A not so modest proposal for pathogens: evolutionary diversification

Due to a ground swell of interest, we recently read Robert Ricklefs inaugural article (Ricklefs 2010) in to the National Academy of Sciences (of the United States of America) in which he proposes a new mechanistic role for parasites and pathogens to generate diversity within the tree of life. In this paper, Ricklefs (2010) distinguishes between two compartments of the ecological niche of a species: 1) the individual niche space and 2) the population niche space. He contrasts these two concepts of niche space by indicating which processes are most involved in defining the boundaries: 1) evolution and adaptation of an individual versus 2) demographic properties of a population in a point in space. Being a fan of processes not patterns, I thought that these definitions were particularly helpful when reading the rest of the article and understanding his proposed novel mechanism of diversification at the end.

Ricklefs asks how different clades occupy population niche space: do more taxon rich clades occupy larger niche space or simply pack the available space more tightly with narrower species niches or larger overlap between species? Among several lines of evidence, the most crucial to his proposal is that there is independence of the diversity of a clade and the total population niche space occupied. That is, larger clades must pack niche space more tightly. But apparently they are doing it not by lowering the species densities because species abundance does not decrease with increasing local diversity. What Ricklefs suggests is it that the tighter packing is achieved via uneven filling of particular population niche space. This uneven filling is due to interactions with pathogens and parasites. The outcome of these interactions being determined by spatially and temporally varying antagonistic interactions that may also vary not just in the composition of those interactions but the diversity of the players involved (conjuring the idea of geographic mosaic of coevolution [Thompson 2005]).

How is this proposal different from adaptive radiations or escape and radiate coevolution? The paper makes the first contrast from adaptive radiations by presenting his mechanism in context of seemingly saturated niches rather than a having diversification happen in a wide open landscape. What about escape and radiate coevolution (Ehrlich and Raven 1964) which also has a role for interacting species in diversification? Again, this is a case were new adaptive zones (Simpson 1953) are opened up and allow species to occupy these new empty niches. Ricklefs' idea is fundamentally different in that pathogen interactions are seen as a mechanism that reduces efficient packing and saturation of population niches. This is achieved by affecting the population demographics which can result in a feedback to evolutionary dynamics at the individual niche level. This last part highlights the importance of linking demographic and evolutionary factors into models of coevolutionary interactions when concerned with patterns of diversification. Others have already pointed out this need in models exploring other evolutionary important traits (Mideo et al 2008)

While I was expecting something different at the conclusion of this article, what Ricklefs does do is lay out a program of study and call for data to defend his proposal. This request does not only extend to the field parasitologist but also to theoreticians as well.

References

Ehrlich, P. R., and P. H. Raven. 1964. Butterflies and plants: a study in coevolution. Evolution 18:586-608.

Mideo, N., S. Alizon, and T. Day. 2008. Linking within- and between-host dynamics in the evolutionary epidemiology of infectious diseases. Trends in Ecology & Evolution 23:511-517.

Ricklefs, R. E. 2010. Evolutionary diversification, coevolution between populations and their antagonists, and the filling of niche space. Proceedings of the National Academy of Sciences of the United States of America 107:1265-1272.

Simpson, G. G. 1953. The major features of evolution. Columbia University Press, New York.

Thompson, J. N. 2005. The Geographic Mosaic of Coevolution. University of Chicago Press, Chicago.

Paper Read

Ricklefs, R. (2010). Inaugural Article: Evolutionary diversification, coevolution between populations and their antagonists, and the filling of niche space Proceedings of the National Academy of Sciences, 107 (4), 1265-1272 DOI: 10.1073/pnas.0913626107

Sex and death: a model of density-dependent virulence

Providing evidence that supports the role of parasites driving the maintenance of sex (i.e. the Red Queen hypothesis) has been a challenge ever since it was proposed. Both theoreticians and empiricists have tackled this hypothesis with vigor to mixed results. This week we read Lively (2009) which focuses on a singular effect to help build a theoretical argument for the Red Queen, density-dependent virulence. Here virulence is defined as the effect of the parasite on the host population growth rate. The density-dependent part comes into play in that the virulence increases with host population size.

The main argument of the paper is that as an asexual population invades a sexual population, the level of virulence changes and this can in turn change the outcome of the overall winner. Parasites with large density-dependent effects can change the balance and allow the maintenance of sexual populations. Presented in several graphs, virulence is a population measure of the effect of the parasites on the hosts. I'm still curious about the magnitude of selection on the individual genotypes in the model. When interpreting the results of this model, I was only able to see what happens when a group of asexual organisms invades a sexual one.

Lively provides an excellent ion description and understanding of the cost of sex. Of course the cost of sex has been detailed before, but the mathematical explanation helps with a basic intuition. The model described in the paper identifies two populations of hosts: asexual and sexually reproducing individuals. What he identifies is that in a sexual population, males provide little and females must produce at least two offspring to replace themselves. These males are using up resources. They are also decreasing the overall density of hosts that could be achieved in a complete female (or asexual) population.

One of the topics that came up during out discussion was how sex ratio may change or evolve during the evolution of sex. The simulation results presented in Lively (2009) assumes a sex ratio of 50/50 which makes sense in an evolutionary context. This has the effect of setting the advantage of the asexual population to be two fold over the sexual population. What happens when instead of two separate populations that do not interbreed, we have females choosing to produce offspring via sex or parthenogenesis? Will rare males in such a population change the early dynamics enough to produce different results?

References

Lively, C. M. 2009. The maintenance of sex: host-parasite coevolution with density-dependent virulence. J Evolution Biol 22:2086-2093.

LIVELY, C. (2009). The maintenance of sex: host-parasite coevolution with density-dependent virulence Journal of Evolutionary Biology, 22 (10), 2086-2093 DOI: 10.1111/j.1420-9101.2009.01824.x

Is the Red Queen showing her face? Evidence of negative frequency dependent selection by parasites

Recently Wolinska and Spaak (2009) provide a survey of Daphnia infections by genotype across a number of lakes in Italy and Switzerland. They present their results as empirical evidence of Red Queen dynamics in which coevolution with virulent parasites generates continued evolution. Although Van Valen (1973) originally presented a macroevolutionary argument where by reciprocal selection of hosts and their parasites generates conditions for continuous change, Bell (1982) narrowed the focus as a mechanistic explanation for the evolution or maintenance of sexual reproduction through cyclical changes in genotype frequencies. Wolinska and Spaak (2009) are not addressing the evolution of sex, but looking for evidence that parasites in Daphnia populations are generating negative frequency dependent selection such that a rare genotype has an advantage. Evidence consistent with the Red Queen has been found in other systems using spatially distributed samples (e.g. Dybdahl and Lively 1995) to look non-random infection rates as well as more directly looking at changes in frequencies of common genotypes (e.g. Dybdahl and Lively 1998).

Wolinska and Spaak (2009) propose three hypotheses to test with their data. The first is that common genotypes should be either over or under infected compared to a random sample. This prediction is based on stereotyped cyclical dynamics of genotypes of hosts and parasites (image two out of sync sine waves). At some points, the common clones will be targeted by the parasites and become overly infected. As a genotype becomes common, parasites haven't started attacking this genotype yet (i.e. time lagged), so it is under infected. In their survey, the found that indeed, some of the populations showed over infection (n = 1) and other showed under infection (n = 11), although the majority of cases did show no significant difference from random infection probabilities which is predicted as being a rare event. Their second hypothesis was that common genotypes should over the course of time decline if they are being tracked by parasites. The previous sample included only different lakes; where as the data needed to test this hypothesis are temporal samples from the same location. Their additional data is consistent with common genotypes declining over time (9 out of 10 cases). However, it is unclear to me how the general trend in this data of common genotypes decreasing over time, leads to the evidence supporting the first hypothesis. Shouldn't they find many more over infected common clones? A third hypothesis that they tested regarded host-parasite interactions maintaining diversity and an evenness of genotype frequencies which their data supported.

When discussing this paper, we were interested in what happens to predictions based on Red Queen dynamics when more than one parasite is involved. Previous empirical papers and theory seems to be generally focused on a host and a common parasite, but we know hosts are attacked by all kinds of parasites and pathogens. The system described by Wolinska and Spaak (2009) involves a host hybrid complex as well as four different parasites and questions about host specialization and hybrid maintenance were addressed in a previous paper (Wolinska et al. 2007). Where is the companion theoretical work to provide testable hypotheses?

References

Bell, G. 1982. The Masterpiece of Nature: The Evolution and Genetics of Sexuality. University of California Press, Berkeley.

Dybdahl, M. F., and C. M. Lively. 1995. Host-Parasite Interactions: Infection of Common Clones in Natural Populations of a Freshwater Snail (Potamopyrgus antipodarum). Proceedings of the Royal Society of London. Series B: Biological Sciences 260:99-103.

Dybdahl, M. F., and C. M. Lively. 1998. Host-parasite coevolution: Evidence for rare advantage and time-lagged selection in a natural population. Evolution 52:1057-1066.

Van Valen, L. 1973. A new evolutionary law. Evolutionary Theory 1:1-30.

Wolinska, J., B. Keller, M. Manca, and P. Spaak. 2007. Parasite survey of a Daphnia hybrid complex: host-specificity and environment determine infection. Journal of Animal Ecology 76:191-200.

Wolinska, J., and P. Spaak. 2009. The cost of being common: evidence from natural Daphnia populations. Evolution 63:1893-1901.

Wolinska, J., & Spaak, P. (2009). The cost of being common: evidence from natural Daphnia populations Evolution, 63 (7), 1893-1901 DOI: 10.1111/j.1558-5646.2009.00663.x

Can the Red Queen keep running? A case against recombination

In 2004, Otto and Nuismer published a theoretical paper on the evolution of sex where they examined a range of stereotyped models (e.g. gene-for-gene) of species interactions (both antagonistic and beneficial) that are often used by theoreticians. Their results indicated that sex and recombination were generally selected against regardless of the model of interaction given the assumptions of the quasi linkage equilibrium (QLE, in this case, weak selection and strong recombination). In their numerical simulations that explored parameter space potentially outside the assumptions of the QLE, they found that some cases of the matching-genotypes model (or a strict matching alleles model) of interactions would favor sex and recombination.

Kouyos et al (2007) looked at a wide range of matching alleles models (MAM) and found that when selection was strong, some models would favor sex and recombination. Salathé et al (2008b) also provide evidence of strong selection favoring recombination under the MAM. However, both did find that the closer these models were to a multiplicative form of the MAM, sex and recombination were selected against. These multiplicative matching alleles models (MMAM) were described by Otto and Nuismer (2004) as the negative control in their numerical simulations because they never favored recombination. Their QLE results also indicated that this model of interaction should not generate linkage disequilibrium and therefore neither favor nor select against recombination. Contrary to this, in a surprising result by Kouyos et al (2007), their simulations found that there was strong selection against recombination (rather than no selection at all) in the parameter space near a MMAM.

It was this surprising result that was explained in the paper that we read this past week for Coevolvers (Kouyos et al 2009). Here the authors investigated why this parameter space shows strong selection again recombination. In a MMAM, there are no epistatic interactions between the loci involved in the fitness of the interaction between host and parasite. Despite this, previous observations (Kouyos et al 2007) and the current simulations have shown that strong linkage disequilibrium is built up and maintained. It turns out that here that an interaction governed by the MMAM can equilibrate to a region of high complementarity. The importance of this is that this equilibrium is such that any recombination among the loci will generate genotypes that have a lower fitness and recombination should be selected against.

I think that this recent paper (Kouyos et al 2009) sheds more light on specific potential microevolutionary mechanisms that drive the maintenance of recombination. We still need empirical test of some more of these new predictions. The challenge for empiricists is to find the right kind of systems and a challenge for the theoreticians is to help design the right kinds of experiments.

While I have just touched on a couple of recent results testing aspects of the Red Queen Hypothesis, Salathé et al (2008a) produced a wonderful review of many of many recent theoretical results on the evolution of sex and recombination driven by host-parasite interactions. In addition, this group has another paper on this topic out recently in the American Naturalist (Salathé et al 2009) that I'm looking forward to reading.

References

Kouyos, R., M. Salathe, and S. Bonhoeffer. 2007. The Red Queen and the persistence of linkage-disequilibrium oscillations in finite and infinite populations. BMC Evolutionary Biology 7:211.

Kouyos, R. D., M. Salathé, S. P. Otto, and S. Bonhoeffer. 2009. The role of epistasis on the evolution of recombination in host-parasite coevolution. Theoretical Population Biology 75:1-13.

Otto, S. P., and S. L. Nuismer. 2004. Species interactions and the evolution of sex. Science 304:1018-1020.

Salathé, M., R. D. Kouyos, and S. Bonhoeffer. 2008a. The state of affairs in the kingdom of the Red Queen. Trends in Ecology & Evolution 23:439-445.

Salathé, M., R. D. Kouyos, and S. Bonhoeffer. 2009. On the Causes of Selection for Recombination Underlying the Red Queen Hypothesis. The American Naturalist 174:S31-S42.

Salathé, M., R. D. Kouyos, R. R. Regoes, and S. Bonhoeffer. 2008b. Rapid parasite adaptation drives selection for high recombination rates. Evolution 62:295-300.

KOUYOS, R., SALATHE, M., OTTO, S., & BONHOEFFER, S. (2009). The role of epistasis on the evolution of recombination in host–parasite coevolution Theoretical Population Biology, 75 (1), 1-13 DOI: 10.1016/j.tpb.2008.09.007