Phylogeographic relationships among Asian eggplants and new perspectives on eggplant domestication
Graphical abstract
Highlights
► Analyses revealed two candidate progenitors of Solanum melongena may be conspecific. ► Gene flow appears to have occurred among domesticated eggplant populations in Asia. ► Results support two main domestication events in India and southern China. ► Results support independent domestication of S. melongena subsp. ovigerum in SE Asia.
Introduction
Eggplants are an economically important food crop available worldwide, with more than 2,000,000 hectares dedicated annually to their cultivation (2008 data; http://faostat.fao.org). The center of diversity for eggplant is tropical Asia (Vavilov, 1951, Lester and Hasan, 1991, Furini and Wunder, 2004), and in many Asian regions eggplant ranks in the top five most important vegetables (Frary et al., 2007). Currently, only a few popular cultivars are represented in global trade (e.g., China’s Hainan eggplant; Muñoz-Falcón et al., 2009). Worldwide, eggplant varietal diversity is decreasing (Ali et al., 2011), however several regions in Asia are rich in number of cultivars. Found throughout the Philippines, southern China. Southeast Asia, and India, thousands of local landraces and heirloom cultivars (collectively referred to as ‘landraces’ throughout this paper) exist and these exhibit a wide range of variability in morphology, flavor, and pathogen resistance (Fig. 1). To conserve landrace diversity, large-scale efforts to collect and maintain germplasm of wild relatives and landraces are underway, currently encompassing more than 15,000 collections in 99 institutions worldwide for landraces alone (WIEWS; http://apps3.fao.org/wiews).
Worldwide landraces harbor alleles that may be potentially useful for plant breeding efforts aimed at improving eggplant stress tolerance, disease resistance, and nutritional quality (Brush, 2000, Smalde, 2006). According to Vavilov, 1926, Vavilov, 1951, greatest allelic diversity may lie in eggplant’s domestication center(s); however, the center of eggplant domestication remains yet unidentified, thus useful germplasm may remain undiscovered. Eggplant biodiversity within regions of Asia has not been sufficiently assessed, so identifying candidate domestication centers on this basis has not been possible. The goal of this project, therefore, was to explore phylogeographic patterns of eggplant in Asia to identify the domestication center(s) and the wild progenitor.
The genus Solanum (ca. 1250 species; Nee, 1999) contains many species of economic value and economic potential for both food and medicine. Most well studied food crops within Solanum are from the New World (e.g., tomato and potato). The origin of eggplant, Solanum melongena L., is Old World. S. melongena belongs to the subgenus Leptostemonum, known as the “spiny solanums.” Subgenus Leptostemonum comprises about a third of Solanum species, and half of the economically significant species in the genus. However, there have been few studies of relationships within subgenus Leptostemonum (Levin et al., 2006, Olmstead et al., 2008, Weese and Bohs, 2010), which includes approximately 500 species among 30 sections (D’Arcy, 1972, Whalen, 1984). Old World subgenus Leptostemonum includes S. melongena (section Melongena) as well as S. macrocarpon L. (section Melongena), the gboma eggplant, and S. aethiopicum L. (section Oliganthes), the scarlet eggplant; all three of these Old World species are grown as food crops. The history of the eggplant lineage may start in Africa, however relationships among African species and their Asian allies are not well resolved.
The closest wild relatives of eggplant are distributed in tropical regions throughout Africa and Asia (Hawkes and Smith, 1965). Two of these species, S. incanum L. and S. undatum Lam., are considered candidate progenitors of the domesticated eggplant. S. incanum occurs in northern Africa, the Middle East, and through India, and S. undatum is mainly found from Burma to the Pacific. These two species are cytogenetically compatible with S. melongena, occur in putative domestication centers, and are morphologically similar to domesticated eggplant (e.g., Lester and Hasan, 1990, Lester and Hasan, 1991, Karihaloo, 2009). The genetic relationship between the two candidate wild progenitors is unclear, and further, whether these species are truly wild or have experienced genetic introgression from domesticated eggplant is unknown because potentially useful phenotypic or genetic markers to delineate them have not been identified. As a result, their status as separate species has been called in to question (Weese and Bohs, 2010).
Most researchers agree that the origin of section Melongena was African, followed by radiation into Asia (reviewed in Levin et al. (2006)). Lester and Hasan (1991) suggested that S. incanum, originating in northern Africa, gave rise to S. undatum as it spread to East Asia, possibly carried by humans; they further suggested that S. undatum might be a semi-domesticated form of eggplant which, under more intense selection, eventually gave rise to our modern eggplant. Alternatively, however, it has been hypothesized that S. undatum is a true wild species, and that eggplant was domesticated directly from S. incanum in India (de Candolle, 1886, Prain, 1903). To further complicate matters, recent evidence suggests there may be differences between African and Asian S. incanum, and that these may be distinct species (Karihaloo, 2009). Phylogenetic and taxonomic analyses have been unable to resolve the relationships among eggplant and its closest relatives, thus complicating the search for the progenitor of domesticated eggplant and hampering identification of the center of origin.
There have been several molecular analyses evaluating the relationships among eggplant and closely allied species using a variety of techniques including RAPD, SAMP, sequence analysis, and others (e.g., Pearce and Lester, 1979, Karihaloo et al., 1995, Stàgel et al., 2008, Muñoz-Falcón et al., 2009, Nunome et al., 2009, Demir et al., 2010, Li et al., 2010, Tümbilen et al., 2011). These studies have been useful in assessing diversity of particular eggplant collections and for better understanding the position of eggplant within Solanum, but they have not resolved the origin and progenitor of domesticated eggplant, which remains disputed. Additionally, most of these studies include only a small number of samples from candidate domestication sites, thus eggplant diversity in Asia remains poorly understood.
Although the specific country of origin has been debated, it is generally accepted that eggplant was domesticated in tropical Asia. Vavilov (1951) proposed an Indo-Burman origin and since then, India has been proposed as a likely center of domestication (Bhaduri, 1951, Zeven and Zhukovsky, 1975, Choudhury, 1976, Martin and Rhodes, 1979, Lester and Hasan, 1991, Lester, 1998, Mace et al., 1999, Doganlar et al., 2002a, Doganlar et al., 2002b); some studies have since supported this hypothesis (e.g., Daunay, 2008, Weese and Bohs, 2010), whereas others have only supported the broader entire Indo-Burma region as a domestication center, and have suggested that eggplant was domesticated multiple times (e.g., Daunay et al., 2001, Doganlar et al., 2002a, Doganlar et al., 2002b, Daunay and Janick, 2007, Sękara et al., 2007, Wu et al., 2009, Polignano et al., 2010, Vilanova et al., 2011). If domestication occurred at the eastern end of this range, S. undatum would be a candidate progenitor (Weese and Bohs, 2010). However, if domestication occurred in India, it may have been directly from S. incanum, or indirectly via S. undatum through the more complex process described by Lester and Hasan (1991).
Ancient written records can often provide clues to the region in which a crop species was first domesticated. The oldest reliable written records for eggplant are 2000 years old, however these records describe already-domesticated eggplants, suggesting that domestication occurred even earlier. These 2000 year-old records exist in India (Charaka and Sushruta Samhitas: Bhishagratna, 1907, Sharma and Dash, 1983, Daunay and Janick, 2007) but also in southern China (Wang et al., 2008), raising the possibility that domestication occurred in China. These equally ancient writings in two different regions imply either a far more ancient domestication history and subsequent spread from one locale to the other, or possibly multiple domestication events.
We used AFLP data and a molecular phylogeny based on the nuclear ribosomal Internal Transcribed Spacer (nrITS), to clarify the phylogeographic relationships among 149 accessions representing Asian landraces, candidate progenitors (S. incanum and S. undatum), and closely related species, to gain insight into the ancestry, origins, and spread of eggplant landraces. These markers sample different levels of genetic diversity and thus have the potential to provide complementary information. We sampled in depth from among landraces from India and China, the putative domestication centers, and also included many landraces from elsewhere in Indochina and Malesia. Written records of eggplant in Asia, specimen distribution data, and accounts of the history of human migration and trade in Asia were also used to test the competing hypotheses that eggplant was domesticated in India vs. China, from a wild ancestor that was either S. incanum or S. undatum.
Section snippets
Plant materials
To investigate the history and landrace diversity of eggplant in Asia, germplasm of S. melongena and wild related species was obtained from fieldwork that included sites of human settlements and natural environments of low and high elevation, as well as from seed catalogs, and germplasm centers, with a focus on three floristic regions (Takhtajan, 1986) with high landrace diversity: Indochina (31%), South Asia (16%), and Malesia (23%). Germplasm of landraces was field-collected in China, in
Results
The aim of this study was to resolve long-standing questions about the domestication history of eggplant, including the identity of the wild progenitor, the location, and the number of domestication events. To address these issues, germplasm of local landraces was collected, the morphological diversity of wild candidate progenitors of eggplant was evaluated, and early written records of domesticated eggplant were searched to contribute information toward building a framework of the
Discussion
Our study used AFLP and nrITS sequence data from 112 accessions of eggplant landraces and candidate progenitors from throughout Asia to evaluate alternative hypotheses regarding the location of domestication and the relationships of putative wild progenitors. Several population genetic studies have focused on diversity of national traditional cultivars, and some have compared these cultivars to commercial varieties (Behera et al., 2006, Isshiki et al., 2008, Raigon et al., 2008,
Acknowledgments
The authors thank the farmers, residents, seed companies, and germplasm centers who donated seeds for this project. We received help from the faculty and staff at the Kunming Institute of Botany, The Philippines National Museum, Mariano Marcos State University, and Deccan College, as well as herbaria visited: BSI, ISBC, HK, KUN, MO, PNH. We acknowledge the help of Drs. C.L. Long, D. Tandang, L. Evangelista, D. Madulid, A. Tungpalan, and J. Sathe, as well as Y.F. Zhao and his family who hosted
References (88)
- et al.
Assessment of genetic diversity and species relationships in eggplant (Solanum melongena L.) using STMS markers
Sci. Hortic.
(2006) - et al.
ISSR variations in eggplant (Solanum melongena L.) and related Solanum species
Sci. Hortic.
(2008) - et al.
Analysis of genetic variation in eggplant and related Solanum species using sequence-related amplified polymorphism markers
Sci. Hortic.
(2010) - et al.
Comparison of eggplant landraces and commercial varieties for fruit content of phenolics, minerals, dry matter and protein
J Food Compost. Anal.
(2008) A new look at the statistical model identification
IEEE Trans. Automat. Contr.
(1974)- et al.
Molecular diversity analysis of eggplant (Solanum melongena) genetic resources
Genet. Mol. Res.
(2011) Empires of the Silk Road: A History of Central Eurasia from the Bronze Age to the Present
(2009)- et al.
Controlling the false discovery rate: a practical and powerful approach to multiple testing
J. R. Stat. Soc. Series. B Stat. Methodol.
(1995) - et al.
Genetic erosion and genetic pollution of crop wild relatives: the PGR Forum perspective and achievements
Inter-relationship of the non-tuberiferous species of SoIanum with some consideration on the origin of brinjal (Solanum melongena L.)
Indian J. Genet. Plant Breed.
(1951)