A Game of Cat and Mouth

See allHide authors and affiliations

Science  17 Jan 2003:
Vol. 299, Issue 5605, pp. 353-354
DOI: 10.1126/science.1081278

Every genome contains within itself a record of its origin and evolutionary history. This record often can be deciphered with the help of the rich repertoire of molecular evolutionary theory. For infectious diseases, evolutionary changes that affect virulence, drug resistance, and evasion of the host immune system have major implications for public health. This issue of Science includes a remarkable example of evolutionary change in the life history strategy of a pathogenic parasite, Toxoplasma gondii (1). The three predominant and most widespread genotypes of T. gondii are known to have originated as the offspring of a single cross between two parental strains (2). On page 414 of this issue, Su, Evans, and collaborators (1) provide evidence that this cross took place only about 10,000 years ago. The progeny were evidently so successful because they acquired an efficient means of oral transmission that promoted their asexual dissemination from one vertebrate host to the next without the need for a detour through the sexual phase.

T. gondii is an apicomplexan protozoan parasite whose life cycle alternates between an intermediate host, in which the organism reproduces asexually, and a definitive host, in which the organism undergoes sexual reproduction. They are designated apicomplexans because of a characteristic cone-shaped configuration of microtubules at the tapering apical end of the cell. In the case of T. gondii, the intermediate host consists of a wide variety of warm-blooded vertebrates, whereas the definitive host is confined narrowly to the cat family Felidae. For transmission to humans, the most important definitive host is the domestic cat. Transmission occurs by handling soil or other materials contaminated with feces containing sporulated oocysts that infected animals excrete in great numbers for 2 to 3 weeks after infection. Transmission can also occur through contaminated water or by eating raw or insufficiently cooked meat that contains the organism in an encysted form.

Estimates based on the presence of antibodies to T. gondii suggest that 30 to 50% of the human population have been exposed to this organism. In the United States alone, this amounts to 90 to 150 million people. Fortunately, most infected people are asymptomatic and are also nontransmitters. However, the consequences of T. gondii infection can be severe among individuals who are extremely immunocompromised because of AIDS, chemotherapy, or immunosuppressive drug therapy. The clinical manifestations include enlargement of the lymph nodes, respiratory distress and heart disease, and disturbances of the central nervous system affecting vision and hearing. The second high-risk group are fetuses, which can be affected congenitally when the mother becomes infected during pregnancy. Annually more than 3000 newborns in the United States have congenital T. gondii infections, but fortunately most cases are asymptomatic.

About 95% of T. gondii infections are due to one of three strains designated types I, II, and III. Each strain type comprises an asexually propagated clonal lineage whose members are virtually identical in DNA sequence. Although types I, II, and III are genetically distinct, each strain carries one or the other of only two alleles at a large number of polymorphic gene loci (see the figure) (2). This observation suggests that the three strains are siblings—the recombinant offspring of a single mating that brought each pair of alleles together briefly in the diploid sexual phase.

Sibling unity.

Proposed origin of the three predominant types of T. gondii, types I, II, and III, which are sibling progeny of a single cross between two parental strains that took place 10,000 years ago (1, 2). The three strains, which are clonal and show little genetic variation, have only one of two possible alleles for genes at many different loci (depicted as two shades of the same color). From the alleles present in the three strains, the genotype of the diploid ancestor can be inferred.

Su et al. (1) reasoned that the clonality of the predominant types may result from successive oral transmission through intermediate hosts, bypassing the sexual phase of the life cycle. To test this hypothesis, they examined the ability of the predominant strains of T. gondii to be transmitted orally as compared with exotic strains. Exotic strains are genetically diverse and contain many unique polymorphisms, but they account for less than 1% of infections. When mice were fed tissue cysts of the predominant strains, 50 to 100% of the animals became infected. In contrast, two of three exotic strains showed inefficient oral transmission (0 to 10%). Taken at face value, these data suggest that as many as one-third of exotic strains may also show efficient oral transmission.

The authors argue that the fateful cross producing the three predominant clonal lineages may have brought together genes promoting efficient transmission by the oral route and perhaps other necessary adaptations as well. In almost all genetically diverse organisms, recombination can produce progeny with traits differing significantly from those of either parent. For example, a laboratory cross between type II and type III strains of T. gondii produced some progeny that were 1000 times as virulent as either parent (2). (A few percent of natural isolates of T. gondii are in fact recombinants between the clonal lineages.)

When did the fateful cross take place? To address this issue, Su et al. studied polymorphisms in noncoding DNA sequences (primarily introns) in 10 isolates of the predominant strains from diverse geographical locations. On the assumption that nucleotide mutations in noncoding DNA are selectively neutral (or nearly so), the number of unique polymorphisms that have arisen in each lineage since the time of the original cross is expected to increase in proportion to the mutation rate. Across 4067 base pairs in each of the 10 strains, the authors found only two new mutations. If the mutation rate in T. gondii is similar to that in the apicomplexan malaria parasite Plasmodium falciparum, then this result implies that the predominant clonal lineages diverged about 10,000 years ago. This estimate contrasts sharply with the time of divergence for exotic lineages (about 1 million years ago).

About 10,000 years ago, human society was undergoing one of its epochal transitions, from hunting and gathering to slash-and-burn agriculture. The population density was increasing, creating new opportunities for endemic or epidemic transmission of many kinds of parasites, and the cat was adopted as a companion animal. These conditions could well have favored strains of T. gondii that traded frequent sex in favor of rapid and efficient oral dissemination.

The finding of little genetic variation in each of the three predominant clonal lineages has important implications for public health. It means that drug resistance or immune evasion in these lineages must be acquired through the occurrence of new mutations, rather than through the selection of rare mutations that already exist. The waiting time for such mutations to occur may be substantial. On the other hand, the large amount of genetic variation present in the exotic strains, as well as in the recombinants between the predominant strains, implies that any sudden increase in the frequency of these strains should be viewed with alarm.


Navigate This Article