Open Access Highly Accessed Research article

Protein-based signatures of functional evolution in Plasmodium falciparum

Kate B Gardner12, Ipsita Sinha1, Leyla Y Bustamante3, Nicholas PJ Day1, Nicholas J White1 and Charles J Woodrow1*

Author affiliations

1 Wellcome Trust Mahidol University-Oxford Tropical Medicine Research Unit (MORU), 420/6 Rajwithi Road, Bangkok, 10400 Thailand

2 Centre for Infection, St. George's, University of London, Cranmer Terrace, London, UK SW17 0RE

3 Sanger Malaria Programme, Wellcome Trust Sanger Institute, Hinxton, Cambridge, UK CB10 1SA UK

For all author emails, please log on.

Citation and License

BMC Evolutionary Biology 2011, 11:257  doi:10.1186/1471-2148-11-257

Published: 14 September 2011

Abstract

Background

It has been known for over a decade that Plasmodium falciparum proteins are enriched in non-globular domains of unknown function. The potential for these regions of protein sequence to undergo high levels of genetic drift provides a fundamental challenge to attempts to identify the molecular basis of adaptive change in malaria parasites.

Results

Evolutionary comparisons were undertaken using a set of forty P. falciparum metabolic enzyme genes, both within the hominid malaria clade (P. reichenowi) and across the genus (P. chabaudi). All genes contained coding elements highly conserved across the genus, but there were also a large number of regions of weakly or non-aligning coding sequence. These displayed remarkable levels of non-synonymous fixed differences within the hominid malaria clade indicating near complete release from purifying selection (dN/dS ratio at residues non-aligning across genus: 0.64, dN/dS ratio at residues identical across genus: 0.03). Regions of low conservation also possessed high levels of hydrophilicity, a marker of non-globularity. The propensity for such regions to act as potent sources of non-synonymous genetic drift within extant P. falciparum isolates was confirmed at chromosomal regions containing genes known to mediate drug resistance in field isolates, where 150 of 153 amino acid variants were located in poorly conserved regions. In contrast, all 22 amino acid variants associated with drug resistance were restricted to highly conserved regions. Additional mutations associated with laboratory-selected drug resistance, such as those in PfATPase4 selected by spiroindolone, were similarly restricted while mutations in another calcium ATPase (PfSERCA, a gene proposed to mediate artemisinin resistance) that reach significant frequencies in field isolates were located exclusively in poorly conserved regions consistent with genetic drift.

Conclusion

Coding sequences of malaria parasites contain prospectively definable domains subject to neutral or nearly neutral evolution on a scale that appears unrivalled in biology. This distinct evolutionary landscape has potential to confound analytical methods developed for other genera. Against this tide of genetic drift, polymorphisms mediating functional change stand out to such an extent that evolutionary context provides a useful signal for identifying the molecular basis of drug resistance in malaria parasites, a finding that is of relevance to both genome-wide and candidate gene studies in this genus.