Mark D. Rausher                (919) 684-2295 (Voice)
        Department of Biology        (919) 660-7293 (FAX)
        Box 90338                          mrausher@duke.edu
        Duke University
        Durham, NC  27708-0338
 
 
 

RESEARCH INTERESTS

    Our laboratory is involved in a number of research projects in the areas of evolutionary biology
and ecology:

     I. Molecular evolution of ecologically important phenotypes
     II. The evolution of biochemical pathways
     III. The evolution  of plant mating systems
      IV. The evolution and ecology of plant-enemy interactions

 Click on a topic or scroll down to learn more about these projects. 

         The goal of this set of projects is to determine whether understanding the evolution of ecologically
     important traits requires analysis at the molecular level of the biochemical and developmental
     pathways that produce those traits.  In attempting to answer this question, we have chosen
     flower color as the ecologically important trait because the anthocyanin pigment biosynthetic
     pathway has been well characterized at the molecular level.  In one project, we have examined the
     genetic changes that have accompanied a shift from blue, bee-pollinated flowers to red, hummingbird
     pollinated flowers in the genus Ipomoea (morning glories).  We have found that the initial genetic
     change that resulted in red flowers was subsequently followed by degenerative changes in other
     anthocyanin pathway enzymes, suggesting that it would be very difficult to re-evolve blue flowers
     from red¹.  We are continuing to explore how general this phenomenon is by examining other lineages
     in which pigment production has been eliminated, resulting in white flowers.

       


         In a second morning-glory project, we are attempting to ascertain whether there are
     “preferred” evolutionary pathways for producing a given phenotype.  Although parallelism at
     the phenotypic level is a common phenomenon, little is known about whether parallelism at
     the phenotypic level is accompanied by parallel genetic change at the genetic level, largely
     because little is known about possible alternative genetic and molecular routes for producing
     the same phenotype.  Because the molecular basis of flower color is well characterized,
     however, it is possible a priori to define possible molecular routes that may be taken to
     produce white flowers.  Specifically, it is well established that white flowers can be produced
     by deactivation of any of  at least four structural genes (CHS, F3H, DFR, ANS) and two
     regulatory genes.  Because evolution of white flowers has occurred repeatedly in the genus
     Ipomoea, it is possible to ask whether these genes have been deactivated with roughly equal
     probability in different lineages that have evolved white flowers, or whether
     one or more of these genes have been activated preferentially.  Such preferential deactivation
     may be expected if, for example, deactivation of regulatory genes has fewer deleterious
     pleiotropic effects than deactivation of structural genes, or if deactivation of downstream
     structural genes has fewer deleterious pleiotropic effects than deactivation of upstream genes
     (which are involved in the production of flavonoids other than anthocyanins).   Our short-
     term goal is to determine whether there are “preferred” pathways to deactivation.
     Our longer-term goal is to understand how patterns of pleiotropy and gene interaction may
     determine any pattern we discover.

          In a third project on Phlox, in collaboration with Donald Levin at the University of Texas, we
    are attempting to identify a gene responsible for prezygotic isolation.  Levin has shown previously
    that in regions where it grows alone, P. drummondii has pink/lavender flowers, but in regions
    where it grows with P. cuspidata it has red flowers.  Levin has also shown that this difference
    in flower color reduces interspecific hybridization in the zone of sympatry.  Our goal is first
    to clone the gene(s) responsible for the red-lavender color difference, then examine patterns
    of variation in this gene to characterize in more detail the evolutionary processes that have
    caused reproductive character displacement in this system.


¹ Zufall, R. A., and M. D. Rausher.  2004. Genetic changes associated with floral adaptation restrict future
            evolutionary potential.  Nature  428: 847-850.

  Zufall, R. A., and M. D. Rausher.  2003.  The genetic basis of a flower-color polymorphism in the common
            morning glory, Ipomoea purpurea.  Journal of Heredity 94: 442-448.

  Tiffin, P., R. E. Miller and M. D. Rausher.  1998.  Control of expression patterns of anthocyanin  structural genes
            by two loci in the common morning glory.  Genes and Genetic Systems 73: 105-110. 

Back to Top 

II. The evolution of biochemical pathways

    One of the most significant discoveries in the field of molecular evolution over the past twenty years has been the documentation of significant variation in the evolutionary rates of proteins.  While this variation has generally been ascribed to differences in selective constraint among proteins, support for this explanation is based largely on anecdotal inferences regarding the degree to which different proteins are thought to be constrained.  We believe that an understanding of rate variation has been impeded by the historical focus on assessing evolutionary rates in individual, unrelated genes.  Most proteins, though, do not operate in isolation, but as components of complex metabolic networks.  Two properties of metabolic networks may be particularly important in determining patterns of variation in evolutionary rates of constituent enzymes: (1) the impact of an individual enzyme of a network on metabolic flux and end-product concentration (and, hence, on the phenotype) depends not only on its own kinetic properties, but on those of other proteins of the network; and (2) some enzymes (those above pathway branch points) potentially influence a greater number of end products than others (those below branch points). These properties suggest that upstream, "regulatory" enzymes in a pathway are likely to evolve more slowly than downstream, "equilibrium" enzymes.  We have recently tested this hypothesis by analyzing evolutionary rate variation among enzymes in the anthocyanin pathway¹ and found it to be upheld.  We have also ruled out the possibility that downstream genes may be evolving faster because of more frequent positive selection².  Our current efforts in this area are directed toward modeling the evolution of control coefficients and devising more direct tests of the hypothesis that upstream enzymes are under greater constraint because of their position in the pathway.

¹Rausher, M. D., R. E. Miller and P. Tiffin.  1999.  Patterns of evolutionary rate variation among genes of the
    anthocyanin biosynthetic pathway.  Molecular Biology and Evolution 16: 266-274.
²Lu, Y., and M. D. Rausher.  2003.   Evolutionary rate variation in anthocyanin pathway genes. Molecular Biology and Evolution 20: 1844-1853..

Back to Top
 

III. The evolution  of plant mating systems

    We are attempting to address this issue by examining the evolutionary processes that prevent the spread and eventual fixation of genotypes with increased selfing affecting two ecologically important traits in natural populations of the common morning glory, Ipomoea purpurea.  One trait is flower color, as influenced by the W locus.  Alleles causing the normally pigmented flowers to be white occur naturally in this species.  Frequency perturbation experiments have shown that some form of balancing selection is acting on this locus¹.   In other experiments, we have sought to understand how this balancing selection arises.  Because white flowers are visited less frequently than pigmented flowers, white-flowered plants have a higher selfing rate than plants with pigmented flowers².  Neither pollen discounting³ nor inbreeding depression⁴, alone or together, is of sufficient magnitude to counteract the transmission advantage associated with the white allele.  At intermediate to high frequencies of the white allele, differential pollinator visitation no longer occurs and plants with white and plants with pigmented flowers have equal selfing rates.  The transmission advantage is thus frequency dependent and selection actively favors an increase in the frequency of the white allele at low frequencies of that allele.  Despite attempts to detect a dark-allele advantage when rare, we still do not know what maintains the dark allele at high frequencies^{2, 5-7}.  We have begun asking similar questions about a white-flower allele at another locus, the A locus^8,9

    A second trait affecting selfing rate in I. purpurea is anther-stigma distance (ASD).  Genotypes with little separation between anthers and stigma (small ASD) have higher selfing rates than those with substantial separation.  We have again found that neither pollen discounting per se nor inbreeding depression is of sufficient magnitude to prevent the fixation of small-ASD genotypes.  However, male outcross success is apparently negatively frequency dependent, with small-ASD genotypes having higher male fitness when rare, and conversely for large-ASD genotypes.  The magnitude of this effect is sufficient to offset the transmission advantage of small-ASD genotypes associated with increased selfing and maintain ASD at an intermediate level¹⁰.  Examination of both traits thus suggests that inbreeding depression and pollen discounting are less important in maintaining variation in selfing rates than theory suggests.  Instead, frequency dependence of mating system parameters (selfing rate or male outcross success), as well as pleiotropic effects (biased inheritance in pollen), seem to be more important in stabilizing a mixed mating system in I. purpurea.  In future work, we hope to examine how generally these conclusions apply to species with mixed mating systems.

    In a theoretical analysis¹¹, we have found that evolutionary stable mixed-mating systems can be favored when the magnitude of inbreeding depression differs for male and female components of fitness in hermaphrodites.  This result suggests that more effort be made to quantify the effects of inbreeding depression on male components of fitness when trying to understand the evolution of selfing rates.

¹Subramaniam, B., and M. D. Rausher.  2000. Balancing selection on a floral polymorphism.  Evolution  54: 691-695.

²Fry, J.D. and M. D. Rausher.  1997.    Selection on a floral color polymorphism in the tall morning glory
        (Ipomoea purpurea L.): transmission success of the alleles through pollen.  Evolution 51: 66-78.
³Rausher, M. D., D. Augustine and A. Vanderkooie.  1993. Absence of pollen discounting in genotypes of
        Ipomoea purpurea exhibiting increased selfing.  Evolution 47: 1688-1695.
⁴Chang, S.-M. and M. D. Rausher. 1999. The role of inbreeding depression in maintaining the mixed mating system
        of the common morning glory, Ipomoea purpurea. Evolution 53: 1366-1376.

⁵Paulsen, S., and M. D. Rausher.  2001.  Floral color polymorphism in Ipomoea purpurea:  biased  inheritance of the
        dark allele is not a general explanation for its maintenance.  Journal of  Heredity 96: 491-495.

⁶Mojonnier, L, and M. D. Rausher.  1997.  Selection on a floral color polymorphism in the common morning glory
        (Ipomoea purpurea): The effects of overdominance in seed size.   Evolution 51: 608-614.

⁷Rausher, M. D., and J. D. Fry.  1993.  Effects of a locus affecting floral pigmentation in Ipomoea  purpurea on female
        fitness components.  Genetics 134: 1237-1247.
⁸Fehr, C.A,. and M. D. Rausher.  2004.  Effect of variation at the flower-colour A locus on mating
    system parameters in Ipomoea purpurea.  Molecular Ecology 13: 1839-1847.
⁹Coberly, L.C., and M. D. Rausher.  2003.  Analysis of a chalcone synthase mutant in Ipomoea purpurea reveals a novel
    function for flavonoids: amelioration of heat stress.  Molecular Ecology 12: 1113-1124.
¹⁰Chang, S.-M. and M. D. Rausher.  1998.  Evolution in mixed mating systems: frequency-dependent selection on
    anther-stigma separation in Ipomoea purpurea. The American Naturalist 152: 671-683.
¹¹Rausher, M. D. and S.-M. Chang.  1999.  Stabilization of mixed-mating systems by differences in the magnitude of
    inbreeding depression for male and female fitness components.  The American Naturalist 155: 242-248.

IV. The evolution and ecology of plant-enemy interactions

    In one project in this area, we are attempting to understand the role of plant enemies inshaping the plant evolution.  Although many evolutionary biologists believe that many plant characters, including trichomes and secondary compounds, have evolved primarily as defenses against pathogens and herbivores, other biologists believe that these characters have evolved primarily to serve other functions, such as protection from ultraviolet light or heat stress.  This disagreement has remained unresolved, primarily because there is little empirical information regarding the extent to which enemies impose selection favoring the evolution of these types of characters.  We are attempting to address this debate by characterizing and quantifying theselection pressures acting on purportedly defensive characters.  In a recent study¹, for example,we have demonstrated that insecticidal elimination of insect herbivores from experimentalpopulations of the annual crucifer Arabidopsis thaliana that alters the pattern of selectionon genetic variation for trichome density and leaf glucosinolate concentration.  By comparingthe pattern of selection in the presence and absence of herbivores, we found that feeding by herbivores favors both increased trichome density and increased glucosinolate concentration.We have found similar results in other plant-enemy systems, suggesting that selection exertedby natural enemies is common in nature².

    A second project involves attempting to ascertain how plant genetic variation and induced resistance influence the population dynamics of herbivorous insects.  This project is motivated by the observation that until recently, most attempts to account for population regulation and fluctuation in insect herbivores have focused on the role of natural enemies of the herbivores.  Recently, however, it has become recognized that host-plant characteristics affecting host quality can influence herbivore growth rates, survival and fecundity--the very life-history characters that determine population dynamics.  Despite this realization there have been few attempts to determine directly whether variation in plant characters influences herbivore population dynamics.  Using soy beans (Glycine max) and Mexican bean beetles (Epilachna verivestis) as a model system, we have recently shown that beetle populations on different soy bean cultivars differ in expected average population size and frequency of fluctuation, as estimated from input-output curves measured under field conditions³.  We have also shown that cultivars with inducible resistance produce
different dynamics than constitutively resistant cultivars⁴.

    In a third recent project, we have examined the selective forces acting on tolerance to herbivory in morning glories⁵.  In particular, we attempted to determine why tolerance levels are not maximal, but kept at intermediate values.  By quantifying the pattern of selection acting on tolerance to herbivbory and to apical meristem damage, we found little evidence of directional or stabilizing selection.  We also found very little indication that costs of tolerance or other genetic constraints act to oppose the benefits of tolerance.  However, the data indicates that at levels of herbivory different from that that occurred in our study, directional selection likely would have operated, suggesting that fluctuating selection resulting from fluctuating herbivore densities may be responsible for maintaining tolerance at intermediate levels.

    In a final ongoing project, we are attempting to determine the nature of coevolution.  Two different types of coevolution
have been proposed: pairwise and diffuse⁶.  In pairwise coevolution, the trajectory of coevolution between a plant and one
of its natural enemies is independent of the presence or absence of other natural enemies.  By contrast, in diffuse
coevolution, that evolutionary trajectory is influenced by whether other natural enemies are present or not.  We have
developed statistical techniques for distinguishing between these two kinds of coevolution⁷.  We have also recently demonstrated
that the pattern of selection on resistance and tolerance to deer herbivory in morning glories is influenced by the presence/absence
of natural enemies, and is thus consistent with the type of selection expected under diffuse coevolution⁸.

¹Mauricio, R. and M. D. Rausher.  1997.  Experimental manipulation of putative selective agents provides evidence
        for the role of natural enemies in the evolution of plant defenses.Evolution 51: 1435-1444.
²Rausher, M. D.  1996.  Genetic analysis of coevolution between plants and their natural enemies. Trends
        in Genetics 12: 212-217.
³Underwood, N. and M. D. Rausher.  2000.  The effects of host-plant quality and plant genotype on herbivore
        populationdynamics in a model system.  Ecology 81: 1565-1576.
⁴Underwood, N., and M. D. Rausher.  2002.  Comparing the consequences of induced and constitutive plant resistance for
            herbivore population dynamics.  The American Naturalist (in press).
⁵Tiffin, P and M. D. Rausher.  1999. Genetic constraints and selection acting on tolerance to herbivory in the common
        morning glory.  The American Naturalist 154: 700-716.
⁶Rausher, M. D.  1996.  Genetic analysis of coevolution between plants and their natural enemies.
        Trends in Genetics 12: 212-217.
⁷Iwao, K. and M. D. Rausher.  1997.  Evolution of plant resistance to multiple herbivores:
        quantifying diffuse coevolution.  American Naturalist 149: 316-355.
⁸Stinchcombe, J. R., and M. D. Rausher.  2001.  Diffuse selection on resistance to deer herbivory
        in the ivyleaf morning glory, Ipomoea hederacea.  The American Naturalist 158: 376-388.

Back to Top
 
 

Rausher Lab Home Page         Duke University Home Page          Department of Biology Home Page