Evolutionary and ecological outcomes of tripartite mutualisms

Ossler, Julia N

Permalink

https://hdl.handle.net/2142/95491 Copy

Description

Title

Evolutionary and ecological outcomes of tripartite mutualisms

Author(s)

Ossler, Julia N

Issue Date

2016-12-01

Director of Research (if dissertation) or Advisor (if thesis)

Heath, Katy

Doctoral Committee Chair(s)

Dalling, Jim

Committee Member(s)

• Paige, Ken
• Yannarell, Tony

Department of Study

Plant Biology

Discipline

Plant Biology

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Mutualisms
• Coevolution
• arbuscular mycorrhizal fungi
• multiplayer
• Rhizobium

Abstract

Many models of coevolution assume that interactions among hosts are mediated by only a single trait or genetic locus (Case, Taper, & Associate Editor: Nicholas, 2000; Gavrilets & Hastings, 1998; S. L. Nuismer, Thompson, & Gomulkiewicz, 1999; Seger & Antonovics, 1988). In mutualistic associations however hosts often rely and interact with multiple species and trophic levels to obtain benefits that will confer a fitness advantage (not strict pairwise interactions) (M. T. Johnson & Stinchcombe, 2007; W. F. Morris et al., 2007; Strauss & Irwin, 2004). Legumes, such as Chamaecrista fasciculata, are an example of just that. Hosts create specialized structures along their root called nodules wherein rhizobia reside and fix atmospheric nitrogen to plant-available forms (Harris, Pacovsky, & Paul, 1985a). Legumes also interact with arbuscular mycorrhizal fungi (AMF) to form multiple intraradical structures (including arbuscules and hyphae) which increase plant nutrition and mineral status, particularly phosphorus (S. Smith & D. Read, 2008). Both traits are regulated in, in part, is regulated through a series of pleiotropic genes known as the common symbiotic pathway, or CSP (Ane et al., 2004; Kistner, Winzer, Pitzschke, Mulder, & Sato, 2005; Murray et al., 2011; Giles E. D. Oldroyd, 2013; Giles E. D. Oldroyd & Downie, 2006; G. E. D. Oldroyd & Long, 2003; Stracke et al., 2002). In my dissertation, I was able to quantify host investment in both rhizobia and AMF by counting structures that represent host investment due to a nutritional demand; the number and weight of nodules as well as the density of intraradical structures AMF produce to exchange nutrients with hosts. I used C. fasciculata as a model legume in both field and controlled greenhouse experiments to shed light on the potential for shared regulatory networks, and genetic variation in those networks, to shape the ecology and evolution of the tripartite symbiosis between legumes, rhizobia, and AMF. I used a remnant prairie site in Kankakee, IL for observations of host plant interactions with each symbiont as well as to collect soil and plants for greenhouse experiments. In the field, both traits, rhizobia nodule number and the density of AMF structures, covary in individuals, suggesting that CSP and shared regulatory networks may couple these two traits tightly in nature. When I manipulated the abundance of rhizobia in a greenhouse study, I found evidence of a trade-off among nodule number and arbuscule density, but this was only seen when hosts were grown with abnormally high densities of rhizobia. Therefore, while host investment in rhizobia may alter investment in AMF, this is not suspected to be common in nature due to lower densities of rhizobia found in the field. Additionally, I found that traits regulated by host CSP (i.e. nodule number and arbuscule density) had ample genetic variation among families and therefore these traits are potential targets of selection. In multiple investigations we found that agents of selection only acted directly on nodule number, not the density of arbuscules. Models of coevolution suggest that selection only acting directly on one symbiont could lead to conflict in certain ecological situations. To discern if this would occur in this system, I conducted another greenhouse experiment, growing 75 families of plants in field soil from Kankakee. Plants were also grown in either a high P or low P environment to lower the benefit of AMF and subsequent host investment. Analysis revealed that host investment in nodules and arbuscules is not genetically correlated. Fertilized plants were able to lower their investment in AMF while subsequently increasing their nodule number. I also found no connection in the plasticity of either trait among families (i.e. families that altered their investment in rhizobia nodule number, did so independently of AMF arbuscule density). This in turn would suggest that each trait is able to evolve independently in hosts and we would not anticipate situations of conflict to arise between host investment in each focal symbiosis. Unexpectedly however, we found that nodule number was genetically correlated with host aboveground biomass. Both traits also exhibited correlated responses to P-fertilizer, such that nodulation changes to phosphorus availability are a strong indicator of plant fitness. Strong host reliance on increased nodule number may be favorable in certain environments, but alternatively could limit host range expansion if partners are in low abundance or not compatible with host genotypes. Nodule number was also highly variable among families which appeared contrary to our observations of direct selection in the field to increase nodule number. To answer this, I investigated the impact of changes in the cost:benefit ratio on host allocation to both symbiosis. By lowering the benefit of the AMF symbiont (increased soil P-content) and raising the cost of both symbiosis (decreased total photosynthate to invest in symbiosis via light limitation) we found that both raised costs and lowered benefits in AMF interacted to influence host investment in both nodule number and arbuscule density. We found that the effect of P fertilization on host investment in rhizobia changed from positive (increased nodulation) to negative (decreasing nodulation) when hosts were light limited. While increased nodulation was still favored in multiple environments, changes in light and P availability altered the strength of selection in the direction that we would predict based on our current understandings of costs and benefits in multiplayer mutualisms. Host investment in AMF also decreased when P was readily available in the soil. These negative effects were exacerbated when hosts were light limited suggesting that both costs and benefits play key roles in host interactions with each symbiosis. Together these results suggest that selection and resulting evolution of multiplayer mutualisms of legumes-rhizobia-and AMF may not be constrained by multiple partners, but agents of selection are anticipated to act differently and independently on each partner which in turn could lead to variation in host investment across different environments.

Graduation Semester

2016-12

Type of Resource

text

Permalink

http://hdl.handle.net/2142/95491

Copyright and License Information

Copyright 2016 Julia N Ossler

Owning Collections