Multilocus Sequence Typing and Population Genetic Analysis of Enterocytozoon bieneusi: Host Specificity and Its Impacts on Public Health

Abstract

Microsporidia comprise a large class of unicellular eukaryotic pathogens that are medically and agriculturally important, but poorly understood. There have been nearly 1,500 microsporidian species described thus far, which are variable in biology, genetics, genomics, and host specificity. Among those, Enterocytozoon bieneusi is the well-known species responsible for the most recorded cases of human microsporidian affections. The pathogen can colonize a broad range of mammals and birds and most of the animals surveyed share some genotypes with humans, posing a threat to public health. Based on DNA sequence analysis of the ribosomal internal transcribed spacer (ITS) and phylogenetic analysis, several hundreds of E. bieneusi genotypes have been defined and clustered into different genetic groups with varied levels of host specificity. However, single locus-based typing using ITS might have insufficient resolution to discriminate among E. bieneusi isolates with complex genetic or hereditary characteristics and to assess the elusive reproduction or transmission modes of the organism, highlighting the need for exploration and application of multilocus sequence typing (MLST) and population genetic tools. The present review begins with a primer on microsporidia and major microsporidian species, briefly introduces the recent advances on E. bieneusi ITS genotyping and phylogeny, summarizes recent MLST and population genetic data, analyzes the inter- and intragroup host specificity at the MLST level, and interprets the public health implications of host specificity in zoonotic or cross-species transmission of this ubiquitous fungus.

Keywords: Enterocytozoon bieneusi; host specificity; microsporidia; multilocus sequence typing; population genetics; public health implication.

FIGURE 1

Phylogenetic analysis of individual and concatenated gene datasets of Enterocytozoon bieneusi. Phylogenies at loci ITS (panel A), MS1 (panel B), MS3 (panel C), MS4 (panel D), MS7 (panel E), and concatenated (panel F) were inferred by the maximum-parsimony method implemented in Mega 7 (http://www.megasoftware.net/). The reliability of cluster formation was evaluated by the bootstrap analysis with 1,000 replicates and the values generated greater than 50% were shown beside nodes. The colors (red, dark green, light green, black, and gray) of the circles specified before specimen codes represent the isolates are from ITS Groups 1 (n = 20), 10 (n = 8), 2 (n = 9), and 6 (n = 2) and an outlier group (n = 1), respectively. The first initials of isolate names, C, I, N, and P, indicate the isolates were sampled in China, India, Nigeria, and Peru, respectively. The letters B, D, H, K, S, and T followed demonstrate the isolates were sampled from bear, deer, human/horse (only human from India, Nigeria, and Peru was sampled, and horse from China likewise), kangaroo, swine, and takin, respectively. ITS genotypes are specified after specimen codes.

FIGURE 2

Multilocus phylogeny and haplotype network of Enterocytozoon bieneusi ITS Group 1 isolates sampled in various hosts and locations. All the E. bieneusi isolates (n = 246, Supplementary Table S2) that were previously genotyped as Group 1 members were included in the analyses except for five isolates (2 for genotype IH and 3 for genotypes D, horse1, and Nig2) with multilocus sequences significantly divergent from the other Group 1 members. The naming of subpopulations is based chiefly on that described previously (Wan et al., 2016). SP7 is highlighted with a dashed border as its host specificity and zoonotic potential are poorly understood. Panel A: Maximum-parsimony phylogeny of 138 unique multilocus genotypes (MLGs) generated considering both single-nucleotide polymorphisms (SNPs) and insertions and deletions from the total isolates using Mega 7 (http://www.megasoftware.net/). The first initials of isolate names, B, C, I, K, N, and P, indicate the isolates were sampled in Brazil, China, India, Kenya, Nigeria, and Peru, respectively. The letters B, C, D, F, H, M, Q, R, S, and T followed indicate the isolates were sampled from bear, cat, deer, fox, human, monkey, squirrel, raccoon dog, swine, and takin, respectively. ITS genotypes are labeled at the ends. Bootstrap values (1,000 replicates) above 50% are shown at the nodes. Panel B: Genetic network analysis of all the 246 isolates using the median-joining method implemented in Network 4.6.1.1 (http://www.fluxus-engineering.com/sharenet_rn.htm). The colors in circles correspond to different hosts sampled. The size of the circles is proportional to the frequency of each of the 98 MLGs obtained with consideration of only SNPs. ITS genotypes are labeled beside the circles. Genotype type IV is abbreviated as IV for better presentation and non-human primates as NHPs likewise. The branches connecting MLGs have a length proportional to the number of SNPs, while the red branches having pairwise differences above 12 SNPs are shortened for better presentation.

FIGURE 3

Multilocus phylogeny and haplotype network of Enterocytozoon bieneusi isolates from different hosts and geographic origins that belong to Group 1 and several other ITS groups. All the E. bieneusi isolates (n = 20) that belong to the ITS groups other than Group 1 were included in the analyses and some of the Group 1 isolates (n = 27) with genotypes being D, EbpC, type IV, horse1, IH, and Nig2 were selected for comparative analysis (Supplementary Table S2). Panel A: Maximum-parsimony phylogeny of 37 unique multilocus genotypes (MLGs) generated considering both single-nucleotide polymorphisms (SNPs) and insertions and deletions from the total isolates using Mega 7 (http://www.megasoftware.net/). The first initials of isolate names, C, I, and N, indicate the isolates were sampled in China, India, and Nigeria, respectively. The letters B, D, H, K, M, Q, S, and T followed indicate the isolates were sampled from bear, deer, human/horse (only human from India and Nigeria was sampled, and horse from China likewise), kangaroo, monkey, squirrel, swine, and takin, respectively. ITS genotypes are labeled at the ends. ITS Groups 1, 2, 6, and 10 and an outlier group that the isolates belong to were indicated by red, light green, blue, dark green, and gray triangles, respectively. Bootstrap values above 50% from 1,000 pseudoreplicates are displayed. Panel B: Median-joining network analysis of all the total 47 isolates using Network 4.6.1.1 (http://www.fluxus-engineering.com/sharenet_rn.htm). The colors in circles represent various host sources. The size of the circles is proportional to the frequency of each of the 26 MLGs obtained based on only SNPs. ITS genotypes are labeled beside the circles. Genotype type IV is abbreviated as IV for better presentation and non-human primates as NHPs likewise. The black branches connecting MLGs have a length proportional to the SNPs, while the red branches having pairwise differences greater than 12 SNPs are shortened for better presentation.