Cestoda is a class of parasitic worms in the flatworm phylum Platyhelminthes that parasitize the intestines of all major groups of vertebrates, including fish (Kuchta et al., 2020), reptiles (Yudhana et al., 2019), birds (Okulewicz, 2014), and mammals (Carlson et al., 2020), and cause severe, mild or no symptoms of infection. Their larvae usually achieve development in one or two intermediate hosts (Kuchta et al., 2020) and can cause severe symptoms and even death in animals and humans; species of Taenia and Echinococcus cause the highest health and economic impact.

To date, there are approximately 4,800 described species in the class Cestoda, belonging to 833 genera and 19 orders, of which the Cyclophyllidea is the most species-rich order in the class, with more than 3,100 valid species, distributed among 16 families (Sharma et al., 2016; Caira and Jensen, 2017; Kuchta et al., 2020). Each free-living metazoan species is considered to harbor at least one protozoan or metazoan parasite species (Poulin and Morand, 2004) with which they interact and usually co-evolve (Ebert and Fields, 2020). Parasites are probably one of the largest groups of undescribed organisms, as most are cryptic whether in their parasitic or free-living forms (Dobson et al., 2008; Larsen et al., 2017; Okamura et al., 2018; Carlson et al., 2020). It is estimated that there are a total of about 100,000–350,000 helminth species in vertebrates around the world, of which 85–95% are undiscovered or recorded to science (Carlson et al., 2020).

Many of the world’s biodiversity hotspots are located in large mountain systems, and their role in the evolutionary diversification of organisms is manifold (Hoorn et al., 2018; Rahbek et al., 2019a,b). The Qinghai-Tibet Plateau (QTP) and its surrounding mountain systems of the Eurasian continent, have yielded arguably the biggest and probably the most biologically diverse area of montane species (Päckert et al., 2020). In the past decade, several new Taeniidae species have been described from wild rodents on the QTP (Xiao et al., 2005; Wu et al., 2021). In terms of species diversity and sheer population sizes, rodents are perhaps the most important intermediate and definitive hosts of tapeworms, and are the most widely distributed and diverse group of mammals; about 43% of all species (Singla et al., 2008; Wu et al., 2018). Changes in climate and vegetation caused by the uplift of the QTP may have promoted local adaptations such as the evolution of cold- and hypoxic-tolerant rodent species, which in turn may have led to the co-evolution and radiation of their parasites (Wang et al., 2018; Wu et al., 2021). We hypothesize that there may be many undiscovered tapeworm species parasitizing the rodents in the QTP. Although the number of tapeworm species has been underestimated in general, some practices may lead to the erroneous proposal of new species (Carlson et al., 2020; Kuchta et al., 2020).

Morphological distinctions have been used for the description of many tapeworms, however, the homoplasy in morphology poses quite a challenge to infer their evolutionary lineage (Scholz et al., 2021), where morphological features may be significantly affected by different intermediate host sources (Lymbery, 1998). Many undescribed species may be genetically different but morphologically indistinguishable. The inability to distinguish such cryptic species affects accurate assessment of host range and estimates of total diversity (Carlson et al., 2020). The 28S and 18S nuclear ribosomal RNA genes (rDNA) are relatively conserved within species and are often used to differentiate different species (Wickström et al., 2005; Nakao et al., 2013; Scholz et al., 2021). The mitochondrial (mt) genome is largely haploid and uniparentally inherited, so their effective population size is four times smaller than that of the nuclear genome (Toews and Brelsford, 2012). Since the process of lineage sorting is inversely proportional to the effective population size, this means mitochondrial (mt) genomes will complete this process faster than nuclear genomes (Toews and Brelsford, 2012). Thus, the genetic nature of a mt genome makes it likely more sensitive than any single nuclear marker to distinguish closely related species and study their phylogenetic relationships (Lee et al., 2007).

Therefore, in this study, we sampled parasites in rodents from the QTP and its surrounding areas. In total, seven Cyclophyllidea specimens were collected and characterized using molecular tools. By comparing new (mitochondrial, nuclear 28SrDNA, 18S rDNA) with published homologs (GenBank) we found four isolates to be markedly different in DNA sequence thus likely representing ‘‘putative new species.’’ The complete mt genomes of these unknown taxa were sequenced and annotated, and their taxonomic status was analyzed and verified through phylogenetic reconstruction of five datasets containing a total of 320 species from 10 families. By combining evolutionary divergence time analyses of mt genes of classified cyclophyllideans in NCBI Taxonomy Database¹ and the paleogeography of QTP, we thus speculate an “out of QTP” theory for cyclophyllideans.

Given the rich diversity and the large rodent population in western China and the few published reports of tapeworms in rodents, except for Taeniidae (Xiao et al., 2005; Zhao et al., 2014; Wang et al., 2018; Zhang et al., 2018; Wu et al., 2021), the present knowledge of tapeworm biodiversity in rodents in western China suggests far greater biodiversity yet to be uncovered.

Here, the mt genes and 18S or 28S rDNA fragments of four (two larvae and two adults) unidentified Cyclophyllidea species differed from their related species (Supplementary Table 10). However, due to the specimen distortion and insufficient specimen encountered in this study, it is not clear whether they have been described morphologically. All four species showed apparent discordance percentage identity with the related species in their mtDNA and 18S or 28S rDNA, which is the common DMD pattern across the animal kingdom, and demonstrated in Echinococcus granulosus sensu stricto (Kinkar et al., 2017), possibly due to the parthenogenetic inheritance of mitochondria, gene flow and recombination in the nuclear genome (Toews and Brelsford, 2012).

The larvae Hydatigera sp. XHPW10 was found to live in the livers of both zokors from northeast of QTP and plateau pika from southwest of QTP (Supplementary Table 1). This adds to the recently described cysticerci species in plateau pika from Qinghai province (Wu et al., 2021). Here the T. caixuepengi larva was also found to be parasitic in plateau pikas from Xietongmen, Saga and Sa’gya county of Tibet and Qilian county of Qinghai (Supplementary Table 1), and so far, not found in other sympatric rodent species. Due to the absence of comparative studies, it is currently not clear if this species has preference only for plateau pikas as its intermediate host. The larvae of T. crassiceps and V. mustelae were, respectively, found in Jeminay and Xinyuan county of Xinjiang autonomous region (Supplementary Table 1), which were also distributed on the northeast QTP (Li et al., 2013; Zhao et al., 2014). Wide geographic distributions of identical species suggest that their endemic geographic range should be far beyond the available survey data. Except for Mesocestoides sp. RKZ08 identified in this study, the Me. litteratus is another species of the genus Mesocestoides reported in Qinghai and Heilongjiang provinces of China (Wang et al., 2006; Li et al., 2013). However, there is no previous record of Paranoplocephala spp. and Mosgovoyia spp. available in China except for Paranoplocephala sp. RKZ13 and Mosgovoyia sp. SQ20 found in this study.

Complete mt genomes of the four “putative new species” were sequenced and annotated, and the sequences were clearly different from all available mt genomes sequences; however, they were similar in length, gene order and composition as other cyclophyllideans with respect to rRNA, tRNA, and protein-encoding genes (Le et al., 2000; Jeon et al., 2005, 2007; Wu et al., 2021). Different arrangement of genes occurred in tRNA genes between nad5 and nad6 genes, but was consistent with their respective most relative species, matching the expectation that the arrangement of mt genes partly determines the genetic relationship of parasites (Gazi et al., 2016). Moreover, repeat copies of tRNA gene between the nad5 and nad6 genes in the mt genome of E. granulosus sequenced by third-generation sequencing have been reported (Kinkar et al., 2019). We also observed a repeat sequence of trnL and trnR genes between nad5 and nad6 genes of the Paranoplocephala sp. RKZ13 mt genome (Figure 3 and Table 1), which indicates that there may be hidden tRNA gene repeats in the mt genome of tapeworms that are hard to identify due to errors in PCR amplification and Sanger sequencing, techniques which often fail to recover repeat regions (Kinkar et al., 2019).

Genetic drift and adaptive differentiation between allopatric populations is responsible for most speciation amongst plants and animals (Turelli et al., 2001). For parasites, however, host association is a key driver in their evolution. Host switching among sympatric populations may lead to ecological isolation, so sympatric speciation of parasites is common (de Meeûs et al., 1998; Paul, 2002; Huyse et al., 2005; Wu et al., 2021). These two models of evolution are not in conflict: the adaptation of parasite to specific host is like the adaptation of animal to specific living environment; so, the evolution of the host, especially its immune system, may be viewed as equivalent to the change of the living environment for the parasite. The environmental and climatic changes caused by the uplift of the QTP are a major driving force for the evolution of associated biotas (Favre et al., 2015). Paleobotanical data suggest that the southeastern margin of the QTP was dominated by a warm and humid subtropical or tropical climate during the Miocene due to the influence of South Asian and East Asian monsoons (Sun and Wang, 2005; Jacques et al., 2011). Since the mid-Miocene, the significant rise of the Himalayas and the Tianshan Mountains, together with worldwide cooling, incurred dramatic changes in air circulation, leading to gradual aridification of the QTP and Central Asia (Miao et al., 2012; Favre et al., 2015). Finally, in the Late Miocene and Early Pliocene, QTP uplift resulted in the accumulation of global ice and the eventual disappearance of the Tethys Sea, which also contributed to the drying of Central Asia (Lu and Guo, 2013).

These timelines of climatic and environmental changes caused by the uplift of the QTP are highly consistent with the timelines of the differentiation rate of Cyclophyllidea species analyzed in this study (Figures 4, 5) and that of plateau pika analyzed in Wang et al. (2020). We speculated in this study that the differentiation of cyclophyllideans may have been driven by host evolution caused by the uplift of the QTP. During the tropical period of QTP, the optimum living environment created the biological diversity (Cai et al., 2020; Päckert et al., 2020), and the species of Cyclophyllidea gradually differentiated. With the rapid uplift of the QTP, the environment changed into a dry and cold climate (Lu and Guo, 2013), and species differentiation of Cyclophyllidea was accelerated by the rapid adaptive evolution of their hosts and geographical isolation caused by the radiation of hosts to the Palaearctic (Favre et al., 2015; Xing and Ree, 2017; Päckert et al., 2020). Finally, in the last 2 million years, Cyclophyllidea differentiation demonstrated an accelerated diversification based on cox1 + nad1 divergence tree (Figure 5), which may be related to the evolution and broad distribution of mammals in Eurasia and the associated population expansion and migration of hominids from Africa to Asia (Dennell, 2004; Rohland et al., 2005; Brugal and Croitor, 2007; Turner et al., 2008; Klein, 2009; Terefe et al., 2014; Wang et al., 2016). The Host-parasite Database of Natural History Museum (HPDNHM) found that most tapeworm were prevalent mainly in the Palaearctic,⁵ which is also consistent with the viewpoint that the order Cyclophyllidea originated from the QTP. The Nearctic is another major endemic area where Cyclophyllidea species may have spread over land Bridges across the Bering Strait. However, some species parasitizing birds and other economic and companion animals tend to show a global epidemic, which may be due to the long-distance migration of birds and the spread of human trade and activities.

Phylogenetic reconstruction reveals that some classifications of Cyclophyllidea species may need to be redefined. Closely related species of tapeworm parasites often have similar host specificities and life history (Scholz et al., 2021), a pattern common in the HPDNHM and in our evolutionary analyses, and may provide a basis for revising the classification of Cyclophyllidea species; for example, the family Hymenolepididae and Anoplocephalidae can be divided into multiple families (Figures 1A–C), and Thysanotaenia congolensis should be reclassified into family Davaineidae (Figures 1A,C). However, considering that there is still a large number of undiscovered species, which may provide better support for classification, the current classification status is likely to remain for some time.

In conclusion, this study expands the biodiversity of Cyclophyllidea in rodents in QTP and its surrounding mountain systems, and suggests an “out of QTP” hypothesis for the Cyclophyllidea, wherein species differentiation was driven by the uplift of the QTP. Although beyond the scope of this study to consider the evolutionary relationships and history of the whole cyclophyllideans, the species analyzed represent 10 of all 16 families, making this the most extensive study of the evolution of Cyclophyllidea order to date. Verifying the taxonomic revision and the “out of QTP” hypothesis requires more sampling and investigation, including data on a wider geographic and host range, and molecular studies uncovering patterns of host-parasite co-evolution.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/genbank/, MZ476188–MZ476193; https://www.ncbi. nlm.nih.gov/genbank/, MW808979–MW808982; https://www. ncbi.nlm.nih.gov/genbank/, OM140661–OM140665.

The animal study was reviewed and approved by the Animal Ethics Procedures and Guidelines of the People’s Republic of China. Animal Ethics Committee of Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences (No. LVRIAEC2012-007).

Y-DW, LL, H-BY, and W-ZJ conceived and designed the experiments. Y-DW, G-DD, L-SZ, A-MG, Y-TW, and YF conducted the sample collection. Y-DW, G-DD, X-WN, and NS performed the experiments. Y-DW and G-DD performed the data analyses. Y-DW prepared the figures and wrote the manuscript. DTJL, JO, W-HL, N-ZZ, and B-QF provided very constructive suggestions for revisions. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.