Host-Pathogen Interaction

Infection by Toxoplasma, at least of mice, results in an acute infection with a major immune response. Interestingly, what happens next ranges between two extremes. In many instances, the animal successfully controls the acute stages but not to the point of a complete cure; they instead become persistently infected for the life of the animal with few if any real symptoms. At the other extreme, the acute infection can overwhelm the animal resulting in death within a matter of days. Giving away a bit of the punch-line, this turns out to be heavily dependent on which strain of Toxoplasma is doing the infecting and we are interested in the following questions that relate to all this:

  1. What is the genetic relationship and population biology of the different strains of Toxoplasma?
  2. What is responsible for their different virulence?
  3. How is the acute stage of infection normally controlled by the host?
  4. Why does the very effective immune response during the acute stages not completely clear the infection.



Below is how things have progressed...

We began, several years ago, by looking at the genetic relatedness between strains as a way of addressing the question of why Toxoplasma infections are so variable in their severity. We were very surprised to find that all 9 mouse-virulent strains that we looked at from various parts of the Americas and Europe were apparently one clonal line eventually dubbed “type I” (Sibley and Boothroyd, 1992).

We went on to show that a supervirulent line could also emerge from a mating between two nonvirulent strains (Grigg et al., 2001). This indicated that virulence may simply be a result of “the shake of the genetic dice” rather than something intrinsic that is due a collection of special genes that are only found in virulent lines (as is the case with some bacteria).

Building on the massive effort of many labs (notably Sibley, Roos, Ajioka, Sanger Center and TIGR) to create databases of sequences from different strains, we began to ask what is the relationship between naturally occurring strains, including the mouse-virulent type I. This yielded the third surprise: one of the most common strains (“type II”) appears to be a parent (or maybe grandparent) for both type I and type III (Boyle et al., 2005). In fact, it looks like literally just two or three cats in the not too distant past have yielded some Toxoplasmarecombinants that have had “just the right mix” of alleles to be super-fit in today’s world.

To figure out what makes these strains so special, we went back to the cross mentioned above that yielded some supervirulent F1 progeny from two nonvirulent parental lines. We applied traditional genetic approaches to map the genes responsible for this virulence in mice. My bias was that this would be a very complex trait with many, many genes involved and, on top of that, they would be interacting in nightmarishly complex ways. Fortunately, others in the lab had more optimistic views and we found that just 5 loci appeared to be key to much of the differences in the virulence of the F1 progeny (Saeij et al., 2006). These were mapped and we have identified the responsible genes in a couple of instances. These turn out to encode polymorphic protein kinases that are injected into the host cell. Now we just have to figure out what these kinases do and why different allelic forms have such a huge impact on disease outcome.

As a complementary approach to identifying genes important to the host-pathogen interaction, we also looked at how different strains impact host gene expression. We knew from previous work (Blader et al., 2001) that, not surprisingly, Toxoplasma infection has a major impact on the infected host cell. The surprise came when we looked and saw major strain-specific differences in this regard. So, going back, to the ever-useful cross described above we asked if these differences were mappable among the F1 progeny. Fortunately, they were and one of the major parasite genes involved co-segregated with one of the virulence genes. This turned out to be another rhoptry protein kinase, ROP16. This protein somehow intersects key signaling pathways in the host cell (STAT3 and STAT6) with the outcome that, depending on which allele of ROP16 a strain possesses, the immune response is either appropriate or excessive; in the latter case, it can be to the point where it may actually be lethal to the host. We still don’t know the actual substrate for this and the other kinases mapped by these approaches and that quest is described further below. [We are also now extending our analyses to include how different strains of Toxoplasmaimpact host microRNAs and how this might feature in strain-specific differences in virulence.]


One final piece to the puzzle is that these kinases are encoded by extensive gene families that appear to be rapidly evolving, perhaps in response to a changing ecology for the parasite (due to human efforts that have radically altered the host landscape by moving sheep from Scotland to New Zealand, rats from Europe to everywhere, infected cats from everywhere to everywhere, etc.)! A couple of fascinating but almost impossible to address questions now are what species is the natural host for these different strains and what factors led to their emergence as the dominant strains we see today. Any ideas...?