Thursday, March 26, 2009

Why doesn’t this pathogen kill me and why is it taking so long to clear?

This week the Coevolvers read a brand new paper by King et al (2009). The authors present a pathogen model that incorporates within host dynamics of pathogen growth as well as multiple forms of transmission among hosts which depend on pathogen load. The authors do motivate the study by telling us about two human disease pathogens, Bordetella pertussis and Bordetella parapertussis (which can cause whooping cough), but model is not meant to be a predictive model of future outbreaks. The main message of the paper is that including within host dynamics in conjunction with SIR models of populations leads to a better understand of disease evolution. Mideo et al (2008) wrote a recent review on including within host dynamics in evolutionary epidemiological models for more general information on this approach.

While the outline of the model was well written, how they combined the multiple different parts was unclear. The model consisted of three components: 1) within host pathogen replication 2) dose dependent transmission and 3) between host/SIR type model. What we found hard to understand was how the model incorporated the variation in pathogen loads among the hosts into the overall transmission rate. It appeared as if the model integrates over a number of classes of hosts (depending on their age of infection), but we felt that this then removed quite a bit of the variation that was being captured by including within host dynamics. A simplifying assumption that the authors made also was that each host was always infected with the same dose of pathogens and that their immune system had to be restarted each time. The authors do state that they have already worked on a stochastic model of this system which hasn't yet been published. We are very interested on the quantitative results from that analysis since some of these problems could be addressed there.

Why not make a population genetics model to address the questions posed by the authors at the beginning of the paper. This was question stimulated by our previous reading of Boots et al (2009) and Day and Gandon (2007) that provide detailed reviews of different modeling approaches as well as addressing specific problems in evolutionary epidemiology. King et al (2009) present their results of how intermediate within host pathogen growth rates can maximize R0 under some transmission models, but what they don't do is present an analysis where they look at how different pathogens might evolve. Is the intermediate growth rate a stable strategy? Given the model framework, there might be complex interactions between different pathogens mediated through hosts. Higher growth rates of an aggressive pathogen could lead to a tragedy of the commons.


Boots, M., A. Best, M. R. Miller, and A. White. 2009. The role of ecological feedbacks in the evolution of host defence: what does theory tell us? Philos. Trans. R. Soc. B-Biol. Sci. 364:27-36.

Day, T and S Gandon. 2007. Applying population-genetic models in theoretical evolutionary epidemiology. Ecology Letters 10 (10), 876–888.

King, A. A., S. Shrestha, E. T. Harvill, and O. N. Bjørnstad. 2009. Evolution of Acute Infections and the Invasion-Persistence Trade-Off. The American Naturalist 173:446-455.

Mideo, N., S. Alizon, and T. Day. 2008. Linking within- and between-host dynamics in the evolutionary epidemiology of infectious diseases. Trends in Ecology and Evolution 23(9): 511-517.

Paper read:

King, A., Shrestha, S., Harvill, E., & Bjørnstad, O. (2009). Evolution of Acute Infections and the Invasion‐Persistence Trade‐Off The American Naturalist, 173 (4), 446-455 DOI: 10.1086/597217

Tuesday, March 17, 2009

Parasites can maintain host diversity

In their recent paper, Morgan et al (2009) look at the role of an antagonistic interaction in promoting coexistence among different hosts. Using a bacteria and phage system (bacterium: Pseudomonas fluorescens and bacteriophage: SBW25Φ2), they determined that in the presence of a coevolving phage, a slower growing, but phage resistant host would persist with a susceptible faster growing host. Without the phage, the better competitor became fixed in experimental lines. This paper did not explicitly demonstrate coevolution between the phage and the bacterial host, however there is previous evidence in this system for reciprocal selection (Buckling and Rainey 2002). The point of this experiment was to demonstrate the cost of parasite resistance.

The authors also presented a second hypothesis that was a little less explicit: "the probability of coexistence would alter through time." This general hypothesis was supported and the authors provided several explanations for the fitness of the resistant mutant changing over time with respect to wild type. They narrow down the field to changes in the cost of resistance and compensatory mutations. Their evidence comparing growth rates from the beginning to the end of the experiment support a change in the cost, but I wasn't completely convinced that this ruled out compensatory mutations.

A disappointing portion of this system is a lack of understanding of the mechanism of phage resistance. This is no fault of the authors, as the paper is just the beginning of an investigation. They have some details about a general reaction (production of "cellulose-like polymer"). It would be very interesting to take this system to the next step and start targeting some genes


Morgan, A. D., R. C. Maclean, and A. Buckling. 2009. Effects of antagonistic coevolution on parasite-mediated host coexistence. J Evolution Biol 22:287-292.

Buckling, A. and Rainey, P.B. 2002. Antagonistic coevolution between a bacterium and a bacteriophage. Proc. R. Soc. Lond. B Biol. Sci. 269: 931–936.

MORGAN, A., CRAIG MACLEAN, R., & BUCKLING, A. (2009). Effects of antagonistic coevolution on parasite-mediated host coexistence Journal of Evolutionary Biology, 22 (2), 287-292 DOI: 10.1111/j.1420-9101.2008.01642.x

Friday, March 13, 2009

Why do I have so many parasites?

This week's paper (Bordes et al 2009) looked for forces that influence the parasite diversity or parasite species richness (PSR) among mammals. While it may seem almost impossible to think that there might be a single factor, there have been many different proposed influences (e.g. body size, geographic range, population density). The host home range, "area used in daily and seasonal movements" (Bordes et al 2009), could be related to the parasite diversity in two distinct ways. Their first prediction is that as home range increases so will PSR because it will result in an increased contact with diverse habitats (and therefore parasites). Their "spatial dispersion model of parasite acquisition" uses parasite transmission and host density to get this relationship to predict the opposite relationship. The results of their analyses supported the second prediction.

The group found the methods and results of this paper relatively straight forward. The use of independent contrasts to control for the effect of phylogeny was very appropriate in this paper. We did find one area of the analysis confusing with respect to the host sampling number. It is well known that sampling intensity may bias the number of parasites found on a host . The more hosts one searches the more parasites will be found up to some saturation point. The authors controlled for the bias by using the residuals of parasite richness and host number. However, we were then confused by the inclusion of "Host sample size" in the regression analyses. While other variables in the regression analyses were significant, it was hard to determine the impact of this highly significant variable on the total fit of the model. We were left wondering how much of the variation in PSR did the home range explain?

The main conclusion of this paper is to confirm a roll for epidemiological factors (density and transmission) on the relationship between home range and PSR. The results show a negative relationship between home range and PSR which is consistent with their second prediction. The strong negative relationship between home range and host density relates their effect to how this can influence the number of parasites. It seems that home range not only describes a complicated trait of a host species, but is perhaps influenced by a complicated set of other factors.

Today's group speculated on broader potential relationships of host traits and parasite diversity. Could there be a more universal law that predicts the parasite species load? This study and many of the citations contained within focus on animals and their macroparasites. Maybe there is a rule that works across such taxonomic divisions? The paper cited previous work on the role of body mass (Arneberg 2002, Lindenfors et al 2007) and parasite diversity with larger hosts being home to a larger number of parasites. Can this relationship be scaled up to incorporate host density? What about the total mass of a host species?


Bordes, F., S. Morand, D. A. Kelt, and D. Van Vuren. 2009. Home Range and Parasite Diversity in Mammals. The American Naturalist 173:467-474.

Arneberg, P. 2002. Host population density and body mass as determinants of species richness in parasite communities: comparative analyses of directly transmitted nematodes of mammals. Ecography 25:88–94.

Lindenfors, P., C. L. Nunn, K. E. Jones, A. A. Cunningham, W. Sechrest, and J. L. Gittleman. 2007. Parasite species richness in carnivores: effects of host body mass, latitude, geographical range and population density. Global Ecology and Biogeography 1:1–14.

Bordes, F., Morand, S., Kelt, D., & Van Vuren, D. (2009). Home Range and Parasite Diversity in Mammals The American Naturalist, 173 (4), 467-474 DOI: 10.1086/597227

Evolution of virulence revisited

In their recent paper, Vigneux et al (2008) address a classic idea in the evolution of virulence. When multiple genotypes of a parasite infect a single host, competition can influence the overall virulence. The paper is examines the interaction of relatedness and virulence. One viewpoint is that with a low level of relatedness, virulence should increase as competition among genotypes overexploits the host. Another hypothesis that the authors test is that different genotypes may engage in a "chemical warfare" inside the host. This would lead to a decrease in virulence as relatedness decreases.

Their overall results are completely consistent with their second hypothesis, increases in virulence with increases in relatedness as mediated through limited migration. Their evidence is that the host shows a quicker mortality in the low migration treatment. More compelling at least in gaining evidence for the role of interference competition is their growth inhibition assay. Bacterial clones from the low migration treatment did not inhibit the growth of other clones from the same host. When the authors tested clones from different hosts did still possessed some ability to inhibit growth.

While the details on the infection protocol in this paper seemed to make the results a little harder to understand, they did gain evidence that clearly support the role of interference competition on virulence. The proposed mechanism seems sound, but obviously could use further investigation. I initially misunderstood the role of migration in this experiment. To my understanding the effect of their different treatments was to reduce the variation among genotypes and increase the relatedness. Previous arguments about the role of transmission and virulence are not completely appropriate in the context of this experiment. Some of the discussion among our group focused on the role of kin selection in the evolution of greater virulence.

Some extra details: This experiment uses a rather complex host-parasite interaction consisting of a nematode (Steinernema carpocapsae) that is a parasite in insect larvae. However, unlike a previous paper (Bashey et al 2007) focusing on the nematode, here the main focus is a symbiotic bacterium of the nematode (Xenorhabdus nematophila) that along with the nematode induces mortality in the insect host. X. nematophila is also known to produce bacteriocins which inhibit the growth of other genotypes. Over the course of 20 host passages, the authors construct two types of experimental treatments. In one treatment (high migration), parasites from several lines are mixed together creating an infection containing bacteria. In the second treatment (low migration), the majority of parasite were transferred from a single host line. These two treatment setup a contrast of the potential relatedness of the bacteria in the current host.


Bashey, F., Morran, L.T. & Lively, C.M. 2007. Coinfection, kin selection, and the rate of host exploitation by a parasitic nematode. Evol. Ecol. Res. 9: 947-958.

VIGNEUX, F., BASHEY, F., SICARD, M., & LIVELY, C. (2008). Low migration decreases interference competition among parasites and increases virulence Journal of Evolutionary Biology, 21 (5), 1245-1251 DOI: 10.1111/j.1420-9101.2008.01576.x