There's a nice Open Access article in PNAS from 2011 which investigates how Y. pestis, the bacterium which causes bubonic plague, can overtake the lungs so quickly. Y. pestis is a fairly recent evolutionary descendant of a larger family of Yersina bacteria. Many of these bacteria can jump to humans from their mammalian reservoirs, but Y. pestis is by far the deadliest. Y. pestis is very unusual in that infected lungs do not get inflamed for up to 36 hours after infection.
In the paper at the link, Price et al. show that Y. Pestis strongly suppresses the lung inflammatory response-- so that not just the plague bacterium, but other resident bacteria start to proliferate like crazy. They were not able to identify which gene or genes of Y. Pestis are responsible for this, but since they have close relative strains of Yersina bacteria, they should be able to figure it out soon. The anti-inflammatory action doesn't work in Y. Pestis mutants which lack the secretory pathway, suggesting that some combination of the known secreted toxins is important to suppress the immune system.
The authors also speculate about why Y. Pestis might have this ability while its recent ancestor does not. Y. Pseudotuberculosis is mainly a food-borne disease, meaning it may be present in large numbers in a prey animal and therefore ingested to successfully infect its next target. Y. Pestis may have been under selective pressure to successfully infect starting from extremely small numbers, such as would be available from a flea-bite. A strong suppression of the immune system would help these bugs reach critical mass-- a feature not required in the ancestor.