Detection and Quantification of the Harmful Dinoflagellate Margalefidinium polykrikoides (East Asian Ribotype) in the Coastal Waters of China

As a marine ichthyotoxic dinoflagellate, Margalefidinium polykrikoides, previously named Cochlodinium polykrikoides, have caused mass mortalities of fish worldwide during blooms. Rapid detection of target species is a prerequisite for the timely monitoring and early warning of harmful algal blooms (HABs). However, it is difficult to achieve rapid identification with traditional methods. The technology of using quantitative real-time PCR (qPCR) to detect and quantify microalgae is relatively mature. Based on the accuracy, rapidity, and sensitivity of qPCR technology, it can be used in the monitoring and development of early warning systems for HABs. From 2017 to 2020, samples were collected from 15 locations off the Chinese coast or from local sea areas. Based on the qPCR detection and analysis, the target species, M. polykrikoides (East Asian ribotype, EAr), was found in samples from Tianjin, Yangtze River estuary, and offshore Fujian (East China Sea). This is the first time that M. polykrikoides (EAr) was detected in the coastal waters of Tianjin. The results reveal a distributive pattern of M. polykrikoides (EAr) along Chinese coastal waters. It is helpful to predict the future diffusion trend of M. polykrikoides (EAr) in the China Sea and provides a practical case for the future construction of monitoring and warning systems for M. polykrikoides and HABs.


Introduction
Margalefidinium polykrikoides [1], formerly known as Cochlodinium polykrikoides [2], is an ichthyotoxic unarmored dinoflagellate that has caused mass mortalities of fish worldwide during blooms, with catastrophic impacts to aquaculture and local economies [1,[3][4][5][6][7]. M. polykrikoides blooms are usually characterized by large spatial scale (10 to 100 km) and high-density aggregation (>10 6 cells·L −1 ) [8]. These blooms are often accompanied by the production of strongly ichthyotoxic compounds, resulting in the death of a large number of marine organisms [9]. The ichthyotoxicity produced by M. polykrikoides can cause high mortality to marine organisms in a short period of time [9]. M. polykrikoides has been

Melting Curve and Standard Curve
Each qPCR product had one informative melting curve. Analyzing the melting curve is useful for the detection of false positives due to primer-dimers or unexpected products. In this study, the melting temperatures were 88 °C and the melting curve showed narrow peaks, indicating that only the target sequence was successfully amplified (Figure 2).
The number of cells corresponding to serial 10 fold dilutions of the DNA extracts was 7.09 × 10 −3 , 7.09 × 10 −2 , 7.09 × 10 −1 , 7.09, and 70.9 cells·µL −1 (cell concentration dissolved in TEbuffer), respectively. There was a strong linear relationship between the CT value and the log10 of the cells number. The regression equation was: y = −3.2417x + 25.52 (R 2 = 0.998) The R 2 values of all the standard curves were over 0.99 ( Figure 3).

Melting Curve and Standard Curve
Each qPCR product had one informative melting curve. Analyzing the melting curve is useful for the detection of false positives due to primer-dimers or unexpected products. In this study, the melting temperatures were 88 • C and the melting curve showed narrow peaks, indicating that only the target sequence was successfully amplified (Figure 2).
The number of cells corresponding to serial 10 fold dilutions of the DNA extracts was 7.09 × 10 −3 , 7.09 × 10 −2 , 7.09 × 10 −1 , 7.09, and 70.9 cells·µL −1 (cell concentration dissolved in TEbuffer), respectively. There was a strong linear relationship between the C T value and the log 10 of the cells number. The regression equation was: y = −3.2417x + 25.52 (R 2 = 0.998) The R 2 values of all the standard curves were over 0.99 ( Figure 3).

Application of qPCR to Field Samples
The applicability of the developed qPCR was tested on the field samples. Three repetitions were set for each sample. M. polykrikoides (EAr) was detected at four sampling sites, including Tianjin (S1), the Yangtze River estuary (S6, September), the Yangtze River estuary (S7, July), and Fujian coastal (S11, May) ( Table 1). The site S11 showed the lowest abundance (1.5×10 3 cells·L −1 ), while S7 displayed the highest concentration with 1.0 × 10 5 cells·L −1 . Overall, 3.6 × 10 3 and 1.9 × 10 4 cells·L −1 were detected at S1 and S6, respectively ( Figure 4). The water temperatures at the stations where M. polykrikoides (EAr) was detected were 27.5 • C (S1), 25.4 • C (S6), 26.2 • C (S7), and 24.7 • C (S11), respectively. The average temperature was 25.9 ± 1.2 • C. The study results showed that M. polykrikoides (EAr) was detected in the Bohai Sea and the East China Sea, while no M. polykrikoides (EAr) was detected in the Yellow Sea and the South China Sea. The sampling time of the stations where M. polykrikoides (EAr) was detected was concentrated from May to September. Station S11, located in Fujian coastal, was sampled in May. Station S7, located in the Yangtze River estuary, was sampled in July. Station S1, located in Tianjin, was sampled in August. As time went on, M. polykrikoides (EAr) was detected from south to north.

Phylogenetic Tree
The qPCR amplification products of field samples with positive results were sequenced, and all sequencing results were an identical sequence. In the clade of M. polykrikoides, specimens were clearly separated into four monophyletic sub-clades. One clade of M. polykrikoides (EAr) was composed of sequences collected from this study and other Asian seas, and all sequences in this clade were completely identical, which was well supported by bootstrap support (BS) values (neighbor-joining (NJ)/maximum likelihood (ML) = 90%/99%). The phylogenetic tree clearly shows the relationships of M. polykrikoides (EAr) and other ribotypes, including the Philippines ribotype, American/Malaysian ribotype, and the Mediterranean Sea ribotype ( Figure 5).
M. polykrikoides (EAr) was composed of sequences collected from this study and other Asian seas, and all sequences in this clade were completely identical, which was well supported by bootstrap support (BS) values (neighbor-joining (NJ)/maximum likelihood (ML) = 90%/99%). The phylogenetic tree clearly shows the relationships of M. polykrikoides (EAr) and other ribotypes, including the Philippines ribotype, American/Malaysian ribotype, and the Mediterranean Sea ribotype ( Figure 5).

Discussion
Although many molecular methods, such as microarrays with molecular probes [35], RFLP [36], high throughput sequencing [37], and FISH-probes [38], have been developed to detect microalgae, qPCR is more widely used to detect and quantify microalgae because of the high accuracy and good sensitivity [27,28,39]. Currently, qPCR has been applied in the field to detect and quantify harmful algae [40][41][42][43]. It may be difficult to detect and SAMITES-S1

Discussion
Although many molecular methods, such as microarrays with molecular probes [35], RFLP [36], high throughput sequencing [37], and FISH-probes [38], have been developed to detect microalgae, qPCR is more widely used to detect and quantify microalgae because of the high accuracy and good sensitivity [27,28,39]. Currently, qPCR has been applied in the field to detect and quantify harmful algae [40][41][42][43]. It may be difficult to detect and quantify microalgae by direct counting with LM at low densities of the target species, but based on the high sensitivity of qPCR, it is possible to detect and quantify the target species even at low densities (<10 cells·L −1 ) [10,44]. The copy numbers of the rRNA gene of dinoflagellates can be up to the order of 10000 [45]. Therefore, even if the extracted DNA contains less than one cell due to dilution, the rRNA gene can be amplified [10]. On the other hand, a high copy number can effectively eliminate PCR inhibitor interference in the qPCR process by diluting the extracted DNA. PCR inhibitors, such as mucopolysaccharides, phenolic compounds, humic acids, and heavy metals in field samples, may cause the qPCR results to be inaccurate or even false negative [46,47]. Based on Park et al. research foundation, the results of this study indicate that the qPCR primer (CPSF2-CPSR3) for the M. polykrikoides (EAr) has a strong specificity and is appropriate for the specific detection and quantification of M. polykrikoides (EAr) [10].
M. polykrikoides is present in tropical, subtropical and temperate waters [10]. Kim et al. [48] research results showed that M. polykrikoides examined in the laboratory exhibited its maximum specific growth rate of 0.41 day −1 at a combination of 25 • C and salinity of 34 psu, and optimum growth rates of >0.3 day −1 were observed at temperatures ranging from 21 to 26 • C and at salinities from 30 to 36 psu. It can be seen that most sea areas in China are suitable for the growth of M. polykrikoides. In 1993, Qi et al. reported the occurrence of red tide of Margalefidinium sp. in Quanzhou Bay, Fujian Province, which caused the death of a large number of marine organisms, but the cause species were not confirmed [15]. In 2009, M. polykrikoides was found in the Pearl River Estuary, South China Sea [49]. In 2014, morphological characterization and phylogenetic analysis of M. polykrikoides isolated from the ECS were carried out by Wang et al. [50]. In 2019, M. polykrikoides was detected in Jiaozhou Bay, Qingdao and demonstrated strong temporal preference with a sharp peak of abundance in early autumn (September), but failed to detect M. polykrikoides from January to May [51]. In this study, we also collected samples in coastal Qingdao on 10 May 2019. It is consistent with the research results of Liu et al. [51], we also failed to detect M. polykrikoides. This may be related to the low water temperature (16.9 • C). However, compared with countries such as Japan and Korea, there are still fewer reports on geographical distribution of M. polykrikoides in China's coastal areas. In 1978, the first outbreak of red tide of M. polykrikoides occurred in the Yatsushiro Sea of Japan, and thereafter, the red tide caused by this species rapidly spread to the extensive waters along the coast of Japan and Korea [22,48,52]. Marine ecosystems are experiencing warming due to global climate change [53]. Seawater warming change the basal metabolic function and species distribution in microalgae [54]. In addition, ballast water is also an important reason for the spread of microalgae species [55]. In a time of global warming and increasingly advanced shipping, the rate of spread of harmful algal bloom species is increasing. Therefore, there is a possibility that M. polykrikoides could cause a massive outbreak in the China Sea. Timely warning is the key to face the outbreak of HABs. qPCR can achieve rapid detection and timely warning.
The results of this study show that M. polykrikoides (EAr) existed in the ECS and Tianjin coastal area, and their concentration can reach 1.0 × 10 5 cells·L −1 at least in the ECS. In this study, field samples were collected from 15 locations, 3 of them were offshore sampling. M. polykrikoides (EAr) was detected in all three samples sampled offshore. However, M. polykrikoides (EAr) was detected in only 1 of the 12 samplings conducted near shore. One of the possible reasons why M. polykrikoides (EAr) were rarely found near shore may be due to the high turbidity of nearshore waters, which can affect the growth and distribution of M. polykrikoides (EAr). Blooms of M. polykrikoides are influenced by prevailing ocean currents [56]. Lee et al. research results showed that the outbreak of M. polykrikoides in coastal areas of Korea was influenced by the Tsushima Warm Current [52]. Large-scale transport of M. polykrikoides blooms by the Tsushima Warm Current also happened in the southwest Sea of Japan [57]. As a branch of the main stem of the Kuroshio in the northeastern waters of Taiwan Island, Taiwan Warm Current (TWC) is a high-temperature, high-salt current that exists year-round in the waters of Fujian and Zhejiang, China [58]. TWC carries high nutritive (phosphate) seawater to the Yangtze River estuary and plays an important role in the hydrology and climate of the ECS [59]. There is no report on the effect of TWC on M. polykrikoides, but there are many studies on the effect of TWC on Prorocentrum donghaiense. The research results of Dai et al. support the hypothesis that P. donghaiense blooms develop from the population at the TWC front in the ECS [60], suggesting the role of the ocean current front as a seed bank to dinoflagellate blooms. Zeng et al.'s research results showed that P. donghaiense blooms first occurred at the northern end of the Taiwan Strait and then moved northward and nearshore with the TWC [61]. Whether TWC will affect M. polykrikoides remains to be investigated. Although the effect of TWC on M. polykrikoides was not addressed in this study, it may be an important direction for future research on the distribution of M. polykrikoides in the coastal waters of China.
In this study, M. polykrikoides (EAr) was detected in the coastal waters of Tianjin for the first time. Since the first reported outbreak of Cochlodinium sp. in Quanzhou Bay in 1993 [15], this study confirmed the presence of M. polykrikoides (EAr) in the offshore of Fujian. M. polykrikoides has a maximum specific growth rate at 25 • C [48]. The water temperature at the time of the M. polykrikoides blooms mostly between 20 to 30 • C [62,63]. In this study, the water temperatures at the stations where M. polykrikoides (EAr) was detected were 27.5 • C (S1), 25.4 • C (S6), 26.2 • C (S7), and 24.7 • C (S11), respectively. The average temperature was 25.9 ± 1.2 • C. Some stations where M. polykrikoides (EAr) was not detected, such as S4, S5, S9, and S10, had water temperatures below 20 • C. Therefore, under suitable environmental conditions, those stations where M. polykrikoides (EAr) was not detected in this study may also exist M. polykrikoides (EAr). The sampling time of the stations where M. polykrikoides (EAr) was detected was mainlyfrom May to September. Following the time and season passing and changing, M. polykrikoides (EAr) was detected from south to north. It suggests that M. polykrikoides (EAr) is likely to keep moving northward as the water temperature rises. In March 2005, massive fish mortalities and water discoloration was reported off the western coast of Puerto Princesa, Palawan, Philippines; phytoplankton analysis revealed a near monospecific bloom of the dinoflagellate, M. polykrikoides [64]. There is a possibility that M. polykrikoides is expanding with seasonal changes from the low to latitude the high latitude sea area, though there still need for further study to clarify this issue.

Conclusions
In this study, qPCR was successfully applied to detect and quantify field samples along the Chinese coast. The target species, M. polykrikoides (EAr), was found in samples from Tianjin, the Yangtze River estuary, and offshore Fujian (East China Sea). This is the first time that M. polykrikoides (EAr) was detected in the coastal waters of Tianjin. Based on the accuracy, rapidity, and sensitivity of qPCR technology, it can be used in the monitoring and early warning system of HABs. It provides a practical case for the future construction of monitoring and warning systems for M. polykrikoides (EAr) and HABs.

Algal Cultures
The algal culture of M. polykrikoides (EAr) was obtained from the Department of Biotechnology at Sangmyung University (Seoul, Korea). The cultures were maintained in f/2 medium at 20 • C ± 1 • C under a light intensity of 65 µmol·Em −2 ·s −1 on a 12:12 h light-dark cycle. The f/2 medium was prepared as described in [65,66]. All seawater at a salinity of 31-33 psu was filtered through GF/F membranes (Whatman, Little Chalfont, UK) and then autoclaved at 121 • C for 30 min. Strains were subcultured with fresh f/2 medium at 20 day intervals to maintain healthy cultures.

LM and SEM
LM images of M. polykrikoides (EAr) in the exponential growth phase were recorded using an inverted microscope (BX53, Olympus, Tokyo, Japan) and analyzed using CellSens Standard 2.3 software (Olympus, Tokyo, Japan) [67]. SEM images were obtained in the following steps. M. polykrikoides obtained from ECS in the exponential growth phase was fixed with 2% glutaraldehyde at 4 • C for 4 h. Following the fixed samples were filtered and collected through a 3 µm polycarbonate membrane (Merck Millipore, Burlington, MA, USA), washed with distilled water, thoroughly removed all fixed reagents and sea salt, dehydrated with a graded ethanol series treatment (30, 50, 70,

Study Area and Field Sampling
From April 2017 to November 2020, samples were collected from 15 locations of the Chinese coast or sea areas, including Tianjin, Beidaihe, Zhangzi Island, Rongcheng, Qingdao, Yangtze River estuary (July), Yangtze River estuary (September), Nanji islands, Fujian coastal (April), Fujian coastal (May), Ningde, Xiamen, Beihai, Weizhou island, and Sanya ( Figure 6). The sampling time and location of each station are listed in Table 1. Seawater ranging from 100 mL to 1200 mL was filtered and collected via filtering with 0.2 µm Millipore filter membrane (Merck Millipore, Burlington, MA, USA). Seawater filtration volume (100-500 mL) were decided according to cell abundance and turbidity of seawater. The filters were placed in a 2 mL microtube containing 800 µL of 2% cetyltrimethylammonium bromide (CTAB) extraction buffer and then stored at −80 • C until DNA extraction.
in the following steps. M. polykrikoides obtained from ECS in the exponential growth phase was fixed with 2% glutaraldehyde at 4 °C for 4 h. Following the fixed samples were filtered and collected through a 3 µm polycarbonate membrane (Merck Millipore, Burlington, MA, USA), washed with distilled water, thoroughly removed all fixed reagents and sea salt, dehydrated with a graded ethanol series treatment (30,50,70,90, 100, and 100%; 30 min per concentration), dried with a critical point drier (Joel Hi-Tech Co., Dalian, China), and gold-coated in a sputter coater. Lastly, the M. polykrikoides cells were observed and photographed under an SEM (TM-1000 Tabletop Microscope, Hitachi High-Technologies Co., Tokyo, Japan) [67,68].

qPCR and Standard Curve Construction
The species-specific qPCR primer used in this study for M. polykrikoides was developed by Park et al. [10]. DNA melting curve analysis was conducted using a CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). Quantitative Real-time PCR assays were performed in a total reaction volume of 20 µL, which contained 10 µL of 1 × SsoFast TM EvaGreen ® Supermix (Bio-Rad, Hercules, CA, USA), 1 µL of each primer (0.1 mM), 1 µL of genomic DNA, and 7 µL double-distilled water. qPCR reactions were run using a CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) at 98 • C for 2 min, followed by 35 cycles at 98 • C for 5 s, then 55 • C for 20 s. The melting curve was increased from 65 • C to 95 • C in 0.5 • C increments, and each step was held for 5 s.
The specificity of a qPCR assay is determined by the primers and reaction conditions used. However, even with well-designed primers, it is always possible to produce primerdimers or unexpected products. The specificity of the qPCR assay can be confirmed using melting curve analysis. The establishment of the standard curve was based on the linear relationship between the C T value and the number of cells. Standard curves were constructed from DNA isolated from 150mL of culture harvested by filtration during the exponential growth phase. M. polykrikoides was counted using a Sedgewick Rafter counting chamber with an LM at 200 × magnification (CKX53, Olympus, Tokyo, Japan) and then filtered to collect the cells. The collected cells were extracted DNA according to the above method. DNA extracts were serially diluted 10-fold and used to construct standard curves.

Phylogenetic Tree
The sequences obtained by qPCR assay in this study and deposited in GenBank (https://www.ncbi.nlm.nih.gov/, accessed on 31 August 2021) were aligned with the sequences obtained from GenBank using BioEdit (North Carolina State University, Raleigh, NC, USA) (Version.7.0.5.3) [71]. The sequences obtained in this study were used together with those in Genbank to construct a phylogenetic tree and using Akashiwo sanguinea, Karenia mikimotoi, and Gymnodinium catenatum sequences as outgroup (Table 2) [72]. This analysis involved 34 nucleotide sequences. All positions containing gaps and missing data were eliminated (complete deletion option). There was a total of 146 positions in the final dataset. Evolutionary analyses were conducted in MEGA X (Pennsylvania State University, State College, PA, USA) (Version.10.2.4) [73]. The sequence obtained in this study is located in the D1-D2 region of the large subunit ribosomal RNA gene (LSU rDNA). ML and NJ phylogenetic tree based on partial LSU rDNA sequences showing the relationships of M. polykrikoides (EAr) and other ribotypes, including the Philippines ribotype, American/Malaysian ribotype, and the Mediterranean Sea ribotype.

Data Analysis
Amplification data were handled in a Bio-Rad CFX Maestro v 3.0 (Bio-Rad, Hercules, CA, USA), with Ct determination mode set to a single threshold and the baseline decided by baseline subtracted curve fit. Unknown cell concentrations were derived directly from the standard calibration curve by a Bio-Rad CFX Maestro v 3.0 (Bio-Rad, Hercules, CA, USA). Raw data were extracted to Microsoft Excel Professional Plus 2010 (Microsoft, Redmond, WA, USA) and OriginPro 2019b (Originlab Co., Northampton, MA, USA), where they were inspected manually.

Data Availability Statement:
No new data were created or analyzed in this study. Data sharing is not applicable to this article.