Polyploidy is common in the ginger family Zingiberaceae. The aims of the present paper are (1) to provide a general introduction on species diversity with emphasis on conservation; (2) to highlight the human-use significance of this family, focusing on the two major genera, Zingiber (ginger) and Curcuma (turmeric); (3) to present chromosome number data from 45 natural and cultivated Curcuma taxa from Thailand, of which polyploids are predominant; and (4) to describe our own work on cytotaxonomy of selected Thai Curcuma species. We obtained somatic chromosome numbers from root tips and analysed meiotic chromosome behaviour from flowers. We also used the molecular cytogenetic method of ribosomal gene mapping on chromosomes to infer mechanism of polyploidization and reveal genomic relationships among closely related species. The main results of our cytogenetic studies include the following. The most sought-after medicinal Curcuma cultivars growing on a large-scale basis are secondary triploids, so as taxa in natural habitats that are harvested for local utilisation. These triploids are sexually deficient, due to meiotic pairing abnormalities, but they are propagated asexually via rhizomes. The ribosomal mapping results indicate natural triploidization process via hybridisation, either within populations or across the species boundaries.
- medicinal plants
Taxonomic classification of the ginger family (Zingiberaceae) is still under revision for many floras, as more than 3000 species names have been used worldwide, but only half of these are accepted. These aromatic herbs grow in moist areas of the tropics and subtropics, including some regions that are seasonably dry. The ginger family comprises about 50 genera and more than 1300 species worldwide, and in Thailand 21 genera with about 200 species have been described. Numerous species are endemic to Thailand, but the majority has a wider distribution, especially over Southeast and South Asia. A few species of this family are commercially cultivated, such as ginger (Zingiber officinale Rosc.), turmeric (Curcuma longa L.) and aromatic ginger (Kaempferia galanga L.). Interestingly, these widely cultivated species are sexually deficient triploid or pentaploid plants—the elite cultivars are therefore propagated by rhizomes. These polyploid species are superior to their diploid relatives in terms of growth and yield, while the sought-after quality characters remain unchanged.
2. The ginger family (Zingiberaceae), with emphasis on Curcuma
The ginger family or Zingiberaceae comprises about 50 genera and more than 1300 species worldwide [1, 2]. The family distribution is pantropical, with centre of species diversity in South and Southeast Asia. Some species are found in America and subtropical and warm-temperate Asia. In China, 20 genera and 216 species (141 endemic, four introduced) have been recorded . Geographically, Thailand is part of the Indochinese region that harbours the highest ginger genetic resources [3, 4]. Several of these species are rare and endemic to Thailand . A large number of Thai taxa of Zingiberaceae are known as edible, ornamental or medicinal plants, from which commercial products beneficial to human can be developed.
Two best known genera in the context of cultivation and human uses worldwide are Zingiber Miller (ginger) and Curcuma Linnaeus (turmeric). The largest genus Zingiber, which comprises 100–200 species, is native to Southeast Asia especially in Thailand , China , the Indian subcontinent and New Guinea . It contains the true gingers, plants grown for their medicinal and culinary value. The best known is Z. officinale, the garden ginger.
Curcuma is a genus of about 120 accepted species in the family Zingiberaceae that contains such species as turmeric (C. longa) and Siam tulip (C. alismatifolia Gagnep.). They are native to Southeast Asia, southern China, the Indian subcontinent, New Guinea and northern Australia [6, 8, 9, 10, 11]. Tropical Asia and South Asia are the diversity hotspots of the genus. Although the species diversity is very high and new species are being discovered regularly, other species are disappearing. Habitat loss, due to global warming, deforestation, agricultural expansion and anthropogenic activities, is one of the main causes of biodiversity loss worldwide [12, 13, 14]. In addition, overharvesting for use in traditional medicine has raised a serious concern that wild plants will be disappearing from nature.
According to the IUCN Red List of Threatened Species, seven Curcuma species have been declared endangered to extinction (EN) and six additionally critically endangered (CR). The endemic EN species are C. caulina J. Graham, India ; C. colorata Valeton, Indonesia ; C. coriacea Mangaly & M. Sabu, India ; C. corniculata Skornick., Lao ; C. prasina Skornick., Thailand ; C. sahuynhensis Skornick. & N.S. Lý, Vietnam ; and C. vitellina Skornick. & H.D.Tran, Vietnam . The endemic CR species are C. bhatii (
3. Recent publications on the ginger family
The survey of recent (2019) publications on Zingiberaceae in the Web-of-Science database, using “ginger” as keyword in titles (Figure 1), shows that the genus Zingiber (ginger) is by far the most investigated worldwide. The most researched topics concern medicinal properties and health benefits of ginger (1); pharmaceutical, biochemical and molecular characterisation (2); applications in food science and chemistry (3); other technologies and industrial applications (4); as well as some effort in improving cultivation (5). On the other hand, research on Zingiber diversity, taxonomy, ecology and genetics (6) is limited.
Ginger (Z. officinale) is a very popular spice used worldwide, whether it be used to spice up meals, or as a medicine . Ginger can be used for a variety of food or medicine items, as vegetables, candy, soda, pickles and alcoholic beverages. It is one of the most versatile, ancient, significant, medicinal, nutritional herbs with several ethnomedical values. This plant is recognised due to its therapeutic properties, including antibiotic, antimicrobial, antioxidant and anti-inflammatory effects . Phenolic acids, diarylheptanoids, terpenoids and flavonoids are reported to exist in ginger rhizomes . A list of 72 gingerols and diarylheptanoids derivatives from ginger rhizomes is presented in Asamenew et al. , and among these compounds, gingerol- and shogaol-related derivatives are the principal medicinally active components contributing to the characteristic pungent flavour of ginger together with essential oil major component, zingerone. These bioactive compounds have been shown in experiments to be effective for inflammatory diseases  and osteoarthritis , to help induce apoptosis in cancer cells  and to show anti-leukaemic effect . Ginger has a great pharmaceutical potential.
Our survey of recent (2019) publications on Zingiberaceae in the Web-of-Science database, using “Curcuma” as keyword in titles (Figure 2), shows that C. longa (turmeric, saffron turmeric) is the single most researched Curcuma species. The results show that this species has received much attention in the area of pharmaceutical research and medicinal applications. Turmeric is commonly used as spice, dye, drug and cosmetics , but recent research efforts have further characterised its medicinal properties and have identified its biochemical components in high resolution and specificity. The genus Curcuma is rich in flavonoids, tannins, anthocyanin, phenolic compounds, oil, organic acids and inorganic compounds . The biological activities of Curcuma have been attributed to the non-volatile ingredients of the rhizome, cucurminoids (e.g. curcumin), as well as to the volatile terpenoids . Curcumin has been shown in experiments to have strong anti-inflammatory and antioxidant effects [29, 36]. The European Union has recommended the use of numerous medicinal plants for the treatment of gastrointestinal disorders, and among them are ginger root (Z. officinale) and turmeric root (C. longa) .
Some 23 other Curcuma species have recently been explored in search for new medicinal applications (Figure 2). The top five Curcuma species investigated are C. zedoaria (Christm.) Rosc. (white turmeric, native to South and Southeast Asia, cultivated in Thailand), C. caesia Roxb. (black turmeric, native to Northeast India, natural species of Thailand), C. xanthorrhiza Roxb. (Javanese ginger/turmeric, originated from Java island, cultivated in Thailand), C. amada Roxb. (mango ginger, originated from East India, natural species of Thailand) and C. aromatica Roxb. (fragrant turmeric, natural species of South Asia, cultivated in Thailand). Bioactive ingredients, including terpenes (more than 40 monoterpenes and sesquiterpenes), antioxidants flavonoids and phenolic compounds are present in all these species [39, 40]. Curcumin from C. zedoaria, as from C. longa, shows good anti-inflammatory effects . Zederone and zedoarondiol, from rhizomes of En-Lueang (Curcuma cf. amada), show strong cytotoxicity in a leukaemic cell line and in peripheral blood mononuclear cells, as well as having antioxidant and haemolysis properties . Dry extracts from rhizomes of C. xanthorrhiza and C. zedoaria have been shown to have anticancer and antiviral properties [42, 43, 44]. Furthermore, volatile oils extracted from leaves of C. caesia have broad antioxidant, anti-inflammatory and antimicrobial effects in vitro [45, 46]. In contrast to C. longa, many of the medicinal Curcuma species are not in large-scale cultivation, and this increases the risk of overharvesting of rhizomes from wild plants. The good news is that researchers are beginning to improve local cultivars and finding suitable methods of micropropagation of these Curcuma species, for example, C. angustifolia Roxb. .
4. The genus Curcuma in Thailand
Forty-five species are found in Thailand or almost 50% of the total species diversity of Curcuma worldwide (Table 1). At least 12 of these species are endemic to Thailand. New species have recently been described. For example, C. saraburiensis Boonma & Saensouk from Saraburi province, Central Thailand  and C. putii Maknoi & Jenjitt . Several species of Curcuma are cultivated throughout Thailand for commercial purposes. The whole plant has economic values: the above-ground part of the plant bears attractive flowers that have been exported worldwide as cut flowers, such as Siam tulip (C. alismatifolia) and C. parviflora Wall. , whereas the below-ground rhizomes are harvested and sold in local markets for use as crude extracts in the traditional medicine or for the production of certified pharmaceutical products. Medicinal species, such as C. comosa Roxb., has received much attention in recent years for being a phytoestrogen-producing plant (e.g. [51, 52]). Products from rhizomes of C. comosa have been developed for use as an anti-inflammation remedy and for treatment of uterine abnormalities and ovarian hormone deficit [53, 54].
|Species of Curcuma L.||2n chromosome number||Ploidy||Natural||Cultivated||References|
|C. aeruginosa Roxb.||63||Triploid||x||4, 6, 9|
|C. alismatifolia Gagnep.||32||x||x||9|
|C. amada Roxb.||42||Diploid||x||1, 2, 4, 9, 10|
|C. angustifolia Roxb.||42 (64)||Diploid||x||4, 6, 9, 10|
|C. aromatica Roxb.||42, 63, 86||Di-, tri-, tetraploid||x||1, 2, 3, 4, 6, 7, 9, 10|
|C. aurantiaca van Zijp||42||Diploid||x||x||4, 6, 9, 10|
|C. bella Maknoi*, K. Larsen & Sirirugsaa|
|C. bicolor J.Mood & K. Larsen||x||x|
|C. caesia Roxb.*b||63||Triploid||x||9, 10|
|C. candida (Wall.) Techapr.*c||42||Diploid||x||12|
|C. cochinchinensis Gagnep.||x|
|C. comosa Roxb.||42, 63||Di-, triploid||x||x||5, 9, 11|
|C. ecomata Craib||x|
|C. elata Roxb.*d||63||Triploid||x||4, 5, 7, 9, 11|
|C. flaviflora S.Q.Tong||42||Diploid||x||7|
|C. glans K. Larsen & J. Mood||x|
|C. glacillima Gagnep.||ca. 32||x|
|C. hermandii Gagnep.||20||x||x||9|
|C. latifolia Roxb.||63, 84||Tri-, tetraploid||x||x||4, 5, 9, 11|
|C. leucorhiza Roxb.||Triploid||x||6,|
|C. longa L.||63 (32, 48, 62-64)||Triploid||x||1, 2, 4, 6, 9, 10|
|C. maehongson C. Maknoi||x|
|C. mangga Val.||42 (63)||Diploid||x||4, 9|
|C. nakornsawan C. Maknoi||x|
|C. parviflora Wall.||28, 30, 32, 36, 42||x||9|
|C. petiolata Roxb.||42, 64||Di-, triploid||x||x||9, 10|
|C. pierreana Gagnep.||x|
|C. putii Maknoi & Jenjitt.*e|
|C. ranong C.Maknoi||x|
|C. rhabdota Sirirugsa & M. Newman||24||x||x||9|
|C. roscoceana Wall.||42||Diploid||x||x||4, 9|
|C. rubescens Roxb.||63||Triploid||x||9|
|C. rubrobracteata Skornickova, Sabu & Prasanth k.||63||Triploid||x||7|
|C. saraburi C.Maknoi||x|
|C. saraburiensis Boonma & Saensouk*f||x|
|C. singularis Gagnep.||42||x||9|
|C. sparganiifolia Gagnep.||x||x|
|C. stenochila Gagnep.||x|
|C. tak C.Maknoi||x|
|C. ubonratchani C.Maknoi||x|
|C. viridiflora Roxb.||x|
|C. woodii N.H.Xia & J. Chen*g||42||Diploid||8|
|C. xanthorrhiza Roxb.||63||Triploid||x||3, 6, 7, 9, 10|
|C. zedoaria (Christm.) Rosc.||63, 64, 84, 105||Tri-, tetra-pentaploid||x||2, 4, 6, 9, 10, 11|
|C. cf. Zedoaroides Chaveer. & Tanee*h||63||Triploid||x||11|
|Total number of taxa = 45||33||19|
We have studied C. comosa and its related species, collectively called wan-chak-motluk in Thai language for its phytoestrogen properties. The plant produces bright colourful flowers in the form of inflorescences (Figure 3a). Its rhizomes are ovoid to ovate spheroidal in shape and about 8–15 cm in diameter (Figure 3b). Wan-chak-motluk belongs to three Curcuma species: C. comosa, C. elata Roxb. and C. latifolia Rosc. . Curcuma comosa is recognised by its inflorescences with short peduncles (Figure 3a), whereas the other two species have long peduncles more suitable for flower arrangements. Curcuma elata and C. latifolia produce large and branchy rhizomes (Figure 3b).
Our chromosome number investigations have shown that the three wan-chak-motluk species can be further separated into five cultivars or cytotypes (; see also Table 1): C. comosa has two cytotypes, diploid with 2n = 2x = 42 (Figure 3d) and triploid with 2n = 3x = 63 (Figure 3f); C. elata (and C. cf. “elata-latifolia”) is triploid with 2n = 3x = 63; but C. latifolia has two cytotypes, triploid with 2n = 3x = 63 (Figure 3g) and tetraploid with 2n = 4x = 84. The group of wan-chak-motluk has been extended to cover more Thai taxa , including triploid C. caesia, triploid C. cf. zedoaroides Chaveer. & Tanee and tetraploid C. cf. zedoaria (Christm.) Rosc.
We have also recently described cytotaxonomy of the white flowering C. candida (Wall.) Techapr., to be diploid with 2n = 2x = 42 (; Figure 3c, e). C. candida is a conservation-vulnerable species, rare and endemic to the Tenasserim Range bordering Thailand and Myanmar. As this species has the potential to be developed as an ornamental or medicinal plant , efforts are being made to promote cultivation rather than harvesting it from the wild.
5. Polyploidy in Curcuma
Our studies and those of others have shown that while most Thai Curcuma species are diploid (2n = 42), other species are polyploid (Table 1). This ploidy level determination is based on the meiotic chromosome pairing in pollen mother cells, i.e. a diploid plant shows 21 bivalents, resulting from a complete synapsis of homologous chromosomes at metaphase-I of the meiotic cell division (e.g. Figure 3e). Therefore, we have concluded that the base chromosome number for Curcuma, at least the Thai species investigated, to be 21 (x = 21), but we have also identified this as “secondary” base chromosome number, possibly deriving from three times primary base number x = 7 [59, 66]. Leong-Skornickova et al.  measured genome size of 51 Indian Curcuma taxa using flow cytometry and obtained chromosome counts from about one-third of the plants. They established that the base number was x = 7 for Indian Curcuma because all the 2n numbers in their study were multiples of seven, from hexaploids (2n = 42) up to 15-ploids. This x = 7 is most likely an ancestral base number of Curcuma. Most angiosperms, woody and herbaceous, are considered being ancient polyploids with the original base numbers x = 6 and x = 7 . The major crop plants of the world are polyploid, for example, wheat, maize, potatoes, banana, cotton, oilseed rape and coffee beans, and most of these highly productive plants are ancient polyploids . Therefore, in this context, all Thai Curcuma species (Table 1) are basically (ancient) polyploids, ranging from 2n = 42 (primary hexaploid, secondary diploid) to 2n = 63 (primary 9-ploid, secondary triploid) and 2n = 84 (primary 12-ploid, secondary tetraploid). However, for the matter of consistency among our studies, we treat all Thai Curcuma taxa based on the secondary base number x = 21. This is in line with most other chromosome studies, whereby the meiotic analysis is used to determine ploidy levels, for example, the most cultivated turmeric species C. longa is triploid with 2n = 63 [60, 64].
The genus Curcuma contains chromosome numbers spanning the full range of the family Zingiberaceae, from 2n = 20 to 105 [10, 51, 59, 61, 63, 64, 70], but is characterised by chromosomes of particularly small sizes, usually less than 2 μm. A large number of Curcuma species (at least 25 species) have the diploid chromosome number 2n = 42 (base number x = 21), several (ca. 12) species have 2n = 63, and other numbers such as 20, 24, 32, 34, 84 and 105 have also been reported. Polyploidy is indeed very common in the ginger family Zingiberaceae.
Fluorescent in situ hybridization (FISH) mapping of the tandemly repeated 45S (18S–25S) ribosomal DNA on chromosomes of wan-chak-motluk supports the occurrence of triploidy among the species and cytotypes with 2n = 63 . Sets of three ribosomal FISH signals (loci) are apparent in the triploid C. comosa (Figure 3h, i) and the triploid C. elata (Figure 3j, k). In addition, the meiotic figure obtained from the triploid cytotype of C. comosa comprises of 21 trivalents; each is a pairing of three homologous chromosomes (Figure 3f). Cytogenetic characteristics of triploidy have been observed in other Curcuma species, such as C. longa  and C. zedoaria .
This triploidization is likely to be the outcome of hybridization between unreduced (2n) and normal (1n) gametes within or between the diploid populations. Such mechanism has been well documented [72, 73]. In Zingiberaceae, multiple occurrences of triploid formation have been shown in the ornamental ginger genus Globba L. from Southeast Asia, based on molecular phylogenetic analysis of both chloroplast and nuclear genes . The situation with Curcuma—wan-chak-motluk—is similar to that of Globba in that tetraploids (2n = 84) are extremely rare (Table 1) and the triploids are variable both morphologically and cytogeographically . The molecular study by Zaveska et al.  has shown that in Curcuma, the genus of palaeopolyploid origin, its evolution is most likely driven by hybridization and polyploidization.
Once a triploid has arisen, it could easily survive because Curcuma, like other genera in Zingiberaceae, reproduces predominantly by vegetative means, i.e. the plants often propagate by rhizomes and numerous bulbils produced on the inflorescence. In the context of cultivation and utilisation of wan-chak-motluk, triploid cultivars (with 2n = 63) are indeed very popular among the growers, for example, for having larger rhizomes. We have also found that triploid and tetraploid plants do have proportionally larger genome sizes compared with the diploid plants . Polyploidization in plants often increases cell size as well as growth rates, giving rise to plant phenotypes having higher physiological capacity and productivity . Increasing the ploidy level is known to be positively correlated with plant production, both biomass and yield . Furthermore, polyploids are often said to have a broader ecological tolerance than their diploid progenitors . This is thought to be due to the effects of increased heterozygosity providing metabolic flexibility to cope with wider arrays of conditions . In addition, the advantages of having more copies of the genes should allow polyploids to thrive in environments that pose challenges to their diploid progenitors . In Zingiberaceae, triploids are highly successful in cultivation, mainly due to their productive rhizomes. In natural environments, triploids may be superior as a likely result of the plant’s fitness as described above. Curcuma triploids are indeed common and widespread over a vast geographical range throughout Asia [9, 10, 62, 63, 70, 71]. Future studies combining cytogenomics, genetics, physiology and ecology should shed light onto the underlying physiological mechanism and its genetic basis of such gains in polyploidy.
The most widely cultivated plants belong to the two largest genera of this family, the ginger genus (Zingiber) and the turmeric genus (Curcuma). They are also the best researched plants from this family, and the most researched topics concern medicinal properties and health benefits, pharmaceutical, biochemical and molecular characterisation, as well as applications in food science and technology. The present study identifies numerous polyploid species in the turmeric genus (Curcuma) from Thailand. In particular, triploid species and/or cultivars are popular for a large-scale cultivation. The plants are easily propagated via underground rhizomes, which are also the part of the plant that contains bioactive compounds with medicinal properties. Rhizomes of triploid cultivars are bigger than those of the diploid, wild relatives and thus are more economically valuable. Triploids are also the most adaptable plants in diverse environments. On the other hand, the overharvesting of wild plants, in search for novel or better bioactive compounds, poses a serious risk of species extinction. Cytogenetic research, such as that presented here, can provide useful information for both types of activities, i.e. in the plant improvement for cultivation and in the conservation of natural biodiversity.
This work was supported by Mahidol University and University of Iceland. We appreciate the contribution in taxonomic identification of plant materials from Dr. Thaya Jenjittikul of Mahidol University. We thank both Dr. Tidarat Puangpairote from Prince of Songkla University and PhD student of Mahidol University, Nattapon Nopporncharoenkul, for their accurate cytogenetic work on Zingiberaceae of Thailand.