Developing Microsatellite Multiplex PCR Panels for Topmouth Culter (Culter alburnus) and their Application in Parentage Assignment

Jianjun Fu1, Shili Liu2, 3, Wenbin Zhu1, Yongyi Jia2, 3, Lanmei Wang1, Yufen Yang4, Zhimin Gu2, 3* and Zaijie Dong1* 1Key Laboratory of Freshwater Fisheries and Germplasm Resources Utilization, Ministry of Agriculture and Rural Affairs, Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences, Wuxi 214081, Jiangsu, China 2Key Laboratory of Healthy Freshwater Aquaculture, Ministry of Agriculture and Rural Affairs, Zhejiang Institute of Freshwater Fisheries, Huzhou 212001, Zhejiang, China 3Key Laboratory of Freshwater Aquaculture Genetic and Breeding, Zhejiang Province, Zhejiang Institute of Freshwater Fisheries, Huzhou 212001, Zhejiang, China 4Wuxi Fisheries College, Nanjing Agricultural University, Wuxi 214081, Jiangsu, China


INTRODUCTION
T he topmouth culter (Culter alburnus) is a carnivorous pelagic fish belonging to the Cyprinidae family and widely distributed throughout large water bodies in China (Chen, 1998). It commonly feeds on small fish species, which plays an important role in the ecological equilibrium of water ecosystems. Its fast growth and distinctive flavor render it a major economic freshwater fish species in Eastern China, especially in the Lake Taihu area. Fish culture production of this species has expanded O n l i n e F i r s t A r t i c l e maintain culture production. However, as phenotypic selection usually targets a few outperforming individuals without knowledge of their pedigree, this can potentially lead to inbreeding and fitness depression (Fu et al., 2013), and subsequently limited genetic improvement. Therefore, accurate pedigree information is imperative in a successful breeding program (Lacy, 2012). Recently, microsatellite markers have shown promising practicability in parentage assignment with few loci and high accuracy (Luo et al., 2017;Sudo et al., 2018;Wang et al., 2018). Moreover, multiplex PCR methods are becoming more utilized for microsatellite loci co-amplification, which has considerably reduced time and costs in parentage analysis procedures. Nowadays, multiplex PCR panels have been successfully developed and utilized for parentage assignment in many aquaculture animals (Fu et al., 2013;Nie et al., 2012;Popa et al., 2015), and considered an effective marker-assisted selection breeding technique. Despite its ecological and economic importance, studies on topmouth culter genetics remain limited (Guo et al., 2018;Liu et al., 2014;Wang et al., 2007) and selection breeding programs non-existent. Therefore, in this study, we developed three multiplex PCR panels involving twelve microsatellite loci with redesigned primers based on released sequences on GenBank database, and conducted parentage analysis using parent and offspring samples from four independent artificial breeding pilots. This approach not only proved effective in monitoring pedigree information, but also demonstrated great potential as an essential tool when developing breeding programs for C. alburnus.

Fish samples and DNA extraction
Culturing and spawning of C. alburnus were carried out at Aquatic Breeding Farm of Zhejiang Institute of Freshwater Fisheries (Zhejiang, China). Thirty-six adult fish (24 females and 12 males) were used for artificial breeding from June 10 to June 15, 2017. Mating designs for four breeding pilots are shown in Table I. Pilot A and Pilot C consisted of eight females and four males, Pilot B and Pilot D consisted of four females and four males, and Pilot A and Pilot B shared four males in practice. Broodfish of four pilots were processed under hormone stimulation and artificial stripping. Milt samples were stripped in advance and stored in an icebox (about 4ºC) for a short time, and fertilized with the eggs stripped out. The fertilized eggs of four pilots were hatched in independent cylinders, and the fry cultured in four different ponds respectively. Thirtyfour offspring were randomly sampled from each pilot after one-month nursery rearing.
Fin clips of broodstock and offspring were collected and stored in absolute ethanol until DNA extraction. Genomic DNA was extracted using a traditional phenolchloroform method. DNA concentration and purity of samples were assessed using the Nano Drop 2000C spectrophotometer (Thermo scientific, US) and 1% agarose gel. Afterward, DNA concentration was adjusted to 20-50 ng/μL and stored at -20ºC for later PCR amplification.

Microsatellite primer design and multiplex PCRs
The original sequences released on GenBank database (Liu et al., 2014) were chosen for designing primers. Novel primers were designed for microsatellite loci of differing product sizes. Amplification efficiency and polymorphism was tested using six parent samples. Primers with different product sizes were used in developing the multiplex sets. In this study, twelve primer pairs were successfully used in the multiplex PCR panels (Table II). Primers were synthesized by Sangon Biotech (Shanghai) Co., Ltd., and each forward primer was labeled with a fluorescent dye (HEX or FAM).

Data analysis
Estimates of genetic variation such as the number of alleles (Na), observed (Ho) and expected heterozygosity O n l i n e (He), polymorphic information content (PIC), exclusion probability for parent pairs (E-PP) and null alleles frequency (Fn) of twelve microsatellite loci were estimated based on the genotype data of 36 candidate parents via allele frequency analysis using Cervus 3.0 software (Marshall et al., 1998;Kalinowski et al., 2007). Correlation analysis for PIC and E-PP of loci were carried out using SAS software (SAS Institute, 1996). Simulation analysis were performed to evaluate the probable performance of the microsatellite loci for parentage identification on C. alburnus breeding programs. Simulation analyses based on genotypes of 36 candidate parents were conducted using the likelihoodbased approach in Cervus 3.0 software (Marshall et al., 1998;Kalinowski et al., 2007) and performed using the following parameters: 10,000 offspring and a pool of candidate parents, varying from 10 to 400 parents and 5 to 200 parent pairs; 98% of candidate parents were sampled and loci typed, 1% error rate in likelihood calculations. Real parentage analysis was performed using Cervus Developing Microsatellite Multiplex PCR Panels

O n l i n e F i r s t A r t i c l e
3.0 software (Marshall et al., 1998;Kalinowski et al., 2007), based on genotype data of parents and offspring from four breeding pilots. Furthermore, both simulation and parentage analysis were carried out based on tested parents (the same candidate parent information was used in the simulation) in different multiplex panel(s), under the methods mentioned above. Pedigree reconstruction was compared to known parental information from the mating design and sampling, and used to examine the accuracy of parentage assignment. Chi-square test was then applied to detect the different progeny contributions of parents and parent pairs using SAS software (SAS Institute, 1996).

RESULTS AND DISCUSSION
In this study, we developed three multiplex PCR panels with four microsatellite loci in each panel. Information and genotyping results of four multiplex PCR panels are shown in Table II and Figure 1. We used different product size designs and fluorescence dyes (FAM or HEX) that avoided alleles overlapping and optimized the genotyping process, respectively. This was highly effective and efficient in both time and cost compared with scoring each locus with independent PCR protocols. We evaluated the genetic diversity parameters of twelve microsatellite loci using 36 candidate parents. The summary statistics are shown in Table II. We revealed that Na, Ho and He ranged from 7 to 22, 0.39 to 0.75, and 0.57 to 0.91, respectively. Genetic diversity parameters of microsatellite loci in C. alburnus were slightly higher in our study than previously reported (Liu et al., 2014), but lower than those reported in another study (Fu et al., 2013). It should be noted that the individuals tested and the detection methods were related to the diversity parameters (Liu et al., 2009). All twelve microsatellite loci were highly informative (PIC > 0.5) (Botstein et al., 1980), and exclusion probabilities for the parent pairs (E-PP) ranged from 0.57 to 0.92, which significantly correlated with the PIC of microsatellite loci (P<0.01; Fig. 2). These correlations are also reported in another study (Ma et al., 2013). Microsatellites with high genetic diversity parameters and hence relatively high statistical powers are therefore included in parentage assignments (Marshall et al., 1998;Tokarska et al., 2009). Simulation assignment approaches estimate the theoretical power of microsatellite markers and allow the adjustment of marker-selection toward the best benefit direction (Hauser et al., 2011). In this study, we conducted simulation analyses (with sexes known or unknown, and varying number of parents) to assess the resolution capability of these twelve loci. According to the simulation results shown in Figure 3, all assignments for mother alone, father alone, and parent pairs were higher than 95% when the number of candidate parents was less than 200 parents (sexes unknown) or 100 parent pairs (100 father and 100 mother, sexes known). This is far greater than the minimum number of parent pairs recommended to prevent inbreeding and obtain a long-term response in a mass selection program (Bentsen and Olesen, 2002). The real parentage analysis for 36 candidate parents and 136 offspring are shown in Table III. The assignment rates for O n l i n e

F i r s t A r t i c l e
Developing Microsatellite Multiplex PCR Panels 5 breeding pilots ranged from 97.06% to 100% (average was 99.26 %). We checked pedigree reconstruction with known mating and sampling information, which we used to assess the accuracy of parentage assignment (Jerry et al., 2006). All pilot assignments agreed with the true mating designs with 100% accuracy.  As shown in Figure 4, simulation assignment rates and parentage assignment rates, based on different multiplex PCR panel or panels, followed a similar trend. In general, we detected a higher assignment rate when we used more panels. Although, the parentage assignment rates were based on real data, they were commonly lower than those detected in simulations, which may due to the null alleles observed at many loci (Carlsson, 2008). Even so, our results highlight the use of simulation analyses to evaluate assignment capacity of multiplex PCR panels (or panel combinations) before breeding practice. We showed that the assignment rate of parentage analysis was higher than 80% when two panels were used (Panels 1 and 3, or Panels 2 and 3), and higher than 95 % when three panels were used. Consequently, using three multiplex PCR panels in this study is considered sufficient and efficient for pedigree reconstruction for C. alburnus. Fig. 4. Assignment rates of simulation and parentage analyses based on 36 candidate parents (24 females and twelve males), using genotype information from different multiplex panels and panel combinations (confidence in 95%).
In aquaculture breeding schemes, equalizing family size is challenging because of the unequal reproductive success of brooders during the breeding practice and differential survival among families (Kong et al., 2015). We detected nine full-sib families (from six dams and three sires) in Pilot A. In Fig. 5, we show different progeny contributions (number and percentage) for the parents and parent pairs in Pilot A. We detected significant unequal progeny contribution of parents and parent pairs in total for each breeding pilots (P<0.01). This phenomenon, which has also been observed in other aquatic species (Fu et al., 2016;Herlin et al., 2008;Loughnan et al., 2013;Rhody et al., 2014;Sudo et al., 2018), may result in higher rates of inbreeding in long-term selection due to a gradual decline of effective population size. Therefore, using molecular markers to monitor pedigree information could be used as an essential tool incorporated into breeding schemes of fish species, including C. alburnus.

CONCLUSIONS
In this study, we developed a multiplex PCR panels for C. alburnus that provides a powerful tool for parentage assignment and population genetic studies of this species. It can be used as an essential tool in developing breeding programs for C. alburnus.

O n l i n e F i r s t A r t i c l e
Developing Microsatellite Multiplex PCR Panels