The bacterium Yersinia pestis (pictured) is notorious as the cause of bubonic plague. When it is breathed in, however, it also causes the rarer but deadlier pneumonic plague. The pathology of this disease in humans and animals is fairly well understood, but much less is known about the earliest stages. Wyndham W. Lathem and colleagues (Proc. Natl Acad. Sci. USA 102, 17786–17791; 2005) have developed a mouse model of pneumonic plague that gives perspectives on these stages as experienced by the host and by the bacterium.

The team infected mice with Y. pestis through the nose, and the animals developed a disease that closely resembled pneumonic plague in humans. Bacterial numbers in the animals' lungs increased massively in the first 24 hours after infection. Yet when the authors studied the levels of inflammatory molecules normally produced during an immune response, there was little change during this period. So the bacteria must have a potent anti-inflammatory activity that allows them to become established before the host immune system detects them. After 48 hours, the levels of inflammatory molecules escalated, showing that the mouse immune response does eventually kick in; but it would seem to be too little, too late.

And what happens in Y. pestis? A microarray analysis showed that there is a change in the expression of about 10% of the bacterium's genes after it infects its host. Notably, many of these genes are associated with virulence, and in particular with the so-called type III secretion system. This system was already known as a potential means for the bacterium to subvert its host's immune system by altering the types and amounts of inflammatory molecules. That the expression of genes for the system is increased in the mice confirms this animal model as biologically valid. Moreover, a comparison with in vitro studies showed that the regulation of this system is more complex in vivo — suggesting that the model will provide greater insight into this devastating infection.