Evolution of rhizobium symbiosis

The evolution of rhizobium symbiosis is studied from several points of view in this thesis. The ultimate goal of the combined approaches is to unravel the genetic constrains of the symbiotic interaction. To this end the legume rhizobium symbiosis is studied in model plant species from the Papilionoideae subfamily such as Medicago truncatula and Lotus japonicus. In these model plants the genetic signaling cascade used for rhizobium symbiosis has been largely unraveled. The cascade is triggered by lipo-chitooligosaccharade-based signal molecules excreted by rhizobia, called Nod factors. In chapter 2 we make use of a whole genome duplication that has occurred at the root of the legume Papilionoideae subfamily to identify maintained paralogous gene pairs. We hypothesized that a substantial fraction of gene pairs which are maintained in distinct Papilionoideae lineages that split roughly 54 million years ago fulfill legume specific functions, among which is rhizobium symbiosis. Furthermore we argue that such approach could identify novel genes as it can also identify genes pairs that are (partially) redundant in function. With applying this approach specifically to the cytokinin phosphorelay pathway we identified a pair of type-A cytokinin Response Regulators that are involved in rhizobium symbiosis. This study provides a proof-of-principle for this strategy. It is known for over fifty years that cytokinin plays an important role in the symbiotic interaction between rhizobia and legume hosts. External application of cytokinin can even result in nodule formation. Only, never had cytokinin levels been quantified in legume root extracts upon symbiotic interaction. In chapter 3 we describe a method for extraction of both cytokinins and auxin from Medicago truncatula roots. We show that cytokinins accumulate in the root zone susceptible to symbiotic interaction upon Nod factor exposure and that this response is dependent on CCaMK; a key gene of the Nod factor signaling cascade. Furthermore, it was found that ethylene signaling has a negative effect on Nod factor induced cytokinin accumulation. The method set up to measure cytokinin as well as auxin provides a tool to further study hormone interactions in rhizobium symbiosis. Parasponia,the only non-legume that can engage the rhizobium symbiosis is also subject of study in this thesis. The genetics of the Parasponia-rhizobium symbiosis had not been studied before. It was therefore unknown whether this independently evolved rhizobium symbiosis makes use of the same symbiotic signaling cascade as legumes. In chapter 4 we provide first evidence that Parasponia indeed makes use of the same signaling cascade as found in legumes. Furthermore, we show that in Parasponia a single Nod factor-like receptor is indispensable for two symbiotic interactions; rhizobium and mycorrhiza, respectively. Therefore we conclude that the rhizobium Nod factor perception mechanism is recruited from the widespread endomycorrhizal symbiosis. Parallel to our studies in Parasponia (Chapter 4), the research team of Jean Dénarié of the French National Institute for Agricultural Research (INRA) published the structure of the signal molecule of the arbuscular endomycorrhizae; theMyc factor (Maillet et al., 2011, Fungal lipochitooligosaccharide symbiotic signals in arbuscular mycorrhiza. Nature, 469, 58-63). It appeared that Myc factors and Nod factors are structurally very similar. In chapter 5 we discuss these findings and present a more thorough phylogenetic analysis of the NFP-like LysM-type receptor kinases. Together, these results suggest that non-legumes that can engage an arbuscular endomycorrhizaesymbiosis can recognize Nod factor-like molecules as well. The last chapter is about a study on the promiscuity and effectiveness of the Parasponia-rhizobium symbiosis. Parasponia uses a single receptor to control entry of rhizobium as well as arbuscular endomycorrhizal fungi and has evolved the rhizobium symbiosis only recently. This made us to hypothesize that Parasponia Nod factor receptors did not coevolve yet with rhizobia and therefore did not diverge from mycorrhizal recognition to develop specificity for the Nod factor. This implies that Parasponia could be a very promiscuous host for rhizobium species. In chapter 6 we describe that Parasponia andersonii can be nodulated by a broad range of rhizobia belonging to 4 different genera, and therefore it is concluded that Parasponia is highly promiscuous for rhizobial engagement. There is a drawback to this high symbiotic promiscuity. Among the strains identified to nodulate Parasponia, a very inefficient rhizobium species, Rhizobium tropici WUR1, was characterized. As this species is able to make effective nodules on two different legume species it suggests that the ineffectiveness of Parasponia andersonii nodules is the result of the incompatibility between both partners. In Parasponia andersonii nodules rhizobia of the ineffective strain become embedded in a dense matrix, but remain vital. This suggests that sanctions or genetic control against underperforming microsymbionts may not be effective in Parasponia. Therefore we argue that the Parasponia-Rhizobium symbiosis is a delicate balance between mutual benefits and parasitic colonization. Parasponiahas been given little attention in the rhizobium symbiosis field over the past two decades but with our efforts renewed interest has been established. We believe that in the end, the comparison of Parasponia to its closest related non-symbiotic sister species Trema, will result in the determination of the genetic constrains of rhizobium symbiosis.