Yush Meng,Chenxing Zheng,Hui Li,Aixin Li,Hong Zhi,Qingmei Wng,Shozhen He,Ning Zho,Hun Zhng,Shopei Go,Qingchng Liu,*

a Key Laboratory of Sweetpotato Biology and Biotechnology,Ministry of Agriculture and Rural Affairs/Beijing Key Laboratory of Crop Genetic Improvement/Laboratory of Crop Heterosis& Utilization and Joint Laboratory for International Cooperation in Crop Molecular Breeding, Ministry of Education,College of Agronomy and Biotechnolog y,China Agricultural

University,Beijing 100193,China

b Crop Research Institute,Shandong Academy of Agricultural Sciences,Jinan 250100,Shandong,China

Keywords:

A B S T R A C T Simple sequence repeat(SSR)markers have been proved to be a very powerful tool for quantitative trait locus(QTL)mapping,marker-assisted selection and comparative genomics research in many crop species.However,a high-density SSR genetic linkage map is still lacking because there are only a few SSR markers available in sweet potato.In this study,a total of 2545 simple sequence repeat(SSR)primer pairs,including 1215 genomic SSR(gSSR)primer pairs and 1330 BES-SSR(bSSR)primer pairs designed from the genome sequence and BAC-end sequence of sweet potato,respectively,were screened with sweet potato cultivars Luoxushu 8 and Zhengshu 20 and their randomly sampled two F1 individuals and 571 of them generated polymorphic bands.The selected 571 polymorphic SSR primer pairs and 35 EST-based SSR(eSSR)primer pairs developed at our laboratory were used to genotype 240 F1 individuals derived from a cross between Luoxushu 8 and Zhengshu 20.A double pseudo-test-cross strategy was applied for linkage analysis.The Luoxushu 8 map included 90 linkage groups with 5057 SSR markers and covered 13,299.9 cM with a marker density of 2.6 cM,and the Zhengshu 20 map contained 90 linkage groups with 3009 SSR markers and covered 11,122.9 cM with a marker density of 3.7 cM.Fifteen homologous groups were identified in both parent maps.These are the first SSR linkage maps consisting of the complete 90 linkage groups and 15 homologous groups,which are consistent with the autohexaploid nature of sweetpotato,and are also the linkage maps with the highest SSR marker density reported to date.These results provide a basis for QTL mapping,marker-assisted breeding and comparative genomics research of sweet potato.

1.Introduction

Sweet potato,Ipomoea batatas(L.)Lam.,is an important food crop worldwide and is cultivated in more than 120 countries,in which China is the largest producer of this crop[1].The important value of this crop has been widely recognized due to its abundant nutritions[2,3].Sweet potato is a clonally propagated autohexaploid species(2n=6x=B1B1B2B2B2B2=90)[4,5].Its available whole genome sequences have been greatly limited because its highly heterozygous genome has proved very difficult to assemble[6].Because of its high heterozygosity and self-incompatibility,this crop poses numerous challenges for the genetic improvement by conventional breeding methods[2,7,8].Molecular approaches have great potential to improve sweet potato;however,a little effort has been devoted to their development and application for the improvement because of its genetic complexity[9].

A high-density genetic linkage map is a powerful tool for quantitative trait locus(QTL)mapping,marker-assisted selection and comparative genomics research.To date,several genetic linkage maps for sweet potato have been constructed using randomly amplified polymorphic DNA(RAPD)[10],amplified fragment length polymorphism(AFLP)[2,9,11],simple sequence repeat(SSR)[2,8,12],sequence-related amplified polymorphism(SRAP)[13,14],retrotrasposon-based insertion polymorphism(RBIP)[3,15]and single nucleotide polymorphism(SNP)[15–17].SNP markers are increasingly used for constructing genetic linkage maps in crop plants.However,SSR markers are broadly distributed in eukaryotic and prokaryotic genomes.In many crop species,such as rice[18],wheat[19],maize[20]and potato[21],SSR markers have been proved to be powerful tools for molecular breeding because of their abundance,high level of polymorphism,codominant Mendelian inheritance and excellent reproducibility.However,a high-density SSR genetic linkage map has not been constructed because there are only a few SSR markers available in sweet potato.To date,the best SSR genetic linkage map of sweet potato was constructed with a mapping population consisting of 300 F1individuals derived from a cross between‘Jizishu 1’and‘Longshu 9’,in which 484 and 573 SSR markers were clustered into 90 linkage groups covering 3974.24 cM and 5163.35 cM and only 8 and 9 homologous groups were identified for both parents,respectively[12].

The objectives of this study are(1)to select polymorphic SSR primer pairs from those designed from the genome sequence and BAC-end sequence of sweet potato,and(2)to develop a highdensity SSR genetic linkage map of sweet potato using the selected polymorphic SSR primer pairs and 240 F1individuals derived from a cross between‘Luoxushu 8’and‘Zhengshu 20’.This study will provide a basis for QTL mapping,marker-assisted selection and comparative genomics research in sweet potato.

2.Materials and methods

2.1.Plant materials

The mapping population consisted of 240 F1individuals derived from a cross between two sweet potato cultivars,Luoxushu 8 as female parent and Zhengshu 20 as male parent.Luoxushu 8 is a low yield and high starch content cultivar and is also resistant to Fusarium wilt.Zhengshu 20 has high yield and low starch content and susceptible to Fusarium wilt.These materials were collected and conserved at Crop Research Institute,Shandong Academy of Agricultural Sciences,Jinan,China.

2.2.DNA extraction

Genomic DNA was extracted from young fresh leaves of the two parents and F1individuals of the mapping population by the cetyltrimethylammonium bromide(CTAB)method[22].The DNA concentration was measured with an ultraviolet spectrophotometer(NanoDrop 2000,Thermo Fisher Scientific,USA)and its quality was determined by electrophoresis on a 1%(w/v)agarose gel.

2.3.SSR primer screening and evaluation

A total of 1215 genomic SSR(gSSR)primer pairs designed from whole genome sequencing data of sweet potato cv.Xushu 18 were kindly provided by Dr.Sachiko Isobe,the Kazusa DNA Research Institute,Japan(Table S1).Meanwhile,1330 BES-SSR(bSSR)primer pairs were designed from the BAC-end sequences of sweet potato line Xu 781 obtained by our laboratory(Table S2)[23].However,the polymorphism of these gSSR and bSSR primer pairs was not previously detected in sweet potato.

Totally,2545 SSR primer pairs(1215 gSSRs and 1330 bSSRs)were synthesized by BGI-Tech,Shenzhen,China.Their polymorphism was detected with Luoxushu 8 and Zhengshu 20 and their randomly sampled two F1individuals.The PCR amplification was performed in a total of 20 μL reaction mixture which consisted of 3 μL(50 ng μL-1)template DNA,2 μL 10×PCR buffer,0.8 μL(10 mmol L-1)dNTPs,1 μL(10 μmol L-1)of each SSR primer,0.2 μL(5 U μL-1)EasyTaq DNA polymerase(TransGen Biotech,Beijing,China)and 12 μL deionized distilled water with the following cycling profile:an initial denaturation at 95°C for 5 min,followed by 35 cycles(95°C for 1 min,55–59°C for 30 s,72°C for 1 min)and final extension at 72 °C for 10 min.The PCR products were separated on a 6%vertical denaturing polyacrylamide gel and visualized by silver staining for the detection of SSR markers.

One hundred of pairs were randomly sampled from the selected polymorphic SSR primer pairs and evaluated for their polymorphism with the 20 sweet potato cultivars with extensive genetic backgrounds,including Caixianshu,Dayeliu,Guangxishu,Guishu 9,Jinguashu,Juhuazhong,Lianchenghuangxinfanshu,Lizixiang,Longshu 14,Shangshu 19,Sushu 20,Wanshu 10,Xiangshu 98,Xu 781,Xushu 18,Xushu 36,Xuzishu 8,Yuzishu 3,Zhanshu 12,and Zixin.PCR amplifications were performed as mentioned above.Polymorphism information content(PIC),effective number of alleles(Ne*),Nei’s gene diversity(H*),and Shannon’s information index(I*)were calculated using PIC_Calc version 0.6[24]and POP-GENE version 1.32[25],respectively.

2.4.Genotyping and marker scoring

The polymorphic gSSR and bSSR primer pairs selected in this study and 35 EST-based SSR(eSSR)primer pairs previously developed by our laboratory(Table S3)were used to amplify Luoxushu 8 and Zhengshu 20 and their 240 F1individuals according to the above-mentioned PCR procedure.Polymorphic markers were visually scored as binary data with presence as‘‘1”and absence as‘‘0”.GeneRuler 100-bp DNA Ladder molecular weight marker was used for sizing DNA fragments.Polymorphic markers were divided into 3 categories:the maternal markers(only present in the parent Luoxushu 8),the paternal markers(only present in the parent Zhengshu 20),and the double-simplex markers(present in both parents).

Marker names were determined by three parts:the corresponding primer,the polymorphic band number and the letters s,d,t,and ds representing simplex,duplex,triplex,and double-simplex markers,respectively(e.g.gSSR0001-1s*).The suffixes*and**showed the segregation-distorted markers with significant difference at the 0.05 and 0.01 probability levels,respectively.

2.5.Segregation of SSR markers

Marker dosage was determined by the segregation ratio of markers(presence‘‘1”:absence‘‘0”)in the mapping population.All markers were analyzed for their goodness-of-fit to the expected segregation ratios using the χ2test(α=0.05).Based on Jones’s cytological hypotheses in sweet potato[26],markers were classified into four types based on their segregation ratios:(1)simplex or single dose markers present in one parent in a single copy with a segregation ratio of 1:1(presence:absence);(2)duplex or double dose markers present in one parent in two copies with a hexasomic(4:1),tetrasomic(5:1),or disomic or tetradisomic(3:1)ratio;(3)triplex or triple dose markers present in one parent in three copies with a hexasomic(19:1),tetradisomic(11:1)or disomic(7:1)ratio;and(4)double-simplex markers present in both parents in a single copy with a 3:1 segregation ratio[9,11].The markers that did not fit the expected ratio of Mendelian inheritance were defined as segregation distortion[2].

2.6.Linkage map construction

A double pseudo-testcross strategy was used to analyze the mapping population consisting of 240 F1individuals derived from a cross between Luoxushu 8 and Zhengshu 20[27].The linkage maps were constructed by JoinMap 4.0 software[28].A framework map of each parent was firstly constructed using simplex markers at LOD 5.0,and then,duplex,triplex,and double-simplex markers(including distorted markers)were inserted into the framework maps one by one to get the final genetic linkage maps[2].The linkage groups were considered as homologous if they had the same multiple-dose markers in each parental map,and those of both maps with the same double-simplex markers were determined as homologous groups between both parents.

Each linkage group name was composed of three parts:(1)the corresponding parent name(LUOXUSHU8 or ZHENGSHU20);(2)a number between 01 and 15 representing the homologous group that the linkage group belongs to for that specific parental map;(3)a number between 01 and 90 corresponding to the linkage group number.For example,LUOXUSHU8(01.06)would refer to the linkage group 06 that belongs to the homologous group 01 in the Luoxushu 8 map.

3.Results

3.1.Screening and evaluation of SSR primer pairs for polymorphism

A total of 2545 SSR primer pairs,including 1215 gSSRs and 1330 bSSRs,were used to amplify Luoxushu 8 and Zhengshu 20 and their randomly sampled two F1individuals.The results showed that 571 pairs of them,including 273 gSSRs and 298 bSSRs,generated clear polymorphic bands and the remaining primer pairs did not exhibit polymorphism(Table S3).

To evaluate the selected 571 SSR primer pairs for their polymorphism,the randomly sampled 100 pairs were applied to amplify the 20 sweet potato cultivars with extensive genetic backgrounds.The results demonstrated that all of them generated clear and reproducible amplicons in the 20 cultivars on 6% denaturing polyacrylamide gels.The number of alleles detected per pair of primers ranged from 3 to 34,with a mean value of 14 alleles.The PIC varied from 0.4118 to 0.9558 with a mean of 0.8406,Ne*ranged from 1.1418 to 1.8205 with a mean of 1.4935,H*was from 0.1223 to 0.4460 with a mean of 0.3001,and I*was from 0.2380 to 0.6368 with a mean of 0.4609(Table S4).These results demonstrated that the selected SSR primer pairs had good polymorphism.

3.2.Scoring and segregation of SSR markers

To develop SSR markers in the mapping population,the 571(273 gSSR and 298 bSSR)polymorphic primer pairs selected in this study were used to amplify Luoxushu 8 and Zhengshu 20 and their 240 F1individuals,and 544 of them generated 3735 scored markers,with an average of 8 markers(Table S3).The remaining 27 pairs were removed from the subsequent analyses because they produced low quality markers.The previously developed 35 eSSR primer pairs produced 209 scored markers,with an average of 6 markers(Table S3).The number of SSR markers scored for Luoxushu 8 was 2678,which contained 1120 simplex,259 duplex,35 triplex and 1264 double-simplex markers,and the number of SSR markers for Zhengshu 20 was 2428,including 950 simplex,190 duplex,24 triplex and 1264 double-simplex markers according to the χ2goodness-of-fit test.Of these markers,1453 and 1539 were found to exhibit distorted segregation in Luoxushu 8 and Zhengshu 20,respectively,at the 0.05 probability level.

Expected and observed proportion of single-to multiple-dose markers was compared in order to evaluate ploidy type in sweet potato[9]andSaccharum spontaneum[29].In this study,the percentage of simplex markers was 79.2%(1120/(1120+259+35))in Luoxushu 8 and 81.6%(950/(950+190+24))in Zhengshu 20,which were roughly in agreement with expectations for an autohexaploid(75% simplex and 25% non-simplex).Only duplex and triplex markers in multiple-dose markers were used to assess this proportion because they are the only informative multiple-dose markers when dealing with dominant markers[2,9,11,12].

3.3.Construction of genetic linkage maps

A framework map of each parent was constructed using simplex markers at LOD 5.0 with Joinmap 4.0 software.Simplex markers were grouped into 90 linkage groups for each parental map.Duplex,triplex and double-simplex markers were then placed on the framework maps to form the final SSR genetic linkage maps of both parents(Figs.S1 and S2).

The genetic linkage map of Luoxushu 8 included 90 linkage groups with 5057 SSR markers,which consisted of 505 simplex,3625 duplex,584 triplex and 343 double-simplex markers.Sweet potato is an autohexaploid,the same duplex or triplex marker can be located on different linkage groups.The number of markers in the Luoxushu 8 map was the cumulative number of markers located on 90 linkage groups.The total map length was estimated as 13,299.9 cM,with a marker density of 2.6 cM.The length of each linkage group varied from 73.1 to 241.3 cM,with an average size of 147.8 cM.The number of mapped markers on each linkage group ranged from 9 to 108,with an average of 56 markers.There were 628(12.4%)distorted markers for Luoxushu 8 at the 0.05 probability level(Table 1).

For Zhengshu 20,the genetic linkage map contained 90 linkage groups with 3009 SSR markers,including 481 simplex,1752 duplex,414 triplex and 362 double-simplex markers.The total map length was 11,122.9 cM,with a marker density of 3.7 cM.The length of each linkage group ranged from 47.8 to 206.1 cM,with an aversge size of 123.6 cM.The number of mapped markers on each linkage group varied from 6 to 103,with an average of 33 markers.A total of 675(22.4%)markers displayed segregation distortion(α=0.05)in the Zhengshu 20 map(Table 2).

3.4.Homologous analysis among linkage groups

Duplex and triplex markers were used to determine the homologous linkage groups in each parental map.As expected,15 homologous groups were identified in each parental map,which were consistent with the chromosome number of the autohexaploid sweet potato(Figs.S1 and S2).Furthermore,the homology of the corresponding linkage groups between both parental maps was detected with double-simplex markers.The corresponding relationships between 49 linkage groups of the Luoxushu 8 map and 50 of the Zhengshu 20 map were identified with 250 doublesimplex markers(Table S5).

4.Discussion

4.1.Screening and evaluation of polymorphic SSR primer pairs

As is commonly known,SSR markers are widely used for researches in genetic diversity[30,31],origin and evolution[32,33],fingerprinting[34,35],linkage mapping[36,37],comparative genomics[38,39]and gene-based association studies[40,41].In sweet potato,SSR markers have also been used in varieties identification,genetic diversity analysis,fingerprinting and linkage map construction[2,8,12,35,42–48].However,the SSR markers are mainly EST-SSRs and the number of available SSRs is very limited in sweet potato.

In this study,the 2545 SSR primer pairs designed from the genome sequence and BAC-end sequence of sweet potato were detected for their polymorphism and 571 of them were found to be polymorphic in both parents(Zhengshu 20 and Luoxushu 8)and their randomly sampled two F1individuals(Table S3).Their polymorphism was further verified by evaluating the randomly sampled 100 pairs with the 20 sweet potato cultivars(Table S4).Therefore,the polymorphic gSSR and bSSR primer pairs selectedin this study are of high practicability,and can be used for genetic linkage map construction,fingerprinting,genetic diversity analysis and comparative genomics research in sweet potato.

Table 1(continued)

Table 1Distribution of SSR markers on each linkage group in the Luoxushu 8 linkage map.

4.2.Construction of genetic linkage maps

Grattapaglia and Sederoff[27]put forward a two-way pseudotestcross strategy on the F1populations for constructing the genetic linkage maps of genetically heterozygous species.Sweet potato is a clonally propagated autohexaploid species with the highly heterozygous genome.This strategy has been applied to construct genetic linkage maps in sweet potato[2,3,8–14].

SSR markers are powerful tools for molecular breeding.In sweet potato,though several SSR genetic linkage maps have been constructed,they can not satisfy the QTL mapping,marker-assisted selection and comparative genomics research due to their low genome coverage and marker density[2,8,12].In this study,using the selected polymorphic SSR primer pairs and two-way pseudotestcross strategy,we successfully developed the first SSR linkage maps consisting of the complete 90 linkage groups and 15 homologous groups,which are consistent with the actual chromosome number and autohexaploid nature of sweet potato(Figs.S1 and S2).The Luoxushu 8 and Zhengshu 20 maps contained 5057 and 3009 SSR markers and covered 13,299.9 cM and 11,122.9 cM with a marker density of 2.6 cM and 3.7 cM,respectively(Tables 1 and 2).These are the linkage maps with the highest SSR marker density reported to date in sweet potato[2,8,12].Both SSR linkage maps have also much better coverage of the sweetpotato genome than those developed by Ma et al.[12],in which 484 and 573 SSR markers were clustered into 90 linkage groups covering 3974.24 cM and 5163.35 cM of‘Jizishu 1’and‘Longshu 9’,respectively.It has been showed that increasing the number of markers resulted in the significant increase of the genomic coverage in sweet potato[2].Kriegner et al.[9]reported that 632 and 435 AFLP markers were ordered in 90 and 80 linkage groups of‘Tanzania’and‘Bikilamaliya’and total map lengths were 3655.6 cM and 3011.5 cM,respectively.A total of 1166 and 960 AFLP markers were placed in 86 and 90 linkage groups covering 5792 cM and 5276 cM of‘Tanzania’and‘Beauregard’,respectively[11].The‘Xushu 18’map included 1936 AFLP and 141 SSR markers and covered 8184.5 cM,and the‘Xu 781’map contained 1824 AFLP and 130 SSR markers and covered 8151.7 cM[2].Thus,the success of this research can be attributed to the development of a large number of SSR markers,which has taken an important step forward in the construction of sweet potato linkage maps.

Furthermore,the length of each linkage group ranged from 73.1 to 241.3 cM in Luoxushu 8 and from 47.8 to 206.1 cM in Zhegnshu 20,respectively(Tables 1 and 2).The number of SSR markers distributed on each linkage group varied from 9 to 108 in Luoxushu 8 and from 6 to 103 in Zhengshu 20,respectively(Tables 1 and 2).Differences in size and marker number of each linkage group might be due to the physical size differences of the chromosomes in sweet potato[2,12].The chromosomes of sweet potato are very small and their sizes vary several-fold[49].Therefore,this work provides a solid framework map for further understanding genetics of sweet potato.

Table 2(continued)

Table 2Distribution of SSR markers on each linkage group in the Zhengshu 20 linkage map.

As mentioned above,the two-way pseudo-testcross strategy and software JoinMap are usually used to construct linkage maps in sweet potato[2,3,8–14].However,they may cause the loss of less informative markers for constructing linkage maps of clonal species[50,51].Recently,Zhang et al.[51]proposed a new method for genetic analysis and map construction in the clonal F1population,in which clonal F hybrids could be regarded as a double cross population.Comparisons with other methods showed that the proposed method built more accurate linkage maps with less time in simulated clonal F1populations and one actual maize double cross.They also developed the software package GACD(Genetic Analysis of Clonal F1and Double cross),which is capable of building highdensity linkage maps and mapping QTL in clonal F1and double cross populations[52].This software may be used to construct linkage maps of sweet potato in the future.Additionally,for complex polyploids,the linkage map construction is restricted mostly to two-point marker analysis.Mollinari et al.[17]built an ultradense multilocus integrated SNP linkage map in sweet potato using their newly developed software,MAPpoly.This will provide an alternative tool for linkage map construction and genetic analysis in complex polyploid species.

4.3.Phenomenon of segregation distortion

The phenomenon of segregation distortion existes in most species,such as rice[53],wheat[54],cotton[55],tomato[56],flax[57],grape[58],and sweet potato[2,3,8,9,11,12].Segregation distortion is thought to be a potent evolutional force in living species[59].

It is reported that distorted segregation ranged from 12.9% to 49.4% in sweet potato[2,9,12].In this study,12.4% and 22.4% of SSR markers exhibited distorted segregation(α=0.05)in Luoxushu 8 and Zhengshu 20,respectively.Several factors,including gametophyte selection,cytological attributes,genetic drift or some biological reasons,might contribute to the segregation distortion[60–62].Moreover,segregation distortion has been proven to have little impact on map order and size[63].

5.Conclusions

The 273 gSSR and 298 bSSR primer pairs with polymorphism were selected from 1215 gSSRs and 1330 SSRs,respectively,in sweet potato.The polymorphic SSR markers were used to genotype 240 F1individuals derived from a cross between Luoxushu 8 and Zhengshu 20,and the SSR linkage maps with the best genome coverage and the highest marker density reported to date were successfully developed in this crop.This study provides a basis for QTL mapping,marker-assisted selection and comparative genomics research of sweet potato.

CRediT authorship contribution statement

Qingchang Liu,Yusha Meng,Chenxing Zheng and Hong Zhaidesigned the experiments;Yusha Meng,Chenxing Zheng,Hui Li,Aixian Li,Qingmei Wang,Ning Zhao,Huan Zhang and Shaopei Gaoconducted the experiments;Qingchang Liu,Yusha Meng and ShaozhenHe wrote the paper.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the National Key Research and Development Program of China(2018YFD1000706/2018YFD1000 700)and China Agriculture Research System(CARS-10,Sweet potato).

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2021.01.003.