American Journal of Botany 87(2): 273±292. 2000. A PHYLOGENY OF THE FLOWERING PLANT FAMILY APIACEAE BASED ON CHLOROPLAST DNA RPL16 AND RPOC1 INTRON SEQUENCES: TOWARDS A SUPRAGENERIC CLASSIFICATION OF SUBFAMILY APIOIDEAE1 STEPHEN R. DOWNIE,2,4 DEBORAH S. KATZ-DOWNIE,2 AND MARK F. W ATSON3 2Department of Plant Biology, University of Illinois, Urbana, Illinois 61801 USA; and 3Royal Botanic Garden Edinburgh, 20A Inverleith Row, Edinburgh EH3 5LR, Scotland, UK The higher level relationships within Apiaceae (Umbelliferae) subfamily Apioideae are controversial, with no widely acceptable modern classi®cation available. Comparative sequencing of the intron in chloroplast ribosomal protein gene rpl16 was carried out in order to examine evolutionary relationships among 119 species (99 genera) of subfamily Apioideae and 28 species from Apiaceae subfamilies Saniculoideae and Hydrocotyloideae, and putatively allied families Araliaceae and Pittosporaceae. Phylogenetic analyses of these intron sequences alone, or in conjunction with plastid rpoC1 intron sequences for a subset of the taxa, using maximum parsimony and neighbor-joining methods, reveal a pattern of relationships within Apioideae consistent with previously published chloroplast DNA and nuclear ribosomal DNA ITS based phylogenies. Based on consensus of relationship, seven major lineages within the subfamily are recognized at the tribal level. These are referred to as tribes Heteromorpheae M. F. Watson & S. R. Downie Trib. Nov., Bupleureae Spreng. (1820), Oenantheae Dumort. (1827), Pleurospermeae M. F. Watson & S. R. Downie Trib. Nov., Smyrnieae Spreng. (1820), Aciphylleae M. F. Watson & S. R. Downie Trib. Nov., and Scandiceae Spreng. (1820). Scandiceae comprises subtribes Daucinae Dumort. (1827), Scan- dicinae Tausch (1834), and Torilidinae Dumort. (1827). Rpl16 intron sequences provide valuable characters for inferring high-level relationships within Apiaceae but, like the rpoC1 intron, are insuf®cient to resolve relationships among closely related taxa. Key words: Apiaceae; Apioideae; Hydrocotyloideae; molecular phylogeny; rpl16 intron; rpoC1 intron; Saniculoideae; Umbelliferae. The ¯owering plant family Apiaceae Lindl. (Umbellif- despite their large size, widespread distribution, and eco- erae Juss.) comprises 300±455 genera and some 3000± nomic importance, no widely acceptable modern classi- 3750 species (Constance, 1971; Pimenov and Leonov, ®cation is available. 1993). It is cosmopolitan, being particularly abundant in The most recent treatment of the family (Pimenov and the northern hemisphere. Daucus carota subsp. sativus Leonov, 1993) is but an adaptation of the century-old (Hoffm.) Arcang., the common cultivated carrot, is by far system of Drude (1898), highly criticized for using subtle its most economically important member. Other familiar or poorly de®ned diagnostic characters (Heywood, vegetables, ¯avorings, or garnishes include angelica, an- 1982a). Radically different classi®cations exist (such as ise (aniseed), caraway, celeriac, celery, chervil, coriander those of Koso-Poljansky, 1916, and Cerceau-Larrival, (cilantro), cumin, dill, fennel, lovage, parsley, and pars- 1962), but have proved unworkable in practice and are nip. Deadly poisonous plants include water hemlock, poi- rarely used. Drude recognized three subfamilies of Api- son hemlock, hemlock water-dropwort, and fool's pars- aceae (Apioideae, Saniculoideae, and Hydrocotyloideae), ley. The obvious distinctive characters of many of these dividing each into a series of tribes and subtribes. Mo- plants, such as herbs with hollow or pith-®lled stems, lecular systematic investigations have con®rmed the pinnately divided leaves with sheathing bases, small un- monophyly of Apioideae and demonstrated its sister- specialized ¯owers in compound umbel in¯orescences, group status to subfamily Saniculoideae, but have also and specialized fruits, make them easily identi®able to shown that all of Drude's tribes (and other reclassi®ca- family (likely making them one of the ®rst families of tions of the family) are largely unsound (Downie and ¯owering plants to be generally recognized). However, Katz-Downie, 1996; Downie, Katz-Downie, and Cho, 1996; Kondo et al., 1996; Plunkett, Soltis, and Soltis, 1 Manuscript received 5 January 1999; revision accepted 3 June 1999. 1996a, b, 1997; Downie et al., 1998; Valiejo-Roman et The authors thank L. Constance, R. Hartman, J. Lahham, B.-Y. Lee, M. Pimenov, A. Troitsky, C. Valiejo-Roman, and the many botanic gar- al., 1998; Katz-Downie et al., 1999; Plunkett and Down- dens cited in the text for generously providing leaf, seed, or DNA ma- ie, 1999). Umbellifers display a remarkable array of mor- terial; K.-J. Cho, B.-Y. Lee, E. Llanas, and J. Luttrell for laboratory phological and anatomical modi®cations of their fruits, assistance; R. Mill for the Latin translations; and K. Spalik and two many of which are adaptations for various modes of seed anonymous reviewers for comments on the manuscript. This work was dispersal. Not surprisingly, these characters are prone to supported by grants to S. Downie from the National Science Foundation convergence, and their almost exclusive use to delimit (DEB-9407712) and the Campus Research Board of the University of Illinois. suprageneric groups has confounded estimates of rela- 4 Author for correspondence (e-mail: [email protected]; FAX: tionship. 217-244-7246). Our goal over the past few years, and that of our col- 273 274 AMERICAN JOURNAL OF BOTANY [Vol. 87 laborators, has been to resolve the ``higher level'' rela- acters, such as nuclear ribosomal DNA ITS (Downie et tionships within subfamily Apioideae. This is necessary al., 1998; Katz-Downie et al., 1999) and chloroplast matK in order to provide the framework for ``lower level'' re- (Plunkett, Soltis, and Soltis, 1996b) sequences, and chlo- visions of particular tribes and complexes of genera, so roplast restriction sites (Plunkett and Downie, 1999). important in such a group of plants where suprageneric Based on consensus of relationship, we take the ®rst steps relationships have been largely speculative and ever towards a ``new Drude'' by formally recognizing seven changing. Eventually, this will lead to the production of groups of apioids at the tribal level and, in so doing, a modern classi®cation (i.e., a ``new Drude''). To achieve provide the requisite framework for ``lower level'' sys- this goal, a variety of molecular characters have been tematic study. used, such as chloroplast gene (rbcL, matK) and intron (rpoC1, rps16), and nuclear ribosomal DNA internal MATERIALS AND METHODS transcribed spacer (ITS) sequences. A recent study ex- amined restriction site variation of chloroplast DNA Plant accessionsÐOne hundred and nineteen species from 99 genera (cpDNA; Plunkett and Downie, 1999); further examina- of Apiaceae subfamily Apioideae, ®ve species (®ve genera) each from tion of chloroplast genomic structure is in progress (G. Apiaceae subfamilies Hydrocotyloideae and Saniculoideae, 11 species (ten genera) of Araliaceae, and seven species (®ve genera) of Pitto- Plunkett and S. Downie, unpublished data). While these sporaceae were examined for rpl16 intron sequence variation (Table 1). characters have been important in providing insight into In total, 147 species representing 124 genera were considered, with 84 evolutionary relationships, not all have been useful at the of these species included in a previous phylogenetic analysis of rpoC1 same hierarchical level. Moreover, because many existing introns (Downie et al., 1998). RpoC1 intron sequences for Billardiera data sets are not parallel in construction, opportunities to scandens and Bursaria spinosa (Pittosporaceae) were procured as part combine data have been few. of this study, for a total of 86 matching rpl16 and rpoC1 intron se- Noncoding regions of cpDNA, such as introns and in- quences (Table 1). With the exception of Anethum graveolens and Crith- tergenic spacers, tend to evolve more rapidly than coding mum maritimum, where different accessions of the same species were loci, both in nucleotide substitutions and in the accumu- examined, both rpl16 and rpoC1 intron data for these 86 species were lation of insertion and deletion events (indels), presum- obtained from precisely the same specimens. ably because they are less functionally constrained (Cur- tis and Clegg, 1984; Palmer, 1991; Clegg et al., 1994). Experimental strategyÐLeaf material for DNA extraction was ob- Because these noncoding regions can potentially supply tained either directly from the ®eld, from plants cultivated from seed in more informative characters than coding regions of com- the greenhouse, or from accessioned plants cultivated at several botanic parable size, they have become popular for phylogenetic gardens (Table 1). For some species, DNAs were extracted from her- studies among taxa that are recently diverged. The chlo- barium specimens or supplied to us directly. All plants cultivated at the roplast gene rpl16, encoding the ribosomal protein L16 University of Illinois at Urbana-Champaign (UIUC), Moscow State (Posno, Van Vliet, and Groot, 1986), is interrupted by an University, and the Royal Botanic Garden Edinburgh (RBGE) are intron in many, but not all, land plants (Campagna and vouchered at ILL, MW, and E, respectively (herbarium acronyms ac- cording to Holmgren, Holmgren, and Barnett, 1990). Details of the Downie, 1998). In most ¯owering plants, this intron is DNA extraction procedures have been presented in Downie and Katz- ;1 kilobase