Potato (Solanum tuberosum L.) is the fourth ranked crops in the world production and a host to diverse pathogens and insects. Therefore, achieving high level resistance to them is one of the most important targets in potato breeding and various wild Solanum species have been used (Pavek and Corsini 2001; Simko et al. 2007).

S. chacoense originating from Argentina is one of wild potato species (Micheletto et al. 2000) and is important as resistant resources to several important pests and pathogens such as Colorado potato beetle, Phytophthora infestans, potato leafroll virus, potato tuberworm, Verticillium wilt, etc. in potato breeding (Brown and Thomas 1993; Cooper et al. 2009; Lynch et al. 1997; Micheletto et al. 2000; Sinden et al. 1986). However, its ploidy level is diploid and endosperm balance number (EBN) is two that are different from those of potato. Potato is tetraploid and its EBN is four. These conditions cause sexual incompatibility between S. chacoense and potato and restrict the utilization of the wild species in classical potato breeding (Cho et al. 1997; Oritz and Ehlenfeldt 1992). Therefore, protoplast fusion was used with the two different species for crop improvement to overcome this sexual barrier for transferring certain interspecific genes from wild species into potato (Bidani et al. 2007; Nouri-Ellouz et al. 2016). Interspecific somatic hybrids between wild species such as S. nigrum, S. brevidens, S. phureja, and S. commersonii and potato were successfully produced in potato breeding (Barsby et al. 1984; Binding et al. 1982; Kim-Lee et al. 2005; Putie et al. 1986) and this technique has also been applied in between S. tuberosum and S. chaconese (Chen et al. 2013). When the products were obtained from the somatic fusion, the identification of the fusion products regarding both nuclear and organelle DNA is essential to confirm the hybridity with molecular markers because a large amount of nuclear and cytoplasmic DNA can be transferred during the somatic hybridization between two sexually incompatible species (Cho et al. 2016; Cho and Park 2016; Guo et al. 2004).

The chloroplast genome (cpDNA) has a circular double- stranded DNA molecule and chloroplast is a photosynthetic intracellular organelle. The size of cpDNA of angiosperm is approximately 115~165 kb and its organization is typically quadripartite consisting of two inverted repeats (IR), the large single copy (LSC) and small single copy (SSC) regions (Yurina and Odintosova 1998). Most cpDNA contain approximately 110~130 genes encoding diverse unique proteins, rRNAs, and tRNAs. Several cpDNA sequences of Solanum species such as S. berthaultii (KY419708, Kim et al. 2018), S. tuberosum (KM489056, Cho et al. 2016) and (NC008096, Gargano et al. 2005), S. commersonii (KM489055, Cho et al. 2016), S. bulbocastanum (NC007943, Daniell et al. 2006), and S. nigrum (KM489054, Cho and Park 2016) have previously been completed (Table 1). Their size and gene organization are very similar and their structures are exactly same with two copies of IRs, LSC, and SSC regions (Palmer 1991; Raubeson and Jansen 2005; Saski et al. 2005; Sugiura et al. 1998). Nevertheless, sufficient information can be obtained from the comparison of complete cpDNA sequences among plant species because a number of mutations caused by structural changes, for instance, rearrangements of gene order and inversions, single nucleotide polymorphism (SNPs), and insertions/ deletions (InDels) in cpDNAs of many plants occurs (Calsa Junior et al. 2004; Cho et al. 2015; Jheng et al. 2012; Kim et al. 2005; Saski et al. 2005).

Table 1 . Summary of cpDNA sequences of S. chacoense and five other Solanum species.

Species	Total Length (bp)	Total No. of genes	No. of rRNA	No. of tRNA	GC content (%)	Accession no.	Reference
S. chacoense	155,532	136	4	36	38	MF471371	In this study
S. berthaultii	155,533	137	4	39	38	KY419708	Kim et al. (2018)
S. nigrum	155,432	139	4	39	38	KM489055	Cho and Park (2016)
S. commersonii	155,525	133	4	33	37.5	KM489054	Cho et al. (2016)
S. tuberosum	155,296	131	4	36	37.5	NC008096	Gargano et al. (2005)
S. tuberosum	155,312	130	4	30	37	KM489056	Cho et al. (2016)
S. bulbocastanum	155,371	133	4	30	38	DQ347958	Daniell et al. (2006)

*The data have been partially adopted from Kim et al. (2018)..

The complete cpDNA sequence of S. chacoense was previously announced (Cho et al. 2017). The detailed information of the cpDNA sequence, the comparison among the cpDNA sequence and those of other species in the Solanaceae family, and PCR-based markers specific to S. chacoense were described in this study.

Plant Materials

18 genotypes of Solanum species such as S. acaule (SA), S. berthaultii (SB1), S. brevicaule (SB2), S. cardiophyllum (SC1), S. chacoense (SC3), S. hjertingii (SH1), S. hougassi (SH2), S. jamesii (SJ), S. kurtzianum (SK), S. microdontum (SM2), S. mochiquense (SM1), S. pinnatisectum (SP), S. stoloniferum (SS), S. tuberosum (PT56 and ‘Tamra’), S. tuberosum subsp. andigenum (ST2), S. verrucosum (SV1), and S. vernei (SV2) were obtained from the HARI (Highland Agriculture Research Institute). Their genebank accession numbers are PI310970, PI310981, PI205394, PI341233, PI201846, PI186559, PI161174, PI578236, PI498422, PI310979, PI338616, PI160224, PI566805, PI160228, and PI230468, respectively, except the S. tuberosum breeding line ‘PT56’ and variety ‘Tamra’. They were used to compare and develop molecular markers.

DNA Extraction

Approximately 100 mg of fresh leaves were collected from in vitro or greenhouse plants and used for the total genome DNA isolation using a Genomic DNA Extraction kit for plants (RBC, New Taipei City, Taiwan).

Chloroplast Genome Sequencing of Solanum chacoense

For the cpDNA sequencing of S. chacoense, the total genomic DNA of the line SC3-12 of S. chacoense was extracted and sequencing was conducted using the Illumina Hiseq2000 (Illumina, SanDiego, CA, USA) platform at Macrogen (Macrogen, Seoul, South Korea). The cpDNA sequence assembly was performed by de novo assembly protocol via the Phyzen bioinformatics pipeline (http:// phyzen.com) (Cho et al. 2015). De novo assembly was conducted using trimmed reads by a 4.06 beta version of CLC genome assembler (CLC Inc, Rarhus, Denmark) with parameters of a minimum (300-1,000 bp) autonomously controlled overlap size. Raw score reads of 20 or less from the overall reads were discarded using the CLC- quality trim tool. The principal contigs of the cpDNA were retrieved by Nucmer (Kurtz et al. 2004). As the reference sequence, the cpDNA sequence of S. commersonii (KM489054) was used for it (Cho et al. 2016). The representative cpDNA contigs were arranged in order with the results from BLASTZ analysis with the reference sequence (Schwartz et al. 2003). A single draft cpDNA sequence of S. chacoense was completed by joining the overlapping terminal sequences and manual editing via a comparison with the S. commersonii (KM489054) cpDNA sequence reported by Cho et al. (2016). Gene annotation was conducted using DOGMA (Wyman et al. 2004) and the circular map of the S. chacoense cpDNA was generated using the software OrganellarGenome DRAW (http://ogdraw.mpimp-golm.mpg.de) (Lohse et al. 2013).

Comparison of cpDNA Sequences with Other Solanaceae Species

The chloroplast coding sequences examined by MEGA 6.0 (Tamura et al. 2013) with eight species including S. chacoense (MF471371) identified in this study and seven Solanaceae species, Capsicum annuum (JX270811), S. berthaultii (KY419708), S. bulbocastanum (NC007943), S. commersonii (KM489054), S. lycopersicum (DQ347959), S. nigrum (KM489055), and S. tuberosum (KM489056 and NC008096) obtained from NCBI (the National Center for Biotechnology Information) were applied to phylogenetic analysis. The same parameters employed in the maximum parsimony methods under PAUP* v4.0b10 were used for the likelihood search (Swofford 2001). A total of 1,000 bootstrap replicates and a 50% cutoff value for the bootstrap consensus tree were applied. Multiple alignment was also conducted using the complete cpDNA sequences of six species such as S. berthaultii (KY419708), S. chacoense (MF471371), S. commersonii (KM489054), S. nigrum (KM489055), S. stoloniferum (MF471373), and S. tuberosum (KM489056) by ClustalW2 in EMBL (https://www.ebi.ac.uk/ Tools/msa/clustalw2).

Development of specific PCR-Based Markers for Solanum chacoenes

The S. chacoense specific InDels and SNPs were identified using the results of multiple alignment with five Solanum species: S. stoloniferum (MF471373), S. berthaultii (KY419708), S. commersonii (KM489054), S. nigrum (KM489055), and S. tuberosum (KM489056). The primers on each InDel region were designed to examine the allele specificity for S. chacoense. PCR was carried out with the 18 genotypes (PT56, ‘Tamra’, SC3, SA, SP, SH1, SB1, SM1, SC1, SV1, SK, SS, SH2, SM2, SB2, SV2, SJ, and ST2) in the 20 ul mixture (20 ng genomic DNA, 0.5 mM dNTP mixture, 10 pMol each primer and 1 U Taq polymerase (Genetbio, Daejeon, South Korea)) as described by Cho and Park (2016) in a thermocycler (Biometra, Göttingen, Germany). The PCR products were analyzed using nucleic acid staining solution RedSafe (Intron Biotechnology, Seongnam, South Korea) on the 1% agarose gel. For the development of SNP-derived specific markers to S. chacoense, the primers on each SNP region were designed after checking whether the proper restriction enzymes, which could digest only a SNP region on the sequence of S. chacoense, but not on the sequences of aligned other five Solanum species or only on the sequences of other five Solanum species, but not on the sequence of S. chacoense existed using dCAPS Finder 2.0 (http://helix.wustl.edu/ dcaps/dcaps.html). PCR and gel-electrophoresis were routinely performed as described above. When certain primer amplifies all 18 genotypes, the PCR products were treated with the selected restriction enzyme. Electrophoresis on 1% agarose gel separated the digested fragments and polymorphic bands were visualized on UV lights.

cpDNA Sequence of S. chacoense

The sequencing of the S. chacoense complete cpDNA was conducted using the NGS technology. The Illumina PE standard protocol produced a library generating a total of 632,775,161bp sequences with an average read length of 297bp. Three representative contigs were generated by assembling de novo paired end sequences of the whole genome that covered the entire cpDNA sequence of S. commersonii (KM489054, Cho et al. 2016). BLASTZ analysis arranged them in order (Schwartz et al. 2003) (Fig. 1).

Figure 1.

Result of assembly of the complete cpDNA sequence of S. chacoense. The four representative contigs of the S. chacoense cpDNA and the corresponding regions of S. commersonii cpDNA (KM489054) are present. Yellow and blue bars indicate contig matching with the reference sequence in reverse and forward orientation, respectively

The putative errors in the assembled sequence were curated by mapping 154.34x raw reads onto the final assembly. PCR and BigDye Terminator Cycle Sequencing with ABI3730 validated further on several regions including border sequences of a pair of inverted repeat regions (IRs), a small single copy region (SSC) and a large single copy region (LSC). As a part of the results has shortly been announced (Cho et al. 2017), the final cpDNA assembly reveals that the cpDNA has double-stranded DNA molecules and is circular as usual in most plant chloroplasts and the size of the complete cpDNA sequence of S. chacoense is 155,532 bp (GenBank accession no. MF471371). Its structure is typically quadripartite and includes 25,592 bp IRs connected with SSC of 18,376 bp and LSC of 85,972 bp. The length is somewhat longer than those of other species in the Solanum genus, but 1 bp shorter than that of Solanum berthaultii (Table 1). The S. chacoense cpDNA contains 156 genes, 23 of which are duplicated in the IRs (Fig. 2). They included 106 protein-coding genes, 48 tRNAs and eight rRNAs. Ten, nine and four of the protein coding genes, the tRNAs and rRNAs are inversely duplicated in the IRs, respectively (Table 1 and Fig. 2). 59.5% of the cpDNA occupied coding regions with an average size of 582.4 bp. There are 51.9% protein coding and 7.6% RNA genes with an average size of 761.7 bp and 223.7 bp, respectively. The GC content of the cpDNA is 37.89% that is highly conserved with those of other Solanum species (Table 1) and the predicted genes, tRNA and rRNA of the S. chacoense cpDNA are almost same in number and order with those of other Solanum species (Cho et al. 2016; Cho and Park 2016).

Figure 2.

Gene map of the S. chacoense cpDNA. Genes on the inside and outside of the map are transcribed in the counterclockwise and clockwise direction, respectively

Phylogenetic and Comparative Analysis of the cpDNA Sequences

The phylogenetic analysis was conducted using the S. chacoense cpDNA coding sequences and those of the seven species in the Solanaceae family (Fig. 3). The maximum likelihood and parsimony analyses produced the same topology and the high bootstrap values supported strongly most nodes in the phylogenetic tree. The results interestingly showed that the closest Solanum species with S. chacoense was S. commersonii originating from Argentina.

Figure 3.

Maximum-likelihood phylogenetic tree of S. chacoense with seven other Solanaceae species. The analysis was conducted with chloroplast protein-coding sequences of each cpDNA. The numbers indicate the bootstrap values from 1,000 replicates on each node

To develop molecular markers which could discriminate S. chacoense cytoplasm, the complete S. chacoense cpDNA sequence was aligned with those of the five other species in the Solanum genus. A number of InDels or SNPs was previously reported by the alignment of the cpDNA sequences in the Solanum species (Cho and Park, 2016; Chung et al., 2006; Kim et al., 2018). Overall 280 and 1,490 InDels and SNPs through the whole cpDNA sequences were detected by the multiple alignment with six species generated using ClustalW2 in EMBL (https://www.ebi.ac.uk/Tools/msa/ clustalw2), respectively.

Discovery and Verification of S. chacoense-Specific Molecular Markers

The alignment of the cpDNA sequences revealed many InDels and SNPs. However, it was mainly caused by S. nigrum species that was farther phylogenetically from other Solanum species (Fig. 3). Therefore, the InDels or SNPs should be limited with criteria. The sequences on the InDels or SNPs should be specific to S. chacoense, but conserved in the five other Solanum species. SNPs were almost equally positioned on coding and non-coding regions through the whole cpDNA sequence, however, InDels on the non-coding regions were positioned more than on the coding regions in the cpDNA of Solanum species (Cho and Park 2016; Chung et al. 2006).

PCR with primer designed on InDel regions could be effectively applied to develop molecular markers that reveals polymorphism and the markers can be used to distinguish plant species (Garcia-Lor et al. 2013; Cho et al. 2015; Yamaki et al. 2013). Only two of the 280 InDels detected on the coding regions, rps4 and accD, were comparable enough to develop markers specific to S. chacoense. Fourteen and four base pairs on the S. chacoense sequence differ from those of other five Solanum species on the first (SC3_InDel_11) and the second (SC3_InDel_16) InDel regions, respectively (Fig. 4). The InDel specific primers were designed on those two regions to develop PCR-based markers (Table 2) and used for PCR with a S. tuberosum breeding line (PT56), a potato variety ‘Tamra’, S. chacoense (SC3), S. acaule (SA), S. pinnatisectum (SP), S. hjertingii (SH1), S. berthaultii (SB1), S. mochiquense (SM1), S. cardiophyllum (SC1), S. verrucosum (SV1), S. kurtzianum (SK), S. stoloniferum (SS), S. hougassi (SH2), S. microdontum (SM2), S. brevicaule (SB2), S. vernei (SV2), S. jamesii (SJ) and S. tuberosum subsp. andigenum (ST2). The results of PCR showed that the two InDels-based primers amplified specifically only S. chacoense as expected (Fig. 5A and 5B).

Table 2 . Primers and restriction enzymes used to develop S. chacoense specific markers.

Marker name	Region	Sa	Primer sequence	Size (bp)b	REc
SC3_InDel_11	rps4 (Intragenic)	F	TTGTATCTTTATCCCGGAGC	300	as
		R	GTTGTTAGTAAAATGTTGTTAG
SC3_InDel_16	accD (Intragenic)	F	CTTTGTTCCGTGTTGAAATA	334	as
		R	AAGCATAGAAATCCAGTGTT
SC3_SNP_7	rpoB (Intragenic)	F	CTAGCAGACTCCCTGCTAAC	397	MboII
		R	GTGAAAGAATCTTCGTATGC
SC3_SNP_15	accD (Intragenic)	F	TATAGGAGAGGACAAATCTC	406	MaeI
		R	TTTCAATAGGACCAAGACTG

^aF and R indicate forward and reverse strand of primers.

^bThe expected size of PCR fragments are measured based on the sequence of S. chacoense.

^cRestriction enzymes generating S. chacoense specific markers. ‘as’ indicates allele specific marker.

Figure 4.

Sequence alignments on the each InDel and SNP regions. The cpDNA sequences of S. stoloniferum (SS16: MF471373), S. chacoense (SC3-12: MF471371), S. berthaultii (Sber: KY419708), S. commersonii (Scom: KM489054), S. nigrum (Snig: KM489055), and S. tuberosum (Stub: KM489056) were included. The sequences on the InDels and SNPs in the sequence of S. chacoense are highlighted. The restriction sites are red-colored and bold

Figure 5.

PCR-based markers for the discrimination of S. chacoense from other Solanum species. A: SC3_InDel_11. B: SC3_InDel_16. C: SC3_SNP_7. D: SC3_SNP_15. The four markers are all positively specific to S. chacoense. M, PT56, TR, SC3, SA, SP, SH1, SB1, SM1, SC1, SV1, SK, SS, SH2, SM2, SB2, SV2, SJ, and ST2 indicate size marker ladder, a S. tuberosum breeding line, a potato variety (Tamra), a line of S. chacoense, S. acaule, S. pinnatisectum, S. hjertingii, S. berthaultii, S. mochiquense, S. cardiophyllum, S. verrucosum, S. kurtzianum, S. stoloniferum, S. hougassi, S. microdontum, S. brevicaule, S. vernei, S. jamesii and S. tuberosum subsp. andigenum, respectively

In total, 37 of the 1,490 SNPs were specific to S. chacoense. Molecular markers can effectively be developed by the application of PCR on the regions covering SNPs followed by the digestion of PCR product with the appropriate restriction enzymes that called the cleaved amplified polymorphism sequence (CAPS) marker (Konieczny and Ausubel 1993). The CAPS markers were widely used to compare with polymorphism in different genotypes of Solanum species (Cho and Park 2016; Cho et al. 2016; Kim et al. 2018; Park et al. 2005; Smilde et al. 2005; Uribe et al. 2014) and other plant species (Komori and Nitta 2005; Uncu et al. 2015; Wang et al. 2017). Therefore, 17 primer pairs which can amplify regions including each SNPs were designed and the proper restriction enzymes had been identified. The enzymes should digest only S. chacoense but not other Solanum species or digest all of other Solanum species, but not S. chacoense. PCR was performed with the primers on 18 different lines of Solanum species as above and the each PCR product was treated with the each proper restriction enzyme. Finally, two CAPS markers were confirmed (Table 2 and Fig. 5). Two primer combinations and restriction enzymes produced unique bands in the four S. chacoense lines, but not in the other Solanum species (Table 2, Fig. 5C and 5D).

The molecular markers characterizing chlorotypes are crucial for not only evolutionary studies, but also potato breeding in Solanum species (Bohs and Olmstead 1997; Hosaka and Sanetomo 2012). During somatic fusion and in vitro plant regeneration, a random distribution of cpDNA was reported in certain genotypes of plant species including Solanum species, although a high frequency of recombination occurred in mtDNA (mitochondrial genome) (Chen et al. 2013; Cho et al. 2016; Lössl et al. 2000; Mohapatra et al. 1998; Smyda-Dajmund et al. 2016; Xiang et al. 2004). Overall, the markers developed based on the InDels and SNPs identified in this study can be used to reliably distinguish S. chacoense from other Solanum species, to select the proper chlorotype from the somatic hybrids between S. chacoense and S. tuberosum and to accelerate potato breeding using S. chacoense.

Golden Seed Project supported this study (Project No. 213009-05-3-WT411). The project was funded by MAFRA, MOF, RDA, and KFS via IPET whose full names are Ministry of Agriculture, Food and Rural Affairs, Ministry of Oceans and Fisheries, Rural Development Administration, Korea Forest Service, and Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries, respectively.