Desktop version

Home arrow Economics arrow American Trypanosomiasis Chagas Disease, Second Edition: One Hundred Years of Research


An indispensable recall of evolutionary genetics

Predominant clonal evolution: what does it mean?

Predominant clonal evolution (PCE) is the basic evolutionary model proposed for T. cruzi1 and other eukaryotic pathogens, with many clarifications and refinements recently proposed.2-7 Many misleading interpretations have been made, of the model designated by this term. Clarification is hence necessary. It can never be emphasized enough that clonality, according to this model, refers to all cases where offsprings have multilocus genotypes that are identical or extremely similar to parental lines, whatever the cytological mechanism of reproduction may be. PCE, in this genetical meaning, is synonymous with lack or rarity of genetic recombination (reassortment of genotypes occurring at different loci). This situation can originate from: (1) mitotic propagation; (2) several cases of parthenogenesis; (3) gynogenesis, hybridogenesis; (4) self-fertilization in the homozygous state; and (5) extreme cases of homogamy.

Mitotic propagation (1) is the usual case observed in many bacterial species and does occur in T. cruzi. However, PCE, again, has a much broader meaning. Cases (2) to (5) are able to generate genetic clones as well. Parthenogenesis (2) is observed in many insects, and even vertebrates.8 Specific cases of parthenogenesis (gynogenesis, hybridogenesis) are recorded in some fish and salamander species. Interestingly, gynogenetic and hybridogenetic females mate and therefore mimic the behavior of sexuality. Only the genetic analysis of their offspring can evidence that they actually generate clonal lines.8,9 Cases (4) and (5) have been presented as an alternative hypothesis to clonal evolution in the case of Leishmania parasites,10-13 while according to the PCE model, they obviously constitute only a specific case of it.2,14,15 Another recurrent source of misunderstanding comes from the fact that the PCE model does not by any means rule out occasional bouts of genetic exchange.16 It only stipulates that such events are rare and interfere only at an evolutionary scale. The PCE model therefore does not amount to absolute clonality, as sometimes claimed.17 It does not state either that its evolutionary and epidemiological impact is negligible (as misunderstood by some authors18-22), but only that it is not frequent enough to break up the predominance of clonal evolution.

Lack of, or severe restrictions to genetic recombination is the only, necessary, and sufficient criterion to settle the working hypothesis of broad-sense clonality (genetic clonality). This definition is broadly accepted by many scientists working on pathogen population genetics (see many examples cited in Refs. [2-7]). The relevant population genetics statistics is linkage disequilibrium (LD), or nonrandom association of genotypes occurring at different loci. By definition, linkage disequilibrium

evaluates the obstacles to recombination and is the only statistical approach able to do this. Unilocus segregation analysis based on Fis, Fst, Hardy—Weinberg equilibrium analysis,23 although very useful to explore mating strategies in depth, by its very nature cannot estimate the strength of recombination inhibition. Moreover, segregation tests are based on the hypothesis of diploidy, which has been recently questioned, since not only Leishmania, but also Trypanosoma cruzi have been hypothesized to undergo widespread aneuploidy. If this hypothesis is confirmed, tests based on diploidy will be irrelevant, while LD tests remain valid.

LD analysis is based on the very simple principle that the expected frequency of multilocus genotypes is the product of the observed frequencies of the unilocus genotypes they are composed of. For example, if two loci A and B are surveyed, and the observed frequencies of genotypes A1 and B2 are 0.3 and 0.5, respectively, the expected frequency of the bilocus genotype A1 + B2 is 0.3 X 0.5 = 0.15. When a large number of loci are surveyed, this analysis becomes very powerful, because the mere fact that a multilocus genotype is recorded more than once could become highly improbable. Calculating this by hand and simple chi-square analysis is possible but rapidly becomes cumbersome. Various indices have been published.16,38 Biases due to time or geographical separation (Wahlund effect) have been previously discussed.39

LD tests evidence severe obstacles to recombination. However, in some cases, they could be positive in situations where the genetic clones in a given species are ephemeral and soon disappear in the common gene pool of the species (epidemic clonality38). From an epidemiological and medical point of view, the important parameter to evaluate is the stability of the genetic clones in space and time. For this, LD analysis is usefully completed by two approaches: (1) direct observation and (2) phylogenetic analysis.

  • 1. Direct observation is done at a limited timescale. The T. cruzi strains have now been characterized for long enough to score recurrent observations of multilocus genotypes that have been repeatedly sampled for more than 30 years over vast geographical areas. This is in itself a strong indication of stable clonal propagation. Such observations cannot be made in highly recombinant pathogens such as Helicobacter pylori.40
  • 2. Phylogenetic analysis addresses a much longer timescale than population genetics and aims at reconstructing the evolutionary past of a given species over thousands or millions of years. This chapter is not the place to provide a comprehensive presentation of phylogenetic analysis. Many valuable textbooks have detailed the matter. Instead, I will present a few general principles that are specifically relevant for surveying the subspecific phylogenetic diversity of T. cruzi.
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >

Related topics