If Mice Are Men...

...then what we think makes community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) so infectious might be wrong. A to-be published article in the Journal of Infectious Diseases (Dec. 15) asks "Is Panton-Valentine Leukocidin the major virulence determinant in community-associated methicillin-resistant Staphylococcus aureus disease?" Before everyone wanders off due to boredom, let me translate: we don't really understand how CA-MRSA can kill you.

The study, conducted by Jovanka Voyich and colleagues, examined the assumption that the Panton-Valentine Leukocidin toxin ('PVL') is what causes CA-MRSA infection. Epidemiological studies have found that PVL is strongly correlated with disease. However, the authors engineered identical CA-MRSA that can not produce PVL, and found that in mice, there is no effect of the presence or absence of PVL to cause disease in either bloodstream infections or skin infections (in the case of bloodstream infections, 'disease' means the mousey dies).

What's particularly interesting is that the function of PVL, which is to lyse human neutrophils when the neutrophils engulf the bacteria, does not appear to be affected by the presence or absence of PVL. While there are some caveats (for example, necrophilizing pneumonia wasn't examined), it seems pretty solid. The authors suggest that a gene (or genes) linked to PVL are responsible for causing disease; PVL might just be along for the ride with the actual cause of disease.

This study highlights the importance of follow-up functional studies. While epidemiology can reveal plausible hypotheses, many of them must be, when possible, tested with good experiments. Also, it's interesting to note that even though PVL might not do anything, it's still an excellent marker for diagnosing the potential to cause disease.

More like this

YOU are a mouse.

YOU moderate your comments but pretend like you don't.

Pussy man.

In your experience (and I don't know if you have any in this area, so please let us know) have studies of S. a. virulence in mice been useful? Because we did our stuff in rabbits for the simple reason that TSS (toxic shock syndrome) doesn't happen in mice, period. Another colleague of mine has told me informally that he cannot inoculate a wounded mouse with enough staph to kill it. So I'm wondering if this is a particularly good model. On the other hand, I always thought PVL was an unlikely virulence factor, and since it is phage borne, I thought maybe the relationship was genetic rather than functional.

By Paul Orwin (not verified) on 14 Nov 2006 #permalink

I haven't done experimental work on SA, but I know the mouse skin infection model is supposed to be pretty good, and is widely used. As to sepsis, no idea; I would hope JID wouldn't publish something that shoddy, if mice can't have TSS. They did seem to be able to kill mice rather well.

I know (by reputation) that if Barry Kreisworth wrote it, it's right :) (yes, argued from authority, but he's a pretty good authority on SA). I don't think that this is shoddy work, but I do think that it probably isn't the last word on CA-MRSA pathogenesis. The big virulence issue with CA-MRSA has been Necrotizing Fasciitis (previously seen primarily associated with Group A Strep, rather than Staph, and associated with the Pyrogenic Exotoxins). The curiosity about the CA-MRSA is that they don't seem to produce any of the standard SA superantigens, such as TSST-1 or SEA/B/C. There are lots of other enterotoxins in the Staph genome (I should know :)) but none of them have been significantly associated with disease.
Overall, though I'm not sure that mice are good models for staph (even though they are widely used). I'd have to read the paper (no sub to JID at my school, unfortunately). However, they are an accepted disease model system, and I'm sure that there are arguments to be made in their favor. I also don't think that they were surprised by the results....

By Paul Orwin (not verified) on 15 Nov 2006 #permalink

Very much! If you wouldn't mind sending it to me. Thanks

By Paul Orwin (not verified) on 15 Nov 2006 #permalink