(C) PLOS One [1]. This unaltered content originally appeared in journals.plosone.org.
Licensed under Creative Commons Attribution (CC BY) license.
url:https://journals.plos.org/plosone/s/licenses-and-copyright

------------



Morphological and genomic characterisation of the Schistosoma hybrid infecting humans in Europe reveals admixture between Schistosoma haematobium and Schistosoma bovis

['Julien Kincaid-Smith', 'Ihpe', 'Univ Perpignan Via Domitia', 'Cnrs', 'Ifremer', 'Univ Montpellier', 'Perpignan', 'Cbgp', 'Ird', 'Cirad']

Date: 2022-01 

Schistosomes cause schistosomiasis, the world’s second most important parasitic disease after malaria in terms of public health and social-economic impacts. A peculiar feature of these dioecious parasites is their ability to produce viable and fertile hybrid offspring. Originally only present in the tropics, schistosomiasis is now also endemic in southern Europe. Based on the analysis of two genetic markers the European schistosomes had previously been identified as hybrids between the livestock- and the human-infective species Schistosoma bovis and Schistosoma haematobium, respectively. Here, using PacBio long-read sequencing technology we performed genome assembly improvement and annotation of S. bovis, one of the parental species for which no satisfactory genome assembly was available. We then describe the whole genome introgression levels of the hybrid schistosomes, their morphometric parameters (eggs and adult worms) and their compatibility with two European snail strains used as vectors (Bulinus truncatus and Planorbarius metidjensis). Schistosome-snail compatibility is a key parameter for the parasites life cycle progression, and thus the capability of the parasite to establish in a given area. Our results show that this Schistosoma hybrid is strongly introgressed genetically, composed of 77% S. haematobium and 23% S. bovis origin. This genomic admixture suggests an ancient hybridization event and subsequent backcrosses with the human-specific species, S. haematobium, before its introduction in Corsica. We also show that egg morphology (commonly used as a species diagnostic) does not allow for accurate hybrid identification while genetic tests do.

In 2013, schistosomiasis reached Southern Europe. Since then, endemic infections were recurrently identified clearly indicating that the parasite has settled and established locally. Using two molecular markers, we had previously demonstrated that the parasite was a hybrid between Schistosoma haematobium and Schistosoma bovis, two species known to infect humans and livestock, respectively. Nevertheless, this method has very low resolution and is not informative for determining the origins and the mechanisms of hybridization events, e.g. if the hybrid had been recently generated in Corsica or long before it’s introduction. The genome-wide sequencing approach, used in this study, allowed us to reveal genomic admixture suggesting that these hybrids are the result of ancient crossing events between S. haematobium and S. bovis, with subsequent backcrossing with S. haematobium. This introgression has prevented compatibility with P. metidjensis snail vectors, which are specific to S. bovis. Additionally, whether in Africa or in Europe, a clear discrepancy exists between the egg shape usually used for species identification and the genomic composition of schistosomes. Therefore, egg shape cannot be used as a good indicator for hybrid detection. Knowing the phenotypic traits and the genetic features of such hybrids may have important implications in terms of diagnostics and disease management either through vector control strategies or treatment of patients.

To this end, next generation whole-genome sequencing is now the tool of choice for a deeper insight into the genomic composition of natural hybrids. In particular, it may enable for a better understanding of these hybridization events, if they are frequent and active, rare and/or ancient together with the direction of the genetic introgression which may result in the inheritance of species specific phenotypic traits [ 11 – 13 ]. This genomic interrogation may also provide valuable insights into reproductive isolating barriers that are at play helping to maintain species integrities and prevent hybrid speciation [ 11 , 14 , 15 ]. Although recent studies on the absence of pre-zygotic isolation mechanisms suggest that hybridization between S. haematobium and S. bovis may be common [ 16 ], current genomic analyses of hybrids recovered from endemic areas indicate that introgression between S. haematobium and S. bovis is the result of ancient events [ 11 , 14 , 15 ]. Only a small proportion of the S. bovis genome appears to be introgressed into the genomic background of S. haematobium, with a potential adaptive significance related to host-pathogen interactions [ 11 , 15 ].

Global changes, including both anthropogenic and environmental modifications, may contribute to modifications in the geographical distribution of species and expand their potential ecological niches [ 4 , 5 ]. Distinct species may thus acquire a new capacity to interact, hybridize and subsequently introgress their genomes by backcrossing with parental species or other hybrids, a phenomenon called “hybrid swarm” [ 6 ]. Hybridization between individuals from two previously reproductively isolated species is generally expected to result in the production of offspring less fit than the parents, sometimes non-viable or sterile. In some cases however, hybridization may lead to viable progeny that can even have a greater fitness than parental species, a genetic effect known as hybrid vigour or heterosis that is generally observed during early generations [ 7 ]. These advantageous combinations of parental genes in offspring may enable progeny to adapt to new environments as potentially exemplified by the recent outbreak of schistosomiasis in Southern Europe (Corsica, France) [ 8 ].

Besides being of fundamental interest for the evolutionary biology of the parasite, this finding could also have immediate consequences for parasite control. One of the few phenotypic features that are of relevance for infection success and for which the genetic basis is known is resistance to Oxamniquine (OXA). In S. mansoni the mutations that confer resistance occur in the SmSULT-OR gene (Smp_089320), encoding a sulfotransferase that is required for drug activation, are p.E142del and p.C35R. [ 23 , 24 ]. In S. haematobium the drug is not efficient due to a F39 Sm > Y54 Sh substitution [ 24 ]. We reasoned that if the corresponding S. bovis allele would have introgressed into the hybrid it would follow the S. mansoni trait making the hybrid more sensitive to OXA. We have undertaken reciprocal homology searches with Smp_089320 against the S. haematobium and the new S. bovis genome to identify orthologues. Orthologues exist in both genomes and both possess the F->Y Sh54 and no mutations in C Sm35 or L Sm256 or deletion in E Sm142.

Interestingly, the mapping of the European hybrid reads against the S. haematobium and S. bovis reference genomes revealed admixture between the parental genomes with a proportion of 76.9% of sequences mapping to the S. haematobium, and 23.1% mapping to the S. bovis genomes for both male and female parasites ( Table 4 ). Alignment to the mitochondrial genomes of both parents also showed results concordant with the previous Sanger sequencing data for this marker, with 97.8% of reads being mapped to the S. bovis mitochondrial genome, and 2.2% to the S. haematobium mitochondrial genome [ 8 ]. To assess the divergence level between the nuclear genomes of the two “pure” species, we identified homologous regions between S. haematobium and S. bovis using CACTUS [ 22 ]. A total of 234.5 Mb sequences aligned between the parental species, representing 64% of the S. haematobium genome length. The mean similarity of these shared sequences was 95.9% compared to ~99% and ~90% for commonly used phylogenetic markers such as ITS and cox1, respectively. The mean similarity between S. haematobium mitochondria (GenBank accession number DQ157222.2) and S. bovis mitochondria (Contig 00439F of our assembly) was 82.1%.

Unfortunately, we were not able to produce enough high molecular weight genomic DNA to directly proceed to long-read sequencing of individual hybrid adult worms. To better characterize the level of hybridization we therefore used Illumina short read high-throughput sequencing and alignment to parental genomes. As the previous S. bovis assembly was highly fragmented (111 328 scaffolds, N50 7kb), and another S. bovis strain from Spain was used to produce the experimental hybrids examined in this study (as below) we resequenced the S. bovis (Villar de la Yegua, Salamanca, Spain) genome using 200 worms with PacBio long reads. This produced 4,102,584 filtered subreads (48,987,175,429 bases). The genome assembly led to a new genome version of 486 scaffolds (N50: 3.1Mb) from which 14,104 protein-coding genes were identified with an average length of 18,725 bp (5.3% of genome length) ( Table 2 ), consistent with the known characteristics of Schistosoma genomes [ 17 , 18 ].

The oesophagus started near to the oral sucker and bifurcated immediately after the acetabulum. The genital pore was situated at the posterior end of the ventral sucker (dorsally) and some eggs could be observed in the uterus ( Fig 3C ). A single ovary measuring 0.3 (±0.01) long and 0.09 (±0.01) wide was situated in the posterior third of the female’s body ( Fig 3A ). The vitelline glands, also called vitellaria were extensive, occupying roughly 50% of the posterior part of the worm and extending further posteriorly than the intestine. The females’ tegument was smooth and uniform without significant projections. The posterior extremity was tapered and rounded.

Four to five testes were observed per male, situated dorsally posterior to the ventral sucker and were round to ovoid (4.2±0.5). Posterior to the acetabulum, in the dorsal region, tubercles with a round extremity began to appear at the level of the gynaecophorial canal and occurred to all the posterior region of the body ( Fig 2C ). The presence of tegument projections on the tubercles where identified with several apical spines which decreased in distribution and size towards the back and sides of the male’s bodies ( Fig 2C ). The females were elongated and filiform measuring 9.6 (±1.2) mm in length and 0.2 (±0.04) mm in width, with the posterior half of body expanded ( Fig 3A ). The anterior regions of the females were smaller compared to males, had a small oral sucker (length: 0.05±0.004 mm; width: 0.05±0.002 mm) and acetabulum (length: 0.06±0.001 mm; width: 0.06±0.001 mm) ( Fig 3B ).

The adult S. haematobium-bovis hybrid males had an elongated body measuring, on average, 5.3 (±1.0) mm in length and 0.3 (±0.07) mm in width at largest part of the body (middle). The male body was dorso-ventrally flattened with folds in the ventral axis forming the gynaecophoral canal ( Fig 2A ). The anterior region was narrower than the rest of the body and presented a sub-terminal mouth formed into the oral sucker (length: 0.2±0.01 mm; width: 0.2±0.006 mm) and a robust ventral sucker or acetabulum (length: 0.3±0.008 mm; width: 0.3±0.01 mm) near the top of the gynaecophoral canal ( Fig 2B ). The oesophagus began at the oral sucker and was extended posteriorly to the acetabulum where it bifurcates into the gut caeca until it reunites at the posterior end of the body.

A total of 44 eggs collected from hamsters infected with the European S. haematobium-bovis hybrid were examined for morphological characterization. The length and width were measured for all 44 eggs, however, the spine length was only measured for a subset of 36 eggs, due to it not being distinctive enough to allow for accurate measurements for eight of the eggs. The eggs of the European S. haematobium-bovis hybrid showed a high variability and ranged between 73.9–170.9 μm in length (mean: 126.4 ± 22.9 standard deviation), 40.9–92.5 μm in width (mean: 60.8 ± 13.0 standard deviation) and 3.95–13.6 μm for the spine length (mean: 8.2 ± 2.1 standard deviation).

Discussion

The emergence of infectious diseases are currently among the greatest concerns of our changing world and has strong outreach effects for society. Besides the important impacts that global changes may have on the spread and transmission of tropical infectious diseases in higher latitudes, other phenomenon may combine and act as driving forces promoting the emergence of novel disease in unsuspected areas. The importance and the frequency of hybridization in infectious agents are certainly underestimated, and very little attention has been given so far to the role of genetic introgression on infectious disease emergence, spread and control [25]. In the genus Schistosoma, several reports have revealed that inter-species hybrids are frequent and are a real concern for human health [9,10]. In particular hybridization between S. haematobium and S. bovis have now been identified with molecular tools in Senegal, Niger, Benin, Mali, Cote D’Ivoire and also Malawi [26–30]. To date genomic evidence indicates that although only a small proportion of S. bovis seems to have introgressed in the genomic background of S. haematobium (i.e. 3–8%), introgression from S. bovis is widespread across S. haematobium populations in endemic areas. The signatures of introgression observed indicate ancient and unidirectional events with a potential adaptive significance related to host-pathogen interactions [11,14,15]. Nevertheless, it is the first time that a hybrid schistosome has been involved in a large-scale outbreak in Europe [8,31,32]. Although usually restricted to tropical areas, schistosomiasis transmission is now persisting in Corsica and the hybrid status of the parasite might have increased its invasive and adaptive capacities. Indeed, our study revealed that the European hybrids established in Corsica are highly introgressed, 77% S. haematobium origin and 23% S. bovis origin. Together with the maternal inheritance of the mitochondria this suggested that these hybrids were generated by an initial cross between a male S. haematobium and a female S. bovis with successive backcrosses with S. haematobium. This seems to confirm an ancient and mostly unidirectional introgression event(s) that could potentially be advantageous for the parasites. The hybrid status of the parasite may thus have important implication for disease control in term of host spectrum, diagnostics, and treatment in endemic areas but also Europe.

Implications for host spectrum and parasite distribution S. haematobium and S. bovis have different intermediate host specificities. S haematobium only infects snails within the genus Bulinus while S. bovis can also infect Planorbarius snails (widely present in the Iberian Peninsula), together with Bulinus species. The potential distribution range of the disease may thus be enhanced if the hybrid is able to infect the intermediate hosts of both parental species. Interestingly, the natural European hybrid schistosomes, that we recovered in Corsica, were not able to infect our laboratory strain of P. metidjensis, but displayed high levels of compatibility with Corsican B. truncatus (24% infection prevalence) which is consistent with previous schistosome-snail compatibility assessments for the hybrid parasite recovered directly from an infected patient in 2014 [8]. This is consistent with the predominance of S. haematobium within the genome of this hybrid, and may open the door to vector based control, e.g. by control strategies targeting these snails. However, although our findings indicate that P. metidjensis is not a host for this European hybrid we cannot exclude the possibility that other strains of P. metidjensis from Europe or Africa could be compatible and act as a host. Moreover, as the tested miracidia were collected after passage through a laboratory host, potentially inducing a population bottleneck, further testing on various hybrid field isolates is warranted. One other fundamental concern is the capacity of such introgressed schistosomes to infect livestock or other animal reservoir hosts. The zoonotic potential of the hybrids would strongly impact the parasite transmission in the field, in and out of endemic areas, and may hamper our capacity to maintain adequate control strategies as schistosomiasis treatment focuses almost exclusively on humans. Recent studies are now showing the presence of not only S. bovis, S. haematobium, and S. mansoni in rodents (hosts in which hybridization may occur), but also the occurrence of S. haematobium x bovis hybrids in such hosts in Senegal [33,34] and Benin [35] although the importance of rodents in transmission dynamics needs further exploration. Moreover, beside widespread investigation of animal reservoir in Senegal [26] to date only one study in Benin suggests that other animals such as cattle may be natural hosts for such hybrid parasites [27]. The situation in Corsica and the role of animal reservoir needs to be precisely investigated as despite ongoing transmission and developing endemicity on the island [36], no infection has been detected in livestock in the region, and the only infected animals found were two rats that do not seem to play a significant role in the transmission for this particular foci [37]. However, we cannot rule out the influence of an undetected animal reservoir such as Ovis aries musimon, a wild sheep native to Corsica, that have never been tested for infections [36].

Implications for diagnostics The hybrid status of the parasite may impair parasitological, serological and molecular diagnostics used to diagnose infections. In endemic countries, parasitological diagnosis (egg detection) is the gold standard, whereas serological tests are commonly used for imported cases of schistosomiasis in non-endemic, developed countries. In humans, schistosome eggs that are partly retained in the tissues are the cause for the disease and host-induced pathology, but are also classical tools for diagnosis and species identification. At first sight, egg morphology and their localization in the urine of infected patients in Corsica strongly suggested an S. haematobium infection [8]. Indeed S. haematobium eggs that are usually voided by the urine have a typical round to oval shape (elliptical or elongated) with a terminal spine. According to previous studies, S. haematobium eggs measure between 100–156 μm long and 40–50 μm wide with usual length between 115–135 μm long [38–40]. A previous analysis of the European hybrid eggs revealed smaller eggs (n = 15) with a mean length of 106.5 μm, a width of 42.8 μm with a spine length of 10.4 μm [32]. According to our results the eggs generally show an ovoid shape measuring 126.4 x 60.8 μm, more similar to S. haematobium eggs. This is also consistent with the introgression levels that show a predominance of S. haematobium-type genetic material (Table 4). However, sometimes eggs appeared intermediate with spindle or diamond shapes, which are characteristic of S. bovis eggs (usually bigger and measuring between 170–223.9 μm long and 55–66.0 μm wide) [38,41,42] (Fig 1). In addition, S. bovis eggs are released in the feces of infected animals, due to S. bovis locating around the mesenteric vessels. Thus, we could expect that hybrid parasite eggs may also be released in part in the feces of human hosts. This could explain, together with the low parasite intensity, why only 30% of patients infected in Corsica had eggs that were able to be detected in their urine [43]. The route of excretion associated with egg shape is the current standard for diagnostic and species determination, however our results confirm earlier publications showing that it is impossible to detect hybridization in schistosome species using egg morphology alone [44]. Although adult worm morphology has a limited interest for diagnostics in humans, principally because worms are not accessible, they may be useful for a taxonomic purpose. Males and females of the European hybrids were generally smaller (in length and width) when compared to both parental species [45–49], but most interestingly, the presence of spines on the tubercles of the males, a trait not found in S. bovis male, [50–52], was concordant with the predominance of S. haematobium in the genetic make up of these hybrids. Concerning serological diagnosis, the majority of commercial tests, ELISA or IHA (indirect hemagglutination) use S. mansoni antigens. A discrepancy between those antigens and the infecting species may induce false negative results [53]. The efficiency of these commercial diagnostic kits thus needs to be reevaluated in a context of different species and hybrid forms. Finally, molecular diagnostic for urogenital schistosomiasis using PCR has already been used in urine or serum, targeting a highly repeated sequence (DraI), which is restricted to the S. haematobium group of schistosomes (including both S. haematobium and S. bovis) [54,55]. We expect that this test would be efficient to detect such infection but not to identify the hybrid status of the parasite.

Implications for treatment Praziquantel (PZQ) is currently the main drug used to treat schistosomiasis and the application of mass chemotherapy programs is the prevailing strategy for schistosomiasis control [56]. PZQ is also efficient for treating bovine schistosomiasis, but the dose needed is quite high (60 mg/kg for 95% deworming efficacy in goats [57,58]) and cannot be considered in endemic areas where treatment capacities are primarily focused on human schistosomiasis. Moreover, it has been shown in humans that a dose of 40 mg/kg of PZQ is only 63.5% efficient for mixed infections, compared to 76.7% and 77.1%, for mono infections of S. mansoni and S. haematobium, respectively [59]. To date, neither experimental nor field trials have tested the sensitivity of hybrid parasites to PZQ. Thus, there is no current evidence that there is any difference in drug response in natural infections and changes associated to PZQ response in hybrids is still theoretical. However, a lower sensitivity to PZQ of S. bovis x S. haematobium hybrid schistosomes compared to pure S. haematobium parasites, although not tested, has been proposed to be at the origin of the spread of the hybrid forms in Senegal [60] and as discussed earlier hybridization may also affect Oxamniquine efficiency [61]. Since the genetic basis of Oxamniquine residence is known, our data suggests that OXA is not a treatment option.

[END]

[1] Url: https://journals.plos.org/plosntds/article?id=10.1371/journal.pntd.0010062

(C) Plos One. "Accelerating the publication of peer-reviewed science."
Licensed under Creative Commons Attribution (CC BY 4.0)
URL: https://creativecommons.org/licenses/by/4.0/


via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/