carbon D05 Sources-41 Tween 40 PM1 Carbon D06 Sources-42α-Keto-Glutaric Acid C-Source, carboxylic acid PM1 Carbon D07 Sources-43α-Keto-Butyric Acid PM1 Carbon D08 Sources-44α-Methyl-DGalactoside PM1 Carbon D09 Sources-45α-D-Lactose PM1 Carbon D10 Sources-46 Lactulose PM1 Carbon D11 Sources-47 Sucrose PM1 Carbon D12 Sources-48 Uridine PM1 Carbon E01 Sources-49 L-Glutamine PM1 Carbon E02 Sources-50 M-Tartaric Acid PM1 Carbon E03 Sources-51 D-Glucose-1-Phosphate PM1 Carbon E04 Sources-52 D-Fructose-6-Phosphate C source; C6 carbohydrate

264 Appendix

Table 1. (cont).

PM1 Carbon E05 Sources-53 Tween 80 PM1 Carbon E06 Sources-54α-Hydroxy Glutaric Acid-γ-Lactone PM1 Carbon E07 Sources-55α-Hydroxy Butyric Acid PM1 Carbon E08 Sources-56β-Methyl-DGlucoside PM1 Carbon E09 Sources-57 Adonitol (= Ribitol) C5 alcohol PM1 Carbon E10 Sources-58 Maltotriose PM1 Carbon E11 Sources-59 2-Deoxy Adenosine PM1 Carbon E12 Sources-60 Adenosine PM1 Carbon F01 Sources-61 Glycyl-L-Aspartic Acid PM1 Carbon F02 Sources-62 Citric Acid PM1 Carbon F03 Sources-63 m-Inositol C source; C6 carbohydrate (polyol of cyclohexane) PM1 Carbon F04 Sources-64 D-Threonine PM1 Carbon F05 Sources-65 Fumaric Acid C-Source, carboxylic acid PM1 Carbon F06 Sources-66 Bromo Succinic Acid C-Source, carboxylic acid PM1 Carbon F07 Sources-67 Propionic Acid PM1 Carbon F08 Sources-68 Mucic Acid PM1 Carbon F09 Sources-69 Glycolic Acid PM1 Carbon F10 Sources-70 Glyoxylic Acid PM1 Carbon F11 Sources-71 D-Cellobiose C source; C12 carbohydrate PM1 Carbon F12 Sources-72 Inosine PM1 Carbon G01 Sources-73 Glycyl-LGlutamic Acid PM1 Carbon G02 Sources-74 Tricarballylic Acid PM1 Carbon G03 Sources-75 L-Serine C-Source, amino acid PM1 Carbon G04 Sources-76 L-Threonine PM1 Carbon G05 Sources-77 L-Alanine PM1 Carbon G06 Sources-78 L-Alanyl-Glycine PM1 Carbon G07 Sources-79 Acetoacetic Acid PM1 Carbon G08 Sources-80 N-Acetyl-β-DMannosamine

265 Table 1. (cont).

PM1 Carbon G09 Sources-81 Mono Methyl Succinate C-Source, carboxylic acid PM1 Carbon G10 Sources-82 Methyl Pyruvate C-Source, carboxylic acid PM1 Carbon G11 Sources-83 D-Malic Acid C-Source, carboxylic acid PM1 Carbon G12 Sources-84 L-Malic Acid C-Source, carboxylic acid PM1 Carbon H01 Sources-85 Glycyl-L-Proline PM1 Carbon H02 Sources-86 p-Hydroxy Phenyl Acetic Acid PM1 Carbon H03 Sources-87 m-Hydroxy Phenyl Acetic Acid PM1 Carbon H04 Sources-88 Tyramine PM1 Carbon H05 Sources-89 D-Psicose C source; C6 carbohydrate PM1 Carbon H06 Sources-90 L-Lyxose C source; C5 carbohydrate PM1 Carbon H07 Sources-91 Glucuronamide C-Source, C6 amide PM1 Carbon H08 Sources-92 Pyruvic Acid C-Source, carboxylic acid PM1 Carbon H09 Sources-93 Dihydroxyacetone C-Source, C3 alcohol PM1 Carbon H10 Sources-94 D-Galacturonic Acid C-Source, C6 carboxylic acid PM1 Carbon H11 Sources-95 Phenylethylamine PM1 Carbon H12 Sources-96 2-Aminoethanol PM2 Carbon A01 Sources-2 Negative Control PM2 Carbon A02 Sources-3 Chondroitin Sulfate C PM2 Carbon A03 Sources-4α-Cyclodextrin PM2 Carbon A04 Sources-5β-Cyclodextrin PM2 Carbon A05 Sources-6γ-Cyclodextrin PM2 Carbon A06 Sources-7 Dextrin PM2 Carbon A07 Sources-8 Gelatin PM2 Carbon A08 Sources-9 Glycogen PM2 Carbon A09 Sources-10 Inulin PM2 Carbon A10 Sources-11 Laminarin PM2 Carbon A11 Sources-12 Mannan PM2 Carbon A12 Sources-13 Pectin C-Source, polymer

266 Appendix

Table 1. (cont).

PM2 Carbon B01 Sources-14 N-Acetyl-DGalactosamine PM2 Carbon B02 Sources-15 N-Acetyl-Neuraminic Acid PM2 Carbon B03 Sources-16β-D-Allose PM2 Carbon B04 Sources-17 Amygdalin PM2 Carbon B05 Sources-18 D-Arabinose C source; C5 carbohydrate PM2 Carbon B06 Sources-19 D-Arabitol PM2 Carbon B07 Sources-20 L-Arabitol PM2 Carbon B08 Sources-21 Arbutin (glucosylated hydroquinone) C source; C12 carbohydrate; glycosylated hydroquinone PM2 Carbon B09 Sources-22 2-Deoxy-D-Ribose C source; C5 carbohydrate PM2 Carbon B10 Sources-23 I-Erythritol PM2 Carbon B11 Sources-24 D-Fucose PM2 Carbon B12 Sources-25 3-0-β-D-Galactopyranosyl-DArabinose PM2 Carbon C01 Sources-26 Gentiobiose PM2 Carbon C02 Sources-27 L-Glucose PM2 Carbon C03 Sources-28 Lactitol PM2 Carbon C04 Sources-29 D-Melezitose PM2 Carbon C05 Sources-30 Maltitol PM2 Carbon C06 Sources-31 a-Methyl-DGlucoside PM2 Carbon C07 Sources-32β-Methyl-DGalactoside PM2 Carbon C08 Sources-33 3-Methyl Glucose PM2 Carbon C09 Sources-34β-Methyl-DGlucuronic Acid PM2 Carbon C10 Sources-35α-Methyl-DMannoside PM2 Carbon C11 Sources-36β-Methyl-DXyloside PM2 Carbon C12 Sources-37 Palatinose (isomaltulose) C source; C12 carbohydrate PM2 Carbon D01 Sources-38 D-Raffinose PM2 Carbon D02 Sources-39 Salicin PM2 Carbon D03 Sources-40 Sedoheptulosan PM2 Carbon D04 Sources-41 L-Sorbose

267 Table 1. (cont).

PM2 Carbon D05 Sources-42 Stachyose PM2 Carbon D06 Sources-43 D-Tagatose C source; C6 carbohydrate PM2 Carbon D07 Sources-44 Turanose PM2 Carbon D08 Sources-45 Xylitol PM2 Carbon D09 Sources-46 N-Acetyl-DGlucosaminitol PM2 Carbon D10 Sources-47γ-Amino Butyric Acid PM2 Carbon D11 Sources-48δ-Amino Valeric Acid PM2 Carbon D12 Sources-49 Butyric Acid PM2 Carbon E01 Sources-50 Capric Acid PM2 Carbon E02 Sources-51 Caproic Acid PM2 Carbon E03 Sources-52 Citraconic Acid PM2 Carbon E04 Sources-53 Citramalic Acid PM2 Carbon E05 Sources-54 D-Glucosamine C source; C6 carbohydrate (amino sugar) PM2 Carbon E06 Sources-55 2-Hydroxy Benzoic Acid PM2 Carbon E07 Sources-56 4-Hydroxy Benzoic Acid PM2 Carbon E08 Sources-57β-Hydroxy Butyric Acid PM2 Carbon E09 Sources-58γ-Hydroxy Butyric Acid PM2 Carbon E10 Sources-59 a-Keto-Valeric Acid PM2 Carbon E11 Sources-60 Itaconic Acid PM2 Carbon E12 Sources-61 5-Keto-DGluconic Acid C-Source, carboxylic acid PM2 Carbon F01 Sources-62 D-Lactic Acid Methyl Ester PM2 Carbon F02 Sources-63 Malonic Acid C-Source, carboxylic acid PM2 Carbon F03 Sources-64 Melibionic Acid PM2 Carbon F04 Sources-65 Oxalic Acid PM2 Carbon F05 Sources-66 Oxalomalic Acid C-Source, C6 carboxylic acid PM2 Carbon F06 Sources-67 Quinic Acid PM2 Carbon F07 Sources-68 D-Ribono-1,4-Lactone PM2 Carbon F08 Sources-69 Sebacic Acid

268 Appendix

Table 1. (cont).

PM2 Carbon F09 Sources-70 Sorbic Acid PM2 Carbon F10 Sources-71 Succinamic Acid C-Source, C4 carboxylic acid PM2 Carbon F11 Sources-72 D-Tartaric Acid PM2 Carbon F12 Sources-73 L-Tartaric Acid PM2 Carbon G01 Sources-74 Acetamide PM2 Carbon G02 Sources-75 L-Alaninamide PM2 Carbon G03 Sources-76 N-Acetyl-LGlutamic Acid PM2 Carbon G04 Sources-77 L-Arginine PM2 Carbon G05 Sources-78 Glycine PM2 Carbon G06 Sources-79 L-Histidine PM2 Carbon G07 Sources-80 L-Homoserine PM2 Carbon G08 Sources-81 Hydroxy-LProline PM2 Carbon G09 Sources-82 L-Isoleucine PM2 Carbon G10 Sources-83 L-Leucine PM2 Carbon G11 Sources-84 L-Lysine PM2 Carbon G12 Sources-85 L-Methionine PM2 Carbon H01 Sources-86 L-Ornithine PM2 Carbon H02 Sources-87 L-Phenylalanine PM2 Carbon H03 Sources-88 L-Pyroglutamic Acid PM2 Carbon H04 Sources-89 L-Valine PM2 Carbon H05 Sources-90 D,L-Carnitine PM2 Carbon H06 Sources-91 Sec-Butylamine PM2 Carbon H07 Sources-92 D.L-Octopamine PM2 Carbon H08 Sources-93 Putrescine PM2 Carbon H09 Sources-94 Dihydroxy Acetone C-Source, alcohol (C3 ketose) PM2 Carbon H10 Sources-95 2,3-Butanediol PM2 Carbon H11 Sources-96 2,3-Butanone PM2 Carbon H12 Sources-97 3-Hydroxy 2-Butanone

269 Table 1. (cont).

PM3 Nitrogen A01 NegativeSources Control PM3 Nitrogen A02 AmmoniaSources N-Source, inorganic PM3 Nitrogen A03 NitriteSources N-Source, inorganic PM3 Nitrogen A04 NitrateSources N-Source, inorganic PM3 Nitrogen A05 UreaSources N-Source, inorganic PM3 Nitrogen A06 BiuretSources PM3 Nitrogen A07 L-AlanineSources N-Source, amino acid PM3 Nitrogen A08 L-ArginineSources N-Source, amino acid PM3 Nitrogen A09 L-AsparagineSources N-Source, amino acid PM3 Nitrogen A10 L-AsparticSources Acid N-Source, amino acid PM3 Nitrogen A11 L-CysteineSources N-Source, amino acid PM3 Nitrogen A12 L-GlutamicSources Acid N-Source, amino acid PM3 Nitrogen B01 L-GlutamineSources N-Source, amino acid PM3 Nitrogen B02 GlycineSources N-Source, amino acid PM3 Nitrogen B03 L-HistidineSources N-Source, amino acid PM3 Nitrogen B04 L-IsoleucineSources N-Source, amino acid PM3 Nitrogen B05 L-LeucineSources N-Source, amino acid PM3 Nitrogen B06 L-LysineSources N-Source, amino acid PM3 Nitrogen B07 L-MethionineSources N-Source, amino acid PM3 Nitrogen B08 L-PhenylalanineSources N-Source, amino acid PM3 Nitrogen B09 L-ProlineSources N-Source, amino acid PM3 Nitrogen B10 L-SerineSources N-Source, amino acid PM3 Nitrogen B11 L-ThreonineSources N-Source, amino acid PM3 Nitrogen B12 L-TryptophanSources N-Source, amino acid PM3 Nitrogen C01 L-TyrosineSources N-Source, amino acid PM3 Nitrogen C02 L-ValineSources N-Source, amino acid PM3 Nitrogen C03 D-AlanineSources N-Source, amino acid PM3 Nitrogen C04 D-AsparagineSources N-Source, amino acid

270 Appendix

Table 1. (cont). PM3 Nitrogen C05 D-AsparticSources Acid N-Source, amino acid PM3 Nitrogen C06 D-GlutamicSources Acid N-Source, amino acid PM3 Nitrogen C07 D-LysineSources N-Source, amino acid PM3 Nitrogen C08 D-SerineSources N-Source, amino acid PM3 Nitrogen C09 D-ValineSources N-Source, amino acid PM3 Nitrogen C10 L-CitrullineSources N-Source, amino acid PM3 Nitrogen C11 L-HomoserineSources N-Source, amino acid PM3 Nitrogen C12 L-OrnithineSources N-Source, amino acid PM3 Nitrogen D01 N-Acetyl-D,LGlutamicSources Acid N-Source, amino acid PM3 Nitrogen D02 N-Phthaloyl-LGlutamicSources Acid N-Source, amino acid PM3 Nitrogen D03 L-PyroglutamicSources Acid N-Source, amino acid PM3 Nitrogen D04 HydroxylamineSources PM3 Nitrogen D05 MethylamineSources PM3 Nitrogen D06 N-AmylamineSources PM3 Nitrogen D07 N-ButylamineSources PM3 Nitrogen D08 EthylamineSources PM3 Nitrogen D09 EthanolamineSources PM3 Nitrogen D10 EthylenediamineSources PM3 Nitrogen D11 PutrescineSources PM3 Nitrogen D12 AgmatineSources PM3 Nitrogen E01 HistamineSources PM3 Nitrogen E02 βSources-Phenylethylamine PM3 Nitrogen E03 TyramineSources PM3 Nitrogen E04 AcetamideSources PM3 Nitrogen E05 FormamideSources PM3 Nitrogen E06 GlucuronamideSources PM3 Nitrogen E07 D,L-LactamideSources PM3 Nitrogen E08 D-GlucosamineSources N-Source, other

271 Table 1. (cont).

PM3 Nitrogen E09 D-GalactosamineSources N-Source, other PM3 Nitrogen E10 D-MannosamineSources N-Source, other PM3 Nitrogen E11 N-Acetyl-DGlucosamineSources N-Source, other PM3 Nitrogen E12 N-Acetyl-DGalactosamineSources PM3 Nitrogen F01 N-Acetyl-DMannosamineSources PM3 Nitrogen F02 AdenineSources PM3 Nitrogen F03 AdenosineSources PM3 Nitrogen F04 CytidineSources PM3 Nitrogen F05 CytosineSources PM3 Nitrogen F06 GuanineSources PM3 Nitrogen F07 GuanosineSources PM3 Nitrogen F08 ThymineSources PM3 Nitrogen F09 ThymidineSources PM3 Nitrogen F10 UracilSources PM3 Nitrogen F11 UridineSources PM3 Nitrogen F12 InosineSources PM3 Nitrogen G01 XanthineSources N-Source, other PM3 Nitrogen G02 XanthosineSources PM3 Nitrogen G03 UricSources Acid PM3 Nitrogen G04 AlloxanSources N-Source, other PM3 Nitrogen G05 AllantoinSources N-Source, other PM3 Nitrogen G06 ParabanicSources Acid PM3 Nitrogen G07 D,L-Sourcesα-Amino-NButyric Acid PM3 Nitrogen G08 γSources-Amino-NButyric Acid N-Source, other PM3 Nitrogen G09 εSources-Amino-NCaproic Acid PM3 Nitrogen G10 D,L-Sourcesα-Amino-Caprylic Acid N-Source, other PM3 Nitrogen G11 δSources-Amino-NValeric Acid N-Source, other PM3 Nitrogen G12 αSources-Amino-NValeric Acid

272 Appendix

Table 1. (cont).

PM3 Nitrogen H01 Ala-AspSources N-Source, peptide PM3 Nitrogen H02 Ala-GlnSources N-Source, peptide PM3 Nitrogen H03 Ala-GluSources N-Source, peptide PM3 Nitrogen H04 Ala-GlySources N-Source, peptide PM3 Nitrogen H05 Ala-HisSources N-Source, peptide PM3 Nitrogen H06 Ala-LeuSources N-Source, peptide PM3 Nitrogen H07 Ala-ThrSources N-Source, peptide PM3 Nitrogen H08 Gly-AsnSources N-Source, peptide PM3 Nitrogen H09 Gly-GlnSources N-Source, peptide PM3 Nitrogen H10 Gly-GluSources N-Source, peptide PM3 Nitrogen H11 Gly-MetSources N-Source, peptide PM3 Nitrogen H12 Met-AlaSources N-Source, peptide PM4 Phosphorus A01 Negative and Sulfur Control Sources PM4 Phosphorus A02 Phosphate and Sulfur Sources PM4 Phosphorus A03 Pyrophosphate and Sulfur Sources PM4 Phosphorus A04 Trimetaphosphate and Sulfur Sources PM4 Phosphorus A05 Tripolyphosphate and Sulfur Sources PM4 Phosphorus A06 Triethyl and PhosphateSulfur Sources PM4 Phosphorus A07 Hypophosphite and Sulfur Sources PM4 Phosphorus A08 Adenosine- and Sulfur 2’-monophosphate Sources P-Source, organic PM4 Phosphorus A09 Adenosine- and Sulfur 3’-monophosphate Sources P-Source, organic PM4 Phosphorus A10 Adenosine- and Sulfur 5’-monophosphate Sources P-Source, organic PM4 Phosphorus A11 Adenosine- and Sulfur 2’,3’-cyclic Sources monophosphate P-Source, organic PM4 Phosphorus A12 Adenosine- and Sulfur 3’,5’-cyclic Sources monophosphate P-Source, organic PM4 Phosphorus B01 Thiophosphate and Sulfur Sources P-Source, inorganic PM4 Phosphorus B02 Dithiophosphate and Sulfur Sources P-Source, inorganic PM4 Phosphorus B03 D,L- andα-Glycerol Sulfur SourcesPhosphate P-Source, organic PM4 Phosphorus B04 β-Glycerol and Sulfur Phosphate Sources P-Source, organic

273 Table 1. (cont).

PM4 Phosphorus B05 Carbamyl and Sulfur Phosphate Sources P-Source, organic PM4 Phosphorus B06 D-2-Phospho-Glyceric and Sulfur Sources Acid P-Source, organic PM4 Phosphorus B07 D-3-Phospho-Glyceric and Sulfur Sources Acid P-Source, organic PM4 Phosphorus B08 Guanosine- and Sulfur 2’-monophaphate Sources P-Source, organic PM4 Phosphorus B09 Guanosine- and Sulfur 3’-monophosphate Sources P-Source, organic PM4 Phosphorus B10 Guanosine- and Sulfur 5’-monophosphate Sources P-Source, organic PM4 Phosphorus B11 Guanosine- and Sulfur 2’,3’-cyclic Sources monophosphate P-Source, organic PM4 Phosphorus B12 Guanosine- and Sulfur 3’,5’-cyclic Sources monophosphate P-Source, organic PM4 Phosphorus C01 Phosphoenol and Sulfur Pyruvate Sources P-Source, organic PM4 Phosphorus C02 Phospho-Glycolic and Sulfur Sources Acid P-Source, organic PM4 Phosphorus C03 D-Glucose-1-Phosphate and Sulfur Sources P-Source, organic PM4 Phosphorus C04 D-Glucose-6-Phosphate and Sulfur Sources P-Source, organic PM4 Phosphorus C05 2-Deoxy-DGlucose and Sulfur Sources 6-Phosphate P-Source, organic PM4 Phosphorus C06 D-Glucosamine-6-Phosphate and Sulfur Sources P-Source, organic PM4 Phosphorus C07 6-Phospho-Gluconic and Sulfur Sources Acid P-Source, organic PM4 Phosphorus C08 Cytidine- and Sulfur 2’-monophosphate Sources P-Source, organic PM4 Phosphorus C09 Cytidine- and Sulfur 3’-monophosphate Sources P-Source, organic PM4 Phosphorus C10 Cytidine- and Sulfur 5’-monophosphate Sources P-Source, organic PM4 Phosphorus C11 Cytidine- and Sulfur 2’,3’-cyclic Sources monophosphate P-Source, organic PM4 Phosphorus C12 Cytidine- and Sulfur 3’,5’-cyclic Sources monophosphate P-Source, organic PM4 Phosphorus D01 D-Mannose-1-Phosphate and Sulfur Sources P-Source, organic PM4 Phosphorus D02 D-Mannose-6-Phosphate and Sulfur Sources P-Source, organic PM4 Phosphorus D03 Cysteamine-S and Sulfur PhosphateSources P-Source, organic PM4 Phosphorus D04 Phospho-LArginine and Sulfur Sources P-Source, organic PM4 Phosphorus D05 o-Phospho-DSerine and Sulfur Sources P-Source, organic PM4 Phosphorus D06 o-Phospho-LSerine and Sulfur Sources P-Source, organic PM4 Phosphorus D07 o-Phospho-LThreonine and Sulfur Sources P-Source, organic PM4 Phosphorus D08 Uridine- and Sulfur2’-monophosphate Sources P-Source, organic

274 Appendix

Table 1. (cont). PM4 Phosphorus D09 Uridine- and Sulfur3’-monophosphate Sources P-Source, organic PM4 Phosphorus D10 Uridine- and Sulfur5’-monophosphate Sources P-Source, organic PM4 Phosphorus D11 Uridine- and Sulfur2’,3’-cyclic Sources monophosphate P-Source, organic PM4 Phosphorus D12 Uridine- and Sulfur3’,5’-cyclic Sources monophosphate P-Source, organic PM4 Phosphorus E01 O-Phospho-DTyrosine and Sulfur Sources P-Source, organic PM4 Phosphorus E02 O-Phospho-LTyrosine and Sulfur Sources P-Source, organic PM4 Phosphorus E03 Phosphocreatine and Sulfur Sources P-Source, organic PM4 Phosphorus E04 Phosphoryl and Sulfur Choline Sources P-Source, organic PM4 Phosphorus E05 O-Phosphoryl-Ethanolamine and Sulfur Sources P-Source, organic PM4 Phosphorus E06 Phosphono and Sulfur Acetic Sources Acid P-Source, organic PM4 Phosphorus E07 2-Aminoethyl and Sulfur Phosphonic Sources Acid P-Source, organic PM4 Phosphorus E08 Methylene and Sulfur Diphosphonic Sources Acid P-Source, organic PM4 Phosphorus E09 Thymidine- and Sulfur 3’-monophosphate Sources P-Source, organic PM4 Phosphorus E10 Thymidine- and Sulfur 5’-monophosphate Sources P-Source, organic PM4 Phosphorus E11 Inositol and HexaphosphateSulfur Sources P-Source, organic PM4 Phosphorus E12 Thymidine and Sulfur 3’,5’-cyclic Sources monophosphate P-Source, organic PM4 Phosphorus F01 Negative and Sulfur Control Sources PM4 Phosphorus F02 Sulfate and Sulfur Sources PM4 Phosphorus F03 Thiosulfate and Sulfur Sources PM4 Phosphorus F04 Tetrathionate and Sulfur Sources PM4 Phosphorus F05 Thiophosphate and Sulfur Sources PM4 Phosphorus F06 Dithiophosphate and Sulfur Sources PM4 Phosphorus F07 L-Cysteine and Sulfur Sources PM4 Phosphorus F08 D-Cysteine and Sulfur Sources S-Source, organic PM4 Phosphorus F09 L-Cysteinyl-Glycine and Sulfur Sources PM4 Phosphorus F10 L-Cysteic and Sulfur Acid Sources PM4 Phosphorus F11 Cysteamine and Sulfur Sources PM4 Phosphorus F12 L-Cysteine and Sulfur Sulfinic Sources Acid

275 Table 1. (cont).

PM4 Phosphorus G01 N-Acetyl-LCysteine and Sulfur Sources PM4 Phosphorus G02 S-Methyl-LCysteine and Sulfur Sources PM4 Phosphorus G03 Cystathionine and Sulfur Sources PM4 Phosphorus G04 Lanthionine and Sulfur Sources PM4 Phosphorus G05 Glutathione and Sulfur Sources PM4 Phosphorus G06 D,L-Ethionine and Sulfur Sources PM4 Phosphorus G07 L-Methionine and Sulfur Sources PM4 Phosphorus G08 D-Methionine and Sulfur Sources PM4 Phosphorus G09 Glycyl-LMethionine and Sulfur Sources PM4 Phosphorus G10 N-Acetyl-D,LMethionine and Sulfur Sources PM4 Phosphorus G11 L- Methionine and Sulfur SulfoxideSources PM4 Phosphorus G12 L-Methionine and Sulfur Sulfoxide Sources PM4 Phosphorus H01 L-Djenkolic and Sulfur Acid Sources PM4 Phosphorus H02 Thiourea and Sulfur Sources PM4 Phosphorus H03 1-Thio- andβ Sulfur-DGlucose Sources PM4 Phosphorus H04 D,L-Lipoamide and Sulfur Sources S-Source, organic PM4 Phosphorus H05 Taurocholic and Sulfur Acid Sources PM4 Phosphorus H06 Taurine and Sulfur Sources PM4 Phosphorus H07 Hypotaurine and Sulfur Sources PM4 Phosphorus H08 p-Amino and Sulfur Benzene Sources Sulfonic Acid PM4 Phosphorus H09 Butane and SulfonicSulfur Sources Acid PM4 Phosphorus H10 2-Hydroxyethane and Sulfur Sources Sulfonic Acid PM4 Phosphorus H11 Methane and Sulfur Sulfonic Sources Acid PM4 Phosphorus H12 Tetramethylene and Sulfur Sources Sulfone

276 Appendix

Appendix VII

1 2 3 4 5 6 7 8 9 10 11 12 13 10,000 bp

3,000 bp

1,500 bp

a

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4

1,500 bp

1,500 bp 800 bp

800 bp b c

Figure 1. Authentication of WSM3693 and Lut5 reisolated from harvested nodules: Gel image of rep-PCR using ERIC primers. a) Lane 1: 1 kbp DNA ladder; lane 2: no template; lane 3: A.s N1; lane 4: A.s N2; lane 5: I.f N1; lane 6: I.f N2; lane 7: I.f N3; lane 8: I.f N4; lane 9: WSM3693 –80 stock; lane 10: L.a N1; lane 11: L.a N2; lane 12 L.a N3; lane 13: 1 kbp DNA ladder. b) Lane 1: 100 bp DNA ladder; lane 2: WSM3693 –80 stock; lane 3: P.v N1; lane 4: P.v N2; lane 5: V.u N1; lane 6: V.u N2; lane 7: V.u N3; lane 8: V.u N4; lane 9: V.u N5; lane 10: V.u N6; lane 11: V.u N7; lane 12: 100 bp DNA ladder; b) Lane 1: 100 bp DNA ladder; lane 2: Lut5 –80 stock; lane 3: A.s N1; lane 4: 100 bp DNA ladder. A.s = Acacia saligna; I.f = Indigofera frutescens; L.a = Listia angolensis; P.v = Phaseolus vulgaris; V.u = Vigna unguiculata; N = nodule

277 Appendix

Appendix VIII

Figure 1. Bayesian tree for concatenated sequences of dnaK, gyrB, recA, rpoB (2427 bp) from seven Microvirga strains and eleven Alphaproteobacterial reference taxa. Posterior probabilities are listed above branches. Scale bar for branch lengths shows 0.05 substitutions per site. (Matt Parker, unpublished data).

278 Appendix

Figure 2. Bayesian tree for nodA sequences (594 bp) from three symbiotic Microvirga strains and 29 proteobacterial reference taxa. The posterior probability was 1.0 for 23 of the 29 internal branches of the tree; for the six other branches, the posterior probability is listed on the tree. Scale bar for branch lengths shows 0.05 substitutions per site. (Matt Parker, unpublished data).

279 Appendix

Figure 3. Bayesian tree for concatenated sequences of nifD and nifH (879 bp) from five Microvirga strains and 17 Alphaproteobacterial reference taxa. Posterior probabilities are listed above branches. Scale bar for branch lengths shows 0.05 substitutions per site. (Matt Parker, unpublished data).

280 Appendix

Appendix IX

Table 1. % G+C and DNA:DNA hybridization results. (Anne Willems and Sofie De Meyer, unpublished data.) Lut 5 Lut 6T WSM WSM % GC 3557T 3693T Lut5 - 61.9 Lut6T 97 - 61.9 WSM3557T 28.5±1.5 34.5±0.5 - 62.9± 0.1 WSM3693T 20 25 33.5±0.5 - 62.6 Microvirga flocculans TFBT 19 26 26 64.0* *From Takeda et al. (2004).

281 Appendix

Table 2. Cellular fatty acid compositions*. (Anne Willems and Sofie De Meyer, unpublished data.) Microvirga Microvirga Chelatococcus FAME on TSA medium WSM WSM flocculans subterranea asaccharovorans Lut5 Lut6T 3557T 3693T LMG 25472T LMG 25504T LMG 25503T 11 methyl 18:1 ω7c 1.1 12:00 Tr 1.27 14:00 1.63 1.11 Tr 15:0 3OH Tr 1.08 Tr 15:1 ω8c Tr Tr 16:00 4.25 2.48 8.77 7.85 9.8 11.5 10.4 16:0 3OH Tr 17:00 Tr Tr 1.04 1.3 Tr 17:0 CYCLO 1.82 2.13 1.4 17:1 ω8c Tr 18:00 1.36 2.55 2.71 2.24 5 4.5 4.3 18:0 3OH 1.00 1.07 1.24 Tr 3.1 2.3 18:1 ω7c 68.94 69.71 52.58 53.00 59.8 63.1 14.5 19:0 10 methyl 1.48 1.85 Tr Tr 19:0 CYCLO ω8c 1.32 1.26 17.65 17.25 6.6 6.3 56.7 20:1 ω7c 2.2 20:2 ω6,9c Tr Unknown 14.959 Tr Tr 1 Tr Summed Feature 1 Tr Tr Summed Feature 2 16.06 15.41 5.60 11.29 11.4 8.3 4.7 Summed Feature 3 3.56 2.29 4.17 3.85 3 3.4 * Summed Feature 1 comprises 15:1 iso H or I / 13:0 3OH; Summed Feature 2 comprises 16:1 iso I, 14:0 3OH / unknown 10.938; Summed Feature 3 comprises 16:1 w7c / 15:0 iso 2OH; Tr = Trace amounts (less than 1%)

282