Analysis of sequence diversity at the highly polymorphic Cpgp40/15 locus among Cryptosporidium isolates from human immunodeficiency virus-infected children in South Africa

Abstract

Cryptosporidium sp. is a significant cause of diarrheal disease, particularly in human immunodeficiency virus (HIV)-infected patients in developing countries. We recently cloned and sequenced several alleles of the highly polymorphic single-copy Cryptosporidium parvum gene Cpgp40/15. This gene encodes a precursor protein that is proteolytically cleaved to yield mature cell surface glycoproteins gp40 and gp15, which are implicated in zoite attachment to and invasion of enterocytes. The most-striking feature of the Cpgp40/15 alleles and proteins is their unprecedented degree of sequence polymorphism, which is far greater than that observed for any other gene or protein studied in C. parvum to date. In this study we analyzed nucleic acid and amino acid sequence polymorphism at the Cpgp40/15 locus of 20 C. parvum isolates from HIV-infected South African children. Fifteen isolates exhibited one of four previously identified genotype I alleles at the Cpgp40/15 locus (Ia, Ib, Ic, and Id), while five displayed a novel set of polymorphisms that defined a new Cpgp40/15 genotype I allele, designated genotype Ie. Surprisingly, only 15 of these isolates exhibited concordant type I alleles at the thrombospondin-related adhesive protein of Cryptosporidium and Cryptosporidium oocyst wall protein loci, while five isolates (all of which displayed Cpgp40/15 genotype Ic alleles) displayed genotype II alleles at these loci. Furthermore, the last five isolates also manifested chimeric genotype Ic/Ib or Ic/II alleles at the Cpgp40/15 locus, raising the possibility of sexual recombination within and between prototypical parasite genotypes. Lastly, children infected with isolates having genotype Ic alleles were significantly older than those infected with isolates displaying other genotype I alleles.

FIG. 1.

PCR products at the Cpgp40/15 locus from representative isolates. PCR was performed on DNA extracted from stool samples using the first set of primers. PCR products for samples 9 to 15 are shown. Note the differences in size of PCR products among isolates.

FIG. 2.

Polymorphisms in the deduced amino acid sequences of the Cpgp40/15 gene. The deduced amino acid sequences of 20 isolates were aligned using the Clustal W algorithm of the Align X program of Vector NTI. The five genotype I allelic subgroups—Ia, Ib, Ic, Id, and Ie—are indicated on the left-hand side. Numbers in parentheses indicate the amino acid position within that particular sequence. Hyphens indicate gaps, and conserved residues are shaded. Putative N-glycosylation sites are indicated in boldface and italic type, and predicted O-glycosylation sites are indicated in boldface and underlined type (note that although several S and T residues are conserved among the same allelic subgroup, not all of them are predicted by the NetOGlyc 2.0 program to be sites of O glycosylation). The putative cleavage sites of gp40 and gp15 (EE) are boxed. The predicted site for addition of a GPI anchor is underlined.

FIG. 2.

Polymorphisms in the deduced amino acid sequences of the Cpgp40/15 gene. The deduced amino acid sequences of 20 isolates were aligned using the Clustal W algorithm of the Align X program of Vector NTI. The five genotype I allelic subgroups—Ia, Ib, Ic, Id, and Ie—are indicated on the left-hand side. Numbers in parentheses indicate the amino acid position within that particular sequence. Hyphens indicate gaps, and conserved residues are shaded. Putative N-glycosylation sites are indicated in boldface and italic type, and predicted O-glycosylation sites are indicated in boldface and underlined type (note that although several S and T residues are conserved among the same allelic subgroup, not all of them are predicted by the NetOGlyc 2.0 program to be sites of O glycosylation). The putative cleavage sites of gp40 and gp15 (EE) are boxed. The predicted site for addition of a GPI anchor is underlined.

FIG. 3.

Comparison of SAAPs in the Cpgp40/15 allelic Ic/Ib and Ic/II subclasses with those of their hypothesized allele Ib and genotype II parents, respectively. The gp15 portion of the Cpgp40/15 deduced amino acid sequences of the Ic and Ib isolates from this study and of genotype II isolates Peru (accession number AAF78348), Brazil (accession number AAF78349), Iowa (accession number AAF78345), and KSU-1 (accession number AAF78351) and genotype Ic isolates 0676I (accession number AAF78357) and SFGH23 (accession number AAF78347), previously deposited in GenBank, were aligned using the Clustal W algorithm of the Align X program of Vector NTI. Hyphens indicate gaps. All SAAPs that distinguish the various Ic alleles are indicated in boldface type, allele Ic/II SAAPs that are shared with hypothesized parent allele II are shaded in yellow, and allele Ic amino acid residues that are not shared with their hypothesized parent are shaded in gray.