Genetic structure of Octopus minor around Chinese waters as indicated by nuclear DNA variations (Mollusca, Cephalopoda)

Abstract Octopus minor is an economically important resource commonly found in Chinese coastal waters. The nuclear gene (RD and ODH) approach of investigation has not reported in this species. Rhodopsin (RD) and octopine dehydrogenase (ODH) genes were used to elaborate the genetic structure collected from eight localities ranging from the northern to the southern coast of China. In total, 118 individuals for the RD gene and 108 for the ODH were sequenced. Overall (RD and ODH) genes resulted in high (0.741±0.032; 0.805±0.038) haplotype and low nucleotide (0.01261±0.00165; 0.00747±0.00086) diversity. Molecular variance displayed higher values among the populations and lower values within the population where the fixation index FST denoted 0.880 and 0.584 in RD and ODH genes respectively. The Dongshan population clustered separately in a phylogenetic tree as in the haplotype networking assessment. The current data suggests that the Dongshan population needs separate management.


Introduction
The class Cephalopoda embraces animals which are exclusively marine inhabitants. They have immense commercial and ecological significance, including a profound contribution as a source of protein for humans. The cephalopod has lobed and folded brain like that of vertebrates and are ingenious, migrant and largest of all molluscs (Mather and Kuba 2013;Cheng et al. 2013;Larson et al. 2015) and it is sensitive to environmental factors (Emery et al. 2016;Wang and Zheng 2017 Among cephalopods, the octopuses contribute 33% to the existing cephalopod assembly. Three hundred species of octopus are thought to exist along the coastal waters of Korea, China, and Japan (Roper et al. 1984;Norman and Sweeney 1997;Kang et al. 2012). One hundred thirty-four species are reported only in Chinese waters including the profit-making species like Octopus minor, Amphioctopus fangsiao, and Cistopus chinensis (Lu et al. 2012).
Previously, mtDNA molecular markers were popular because of their high mutation rate, maternal inheritance, and non-recombination (Vaseeharan et al. 2013). In the last decade, the nuclear DNA markers have been widely used for various investigations including identification, population genetics, comparisons between wild and captive populations, demographic evaluations, and rehabilitation projects (Chauhan and Rajiv 2010). Carlini et al. (2000) was first who used nuclear DNA (Actin) for phylogenetic analysis of coleoid species subsequently other authors continued nuclear approach of investigation such as, (Warnke et al. 2003;Lindgren et al. 2004;Strugnell et al. 2005).
The nuclear gene provides more information than mtDNA genes (Graybeal 1994). The rhodopsin and octopine dehydrogenase genes were used for phylogenetic and phylogeographic analysis of octopuses (Strugnell and Nishiguchi 2007;Toussaint et al. 2012). However, meagre information is available on population genetics of octopuses using RD and ODH genes. The RD and ODH genes are less complex and can provide better results than 18S rDNA. Our present study aims to focus on collecting fundamental information about the population structure of this species using nuclear genes.

Materials and methods
Samples were collected from eight locations (Fig. 1). Thereafter, they were preserved in 95% ethanol and transported to the laboratory. The total genomic DNA was isolated from muscle tissues using a standard protocol. Target genes were amplified by PCR using primers (Table 1). The total 25 µl PCR mixture included (DNA template 1.25 µl, each primer 1.25 µl, ES Taq 12.5 µl, and 8.75 µl water). Thermocycler conditions were as follows: denaturation at 94 °C for 5 min, 35 cycles 94 °C for 30 s, annealing at 55 °C for 30 s, extension at 72 °C for 30s, and the final extension at 72 °C for 7 min. Electrophoresis was performed on a 1.2% agarose gel and was sequenced using the same oligonucleotide primers. Sequences were aligned using MEGA 6 software (Tamura et al. 2013). Analysis of genetic Table 1. Oligonucleotide primers used for the amplification of nuclear rhodopsin and octopine dehydrogenase genes in Octopus minor. differentiation, AMOVA, molecular diversity indices, genetic differentiation values, and F ST values were determined with the ARLEQUIN software (Excoffier and Lischer 2010). Calculations of gene flow (N m ) were performed using formula N m = (1-F ST )/2F ST . Haplotype and nucleotide diversity were estimated using DnaSP (Librado and Rozas 2009). The neighbor joining tree was constructed to check the genetic relationship between populations using MEGA 6 (Tamura et al. 2013). The haplotype networking was created using NETWORK software version 5.0.0.1 (Bandelt et al. 1999).

Rhodopsin Gene (RD):
A fragment of 637 bp of RD was sequenced from 118 individuals. RD gene showed 33 polymorphic sites and 13 haplotypes. Haplotype diversity (HD) ranged from 0.090 to 0.833, nucleotide diversity (π) remained in the lower range (0.00027-0.001), where the average number of differences (k) ranged from 0.166 to 1.388 (Table 2). Molecular variance revealed a higher percentage among the populations (88.04%) while lower values were retrieved within (11.96%) populations; fixation index F ST was 0.880 (Table 3). The pairwise F ST values ranged from 0.014 to 0.989. The lowest gene flow was observed in Dongshan population ( Table 4). The values of Tajima's D tests generally showed negative values with exception to Xiamen population. The Wenzhou and Dongshan populations were statistically significant  (Table 6). The phylogenetic analysis of eight populations separated Dongshan population with 99% bootstrap support (Fig. 2). The networking analysis revealed that haplotype three (Hap 3) was shared by six populations (Dalian, Nantong, Zhoushan, Qingdao, Shanghai, and Xiamen), Hap 4 was contributed by five populations (Dalain, Nantong, Qingdao, Xiamen, Zhoushan), Hap 2 shared between Dalian, Zhoushan, and Nantong, while Hap 7 appeared in two populations (Qingdao, Nantong), Hap 11 and Hap 12 were present in the Wenzhou and Xiamen populations, and Hap 5 and Hap 6 were independently represented in the Dongshan population (Fig. 4). Octopine dehydrogenase (ODH): A fragment of 597 bp was sequenced from 108 individuals, revealed 27 polymorphic sites and 29 haplotypes. Haplotype diversity remained higher (0.222-1.000) than nucleotide diversity (π) (0.001-0.004). The average number of differences (k) ranged from 0.444 to 2.733 (      (Table 6). The neighbour-joining phylogenetic tree described two distinct clades where Dongshan population clustered separately with 100% bootstrap support (Fig. 3). The median joining network analysis described that Hap1 shared by seven populations followed by Hap2, which appeared in four populations, namely Dalian, Nantong, Qingdao, and Zhoushan. Hap 13 contributed by three populations (Dalian, Shanghai, and Zhoushan) similarly Hap 6 appeared in Dalian, Shanghai and Xiamen populations, Hap 17 shared by Nantong, Zhoushan, and Xiamen populations, Hap 16 shared by Nantong and Xiamen, Hap 19 appeared in Qingdao and Shanghai populations while Hap, 25 shared between Zhoushan and Wenzhou populations. Hap 8-12 were independently representing Dongshan population (Fig. 5).

Discussion
Octopus minor is a bottom-dwelling inhabitant and its migration is limited. The dispersal capacity has immense influence on population genetics. Tag-recapture investigations of O. vulgaris reported to be restricted within one km from the point of release (Melis et al. 2018 and references therein). Conversely, dispersal range of O. minor juveniles and adults is not precisely known. The basic information of genetic variation and population structure is valuable for stocking, fisheries management, and conservation (Feng et al. 2017). Several divergent forces cause genetic differentiation, including geographic isolation, current and life history characteristics (Gao et al. 2016). In fact, oceanic processes are complex and a single reason cannot be claimed as source of divergent force. There are many islands and gulfs in China's sea, which can contribute to the gene flow complications of the populations (Gao et al. 2016). Earlier several studies were under taken including complete mitochondrial genome of this species (Cheng et al. 2012). Previous investigations on population genetics of O. minor present subtle to significant differences (Gao et al. 2016;Yang et al. 2015;Lü et al. 2013;Kang et al. 2012;Xu et al. 2011;Sun et al. 2010;Li et al. 2010). Our present analysis of two nuclear DNA gene sequences in O. minor collected from eight locations imparted variation. The haplotype diversity of RD gene ranged (0.090-0.833), the higher haplotype diversity was noted in Xiamen population and lowest haplotype diversity showed in Dongshan population. The ODH gene haplotype diversity was higher in Shanghai population (1.000) and was lowest in Wenzhou population (0.222). Comparative studies of Kang et al. (2012) between Korean and Chinese populations showed less haplotype diversity in Korean population than Chinese populations. However, only three Chinese populations were   (Weir and Cockerham 1996). Based on this standard, the results obtained in this study showed high differentiation between Dongshan and other counterpart populations (Tables 4, 5). The Tajima's D analysis of the RD gene showed negative values except for the Xiamen population, unlike to the results of the CO1 studies . The ODH gene revealed negative values for Tajima's D except for three populations, Shanghai, Zhoushan, and Dongshan. Results are similar to mtDNA COI investigations  where the Zhoushan population had positive values; Shanghai and Nantong populations were not included in COI studies. The RD gene revealed 13 haplotypes; among them Hap 3 was common, shared by 6 populations. None of the haplotypes of the Dongshan population was shared by other counterpart populations with similar COII and Cytb investigations Li et al. 2013). Nevertheless, it differs with COII and Cytb Li et al, 2013) where Wenzhou, Zhoushan, and Wenzhou, Xiamen populations remained isolated populations respectively. The ODH gene described 29 haplotypes among which Hap1 remained common, which was shared by seven populations. It is consistent with Lü et al. (2013) and Li et al. (2013) with respect to Dongshan population whereas differ with reference to Wenzhou and Xiamen populations. Studies of Kang et al. (2012) described sharing of one haplotype between Chinese and Korean populations. The neighbour-joining phylogenetic tree of both the genes distinctly clustered Dongshan population separately as in Cyt b, and COII studies Lü et al. 2013). However, COII unveiled Wenzhou and Zhoushan populations as separate clade ; similarly, 16S rRNA showed the Xiamen population as a separate clade .  reported Wenzhou and Xiamen populations as separate clade. Kang et al. (2012) investigated three clades where only one Chinese population (Dalian) parted as sub-clade along with Korean populations. Yang et al. (2015) using AFLP and Sun et al. (2010) using CO1 reported two clades, which is inconsistent with present study. The discrepancy of data is consequences of various unknown oceanic process, genetic markers used and range of sampling locations ). To understand the data variation of O. minor along the Chinese coast, it is imperative to study the larval and adult dispersal range along with seasonal oceanic process during ontogenesis. Present data does not support the isolation by distance (IBD) because the geographic coastal distance between Dongshan and Xiamen is only 157 KM while the gene flow between them is meagre as shown by RD and ODH (0.018; 0.050), whereas the coastal distance between Dalian and Xiamen is approximately 3700 Km but the gene flow is higher (7.436; 3.632). Lü et al. (2013) was also noted gene flow discordance but the concrete reason is unknown. Melis et al. (2018) emphasized the causes of meagre larval dispersal in O. vulgaris including high mortality, philopatric behaviour of larvae and potential cryptic berries. Nevertheless, the above-mentioned factors are difficult to be disentangled, furthermore, oceanic fronts (temperature, salinity, density, turbidity, nutrients, velocity), upwelling and current systems can also influence the larval dispersal (Melis et al. 2018). It has long been entrenched that genetic structure of populations affected by glacier activities where sea level encountered climatic fluctuations during the Pleistocene period and caused gene flow restrictions in marine organisms (Imbrie et al. 1992). However, relatively high haplotype diversity reported in freshwater fish inhabiting nonglaciated regions or temperate regions (Bernatchez and Wilson 1998;Tan et al. 2015).
Most of the studies related to O. minor gave high haplotype diversity in this region, including the present study.
Our present investigation has significant implications for conservation and favourable management of O. minor along the Chinese coast and it also shed light on the need for separate management of the Dongshan population: once an evolutionary lineage is lost, there is no possibility to be recover it (Mortiz 2002). We also recommend infield ontogenetic studies and larval migration ranges observations along with physical oceanographic parameters in future studies to understand the population genetics data inconsistency in O. minor reported along Chinese coastal waters.