J Plant Res (2012) 125:55–69 DOI 10.1007/s10265-011-0422-1 REGULAR PAPER Chromosome diversity and evolution in tribe Lilieae (Liliaceae) with emphasis on Chinese species Yun-Dong Gao • Song-Dong Zhou • Xing-Jin He Juan Wan • Received: 4 January 2011 / Accepted: 17 March 2011 / Published online: 11 May 2011 Ó The Botanical Society of Japan and Springer 2011 Abstract In this paper, karyotype data of the tribe Lilieae in China were analyzed and been superimposed onto a phylogenetic framework constructed by the internal transcribed spacer to investigate the karyotype evolution. Ten parameters for analyzing karyotype asymmetry were assessed and karyotypic idiogram of five genera of Lilieae were illustrated. The results showed that, the relationship of genera in Lilieae that inferred from Maximum Parsimony criteria and Bayesian Inference were congruent with previous studies, which focused on higher level of Liliales. The karyotype showed distinctive among genera, mainly expressed on the location and amount of secondary constrictions and intercalary satellites: the genus Notholirion have neither of them, and the genera Cardiocrinum and Fritillaria have the secondary constriction alone; the genera Lilium and Nomocharis showed both features, and the distribute pattern of the intercalary satellites showed similarity among related clades. The asymmetry that assessed by several methods indicated that the evolution trend of Lilieae did not follow a single direction, but different in each genus. On the sectional level of the genus Lilium (including Nomocharis) the karyotype evolution included three major periods. Combining the chromosomal structure variations and karyotype asymmetry, the chromosome Electronic supplementary material The online version of this article (doi:10.1007/s10265-011-0422-1) contains supplementary material, which is available to authorized users. Y.-D. Gao  S.-D. Zhou (&)  X.-J. He (&)  J. Wan Laboratory of Systematic and Evolutionary Botany, College of Life Science, Sichuan University, Chengdu 610064, China e-mail: songdongzhou@yahoo.com.cn X.-J. He e-mail: xjhe@scu.edu.cn diversity and evolution in Lilieae were quite clear in the light of molecular inference. Keywords China  ITS  Karyotype asymmetry  Secondary constrictions  Lilieae Introduction The tribe Lilieae sensu Tamura (Tamura 1998) belongs to Liliaceae sensu APG III (Angiosperm Phylogeny Group 2009), and contains five genera: Lilium L., Nomocharis Franch., Fritillaria L., Notholirion Wallich ex Boissier and Cardiocrinum (Endlicher) Lindley. Among them, Lilium and Fritillaria are widely distributed in the North hemisphere, while Nomocharis, Notholirion and Cardiocrinum are endemic to East Asia and Himalayas (Liang 1995; Liang and Tamura 2000). All of these five genera are occurred in China, and the proportion of Chinese species in each genus is quite high. To Lilium, there are 55 species in China compared with about 110 species in the world. To other genera, 2 out of 3 for Cardiocrinum, 6 out of 7 for Nomocharis, and 3 out of 5 for Notholirion (Liang and Tamura 2000). The proportion of Fritillaria is the lowest compared with these four genera. Eastern Asia and Himalayas have all genera of Lilieae, which made it a diversity center of this tribe. In fact, Patterson and Givnish (2002) considered Himalayas is also the origin center of it. The tribe Lilieae contains many beautiful flowers in the world, namely, the lilies. Indeed, all five genera were considered worth to cultivate, not just for ornamental but also for medical use. For example, the bulbs of Fritillaria cirrhosa, F. unibracteata, F. przewalskii and F. delavay are used as a traditional Chinese medicine named Chuanbeimu in China. Notholirion bulbliferum is also used medically 123 56 (The State Pharmacopoeia Commission of the People’s Republic of China 2000). The rest three genera are treated as resources of fresh cut flowers or garden plants, for their showy flowers and fragrance. Horticulturists are very interested in these species of Lilieae, and more and more cultivars were produced in the past years (Royal Horticultural Society http:\\www.rhs.org.uk). The Liliaceae sensu stricto was firstly proposed by Takhtajan (1997) and consisted of nine genera which were then subdivided into two tribes: Erythronium and Tulipa (tribe Tulipeae); Cardiocrinum, Lilium, Notholirion, Nomocharis, Fritillaria and Rhinopetalum (tribe Lilieae). Tamura (1998) recognized two subfamilies in the Liliaceae sensu stricto: Medeoloideae (Clintonia and Medeola) and Lilioideae (Erythronium, Tulipa, Gagea and Lloydia (tribe Tulipeae); Cardiocrinum, Lilium, Fritillaria, Nomocharis and Notholirion (tribe Lilieae)). Recently researches using molecular analysis confirmed that Liliaceae sensu Tamura is monophyletic (Rudall et al. 2000; Patterson and Givnish 2002; Fay et al. 2006; Tamura et al. 2004). The genera Lilium and Nomocharis are the members of the tribe Lilieae, with the genera Cardiocrinum, Fritillaria and Notholirion as their sister groups. Hayashi and Kawano (2000) suggested that the genus Fritillaria was the closest affinity of Lilium-Nomocharis group, while Cardiocrinum and Notholirion were relative more remote than expected before. Meanwhile, all previous researches (Patterson and Givnish 2002; Tamura et al. 2004; Fay et al. 2006) proved that the Lilium–Nomocharis complex were the most advanced group in the tribe Lilieae of Liliaceae sensu stricto. Rønsted et al. (2005) investigated the delimitation and infragenetic classification of Fritillaria, combining with 15 species of Lilium. Use molecular analysis they have proved the monophyly of Fritillaria and Lilium, as well as their sister relationship. With these achievements above a well resolved phylogenetic framework of Lilieae had accomplished. Karyotype comparison analysis has been long used to describe chromosome patterns and moreover, the evolutionary direction in some closely related groups (Stebbins 1971; Stace 1978; González-Aguilera and FernándezPeralta 1984; Hong 1990; Watanabe et al. 1995; De Melo Nationiel et al. 1997; Vanzela et al. 1997; Das et al. 1999; Shan et al. 2003). However, for technical defects and the low resolution karyotype analysis have much limitation in use. The usage of this technical alone was questioned in recent years (Peruzzi et al. 2009). Lately, karyotype analysis was more used in the light of phylogenetic framework. For example, Peruzzi et al. (2009) based on karyotypic data accumulated in the past years, by superimposing them to a well resolved phylogeny framework of Liliaceae sensu APG (Angiosperm Phylogeny Group 2009), found the main evolution trend on karyotype in Liliaceae. Their study 123 J Plant Res (2012) 125:55–69 also showed that tribe Lilieae have the similar genome size, karyotype idiogram and the same basic chromosome numbers. Therefore, Lilieae was monophyletic and have been supported by both molecular and cytotaxonomic evidences (Rudall et al. 2000; Patterson and Givnish 2002; Fay et al. 2006; Tamura et al. 2004; Peruzzi et al. 2009). However, the lacking of phylogenetic framework making the karyotype evolution not so clear on the genus level of Lilieae, and this need further research. Noda (1991) suggested the major differences among species of the genus Lilium corresponded to the number and position of secondary constrictions (SCs) and the activity of nuclear organization regions (NORs). He concluding about the type of speciation in the genus Lilium, which included long-term accumulation of numerous small structural chromosome changes. Using molecular cytogenetic characters Muratović et al. (2010a) performed a karyotype evolution scenario on Lilium sect. Liriotypus, and suggested the problem of taxonomic misunderstandings of very close and evolutionary young species could be overcome by the use of molecular cytogenetic markers. Their results also supported Noda’s (1991) findings described above. This indicates the cytological researches in Lilium and its allied genera may find a better resolution in karyotype instead of molecular analysis, since previous works (Nishikawa et al. 1999, 2001; Hayashi and Kawano 2000) and ours (Gao et al. unpublished data) suggested the genus Lilium, including Nomocharis showed low resolution in both nuclear (nrITS) and chloroplast sequences phylogeny. We hope the karyotype analysis in section level of those two genera will provide deeper insights in phylogenetic and biogeographic inferences. As mentioned above, the eastern Asia contains all of these five genera and most of them were distributed in China. With the achievements on both molecular and cytotaxonomy before and present, by putting Lilieae in the worldwide background and using molecular phylogenetic methods, the goals of present work were to (1) find out the karyotype evolution pattern and trend of Lilieae on the generic level, and (2) since our data included a lots of Lilium and Nomocharis species, we want to infer how does the geographic distribute pattern associate with karyotype variation as well as the section level karyotype evolution in these two genera. Materials and methods Sample selection Most samples of China in our study were collected in the wild, and some other species that are not distributed in China were sampled by surveying the literatures before. J Plant Res (2012) 125:55–69 Chromosome and molecular data of species in China were relied exclusively on our research (except for that Lilium speciosum var. gloriosoides, L. henrici, Fritillaria thunbergii and F. yumingensis in which cytological data were adopted based on the previous results, for these species were failed in getting a well karyotype in our study), while other species out of China were chosen from previous works (STable 1). There are thirty-three species (including variants) of Lilium, plus 9 species out of China: North America (3), Japan (2) and Europe (4). The rest accessions consisted of five Nomocharis species, two Cardiocrinum and two Notholirion, and finally, 10 Fritillaria species with 4 added outside China. Each one species of genera Gagea and Lloydia were chosen as outgroups, which belonged to the tribe Tulipeae and were suggested as sister groups of Lilieae in the previous research (Patterson and Givnish 2002). The details of materials are listed in the Supplementary Data (STable 1). Molecular analysis DNA extraction Total DNA was isolated from fresh or silica gel-dried leaf tissue using modified cetyltrimethyl-ammonium bromide (CTAB) protocol by Doyle and Doyle (1987), and several times with Plant Genomic DNA Kit (TIANGEN Biotech, Beijing, China). Polymerase chain reaction We amplified the entire internal transcribed sequences of nrDNA sequences (including ITS1, 5.8S and ITS2) in one piece using ITS4 (50 -TCCTCCGCTTATTGATATGC-30 ) and ITS5 (50 -GGAAGTAAAAGTCGTAACAAGG-30 ) universal primers. The polymerase chain reaction (PCR) profile was as follows: the initial denaturation was 94°C for 2 min and the final extension was 72°C for 10 min. The reactions were subjected to 35 cycles. Each cycle consisted of a denaturation at 94°C for 45 s, primer annealing at 55°C for 45 s, and extension at 72°C for 60 s. PCR reactions were acted on GeneAmp PCR System 9700 (Applied Biosystems, USA). The PCR products were then sent to Invitrogen Biotech Co. Ltd. (Shanghai, China) for purifying and sequencing. Sequencing was done using an ABI-3730XL DNA sequencer. For each sampled specimen, forward and reverse sequencing reactions were performed for confirmation. Phylogeny analysis The boundaries of ITS1 and ITS2 were determined by comparing the aligned sequences with previously published 57 Lilium sequences (Nishikawa et al. 1999, 2001). All the sequences have been deposited in GenBank (see Supplementary Data for accession numbers). Resulting DNA sequences of the entire ITS region of all samples were multiply aligned using ClustalX (Thompson et al. 1997) for alignment initially and then by eye following the guidelines of Kelchner (2000) and Morrison (2009) in MEGA 4.0 (Tamura et al. 2007), and these alignments were used for further analysis. Some (13 out 78) of the sequences used in the study were taken from previous studies (Nishikawa et al. 1999; Rønsted et al. 2005; İkinci et al. 2006; Dubouzet and Shinoda unpublished; Cui et al. unpublished). Gaps were positioned to minimize nucleotide mismatches and were treated as missing data in the phylogenetic analysis. Uncorrected pairwise nucleotide differences were determined using PAUP* version 4.0b10 (Swofford 2003). Maximum Parsimony analysis were also carried out using PAUP*. For each analysis, Maximum Parsimony trees were sought using the heuristic search strategies of PAUP* (with 1,000 replicate analyses, random stepwise addition of taxa, TBR branch swapping, and setting the maximum number of trees to 10,000). Bootstrap values were calculated from 1,000,000 replicate analyses using fast stepwise addition of taxa, and only those values compatible with the majority-rule consensus tree were recorded. Bayesian Inference analyses of ITS dataset were conducted using MrBayes version 3.1.2 (Ronquist and Huelsenbeck 2003). Prior to these analyses, the program MrModeltest version 2.2 (Nylander 2004) was used to select an evolutionary model of nucleotide substitution that best fit these data, and the GTR?I?G model under the Akaike information criterion (AIC) was selected. From a random starting tree, the Bayesian analysis was run for 10 million generations (when the average standard deviation of split frequencies reached 0.01 or below, and the potential scale reduction factor of all parameters approached 1.0) and the trees were saved to a file every 1,000 generations. Four simultaneous Markov chain Monte Carlo (MCMC) chains were run, and the temperature was adjusted to 0.2 in order to keep an appropriate heat range for the 4 chains. Branch lengths of the trees were saved. The first 30% trees were discarded as the ‘‘burn-in’’ and a majority-rule consensus tree was calculated based upon the remaining trees. Cytological studies Root tips were collected in morning, then pretreated with a solution of 0.1% aqueous colchicine and saturated dichlorobenzene (1:1) for 6–7 h at room temperature. Then root tips were fixed in Carnoy’s solution I (ethanol:acetic acid = 3:1) for at least 1 day, and stored in 70% ethanol at 4 ± 2°C for further studies. For chromosome observation, 123 58 123 Table 1 Measures of karyotype asymmetry used in the present work Asymmetry index Equation Reference Coefficient of variation (CV) of the centromeric index (CVCI) Standard deviation of centromeric index ; Mean of centromeric index Standard deviation of haploid chromosome length CVCL ¼  100 Mean chromosome length CVCI  CVCL AI ¼ 100 Pn Mean length for short arms in every homologous chromosome pair i¼1 Mean length for long arms in every homologous chromosome pair A1 ¼ 1  n Standard deviation of haploid chromosome length A2 ¼ Mean chromosome length Length of short arms in chromosome set  100 TF % ¼ Total chromosome length in set Mean length of the short arms Syi ¼  100 Mean length of the long arms Pn Mean of the ratios of the length of each chromosomeðCLi Þ i¼1 Length of the longest chromosomeðLCÞ Rec ¼  100 n Length of long arms in chromosome set  100 As K% ¼ Total chromosome length in set Pn Length of long arm of chromosome i  length of short arm of chromosome i i¼1 Length of long arm of chromosome i þ length of short arm of chromosome i A¼ n Paszko (2006) Coefficient of variation (CV) of chromosome lengths (CVCL) Asymmetry index (AI) Intrachromosomal asymmetry index (A1) Interchromosomal asymmetry index (A2) Total form% (TF %) Index of karyotype symmetry (Syi) Index of chromosome size resemblance (Rec) Percentage karyotype asymmetry index (As K %) Degree of karyotype asymmetry (A) CVCI ¼ Paszko (2006) Paszko (2006) Romero Zarco (1986) Romero Zarco (1986) Huziwara (1962) Greilhuber and Speta (1976) Greilhuber and Speta (1976) Arano (1963) Watanabe et al. (1999) J Plant Res (2012) 125:55–69 J Plant Res (2012) 125:55–69 59 Mean haploid idiograms Fig. 1 Karyograms of Lilium davidii var. davidii (Voucher No. G07021). Arrows indicate intercalary satellites and Asterisks locate secondary constrictions. Scale bar 10 lm All accessions in this research showed the basic chromosome number as x = 12. The haploid idiograms were constructed by the average length of each pair of homologous chromosomes, and arranged according to their decreasing length. Besides the arm length, there were other characters were added on the idiograms, including the locations of the secondary constrictions and the intercalary satellites. The intercalary satellites were considered to be important features in previous researchers (Stewart 1947; Gao et al. 2009; Gao Yundong et al. unpublished data; see Fig. 1 for these two features). Under this the idiograms of Notholirion, Cardiocrinum, Nomocharis and Fritillaria were constructed. Only in Lilium the subgenus level were adopted based on the phylogenetic results in present, by which the karyotype evolution can be illustrated and discussed on the subgenus level in the genus Lilium (including Nomocharis). Results the root tips were macerated in a mixture of 1N HCl for 8–10 min at 60°C, stained by 1% Carbolic acid Fuchsin for about 10 min, and squashed on a glass slide. For each population, there were at least three individuals were studied, and the chromosomes of at least 30 metaphase plates from each individual were studied for counting chromosome numbers. Number, size and shape of chromosomes were observed, and karyotypic asymmetry was evaluated. Satellites were taken into calculation of chromosome size and the highest number of satellite chromosomes in each plate was recorded, following Lifante (1996). Asymmetry assessment Karyotypic asymmetry was assessed by 10 methods that proposed by former researchers, which were discussed and assessed by Paszko (2006), and also adopted by Peruzzi et al. (2009). The indexes adopted in present were list in Table 1, as well as by which parameter (e.g. length of long arms, ratios of short arms against long arms, total haploid chromosome length) they are expressed, respectively. The correlations between each asymmetry index were illustrated using a matrix scatter. And finally the asymmetry indexes and parameters were performed by genus using box plots in order to show the karyotypic asymmetry tendency in the tribe Lilieae. With the aim for better understanding of karyotype evolution on section level of Lilium, the box plots were also performed by clades of Lilium and Nomocharis resolved in molecular analysis. Sequence information The total data set contains 78 accessions or 68 species (because for several species, more than one accessions were included). Among the ingroup sequences analyzed, the ITS region varied in length from 607 bp (Fritillaria maximowiczii) to 633 bp (Lilium speciosum). The two outgroup species have a moderate length of 615 bp (Lloydia serotina) and 618 bp (Gagea filiformis). The alignment length of the internal transcribed spacer (ITS) was 670, including 347 variety sites, in which 259 (75%) sites were parsimony informative. The mean G?C content of the ITS region was 61.22%. Phylogenetic analyses Each of the strict consensus trees resulting from these analyses in Maximum Parsimony showed relationships highly consistent to those inferred in the Bayesian trees, although resolution of relationships was generally much poorer in the former. However, clades supported with 1.00 posterior probability values in the Bayesian trees were also well-supported (with bootstrap values [70%) in the consensus tree. Based on this we only performed the Maximum Parsimony 50% majority consensus tree and indicated the differences between those two analyses (Fig. 2). The tree length = 973, and consistency index (CI) = 0.52, homoplasy index (HI) = 0.48, retention index (RI) = 0.767, rescaled consistency index (RC) = 0.40. 123 60 123 J Plant Res (2012) 125:55–69 J Plant Res (2012) 125:55–69 b Fig. 2 Phylogram of the 50% major consensus tree resulting from the Maximum Parsimony analysis of ITS dataset, accompanied by the karyotype idiogram of major clades. Values along branches represent Bayesian posterior probabilities (PP) and parsimony bootstrap (BS), respectively. The oval spots on the karyotype idiogram indicating the intercalary satellite while the hollow bars represented the secondary constriction. The shadow part illustrating the Bayesian topology that compared with Maximum Parsimony analysis The tribe Lilieae was proved to be monophyletic in this research, which was congruent with previous studies (Rudall et al. 2000; Patterson and Givnish 2002; Fay et al. 2006; Tamura et al. 2004). The phylogenetic framework of Lilieae showed that: the genus Notholirion was the basal group, with Cardiocrinum as its sister; the genus Fritillaria was closely related to Lilium–Nomocharis; the latter was the most advanced group in Lilieae, and based on previous research (Hayashi and Kawano 2000; Nishikawa et al. 1999, 2001; Rønsted et al. 2005; Peruzzi et al. 2009) and the results in present the genus Nomocharis should be combined into Lilium. This combination was supported by more evidences such as morphology (Gao Yundong et al. unpublished data). The subdivision in the genus Lilium is showed in Fig. 2. Ten clades were found and the genus Nomocharis was 61 found to be monophyletic, however, it’s totally embedded in the Lilium clade. The Nomocharis group was found closely related to the European Lilium species, and they are the basal group in Lilium (Fig. 2). The rest part of the Lilium was more or less congruent with the subdivision suggested by Comber (1949), while two clades were also found agreed with other authors, i.e. the section Lilium Wang et Tang and the section Lophophorum Wang et Tang (Wang and Tang 1980). Cytological data The basic chromosome number of tribe Lilieae is x = 12. Polyploidy was not significant except a few triploidy populations were found in Notholirion bulbliferum and Lilium tigrinum. The basic chromosome number and ploidy are quite stable. The asymmetry indexes that assessed by previous methods were listed in the supplementary data, and their correlations were illustrated by a matrix scatter in SFig. 1. The positive correlations were found between several couples, e.g. CVCL and A2, TF% and Syi, A1 and As K%. The box plots of the asymmetry indexes by genus were illustrated in Fig. 3 except that A2 was excluded for it’s Fig. 3 Box plots of nine asymmetry indexes that performed by genus included in this study. Taxa are ordered by phylogenetic grouping (according to the phylogenetic tree in Fig. 2) 123 62 J Plant Res (2012) 125:55–69 Fig. 4 Box plots of ten asymmetry indexes that performed by clades of Lilium (including Nomocharis) in this study. Taxa are ordered by phylogenetic grouping (according to the phylogenetic tree in Fig. 2). Taxa are arranged as Liriotypus, Nomocharis, Archelirion, Leucolirion, Pseudolirion, Martagon, Sinomartagon2, and Lilium from left to right actually the same as CVCL (CVCL = A2 9 100). Figure 4 showed the box plots of major clades in Lilium (including Nomocharis). Mean haploid idiograms for all genera that alongside with the consensus phylogenetic framework of Lilieae were shown in Fig. 2. The secondary constrictions and the intercalary satellites were illustrated beside the phylogenetic tree, according to their respective phylogeny positions (Fig. 2). The hollow bars and reddish-oval spots on karyotype idiograms stood for secondary constrictions and intercalary satellites, respectively. The most ancient genus, namely Notholirion, was found without any secondary constriction, and no intercalary satellite. Cardiocrinum and Fritillaria showed secondary constrictions, however, 123 without any intercalary satellite. In the genus Lilium the situation was quite different, for both features were existed. Though the pattern in the genus Lilium is quite complicate, the karyotype acted distinctively on the subgenus level, which were delimited by the molecular data in present and mostly agree with the section proposed by Comber (1949). However, some subdivisions proposed by other authors in the genus Lilium were proved to be acceptable, like section Lophophorum Wang et Tang (Wang and Tang 1980). Wang and Tang (1980) recognized Sect. Lophophorum (Bur. et Franch.) Wang et Tang out of Sect. Sinomartagon Comber, and placed campaniform-flowered species in it. In our research, all species belongs to Sect. Lophophorum (Bur. et Franch.) Wang et Tang show a very similar and J Plant Res (2012) 125:55–69 unique karyotype (STable 1). They hold a relatively low asymmetry indexes and typed 3A (Stebbins 1971), which made them easy to identified in section Sinomartagon Comber. Section Leucolirion Comber is a group with showy-trumpet flowers. In present study some species showed significant differences between species on the first two pairs of chromosomes. Whether the intercalary satellites on the first two pairs of chromosomes exist or not would split this section into two parts (Fig. 2). Wang and Tang erected a section including L. brownii, L. longifolium, L. formosanum as section Lilium out of Leucolirion. From our research section Lilium did show distinctiveness compared with the rest of Luecolirion (Fig. 2). The molecular data above also supported their independence of these two sections. Discussion The reliability of inferring phylogeny framework from nrITS In the tribe Lilieae, previous researches had shown that these five genera were closely related. Among them, the genus Notholirion was considered to be the basal group (Rudall et al. 2000; Patterson and Givnish 2002; Fay et al. 2006; Tamura et al. 2004). The genus Cardiocrinum was later-branching group and it was considered to be ancient than the rest ones. The genus Fritillaria was believed to be Lilium’s most close affinity, and diversified earlier than the latter (Hayashi and Kawano 2000). Finally, the genus Nomocharis was treated as the youngest group in the tribe by some authors (Liang 1995; Wu et al. 1994), however, recent findings tend to reduce it to a subgenus level and put it into the genus Lilium (Hayashi and Kawano 2000; Nishikawa et al. 1999, 2001; Rønsted et al. 2005; Peruzzi et al. 2009). Our result based on the ITS sequences was congruent with the former result which constructed by several authors and was considered quite robust (Peruzzi et al. 2009). Karyotype structure in Lilieae In Fig. 2, which the mean haploid karyotype was given by clades herein, and the data that showed the diversity of karyotype structure were gathered in the Supplementary Data (STable 1). There was considerable variation in karyotype asymmetry. The secondary constrictions were variable and some authors (Stewart 1947; Gao Yundong et al. unpublished data) have recognized their probable functions in the genus Lilium. By superimposing these data onto the phylogenetic framework, the trends and patterns in karytotype evolution in Lilieae could be seen. 63 Basic chromosome number and the total haploid length All the previous researches (Stewart 1947; Smyth et al. 1989; Siljak-Yakovlev et al. 2003; Muratović et al. 2005; Gao et al. 2009) and present study confirmed that the tribe Lilieae has the basic chromosome number as x = 12. It’s quite stable and the pattern of these 12 pairs of chromosomes was almost the same. The species in present all contained two pairs of larger metacentric chromosomes, and the rest of them are all subtelocentric to telocentric (Fig. 2). However, the stability also caused difficulty on the phylogenetic investigation, for this karyotype pattern among the genera of Lilieae can’t provide enough information. Obviously, the basic chromosome number variation was not the main evolution power in this tribe. The total haploid length (THL) in the tribe Lilieae were increased by the order of Notholirion (109.4 lm), Cardiocrinum (130.62 lm), Fritillaria (153.1 lm) and Lilium (174.38 lm) (including Nomocharis) (Peruzzi et al. 2009). The results of ours showed the same trend as the THL of these four groups as 118.03, 123.43, 149.94 and 155.39 lm (STable 1). This was congruent with the phylogenetic framework that concluded by the molecular analysis (Fig. 2). The THL was considered to be a proxy for genome size (Narayan and Rees 1976; Raina and Rees 1983; Ceccarelli et al. 1995), therefore, the genome size in the tribe Lilieae was increased by the order of phylogeny. Asymmetry In the past years, besides the basic chromosome number, karyotype analysis focused mainly on the structure of chromosomes. The most important method was the asymmetry assessment. Karyotype asymmetry was a good expression for the general morphology of karyotype in plants. The changes of chromosome morphological characters were believed relating to the evolution in higher plants (Zarco 1986). Over the past years, many methods were proposed to assess the karyotype asymmetry. Paszko (2006) revised all the asymmetry indexes before and proposed three new indexes: CVCI, CVCL and AI. As Peruzzi et al. (2009) suggested, the single asymmetry index (AI) didn’t adequately reflect all aspects of karyotype asymmetry in Liliaceae. To retain more indexes can provide more aspects of karyotype asymmetry. This was congruent with our results that focused on Lilieae (Table 1). Therefore, we adopt 10 different parameters or indexes for asymmetry assessment. Box plots summarized the values of 10 parameters for each genus (Fig. 3), according to their phylogenetic position (by the order Notholirion, Cardiocrinum, Fritillaria, Nomocharis and Lilium). We will discuss the karyotype changes by index in the light of phylogeny as follows. 123 64 The total form percent (TF%), was described by Huziwara (1962) to analyze the karyotypes in the genus Aster. It expressed the proportion of total length of short arms in the complement. Our results showed that the asymmetry was increased initially and then decreased in the genera Nomochais and Lilium. The latter two were considered more reasonable to be a single genus (Hayashi and Kawano 2000; Nishikawa et al. 1999, 2001; Rønsted et al. 2005; Peruzzi et al. 2009). In fact the median and range values (IQR) of TF% of the genera Lilium and Nomocharis were identical, which confirmed their closed affinity in the karyotype way. The pattern showed by TF% means the additional DNA were added mainly on short arms in Cardiocrinum and Fritillaria. However, in the latter two genera Nomocharis and Lilium the proportion became lower which indicated a reverse pathway happened. This means the additional DNA was added on the long arm in genera Lilium and Nomocharis. The Syi (Greilhuber and Speta 1976; Venora et al. 2002) value indicates the ratio of the mean length of the short arms against the mean length of the long arms in a chromosome set. It’s resemble to TF%, and these two were positive correlated (SFig. 1). The Rec index (Greilhuber and Speta 1976; Venora et al. 2002) expresses the mean of the ratios of the length of each chromosome to that of the longest one, and been suggested as a incorrect parameter (Paszko 2006). Our result was congruent with Paszko for the box plots of this index showed nearly no differences among all genera herein. The As K% index (Arano 1963) is expressed by the ratio of the sum of the lengths of the long arms of individual chromosomes to the total haploid length of the chromosome complement. Then it’s easy to understand that this index is negative correlated with TF% (SFig. 1). The box plots of As K% was opposite when compared with the one of TF% (Fig. 3a, d), however, what they represented were the same. Zarco (1986) proposed the intrachromosomal asymmetry index (A1) and the interchromosomal asymmetry index (A2) to estimate karyotype asymmetry. The A2 index was identical as the relative variation in chromosome length (CVCL), therefore no discussion was made in this context. The A1 index does not depend on chromosome number or chromosome size (Zarco 1986), and it expressed the intrachromosomal asymmetry of all homologous chromosome pairs. The box plots showed that the genus Notholirion has a highest value comparing with the rest, and the value decreased in Cardiocrinum and Fritillaria. In contrast, the values increased again in Lilium and Nomocharis, indicating that a different process happened in the latter two genera. The pattern showed by A1 in box plots demonstrated that the intrachromosomal variation in the genus Notholirion was pretty great. Then the genera 123 J Plant Res (2012) 125:55–69 Cardiocrinum and Fritillaria have a lower A1 value which probably caused by increasing the short arms length in single chromosomes, especially the second pair (Fig. 2). The short arm of the second homologous chromosome pair in the genus Notholirion was significant shorter than that in the rest four genera. The A1 value were increased in the genera Lilium and Nomocharis, which indicated the differences between the short and the long arms in these two genera were raised, probably by additional DNA that added on the long arms of the smaller chromosomes (Peruzzi et al. 2009). This is in accordance with the conclusion that made by indexes TF% and As K% we discussed above. The degree of karyotype asymmetry (A) defined by Watanabe et al. (1999) was a index that reflex the degree of difference of long and short arms of chromosome in a complement. It decreased by the order of Notholirion, Cardiocrinum and Fritillaria, indicating the length differences were became not significant in these three genera. In contract the value in genera Lilium and Nomocharis increased, and along with the evolution the karyotype features of Lilium and Nomocharis were different comparing with its affinities. These variations in chromosome structures have achieved either by the shift of centromere positions or the addition of supernumerary DNA on the long arms of chromosomes (Hong 1990; Peruzzi et al. 2009). The relative variation in chromosome length (CVCL) and the relative variation in centromeric index (CVCI) were recently proposed by Paszko (2006), who also proposed a new index AI based on these two parameters. The AI index have been proved not adequately to reflect all aspects of karyotype asymmetry in Liliaceae, while CVCL and CVCI were more informative (Peruzzi et al. 2009). The CVCL was correlated with A2 because CVCL = A2 9 100. The CVCL and CVCI showed the same trend for different genera (Fig. 3g, i), and both fluctuant. The genus Cardiocrinum hold the highest CVCL value revealed its more variable in chromosome length. In contrast, the rest four genera were balanced by the additional DNA added on the long arms of small chromosomes, as Peruzzi et al. (2009) proposed. The CVCI showed a same pattern among genera as CVCL. Thus, as the evolution the position of centromeres in the genus Cardiocrinum shifted, while the additional DNA were added on the larger chromosomes instead of the smaller ones, which happened in the rest of genera in Lilieae. Based on the results of asymmetry indexes there is no uniform trend or direction that increase or decrease along with evolution in this tribe. Chromosome variation might have effect in the evolution in the tribe Lilieae, however, these changes could cause much change in chromosome fine structure besides karyotype asymmetry. As mentioned above, the secondary constrictions and intercalary satellites also associated with the evolution pattern in Lilieae, while J Plant Res (2012) 125:55–69 they did not change the karyotype asymmetry. For the DNA amount in these five genera did not significantly increase, the karyotype evolution might largely based on pericentric inversions and/or differential translocations of DNA between smaller and larger chromosomes (Peruzzi et al. 2009). Secondary constrictions and intercalary satellites As mentioned above, the chromosome fine structure variation was also revealed by other features. The most significant ones were the secondary constrictions and what been called intercalary satellites. Figure 1 showed all secondary constrictions and intercalary satellites that found in chromosomes. Previous studies (Stewart 1947; Gao Yundong et al. unpublished data) noticed that variations in the position of secondary constrictions were correlated with the type of chromatin distribution and the geographical distribution. Considering that the geographical distribution was always in accordance with their systematic positions, i.e. their phylogeny framework, it’s not difficult to understand that variations in position of secondary constrictions can reflect their phylogenetic relationships. In the tribe Lilieae, the secondary constrictions were emerged and evolved as the evolution of genera. It’s hard to see them in the base group—Notholirion contains no secondary constrictions (Fig. 2). They emerged in genus Cardiocrinum on few positions of several chromosomes, including one on the second pair of the complement. Then Fritillaria showed approximate distribution pattern with Cardiocrinum. No intercalary satellites showed until it goes to the genus Lilium. The basal group of Lilium– Nomocharis were European lilies (sect. Liriotypus) plus the genus Nomocharis. The locations of intercalary satellites in sect. Liriotypus were identical (pair 7) as it did in Nomocharis, however, the latter have an additional one on the first pair of chromosome. The amount of intercalary satellites increased in the rest groups, while the secondary constrictions reduced. Although not strongly supported, the resemblance on karyotype could found in related groups, such as sect. Archelirion and sect. Leucolirion which showed the same feature on the first two pairs also showed closely relationship on the molecular tree (Fig. 2). Besides, Sect. Pseudolirion which distributed in North America has a karyotype resembled to Sect. Lophophorum Wang et Tang and a clade formed by Lilium duchartrei and L. lankongense, which was in accordance with their phylogenetic relationships in our research. Stewart (1947) investigated most of Lilium species in North America (The New World) and found that it was easy to differentiate them from the Old world’s species. He pointed out that secondary constrictions correlated with chromatin distribution as well as with geographic 65 distribution. In his study, all North American species were found not to carry any intercalary satellites on the first two pairs of submedian centromeres, while the most of the species from Eastern Asia and Himalayas, in contrast, were found to carry such intercalary satellites on the same positions. Also in European species, researchers (Stewart 1947; Smyth et al. 1989; Siljak-Yakovlev et al. 2003; Muratovic et al. 2005) did not find any intercalary satellites on these first two pairs of chromosomes. The situation of East Asian species appears to be more complicated for they seem to have both types, even in the same section (Stewart 1947; and present study). Several species like L. tsingtauense (Stewart 1947; Smyth et al. 1989) in Sect. Martagon Comber and L. longiflorum or L. formosanum in Sect. Leucolirion Comber were found without any intercalary satellites on the first two pairs of chromosomes, while the rest of the species in their sections showed them. These species above were distributed in Far East and Japanese Archipelago (Liang and Tamura 2000), which are more adjacent to North America. This distribution pattern that related to karyotype might suggest multiple dispersal events had happened. The North American and European descendents were more like ancient groups in the genus Lilium, and they colonized their distribution areas earlier than these groups in eastern Asia. Then the ancestors of eastern Asia species colonized and dispersed in Asia until reached Far East. This could explain why American Lilium species were related to these in Himalayas instead of Far East in the molecular analysis. This kind of disjunction distribution pattern have been seen in other Eurasia–North America distributed groups (Wen and Zimmer 1996; Wen 1999, 2001, 1998; Xiang et al. 2000; Xiang and Soltis 2001). Karyotype evolution on section level of Lilium (including Nomocharis) In the phylogeny of the genus Lilium (including Nomocharis) of present study and previous work (Nishikawa et al. 1999, 2001; İkinci et al. 2006), the major clades were resolved as: Liriotypus, Nomocharis, Archelirion, Luecolirion, Sinomartagon1 (Lilium duchartrei and L. lankongense), Lophophorum, Pseudolirion, Martagon, Sinomartagon2 and Lilium (Fig. 2). These clades were further analyzed by using box plot on each asymmetry indexes, as well as considering the chromosome structures. Under this the karyotype evolution in the section level of Lilium is clear. In all ten asymmetry indexes, several of them showed no significant trend; such as As K%, A, Syi, Rec, A1 and TF% (Fig. 4a, d, g, h, i, j). These indexes have been proved not sufficient to indicate the phylogeny relationships (Paszko 2006). The rests contain the CVCI, CVCL, AI and A2 (Fig. 4b, c, e, f), in which A2 is the same as CVCL as 123 66 mentioned before. Therefore, only the parameters proposed by Paszko (2006) have phylogenetic sense in this group. This is accordance with the situation in Liliaceae in which researchers found that only these parameters can provide enough messages to conduct the karyotype evolution (Peruzzi et al. 2009). The latter research also indicated the AI can’t provide much data as expected, instead the separation analysis on CVCI and CVCL were more useful. Thus, the karyotype evolution in section level of Lilium (including Nomocharis) were conducted by these two indexes as follows: (1) CVCI increased initially, and reached the highest in section Archelirion and then decreased to the lowest in Lophophorum; then increased again in the rest of groups (Fig. 4e). This process indicates the additional DNA were added on the long arms of smaller chromosomes at first (before the emergence of Achelirion), then added on the short arms until Lophophorum raised. After this, the additional DNA were added on the long arm again in the rest groups. (2) CVCL decreased from the beginning and reached lowest at Lophophorum, then this value keep increasing in the rests (Fig. 4f). The result indicates the additional DNA were added on the short chromosomes firstly, and after the emergence of Lophophorum the added DNA preferred the longer chromosomes in the rest groups. Considering both two parameters, there were basically three phrases in the karyotype evolution of Lilium: (1) additional DNA added on, or transferred from the larger chromosome to the smaller chromosomes’ long arms, which will cause the increasing of CVCI and decreasing of CVCL the same time; then (2) both CVCI and CVCL were decreased as a result of that additional DNA or translocation of chromosome added on the short arms of smaller chromosomes; at last (3) additional DNA or translocation of chromosome added on the long arms of the longer chromosomes (the biggest two homologous chromosome pairs), which can explain the increasing of both two parameters. To summarize, the karyotype evolution in Lilium is not followed a single trend but been separated in three major periods. The turning point of these periods may stand for environment change or speciation events in the evolution process of Lilium. Considering the fact that chromosome structure changes on amount and location of secondary constrictions and intercalary satellites were associated with the geography distribution pattern (discussed above), the karyotype evolution can be illustrated by using the method in present as evolution goes (Fig. 2). Noda (1991) had already concluded the main aspect of karyotype evolution in Lilium as accumulation of numerous small structural chromosome changes. Such changes produced various chromosome features like amount and location of secondary constrictions or intercalary satellites were found useful in taxonomic researches in our study. Muratović et al. (2010a) confirmed the changes such as translocation of 123 J Plant Res (2012) 125:55–69 rDNA loci, inversion, deletion and deactivation of rDNA loci happened in the first two pairs of chromosomes. However, for the technical defeats, we failed in getting more phylogenetic implications from observable chromosome characters. A comprehensive analysis on karyotype of Lilium using advanced technologies such as chromosome banding or fluorescent in situ hybridization (FISH) are needed for a better understanding of karyotype evolution and speciation in Lilium. Muratović et al. (2010a, b) studied European lilies and found the molecular cytogenetic characters were more useful compared with molecular phylogeny, and the chromosome fine structures observed can provide rational phylogenetic framework in this group. We believe these methods will favor the phylogenetic and taxonomic studies in Lilium. The low resolution of molecular phylogeny and reticulate evolution were already detected in this group (Gao Yundong et al. unpublished data), and Muratović et al. (2010a) provided a new solution which is hopeful to overcome these problems. Trends and patterns of karyotype evolution in Lilieae Based on karyotype asymmetry, the distribution pattern of secondary constrictions and intercalary satellites, and the previous and present molecular studies, the karyotype evolution trends and patterns were clarified: it did not simply followed a uniform direction or tendency, but acted on several aspects of karyotype variations. Indeed, the variations happened on karyotypes in different ways by genus, which shaped the extant genera of Lilieae. Notholirion In this genus no secondary constrictions were found, and its basal position in the molecular tree showed its relatively ancestral status. However, the asymmetry reflecting by TF%, Syi, A1, A and As K% proved that it was more asymmetry. Though in CVCI and CVCL it’s not the highest value, it occupied a moderate position (Fig. 3g, i). This indicated that the karyotype evolution in Lilieae was not followed a simple ascent or descent direction. Notholirion’s intrachromosomal asymmetry raised mainly under the pattern that additional DNA were added on the long arms of smaller chromosomes (Peruzzi et al. 2009). By ‘smaller’ we mean the first two pairs were not included. In Notholirion its intrachromosome variation were also caused by the second pair, for it hold a unordinary short arm comparing with other genera (Fig. 2). Cardiocrinum This genus have a lower intrachromosome asymmetry degree than Notholirion and little higher than Fritillaria. J Plant Res (2012) 125:55–69 Considering the highest CVCI and CVCL values in Lilieae, this genus was evolved along with the shift of centromeres and/or the lost of partial DNA on the short arms in smaller chromosomes. This will cause the intrachromosome asymmetry arise, while the increase of CVCL indicated the additional DNA was not totally added follow the way in Notholirion, instead it’s more likely added on the larger chromosomes’ short arms. It’s clearly that the second pair in this genus was significantly longer than that in Notholirion, which might be in charge for the highest value of CVCL. Fritillaria In this genus the intrachromosomal asymmetry became further lower when compared with that in Cardiocrinum, but not significant. However, the CVCI and CVCL values were down to the lowest which completely in a reverse direction against its sister group. This means in the genus Fritillaria there were additional DNA added on the smaller chromosomes as it did in Notholrion. The difference was the locations where it added. Apparently more additional DNA was added on the short arms, which explained why CVCI was decreased. Therefore, in the genus Fritillaria the karyotype evolution mainly depended on the increase of the short arms length. Lilium ? Nomocharis This was a controversy group, for many research suggested to accommodate Nomocharis into Lilium (Hayashi and Kawano 2000; Nishikawa et al. 1999, 2001; Rønsted et al. 2005; Peruzzi et al. 2009). Indeed every parameter or index in present suggested their closed relationship. The median value and the interquartile range (IQR) in this research showed that Nomocharis were totally embedded in Lilium (Fig. 3). In our study more species of Nomocharis were included and the result showed that they formed a monophyletic group with other Lilium species. The Lilium– Nomocharis group had an intrachromosome asymmetry value higher than Cardiocrinum and Fritillaria, but lower when compared with Notholirion, while the CVCI and CVCL values were moderate. Based on these the changes in this group might resulted from the additional DNA added on the long arms of smaller chromosomes, as it did in Notholirion. At the same time, pericentric inversions and/or differential translocations of DNA between smaller and larger chromosomes happened (Peruzzi et al. 2009), which might cause the emergence of intercalary satellites adjacent to the centromeres. To sum up, in tribe Lilieae there was no uniform trend in karyotype evolution, and the genera under it were evolved in different ways. The conclusions made here were based 67 on the extant species of Lilieae, and the origin of these five genera was still a mystery, especially the giant chromosomes compared with its sister groups (tribe Tulipeae). However, combining more data including karyotype and molecular could shed more light on systematic researches in a special group, which will further give more insights on phylogeny. Acknowledgment We thank Dr. Yan Yu for providing the karyotype analyses tool package (NucType ver. 1.10, http://mnh.scu. edu.cn/soft/blog/nuctype/) for this study. This work was supported by the National Natural Science Foundation of China (31070166), Doctoral Fund of Ministry of Education of China (20090181110064), the Basic Research Program from the Ministry of Science and Technology of China (Grant No. 2007FY110100) and the Research Fund for the Large-scale Scientific Facilities of the Chinese Academy of Sciences (2009-LSF-GBOWS-01). References Angiosperm Phylogeny Group (2009) An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG III. Bot J Linn Soc 161(2):105–121 Arano H (1963) Cytological studies in subfamily Carduoideae (Compositae) of Japan. IX. The karyotype analysis and phylogenic considerations on Pertya and Ainsliaea. Botanical Magazine (Tokyo) 76:32–39 Ceccarelli M, Minelli S, Maggini F, Cionini PG (1995) Genome size variation in Vicia faba. Heredity 74:180–187 Comber HF (1949) A new classification of genus Lilium. Royal Horticultural Society of the Liliy Year Book, vol 13. p 85 Das AB, Mohanty S, Marrs RH, Das P (1999) Somatic chromosome number and karyotype diversity in fifteen species of Mammillaria of the family Cactaceae. Cytobios 97:141–151 De Melo Nationiel F, Guerra M, Benko-Iseppon AM, De Menezes NL (1997) Cytogenetics and cytotaxonomy of Velloziaceae. Plant Syst Evol 204:257–273 Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem Bull 19:11–15 Fay MF, Chase MW, Rønsted N et al (2006) Phylogenetics of Liliales: summarized evidence from combined analyses of five plastid and one mitochondrial loci. In: Columbus JT, Friar EA, Porter JM, Prince LM, Simpson MG (eds) Monocots: comparative biology and evolution (excluding Poales). Rancho Santa Ana Botanic Garden, Claremont, pp 559–565 Gao YD, Zhou SD, He XJ (2009) Karyotype of four genera in Liliaceae (s. str.) from Hengduan Mountains of Southwest China. Acta Botanica Yunnancia 31:399–405 González-Aguilera JJ, Fernández-Peralta AM (1984) Phylogenetic relationships in the family Resedaceae. Genetica 64:185–198 Greilhuber J, Speta F (1976) C-banded karyotypes in the Scilla hohenackeri group, S. persica and Puschkinia (Liliaceae). Plant Syst Evol 126:149–188 Hayashi K, Kawano S (2000) Molecular systematics of Lilium and allied genera (Liliaceae): phylogenetic relationships among Lilium and related genera based on the rbcL and matK gene sequence data. Plant Species Biol 15:73–93 Hong DY (1990) Plant cytotaxonomy. Science Press, Beijing Huziwara Y (1962) Karyotype analysis in some genera of Compositae. VIII. Further studies on the chromosome of Aster. Am J Bot 49:116–119 123 68 İkinci N, Oberprieler C, Güner A (2006) On the origin of European lilies: phylogenetic analysis of Lilium section Liriotypus (Liliaceae) using sequences of the nuclear ribosomal transcribed spacers. Willdenowia 36:647–656 Kelchner SA (2000) The evolution of non-coding chloroplast DNA and Its application in plant systematics. Ann Missouri Bot Garden 87:482–498 Liang SY (1995) Chorology of Liliaceae (S. Str.) and its bearing on the Chinese flora. Acta Phytotaxonomica Sinica 33:27–51 Liang SY, Tamura M (2000) In: Wu ZY, Raven PH (eds.) Flora of China, vol 24. Science Press/Missouri Botanical Garden Press, Beijing/St. Louis, pp 135–159 Lifante ZD (1996) A karyological study of Asphodelus L. (Asphodelaceae) from the Western Mediterranean. Bot J Linn Soc 121:285–344 Morrison DA (2009) A framework for phylogenetic sequence alignment. Plant Syst Evol 282:127–149 Muratović E, Bogunić F, Šoljan D, Siljak-Yakovlev S (2005) Does Lilium bosniacum merit species rank? A classical and molecularcytogenetic analysis. Plant Syst Evol 252:97–109 Muratović E, Robin O, Bogunić F, Šoljan D, Siljak-Yakovlev S (2010a) Karyotype evolution and speciation of European lilies from Lilium sect. Liriotypus. Taxon 59:165–175 Muratović E, Hidalgo O, Garnatje T, Siljak-Yakovlev S (2010b) Molecular phylogeny and genome size in European Lilies (Genus Lilium, Liliaceae). Adv Sci Lett 3:180–189 Narayan RKJ, Rees H (1976) Nuclear DNA variation in Lathyrus. Chromosoma 54:141–154 Nishikawa T, Okazaki K, Uchino T, Arakawa K, Nagamine T (1999) A molecular phylogeny of Lilium in the internal transcribed spacer region. J Mol Evol 49:238–249 Nishikawa T, Okazaki K, Arakawa K, Nagamine T (2001) Phylogenetic analysis of section Sinomartagon in genus Lilium using sequences of the internal transcribed spacer region in nuclear ribosomal DNA. Breed Sci 51:39–46 Noda S (1991) Chromosomal variation and evolution in the genus Lilium. In: Tsuchiya T, Gupta PK (eds) Chromosome engineering in plants: genetics, breeding, evolution, part B. Elsevier, Amsterdam, pp 507–524 Nylander JAA (2004) MrModeltest 2.0. Program distributed by the author. Department of Systematic Zoology, EBC, Uppsala University, Uppsala Paszko B (2006) A critical review and a new proposal of karyotype asymmetry indices. Plant Syst Evol 258:39–48 Patterson TB, Givnish TJ (2002) Phylogeny, concerted convergence, and phylogenetic niche conservatism in the core Liliales: Insights from rbcL and ndhF sequence data. Evolution 56:233– 252 Peruzzi L, Leitch IJ, Caparelli KF (2009) Chromosome diversity and evolution in Liliaceae. Ann Bot 103:459–475 Raina SN, Rees H (1983) DNA variation between and within chromosome complements of Vicia species. Heredity 51:335–346 Ronquist F, Huelsenbeck JP (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572– 1574 Rønsted N, Law S, Thornton H, Fay MF, Chase MW (2005) Molecular phylogenetic evidence for the monophyly of Fritillaria and Lilium (Liliaceae; Liliales) and the infrageneric classification of Fritillaria. Mol Phylogenet Evol 35:509–527 Rudall PJ, Stobart KL, Hong WP, Conran JG, Funess CA, Kite GC, Chase MW (2000) Consider the lilies-systematics of Liliales. In: Wilson KL, Morrison DA (eds) Monocots: systematics and evolution. CSIRO Publishing, Australia, pp 347–359 Shan F, Yan G, Plummer JA (2003) Karyotype evolution in the genus Boronia (Rutaceae). Bot J Linn Soc 142:309–320 123 J Plant Res (2012) 125:55–69 Siljak-Yakovlev S, Peccenini S, Muratovic E, Zoldos V, Robin O, Vallés J (2003) Chromosomal differentiation and genome size in three European mountain Lilium species. Plant Syst Evol 236:165–173 Smyth DR, Kongsuwan K, Wisudharomn S (1989) A survey of C-band patterns in chromosomes of Lilium (Liliaceae). Plant Syst Evol 163:53–69 Stace HM (1978) Cytoevolution in the genus Calotis R. Br. (Compositae: Astereae). Aust J Bot 26:287–307 Stebbins GL (1971) Chromosomal evolution in higher plants. Edward Arnold, London Stewart RN (1947) The morphology of somatic chromosomes in Lilium. Am J Bot 34:9–26 Swofford DL (2003) PAUP*. Phylogenetic analysis using Parsimony (*and other methods). Version 4. Sinauer Associates, Sunderland Takhtajan A (1997) Diversity and classification of flowering plants. Columbia University Press, New York Tamura MN (1998) Liliaceae. In: Kubitzki K (ed) The families and genera of vascular plants. III. Flowering plants-monocotyledons, Lilianae (except Orchidaceae). Springer, Berlin, pp 343–353 Tamura MN, Yamashita J, Fuse S, Haraguchi M (2004) Molecular phylogeny of monocotyledons inferred from combined analysis of plastid matK and rbcL gene sequences. J Plant Res 117:109–120 Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol Biol Evol 24:1596–1599 The State Pharmacopoeia Commission of the People’s Republic of China (2000) Pharmacopoeia of the People’s Republic of China. Chemical Industry Press, Beijing Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The CLUSTAL_X windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25:4876–4882 Vanzela ALL, Ruas PM, Marin-Morales MA (1997) Karyotype studies of some species of Dalechampia Plum. (Euphorbiaceae). Bot J Linn Soc 125:25–33 Venora G, Blangiforti S, Ruffini Castiglione M, Pignone D, Losavio F, Cremonini R (2002) Chromatin organisation and computer aided karyotyping of Triticum durum Desf. cv Timilia. Caryologia 55:91–98 Wang FZ, Tang J (1980) Lilium L. In: Flora Reipublicae Popularis Sinicae, vol 14. Science Press, Beijing, pp 116–157 Watanabe K, King RM, Yahara T, Ito M, Yokoyama J, Suzuki T, Crawford DJ (1995) Chromosomal cytology and evolution in Eupatorieae (Asteraceae). Ann Missouri Bot Garden 82:581–592 Watanabe K, Yahara T, Denda T, Kosuge K (1999) Chromosomal evolution in the genus Brachyscome (Asteraceae, Astereae): statistical tests regarding correlation between changes in karyotype and habit using phylogenetic information. J Plant Res 112:145–161 Wen J (1999) Evolution of eastern Asian and eastern North American disjunct distributions in flowering plants. Annu Rev Ecol Syst 30:421–455 Wen J (2001) Evolution of Eastern Asian–Eastern North American biogeographic disjunctions: a few additional issues. Int J Plant Sci 162(6): S117–S122 (Supplement: historical biogeography of the Northern Hemisphere (Nov. 2001)) Wen J, Zimmer EA (1996) Phylogeny and biogeography of Panax L. (the Ginseng Genus, Araliaceae): inferences from ITS Sequences of nuclear ribosomal DNA. Mol Phylogenet Evol 6(2):167– 177 Wen J, Shi SH, Jansen RK, Zimmer EA (1998) Phylogeny and biogeography of Aralia sect. Aralia (Araliaceae). Am J Bot 85:885–875 J Plant Res (2012) 125:55–69 Wu ZY, Li H, Yang CR (1994) Cytogeography and phylogeny of Lilieae. Acta Botanica Yunnanica 1994(Suppl. VI):101–112 Xiang QY, Soltis DE (2001) Dispersal-vicariance analyses of intercontinental disjuncts: historical biogeographical implications for angiosperms in the Northern Hemisphere. Int J Plant Sci 162: S29–S39 (Supplement: Historical Biogeography of the Northern Hemisphere (Nov. 2001)) 69 Xiang QY, Soltis DE, Soltis PS, Manchester SR, Crawford DJ (2000) Timing the eastern Asian–eastern North American floristic disjunction: molecular clock corroborates paleontological estimates. Mol Phylogenet Evol 15:462–472 Zarco CR (1986) A new method for estimating karyotype asymmetry. Taxon 35:526–530 123