Plant Syst Evol (2011) 292:41–49 DOI 10.1007/s00606-010-0408-9
ORIGINAL ARTICLE
Ribosomal FISH mapping reveals hybridity in phytoestrogen producing Curcuma species from Thailand Puangpaka Soontornchainaksaeng Kesara Anamthawat-Jo´nsson
•
Received: 3 December 2009 / Accepted: 20 December 2010 / Published online: 19 January 2011 Ó Springer-Verlag 2011
Abstract Species in the genus Curcuma (Zingiberaceae) that are cultivated widely in Thailand for their phytoestrogen-producing rhizomes are called wan-chak-motluk. Five cultivars belonging to Curcuma comosa (cultivars with 2n = 42 and 63) and Curcuma elata (2n = 63) were examined using the molecular cytogenetic method of fluorescence in situ hybridisation (FISH) in order to identify genetic relationships among these cultivars based on chromosomal maps of the 18S–25S ribosomal loci. The results revealed hybrid features in this Curcuma species group and a significant similarity among wan-chak-motluk cultivars. The main features included: (1) the presence of the single largest ribosomal site, assigned the Cc1 marker site, in the somatic 2n complement of all cultivars, and (2) the odd numbers of ribosomal sites in the complements, most often in sets of three. We therefore propose that the cultivar with 2n = 42 (C. comosa) is a homoploid hybrid species comprised of two different ancestral genomes and has a diploid status with the basic chromosome number x = 21. The cultivars with 2n = 63 (C. comosa and C. elata) are most probably triploids arising within the 2n = 42 diploid species/cultivars via a meiotic modification, rather than from hybridisation between diploid and tetraploid plants. The knowledge about genetic and genomic relationships among wan-chak-motluk cultivars will be
P. Soontornchainaksaeng (&) Department of Plant Science, Faculty of Science, Mahidol University, Rama VI, Bangkok 10400, Thailand e-mail:
[email protected] K. Anamthawat-Jo´nsson Institute of Biology, University of Iceland, Askja, Sturlugata 7, 101, Reykjavı´k, Iceland
important in the research projects that aim to explore and promote new plant materials for cultivation. Keywords Chromosomes Curcuma C. comosa C. elata Fluorescence in situ hybridisation (FISH) 18S–25S ribosomal gene (rDNA)
Introduction Species in the genus Curcuma L. (Zingiberaceae) that are cultivated widely in Thailand for their phytoestrogen-producing rhizomes are called wan-chak-motluk. Rhizomes of wan-chak-motluk, which are ovoid to ovate-spheroidal in shape and about 8–15 cm in diameter, have long been used in Thai traditional medicine for treatment of uterine abnormalities and ovarian hormone deficit. Products from cultivars of wan-chak-motluk are extensively used for this purpose, especially those belonging to the species Curcuma comosa Roxb. Estrogenic activities of wan-chak-motluk have been detected in rhizome extracts of C. comosa and in several diarylheptanoids isolated from rhizomes of this species (Suksamrarn et al. 2008; Winuthayanon et al. 2009). These diarylheptanoids and rhizome extracts from C. comosa have also been shown to have anti-inflammatory properties (Sodsai et al. 2007; Thampithak et al. 2009). Other potential properties of C. comosa rhizome extracts have been documented: for example, they have been used as nematocidal agents (Jurgens et al. 1994) and have been shown to reduce plasma cholesterol in hamsters (Piyachaturawat et al. 1999). Curcuma comosa from Thailand has received much attention recently for its naturally producing metabolites. Numerous new compounds from its rhizome have been isolated and described, including monoterpenes (Nakamura et al. 2008) and sesquiterpenes (Qu et al. 2009;
123
P. Soontornchainaksaeng, K. Anamthawat-Jo´nsson
42
Xu et al. 2009). Terpenoids from Curcuma are known to be important in the perfumery and flavour industries (Nahar and Sarker 2007). Due to the medicinal properties of C. comosa, wan-chakmotluk is widely cultivated in Thailand for economic purposes. Wan-chak-motluk in cultivation is extremely variable, both in rhizome morphology and overall plant appearance. This has often resulted in wrong taxonomic assignment of cultivars—the feature which is used by local growers to indicate medicinal properties. Recent botanical and cytotaxonomical study aiming to clarify species and cultivars of wan-chak-motluk collected from major cultivation sites in 16 provinces in Thailand (Soontornchainaksaeng and Jenjittikul 2010) revealed that wan-chakmotluk belongs to three Curcuma species: C. comosa, C. elata Roxb., and C. latifolia Rosc. The characters that can differentiate these species are peduncle length (long or short), lower leaf surface (pubescent or glabrous) and a red patch along the leaf midrib (present or absent). Curcuma comosa, which produces lateral inflorescences with short peduncle, is further divided into two morphologically different cultivars: one with the 2n chromosome number of 42 and the other 63. The cultivars with 2n = 63 have bigger inflorescence and produce larger rhizome than the cultivars with 2n = 42. The other two species, i.e. C. elata and C. latifolia, are botanically identified as having inflorescences with long peduncle. The two species differ in that the leaf of C. latifolia has a red patch along the midrib, which is absent in the leaf of C. elata. Curcuma elata has 2n = 63 and is a cultivar of its own. On the other hand, C. latifolia consists of two chromosomal cultivars, with 2n = 63 and 84. These two species of wan-chak-motluk, C. elata and C. latifolia, have not been systematically explored for their medicinal properties. Chromosome number and genome size are known to be good supportive markers for delimitation of plant species (Guerra 2008), especially species with small chromosomes like Curcuma (Leong-Skornickova et al. 2007). But the situation is more complex in wan-chak-motluk (Soontornchainaksaeng and Jenjittikul 2010). Different wan-chakmotluk species have the same chromosome number 2n = 63, yet only C. comosa has been used in Thai traditional medicine and pharmaceutical research. Different chromosomal cultivars are also found within the same species: for example, C. comosa consists of two cultivars with 2n = 42 and 63, and C. latifolia includes cultivars with 2n = 63 and 84. To our knowledge there is no comparative analysis of biochemical properties among wanchak-motluk species or between cultivars of the same species. It is reasonable to assume that such properties have a genetic basis and therefore it is important to resolve genetic and evolutionary relationships in wan-chak-motluk. Species with a larger genome (higher ploidy) generally
123
have larger cells than species with a smaller genome, and hence they are likely to produce more secondary metabolites and sought-after compounds. The Curcuma species that have been extensively investigated for their phytochemical properties are C. longa L. (turmeric), C. xanthorrhiza Roxb. and C. zedoaria (Christm.) Rosc. (Nahar and Sarker 2007). All of these species are polyploid with 2n chromosome number of 63. The aim of the present study was to compare wan-chakmotluk cultivars with different chromosome numbers (2n = 42 and 63) belonging to the same species and to examine if different cultivars having the same chromosome number (2n = 63) were differentiable using a molecular cytogenetic approach. Fluorescence in situ hybridisation (FISH) method was used in this study to localize the 18S– 25S ribosomal genes on chromosomes of wan-chak-motluk cultivars belonging to two Curcuma species: C. comosa and C. elata from Thailand. This approach should also be able to resolve some of the unknown aspects of this polyploidy, for example if the species were auto- or allopolyploids, and how they were related genomically. A number of repetitive sequences, both highly conserved sequences (e.g. 18S–25S and 5S ribosomal genes and telomeres) and fast evolving species-specific sequences (e.g. pericentromeric sequences, mobile elements, heterochromatic and satellite sequences) have been used as chromosome landmarks to study karyotype evolution in numerous plant species, for example Arabidopsis (Siroky 2008), cereals and grass species (Contento et al. 2005; Anamthawat-Jo´nsson and Bo¨dvarsdo´ttir 2001; Anamthawat-Jo´nsson et al. 2009), orchids (Begum et al. 2009) and tree species (Chokchaichamnankit et al. 2008). The knowledge about genetic and evolutionary relationships among taxonomically related wan-chak-motluk cultivars will be important in the botanical, biochemical and pharmaceutical research projects that aim to explore and promote new plant materials for cultivation.
Materials and methods Plant materials Plant materials used in this study were part of the living plant collection established at Mahidol University, Salaya Campus (by P. Soontornchainaksaeng and T. Jenjittikul), from rhizomes of wan-chak-motluk collected in cultivated sites in 16 provinces throughout Thailand, 10–20 plants from each province. Taxonomic identification of these accessions, together with their chromosome numbers, was described (Soontornchainaksaeng and Jenjittikul 2010). Samples used in the present study were individual plants belonging to two Curcuma species, i.e. C. comosa with the
Ribosomal FISH mapping in Curcuma
43
2n chromosome numbers of 42 and 63, and C. elata with 2n = 63 (Table 1). Chromosome preparation For the FISH experiments, chromosome preparation was performed using either an enzymatic squash method similar to that described in Anamthawat-Jo´nsson and HeslopHarrison (1995) or a protoplast dropping method as in Anamthawat-Jo´nsson (2003). Root tips and young shoot buds were collected from the germinating rhizomes grown in pots. For the pre-treatment, the root tips were collected in ice water to arrest metaphases, whereas the young buds or shoot tips were placed directly in saturated paradichlorobenzene solution. They were pre-treated for 26 h in a refrigerator. The materials (both root tips and buds) were then fixed in a 3:1 mixture of glacial acetic acid and 96% ethanol for at least 2 h at room temperature. With the enzymatic squash method, the pre-treated or fixed materials were rinsed for 20 min at room temperature in a citrate buffer containing 4 mM citric acid monohydrate and 6 mM trisodium citrate dihydrate. The materials were then digested for 1 h at 37°C in an enzyme mixture containing 800–1,000 U of cellulase (Sigma, from Aspergillus niger) and 500–650 units of pectinase (Sigma, also from A. niger) prepared in the same citrate buffer as used for rinsing. After the enzymatic digestion, the materials were rinsed again in the citrate buffer. Each root tip or bud was trimmed to about one mm in size and dissected for the meristem, which was then squashed in 45% acetic acid on an acid-cleaned microscopic slide. The coverslips were removed after freezing the slides on dry ice or in liquid nitrogen. The chromosome spreads were evaluated under an interference microscope. For the protoplast dropping method used in this study, the materials (buds or shoot tips) were trimmed down to about 5 mm in size by removing outermost leaves from the youngest part containing meristems. The buds were then rinsed for 20 min in distilled water, after which each bud was digested for 2–3 h at room temperature in a 100 ll aliquot of protoplast enzyme mixture containing 500 units of cellulase
(Onozuga R10, Merck) and 280 units of Pectinase (Sigma) in a buffer containing 75 mM KCl and 7.5 mM EDTA, pH 4. The digested bud was minced gently in a microtube using a pipette tip to break up large pieces and to release protoplasts into suspension. The suspension was filtered through a nylon mesh to remove tissue debris and then treated with hypotonic solution (1.5 ml of cold 75 mM KCl) for 10 min at room temperature. The suspension was then centrifuged at 5,000g for 5 min to collect the cells (protoplasts), while the supernatant was discarded. The protoplasts were resuspended in 1 ml of fresh fixative and spun down; this step was repeated twice. The final protoplast suspension, in ca. 50 ll of fresh fixative, was dropped onto ice-cold, wet, acidcleaned microscope slides, one drop per slide, and air-dried. The chromosome spreads were examined under a phase contrast or an interference microscope. Additional plant materials from the same species as used in FISH experiments (Table 1), both root/shoot tips for mitosis and newly developed flower buds for meiosis, were prepared using standard Feulgen squash methods. The mitotic chromosomes were prepared as described in Soontornchainaksaeng and Jenjittikul (2010), whereas the meiotic materials were prepared according to Soontornchainaksaeng et al. (2003). Root tips and young shoot buds were pre-treated in saturated paradichlorobenzene solution for 6 h, followed by hydrolysis in 1 N HCl at 60°C for 10 min, stained and squashed in aceto-orcein. Young flower buds were fixed in Canoy’s solution for 24–48 h, after which they were stained and squashed in propionocarmine. The chromosomes were examined under 1,0009 magnification in Olympus BX50 light microscope. Wellspread metaphases were photographed. The mitotic chromosome numbers and the meiotic pairing were determined in the microscope and from the photographs, from at least ten slides each plant. Fluorescence in situ hybridisation The FISH procedure as conducted in this study was a modification of Chokchaichamnankit et al. (2008). The plasmid clone pTa71 (Gerlach and Bedbrook 1979), a 9-kb
Table 1 Plant materials examined in this study, their origin and identification Plant identity
Province name
Province location
Species
Chromosome number (2n)
Method of study
KB-803-1
Kanchanaburi
Central Thailand
C. comosa
42
rDNA-FISH
KB-803-2
Kanchanaburi
Central Thailand
C. comosa
63
rDNA-FISH
KB-5201
Kanchanaburi
Central Thailand
C. comosa
42
Feulgen-staining (meiosis)
PB-2
Phetchaboon
Northern Thailand
C. elata
63
rDNA-FISH
PR-202
Prae
Northern Thailand
C. elata
63
rDNA-FISH
SK-102
Sakon Nakhon
North-eastern Thailand
C. elata
63
rDNA-FISH
PK-341
Prachuap Khirikhan
Central-southern Thailand
C. elata
63
Feulgen-staining (mitosis)
123
44
fragment from wheat which contained a part of 18S and the entire 5.8S and 25S coding region, was used as an 18S–25S rDNA probe. The rDNA probe was labelled with the red label Spectrum Red-dUTP (Vysis) or the green label Fluorescein-12-dUTP (Roche Applied Science), by standard nick translation. The labelled probes were purified through ProbeQuant G-50 Micro Column (GE Healthcare) following the manufacturer’s protocol. Before performing FISH experiments, the chromosome preparations were first treated with fresh fixative for 10 min at room temperature, washed twice with 96% ethanol and air-dried. The preparations were then treated with RNase-A (5 lg/ml) for 1 h at 37°C, washed twice in 29 SSC buffer (3 M NaCl and 30 mM trisodium citrate, pH 7), after which they were treated with paraformaldehyde (4%, w/v) for 20 min at room temperature. The slides were then washed briefly in 29 SSC, dehydrated in a series of 70, 90 and 96% ethanol and air-dried. In FISH experiments, a 20 ll of pre-boiled probe mixture containing 50 ng of the labelled rDNA probe, 50% formamide, 20% dextran sulphate, 29SSC and 0.5% SDS was applied to each chromosome preparation and a coverslip placed over. The probe and the slide were denatured together at 87–88°C for 10 min in a Hybaid FISH thermocycler, after which the DNA:DNA in situ hybridisation was allowed to take place in a moist chamber overnight at 37°C. The post-hybridisation washing steps included a stringent wash in 0.19 SSC at 60°C for 15 min, a brief wash in pre-warmed 29 SSC and then equilibration in 49 SSC with 0.2% Tween 20 at room temperature. The chromosomes were stained for 1 min with a 1 lg/ml solution of fluorochrome DAPI (4, 6-diamidino-2-phenylindole), washed briefly with distilled water and mounted with the antifade Citifluor AF1 (Citifluor Ltd, UK). The FISH signal on chromosomes was examined under 1,0009 magnification in the Olympus BX50 epifluorescence microscope using appropriate filters and the images captured with the Olympus DP50 digital camera.
Results FISH mapping of the tandemly repeated 18S–25S ribosomal DNA on chromosomes of wan-chak-motluk cultivars revealed hybridity in the genome of this Curcuma species group. All five wan-chak-motluk cultivars belonging to two Curcuma species, C. comosa and C. elata, showed a similar pattern of the rDNA localisation on chromosomes and in interphase nuclei (Fig. 1). Each cultivar was subjected to two FISH experiments and the results were drawn from images of 6–19 cells each. Chromosomes of wan-chak-motluk species in this study were very small—the metaphase chromosomes were on
123
P. Soontornchainaksaeng, K. Anamthawat-Jo´nsson Fig. 1 Chromosome preparations and fluorescence in situ hybridisa- c tion (FISH) of the 18S–25S rRNA gene to chromosomes of Curcuma comosa (a–h) and C. elata (i–w). Red and green fluorescently labelled rDNA probes were used in the FISH experiments. The scale bar represents approximately 2 lm. Note that cells shown in this figure are only examples of the cells analysed in this study. Wan-chakmotluk cultivar KB803-1, Curcuma comosa with 2n = 42 (a–c): the somatic 2n number of 42 is shown in (a), whereas b shows the typical rDNA map on metaphase chromosomes comprising the Cc1 marker site (arrowed) and six smaller sites, probably in two sets of three homologous/homoeologous sites. A meiotic metaphase of this cultivar shows 21 bivalents (c). The cultivar KB803-2, Curcuma comosa with 2n = 63 (d–h): the somatic 2n number of 63 is shown in (d). This cultivar also shows the marker site Cc1 (e, arrowed). Other sites were variably detectable depending on the local stringencies of FISH experiments—a total of three major sites can be seen in (d), whereas 3–5 major sites and 3–4 minor sites can be seen in these examples of interphases (f–h). Wan-chak-motluk cultivars belonging to C. elata (i–w): the somatic complement 2n = 63 characteristics of C. elata is shown in (i). The cultivar SK102 (j–m) also has the Cc1 marker site (arrowed) in metaphase (j–k), whereas three major and four minor ribosomal sites can be seen in these interphases (l–m). The cultivar PB2 is seen here (n–r) having 3–5 major and 1–4 minor rDNA sites. The cultivar PR202 (s–w) clearly show the typical three major sites in most cases (s–t), but the overall range includes 3–4 major and 3–4 minor ribosomal sites
average about 1 lm long. As expected the two different cultivars of C. comosa from the same accession, Kanchanaburi (KB803), showed different somatic chromosome numbers 2n = 42 (Fig. 1a) and 2n = 63 (Fig. 1d), whereas C. elata had 2n = 63 (Fig. 1i). At the first meiotic metaphase, the cultivar of C. comosa with 2n = 42 formed 21 bivalents (Fig. 1c). The somatic chromosome number of all accessions under study was determined from the standard Feulgen-squash preparations (Table 1). In C. comosa with 2n = 42 (cultivar KB803-1), the most common number of 18S–25S ribosomal sites was found to be 7 (Fig. 1b). The rDNA site which showed the strongest FISH signal was a single site, i.e. not present in a homologous pair, and therefore was designated Cc1 (arrowed in Figs. 1a, b). This enlarged site appeared to occupy the whole short arm of a sub-metacentric chromosome. All of the other sites were also short-arm submetacentric and seemed to be localized in two sets of three morphologically similar (possibly homologous or homoeologous) chromosomes (Fig. 1b). The cultivar of C. comosa with 2n = 63 (cultivar KB803-2, Fig. 1e–h) also showed odd numbers of the rDNA loci in most cases, including 3, 5 and 7. In a typical metaphase (Fig. 1e), the largest site (arrowed) appeared to be the same as the Cc1 site in the cultivar with 2n = 42. It occupied the whole short arm of a sub-metacentric chromosome and had the typical double-knob morphology of this site (compare Fig. 1b, 1e). Two other major sites were apparent in the metaphase shown (Fig. 1e). The number of FISH signals in the interphases of this cultivar was 3–7,
Ribosomal FISH mapping in Curcuma
with the most common pattern being 1 ? (1–3) ? (1–3) based on the signal strength. The smallest (minor) sites were not always detectable, as that depended on the FISH experimental conditions. In general the 18S–25S ribosomal gene maps in both cultivars of C. comosa were very similar.
45
The 18S–25S ribosomal gene maps of C. elata cultivars were not very different from those of C. comosa. The single most enlarged rDNA site similar to the Cc-1 locus could be seen for example in the metaphase of cultivar SK102 (Fig. 1j–k, indicated with an arrow). Other sites in this cell were not detectable, but the FISH signal was weak
123
46
in this particular experiment. Based on FISH signals in the interphases, the cultivar SK102 is probably most similar to C. curcuma in that it had 3–8 ribosomal sites (Fig. 1l, m), with the most common pattern (1–2) ? (1–3) ? (1–3). The other two cultivars of C. elata were more divergent from C. comosa. The cultivar PB2 (Fig. 1n–r) had 5–8 rDNA sites including 3–5 major and 1–3 minor sites, whereas the cultivar PR202 (Fig. 1s–w) generally had 5–9 sites comprising 3–4 major and 3–4 minor sites. This particular cultivar in fact differed in rhizome morphology from other C. elata cultivars. The interphase results are very consistent within cultivars. We did not see, for example, fusion of NORs (nucleolar organizing regions) or separation of hybridization loci from within the same NOR. In fact there was very little of nucleolus-forming expression of NOR in these nuclei, and this has made the interphase interpretation straightforward.
Discussion Wan-chak-motluk cultivar with 2n = 42 (Curcuma comosa) The rDNA-FISH mapping results indicate that the phytoestrogen-producing C. comosa with 2n = 42 is a hybrid species. Ribosomal sites in this cultivar are in most cases odd in number (7 sites shown in Fig. 1b), with the Cc1 marker site being non-homologous. First of all, the odd number of 18S–25S ribosomal sites can indicate that the species derived from hybridisation of ancestral species with different genomes, which had different number of ribosomal loci. A clear case of hybridity has been shown in Crocus (Orgaard et al. 1995), whereby unequal number of nucleolar chromosome pairs in the parental genomes resulted in an odd number of ribosomal sites in the hybrid cultivar. In that study, the rDNA-FISH mapping was performed on the same cells where parental genomes were discriminated unequivocally using the GISH (genomic in situ hybridization) method of Anamthawat-Jo´nsson et al. (1990). We plan to use the same approach with Curcuma cultivars to confirm the parental origin of all nucleolar chromosomes. Wan-chak-motluk in the present study could also be non-hybrid having polymorphic NORs, as has been observed for example in wild species of Allium (Sharma and Gohil 2008). However, there was no further evidence for or against hybrid nature of the Allium materials. Secondly, the odd number of major ribosomal sites in this wan-chak-motluk is consistently due to the presence of the single enlarged Cc1 marker site. The single Cc1 site presumably belongs to one of the two ancestral genomes in this hybrid cultivar, deriving as the result of extensive
123
P. Soontornchainaksaeng, K. Anamthawat-Jo´nsson
amplification at one NOR of the homologous pair, or from a fusion of NORs via translocation or other means of chromosomal rearrangement. The double-knob appearance of the Cc1 site at metaphase (see Fig. 1b, e) could support the latter explanation, i.e. due to chromosomal changes. In Allium species showing substantial karyotypic variation, the nucleolar chromosomes are the most unstable members of the complements and these chromosomes exhibit variations both in number as well as structure (Stevens and Bougourd 1991; Sharma and Gohil 2008). In this wan-chak-motluk cultivar, the other two sets of three rDNA sites could have come from one or both parental species. The most likely scenario is that two different species, each with 2n = 42, hybridised and as a result, two n = 21 gametes combined again into 2n = 42 in the hybrid, a homoploid hybrid. Homoploid hybrid speciation (Rieseberg 1997) is the process by which an independent lineage arises through hybridisation and the combination of parental genomes, but without an increase in ploidy. This has been reported to occur in a number of plant groups, for example Helianthus sunflower (Gross and Rieseberg 2005; Buerkle and Rieseberg 2008), Iris (Arnold 1992), Scaevola (Howarth and Baum 2005) and Quercus oak (Mir et al. 2006). A homoploid hybrid species could become established in a relatively short time if the hybrids are reproductively isolated from the parental types, for example with karyotypic divergence and genomic rearrangements, or by genetic means of incompatibility (Rieseberg and Willis 2007). A homoploid hybrid species can become established in new ecological niches, or the hybrids may be more adaptive and have high fitness which allows them to increase in abundance. In the case of Curcuma in the present study, wan-chak-motluk cultivars (genotypes) are cultivated on a large scale via vegetative (rhizome) propagation. Growers might have selected and preferred homoploid hybrids, either based on desirable agronomic features or medicinal properties. This C. comosa cultivar with 2n = 42 is therefore considered as being a homoploid hybrid species/cultivar with diploid status if based on the basic chromosome number x = 21 proposed by Soontornchainaksaeng and Jenjittikul (2010) for wan-chak-motluk. This basic number has been inferred in a number of studies of Curcuma from Bangladesh, Indonesia and elsewhere (e.g. Ramachandran 1961; Prana et al. 1978). However, rDNA-FISH in the present study also supports the basic number x = 7. The number of 18S–25S ribosomal sites in this wan-chakmotluk cultivar tends to be in sets of three, indicating that x = 21 is possibly a secondary basic number deriving from three times x = 7. Leong-Skornickova et al. (2007) 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 basic
Ribosomal FISH mapping in Curcuma
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. There were no ploidy levels found below 69 except for one unique diploid species with the basic number x = 11. This means that x = 7 in Curcuma is most likely an ancestral basic number of this genus. Stebbins (1971) explained that basic numbers of modern genera were derived by ancient polyploidy, and that the original basic numbers of angiosperms, both woody and herbaceous, were x = 6 and x = 7. We therefore propose that wan-chak-motluk C. comosa with 2n = 42 is a diploid species based on the ‘‘secondary’’ basic number x = 21. Meiotic analysis of this material has indeed confirmed the diploid status with 21 bivalents (Fig. 1c). Ramachandran (1961) observed regular bivalent formation during meiosis in C. decipiens (2n = 42), but a high frequency of trivalents in C. longa (C. domestica) with 2n = 63. Wan-chak-motluk cultivars with 2n = 63 (C. comosa and C. elata) The ribosomal FISH mapping reveals significant similarity between wan-chak-motluk cultivars with 2n = 42 and 63. The single marker site Cc1 characteristics of C. comosa cultivar with 2n = 42, is also present in the cultivars with 2n = 63 from both C. comosa and C. elata (see metaphases in Fig. 1e, k). This is an evidence which indicates that cultivars with 2n = 63 have originated from the diploid cultivar(s) with 2n = 42. The odd numbers of 18S–25S ribosomal sites in wan-chak-motluk cultivars with 2n = 63 also tend to be in sets of three, similar to that in the 2n = 42 cultivar, supporting the ancestral-descendant relationship. Wan-chak-motluk cultivars with 2n = 63 are therefore considered here to be triploids, probably arising from independent events of triploid formation via meiotic non-disjunction and fusion of reduced and unreduced gametes in the diploid cultivar(s). Characteristics of triploidy have been observed in other Curcuma species. Ramachandran (1961) reported a high frequency of trivalents in meiosis of C. longa (C. domestica) with 2n = 63, whereas Islam et al. (2007) reported three (homologous) satellite chromosomes in C. zedoaria with 2n = 63 from Bangladesh. As the ribosomal FISH mapping also reveals a distinct variation among the triploid cultivars (2n = 63) in the present study, the triploids must have had different evolutionary histories. The diploid and triploid cultivars of C. comosa (KB803-1 and KB803-2 respectively) show almost the same rDNA-FISH patterns, therefore this triploidization has probably arisen within C. comosa and at the same cultivation site (Kanchanaburi, central Thailand). The other triploid cultivars are members of a different
47
Curcuma species, i.e. C. elata, and the samples came from other geographical regions. Based on the rDNA–FISH maps, the cultivar SK102 from Sakon Nakorn (northeastern Thailand) appears to be more similar to the triploid C. comosa, whereas triploid cultivars of C. elata from northern Thailand (PB2 and PR202) are more distant from this C. comosa, although they are relatively similar to one another. Future studies of wan-chak-motluk should include a much larger sample size as well as greater taxonomic and cytogeographic diversity of the samples. Significant variation in the genome size (most probably in the heterochromatin content) was detected among different geographic populations of triploid C. zedoaria from Bangladesh (Islam et al. 2007). Based on genome size variation, Leong-Skornickova et al. (2007) confirmed by correlation analysis a link between genome size and geographical location of Indian Curcuma taxa. Such correlation is more likely to be due to different evolutionary histories rather than ecological conditions as interpreted in some papers. Multiple occurrences of triploid formation have been shown in the ornamental ginger genus Globba (Zingiberaceae) from SE Asia, based on molecular phylogenetic analysis of both chloroplast and nuclear genes (Takano and Okada 2002). In this study the triploids were scattered into several clades, clearly indicating multiple occurrences of triploids and that such triploid formation was an important force for speciation of Globba. Furthermore, certain diploid and triploid individuals shared the same sequences, whereas tetraploid species were rare and phylogenetically distant from the triploids. The triploid individuals in Globba might be derived from diploid relatives by triploidization, e.g. from the fertilisation between unreduced (2n) and reduced (n) gametes, rather than from hybridisation between diploid and tetraploid species. The latter mechanism is known to be more common among higher plants, especially where the species complex contains both diploids and tetraploids (Grant 1982). The situation with wanchak-motluk is similar to that of Globba (Takano and Okada 2002) in that tetraploids (2n = 84) are extremely rare and the triploids are variable both morphologically and cytogeographically (Soontornchainaksaeng and Jenjittikul 2010). The triploidization in wan-chak-motluk (Curcuma) has almost certainly happened the same way as in the ornamental ginger Globba. 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. Furthermore, Curcuma triploids may also be able to reproduce sexually, which could be the reason behind the formation of high-ploid cultivars, as postulated by Leong-Skornickova et al. (2007) for the formation of Curcuma taxa with 2n = 84 and 105.
123
48
The occurrence of tetraploid (2n = 84) wan-chak-motluk cultivated in Thailand could be explained the same way. Among several plant species, triploids are known to be fertile by means of apomixis (Stebbins 1971), whereby their meiosis is modified in such a way that viable, unreduced gametes are also produced. Our future studies will include a detailed analysis of meiosis in these Curcuma cultivars. At this stage we want to be able to initially sort out accessions of wan-chak-motluk that are most widely cultivated by this simple cytotaxonomical approach. The result of this paper has an immediate application because right now the growers only rely on rhizome morphology. On the scientific side, our current/future work includes the generation of detailed and accurate rDNA–FISH maps (of both 5S and 18S–25S ribosomal genes) using materials from all major cultivars grown commercially and then we will be able to dissect the species and genomic relationships more meaningfully. Acknowledgments This work was supported by Mahidol University, the National Research Council of Thailand, the Office of the Higher Education Commission of Thailand and the University of Iceland. We are grateful to Prof. Pawinee Piyachaturawat of Mahidol University for valuable comments on the project. We thank Dr Thaya Jenjittikul of Mahidol University for her contribution in sample collection and taxonomic identification of plant materials and Dr Ploenpit Chokchaichamnankit of Chulalongkorn University for assistance with the optimization of molecular cytogenetic protocols. We also thank the staff of the Cytogenetic Research Laboratory of the Botany Department, Mahidol University, for their assistance, especially Jatuporn Chandrmai and Tidarat Puangpairote.
References Anamthawat-Jo´nsson K (2003) Preparation of chromosomes from plant leaf meristems for karyotype analysis and in situ hybridization. Met Cell Sci 25:91–95 Anamthawat-Jo´nsson K, Bo¨dvarsdo´ttir SK (2001) Genomic and genetic relationships among species of Leymus (Poaceae: Triticeae) inferred from 18S.26S ribosomal genes. Amer J Bot 88:553–559 Anamthawat-Jo´nsson K, Heslop-Harrison JS (1995) Molecular cytogenetics of Icelandic birch species: physical mapping by in situ hybridization and rDNA polymorphism. Can J Forest Res 25:101–108 Anamthawat-Jo´nsson K, Schwarzacher T, Leitch AR, Bennett MD, Heslop-Harrison JS (1990) Discrimination between closely related Triticeae species using genomic DNA as a probe. Theor Appl Genet 79:721–728 Anamthawat-Jo´nsson K, Wenke T, ÆTh Tho´rsson, Sveinsson S, Zakrzewski F, Schmidt T (2009) Evolutionary diversification of satellite DNA sequences from Leymus (Poaceae: Triticeae). Genome 54:381–390 Arnold ML (1992) Natural hybridization as an evolutionary process. Annu Rev Ecol Syst 23:237–261 Begum R, Alam SS, Menzel G, Schmidt T (2009) Comparative molecular cytogenetics of major repetitive sequence families of
123
P. Soontornchainaksaeng, K. Anamthawat-Jo´nsson three Dendrobium species (Orchidaceae) from Bangladesh. Ann Bot 104:863–872 Buerkle CA, Rieseberg LH (2008) The rate of genome stabilization in homoploid hybrid species. Evolution 62:266–275 Chokchaichamnankit P, Anamthawat-Jo´nsson K, Chulalaksananukul W (2008) Chromosomal mapping of 18S–25S and 5S ribosomal genes on 15 species of Fagaceae from northern Thailand. Silvae Genet 57:5–13 Contento A, Heslop-Harrison JS, Schwarzacher T (2005) Diversity of major repetitive DNA sequence in diploid and polyploidy Triticeae. Cytogenet Genome Res 109:34–42 Gerlach WL, Bedbrook JR (1979) Cloning and characterization of ribosomal RNA genes from wheat and barley. Nucleic Acids Res 7:1869–1885 Grant V (1982) Periodicities in the chromosome numbers of the angiosperms. Bot Gaz 143:379–389 Gross BL, Rieseberg LH (2005) The ecological genetics of homoploid hybrid speciation. J Hered 96:241–252 Guerra M (2008) Chromosome numbers in plant cytotaxonomy: concepts and implications. Cytogenet Genome Res 120:339–350 Howarth DG, Baum DA (2005) Genealogical evidence of homoploid hybrid speciation in an adaptive radiation of Scaevola (Goodeniaceae) in the Hawaiian Islands. Evolution 59:948–961 Islam MA, Meister A, Schubert V, Kloppstech K, Esch E (2007) Genetic diversity and cytogenetic analysis in Curcuma zedoaria (Christm.) Roscue from Bangladesh. Genet Resources Crop Evol 54:149–156 Jurgens TM, Frazier EG, Schaeffer JM, Jones TE, Zink DL, Borris RP, Nanakorn W, Beck HT, Balick MJ (1994) Novel nematocidal agents from Curcuma comosa. J Nat Prod 57:230–235 Leong-Skornickova J, Sida O, Jarolimova V, Sabu M, Fer T, Travnicek P, Suda J (2007) Chromosome numbers and genome size variation in Indian species of Curcuma (Zingiberaceae). Ann Bot 100:505–526 Mir C, Toumi L, Jarne P, Sarda V, Di Giusto F, Lumaret R (2006) Endemic North African Quercus afares Pomel originates from hybridisation between two genetically very distant oak species (Q. suber L. and Q. canariensis Willd.): evidence from nuclear and cytoplasmic markers. Heredity 96:175–184 Nahar L, Sarker SD (2007) Phytochemistry of the genus Curcuma. In: Ravindran PN, Babu KN, Sivaraman K (eds) Turmeric: the genus Curcuma. CRC Press, USA, pp 71–106 Nakamura S, Qu Y, Xu F, Matsuda H, Yoshikawa M (2008) Structures of new monoterpenes from Thai herbal medicine Curcuma comosa. Chem Pharm Bull 56:1604–1606 Orgaard M, Jacobsen N, Heslop-Harrison JS (1995) The hybrid origin of two cultivars of Crocus (Iridaceae) analysed by molecular cytogenetics including genomic Southern and in situ hybridizaiton. Ann Bot 76:253–262 Piyachaturawat P, Charoenpiboonsin J, Toskulkao C, Suksamrarn A (1999) Reduction of plasma cholesterol by Curcuma comosa extract in hypercholesteromic hamsters. J Ethnopharm 66:199– 204 Prana MS, Sastrapradja S, Hawkes JG, Lubis I (1978) A cytological study of some Indonesian Curcuma species. J Root Crops 4:31–35 Qu Y, Xu F, Nakamura S, Matsuda H, Pongpiriyadacha Y, Wu L, Yoshikawa M (2009) Sesquiterpenes from Curcuma comosa. J Nat Med 63:102–104 Ramachandran K (1961) Chromosome numbers in the genus Curcuma Linn. Curr Sci 30:194–196 Rieseberg LH (1997) Hybrid origins of plant species. Annu Rev Ecol Syst 28:359–389 Rieseberg LH, Willis JH (2007) Plant Speciation. Science 317: 910–914
Ribosomal FISH mapping in Curcuma Sharma G, Gohil RN (2008) Intrapopulation karyotypic variability in Allium roylei Stearn—a threatened species. Bot J Linn Soc 158:242–248 Siroky B (2008) Chromosome landmarks as tools to study the genome of Arabidopsis thaliana. Cytogenet Genome Res 120:202–209 Sodsai A, Piyachaturawat P, Sophasan S, Suksamrarn A, Vongsakul M (2007) Suppression by Curcuma comosa Roxb. of proinflammatory cytokine secretion in phorbol-12-myristate-13acetate stimulated human mononuclear cells. Int Immunopharm 7:524–531 Soontornchainaksaeng P, Jenjittikul T (2010) Chromosome number variation of phytoestrogen-producing Curcuma (Zingiberaceae) from Thailand. J Nat Med 64:370–377 Soontornchainaksaeng P, Chantaranothai P, Senakun C (2003) Genetic diversity of Croton L. (Euphorbiaceae) in Thailand. Cytologia 68:379–382 Stebbins GL (1971) Chromosome evolution in higher plants. Addison– Wesley, Massachusetts Stevens JP, Bougourd SM (1991) The frequency and meiotic behaviour of structural chromosome variants in natural populations of Allium schoenoprasum L. (wild chives) in Europe. Heredity 66:391–401
49 Suksamrarn A, Ponglikitmongkol M, Wongkrajang K, Chindaduang A, Kittidanairak S, Jankam A, Yingyongnarongkul B, Kittikanumat N, Chokchaisiri R, Khetkam P, Piyachaturawat P (2008) Diarylheptanoids, new phytoestrogens from the rhizomes of Curcuma comosa: isolation, chemical modification and estrogenic activity evaluation. Bioorg Med Chem 16:6891–6902 Takano A, Okada H (2002) Multiple occurrence of triploid formation in Globba (Zingiberaceae) from molecular evidence. Plant Syst Evol 230:143–159 Thampithak A, Jaisin Y, Meesarapee B, Chongthammakun S, Piyachaturawat P, Govitrapong P, Supavilai P, Sanvarinda Y (2009) Transcriptional regulation of iNOS and COX-2 by a novel compound from Curcuma comosa in lipopolysaccharideinduced microglial activation. Neurosci Lett 462:171–175 Winuthayanon W, Piyachaturawat P, Suksamrarn A, Ponglikitmongkol M, Arao Y, Hewitt S, Korach KS (2009) Diarylheptanoid phytoestrogens isolated from the medicinal plant Curcuma comosa: biologic actions in vitro and in vivo indicate estrogen receptordependent mechanisms. Environ Health Perspect 117:1155–1161 Xu F, Nakamura S, Qu Y, Matsuda H, Pongpiriyadacha Y, Wu L, Yoshikawa M (2009) Structures of new sesquiterpenes from Curcuma comosa. Chem Pharm Bull 56:1710–1716
123
