1 Introduction
Biological invasion denotes the phenomenon that an alien species spreads outside its natural range and proliferates and persists in new habitats. This is increasingly occurring as a result of human activities and causes negative impacts on native ecosystems (Mack et al., 2000; Reid et al., 2005). Nevertheless, invasive populations are also interesting models to investigate rapid genetic response and adaptation to novel environments, thus providing valuable insights into basic biological processes (Prentis et al., 2008; Sakai et al., 2001). In general, an introduced population tends to lose its genetic diversity because of bottlenecks which can reduce the fitness and evolutionary potential (Lee, 2002). The genetic paradox of invasion appears when a bottlenecked introduced population becomes invasive (Estoup et al., 2016). However, recent studies provide evidence that the genetic paradox of invasion is to a large extent overrated. In some cases, no paradox exists, as introduced populations shows no loss of genetic diversity possibly owing to multiple introductions, or no adaptive challenge in the introduced habitat (Blumenfeld et al., 2021; Facon et al., 2006; Rius et al., 2015). In other cases, an apparent paradox is spurious, when the loss in genetic diversity of introduced populations mainly occurs at neutral genetic markers which are not reflected in adaptive traits, or the diversity loss is a consequence of a successful response to strong selection (Dlugosch & Parker, 2008; Estoup et al., 2016; Gonzalez et al., 2013). Genetic analysis of the genetic architecture and composition of introduced populations would provide insights into the evolutionary mechanisms facilitating the invasion success.
High-throughput sequencing technologies, such as whole genome resequencing and reduced representation sequencing, provide huge number of molecular markers for genome-wide population genetic studies in non-model organisms (Ellegren, 2014). Genome-wide markers, such as single nucleotide polymorphisms (SNPs), can be used for accurately estimating genome-wide genetic diversity both within individuals and at population levels and provide powerful tools to uncover population genetic structure and reconstruct invasion history (Austin et al., 2006; Baltazar-Soares et al., 2020; Le Moan et al., 2021; Liu et al., 2018). Moreover, genome scans allow detecting possible footprints of selection associated with local adaptation, albeit not without challenges (Haasl & Payseur, 2016). Finally, significant progress has been made in using genomic data for reconstructing long-term demographic history and estimating genomic inbreeding level and recent demographic history which affect genetic diversity (Beichman et al., 2018; Ceballos et al., 2018; Dong et al., 2021).
The Tibetan Plateau is one of the largest and highest plateaus on Earth. The fish diversity of the Tibetan Plateau is very sensitive and vulnerable to biological invasion due to the fragile ecosystem and unique fish fauna (Favre et al., 2015; He et al., 2020; Jia et al., 2019; Tao et al., 2018). One of the most widespread invasive species in the region is the goldfish Carassius auratus which is originally distributed throughout the East Asian region except for the Tibetan Plateau (Luo & Yue, 2000). It represents a remarkable species complex containing individuals with different ploidy levels in natural waters (mainly diploid and triploid, and in rare cases tetraploid) (Liu et al., 2017b; Xiao et al., 2011). They have different reproduction modes, with diploids exhibiting sexual reproduction and triploids exhibiting unisexual gynogenesis (Gui & Zhou, 2010; Zhou et al., 2000). Triploids produce chromosome number-unreduced eggs by suppression of the first meiotic division. Eggs are subsequently activated by the sperm of sympatric sexual species to initiate embryogenesis, resulting in clonal offspring from their mothers (Wang et al., 2022). Such parthenogenesis in vertebrates is also observed in a few other fishes, such asPoecilia formosa (Warren et al., 2018) and Phoxinus eos-neogaeus (Angers & Schlosser, 2007). Unisexual vertebrates are generally predicted to have low evolutionary potentials due to lack of meiotic recombination which thus result in the accumulation of deleterious mutations and hindrance of the creation of genetic diversity (Butlin, 2002). On the other hand, asexual organisms may have advantages to be better colonizers than sexual organisms, because they can populate new habitats without mates thus avoiding inbreeding (Avise, 2008). The unisexual triploid C. auratus derives from sympatric diploids by multiple independent polyploidization events and possesses a comparable or slightly higher genetic diversity compared with diploids (Liu et al., 2017c; Luo et al., 2014; Ren et al., 2018). If both sexual diploids and unisexual triploids have been introduced into the Tibetan Plateau, this raises an interesting possibility to compare the genetic responses of the two forms.
C. auratus is one of the most popular fishes in Tibet markets and are bought and released into local waters for religious reasons (Jia et al., 2019), resulting in repeated introductions of the species into the waters of Tibet. Based on our preliminary market survey (data not published), C. auratus in Tibet is directly imported from the aquatic product markets in Ningxia (NX: located on the upper Yellow River) and Sichuan (SC: located on the upper Yangtze River) Provinces mainly by the two main transportation routes (Qinghai-Tibet Highway and Sichuan-Tibet Highway) connecting Tibet and its east (Figure 1). In addition, C. auratus in Ningxia was also imported from the central and eastern regions along the Yangtze River and Huai River. Hence, invasive populations of C. auratus in Tibet came from diverse sources which may be genetically distinct. Previous studies of multiple invasive species have shown that multiple introductions from the same or distinct source populations may have prevented or even reversed loss of genetic diversity within the invasive range owing to founder events (Blumenfeld et al., 2021; Kelager et al., 2013; van Boheemen et al., 2017). However, given the extreme environmental conditions on the Tibetan Plateau (Feng et al., 2019), this begs the question to which extent genetic diversity of C. auratus has been shaped by the extreme environments that it has been transplanted into during multiple introductions.
In this study, by analyzing whole-genome sequence data of 151 goldfish individuals from two invasive and 11 native populations, we investigated how genomic changes occurred in invasive populations in Tibet derived from diverse sources. We specifically tested the hypothesis that different genetic imprints of the colonization process would be detected between invasive diploid and triploid populations, with loss of variation and inbreeding occurring in diploids, but with no loss of variation in triploids. To address this, we first determined the ploidy levels of each individual and called SNPs for diploids and triploids separately. Subsequently, genetic diversity and population genetic structure were analyzed in both invasive and native populations, with a specific focus on detecting inbreeding and footprints of recent founder events in invasive populations. Finally, putative signatures of selection in invasive populations were analyzed, with the expectation that such patterns would be evident only in diploids. Our results provide insights into evolutionary processes facilitating the invasion success in Tibet and also insights into different genetic responses to novel habitats between sexual and asexual forms of vertebrate species.