Cottonwood IRCEB Grant

Plant genetic structure as a controlling factor in community and ecosystem functioning: studies using natural and synthetic hybrids of a dominant riparian tree

This material is based upon work supported by the National Science Foundation under Grant DEB-0078280
Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation (NSF).

INTRODUCTION

A major issue in community ecology and conservation biology is to understand the links that tie species together to form communities, and to understand how these communities may affect ecosystem function. Ecological genetics studies have shown that genetic variation within species has important ecological and evolutionary implications for individual species and two species interactions (Charlesworth & Charlesworth 1987, Lande 1992, Pusey & Wolf 1996, Lacy 1997). However, at the community and ecosystem level, little is known about the consequences of genetic variation. We believe that a genetic perspective of community and ecosystem functioning is important to understand these complex processes.

We propose that genetic variation in primary producers is a major factor that structures communities and influences ecosystems. We also propose that plant hybrid zones represent an ideal arena in which to examine the effects of genetic variation on communities and ecosystems.

We argue that:
1) Genetic variation among primary producers influences community structure through impacts on competition, bottom-up forces, and the distribution of keystone species.

2) Genetic variation influences ecosystem processes by influencing traits such as leaf chemistry, and by mediating the distribution of herbivores that modify plant traits.

3) Hybrid populations are characterized by increased genetic variability, making hybrid zones ideal systems in which to study the consequences of genetic variation.

BACK TO TOP

Research Hypotheses:

We address the basic premise that genetic variation in a dominant plant structures communities and significantly affects ecosystem processes. We have developed five interrelated hypotheses that explore patterns and mechanisms that scale from individual plants to landscape levels.

Hypothesis #1: Biodiversity and stability are correlated with genetic diversity at local to landscape levels.

Hypothesis #2: Cross-types vary greatly in their defensive chemistry, which in turn is associated with the distributions of common and rare insect species .

Hypothesis #3: Genetic effects of keystone species are major factors that structures communities.

Hypothesis #4: Genetic variation among cross-types and herbivory affect ecosystem-level processes such as litter decomposition, nutrient cycling, and NPP.

Hypothesis #5: Using molecular trait mapping techniques, important community traits may be mapped just as traditional morphological plant traits are mapped.

BACK TO TOP

GENETIC VARIATION AND COMMUNITY STRUCTURE

Three well-known factors that are important in structuring communities are competition, bottom-up forces (i.e., resource availability and diversity), and keystone-species which, through a variety of mechanisms, have community-wide impacts. Because these factors can be influenced by genetic variation in plants, plant genetic variation may play a large role in determining community structure. Competition can structure communities through niche partitioning and by excluding inferior competitors from patches occupied by superior competitors (Roughgarden 1986, McLaughlin & Roughgarden 1993). Field studies of competition commonly determine the effect of one species on another, but they seldom address underlying factors influencing the frequency or intensity of competition (Tilman 1987). Genetic variability in host plants and the spatial distribution of genotypes may set an arena for interspecific competition among herbivores, or mediate the outcome of competition (Fritz 1990, Mopper et al. 1990, Moran & Whitham 1990).

Bottom up effects such as host plant quality and top-down effects such as predation are thought to be important facotrs influencing community structure (Hunter & Price 1992, Power 1992, Strong 1992). Plant variability in herbivore resistance and resource quality is known to affect herbivore population dynamics (Karban 1992, Fritz 1992, Jervis et al. 1993, Siemann et al. 1998). Bottom-up influences of plant variation on predators and parasites may in turn influence top-down effects on communities. One mechanism through which plant genetic diversity may structure communities is by first influencing the distribution of herbivores associated with resistant and susceptible genotypes. This spatial variability in herbivore communities on different plant genotypes could alter predator communities by influencing predator abundance and searching abilities. Thus, plant genetic variation is likely to influence both herbivores and higher trophic levels.

Communities are also structured by the presence of keystone species. Relative to their abundances, keystone species have large impacts on communities (Power et al. 1996). Examples of the impacts of keystone species include predators whose removal from a system alters the density of prey, thus affecting a wide array of species (Paine 1966, Bond 1993, Mills et al. 1993), and herbivores that alter the condition of plant resources, thereby affecting other herbivores (Hunter & Price 1992, Bond 1993). Because keystone
species have dramatic effects on communities, the strongest effects of plant genetic diversity for community structure may occur through influences on keystone species. Our studies of hybridizing poplars have established that the distribution of keystone herbivores that structure arthropod communities are influences by genetic-based plant resistance traits (Dickson & Whitham 1996, Waltz & Whitham 1997, Whitham et al. 1999, Wimp & Whitham 2000).

BACK TO TOP

GENETIC VARIATION AND ECOSYSTEM FUNCTIONING

The role of species diversity in ecosystems has received considerable attention in the last ten years and has paramount importance in conservation and restoration strategies (Ehrlich & Ehrlich 1981, Jones & Lawton 1995, Lawton 1994). Key studies have shown that species, or species combinations impact various components of ecosystem function, such as litter decomposition, nitrogen mineralization and net primary productivity (NPP), through plant traits such as leaf chemistry (Wedin & Tilman 1990, Hobbie 1992, Naeem et al. 1994, Hooper & Vitousek 1998). However, no study has specifically examined how plant genetic variation can impact theses processes. Since plant genotype can influence traits such as leaf chemistry and nutrient uptake (Tilman & Wedin 1991, Wardle et al. 1998), genetic variation in dominant plants is likely to influence ecosystem functioning.

Genetic variation among plants may also indirectly impact ecosystem function by influencing the distribution of keystone herbivores. Herbivores can induce chemical changes in leaves that feed back to ecosystem processes ( Choudhury 1988, Findlay et al. 1996). Thus, genetic variation in herbivore resistance may control the distribution of organisms that alter ecosystem processes. The combined roles of plant genetic variation and keystone herbivores represent major components that are missing from theories of biodiversity and ecosystem functioning (Solbrig 1991, Mooney et al. 1995).

BACK TO TOP

PLANT HYBRID ZONES AS MODEL SYSTEMS TO STUDY THE EFFECTS OF GENETIC DIVERSITY

Our research and others argue that studies of hybridizing systems can tell us much about communities and ecosystems. Although much work has focused on the evolutionary implications of hybridization from the plant perspective (e.g., Stebbins 1950, Stace 1987, Wendel et al. 1991, Rieseberg 1991, Arnold 1992, Rieseberg & Wendel 1993, Arnold & Hodges 1995, Rieseberg 1997) and serious conservation issues have been raised about hybridization (e.g., Levin et al. 1995), we are just now beginning to appreciate that hybridization in plants can affect all trophic levels and diverse taxa from microbes to vertebrates. Using the genetic variation found in natural and synthetic populations of hybrids, recent studies have begun to explore the genetic mechanisms that affect plant defensive chemistry (Orians & Fritz 1995) and decomposition of leaf litter (Driebe & Whitham 2000). These studies have also demonstrated effects of plant hybridization on three-trophic level interactions (e.g., Ericson et al. 1993, Preszler & Boecklen 1994) and arthropod communities (e.g., Aguilar & Boecklen 1992, Fritz et al. 1994, 1996, Messina et al. 1996, Mattson et al. 1996, Dungey 1996).

Recent review articles (Strauss 1994, Whitham et al. 1999) show that the genetic variation in hybrid zones has community-level consequences. In 152 case studies of diverse hybridizing systems, 28% of the hybrids exhibited greater susceptibility to herbivores or pathogens than either parental species, 5% were more resistant, 23% showed additive genetic effects, 23% exhibited dominant genetic effects, and only 21% showed no response to hybridization. Thus, most organisms exhibit significant responses to hybridization. Hybridization results in a combination of inferior, intermediate, and superior genotypes that provide the genetic variation needed to explore the genetic basis of community structure.

Our group has contributed extensively to this literature, and the following patterns have emerged:

1. Genetic variation within hybrids (e.g., F1 s& backcrosses) is far greater than that found within species (Whitham et al. 1999, Martinsen et al. 2000).
See Figure 1 below:

Figure 1:
Based upon 40 RFLP markers, the genetic distances among F1 and backcross hybrids are about 3X greater than each parental species.

From Fisher's Fundamental Theory of Evolution, such variation is extremely important because the rate of evolution is directly proportional to the genetic variation upon which selection can act. This genetic variation represents a largely unexploited tool we can use to study community and ecosystem level processes.

2. Herbivores are 3x more likely to discriminate among hybrids than to discriminate among plant species ( Morrow et al. 1994, Whitham et al. 1994, 1999).
3. At the community level, different hybrid classes support different communities that also differ from the communities of the parental species (Floate & Whitham 1995).
4. Generalist and specialist herbivores predictable vary in their responses to hybrids (Whitham et al. 1994).
5. Plant hybrid zones may represent essential habitat for rare species and represent centers of biodiversity (Whitham et al. 1991, 1994 Dungey et al. 1994, 2000).
6. Even nesting birds respond to hybridizing plants (Martinsen & Whitham 1994).
7. Including multiple trophic levels and taxa from fungi to vertebrates, susceptible hybrid genotypes support greater biodiversity than resistant genotypes (Dickson & Whitham 1996).
8. The effects of hybridization on keystone species can either positively or negatively affect biodiversity (Dickson & Whitham 1996, Waltz & Whitham 1997, Wimp & Whitham 2000) and can affect the distribution of species at both local and regional levels (Floate et al. 1997).
9. Although hybrids are often maligned, native hybrids are part of natural ecosystems and should be conserved (Whitham & Maschinski 1996). Combined with studies from other lab groups (e.g. see Special Features in Ecology, 1999), we are rapidly gaining insights into a field previously avoided by community ecologists because hybrids were viewed as "freaks" of nature with no ecological or evolutionary consequence.

BACK TO TOP

WHY HYBRID POPLARS ARE A MODEL SYSTEM:

1. The genetic variation among naturally occuring hybrids is approximately 3X greater than among trees of either parental species.
2. Natural hybrid zones are found on virtually every river in the western United States (Eckenwalder 1984a,b) and range in size from small isolated relics to huge hybrid swarms over 1000km in length (Whitham unpub. data).
3. Using 35 species specific RFLP markers (nuclear, mitochondrial and chloroplast), analyses demonstrate introgression from the hybrid zone into the "pure" narrowleaf zone. However, 80% of the markers show no introgression (Martinsen et al. 2000).
See Figure 2:

Figure 2:
Figure 2 shows an example of a marker that has introgressed into the otherwise "pure" narrowleaf zone. No gene flow occurs in the opposite direction, so introgression is unidirectional (Keim et al. 1989). This analysis is important for three reasons. It demonstrates that we have numerous species specific markers that allow us to definitively identify pure and hybrids trees. It also demonstrates that hybrids can facilitate gene flow between species. Last, markers that do not cross species boundaries are probably associated with negative genes, while those that introgress far into the narrowleaf zone may be associated with traits that increase fitness and are under positive selection.

A marker that show introgression--this occurs only 20% of the time.

4. Within the hybrid zone, molecular analyses show that different stands exhibit different genetic structures (Martinsen et al. 2000), which affects common or keystone species.
5. At the landscape level, molecular analyses show that the hybrid zones of different river systems from New Mexico to Alberta have very different genetic structures, which may support different arthropod communities.
6. Based upon reproduction success, the fossil record, and a hybrid speciation event, hybrid cottonwoods are at least equal in fitness to their parental species.
7. Plant defenses and nutrient content vary greatly among species and hybrid cross-types.
8. Plant genetics affects populations, communities, and 3-trophic level interactions.

BACK TO TOP