The way this question is typically being answered is by giving a long and tedious exposition on the operation of virulent factors of Yersinia pestis, the bacteria that causes the malady. This bacterium is a nasty piece of work, because it kills the very macrophages that normally destroy pathogenic bacteria. It injects a factor into the macrophages that subverts their own capsases (that activate regulated death of our body cells, the apoptosis) into killing these cells. Meanwhile, the bacterium joyrides the dying macrophages back to the lymph nodes, where Y. pestis multiplies unharassed and then gets out to infect the blood. And yet this is the clear-cut case when "how" does not help to understand "why." The "how" only tells how Y. pestis kills mammals, which is not too important, because killing a mammal is no brainer for a bacterium. Any bug can do that with very little effort. The skill of the pathogen is precisely in NOT killing its host while using its resources and putting other bugs out of business. Y. pestis is not doing its job right. The question is why.
The first thing to realize is that Y. pestis is a very young bacterium that evolved from a much less virulent bacterium, Y. pseudotuberculosis, less than 20 kya. Both of these bacteria are not interested in humans; their targeted hosts are wild rodents, like marmots. Y. pseudotuberculosis lives in the gut and does everything right: it is a mild infection that does not bother the animal greatly, but gives the bacterium a beachhead in the most desirable part of the host, where nutrients are aplenty; it makes energy by fermentation. Alas, there is harsh competition there between various bacteria in the gut, and Yersinia is not doing particularly well; E. coli is doing much better. As all other bacteria in the gut, Yersinia's chosen method of infecting mammals is oral-fecal. Y. pestis is a daughter strain which is 97% identical (genetically) to its mother strain. Y. pestis is trying something new: leaving the conjested gut and colonizing the body, where there is less competition. Another innovation is transmission through blood. To do that, Y. pestis relies on an insect vector: the flea. The flea bites a mammal, the bacteria infect the flea, and then infect another mammal. These mammals are not supposed to be humans, as humans have too low immunity to Y. pestis, die rapidly, and so they are of little use; we are collateral damage. The supposed host is a rodent that has reasonably strong immunity to Yersinia, as the rodents have been co-existing with it from time immemorial. Y. pestis' problem is that it chose extremely inefficient strategy of spreading across rodent population and to compensate for this poor efficiency it must be supervirulent. As at this level of virulence it is difficult to adjust virulence just right, the bacteria begins to kill both of its hosts without much benefit to itself. It is a failure.
...The genus Yersinia consists of 15 species, and only three of them, Y. pestis, Y. pseudotuberculosis, and Y. enterocolitica, are pathogenic to mammals. The Y. pseudotuberculosis-Y. pestis evolutionary linkage diverged from Y. enterocolitica between 41 and 186 million years ago, while Y. pestis diverged from Y. pseudotuberculosis within the last 1,500 to 20,000 years. Very close genetic similarity is found between Y. pseudotuberculosis and Y. pestis. Y. pseudotuberculosis causes only nonfatal gastrointestinal disease in mammalian hosts, including humans, and the disease is transmitted by the food-borne route. Y. pestis causes plague, which is one of the most deadly diseases. Three pandemics of plague have been recorded in human history and have claimed hundreds of thousands of lives. Plague is a typical enzootic disease (an infection of the animal population[s] in one or more confined natural foci without the need for external inputs), and epidemics of rodent plague are restricted in various enzootic plague foci especially in Asia, the Americas, and Africa. Compared to its progenitor Y. pseudotuberculosis, Y. pestis utilizes a radically different mechanism of transmission in rodent reservoirs that relies primarily upon biting by flea vectors.
Is it at least using its flea vectors efficiently? Absolutely not:
...arthropodborne transmission is an evolutionarily recent adaptation in Yersinia pestis. We show that the infectivity of Y. pestis to its most proficient vector, the rat flea Xenopsylla cheopis, and subsequent transmission efficiency are both low. The poor vector competence of fleas likely imposed selective pressure that favored the emergence and continued maintenance of a hypervirulent Y. pestis clone. In particular, the rapidly fatal gram-negative sepsis that typifies plague is a consequence of the high threshold bacteremia level that must be attained to complete the transmission cycle. (J Infect Dis. 2005 Jun 1;191(11):1907-12, Poor vector competence of fleas and the evolution of hypervirulence in Yersinia pestis.)
The bacterium needs to infect blood massively ("bacteremia") to make itself available to its flea vectors. This causes septic reaction that easily kills its host before the fleas have a chance to bite. Furthermore, when the flea bites, this is not enough, because it could take a long time before it will bite again. To force it to bite again, Y. pestis makes polysaccharide biofilm in the gut of the flea that is blocking the digestion, so the flea feels hungry and bites again and again. Blocking a flea takes full two weeks, slowing the infection down. Even this ingenious trick is not working too well, as the infection rate is low. Worse, there are emerging fleas that cannot be blocked, as it is not only Y. pestis that is evolving. In fact, how plague had been spreading during the pandemics is a puzzle, precisely due to these low rates of infection. One idea is that rapid transmission may occur throughthe fleas that are not being blocked:
...Dissemination of Y. pesti by blocked fleas has been the accepted paradigm for flea-borne transmission. However, this mechanism, which requires a lengthy extrinsic incubation period before a short infectious window often followed by death of the flea, cannot sufficiently explain the rapid rate of spread that typifies plague epidemics and epizootics. Inconsistencies between the expected rate of spread by blocked rat fleas and that observed during the Black Death has even caused speculation that plague was not the cause of this medieval pandemic. We used the primary vector to humans in North America, Oropsylla montana, which rarely becomes blocked, as a model for studying alternative flea-borne transmission mechanisms. Our data revealed that, in contrast to the classical blocked flea model, O. montana is immediately infectious, transmits efficiently for at least 4 d postinfection (early phase) and may remain infectious for a long time because the fleas do not suffer block-induced mortality. These factors match the criteria required to drive plague epizootics as defined by recently published mathematical models. (PNAS 2006 103(42):15380-5. Early-phase transmission of Y. pestis by unblocked fleas as a mechanism explaining rapidly spreading plague epizootics.)
Y. pestis can handle mammals, but dealing with fleas and interaction between the fleas and the mammals is well over this bug's head. It is just beginning to learn the tricks. In fact, even its chief skill - making biofilm in the gut of the flea is undeserved luck. This skill is not the result of some biochemical wizardry, but rather the consequence of deleterious mutation (!) knocking down a particular gene. How this gene affects the biosynthesis of the biofilm is a mystery, but this is all that was required to block a flea. [All yersiniae use this biofilm to protect themselves from phagocytes in their rodent hosts.]
...The rapid evolutionary transition of Y. pestis to flea-borne transmission involved at least three steps: acquisition of the two Y. pestis-specific plasmids by horizontal gene transfer; and recruitment of endogenous chromosomal genes for new functions. Perhaps reflective of the recent adaptation, transmission of Y. pestis by fleas is inefficient, and this likely imposed selective pressure favoring the evolution of increased virulence in this pathogen. (Curr Issues Mol Biol. 2005 7(2):197-212. The evolution of flea-borne transmission in Yersinia pestis.)
...Not all infected fleas become blocked, and unblocked fleas are also capable of spreading disease, but it is nevertheless believed that blockage is crucial for long-term enzootic persistence of Y. pestis. Blockage is due to a biofilm that binds tightly in the proventriculus, an organ in the flea digestive tract; this makes the infection less susceptible to clearance by peristalsis and defecation and thus enhances high-titer colonization for extended periods. The in vivo biofilm, consisting of bacteria in a self-synthesized, exopolysaccharide-rich matrix, requires the hmsHFRS operon of polysaccharide biosynthetic genes. Y. pseudotuberculosis contains a functional hmsHFRS operon, but it is unable to make biofilms in fleas (11). This suggests that the different biofilm capabilities in fleas are due to functions other than exopolysaccharide (EPS) biosynthesis. Bacterial phosphorelays are signal transduction pathways. A sensor histidine kinase autophosphorylates, then transfers the phosphate to an aspartate in the receiver domain of a response regulator. The Rcs phosphorelay is conserved in E. coli and Y. pestis.
In this report, we show that Y. pestis rcsA is a pseudogene. Substitution with the functional homologue from Y. pseudotuberculosis abolishes biofilms in fleas, evidence that the Y. pestis pseudogene arose by negative selection.
To round it up, Y. pestis kills its mammalian hosts because it chose singularly inefficient strategy for transmission and even this strategy is executed very sloppily. The bacteria harm their own vectors so that they cannot even infect their mammalian hosts at high rate. In a futile effort to improve transmission, it induces sepsis putting the life of its host in grave danger. And it seems to be doing it all for only one reason: it is trying to avoid the competition in the gut, where its mother strain has never been able to outcompete other bacteria. Y. pestis is the evolutionary dead end that cannot put its act together, going from bad to worse. Alas, before it puts itself out of misery, it is fully capable of bringing all of us down. Our main hope is that the fleas on the rodents will learn how to break down the biofilm, as some of them have already been quitely doing. Then Y. pestis will be memory.
Why does the bubonic plague kill?