Common carp (Cyprinus carpioL.) is an important economic fish species with very rich resources,which have been used to develop many new varieties,such as Jian carp (Cyprinus carpiovar. Jian) ,the first artificially bred aquatic species in China. Its selfreproduction,stable genetic traits,good meat quality and strong disease resistance make it commercially important in aquaculture. It is now cultured all over China (Zhang and Sun, 2006; Zhang and Sun, 2007) . Despite its economic importance,relatively little progress has been made in genetic and genomic research. Only a few genome resource studies have been performed on the development of SNPs in genes (Tao et al., 2011; Li et al., 2012) , and no genetic linkage maps have been reported for Jian carp.

Genetic linkage maps are essential for mapping quantitative trait loci (QTLs) to complete markerassisted selection (MAS) ,which represents an excellent way to improve the efficiency of breeding. They also represent tools for underst and ing genome organization and evolution through comparative mapping. Different markers such as r and om amplified polymorphic DNA (RAPD) ,amplified fragment length polymorphism (AFLP) ,microsatellites, and single nucleotide polymorphisms (SNPs) can be used in linkage map construction. Microsatellites are among the most popular markers,which can not only carry out MAS but can also be used for transferringgenetic information across species,in addition to their co-dominant inheritance,high-throughput genotyping,abundance and even genomic distribution. They also require no sophisticated technology and are low cost compared with SNP markers,making them suitable for the construction of junior maps. A large number of microsatellites have been developed in common carp,from anonymous genomic regions using microsatellite-enriched libraries (Wei et al., 2001; Sun et al., 2005; Hou et al., 2008; Lu et al., 2009) , and mining from expressed sequence tag (EST) databases (Lu et al., 2007) or from bacterial artificial chromosome (BAC) end sequences (Xu et al., 2011) ,which should allow significant advances in common carp genetic research.

Recently genetic mapping has progressed rapidly since the use of microsatellites and SNPs. Several linkage maps were constructed using different mapping populations of common carp by microsatellite and SNP markers (Zheng et al., 2011; Jin et al., 2012; Xu et al., 2013) . Combining these maps,two consensus linkage maps were created using four F₁populations comprising 257 microsatellites and 421 SNPs, and using two F₁populations and 627 microsatellites and 105 SNPs,respectively (Zhang et al., 2013; Zheng et al., 2012) . In addition,comparative genome analyses have been performed for this species. Comparison of the common carp map and five model teleost fish genomes (Zheng et al., 2011) revealed extensive synteny among them. Synteny was also observed by comparative mapping between two varieties of common carp (Zhang et al., 2013) . The comparative studies help to transfer genome information from better-studied species to lessstudied species.

Target QTLs influencing important economic traits have been mapped including growth-related traits (Zhang et al., 2011; Zheng et al., 2012) ,muscle fiberrelated traits (Zhang et al., 2011) ,physiological traits (Xu et al., 2013) and shape-related traits (Zhang et al., 2013) . Although QTLs influencing the same traits have been detected in different populations,it is difficult to make a detailed comparison of these QTLs because of lack of sufficient common markers among these maps. Thus,it is unclear whether the QTLs detected in one panel are applicable to MAS in other populations. However,most of these QTLs were verified in mirror carp populations. Significant differences in body shape,growth speed and scale have been found between Jian carp and mirror carp. Jian carp has a long shape with a body length to body height ratio 1.4 times that of mirror carp. The growth rate of Jian carp is higher than other hybrid carp by 30%,on average (Sun et al., 1995) . To date,no QTL mapping has been reported in Jian carp. Body weight (BW) is a comprehensive trait,which directly affects fish production. Here,we present the first genetic linkage map of Jian carp and QTL detection for BW trait,which will be useful for MAS breeding plans in Jian carp. Furthermore,we compared our results with previous findings in mirror carp (Xu et al., 2013) ,which revealed syntenic relationships between the Jian carp and mirror carp maps. QTLs were detected in the syntenic segments of homologous linkage groups. This research might increase efficiency in genetic research as well as fish breeding of Jian carp. 2 MATERIAL AND METHOD 2.1 Mapping materials

Ten pairs of brooders were selected for artificial insemination to produce 10 full-sib families according to pairwise genetic distances. Two hundred fry from each family were taken to polyculture in the large tank of the Freshwater Fisheries Research Center. After rearing for 8 months,a total of 647 progeny survived and were collected. Four growth related traits including body weight (BW) ,body length (BL) ,body height (BH) , and body thickness (BT) were measured and blood samples of parents and their progeny were immediately collected and frozen for further analysis. Individuals from the same full-sib families were identified through paternity analysis,as previously described (Gu et al., 2012) . Ninety-four r and om progeny from one full-sib family were analyzed for genetic linkage mapping. 2.2 DNA extraction and microsatellite genotyping

Genomic DNA was extracted individually from 20 μL of blood following the st and ard phenolchloroform method (Sambrook and Russell, 2001) . The DNA was adjusted to 50 ng/μL and arrayed into 96-well PCR plates for later work.

A total of 772 common carp simple sequence repeat (SSR) markers,chosen from two previously generated linkage maps of mirror carp,were detected in our mapping population. Among them,505 with the ‘HLJ’ prefix were developed from genomic libraries enriched for SSRs using a combination of biotin capture method and radioactive labeling hybridization in our laboratory (Wei et al., 2001; Lu et al., 2009) . Two hundred and thirty-five microsatellites with the ‘CA’ prefix were derived from the Chinese Academy of Fisheries Sciences database and 32 expressed sequence tag (EST) -derived microsatellites named with the ‘HLJE’ prefix were identified from the common carp EST database in GenBank.

PCR reactions were prepared in a 15 μL volume containing 1 PCR buffer (10 mmol/L Tris-HCl (pH 8.3) ,50 mmol/L KCl,2.0 mmol/L MgCl²,0.01% gelatin,0.1% Triton X-100,0.1% NP-40 and 2 mmol/L of each dNTP) ,1U of TaqDNA polymerase (TaKaRa,Dalian) ,5 μmol/L of forward and reverse primer and 50 ng genomic DNA. Amplifications were carried out in a Perkin-Elmer 9700 DNA Thermal Cycler (Perkin-Elmer Corporation,Norwalk,Conn.) using the following thermal cycling profile: initial denaturation at 94℃ for 3 min,followed by 25 cycles at 94℃ for 30 s,annealing at locus-specific temperatures for 30 s,at 72℃ for 30 s; and a final extension step at 72℃ for 5 min. Genotyping was carried out on 8% polyacrylamide gels using DL2000 (TaKaRa) as a reference marker to identify alleles. Gel-Pro analyzer software 4.5 was used to analyze the genotypes and fragment sizes. 2.3 Linkage mapping and QTL analysis

A linkage map was constructed using JoinMap 4.0 software. The three expected segregation types that can be used here for genotyping were calculated as follows: 1:1 (type ab×aa or aa×ab) ,1:2:1 (type ab×ab) and 1:1:1:1 (type ab×ac or ab×cd) . Segregation ratios were analyzed for all the loci by a Chi-square test and the markers that deviated from the expected Mendelian segregation patterns (P<005) were excluded from the linkage analyses. Linkage analysis was performed with a maximum recombination threshold of 0.4 and a maximum distance between adjacent markers of 50 cM. The default value of the logarithm of odds (LOD) score was from 2.0 to 10.0,with steps oF₁.0 and finally a maximum LOD score of 4.0 was selected to group markers by approaching the chromosome number of common carp. The recombination rate was converted to map distance using the Kosambi map function. Genome size of common carp was estimated using the methods described by Zheng et al. (2011) .

Interval mapping (IM) ,based on the linkage maps using the MapQTL5.0 program,was used to test for QTLs. Threshold of LOD scores were calculated using 1 000 permutation tests in MapQTL 5.0,with a linkage group-wide significance level of P<05 for significant linkages (Churchill and Doerge, 1994; Doerge and Churchill, 1996) . 2.4 Mapping and QTL data for mirror carp used for comparing analysis

Xu et al. (2013) described the genetic map of mirror carp,which was used for comparative analysis. An F₁population,comprising 190 individuals,was mapped using 992 microsatellite markers. The map spanned 5 183.9 cM in 51 linkage groups,which were named C1-C51,with an average marker spacing of 5.226 cM. The QTLs for body weight (BW) based on this mirror carp map will be published in another paper. Seventeen QTLs for BW were detected with LOD values ranging from 2.0 to 4.06, and explained from 4.8% to 25.7% of the phenotypic variation. We compared the two maps of Jian carp and mirror carp using 198 common SSR markers. The conservation of SSR markers between the two maps made it possible to compare the location of QTLs for the BW trait. 3 RESULT 3.1 Phenotype analysis

The distribution of BW trait in this F₁ mapping family was continuous and normal with P=0.132 (>0.05) based on the Shapiro-Wilk test. BW values ranged from 469.7 to 792.3 g with an average of 624.738±71.864 g. There was significant variation among the 94 individuals,with a coefficient of variation (CV) of 0.115. 3.2 Marker polymorphism and linkage analysis

The 772 SSR markers were tested on the parents and a small set of the progeny of this mapping family, and 323 (41.84%) markers showed polymorphisms,which were used for genotyping the F₁ population. Ultimately,265 markers were used for mapping after eliminating those markers that deviated from expected Mendelian segregation and some null alleles.

Of the 265 SSR markers used in the linkage analysis,254 (164 with the ‘HLJ’ prefix,80 with the ‘CA’ prefix and 10 with the ‘HLJE’ prefix) were ordered in 44 linkage groups,which were named LG1 to LG44,including nine LGs with three markers and three LGs with only two markers. Eleven markers were not linked to any groups. The total map,including the 44 LGs,spanned 1 381.592 cM,with an average marker distance of 6.58 cM,ranging in length from 0 to 107.224 cM. The number of markers in these LGs varied from 2 to 12,with an average of 5.77 markers per LG. There were eight gaps larger than 30 cM (Fig. 1) .

Fig. 1 Genetic linkage map of Cyprinus carpiovar. Jian based on microsatellite markers The marker name and map distances in centiMorgans (cM) are indicated on the right and left side of the linkage groups (LGs) ,respectively. Quantitative trait loci (QTLs) for body weight (BW) ,detected at the linkage group level,are indicated by bars on the right side of the LGs.

Figure options

Scanning each linkage group with phenotypic data in MapQTL5 was used to perform QTL analysis. The LOD thresholds varied from 2.0 to 2.6,which were obtained by 1 000 permutation tests on the linkagegroup level of P<05. For Jian carp,eight QTLs for BW were mapped on seven linkage groups (LG10,LG20,LG24,LG25,LG32,LG40, and LG41) ,of which two QTLs were mapped on LG20 (Fig. 1) . These QTLs explained from 12.6% to 17.3% of the phenotypic variation with LOD scores of 2.1–3.07. Of these QTLs,BW- 10- 1showed the largest LOD value of 3.07 and accounted for 14.4% of the variance,with HLJ3311 as the peak marker. QTL BW- 40- 1showed the smallest LOD value of 2.17 and explained 12.6% of the variance,with the shortest distance (1.04 cM) between flanking markers. Six marker loci were associated with QTLs for BW in Jian carp (P<05) , and three of them (HLJ3311,HLJ2772, and HLJ2544) were as the peak marker. The other three loci (HLJ229,HLJ1132-2, and HLJ2315) are located only 1.04–3.37 cM from the peak of the QTL. Summary statistics for the identified QTLs are presented in Table 1.

Table 1 Quantitative trait loci (QTLs) results for body weight (BW) in Jian carp

Table options

Of the 254 marker loci mapped in Jian carp,198 were found in common with the mirror carp map,with 186 (93.94%) showing synteny. The 198 common markers were distributed across all 44 Jian carp LGs,which allowed the identification of homologous linkage groups between the two maps. Each of the 44 Jian carp linkage groups aligned to a single linkage group in mirror carp. All Jian carp LGs shared two or more markers with the homologous groups in mirror carp,except for LG25,which shared only one common marker with group C5 in mirror carp. The number of shared loci varied from 1 to 10 per homologous linkage group pairs,with an average of 4.2. A large block of synteny and collinearity was observed between the two varieties on this comparative map. Six pairs of homologous linkage groups,LG18/C39,LG20/C4,LG21/C19,LG22/C46,LG23/C40, and LG27/C24,shared three or more common markers arranged in the same order,while the other homologous groups shared three or more common markers with variations in the marker order. In some cases,the recombination rates of the common marker pairs were significantly different between the two populations,which were higher in the mirror carp relative to Jian carp. In addition,two or three markers mapping to a single location with zero recombination were observed in Jian carp in 11 positions; however,these markers did not show zero recombination in mirror carp; the distance between these markers varied from 1.3 to 41.065 cM in mirror carp (Fig. 2) .

Fig. 2 Synteny relationships between Jian carp and mirror carp map The Jian carp linkage groups (LGs) are on the left side and mirror carp linkage groups (C) are on the right side of the comparative map. Homologous markers joining Jian carp and mirror carp linkage groups are colored and connected by lines.

Figure options

Fig.2 Continued

Based on the homology of microsatellite markers between the two maps,QTLs for BW were compared between Jian carp and mirror carp. Of the eight QTLs for BW identified in Jian carp,three were homologous in mirror carp. The region of LG10 in Jian carp,where QTL BW- 10- 1was detected,appears to be homologous to a region of C30 where QTL BW'- 30- 1has also been detected. The region of LG24 in Jian carp,where QTL BW- 24- 1was detected,appears to be syntenic with C15 where QTL BW'- 15- 1was detected. LG40 shared homology to C9; two QTLs for BW were located in similar positions in the two linkage groups. QTL confidence intervals in the mirror carp were more precise than the homologous interval in Jian carp,which were usually contained in QTL intervals in Jian carp (Table 2,Fig. 3) .

Fig. 3 The proposed homologous relationships between quantitative trait loci (QTLs for body weight (BW) ) on the Jian carp and mirror carp maps

Figure options

Table 2 Comparative analysis of QTLs for body weight (BW) between homologous linkage groups in Jian carp and mirror carp

Table options

Microsatellites are one of the best choices of markers in map construction and comparisons. Here,we reported the first genetic map of Jian carp using microsatellites. Among the 772 microsatellite markers developed from common carp genome,previously mapped in mirror carp,323 (41.84%) showed polymorphisms in this Jian carp population. These markers had been scanned in the mirror carp map and showed clear b and s and polymorphisms; therefore,the level of polymorphisms in this mapping family is high. According to previous studies,the polymorphism rates of common carp SSRs are about 40% to 70% in different varieties of common carp genome (Zhang et al., 2007; Liu et al., 2009; Lu et al., 2009) and similar rates of polymorphism were observed in our study,indicating high conservation of intra-specific microsatellites. This high level of polymorphisms should facilitate the development of comparative mapping in common carp. 4.2 Genetic and comparative mapping

Jian carp and mirror carp represent two varieties of common carp with the same chromosome number of 50 (Mao et al., 2007) ; therefore,the linkage group number should correspond to the chromosome number. The genetic map presented here consists of 254 microsatellite markers distributed in 44 linkage groups,including three doublets and nine triplets,indicating that the map does not cover the entire genome: some chromosomes are not covered by SSR markers. There were eight gaps larger than 30 cM; however,these LGs with gaps should not be split into two groups because of the one-to-one aligned relationship to the mirror carp LGs. In addition,the lack of suitable markers may result in the presence of split linkage groups that belong to the same chromosome; thus,some groups might be linked to others if the marker densities in them were higher. Xu et al. (2013) found 51 linkage groups in the mirror carp map,which included at least one extra linkage group. Since the lower genome coverage of the Jian carp map is caused by the limited marker number,more markers are needed to position the unlinked markers and obtain full genome coverage. Fluorescent in situ hybridization (FISH) is required to anchor the markers to the physical map.

Intra-specific and inter-specific comparative genetic mapping have generally revealed extensive genome synteny chromosomal segments. Jian carp and mirror carp both belong to the same species of common carp,but the two varieties show significant differences in shape,growth and scales. In this study,we compared the genetic maps between the Jian carp and mirror carp using 198 common markers that showed a high degree of synteny (93.94%) between the two maps. And there are 12 non-synteny markers mapped to different LGs relative to the comparable position on the mirror carp map. All 44 linkage groups of the Jian carp showed a one-to-one alignment to the 44 linkage groups of the mirror carp map,sharing two or more common loci in most cases,except for LG25,indicating that the 44 linkage groups may cover the 44 chromosomes of Jian carp. For six pairs of homologous linkage groups,the common markers were arranged in the same order in the conserved collinearity blocks,while the common markers on other homologous groups showed variations in marker order. In addition,the recombination rates between the common markers were higher in mirror carp relative to Jian carp. Markers with zero recombination in Jian carp did not show zero recombination in mirror carp. These variations in marker order may have been caused by chromosome rearrangements (translocations,duplication,inversions or deletions) ,which are frequently found in animal and plant comparative mapping analysis (Kalo et al., 2004; Ellwood et al., 2008) . However,it is most likely that the differences in marker order and recombination are a result of the mapping population size. The population size in Jian carp is 94,which is much smaller than that in the mirror carp (190) ; therefore,the linkage relations were more precise for mirror carp than for Jian carp. This was consistent with previous reports (Zhang et al., 2013) . Hence,the Jian carp genetic maps available so far are partially uncovered and share relatively few bridge markers with the mirror carp map. More efficient markers are required for comparative mapping to gain new insights into the orthologous relationship between the two varieties. 4.3 Comparative QTL analysis

The alignment of the Jian carp and mirror carp maps made it possible to compare the location of QTLs controlling the same traits. We detected eight QTLs for BW in Jian carp at the significance of linkage-group level,three of which were mapped to the homologous regions between the Jian carp and mirror carp linkage groups. The homologous QTL confidence intervals identified in the mirror carp were smaller than those in the Jian carp,which were usually contained within the QTL interval in Jian carp. The large difference in population size and map density between Jian carp and mirror carp are the most likely reasons for this. Thus,further reduction in the size of confidence intervals will require the use of larger populations and higher density markers of genetic maps.

The markers tightly linked to these QTLs can be used in breeding programs for MAS. Here,we found six SSR loci that were tightly linked to these QTLs for BW in Jian carp (P<05) ,three of which were mapped to the homologous regions of mirror carp map. The orthologous markers tightly linked to these QTLs are potentially useful in MAS for Jian carp.

Although the present comparative analysis was of low resolution because of the incomplete genetic map and shared markers,it has clearly identified syntenic relationships between the two varieties. The identification of syntenic relationships allows the transfer of genetic information (markers,genes and QTLs) from one species to another. When the genome sequences of some model fish and common carp become available,the utility of intra- and inter-specific comparative mapping will aid the identification of c and idate genes associated with important traits. Further study,including increasing the marker density to conduct fine mapping, and additional QTL analysis with more mapping panels,will allow for a more detailed study to determine syntenic relationship and reduce the homologous QTL regions. The conserved genes and QTLs will facilitate the breeding of different varieties of common carp.