Experimental manipulation of immune-mediated disease and its fitness costs for rodent malaria parasites

Abstract

Background: Explaining parasite virulence (harm to the host) represents a major challenge for evolutionary and biomedical scientists alike. Most theoretical models of virulence evolution assume that virulence arises as a direct consequence of host exploitation, the process whereby parasites convert host resources into transmission opportunities. However, infection-induced disease can be immune-mediated (immunopathology). Little is known about how immunopathology affects parasite fitness, or how it will affect the evolution of parasite virulence. Here we studied the effects of immunopathology on infection-induced host mortality rate and lifetime transmission potential - key components of parasite fitness - using the rodent malaria model, Plasmodium chabaudi chabaudi.

Results: Neutralizing interleukin [IL]-10, an important regulator of inflammation, allowed us to experimentally increase the proportion of virulence due to immunopathology for eight parasite clones. In vivo blockade of the IL-10 receptor (IL-10R) with a neutralizing antibody resulted in a shorter time to death that was independent of parasite density and was particularly marked for normally avirulent clones. This suggests that IL-10 induction may provide a pathway to avirulence for P. c. chabaudi. Despite the increased investment in transmission-stage parasites observed for some clones in response to IL-10R blockade, experimental enhancement of immunopathology incurred a uniform fitness cost to all parasite clones by reducing lifetime transmission potential.

Conclusion: This is the first experimental study to demonstrate that infection-induced immunopathology and parasite genetic variability may together have the potential to shape virulence evolution. In accord with recent theory, the data show that some forms of immunopathology may select for parasites that make hosts less sick.

Figure 1

Effect of IL-10R blockade on the extent decreased survival time depended on P. c. chabaudi parasite genotype. C57BL/6J mice were administered neutralizing αIL-10R mAb or control IgG, 1 day before and on days 1, 2, 3 and 4 post infection with 10⁶ parasites of one of eight distinct P. c. chabaudi clones (CW, AS, AD, AQ, BC, AJ, AT or ER). Each pie-chart symbol represents the proportion of mice surviving until day 21 (white fill), while the position of the pie-chart along the y-axis indicates mean time to death in that treatment group (based on the mean of five mice ± S.E.). Day 21 was chosen as the end-point, to ensure that the acute phase of malaria infection was captured. Thus a mean of 21 without variance and with solid white fill indicates no mice died from that treatment group. Relative to control mice, αIL-10R treated mice suffered a higher mortality rate (~15% versus ~100%, respectively) and a quicker time to death (average time to death: 18.1 +/- 0.1 versus 7.5+/-0.2 days, respectively) and normally avirulent P. c. chabaudi clones were brought to lethality when IL-10R was blocked.

Figure 2

Total parasite densities were lower during IL-10R blockade, regardless of P. c. chabaudi parasite genotype. Line graphs represent the dynamics of total parasite density in αIL-10R (filled symbols) or control IgG (open symbols) treated mice during single-clone infections with each of eight P. c. chabaudi clones. Each line represents the mean of five mice (± S.E.), except where deaths had occurred. Regardless of P. c. chabaudi parasite genotype, neutralizing IL-10R resulted in lower total parasite densities between days 3 to 21 p.i. and lower early parasite densities (days 3–8 p.i.) compared to control mice.

Figure 3

Effect of IL-10R blockade on the relationship between virulence and asexual parasite density. Virulence, as measured in terms of mortality rate (1/day of death), is plotted against maximum parasite density for each of 8 parasite genotypes in αIL-10R (filled symbols) versus IgG control (open symbols) treated mice. Pearson correlation analyses revealed a significant negative relationship between virulence and asexual parasite density when IL-10R was blocked, but not in control mice. This suggests that the αIL-10R-driven increase in mortality was caused by uncontrolled inflammatory responses, and not increases in parasite load. Indeed, the same pro-inflammatory molecules that kill parasites also kill hosts [29]. Each large symbol represents the mean of five mice (± S.E) and small symbols represent individual mice.

Figure 4

The mortality caused by IL-10R blockade incurred a universal cost to lifetime transmission potential, but early in infection clones differed in their response to IL-10R blockade. (A) Line graphs compare the dynamics of gametocyte density during IL-10R blockade (filled symbols) or control IgG treatment (open symbols) during single-clone infections with each of eight P. c. chabaudi clones. Insets show day to death for each of the clones in αIL-10R (filled bars) versus IgG (open bars) treated mice. Bar graphs represent the least squares mean of total gametocyte density (days 4–21 p.i. inclusive, an estimate of lifetime transmission potential) (B), or early gametocyte density (days 3–8 p.i. inclusive) (C), broken down by treatment and P. c. chabaudi parasite clone. Each line or bar represents the mean of five mice (± S.E.), except where deaths had occurred. Regardless of P. c. chabaudi parasite genotype, blocking IL-10R reduced the lifetime transmission potential relative to control mice (A and B). However, the extent to which IL-10R blockade directly affected early gametocyte density depended on P. c. chabaudi parasite clone (C).