Characterisation of large and small subunit rRNA and mini-exon genes further supports the distinction of six Trypanosoma cruzi lineages

Abstract

It has been proposed that isolates of Trypanosoma cruzi, the agent of American trypanosomiasis, can be ordered into two primary phylogenetic lineages, first based on multilocus enzyme electrophoresis and random amplified polymorphic DNA, and subsequently based on the 24Sα rRNA and mini-exon genes. Recent multilocus enzyme electrophoresis and random amplified polymorphic DNA data have additionally shown that the major multilocus enzyme electrophoresis/random amplified polymorphic DNA lineage II is further subdivided into five smaller lineages, designated IIa–IIe. In this study, the precise correspondence between the multilocus enzyme electrophoresis/random amplified polymorphic DNA and rRNA/mini-exon lineages was investigated. Using the 24Sα rRNA and mini-exon markers in combination, five sets of strains were distinguished, corresponding to the multilocus enzyme electrophoresis/random amplified polymorphic DNA lineages I, IIa, IIc, IId and to lineages IIb/IIe together, respectively. The previous categorisation into only two primary lineages based on 24Sα rRNA and mini-exon characterisation is explained, in part, by the lack of representativeness of the breadth of T. cruzi diversity in earlier study samples. Additionally, a PCR assay based on a length-variable region of the 18S rRNA gene distinguished lineage IIe from lineage IIb. Thus, the six multilocus enzyme electrophoresis/random amplified polymorphic DNA lineages could be readily identified by combining data from the 24Sα rRNA, mini-exon and 18S rRNA characterisation assays, further supporting the relevance of these genetic units for T. cruzi strain classification and subspecific nomenclature. The recently proposed groups T. cruzi I and T. cruzi II correspond to multilocus enzyme electrophoresis/random amplified polymorphic DNA lineages I and IIb, respectively. Our findings show that T. cruzi lineage characterisation based on a single marker (either mini-exon or 24Sα rRNA) has insufficient resolution, and leads to important reinterpretations of recent epidemiological and evolutionary studies based on the oversimplified rRNA/mini-exon dichotomic classification of T. cruzi isolates.

Introduction

Trypanosoma (Schizotrypanum) cruzi is the aetiological agent of Chagas disease, which remains a major health problem in Central and South America, with an estimated 16–18 million infected (World Health Organization, 1991). Trypanosoma cruzi is widespread in wild mammals and vectors from the USA to the southern cone countries of South America. In humans, Chagas disease is characterised by a wide spectrum of clinical outcomes, ranging in severity from asymptomatic infections to severe cardiac and digestive tract damage.

Understanding the complex epidemiology and variety of pathogenic behaviours of T. cruzi requires a clear picture of the parasite's genetic diversity and easy strain characterisation schemes. The tremendous genetic diversity of T. cruzi populations was recognised long ago, based mostly on multilocus enzyme electrophoresis (MLEE; Miles et al., 1978, Tibayrenc et al., 1986), and their predominantly clonal population structure was demonstrated (Tibayrenc et al., 1986, Tibayrenc and Ayala, 1988). Successive uncoordinated proposals to describe T. cruzi genetic variants have resulted in a confusing profusion of independent descriptions (Momen, 1999). In recent years, a dominant focus has been put on the subdivision of T. cruzi into two primary phylogenetic lineages (Tibayrenc, 1995), initially based on MLEE and random amplified polymorphic DNA (RAPD) data (Tibayrenc et al., 1993). A similar conclusion was drawn after analysis of the 24Sα rRNA genes and the non-transcribed spacer of the mini-exon genes (Souto et al., 1996), of the promoter regions of the rRNA and mini-exon genes (Nunes et al., 1997), and of the 195 bp DNA repeat (Bastrenta et al., 1999). This apparent convergence lead to efforts to harmonise T. cruzi lineages denomination (Anonymous, 1999). However, the exact correspondence between the MLEE/RAPD lineages (Tibayrenc, 1995) and the rRNA/mini-exon lineages (Souto et al., 1996) has never been carefully checked by parallel analysis of a representative set of strains. In particular, it is important to realise that MLEE/RAPD lineage 2 was defined in a broader way than the corresponding rRNA/mini-exon lineage (Table 1). Moreover, recent MLEE and RAPD analysis of a larger set of T. cruzi isolates demonstrated the distinction of five lower phylogenetic subdivisions, designated IIa–IIe, within the MLEE/RAPD lineage 2, whereas no clear subdivision was found within the first lineage (Barnabé et al., 2000, Brisse et al., 2000a). Thus, six phylogenetic subdivisions can be distinguished within T. cruzi, but the correspondence between this refined MLEE/RAPD classification scheme and the two major rRNA/mini-exon lineages is unknown, rendering it very difficult to interpret epidemiological or evolutionary studies based on the widely used rRNA/mini-exon characterisation. In order to investigate this correspondence and help further harmonisation of T. cruzi infra-specific nomenclature, and to determine whether the six lineages can also be demarcated by these markers, we analysed representative strains of each of the six RAPD/MLEE lineages using the 24Sα rRNA and mini-exon PCR markers (Souto et al., 1996). In addition, we investigated the relevance of the size polymorphism of the 18S rRNA genes (Clark and Pung, 1994) for lineage identification.