Introduction
Interest in the preservation of the environment has been heightened by
environmental disruptions. Investigating biodiversity, population size,
and time-course changes associated with environmental change is
important for the conservation of biodiversity. Traditionally,
monitoring of species has been carried out by counting individuals using
distinct morphological characters. Although the direct monitoring of
species is a reliable approach, it is laborious and relies on intense
monitoring and sampling efforts. Moreover, certain species, such as
nocturnal species, are difficult to investigate. Furthermore, direct
monitoring is sometimes invasive to the environment and individual
organisms.
Environmental DNA (eDNA) is a mixture of DNA released to the environment
from many different species in the form of mucus, saliva, faeces, urine,
gametes, and skin (Barnes and Turner, 2016; Thomsen and Willerslev,
2015; Deiner et al., 2017). By monitoring eDNA, we can monitor species
in the environment without the need for sampling efforts or visual
identification expertise. In the past decade, eDNA has been used to
investigate the biodiversity of various species such as fish, plants,
fungi, birds, and mammals (Tsuji, Takahara, Doi, Shibata, and Yamanaka,
2019). eDNA analysis is not only applied to the present environment, but
also to studies of past biodiversity in ice cores and sediments
(Parducci et al, 2017; Willerslev et al., 2007).
For ecological community analysis, DNA fragments (e.g., mtDNA and
nuclear rRNA genes) used for phylogenetic identification are amplified
by PCR followed by high-throughput sequencing (HTS). Species abundances
in the community are often determined according to the obtained number
of sequences or relative abundance of the sequences of each taxon in a
sequence library. However, these methods are not quantitative because
efficiencies of PCR amplification vary depending on several factors,
such as amplified DNA sequences (e.g. GC content or the base adjacent to
primers) and primer sequences when degenerate primers are used (Ben‐Dov,
Shapiro, and Kushmaro, 2012; Ruijter et al., 2009; Salipante et al.,
2014; Sipos et al., 2007).
Quantitative eDNA data from each taxon is useful for estimating species
biomass/abundance and determining the effects of environmental
disruption. DNA quantification has been conventionally performed by
quantitative PCR (qPCR) (Takahara, Minamoto, Yamanaka Doi, and Kawabata,
2012; Doi et al., 2017) and more recently by digital PCR (dPCR), which
is more accurate than the former because of tolerance to PCR inhibitory
substances (Hoshino and Inagaki, 2012). For quantification by qPCR or
dPCR, establishing an assay for each target is needed, including the
design of PCR primers, preparation of standards (if qPCR is used), and
optimization of PCR conditions. Therefore, the simultaneous
quantification of many species is not straightforward.
Quantitative sequencing (qSeq), which has been recently developed
(Hoshino and Inagaki, 2012; Hoshino and Hamada, 2017), enables
simultaneous sequencing and quantification of many species in a single
HTS run. In qSeq analysis, a random sequence tag is added to the 5′ end
of the target sequence during single primer extension prior to PCR
amplification to prepare the sequence library. If the variety of random
tag sequences is sufficiently large relative to the number of targeting
DNA molecules, the distribution of the random tag to DNA molecules
follows the Poisson statistic. Therefore, after high-throughput
sequencing, the number of DNA molecules in a sample can be estimated by
counting the variety of random tags at the 5′ end of the targeted
sequence without being affected by PCR bias.
In this study, we applied qSeq to eDNA analysis of five species of fish
in aquaria and compared the results with dPCR and relative
quantification using conventional HTS.