Fresh leaves of S. viarum were collected in July 2022 from Jejudo Island, Korea (33°19′40.8″N, 126°19′56.6″E) (Fig. 1). The voucher specimen was deposited at the Korea National Arboretum Herbarium (KH) under voucher number KHB1662054 (Fig. 2); the contact person is Dong Chan Son (E-mail: sdclym@korea.kr).

Total DNA was extracted using a DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) following the manufacturer’s protocol. The extracted DNA was detected on agarose gel (2%), and high-quality gDNA was sequenced using an Illumina MiSeq platform (Illumina Inc., San Diego, CA, USA) with a 301-bp insert size. In total, 7,061,852 reads were obtained and combined using Geneious v.8.0.5. (https://www.geneious.com), with reference to S. aculeatissimum Jacq. (OR381845.1), which is registered in the National Center for Biotechnology Information (NCBI) database. Thereafter, to verify the genome assembly, read coverage depth was assessed using the Draw_SequencingDepth.py script as described by Ni et al. (2023). Finally, GeSeq (Tillich et al., 2017) was used for genome sequence annotation. A circular map of the complete chloroplast genome was generated using OrganellarGenomeDRAW (Greiner et al., 2019). The complete sequenced chloroplast genome of S. viarum was submitted to GenBank (accession no. PV866731).

Maximum likelihood (ML) analysis was performed with 34 subgenus of Leptostemonum (genus Solanum) and two outgroups (genus Calystegia in Convolvulaceae) to investigate the phylogenetic relationships. The chloroplast genome sequences of 36 species excluding S. viarum were downloaded from the NCBI GenBank database. In total, 78 coding DNA sequences (CDSs) were concatenated and aligned using MAFFT in PhyloSuite v.1.2.2 (Zhang et al., 2020); this process was performed using Geneious. The analysis model was tested using PhyloSuite v.1.2.3 (Zhang et al., 2020; Xiang et al., 2023). ML analysis was performed with IQ-tree v.2.2.0 (Nguyen et al., 2015) by setting the transversion model (TVM) + freerate model parameters with three of categories (R3) + empirical base frequency (F) model with 5,000 ultrafast bootstraps (Minh et al., 2013).

The complete chloroplast genome of S. viarum has a sequence length of 156,209 bp and a 37.8% GC content (Fig. 3), with an average read coverage depth of 165.79× (Fig. 4). The genome sequence was deposited in GenBank under the accession number PV866731, with associated BioProject, Sequence Read Archive, and Bio-Sample numbers of PRJNA1285601, SUB15430035, and SAMN49770012, respectively. The genome displayed a typical quadripartite structure comprising a pair of inverted repeats (IR, 25,419 bp each) separating a large single-copy region (LSC, 86,877 bp) from a small single-copy region (18,494 bp). The genome comprised 130 genes, including 84 CDSs, 37 tRNA genes, 8 rRNA genes, and 1 pseudogene (infA) (Table 1). The IR contained 18 genes, including 6 CDS genes, 7 tRNA genes, and 4 rRNA genes (CDS: ndhB, rpl2, rpl23, rps12, rps7, ycf2; tRNA: trnA-UGC, trnl-CAU, trnl-GAU, trnL-CAA, trnNGUU, trnR-ACG, trnV-GAC; rRNA: rrn16, rrn23, rrn4.5, rr5). Of the genes, 17 had one intron (10 CDS: atpF, ndhA, ndhB, petB, petD, rpl16, rpl2, rpoC1, rps12, rps16; 6 tRNA: trnAUGC, trnG-UCC, trnI-GAU, trnK-UUU, trnL-UAA, trnVUAC; 1 rRNA: rrn23) and two had two introns (clpP1 and pafI) (Fig. 5A). The rps12 gene is a trans-spliced gene with the 5′ end located in the LSC region and the duplicated 3′ end located in the IR regions (Fig. 5B).

An ML phylogenetic tree was constructed from the chloroplast genome of 78 CDSs from 37 species, including 35 species from the subgenus Leptostemonum of the genus Solanum (family Solanaceae) and two species from the genus Calystegia (family Convolvulaceae) as outgroups (Fig. 6). ML analysis supported the monophyly of subgenus Leptostemonum (bootstrap value = 100). The resulting phylogenetic tree was structured according to the classification systems described by Levin et al. (2006) and Stern et al. (2011). However, S. hieronymi, S. multispinum, and S. crotonoides were treated as exceptions because their phylogenetic relationships have not been clearly resolved in previous studies (Whalen, 1984; Levin et al., 2006; Stern et al., 2011). Specifically, S. viarum, S. aculeatissimum, and S. capsicoides (Acanthophora clade) formed a strongly supported clade along with S. lasiocarpum (Lasiocarpa clade) (bootstrap value = 99), which was placed at the basal position in the tree. Within this clade, members of the Acanthophora clade were grouped into a well-supported subclade (bootstrap value = 100), with S. viarum and S. aculeatissimum showing a close relationship strongly supported by a high bootstrap value (Fig. 6). This result is consistent with previous phylogenetic analyses (Levin et al., 2006) based on two nuclear gene regions (internal transcribed spacer and granule-bound starch synthase gene [GBSSI or waxy]) and one chloroplast spacer region (trnS–trnG), which also identified S. viarum and S. aculeatissimum as sister taxa. S. viarum and S. aculeatissimum were previously treated as S. khasianum (Babu and Hepper, 1979), but a subsequent monographic revision by Nee (1979) clarified their status as distinct species. The results of this study confirmed that S. viarum is closely related to S. aculeatissimum but remains a separate taxon distinguished by phylogenetic analysis of the chloroplast genome. In addition to clarifying the phylogenetic placement of S. viarum, this study also revealed that the Androceras/Crinitum clade exhibits paraphyly. This finding contrasts with Stern et al. (2011), who reported weak support for the clade (bootstrap value = 63) based on limited nuclear and chloroplast markers. By contrast, a more recent chloroplast genome study (Zhang et al., 2024) produced results consistent with ours, and in our analyses S. wrightii and S. rostratum were clearly separated with strong support (bootstrap value = 100). These results underscore the utility of complete chloroplast genomes for resolving problematic relationships within Solanum. This study reports the complete chloroplast genome of S. viarum for the first time, as well as a phylogenetic analysis of the subgenus Leptostemonum. These findings provide a molecular basis for future phylogenetic studies within this subgenus.