Rigorous understanding of the freshwater zooplankton fauna using molecular ecological methods: a case study in Japan

概要

Zooplankton play fundamental roles in the food webs of freshwater lakes and ponds. Their community structure has a direct influence on water quality characteristics such as Secchi Depth transparency (Gannon and Stemberger, 1978; Horn, 1991; Makino et al., 2001; Sprules, 1977, Takamura et al. 1999). To keep freshwater lakes and ponds in good conditions for human use, careful monitoring on zooplankton is very important.

Zooplankton taxonomy has improved significantly in recent years as a result of using molecular ecological analyses and having more detailed morphological analyses (Table 1). Such molecular analyses are able to separate even morphologically-indistinguishable-but- genetically-separable cryptic species complexes (Table 2), which results in a more accurate description of zooplankton fauna. However, such updated information on zooplankton taxonomy and the advantages of molecular ecological methods (MEMs) are largely unacknowledged in most nations including in Japan, despite the monitoring efforts that have been made for years. Furthermore, recent taxonomic updates have been biased to cladocerans (e.g., Galibian et al., 2021; Kotov et al., 2009; Maruoka et al., 2018), merely because taxonomic studies with MEMs are more prevailing in cladocerans than copepods. Therefore, it is essential that the methodology used in zooplankton monitoring becomes up to date with the current technological advances.

The purpose of this study is to highlight the importance of MEMs for deepening our understanding of freshwater zooplankton. To lucidly demonstrate that, the first two chapters will concentrate on copepods, since freshwater copepods in Japan are less focused on compared to cladocerans (Tables 1 and 2). The emphasis will specifically be on two closely related species of the genus Cyclops (Copepoda: Cyclopoida).

Chapter 1 resolves issues related to the taxonomic ambiguity of the Japanese Cyclops vicinus Uljanin, 1875 and Cyclops kikuchii Smirnov, 1932 using both MEMs and traditional morphological methods. In Japanese literature, the nomenclature and distinction between these two congeners are still questionable and they are both identified as “C. vicinus” in most of the cases (thus, hereafter C. vicinus s.l.). This chapter determined the taxonomic position of the Japanese C. vicinus s.l. by morphological analyses and by sequencing nuclear internal transcribed spacer 1 (ITS-1) and the mitochondrial 12S (mt12S) genes. The ITS-1 phylogeny separated Japanese C. vicinus s.l. into two genetic groups; one clustered with European C. kikuchii (thus, CkikJPN), while the other clustered with European C. vicinus (thus, CvicJPN). Morphological analyses using the lengths of terminal furcal setae revealed that CkikJPN and CvicJPN correspond to C. kikuchii and C. vicinus, respectively. These results showed that C. vicinus and C. kikuchii in Japan were confidently the same species as those in Europe. The degree of divergence in the mt12S between Japanese and European populations varied largely between the species, as it was nearly at the interspecific level in C. kikuchii, while it was at the intraspecific level in C. vicinus; it is the first case in which the genetic divergence between the European and Asian population is greatly different among closely related species of freshwater copepods (Sioud et al., 2021).

Chapter 2 also results from applying MEMs, with more extensive sampling efforts across the country compared to those in Chapter 1. The aim was to have a better grasp of the two Japanese species’ spatial distribution and to inspect whether they exhibit similar phylogeographic patterns (in terms of mitochondrial cytochrome c oxidase subunit I, mtCOI) and to subsequently infer any differences in their ecological characteristics. The results showed a distinct latitudinal structure separating the two congeners; even when their occurrence zones overlapped. C. vicinus was more frequently found in the south, while C. kikuchii dominated in the north. This suggests that C. kikuchii has a stronger spatial competition than C. vicinus in the north and vice versa. Therefore, the odds of C. “vicinus” being rightly C. kikuchii in previous literature are higher in specimens which were collected from the north. Furthermore, results showed that the Japanese C. kikuchii was comprised of two phylogroups with a large genetic differentiation, suggesting a case of secondary contact. Both species did not exhibit any strong sign of demographic expansion after bottleneck events in Japan. Additionally, C. kikuchii had many private haplotypes across Japan, while C. vicinus possessed several geographically widespread haplotypes. This indicates that the realized range of dispersal and frequency of effective dispersal of C. vicinus are likely larger than those of C. kikuchii. Consequently, we argued that the dispersal ability was stronger in C. vicinus than it was in C. kikuchii.

Chapters 1 and 2 collectively show the inevitable necessity and evident usefulness of MEMs for freshwater zooplankton. It is expected that the utility of MEMs in future monitoring and studies would be maximized by adopting high throughput sequencing technologies such as “metabarcoding”, which allows the simultaneous identification of many taxa within the same sample. Nevertheless, there are limitations to these technologies such that the amplicon size should be kept shorter than in the case of Sanger sequencing technology and that the PCR efficiency may vary largely between different zooplankton species within the same sample. As for the PCR efficiency in zooplankton metabarcoding, the results of previous studies (Clusa et al., 2021; Zhan et al., 2014; Zhao et al., 2021) are not helpful, since they usually ignore the differences in PCR efficiency between species, adding to the fact that species composition of zooplankton in Japan must be vastly different from those in the previous studies due to the presence of endemic species in Japan (e.g., Makino et al., 2018). Therefore, in Chapter 3, I seek the best gene region(s) for freshwater zooplankton in terms of metabarcoding with high throughput sequencing technologies. Specifically, mock community experiments with defined mixtures of different and common zooplankton species in Japan are conducted.

In the experiments, extracted DNA from each mock community was individually amplified in PCR for three gene regions [mt12S, mtCOI, and the D6 domain of nuclear ribosomal DNA (nr28S)]. PCR products from different mock communities were sequenced with a high throughput sequencer (HTS). The results showed substantial differences in PCR efficiency among the species in all gene regions examined, such that different species were nearly eliminated in different gene regions. For example, Eodiaptomus japonicus, which is a common copepod species in Japan (along with C. vicinus), was rarely detected when using mtCOI, while it was evidently detected when nr28S was used. On the other hand, another common zooplankton taxa, Thermocyclops spp., was rarely detected in nr28S, while it was distinctly detected in mtCOI. It also appeared that mt12S detected copepods successfully, but failed to detect cladocerans. Based on these results, it is concluded that using two gene regions, namely mtCOI and nr28S, simultaneously in the metabarcoding is the best choice for Japanese freshwater zooplankton communities. Finally, the present experiment uncovered that there were taxa which were over- and under-represented compared with their biomass in all gene regions, and such over- and under-represented taxa varied with gene regions used in the HTS analysis. In other words, the metabarcoding did not produce quantitative data. Therefore, the quantitative nature of zooplankton monitoring data should be ensured by traditional quantitative collection methods and traditional counting operations using a microscope.

この論文で使われている画像