Biologia 65/6: 997—1003, 2010 Section Botany DOI: 10.2478/s11756-010-0105-8
Molecular phylogeny of Ranunculaceae based on rbc L sequences Ying-fan Cai1*† , Sheng-wei Li2, Min Chen2, Ming-feng Jiang2† , Yi Liu1, Yong-fang Xie1, Quan Sun1, Huai-zhong Jiang1, Neng-wen Yin1, Ling Wang1, Rui Zhang1, Cheng-lin Huang1 & Kairong Lei3 1
Chongqing University of Posts and Telecommunications, Chongqing 400065, People’s Republic of China; e-mail: [email protected]
2 Southwest University for Nationalities, Chengdu 610041, People’s Republic of China 3 Chongqing Key Laboratory of Adversity Agriculture, Chongqing 401329,People’s Republic of China
Abstract: A phylogenetic tree was constructed by sequencing rbcL genes of 33 species representing 19 genera of Ranunculaceae, and three related species, Mahonia bealei, Mahonia fortunei and Nandina domestica. The results showed that the rbcL sequences of these Ranunculaceae range from 1,346 bp to 1,393 bp. The results based on the phylogenetic tree indicated that Caltha and Trollius should not be put in the same tribe, and a close relationship betweenAdonis and Trollius is supported by our research, while Aquilegia should be in Thalictroideae. In combination with the morphological and chemical evidence, the generic classiﬁcation of Ranunculaceae should be revised into ﬁve subfamilies: Hydrastidoideae, Coptidoideae, Helleboroideae, Thalictroideae and Ranunculoideae. We demonstrate that the rbcL gene is of great value for investigating generic to subfamilial relationships in Ranunculaceae. Key words: phylogeny; Ranunculaceae; rbcL
Au th or
Abbreviations: rbcL, ribulose-1,5-bisphosphate carboxylase/oxygenase; IPTG, isopropyl β-D-1-thiogalactopyranoside; XGal, 5-bromo-4-chloro-3-indolyl-beta-D-galactopyranoside
The Ranunculaceae comprises about 2,500 described species distributed amongst 59 genera throughout the world, but mostly in temperate and cold areas of the northern hemisphere (Tamura 1993; Wu et al. 2003). Ranunculaceae, which is considered pharmaceutically important, is also of phylogenetic importance (Tamura 1993). More than 30 genera and about 220 species have been used as herbal medicine in China, and in many other countries for a variety of uses (e.g. antibiosis, congestion, fever, cancer, arrhythmia, malaria; Xiao 1980). In China, Ranunculaceae distribute mainly in the southwest (Delectis Florae Reipublicae Popularis Sinicae Agendae, Academiae Sinicae Edita 1979). The chemistry and taxonomy of Ranunculaceae is varied and complex within and amongst species. To clarify the relationships amongst subfamilies of Ranunculaceae, and correlations amongst their phylogeny, chemistry and pharmacology, many studies have examined taxonomic characters. According to chromosome number and ﬂoral characteristics, Tamura et al. (1966, 1993) recognised six subfamilies, namely Hydrastidoideae, Thalictroideae, Ranunculoideae, Helle-
boroideae, Coptidoideae and Isopyroideae. Based on nuclear 26S ribosomal DNA, Ro et al. (1997) suggested classiﬁcation into four subfamilies, Hydrastidoideae Raﬁnesque, Coptidoideae Tamura, Thalictroideae Raﬁnesque and Ranunculoideae Arnott. Peng’s (2006a) results, based on pharmaphylogenetic research, were in accordance with the phylogenetic analysis of Tamura (1966, 1993), and also supported the establishment of Cimifugoideae in light of their chemical composition, with Isopyrum as a transitional group (Peng et al. 2006b). Other studies have used cytology (Yang 2001; Lehnebach 2007), serological approaches (Jensen 1966, 1968) and cladistic analyses (Leconte & Estes 1989; Hoot 1991). Some of these studies are roughly compatible with the current classiﬁcation of Ranunculaceae, but this information should be carefully evaluated in light of independent phylogenetic estimates, especially molecular sequencing. In recent years, plant molecular systematics and phylogenetics have been used to supplement classical taxonomy. Because of diﬀerences in evolutionary rates, plant DNA sequences can be used to examine classiﬁcation. However, only a small number of molecular markers have currently proved useful for phylo-
* Corresponding author † Ying-fan Cai and Sheng-wei Li contributed equally to this work.
c 2010 Institute of Botany, Slovak Academy of Sciences
Y.-F. Cai et al.
998 Table 1. RbcL sequences identiﬁed in this study. Species
Aconitum carmichaeli Aconitum racemulosum Aquilegia vulgaris Clematis armandi Clematis finetiana Clematis gratopsis Clematis montona Coptis chinensis Mahonia bealei Mahonia fortunei Nandina domestica Ranunculus cantoniensis Ranunculus japonicus Ranunculus sieboldii Thalictrum simplex
Mt. JINFO, Chongqing, China Mt. JINFO, Chongqing, China Mt. JINFO, Chongqing, China Mt. JINFO, Chongqing, China Mt. JINFO, Chongqing, China Mt. JINFO, Chongqing, China Mt. JINFO, Chongqing, China Shizhu Chongqing, China Chongqing Academy of Chinese Chongqing Academy of Chinese Chongqing Academy of Chinese Mt. JINFO Chongqing, China Chongqing Academy of Chinese Mt. JINFO, Chongqing, China Jianyang, Sichuan, China
S. Q. Sun T. Y. Zhang T. Y. Zhang T. Y. Zhang S. Y. Chen W. T. Wang C. Ho W. T. Wang Z. O. Gu Z. O. Gu T. Y. Zhang H. X. Luo J. L. Li C. Ho S. Y. Chen
Materia Medica Materia Medica Materia Medica Materia Medica
SM SM SM SM SM SM SM SM SM SM SM SM SM SM SM
0322 1043 793 0061 2653 0574 96482 1137 543 589 0348 0027 26297 1751 6572
FJ449849 FJ449850 FJ449851 FJ449852 FJ449853 FJ449854 FJ449855 FJ449856 FJ449858 FJ449857 FJ449859 FJ449861 FJ449862 FJ449860 FJ449863
Genomic DNA extraction and PCR amplification Genomic DNA was extracted from freshly frozen leaf material using Genomic DNA extraction reagent kits. According to the rbcL gene sequences in GenBank data of relative plants the primer of rbcL sequences were designed with Primer 5 software. which are primer rbcL-F: 5’-TTC AAA GCG GGT GTT AAA GAT TA-3’ and primer rbcL-R: 5’GAT TGG GCC GAG TTT AAT TGC-3’. The ampliﬁcation reactions were carried out in a ﬁnal volume of 20 µL containing 2 µL DNA (100 ng), 5 pmol each primer and 0.5 U Taq DNA polymerase in 1.8 mM MgCl2 , 100 mM dNTP and 10× reaction buﬀer. The cycling parameters were: 94 ◦C for 5 min followed by 35 cycles of 94 ◦C 1min, 55 ◦C 1.5min, 72 ◦C 1min, and a ﬁnal extension for 5 min at 72 ◦C.
Au th or
genetic inference within ﬂowering plants (Ro 1997). The ribulose-1,5–bisphosphate carboxylase/oxygenase (rbcL) gene from the chloroplast genome has proved suitable for phylogenetic analyses (Ritland & Clegg 1987; Zurawski & Clegg 1987), with information now available on its structure and function (Kellogg & Juliano 1997), evolutionary rate (Bousquet et al. 1992) and systematic signiﬁcance in classiﬁcation (Kellogg & Juliano 1997). rbcL sequences are now commonly applied to study molecular plant phylogeny. Coding genes such as rbcL are likely informative to resolve phylogenetic issues ranging from higher taxonomic ranks to relationships amongst seed plant lineages (Tian & Li 2002). In this research, we analysed the cpDNA rbcL sequences of Ranunculaceae and the related genera Mahonia, Nandina and Hydrastis. A preliminary molecular basis was provided to distinguish Ranunculaceae containing diﬀerent chemical components, such as magnoﬂorine and ranunculin. The research will be useful to ﬁnd new drugs, as well as new information on relationships within the Ranunculaceae.
Specimen No. GenBank Accession No.
Material and methods
Plant materials The materials used in the present study are listed in Table 1. Coptis Chinensis was from Shizhu of Chongqing provided by Chen Daxia, Ranunculus japonicus, Mahonia fortunei, Mahonia bealei and Nandina domestica were collected from the Medical Plants Garden of Chongqing Academy of Chinese Materia Medica, and all the other species were collected from Mt. Jinfo, Chongqing. In addition, rbcL sequences of 21 species were downloaded from GenBank. Totally 36 species out of 19 genera were included in the ﬁnal analyses (Tables 1, 2). Strain and agent Escherichia coli DH5α were preserved in the laboratory, pMD18-T Vector purchased from TaKaRa company. Genomic DNA extraction reagent kits, Taq DNA polymerase, dNTPs and MgCl2 were purchased from HUASHUN Biotechnology Company in SHANGHAI; DNA Ladder was purchased from TIANGEN Biotechnology Company in Beijing.
T/A clone and sequencing Using the Agarose Gel DNA Puriﬁcation Kit (TaKaRa) to purify the gene fragments, isolated by 1.0% agarose gel electrophoresis, quantitative PCR puriﬁcation products insert to the pMD18-T cloning vector and transform into the competent E. coli DH5α, then the cells were incubated in a growth medium and ﬁnally spread on an agar LB medium plate which contained ampicillin, IPTG and X-gal then incubated at 37 ◦C for 12 ∼ 16 h. Picking the white monoclonal colony to incubate in the LB liquid culture medium, as well, extracting the plasmid which was digested by restriction enzymes and sequenced. Plasmid DNA sequence was completed by Beijing Sunbiotech co., Ltd. (ABI3730XL DNA Sequencer). Analysis of DNA sequence data and construction of a phylogenetic tree The NCBI website and Blastn online searches were used to analyse the target gene, and sites with missing data or gaps were excluded from all following analyses. Neighbourjoining (NJ) analysis (Saitou & Nei 1987) was performed using the MEGA software (Kumar et al. 1993). The Jukes– Cantor model of nucleotide substitution was selected for analyses based on Nei’s (1991) guidelines for choosing the most appropriate distance measure. Maximum parsimony (MP) analysis was used to implement the heuristic search procedure, with 100 replications with random addition of taxa to reduce possible bias from the input order. The reliability of clustering patterns in trees was tested by bootstrapping, in the case of both NJ and MP trees, and by standard error testing for internal branches of NJ trees (Rzhetsky & Nei 1992). In total, 2,000 bootstrap replications were used
Au th or
Molecular phylogeny of Ranunculaceae based on rbcL sequences
Fig. 1. Ranunculaceae group relationships inferred from Neighbor-Joining tree using Jukes-Cantor distances based on all pairwise comparison of rbcL sequences (1389bp). Integers above branches are bootstrap support values.
for the NJ and MP trees. Ranunculaceae group relationships were inferred from the NJ tree using Jukes–Cantor distances based on all pair-wise comparisons of rbcL sequences from 33 species and three out-group taxa in the Berberidaceae. Berberidaceae was chosen as the out-group because it contains several protoberberine compounds that are common to both families (Xiao 1980; Ro et al. 1997).
Result PCR amplification and sequencing of the rbcL gene The ampliﬁed rbcL sequences of all species were submitted to GenBank (Table 1). The length of the rbcL sequences included in the ﬁnal data matrix ranged from 1,346 to 1,393 bp. For the 36 species studied, we obtained 1,395 aligned sequences. Amongst these, 289 were variable and 203 informative for parsimony analysis. The average A:T:C:G was 27.3:28.1:19.9:24.7, with a narrow standard error around the means. There were 1,330 identical pairs and the average transition/transversion (ts/tv) ratio was 2.6, which is higher
than the expected ratio of 0.5 when all types of substitution are assumed to be equally likely. Phylogenetic analysis Clustal X software was applied to align the sequences of rbcL and a phylogenetic tree was constructed using the rbcL sequences and MEGA software. Ranunculaceae group relationships were inferred from the NJ tree using Jukes–Cantor distances based on all pairwise comparisons of rbcL sequences from 33 species of Ranunculaceae and three out-group taxa in Berberidaceae. The inferred phylogenies with NJ were nearly consistent (Fig. 1); interior branch P values (Pc) from the standard error test tended to be higher than bootstrap P values (Pb), and we considered values greater than or equal to 95% as statistically signiﬁcant. Pb values above 70% are generally considered informative (Hillis & Bull 1993; Ro et al. 1997). In Ranuculoideae groups, nearly all the species were concentrated together into genera (Figs 1, 2). Caltha and Trollius should not be put in the same tribe, according to NJ
Y.-F. Cai et al.
Au th or
Fig. 2. Ranunculaceae group relationships inferred from Maximum parsimony tree based on all pairwise comparison of rbcL sequences (1389bp). Integers above branches are bootstrap support values.
and MP trees. Groups 13, 14 and 15 were placed in Ranunculoideae, and group 18 was placed in the centre, between groups 17 and 19 (Fig. 2). However, all distance measures (P distance, Jukes–Cantor, Kimura two-parameter, Tajama–Nei and maximum composite likelihood) generated identical NJ topologies. Furthermore, the NJ tree topologies were closest to the Tamura classiﬁcation (Table 3), so we chose the NJ method for further analysis (Fig. 1). Discussion
Phylogenetic studies are based mainly on morphological and molecular characters. In this research, we sequenced and analysed rbcL genes of 36 species of Ranunculaceae and the related Mahonia bealei, Mahonia fortunei and Nandina domestica in the Berberidaceae. We thus obtained an initial understanding of the evolutionary relationship and process in Ranunculaceae and related plants.
Diﬀerences were found between our rbcL phylogeny and previous morphology-based taxonomies (Figs 1, 2). For the following discussion, Tamura’s (1993) classiﬁcation (Table 3) and nomenclature are followed throughout. Most clades found in NJ analyses (numbered in numerals) can be characterized by morphological or karyological features reported in the literature and might represent a basis for future revised classiﬁcation. Many species of the Ranunculaceae have interesting morphology and chemical structure, which could be mapped on an inferred topology derived from rbcL sequences. Analysis of rbcL sequences with in-groups and out-groups has proven useful for their classiﬁcation. The result showed that the genetic distance between Hydrastis and Ranunculoideae is large. However, the nearest genus to Hydrastis is Coptis in Ranunculoideae (Fig. 1); these have similar chemical composition, both contain berberine and isoquinoline alkaloids and other common chemical constituents, suggesting
Molecular phylogeny of Ranunculaceae based on rbcL sequences
GenBank Accession No
Actaea laciniata Adonis amurensis Aquilegia brevistyla Aquilegia ecalcarata Asteropyrum cavaleriei Beesia calthifolia Caltha appendiculata Caltha palustris Coptis trifolia Dichocarpum sutchuenense Enemion raddeanum Halerpestes cymbalaria Helleborus niger Helleborus thibetanus Hydrastis canadensis Myosurus minimus Pulsatilla cernua Ranunculus macranthus Thalictrum cultratum Thalictrum javanicum Trollius laxus
Anderson C.L. Wang W. Anderson C.L. Wang W. Wang X.Q. Wang X.Q. Wardle P. Silvertown J. Hoot S.B. Wang W. Wang W. Wang W. Anderson C.L. Wang W. Hoot S.B. Anderson C.L. Wang W. Leebens-Mack J. Anderson C.L. Wang W. Wang W.
DQ099449 AY954487 DQ099444 AY954495 AF079453 AF079452 AF307908 AY395532 AF093730 AY954493 AY954494 AY954490 DQ099436 AY954485 AF093725 DQ099441 AY954492 DQ069502 DQ099447 AY954496 AY954486
Table. 3. The classiﬁcation by Tamura (1993) for Ranunculaceae. Genus
Caltha Trollius Beesia Actaea Helleborus Aconitum
Group Group Group Group Group Group
17 8 15 14 16 6
Clematis Pulsatilla Adonis Ranunculus Myosurus Halerpestes
Group Group Group Group Group Group
5 4 7 1 3 2
Cimicifugeae Helleboreae Delphineae Ranunculoideae
Au th or
Group 12 Aquilegia Dichocarpum Group 10 Group 11 Enemion
Coptis Group 18 Asteropyrum Group 13
port that place at the base of phylogenetic trees. These authors consider that Hydrastis is a highly autapomorphic lineage that should be retained within the Ranunculaceae, while Hoot (1991) advocates placement of Hydrastis in the monotypic family Hydrastidaceae, closely related to the Ranunculaceae (Delectis Florae Reipublicae Popularis Sinicae Agendae, Academiae Sinicae Edita 1979). Combining pharmaphylogeny analysis, we also believe that Hydrastis should be retained within the Ranunculaceae, congruent with the results of Ro and colleagues (1997). Based on chromosome and ﬂoral characteristics, Tamura (1993) recognised ﬁve subfamilies in the Ranunculaceae (Hydrastidoideae, Thalictroideae, Isopyroideae, Ranunculoideae, Helleboroideae). The results of Peng (2006a) using pharmaphylogenetic analysis are in accordance with the phylogenetic system presented by Tamura (1993), and support the establishment of Cimifugoideae. However, our phylogenetic tree based on rbcL sequences shows that Actaea and Beesia were in a low bootstrap value branch, with only 15% support in the NJ tree and 25% in the MP tree. Hence, these results do not support the establishment of Cimifugoideae. Beesia, which is more primitive than Actaea, is included in our data and other research (Wang et al. 1993, 1998; Wang 1999). Caltha is generally considered the most original group of Helleboroideae (Smith 1928; Tamura 1987, 1990), as also supported in later research (Loconte & Estes 1989; Hoot 1991), and the two trees used in our research gave similar results. Caltha is not closely related to Trollius and these two genera should not be assigned to the same tribe (Xi et al. 1993). Furthermore, our phylogenetic tree supported a distance genetic relationship between Caltha and Trollius, consistent with Song et al. (2007) who used ﬂoral morphology. Our trees suggested that Adonis and Trollius are sister groups and that they have a close genetic distance (Ro et al. 1997), conﬁrming Tamura’s (1966, 1995) classiﬁcation. However, results from traditional taxonomy and molecular phylogeny are still disputed and further work is needed to clarify the relationship between Adonis and Trollius. A closely genetic relationship was observed among Thalictrum, Enemion and Aquilegia, strongly consistent with the cladogram of Ro et al. (1997) and the accepted Thalictroideae classiﬁcation (Delectis Florae Reipublicae Popularis Sinicae Agendae, Academiae Sinicae Edita 1979). Most Isopyroideae contain benzylisoquinoline and bisbenzylisoquinoline alkaloids, berberine and saponin. However, Thalictroideae mainly contain benzylisoquinoline and bisbenzylisoquinoline alkaloids, and most Coptidoideae contain berberine. Isopyroideae sits between Thalictroideae and Coptidoideae from an evolutionary viewpoint, and is considered a transitional group between subfamilies (Peng et al. 2006b). Combined with chemical constituents and the result of phylogenetic trees, we infer that Isopyroideae should not be treated as a subfamily and should be retained in Thalictroideae. Most Aquilegia species contain ﬂavonoids, saponins
Table 2. GenBank accession numbers of downloaded DNA sequences in this study.
they may have the same origin (Xiao 1980; Bill et al. 2007). Ro et al. (1997) used 26S ribosomal DNA to study intrafamilial relationships of the Ranunculaceae. Gu & Ren (2007) found that Coptidoideae was an early clade next to Hydrastis in Ranunculaceae and our results sup-
Y.-F. Cai et al.
Drummond J.R. & Hutchinsom J. 1920. A revision of Isopyrum (Ranunculaceae) and its nearer allies. Kew Bull. 1920: 145– 169. Fu D.Z. 1990. Phylogenetic considerations on the subfamily Thalictroideae (Ranunculaceae). Cathaya 2: 181–190. Gu T.Q. & Ren Y. 2007. Floral Morphogenesis of Coptis (Ranunculaceae) 24: 80–86. Hillis D.M. & Bull J.J. 1993. An empirical test of bootstrapping as amethod for assessing conﬁdence in phylogenetic analysis. Syst. Biol. 42: 182–192. Hoot S.B. 1991. Phylogeny of the Ranunculaceae based on epidermal microcharacters and macromorphology. Syst. Bot. 16: 741–755. Hoot S.B. 1995. Phylogeny of the Ranunculaceae based on preliminary atpB, rbcL and 18S nuclear ribosomal DNA sequence data. Pl. Syst. Evol. (Suppl.) 9: 241–251. Hutchinson J. 1923. Contributions towards a phylogenetic classiﬁcation of ﬂowering plants I. Kew Bull. 1923: 65–89. Jensen U. 1966. Die Verwandtschaftsverhaltnisse innerhalb der Ranunculaceae aus serologischer Sicht. Ber Deutsch Bot. Ges. 79: 407–412. Jensen U. 1968. Serologische Beitrage zur Systematik der Ranunculaceae. Bot. Jahrb. 88: 269–310. Kellogg E.A. & Juliano N.D. 1997. The structure and function of RuBisCO and their implications for systematic studies. Amer. J. Bot. 84: 413–428. Kumar S., Tamura K. & Nei M. 1993. ‘MEGA: Molecular Evolutionary Genetics Analysis’ ,Version 1.01. The Pennsylvania State Univ., University Park, PA. Lehnebach C.A., Cano A., Monsalve C., McLenachan P., Horandl E. & Lockhart P. 2007. Phylogenetic relationships of the monotypic Peruvian genus Laccopetalum (Ranunculaceae). Pl. Syst .Evol. 264: 109–116. Loconte H. & Estes J.R. 1989. Phylogenetic systematics of Berberidaceae and Ranunculales (Magnoliidae). Syst. Bot. 14: 565–579. Loconte H., Campbell L.M. & Stevenson D.W. 1995. Ordinal and familial relationships of Ranunculid genera. Pl. Syst. Evol. (Suppl.) 9: 99–118. Nei M. 1991. Relative eﬃciencies of diﬀerent tree-making methods for molecular data, pp. 90–128. In: Miyamoto M.M. & Cracraft J., (eds), Phylogenetic Analysis of DNA Sequences, Oxford Univ. Press, New York. Peng Y., Chen S.B., Chen S.L. & Xiao P.G. 2006a. Preliminary pharmaphylogenetic study on Ranunculaceae. China Journal of Chinese Materia Medica 31: 1124–1128. Peng Y., Chen S.B., Liu Y., Wang L.W. & Xiao P.G. 2006b. Preliminary pharmaphylogenetic study on Isopyroideae (Ranunculaceae). China Journal of Chinese Materia Medica 31: 1210–1214. Ritland K. & Clegg M.T. 1987. Evolutionary analysis of plant DNA sequences. Amer. Naturalist 30: 74–100. Ro K.E., Keener C.S. & Mcpheron B.A. 1997. Molecular phylogenetic study of the Ranunculaceae: Utility of the nuclear 26S ribosomal DNA in inferring intrafamiliar relationships. Mol. Phyl. Evol.8: 117–127. Ro K.E. & Mcpheron B.A. 1997. Molecular phylogeny of the Aquilegia group (Ranunculaceae) based on internal transcribed spacers and 5.8S nuclear ribosomal DNA. Biochem. Syst. Ecol 25: 445–461. Rzhetsky A. & Nei M. 1992. A simple method for estimating and testing minimum-evolution trees. Mol. Biol. Evol. 9: 945–967. Saitou N. & Nei M. 1987. The neighbor-joining method: A new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4: 406–425. Smith G.H. 1928. Vascular anatomy of ranalian ﬂowers, II. Ranunculaceae (continued), Menispermaceae, Calycanthaceae, Annonaceae. Bot. Gaz.85: 152–177. Song P. Tian X.H. & Ren Y. 2007. Floral morphogenesis of Caltha and Trollius (Ranunculaceae) and its systematic signiﬁcance. Acta Phytotax. Sin. 45: 769–782. Sun A.C. & Wang F.X. 1983. Contribution to the morphology and embryology of Asteropyrum peltatum. Bot. Res. 1: 85–90. Tamura M. 1966. Morphology, ecology and phylogeny of the Ranunculaceae. VI. Sci. Rep. Osaka Univ. 15: 13–35.
Au th or
and fewer benzylisoquinoline alkaloids and may be related to Thalictrum, which also contains saponins (Peng et al. 2006a). Our results supported that they are sister groups, with a bootstrap value of 99%, but the appearance of cyanogenic glycosides in Aquilegia showed that the genetic location of this genus is closer to Ranunculoideae. However, our results did not infer this relationship, and Aquilegia should belong to Thalictroideae. The genus Asteropyrum has not been disputed since it was established by Drummond & Hutchinson (1920), but its systematic position within Ranunculaceae has long been disputed. Based on fruit type and leaf characters, Hutchinson (1923) placed it in Helleboreae of Helleboroideae. This was supported by later studies (Yang et al. 1993, 1994; Ro & McPheron 1997). However, other studies consider Asteropyrum as a member of Thalictroideae (e.g. Fu 1990; Tamura 1992, 1993, 1995; Loconte et al. 1995) or Coptidoideae (e.g. Xiao & Wang 1964; Tamura 1968; Xiao 1980; Sun & Wang 1983). Moreover, a new tribe, Asteropyreae, has been established (Zhang 1982; Wang et al. 2005), but the NJ tree suggests that Asteropyrum may have undergone convergent evolution with Actaea and Beesia and should be placed within Helleboroideae (Fig. 1). Based on our current knowledge of morphological, karyological, chemical and molecular characters, we have demonstrated that the classiﬁcation of Ranunculaceae should be revised into at least ﬁve subfamilies: Hydrastidoideae (group 19), Coptidoideae (group 18), Helleboroideae (groups 13 to 17), Thalictroideae (groups 9 to 12) and Ranunculoideae (groups 1 to 8), which are clearly indicated in our results (Fig. 1). Our and previous research into molecular phylogeny clearly showed that application of molecular genetics, such as rbcL, is not only very useful in reconstructing taxonomic groups, but also reﬂects the evolutionary history of ﬂowering plants and provides important developmental information (Tian & Li 2002).
This work was supported by grants from the National Nature Science Foundation of China (No. 30771311), Natural Sciences Foundation of Chongqing, China (No. cstc2007 BB1328) and project of Chongqing Education Committee, China (KJ050510, KJ080504).
References Bill J.G., Ashley S., Gary W.B., Keith W., Philip B., Ryan Y., Leslie B.S., Martha A.H., Yudong T. & Sreekhar C. 2007. Eﬀect of goldenseal (Hydrastis canadensis) and kava kava (Piper methysticum) supplementation on digoxin pharmacokinetics in humans. Drug Metab. Dispos. 35: 240–245. Bousquet J., Strauss S.H. & Doerksen A.H. 1992. Extensive variation in evolutionary rate of gene sequences among seed plants. Proc. Natl. Acad. Sci. USA 89: 784–788. Delecctis Florae Reipublicae Popularis Sinicae Agendae Academiae Sinicae Edita 1979. Flora Reipublicae Popularis Sinicae, Tomus 27, Science Press, Beijing, 502 pp.
Molecular phylogeny of Ranunculaceae based on rbcL sequences
Wu Z.Y., Lu A.M. & Tang Y.C. 2003. The Families and Genera of Angiosperms in China, A Comprehensive Analysis. Science Press, Beijing, 378 pp. Xi Y.Z., Ning J.C. & Fu X.P. 1993. Pollen morphology of the tribe Trollieae and its taxonomic signiﬁcance. Cathaya 5: 115–130. Xiao P.G. . & Wang W.T. 1964. A new genus of Ranunuclaceae – Dichocarpum. Acta Phytotax. Sin.9: 315–333. Xiao P.G. 1980. A preliminary study of the correlation between phylogeny, chemical constituents and pharmaceutical aspects in the taxa of Chinese Ranunculaceae. Acta Phytotax. Sin. 18: 143–153 Yang Q.E., Gong X., Gu Z.J. & Wu Q.A. 1993. A karyomorphological study of ﬁve species in the Ranunculaceae from Yunnan, with a special consideration on systematic positions of Asteropyrum and Calathodes. Acta Bot. Yunnan. 15: 179– 190. Yang Q.E., Luo Y.B. & Hong D.Y. 1994. A karyotypic study of six species in the Ranunculaceae from Hunan in China. Guihaia 14: 27–36. Yang Q.E. 2001. Cytology of 12 species in Aconitum L. and of 18 species in Delphinium L. of the tribe Delphineae (Ranunculaceae) from China. Acta Phytotax. Sin. 39: 502–514 Zurawski G. & Clegg M.T. 1987. Evolution of higher-plant chloroplast DNA-coded genes: Implications for structure-function and phylogenetic studies. Ann. Rev. Pl. Phys. 38: 391–418.
Au th or
Tamura M. 1968. Morphology, ecology and phylogeny of the Ranunculaceae VIII. Sci. Rep. Osaka Univ. 17: 41–56. Tamura M. 1987. A classiﬁcation of genus Clematis. Acta Phytotax. Geobot. 38: 33–44. Tamura M. 1990. A new classiﬁcation of the family Ranunculaceae I. Acta Phytotax. Geobot. 41: 93–101. Tamura M. 1992. A new classiﬁcation of the family Ranunculaceae 3. Acta Phytotax. Geobot. 43: 53–58. Tamura M. 1993. Ranunculaceae, pp. 563–583. In: Kubitzki K. et al. (eds), The Families and Genera of Vascular Plants, Vol.2. Springer-Verlag, Berlin. Tamura M. 1995. Angiospermae. Ordnung Ranunculales. Fam. Ranunculaceae. II. Systematic Part, pp. 223–519. In: Hiepko P. (ed.), Nat¨ urliche Pﬂanzenfamilien, second ed., 17aIV. Duncker & Humblot, Berlin, Germany. Tian X. & Li D.Z. 2002. Application of DNA sequences in plant phylogenetic study. Acta Bot. Yunnan. 24: 170–184. Wang W., Li R.Q. & Chen Z.D. 2005. Systematic position of Asteropyrum (Ranunculaceae) inferred from chloroplast and nuclear sequences. Pl. Syst. Evol. 225: 41–54. Wang W.T., Li L.Q. & Wang Z. 1999. Notulae de Ranunculaceis sinensibus (XI). Acta Phytotax. Sin. 37: 209–219. Wang X.Q., Hong D.Y. & Li Z.Y. 1993. A study on pollen and seed coat in the tribe Cimicifugeae and some allied genera (Ranuncutaceae). Cathaya 5: 131–149. Wang X.Q., Deng Z.R. & Hong D.Y. 1998. The systematic position of Beesia: evidence from ITS (nrDNA) sequence analysis. Acta Phytotax. Sin. 36: 403–410.
Received September 3, 2009 Accepted June 8, 2010