The 18S rRNA genes of Haemoproteus (Haemosporida, Apicomplexa) parasites from European songbirds with remarks on improved parasite diagnostics

Abstract

Background: The nuclear ribosomal RNA genes of Plasmodium parasites are assumed to evolve according to a birth-and-death model with new variants originating by duplication and others becoming deleted. For some Plasmodium species, it has been shown that distinct variants of the 18S rRNA genes are expressed differentially in vertebrate hosts and mosquito vectors. The central aim was to evaluate whether avian haemosporidian parasites of the genus Haemoproteus also have substantially distinct 18S variants, focusing on lineages belonging to the Haemoproteus majoris and Haemoproteus belopolskyi species groups.

Methods: The almost complete 18S rRNA genes of 19 Haemoproteus lineages of the subgenus Parahaemoproteus, which are common in passeriform birds from the Palaearctic, were sequenced. The PCR products of 20 blood and tissue samples containing 19 parasite lineages were subjected to molecular cloning, and ten clones in mean were sequenced each. The sequence features were analysed and phylogenetic trees were calculated, including sequence data published previously from eight additional Parahaemoproteus lineages. The geographic and host distribution of all 27 lineages was visualised as CytB haplotype networks and pie charts. Based on the 18S sequence data, species-specific oligonucleotide probes were designed to target the parasites in host tissue by in situ hybridization assays.

Results: Most Haemoproteus lineages had two or more variants of the 18S gene like many Plasmodium species, but the maximum distances between variants were generally lower. Moreover, unlike in most mammalian and avian Plasmodium species, the 18S sequences of all but one parasite lineage clustered into reciprocally monophyletic clades. Considerably distinct 18S clusters were only found in Haemoproteus tartakovskyi hSISKIN1 and Haemoproteus sp. hROFI1. The presence of chimeric 18S variants in some Haemoproteus lineages indicates that their ribosomal units rather evolve in a semi-concerted fashion than according to a strict model of birth-and-death evolution.

Conclusions: Parasites of the subgenus Parahaemoproteus contain distinct 18S variants, but the intraspecific variability is lower than in most mammalian and avian Plasmodium species. The new 18S data provides a basis for more thorough investigations on the development of Haemoproteus parasites in host tissue using in situ hybridization techniques targeting specific parasite lineages.

Keywords: Birth-and-death evolution; Parahaemoproteus; Ribosomal genes; Semi-concerted evolution.

Fig. 1

Median‑Joining DNA haplotype network of partial (474 bp) CytB sequences belonging to the Haemoproteus majoris group. The two figures show the distribution in A bird families and B geographic areas according to the United Nations geoscheme. Each circle represents a unique haplotype/lineage. The frequency is indicated for all haplotypes with more than one record and roughly corresponds to the size of circles. Bars on branches indicate the number of substitutions between two haplotypes. Small white circles represent median vectors, which are hypothetical (often ancestral or unsampled) sequences required to connect existing haplotypes with maximum parsimony. The lineages analysed in the present study are marked with asterisks

Fig. 2

Median‑Joining DNA haplotype network of partial (474 bp) CytB sequences belonging to the Haemoproteus belopolskyi group. The two figures show the distribution in A bird families and B geographic areas according to the United Nations geoscheme. Each circle represents a unique haplotype/lineage. The frequency is indicated for all haplotypes with more than one record and roughly corresponds to the size of circles. Bars on branches and numbers in squares indicate the number of substitutions between two haplotypes. Small white circles represent median vectors, which are hypothetical (often ancestral or unsampled) sequences required to connect existing haplotypes with maximum parsimony. The lineages analysed in the present study are marked with asterisks

Fig. 3

Pie charts showing the distribution of parasite lineages in geographic areas according to the United Nations geoscheme (left) and bird hosts (right)

Fig. 4

Bayesian inference tree of Haemoproteus 18S sequences. Posterior probabilities and maximum likelihood bootstrap values are indicated at most nodes. The scale bar indicates the expected mean number of substitutions per site according to the model of sequence evolution applied. The tree was midpoint-rooted, no outgroup was used

Fig. 5

Bayesian inference tree of Haemoproteus CytB sequences (885 bp). Posterior probabilities and maximum likelihood bootstrap values are indicated at most nodes. The scale bar indicates the expected mean number of substitutions per site according to the model of sequence evolution applied. The tree was rooted with a sequence of Plasmodium matutinum pLINN1