Multilocus sequence typing of Giardia duodenalis genotypes circulating in humans in a major metropolitan area

Abstract

Giardia duodenalis is an intestinal protozoan parasite of humans and animal hosts and comprises eight microscopically indistinguishable molecularly-diverse lineages designated as assemblages A-H. Assemblages A and B are the primary sources of infections in humans and a wide range of mammals. Here, we identified assemblages, and inter-/intra-assemblage genetic diversity of human G. duodenalis isolates based on the multilocus sequence typing of the triosephosphate isomerase (tpi), β -giardin (bg), and glutamate dehydrogenase (gdh) loci. Multilocus sequence analysis of 62 microscopically-positive G. duodenalis fecal samples identified 26 (41.9%), 27 (43.5%), and nine (14.5%) isolates belonging to assemblages A, B, and discordant assemblages, respectively. The tpi locus assemblage-specific primers identified dual infections with A and B assemblages (45.2%). The sequence analysis of multiple alignments and phylogenetic analysis showed low genetic polymorphism in assemblage A isolates, classified as sub-assemblage AII at three loci, subtype A2 at tpi and gdh loci, and subtype A2 or A3 at bg locus. High genetic variations were found in assemblage B isolates with 14, 15, and 23 nucleotide patterns at tpi, bg, and gdh loci, respectively. Further concatenated sequence analysis revealed four multilocus genotypes (MLG) in 24 assemblages A isolates, two previously-identified (AII-1 and AII-5), with one novel multilocus genotype. However, the high genetic variations observed in assemblage B isolates among and within the three genetic loci prevented the definitive designation of specific MLGs for these isolates. Multilocus sequence typing may provide new insight into the genetic diversity of G. duodenalis isolates in Tehran, suggesting that humans are likely a potential source of G. duodenalis infection. Further host-specific experimental transmission studies are warranted to elucidate the modes of transmission within multiple host populations.

Keywords: Giardia duodenalis; Iran; Tehran; glutamate dehydrogenase (gdh); multilocus sequence typing; triosephosphate isomerase (tpi); β-giardin (bg).

Figure 1

The phylogenetic analysis of the tpi gene nucleotide sequences of Giardia duodenalis using the Maximum Likelihood (ML) method and Tamura 3-parameter model (38) (T92 + G + I). The analysis involved 23 nucleotide sequences: Nine representative nucleotide sequences of tpi retrieved from this study (LC183913–LC183966) compared with 13 reference sequences of known assemblages from Genbank (indicated in bold) with Giardia microti as an outgroup. The percentage of trees in which the associated taxa clustered together (achieved from 1000 replicates) is shown next to the branches, only bootstrap values >50% are demonstrated. A discrete Gamma distribution was used to model evolutionary rate differences among sites [5 categories (+G, parameter = 9.6789)]. The rate variation model allowed for some sites to be evolutionarily invariable [(+I), 27.96% sites]. The scale bar represents substitutions per nucleotide. The final dataset contained 456 positions. Evolutionary analyses were conducted in MEGA X (39).

Figure 2

The phylogenetic analysis of the bg gene nucleotide sequences of Giardia duodenalis using the Maximum Likelihood method (ML) and Tamura 3-parameter model (38) (T92 + G + I). The analysis involved 30 nucleotide sequences: 14 representative nucleotide sequences of bg retrieved from this study (LC183967–LC184028) compared with 16 reference sequences of known assemblages from Genbank which are indicated in bold. The percentage of trees in which the associated taxa clustered together (achieved from 1,000 replicates) is shown next to the branches, only bootstraps values >50% are demonstrated. A discrete Gamma distribution was used to model evolutionary rate differences among sites [5 categories (+G, parameter = 0.1294)]. The rate variation model allowed for some sites to be evolutionarily invariable [(+I), 42.59% sites]. The scale bar represents substitutions per nucleotide. The final dataset included 506 positions. Evolutionary analyses were conducted in MEGA X (39).

Figure 3

The phylogenetic analysis of the gdh gene nucleotide sequences of Giardia duodenalis using the Maximum Likelihood method and Tamura 3-parameter model (38) (T92 + G + I model). This analysis involved 29 nucleotide sequences: Nine representative nucleotide sequences of gdh retrieved from this study (LC184423–LC184474) compared with 19 reference sequences of known assemblages from Genbank (indicated in bold) with Giardia ardeae as an outgroup. The percentage of trees in which the associated taxa clustered together (achieved from 1,000 replicates) is shown next to the branches, only bootstraps values >50% are demonstrated. A discrete Gamma distribution was used to model evolutionary rate differences among sites [5 categories (+G, parameter = 0.4823)]. The rate variation model allowed for some sites to be evolutionarily invariable [(+I), 36.49% sites]. The scale bar represents substitutions per nucleotide. A total of 433 positions were evaluated in the final dataset. Evolutionary analyses were conducted in MEGA X (39).

Figure 4

The phylogenetic analysis of the concatenated tpi, bg, and gdh nucleotide sequences of Giardia duodenalis using the maximum likelihood method (ML) and Tamura 3-parameter model (38) (T92 + G + I). The analysis involved 40 nucleotide sequences: 29 concatenated (tpi + bg + gdh) nucleotide sequences retrieved from this study compared with 11 known multilocus genotype reference sequences reported in previous studies (17, 40, 41) are indicated in bold. The red-filled square represents the new MLG of assemblage A reported in this study and the blue-filled circles indicate the one nucleotide substitution in A3 compared with the AII-5 MLG of assemblage A based on the modified numerical MLG reviewed in Cai et al. (4). The final dataset contained 1,395 positions. The percentage of trees in which the associated taxa clustered together (achieved from 1,000 replicates) is shown next to the branches, only bootstraps values >50% are demonstrated. A discrete Gamma distribution was used to model evolutionary rate differences among sites [5 categories (+G, parameter = 0.3229)]. The rate variation model allowed for some sites to be evolutionarily invariable [(+I), 37.78% sites]. The scale bar represents substitutions per nucleotide. Evolutionary analyses were conducted in MEGA X (39).