Association of clonal diversity and population growth in the small-type rotifer Brachionus koreanus during hatchery mass production

Aquaculture 465 (2016) 296–302

Contents lists available at ScienceDirect

Aquaculture journal homepage: www.elsevier.com/locate/aquaculture

Association of clonal diversity and population growth in the small-type rotifer Brachionus koreanus during hatchery mass production Eitaro Sawayama a, Wilma Moka b, Daiki Noguchi c, Motohiro Takagi d,⁎ a

R&D Division, Marua Suisan Co., Ltd., 4472 Iwagi, Kamijima, Ehime 794-2410, Japan The United Graduate School of Agricultural Sciences, Ehime University, 3-5-7 Tarumi, Matsuyama, Ehime 790-5866, Japan Nippon Total Science, Inc., 456-2 Minomi cho, Fukuyama, Hiroshima 720-0832, Japan d South Ehime Fisheries Research Center, Tarumi Branch, Ehime University, 3-5-7 Tarumi, Matsuyama, Ehime 790-8566, Japan b c

a r t i c l e

i n f o

Article history: Received 10 February 2016 Received in revised form 1 September 2016 Accepted 11 September 2016 Available online 12 September 2016 Keywords: Rotifer Brachionus koreanus Clone Haplotype Microsatellite Multi-locus genotype Multi-locus lineage

a b s t r a c t In this study, we identify clones of S-type rotifers Brachionus koreanus based on haplotypes of mitochondrial gene cytochrome c oxidase subunit I (mtCOI) and microsatellite DNA genotypes. We then study how clonal diversity affects the population growth of rotifers in a mass culture at a hatchery using a lot of B. koreanus rotifers for analysis. Microsatellite DNA markers were also developed to identify clones of B. koreanus. Clones of B. koreanus were identified based on haplotypes of mtCOI, microsatellite genotypes, as well as a combination of these haplotypes and genotypes. Three haplotypes of mtCOI, three clones based on microsatellites, and six clones based on a combination of mtCOI haplotypes and microsatellite genotypes were identified. The population growth rate of masscultured rotifers was monitored for a month, and the correlation between population growth rate and these clonal diversities was analyzed. Observed was a significant positive correlation between the population growth rate and haplotype diversity (r = 0.695, P = 0.004), however no correlations were found between the population growth rate and clonal diversity based on either microsatellite genotypes (r = 0.320, P = 0.245) or a combination of mtCOI haplotype and microsatellite genotypes (r = 0.435, P = 0.105). Some clones shared mtCOI haplotypes and microsatellite genotypes suggesting sexual reproduction occurred in the hatchery stock of B. koreanus. Statement of relevance: This study showed the correlation between population growth rate and clonal diversities based on mtCOI haplotypes and microsatellite DNA genotypes in Brachionus koreanus. There was a significant correlation between the population growth rate and haplotype diversity and suggested genetic factor is one of the possible causes affecting population growth rate. Our results will be useful in mass-production and maintenance of Brachionus koreanus at hatcheries. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Small-morphotype (S-type) rotifers belonging to the Brachionus species have been used as larval food in the larviculture of many marine fish species (Hagiwara et al., 2001). Rotifers are used as an initial food in larval rearing for 10–30 days after mouth opening (Lubzens et al., 1989), before feeding with Artemia and formula diets. S-type rotifers are commonly used in hatcheries worldwide because this morphotype of rotifers is easily adopted and grown in hatchery environments (Papakostas et al., 2006). Several studies based on DNA barcoding analysis revealed that hatchery-used S-type rotifers in Europe and Japan belong to B. koreanus, formally called as B. sp. “Cayman” (Papakostas et al., 2006; Baer et al., 2008; Hwang et al., 2013; Moka et al., 2016; Mills et al., 2016). Rotifers can suddenly die in very large numbers with a subsequent decrease in culture density, usually called a “Crash”. Such Crashes ⁎ Corresponding author. E-mail address: [email protected] (M. Takagi).

http://dx.doi.org/10.1016/j.aquaculture.2016.09.020 0044-8486/© 2016 Elsevier B.V. All rights reserved.

have been one major obstacle to rotifer production since rotifer culture started. Once rotifer production has Crashed, sufficient numbers of rotifers for larval rearing are not produced in a hatchery, therefore, seed production is not able to continue. Crashes are thought to be caused by environmental factors such as: food quality and the absence of vitamin B12 (Scott, 1981; Hirayama, 1987), or competition with other organisms such as viruses, bacteria, and ciliates (Hagiwara et al., 1995; De Araujo et al., 2000). However, based on data, a few genetic factors has been considered a cause of such Crashes (Papakostas et al., 2007). We previously identified three haplotypes of mitochondrial gene cytochrome c oxidase subunit I (mtCOI) in the S-type rotifer B. koreanus of hatchery strains in Japan, and usually several haplotypes were mixed in a mass culture (Moka et al., 2016). This finding suggests that the genetic diversity of S-type rotifers changes during mass culturing and it may be one cause of the Crashes. Billions of rotifers are cultured in mass-production tanks every day. Although sexual recombination may occur in continuous and batch culture systems (Declerck et al., 2015), standing genetic variation may be a key factor affecting population growth in mass culture conditions. Therefore, genetic diversity using

E. Sawayama et al. / Aquaculture 465 (2016) 296–302

Nomenclature S-type small-morphotype mtCOI mitochondrial gene cytochrome c oxidase subunit I MLG multi-locus genotype MLL multi-locus lineage mtMLL mtCOI haplotype msMLL microsatellite DNA based-MLL ms + mtMLL MLL combining information on both mtMLL and msMLL

maternal DNA such as mtCOI may be underestimated. If rotifers in a mass culture possess large genetic diversity, genetic factors could affect population growth and be considered a cause of Crashes. Microsatellite DNA markers are commonly used as a tool to identify clones of many living organisms (Guo and Gui, 2008; Halkett et al., 2005) including Brachionus rotifers (Gómez and Carvalho, 2000; Campillo et al., 2009; Papakostas et al., 2009; Declerck et al., 2015) because of high rates of polymorphisms. Papakostas et al. (2009) successfully identified the clonal composition of L-type rotifers used in aquaculture based on microsatellites. Therefore, microsatellite DNA markers are also expected to be a useful tool in identifying the clonal structure of the S-type rotifer B. koreanus. In this study, we developed microsatellite DNA markers for the S-type rotifer B. koreanus, and clonal lineages were estimated based on the polymorphisms of microsatellites. In addition, the possibility of sexual reproduction was estimated by comparing mtCOI haplotype with clonal lineages composed of microsatellite DNA polymorphisms. Correlations between population growth and clonal diversity in mass culturing were analyzed based on mtCOI and microsatellite genotype data. 2. Materials and methods 2.1. Rotifer culture and sampling Rotifers were intensively cultured, at Marua Suisan Co., Ltd., in 3 kL tanks with a 3-day interval of batch culturing. Water temperature and salinity were set at 27 °C and 25 ppt, respectively. Rotifers were fed concentrated freshwater Chlorella (V12; Chlorella Industry Co., Ltd., Tokyo, Japan) twice a day. To estimate population growth rate, the number of rotifer individuals per mL was manually calculated under a profile projector, V12 (Nikon, Tokyo, Japan). 1 mL of culturing media was sampled and the number of rotifer individuals counted. Rotifer individuals were counted three times, and the average number of rotifer individuals/mL was used as a measure of rotifer density (individuals/mL). Rotifer density was set at approximately 500 individuals/mL (1.5 × 109 individuals in 3 kL) on the first day. Population growth rate of 3 days per batch was also calculated based on the following ratio: Population growth rate ¼ Nt=No

where: No is the initial rotifer density (individuals/mL) Nt is the rotifer density on day t. Sampling of rotifer individuals for further population genetic analysis was conducted at the end of each batch culturing of 31 days (batches I to XV). Culture media (1 mL) containing rotifers was taken from the culture tank, and 24 individuals per batch (total 356 individuals) were individually collected into a PCR tube, DNA was then extracted according to the method by Moka et al. (2015) and stored at − 20 °C until genotyping.

297

2.2. Neutral marker analysis We used two types of DNA markers to identify rotifer clones: mtCOI and microsatellite. PCR of mtCOI was performed according to the method by Moka et al. (2015). A partial mtCOI sequence (534 bp) was used for the genetic analysis. Partial COI sequences were aligned with ClustalW (Thompson et al., 1994), and the number of haplotypes was identified using MEGA 5.1 software (Tamura et al., 2011). Haplotype diversity (h) was also calculated using GenAlEx. To conduct microsatellite analysis for S-type rotifer, we developed species-specific microsatellite DNA markers for B. koreanus. A hatchery strain of B. koreanus was collected from a private hatchery (Marua Suisan Co., Ltd., Ehime, Japan). This hatchery strain contained three haplotypes of B. koreanus (accession numbers LC004288 to LC004290) (Moka et al., 2016). Rotifers were stored in a 10 L container, and starved for one day before DNA extraction. Total DNA was extracted from 50 μg of well-washed rotifers using an Isotissue DNA extraction kit (Nippon Gene, Toyama, Japan). Di-nucleotide repeats (GT)n were isolated using dual-suppression polymerase chain reaction (PCR) (Lian et al., 2001). In brief, the DNA was separately digested with AluI, EcoRV, HaeIII, SspI, HinCII, and AfaI blunt-end restriction enzymes. The DNA fragments were then ligated with a blunt adaptor using a DNA Ligation Kit (Takara Bio, Shiga, Japan). As the first step, fragments flanked by a microsatellite at one end were amplified by the (GT)10 primers and the adaptor primer designed from the longer strand of the adaptor. The amplified fragments were then cloned and sequenced. Next, two primers were prepared: primer IP1, designed from the sequenced region flanking the microsatellite, and, for nested PCR, primer IP2, based on the sequence between IP1 and the microsatellite. The primary PCR reaction was conducted with each constructed DNA library using IP1 and AP1 primers. The secondary PCR reaction was conducted using a 100-fold dilution of the primary PCR products using IP2 and AP2 primers, on the basis that nested PCR dramatically improves the success rate of amplifying the microsatellite flanking regions. The single-banded fragments were purified and then directly sequenced. All primer pairs for amplifying the microsatellites were designed by Primer3 (Rozen and Skaletsky, 2000). In addition to B. koreanus, we also used another S-type rotifer strain belonging to the SM3 clade (B. sp. SM3) (Mills et al., 2016). This rotifer strain was previously genotyped (accession numbers LC004291 and LC004292) (Moka et al., 2016). The procedures to develop microsatellite DNA markers for B. sp. SM3 were the same as the method described above. All microsatellite sequences developed were deposited in GenBank (LC125210 – LC125217). We performed PCR in 5 μL reaction mixture containing 1 μL of extracted template DNA, 2 × KOD Fx buffer, 0.2 mM dNTPs, 0.1 U of KOD Fx polymerase (Toyobo, Osaka, Japan) and 0.1 μM each of the designed primers. The PCR reaction was performed using the following cycling conditions: denaturation at 94 °C for 2 min, followed by 40 cycles of denaturation at 98 °C for 10 s, annealing at 50 °C for 30 s, and extension at 68 °C for 30 s. The PCR products of non-labeled primers were separated using Fragment Analyzer (Advanced Analytical, Iowa, USA) at 8 kV for 70 min using a DNF-900 dsDNA Reagent Kit, and the band sizes calculated using PRO Size™ software (Advanced Analytical). Five out of ten primer pairs isolated from the B. koreanus genome DNA were successfully amplified. Three out of eight primer pairs isolated from B. sp. SM3 were successfully cross-amplified with B. koreanus. These forward and reverse primers of microsatellite DNA markers were then labeled with fluorescent dyes and tailed (Life Technologies). These eight primers were amplified using 46 individuals of the S-type rotifer population mass cultured at the hatchery, and the minimum combination of markers for identifying multi-locus genotypes (MLGs) was determined using GenClone 2.0 software (Arnaud-Haond and Belkhir, 2007). The PCR protocol using fluorescent-labeled primers was the same as that used for non-labeled primers, as described as above. The PCR products with fluorescent dyes were electrophoresed and quantified using an

298

E. Sawayama et al. / Aquaculture 465 (2016) 296–302

Culturing batch I

II III IV

V VI VII VIII IX

X XI XII XIII XIV XV

Population growth rate

4.0

3.0

2.0

1.0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 Culturing period (day)

could be represented by K separate genetic clusters for K = 1 to 20. The ΔK-value (Evanno et al., 2005), based on the rate of change in the log probability of data between successive K, was then estimated using STRUCTURE HARVESTER (Earl and von Holdt, 2012). Rotifer individuals sharing N 50% of each membership coefficient were assigned to the same cluster as msMLL. We also constructed MLLs by combining information on both maternal DNA (mtCOI haplotypes as mtMLL) and nuclear DNA (msMLLs), which we defined as ms + mtMLLs. Pearson's correlation coefficients and its significances between population growth rate and haplotype, msMLL, and ms + mtMLL diversities were calculated using PAST version 3.04 (Hammer et al., 2001). Haplotype, msMLL, and ms + mtMLL diversities were calculated following Nei's genetic diversity model (Nei, 1987). Frequencies of haplotype, msMLL, and ms + mtMLL in each batch were calculated using GenAlEx (Peakall and Smouse, 2006).

3. Results Fig. 1. Growth rate of rotifers used in this study during culturing (30 days). Gray bars indicate culturing batches with bad growth performances (growth rate around 2).

ABI 310 genetic analyzer and a PeakScanner ver. 1.0 (Life Technologies), respectively. The number of alleles per microsatellite DNA marker, allele frequencies, as well as observed and expected heterozygosity of 46 individuals of the S-type rotifers were also calculated using GenAlEx (Peakall and Smouse, 2006). 2.3. Clonal identification, clonal diversity and association with population growth We selected four newly developed microsatellite DNA markers (WamS kor1, WamS kor6, WamS SM3-3, and WamS SM3-5) and used them for genotyping. Twenty-four rotifer individuals per batch were used for genotyping. The large number of MLGs found caused analytical difficulty, thus we decided to use multi-locus lineages (MLLs) as units of clones and for further clonal analysis of rotifers in this study. To determine the microsatellite DNA based-MLLs (msMLLs), we estimated population genetic structure using the Bayesian clustering procedure implemented in STRUCTURE (Pritchard et al., 2000) using the admixture method and correlated the allele frequency version of the program. We conducted 20 runs for each different value of K using 100,000 iterations following a burn-in period of 10,000, assuming that the data set

3.1. Population growth of mass-cultured S-type rotifer Population growth rates during the culturing batch are shown in Fig. 1. Population growth rates ranged from 1.9 (batch VIII) to 4.0 (batch XV), and the average population growth rate was 2.9. Low population growth of around 2 was observed in batches II, V, and VII to IX.

3.2. Microsatellite isolation and marker selection for clonal study of S-type rotifer Microsatellite markers successfully amplified in B. koreanus are shown in Table 1. The eight microsatellite markers amplified with B. koreanus were tested to determine whether they were polymorphic, or not, using eight individuals of B. koreanus. Two markers (WamS kor6 and WamS SM3-7) were monomorphic and one marker (WamS kor2), displaying tri-allelic patterns in some rotifer individuals were difficult to score. The remaining five markers (WamS kor1, WamS kor5, WamS kor6, WamS SM3-3, and WamS SM3-5) were polymorphic and used for amplification using 46 individuals of the S-type rotifer population cultured at the hatchery, and a combination of four microsatellite markers (WamS kor1, WamS kor5, WamS SM3-3, and WamS SM3-5) was used to successfully identify MLGs using GenClone 2.0 software (Arnaud-Haond and Belkhir, 2007).

Table 1 Characterization of microsatellite loci developed for B. koreanus. Locus ID

Primer sequence (5′-3′)

Origin

Repeat

Size (bp)

Na

Ho

He

Accession no.

WamS kor1

F: FAM–CAAATCTTAAGCATTGTGAG R: Tail–AGATTCACGATCCAGTTAAA F: VIC–AATTCATTCAGCAAAAAGAC R: Tail–CATCATGCAACTCAAATAGA F: VIC–ACAGTCGGAAAATCTCTTAG R: Tail–ATTTTCAATGTGCGATTTAT F:NED–AGAGTCTGTATATGTGCTAGTG R: Tail–AGGCGTCAGAGATTTGTTC F: FAM–ACTTCCTGCTTTTCTCATTA R: Tail–TGAGTGCATTAAAATCAAAA F: VIC-GCGTTTTGGATATATAGTTT R: Tail-ATGAAAATAAACCAACTCAA F: NED-TTCATTAGCAAGTTTCTAGC R: Tail-CTCGTGAAATGAATGTTAAG F: PET-TAGTTGCAGATCTAAAGGAA R: Tail-CTCATACACAAACACTACCA

B. koreanus

(GT)9

191–209

5

0.917

0.657

LC125210

WamS kor2 WamS kor3 WamS kor5 WamS kor6 WamS SM3-3 WamS SM3-5 WamS SM3-7

B. koreanus

(GT)7

153–161

3

–

a

LC125211

b

LC125212

B. koreanus

(GT)11

178

1

–

B. koreanus

(GT)9

130–134

3

0.083

0,190

LC125213

B. koreanus

(GT)11

170–174

2

0.000

0.305

LC125214

B. sp. SM3 clade

(GT)20

123–129

3

0.386

0.167

LC125215

B. sp. SM3 clade

(GT)14

148–160

4

0.813

0.619

LC125216

B. sp. SM3 clade

(GT)14

109

1

–b

Number of samples genotyped: 46 individuals collected from a mass-production tank at a hatchery, Marua Suisan Co., Ltd. Na number of alleles, Ho observed heterozygosity, He expected heterozygosity. a Showing tri-allelic pattern. b No polymorphic alleles were observed.

LC125217

E. Sawayama et al. / Aquaculture 465 (2016) 296–302 Table 2 Average genetic parameters of four microsatellite loci in each culturing period. Culturing batch

N

Na

Ne

Ho

He

I II III IV V VI VII VIII IX X XI XII XIII XIV XV Overall

23 24 24 24 24 24 24 24 23 24 24 22 24 24 24 356

3.5 3.8 2.5 3.5 2.0 3.0 2.0 1.8 2.8 3.0 2.8 3.3 3.0 3.3 3.0 4.5

1.932 2.129 1.693 2.023 1.570 1.754 1.567 1.637 1.571 1.795 1.807 2.047 1.880 1.906 2.063 1.819

0.587 0.563 0.542 0.510 0.521 0.542 0.510 0.552 0.478 0.490 0.479 0.489 0.583 0.500 0.500 0.523

0.448 0.499 0.363 0.429 0.306 0.382 0.307 0.338 0.318 0.410 0.380 0.436 0.433 0.429 0.482 0.407

N, number of specimens; Na, number of alleles per locus, Ne, number of effective alleles, Ho, observed heterozygosity, He, expected heterozygosity.

3.3. Population genetic parameters of microsatellite DNA markers Population genetic parameters based on the microsatellite markers are shown in Table 2. Using both microsatellite and mtCOI markers, 356 individuals were successfully genotyped. Average numbers of alleles and effective alleles in the four microsatellite markers ranged from 1.8 (batch VIII) to 3.8 (batch II), and 1.6 (batch IX) to 2.1 (batch II), respectively. The average number of alleles and effective alleles overall in the culturing batch was 2.9 and 1.8, respectively. Observed and expected heterozygosity ranged from 0.478 (batch IX) to 0.587 (batch I), and 0.306 (batch V) to 0.499 (batch II), respectively. The averages of observed and expected heterozygosity were 0.523 and 0.397, respectively.

3.4. Clonal identification using molecular markers We identified 51 MLGs among 356 individuals using four microsatellite marker polymorphisms. One MLG was dominant with 198 individuals belonging to that particular MLG (data not shown). Twentynine MLGs were identified only once. As so many MLGs were observed, we applied multi-locus lineages in order to identify clones. Identification of the multi-locus lineages was based on recognizing K = 3 as the most likely number of genetically distinct groups of multilocus genotypes (Fig. 2A). A graph of membership coefficients constructed using STRUCTURE indicated three main clusters (Fig. 2B). Frequencies of clones constructed using two different molecular markers and these combinations are shown in Table 3. Three haplotypes

50

K=3

40 Delta K

(#1, #2 and #3) belonging to two lineages of B. koreanus were identified in 356 individuals as previously reported by Moka et al. (2016). Haplotype #1 was dominant and observed in 92.4% of the analyzed samples. Frequencies of haplotypes #2 and #3 in the analyzed samples were 6.5% and 1.1%, respectively. Of the three msMLLs observed in the analyzed samples: msMLL #A was dominant (73.9%); msMLL #B was the second major lineage (15.7%), and msMLL #C a minor lineage (10.4%). We combined three haplotypes with the msMLLs, resulting in six combinations (see Table 3). The clonal numbers for ms + mtMLLs are represented by the haplotype number added to the msMLL number (e.g. for haplotype #1 and msMLL #A, the resulting ms + mtMLL number is #1A). Haplotype #1 was shared with msMLL #A, B and C, and ms + mtMLL #1A was dominant (72.5%). Haplotype #2 was shared with msMLL #A and C. Haplotype #3 was shared with only msMLL #C. msMLL #B was shared with only haplotype #1. 3.5. Clonal diversity and relationships to population growth The frequency of haplotypes during culturing is shown in Fig. 3A. One to three haplotypes were observed in each batch, and the average number of haplotypes per batch was 1.9. Haplotype #1 was the most dominant during culturing, ranging from 80% to 100%. Haplotype #2 was second and ranged from 0% to 20%. Haplotype #3 ranged from 0% to 5%. A combination of three haplotypes (haplotypes #1, #2, and #3) was observed in culturing batches I and IV. A combination of haplotypes #1 and #2 was observed in culturing batches II, VI, IX, X, XI, XII, XIII, XIV, and XV. A single haplotype was observed in culturing batches V, VII, and VIII. The frequency of msMLLs during culturing is shown in Fig. 3B. Two to three msMLLs were observed in each batch, and the average number of msMLLs per batch was 2.8, with msMLL #A being dominant ranging from 58% to 96%. The frequency of ms + mtMLLs during culturing is shown in Fig. 3C. Two to five ms + mtMLLs were observed in each batch, and the average number of ms + mtMLLs per batch was 3.5. One MLL (ms + mtMLL #1A) was dominant and ranged from 58% to 92%. The minimum number of clones (n = 2) was observed in culturing batches V, VII and VIII, which displayed a low growth rate. The relationships between population growth and genetic diversities in each batch are shown in Fig. 4. As shown in Fig. 4A, haplotype diversity (h) ranged from 0.000 (batches V, VII, and VIII) to 0.351 (batch IV), and the average value of h through the culturing was 0.134. There was a significant correlation between population growth rate and haplotype diversity (r = 0.695, P = 0.004). The diversity of msMLL ranged from 0.083 (batch IX) to 0.569 (batch II), with an average value through the culturing phase of 0.393 (Fig. 4B). There was no correlation between population growth rate and msMLL diversity (r = 0.320, P = 0.245). The diversity of ms + mtMLL ranged from 0.153 (batch V) to 0.583 (batch II), with an average value through the culturing phase of 0.422

B Membership coefficient

A

299

30 20 10 0

1.00 0.80 0.60 0.40 0.20 0.00

1 4 7 10 13 16 19 K Fig. 2. Bayesian clustering of all rotifer individuals (n = 356) used in this study using four microsatellite markers. The ΔK values (K = 2 to 19) of each clustering are shown on the left (A). Clustering was obtained from STRUCTURE, for a model with admixture and correlated allele frequencies between samples; each individual is represented by a single vertical line broken into different shades of gray and sorted by membership coefficient, with lengths proportional to the estimated membership of the inferred cluster. Major clusters (above) and sub-clusters (below) are indicated (B).

300

E. Sawayama et al. / Aquaculture 465 (2016) 296–302

Table 3 Number of mtCOI haplotypes, msMLLs, and the combination (mt + msMLLs) observed in this study (n = 356). msMLL

#A #B #C Total (haplotype) 1

mtCOI haplotype

Total (msMLL)

#1

#2

#3

258 (72.5) 56 (15.7) 15 (4.2) 329 (92.4)

5 (1.4) 0 18 (5.1) 23 (6.5)

0 0 4 (1.1) 4 (1.1)

263 (73.9) 56 (15.7) 37 (10.4) 356 (100)

mt + msMLL #1A, 2mt + msMLL #1B, 3mt + msMLL #1C, 4mt + msMLL #1E, 5mt +

msMLL #2A, 6mt + msMLL #2B, 7mt + msMLL #2D, 8mt + msMLL #3C. Number in parentheses shows frequencies (%) of each clone observed in this study.

(Fig. 4C). There was no correlation between population growth rate and ms + mtMLL diversity (r = 0.435, P = 0.105). 4. Discussion Using newly developed microsatellite markers, large numbers of MLGs (51 MLGs in 356 rotifer individuals) were found in the masscultured tank based on our microsatellite marker analysis, in contrast to the low clonal composition of a hatchery strain of L-type rotifers

(Papakostas et al., 2009). Genotyping errors and somatic mutation are one cause of a high level of MLG construction (Meirmans and van Tienderen, 2004). Allele peaks of the four microsatellite markers were very clear and easy to score. We conducted PCR twice using the same 48 samples and no scoring errors were observed (data not shown). There is no data on how often somatic mutation occurs in B. koreanus, however the occurrence of somatic mutation cannot be ignored in mass culture because billions of rotifer individuals are cultured and reproduced every day. The rotifer strain monitored in this study was first introduced from a distributor and contained only one haplotype (Moka et al., 2016). Contamination, somatic mutation, and sexual reproduction possibly occurred during long-term culturing in a mass-culture tank at the hatchery, and the genetic diversity of the hatchery-cultured rotifers may be constructed. However, one haplotype, msMLL, and ms + mtMLL were each dominant during the one month of culturing. Rotifers mainly undergo parthenogenic reproduction, and their numbers dramatically increase clonally when the rearing environment is suitable for rotifers (Suzuki and Ohno, 1996). The rearing environment of the hatchery may be suitable for the dominant rotifers observed. In this study, the genetic monitoring of B. koreanus in the hatchery stain was conducted for only one month. Long-term genetic monitoring of

A Haplotype frequency

0.900 0.800 0.700 0.600 0.500 I

Growth rate

5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0

1.000

#3 #2 #1

II III IV V VI VII VIII IX X XI XII XIIIXIV XV

B 5.0

0.900

4.0

0.800 3.0 0.700 2.0

0.600 0.500

Growth rate

msMLL frequency

1.000

#C #B #A

1.0 I

II III IV V VI VII VIII IX X XI XII XIII XIV XV

5.0

0.900

4.0

0.800 3.0 0.700 2.0

0.600 0.500

Growth rate

mt+msMLL frequency

C 1.000

#3C #2D #2A #1C #1B #1A

1.0 I

II III IV V VI VII VIII IX X XI XII XIII XIV XV Culturing batch

Fig. 3. Frequency of clones in the sampling batches: (A) haplotypes of mtCOI (n = 3), (B) msMLL identified by four microsatellite genotypes (n = 3), (C) mt + msMLL identified by a combination of mtCOI haplotype and four microsatellite genotypes (n = 6). The left side of the Y-axis in each graph indicates frequency (%) of each clone. The right side of the Y-axis and polygonal line in each graph indicates growth rate. The X-axis indicates culturing batches during the month (n = 15).

E. Sawayama et al. / Aquaculture 465 (2016) 296–302

Haplotype diversity

0.600

A

0.500

r=0.695

0.400 0.300 0.200 0.100 0.000 1.0

2.0

3.0

4.0

5.0

msMLL diversity

0.600

B

0.500 0.400 0.300

r=0.320

0.200 0.100 0.000 1.0

2.0

3.0

4.0

5.0

mt+msMLL diversity

0.600

C

0.500 0.400 0.300 0.200

r=0.435

0.100 0.000 1.0

2.0

3.0

4.0

5.0

Growth rate Fig. 4. The relationship between population growth rate and clonal diversity: (A) haplotypes of mtCOI (n = 3), (B) msMLL identified by four microsatellite genotypes (n = 3), (C) mt + msMLL identified by a combination of mtCOI haplotypes and four microsatellite genotypes (n = 6). Correlation coefficient (r) is also shown on each graph.

hatchery-cultured rotifers will be needed in order to confirm whether the dominant types of rotifers have been replaced. Some ms + mtMLL identified in this study shared somatic origin DNA (microsatellite genotypes) and maternal origin DNA (COI haplotypes). If rotifers perform only parthenogenetic reproduction, these two types of DNA will not be shared in one MLL. This suggests that sexual reproduction has occurred in the S-type rotifers in the masscultured tank in the past. The probability of sexual reproduction, Psex(FIS) (Parks and Werth, 1993) was also calculated using GenClone 2.0 software: Psex(FIS) of the major MLG (n = 198) was not significant (P b 0.05), supporting sexual reproduction (data not shown). We also found a second peak of ΔK at K = 5 (Fig. 3A), and genetic distances of the additional two clusters were very close to cluster 3 (data not shown). These two clusters also showed admixture clustering (Fig. 3B) and possibly represent a cross between individuals across clusters. The rate of sexual reproduction and resting egg formation is different between rotifer strains, and the genetic distance between mated

301

pairs correlates to the success of resting formation (Fu et al., 1993). There is no data on the sexual reproduction of this lineage of S-type rotifer, but environmental factors may inhibit the triggers of sexual reproduction during long term intensive culturing at the hatchery. However, our results suggest that sexual reproduction may have occurred in the past. Further study will be required in order to identify the frequency of sexual reproduction in the rotifer strain and to determine the causes of sexual reproduction in a hatchery-culture of S-type rotifers. Population growth and mtCOI haplotype diversity displayed significant correlation (r = 0.695, P = 0.004). Two reasons that could explain this phenomenon are: 1) cryptic lineages grow well because the rearing environment is suitable for all the rotifers, and 2) rotifer numbers increase because genetic diversity is high. Contamination of S-type rotifers into an L-type rotifer culture reduces the number of L-type rotifers because of competition for habitat and food (Oka, 1991; Fernández-Araiza et al., 2005; Papakostas et al., 2007; Baer et al., 2008). If physiological characters differ between maternal lineages then some lineages grow more rapidly under specific environmental conditions. Rearing experiments in which several rotifer lineages based on mtCOI haplotype are mixed will be required to answer this question. Also, it is important to identify the best rearing conditions such as water temperature, salinity and food for each maternal rotifer lineage. However, no correlations were found between population growth and msMLLs and ms + mtMLLs. The genetic background of msMLLs is admixed with three maternal lineages, and physiological responses to culturing conditions may not differ among msMLLs. Therefore, growth may not correlate with each msMLL. Several studies have reported that Brachionus rotifers have higher growth rates in monocultures than in mixes (Hagiwara et al., 1995; Fernández-Araiza et al., 2005). Also, Papakostas et al. (2007) observed good growth performance of rotifers when S-type rotifers (B. sp. “Cayman”) were contaminated into a L-type rotifer (B. sp. “Austria”) tank and were replaced by the S-type rotifers. While these studies dealt with a combination of different morphotypes, there have not been any studies on a mixed culture of clones in the same morphotypes. Therefore, no information is available on whether a mixed culture or monoculture of S-type rotifer clones is better suited for stable culturing of rotifers. Tortajada et al. (2009) reported that inbreeding depression affects several fitness components of B. plicatilis, especially in the asexual phase. This suggests that inbreeding depression may also occur in B. koreanus in mass cultures at hatcheries and thus affect population growth. More experiments will be needed to clarify association with sexual reproduction, genetic diversity and population growth in B. koreanus. Microsatellite DNA markers have become a powerful tool to identify clonal structures in hatchery stock and natural populations of Brachionus rotifers (Gómez and Carvalho, 2000; Campillo et al., 2009; Papakostas et al., 2009). In our study, genetic structure based on microsatellites is more sensitive than that of mtCOI markers. In addition, MLGs and MLLs constructed with microsatellites alone do not reveal footprints of sexual reproduction, therefore using both mtCOI and microsatellite DNA markers is more efficient for undertaking a clonal study of the S-type rotifer B. koreanus. However, we only developed a small number of rotifer microsatellite DNA markers, moreover a fourmicrosatellite dataset is too low to extract confident solutions of population structure. Clonal analysis using a large number set of DNA markers should improve the accuracy of clonal identification and provide a deeper insight into the population genetics of B. koreanus in both hatcheries and the wild environment.

Acknowledgements This work was supported by a JSPS KAKENHI Grant-in-Aid for Encouragement of Scientists (Grant Number 26925023) to ES. We also thank Dennis Murphy for editing this manuscript.

302

E. Sawayama et al. / Aquaculture 465 (2016) 296–302

References Arnaud-Haond, S., Belkhir, K., 2007. GENECLONE: a computer program to analyse genotypic data, test for clonality and describe spatial clonal organization. Mol. Ecol. Notes 15–17. Baer, A., Langdon, C., Mills, S., Schulz, C., Hamre, K., 2008. Particle size preference, gut filling and evacuation rates of the rotifer Brachionus “Cayman” using polystyrene latex beads. Aquaculture 282, 75–82. Campillo, S., García-Roger, E.M., Carmona, M.J., Gómez, A., Serra, M., 2009. Selection of lifehistory traits and genetic population divergence in rotifers. J. Evol. Biol. 22, 2542–2553. De Araujo, A.B., Snell, T.W., Hagiwara, A., 2000. Effect of unionized ammonia, viscosity and protozoa contamination on the enzyme activity of the rotifer Brachionus plicatilis. Aquac. Res. 31, 359–365. Declerck, S.A.J., Malo, A.R., Diehl, S., Waasdorp, D., Lemmen, K., Proios, K., Papakostas, K., 2015. Rapid adaptation of herbivore consumers to nutrient limitation: ecoevolutionary feedbacks to population demography and resource control. Ecol. Lett. 18, 553–562. Earl, D.A., von Holdt, B.M., 2012. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 4, 359–361. Evanno, G., Regnaut, S., Goudet, J., 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol. Ecol. 14, 2611–2620. Fernández-Araiza, A.M., Sarma, S.S.S., Nandini, S., 2005. Combined effects of food concentration and temperature on competition among four species of Brachionus (Rotifera). Hydrobiologia 546, 519–534. Fu, Y., Hagiwara, A., Hirayama, K., 1993. Crossing between seven strains of the rotifer Brachionus plicatilis. Nippon Suisan Gakkaishi 59, 2009–2016. Gómez, A., Carvalho, G.R., 2000. Sex, parthenogenesis and genetic structure of rotifers: microsatellite analysis of contemporary and resting egg bank populations. Mol. Ecol. 9, 203–214. Guo, W., Gui, J.F., 2008. Microsatellite marker isolation and cultured strain identification in Carassius auratus gibelio. Aquac. Int. 16, 497–510. Hagiwara, A., Gallardo, W.G., Assavaaree, M., Kotani, T., de Araujo, A.B., 2001. Live food production in Japan: recent progress and future aspects. Aquaculture 200, 111–127. Hagiwara, A., Jung, M.M., Sato, T., Hirayama, K., 1995. Intraspecific relations between marine rotifer Brachionus rotundiformis and zooplankton species contaminating in the rotifer mass culture tank. Fish. Sci. 61, 623–627. Halkett, F., Simon, J.C., Calloux, F., 2005. Tackling the population genetics of clonal and partially clonal organisms. Trends Ecol. Evol. 20, 194–201. Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: paleontological statistics software package for education and data analysis. Palaeontol. Electron. 4, 9. Hirayama, K., 1987. A consideration why mass culture of the rotifer Brachionus plicatilis with baker's yeast is unstable. Hydrobiologia 147, 269–270. Hwang, D.S., Dahms, H.U., Park, H.G., Lee, J.S., 2013. A new intertidal Brachionus and intrageneric phylogenetic relationships among Brachionus as revealed by allometry and CO1-ITS1 gene analysis. Zool. Stud. 52, 13. Lian, C., Zhou, Z., Hougetsu, T., 2001. A simple method for developing microsatellite markers using amplified fragments of inter-simple sequence repeat (ISSR). J. Plant Res. 114, 381–385. Lubzens, E., Tandler, A., Minkoff, G., 1989. Rotifers as food in aquaculture. Hydrobiologia 186/187, 387–400.

Meirmans, P.G., van Tienderen, P.H., 2004. GENOTYPE and GENODIVE: two programs for the analysis of genetic diversity of asexual organisms. Mol. Ecol. Notes 4, 792–794. Mills, S., Alcántara-Rodríguez, J.A., Ciros-Pérez, J., Gómez, A., Hagiwara, A., Galindo, K.H., Jersabek, C.D., Malekzadeh-Viayeh, R., Leasi, F., Lee, J.S., Welch, D.B.M., Papakostas, S., Riss, S., Segers, H., Serra, M., Shiel, R., Smolak, R., Snell, T.W., Stelzer, C.P., Tang, C.Q., Wallace, R.L., Fontaneto, D., Walsh, E.J., 2016. Fifteen species in one: deciphering the Brachionus plicatilis species complex (Rotifera, Monogononta) through DNA taxonomy. Hydrobiologia ROTIFERA. XIV, pp. 1–20. Moka, W., Sawayama, E., Noguchi, D., Takagi, M., 2016. Genetic identification of S-type rotifer Brachionus plicatilis sp. complex based on mtDNA COI of hatchery strains used in Japan. Fish Genet. Breed. Sci. 45, 1–10. Moka, W., Sawayama, E., Takagi, M., 2015. Optimization of DNA Barcoding Method for Small-type Rotifer, Brachionus plicatilis sp. Fish Genet. Breed. Sci. 43, 69–74. Nei, 1987. Molecular Evolutionary Genetics. Columbia University Press, New York. Oka, A., 1991. Biological characteristics of Brachionus plicatilis-breeding environment. In: Fukusho, K., Hirayama, K. (Eds.), The First Live Feed-Brachionus plicatilis. Koseisha Koseikaku, Tokyo, pp. 28–38 (In Japanese). Papakostas, S., De Wolf, T., Triantafyllidis, A., Vasileiadou, K., Kanellis, D., Cecconi, P., Kappas, I., Abatzopoulos, T.J., 2007. Follow-up of hatchery rotifer cultures with regard to their genetic identity. J. Biol. Res. 7, 41–49. Papakostas, S., Dooms, S., Triantafyllidis, A., Delloof, D., Kappas, I., Dierckens, K., De Wolf, T., Bossier, P., Vadstein, O., Kui, S., Sorgeloos, P., Abatzopoulos, T.J., 2006. Evaluation of DNA methodologies in identifying Brachionus species used in European hatcheries. Aquaculture 255, 557–564. Papakostas, S., Trianafyllidis, A., Kappas, I., Abatzopoulos, T.J., 2009. Clonal composition of Brachionus plicatilis s.s. and B. sp. ‘Austria’ hatchery strains based on microsatellite data. Aquaculture 296, 15–20. Parks, J.C., Werth, C.R., 1993. A study of spatial features of clones in a population of bracken fern, Pteridium aquilinum (Dennstaedtiaceae). American Journal of Botany]–>Am. J. Bot. 5, 537–544. Peakall, R., Smouse, P.E., 2006. GENALEX 6: genetic analysis in excel. Population genetic software for teaching and research. Mol. Ecol. Notes 6, 288–295. Pritchard, J.K., Stephens, M., Donnelly, P., 2000. Inference of population structure using multilocus genotype data. Genetics 155, 945–959. Rozen, S., Skaletsky, H.J., 2000. Primer3 on the WWW for general users and for biologist programmers. In: Krawetz, S., Misener, S. (Eds.), Bioinformatics Methods and Protocols: Methods in Molecular Biology. Humana Press, Totowa, pp. 365–396. Scott, J.M., 1981. The vitamin B12 requirement of the marine rotifer Brachionus plicatilis. United Kingdom]–>J. Mar. Biol. Assoc. U. K. 61, 983–994. Suzuki, S., Ohno, A., 1996. Population dynamics of the S-type rotifer, Brachionus rotundiformis, under laboratory conditions. Aquat. Sci. 44, 45–52. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731–2739. Thompson, J.D., Higgins, D.J., Gibson, T.J., 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positionspecific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680. Tortajada, A.M., Carmona, M.J., Serra, M., 2009. Does haplodiploidy purge inbreeding depression in rotifer population? PLoS One 4, e8195.