The Evolution of Plagues

      “A recurrent problem for all parasites … is how to get from one host to another in a world in which such hosts are never contiguous entities,” wrote the historian William McNeill. He went on: “Prolonged interaction between human host and infectious organisms, carried on across many generations and among suitably numerous populations on each side, creates a pattern of mutual adaptation to survive. A disease organism that kills its host quickly creates a crisis for itself since a new host must somehow be found often enough and soon enough, to keep its chain of generations going.” Based on this view, it would seem obvious that the longer the host lives, the greater the possibility for the parasite to grow, reproduce, and disperse its infective stages to new hosts. The conventional wisdom, therefore, is that the most successful parasites are those that cause the least harm to the host, and over time virulent parasites would tend to become benign.

At first glance, it would appear that the progress of the disease myxomatosis in Australia supports this evolutionary perspective. The story of myxomatosis begins in 1839, when the Austin family migrated from England to Australia. Over time they became rich from sheep farming. To reestablish their English environment, the Austins imported furniture, goods, and a variety of animals. In 1859 a ship came from England to Australia with rabbits. Since the rabbits had no natural predators, they multiplied rapidly, destroying plants and the native animals. The Austins began to wage war on the rabbits. By 1865, >20,000 rabbits were killed on the Austin estate. And still the rabbits continued to spread, traveling as much as 70 miles per year. Control measures such as fences, barbed wire, ditches, and the like did not work. A viral disease of wild rabbits from South America, called myxomatosis and lethal to domestic rabbits, was introduced into Australia in the 1950s to act as a biological control agent. The vector for the myxoma virus is a mosquito. In 1950, 99% of the rabbits died of myxomatosis. Several years later the virus killed only 90%, and it declined in lethality with subsequent outbreaks. It was also found that the viruses from the later epidemics were less virulent than the earlier forms and that these less virulent forms were much better at being transmitted by mosquitoes—the rabbits lived longer and the number of infected rabbits was higher with milder disease. One may conclude that the virus had evolved toward benign coexistence with the rabbit host. McNeill, impressed by the results of the introduction of the myxoma virus into Australia, wrote: “from an ecological point of view … many of the most lethal disease-causing organisms are poorly adjusted to their role as parasites … and are in the early stages of biological adaptation to their human host; though one must not assume that prolonged co-existence necessarily leads toward mutual harmlessness. Through a process of mutual accommodation between host and parasite … they arrive at a mutually tolerable arrangement … (and based on myxomatosis) … some 120-150 years are needed for a human population to stabilize their response to drastic new infections.” There is, however, reason to question McNeill’s conclusions.

      A recent reexamination of myxomatosis in Australia shows that the mortality of the rabbits, after the decrease in the virulence of the virus and the increase in rabbit resistance, was comparable to the mortality of most vector-borne diseases of humans, such as malaria. In other words, the virus was hardly becoming benign. Further, the decrease in virulence observed over the first 10 years of the study did not continue, but reversed. It appears that myxomatosis is not an example of benign evolution.

An alternative to the contention that parasites evolve toward a harmless state is that natural selection favors an intermediate level of virulence. This intermediate level is the result of a trade-off between parasite transmission and parasite-induced death. Since the value for R₀ increases with the transmission rate as well as the duration of the host’s infectiousness, an increase in transmission would reduce the duration of infection, and then selection may favor intermediate virulence. And because R₀ depends directly on the density of susceptible hosts, if the number of susceptible individuals is great, then a parasite may benefit from an increased rate of transmission even if this kills the host sooner and prevents transmission at a later time. If susceptible hosts are not abundant, however, then the parasite that causes less harm to the host (i.e., is less virulent) may be favored since that would allow the host to live longer, thereby providing more time for the production of transmission stages. The hypothesis that virulence is always favored when hosts are plentiful and is reduced when there are fewer hosts neglects the fact that a feedback exists in the host-parasite interaction: a change in parasite virulence impacts the density of the host population, and this in turn alters the pressures of natural selection on the parasite population, and so on. Thus, although parasite virulence generally tends to decline over evolutionary time, it never becomes entirely benign, and in the process the parasite population becomes more efficient in regulating the size of the susceptible host population.

      The view that parasites evolve toward becoming benign suggests that parasites are inefficient if they reproduce so extensively that they leave behind millions of progeny in an ill or dead host. Indeed, some biologists have contended that enhanced virulence is the mark of an ill-adapted parasite or of one recently acquired by the host. This is not true. The number of parasite progeny lost is not of evolutionary significance; rather, it is the number of offspring that pass on their genes to succeeding generations that determines evolutionary success. Natural selection does not favor the best outcome for the greatest number of individuals over the greatest amount of time, but instead favors those characteristics that increase the passing-on of a specific set of genes. Consider a particular species of weed that is growing in your garden. The production of 1,000 seeds that yields only 100 new weed plants might be considered wasteful in terms of seed death and the amount of energy the weed put into seed production, but if the surviving seeds ultimately yield more weed plants in succeeding generations, then that weed species is more efficient in terms of evolutionary success. Parasites are like weeds. They have a high biotic potential, and those that leave the greatest number of offspring in succeeding generations are the winners, evolutionarily speaking. Evolutionary fitness, be it for a parasite, human, bird, or bee, is a measure of the success of the individual in passing on its genes into future generations through survival and reproduction. When the fitness of the host is reduced by a parasite, there is harm, illness, and an increased tendency toward death. Host resistance is the counterbalance to virulence or the degree of harm imposed on the host by the presence of the parasite. If host resistance is lowered, a disease may be more pathogenic although the parasite’s inherent virulence may be unchanged. How negatively a host will be affected, i.e., how severe or how pathogenic is the disease, is thus determined by two components: virulence and host resistance. In addition, virulence is not so much a matter of a particular mutation but rather how that mutation is filtered through the process of natural selection; it is through natural selection that the final outcome may be a lethal outbreak or a mild disease, and, of course, when a new pathogen emerges, R₀ must be a number >1.

Since parasite survival requires reaching and infecting new hosts, effective dispersal mechanisms may require that the host become sick: sneezing, coughing, and diarrhea may assist in parasite transmission. The conventional wisdom is that it takes a prolonged period of time for virulence to evolve; the evolution of parasite virulence, however, may be quite rapid (on the order of months) and need not take years, as was the case with the myxoma virus. The basis for this is that a parasite may go through hundreds of generations during the single lifetime of its host. Then, too, because of competition between different parasites living in a single host, it might be advantageous for one kind of parasite to multiply as rapidly as it can before the host dies from the other infectious species. Succinctly, the victorious parasite is the one that most ruthlessly exploits the pool of resources (food) provided by the host and produces more offspring, thus increasing its chances to reach and infect new hosts.

      If parasite dispersal depends on the mobility of the host as well as host survival, then severe damage inflicted on the host by enhanced virulence could endanger the life of the parasite.

      Consider, for example, the common cold. It would be very much in the interest of the cold virus to avoid making you very sick, since the sicker you become, the more likely you are to stay at home and in bed; this would reduce