Arias & Pires • Phylogeny of brassica crops and wild relatives TAXON 61 (5) • October 2012: 980–988 A fully resolved chloroplast phylogeny of the brassica crops and wild relatives (Brassicaceae: Brassiceae): Novel clades and potential taxonomic implications Tatiana Arias & J. Chris Pires Division of Biological Sciences, University of Missouri, 311 Bond Life Sciences Center, 1201 Rollins Street, Columbia, Missouri 65211, U.S.A. Author for correspondence: Tatiana Arias, tatiana.tatianaarias@gmail.com Abstract Brassicas boast an enormous diversity of economically important products. Strikingly, relationships among main lineages are still unknown. The phylogeny of the tribe Brassiceae (Brassicaceae) was reconstructed for 89 species using four plastid regions (rpl32-trnL, atpI-atpH, psbD-trnT, ycf6-psbM ). Representatives of almost all genera were sampled, covering the entire natural range of the tribe from Central Asia to the western Mediterranean, including four previously unsampled genera (Ammosperma, Eremophyton, Fezia, Pseudofortuynia). Phylogenetic analysis recovered eight well-supported clades in the tribe including a new African clade (Henophyton) comprised of genera that have not been previously sampled. Relationships within and between the eight major clades are strongly supported for the first time. The earliest-divergent lineages in the tribe are the clades Vella and Zilla. Henophyton is sister to the clade that includes the lineages Nigra, Crambe, Cakile, Savignya, and Oleracea. Core Brassiceae—a new clade defined here—is comprised of two subclades: (1) Nigra, Crambe, and Cakile and (2) Savignya and Oleracea. Within the Oleracea lineage, a new Core Oleracea clade is defined. Several genera were confirmed to be polyphyletic or paraphyletic, including Brassica, Erucastrum, Sinapis, Diplotaxis and Cakile. Economically important species belonging to the genera Brassica were primarily distributed across the two most species-rich lineages, Nigra or Oleracea. Collectively, the finding of these novel clades has numerous taxonomic implications. Keywords Brassica; Brassiceae; Brassicaceae; Brassica oleracea; Cruciferae; Henophyton; mustards; phylogenetics; U triangle Supplementary Material Alignments are available at TreeBASE (http://treebase.org, 13066). INTRODUCTION Understanding the evolution of the Brassica L. crops and wild relatives (Brassicaceae: Brassiceae), which boasts an enormous diversity of economically important products in a single tribe (Tatout & al., 1999; Lysak & al., 2005; Schranz & al., 2006), is perhaps one of the most enigmatic and fascinating questions in plant evolutionary biology. Brassicas are distributed throughout the Mediterranean Basin; a region where geological activity, climatic change and human civilization have had major impacts on plant biodiversity (Thompson, 2005). Reconstructing the phylogenetic relationships among members of the tribe is essential for understanding the taxonomy of this ecologically and economically important group of angiosperms. During the past few years, three major lineages (I, II, III) have been recognized in the Brassicaceae using chloroplast and nuclear markers (Beilstein & al., 2006, 2008). The number of tribes recognized in the mustard family has exploded from 25 (Beilstein & al., 2008) to 49 (Al-Shehbaz, 2012). The tribe Brassiceae is firmly placed within lineage II together with the tribes Thelypodieae, Sisymbrieae, and Isatideae (Beilstein & al., 2008; Warwick & al., 2009b, 2011). The tribe has long been identified as monophyletic among the tribes of Brassicaceae based on morphological (Hedge, 1976; GómezCampo, 1980; Al-Shehbaz, 1985; Gómez-Campo & Prakash, 980 1999; Koch & al., 2001; Appel & Al-Shehbaz, 2003; Koch, 2003) and molecular evidence (Warwick & Black, 1991, 1993, 1994, 1997; Anderson & Warwick, 1999; Bailey & al., 2006; Beilstein & al., 2006). Morphologically, members of the tribe Brassiceae (47 genera and ca. 227 spp.; Al-Shehbaz, 2012) are distinguished from other members of the family by cotyledons that are longitudinally folded around the radicle (conduplicate cotyledons), transversely segmented fruits that contain seeds in one or both segments (heteroarthrocarpic; Appel, 1999) and, if present, simple hairs (Gómez-Campo, 1980; Al-Shehbaz, 1985; Gómez-Campo & Prakash, 1999). Massive evolutionary radiation in the Brassiceae has taken place in the southwestern Mediterranean region (Gómez-Campo, 1980). More than twenty genera reach maximum taxonomic diversity here and levels of endemism are particularly high in Algeria, Morocco and Spain (Hedge, 1976; Gómez-Campo, 1980; Al-Shehbaz, 1985, Gómez-Campo & Prakash, 1999). Few members of the tribe extend eastward to India and Pakistan (Crambe L., Orychophragmus Bunge, Physorhynchus Hook., Schouwia DC.) and native species representation in the New World is known only for five species of Cakile DC. (Warwick & al., 2009a). Schulz (1919, 1923, 1936) recognized seven Brassiceae subtribes using morphology: Brassicinae, Cakilinae, Moricandiinae, Raphaninae, Savignyinae, Vellinae, and Zillinae. Later, Gómez-Campo (1980) recognized six subtribes by merging Version of Record (identical to print version). TAXON 61 (5) • October 2012: 980–988 Arias & Pires • Phylogeny of brassica crops and wild relatives Savignyinae and Vellinae on the basis of their seed wing. Although Schulz (1919, 1923, 1936) and Gómez-Campo (1980) both grouped all Brassica species into subtribe Brassicinae, later molecular studies showed the genus to be polyphyletic (Yanagino & al., 1987; Song & al., 1990). Warwick & Black (1991, 1993, 1994, 1997) and Warwick & Sauder (2005) have suggested several sub-tribal classifications based on ITS, restriction site polymorphisms and the chloroplast region trnL. In these analyses seven major groups are moderately supported: the Oleracea, Nigra, Cakile, Crambe, Savignya, Zilla, and Vella lineages (Warwick & Hall, 2009). However relationships among the main lineages and genera in the tribe remain poorly understood. In addition to the unresolved phylogenetic resolution among the subtribes, several genera within the Brassiceae are also poorly circumscribed. For example, the genera Erucastrum C. Presl., Sinapis L. and Diplotaxis DC. have also been shown to be polyphyletic and morphology has proven to be a poor estimator of the evolutionary relatedness among species in the tribe (Warwick & Hall, 2009; Hall & al., 2011). Evolutionary placements of the genera Ammosperma Hook. f., Douepia Cambess., Eremophyton Bég., Fezia Pit., Orychophragmus, Pseudofortuynia Hedge and Quezeliantha H. Scholz, are still controversial or unknown (Warwick & Sauder, 2005; Bailey & al., 2006; Warwick & Hall, 2009). Progress toward resolving the tribal phylogeny and establishing the monophyly of genera has been slow due to (1) the selection of few and poorly informative molecular markers for phylogenetic studies; and because (2) classification schemes proposed solely on morphological characters have not been supported by modern molecular systematic data. A strong phylogenetic framework is required in order to restructure the classification of members of the tribe, in particular the most species-rich and polyphyletic genera. Many questions about the genomic relationships among the six cultivated species of Brassica, known as the triangle of U, also remain unresolved (U Nagaharu, 1935). For example, the evolutionary relationships among the “diploid” crops, B. rapa L.(e.g., Pak-choi, Chinese cabbage, turnip), B. nigra (L.) W.D.J. Koch (e.g., black mustard), and B. oleracea L. (e.g., cabbage, broccoli, Brussels sprouts, cauliflower, kale), and their allotetraploid hybrid crops B. juncea L. (Coss.) (Indian mustard), B. napus L. (e.g., canola, rapeseed, rutabega), and B. carinata A. Braun (e.g., Ethiopian mustard) are still obscure. The domesticated species B. rapa and B. oleracea, members of the Oleracea clade, seem to be closely related (Warwick & Hall, 2009) and show close parallels in morphological traits selected during domestication (Mizushima & Tsunoda, 1967). In this study, the principle aims were to (1) resolve the relationships within and among clades comprising the tribe Brassiceae and (2) place previously unsampled genera and species. To achieve these goals, the taxonomic sampling within Brassiceae was increased to represent the current classification at generic and subgeneric levels. In addition, data from four rapidly evolving non-coding chloroplast regions (rpl32-trnL, atpI-atpH, psbD-trnT, ycf6-psbM) were used. Parsimony, likelihood and Bayesian analyses were performed to obtain the phylogeny. The taxonomic implications for the discovery of novel clades are discussed. MATERIALS AND METHODS Taxon sampling. — The initial taxon sampling included 142 samples representing 89 species of the tribe and 24 outgroups. The final taxon sampling set (deposited in GenBank) included 110 samples representing 89 species and 39 genera of the tribal ingroup, and 15 species belonging to different tribes in the Brassicaceae used as outgroups (Appendix). Leaf material for DNA extraction was collected from (1) plants grown from seed, (2) herbarium specimens, and (3) plants collected in the wild. Seeds came from the United States Department of Agriculture (USDA) and the specialized mustard seed bank at La Universidad Politécnica de Madrid, Spain. Plants were grown in the greenhouses at the Bond Life Sciences Center (University of Missouri, Columbia). Herbarium vouchers were collected when plants flowered and germination time and flowering time was recorded for all specimens. Fieldwork was conducted in southern Spain (Granada, Almeria, Murcia, Alicante) in the spring of 2010 covering 3820 km across the center of diversity for the tribe; 114 specimens were collected (including herbarium material and seeds) representing 32 species. Herbarium vouchers were deposited at MA, MO, and UMO. Specimens of genera and species that have not been previously sampled in the tribe were collected for DNA extraction in MA. These samples were mostly from rare and endemic North African species. DNA extraction and selection of molecular markers. — Total DNA was isolated from fresh leaf material or silica-dried leaves using DNeasy Plant Mini Kit (Qiagen, Valencia, California, U.S.A.) and standard CTAB protocols (Doyle & Doyle, 1987). A pilot study of chloroplast regions was conducted using species representing the different subtribes and lineages previously identified in the tribe and several outgroups. Nine noncoding regions of the chloroplast genome (Shaw & al., 2005, 2007) were sequenced and from these, four phylogenetically informative and variable chloroplast markers were selected (rpl32-trnL, atpI-atpH, psbD-trnT, ycf6-psbM). In addition, three low-copy nuclear markers were sequenced for variability based on their utility in previous studies in the Brassicaceae (Duarte & al., 2010): SCG4609: carboxymethylenebutenolidase family, SCG4474: aminotransferase, class V, SCG5585: putative c-myc-binding protein. However, none of these regions showed high variation within the tribe Brassiceae. DNA amplification, sequencing and alignment. — Amplification and sequencing primers for all regions were based on previous studies (Shaw & al., 2005, 2007). PCR reactions for all chloroplast regions mixes were: 10 mL ddH2O, 100 ng template DNA, 20 mM Econo-Taq PLUS GREEN 2X Master mix (Lucigen, Radno, Pennsylvania, U.S.A., Cat. No. 30033-1), 10 mM forward primer, and 10 mM reverse primer. Regions were amplified in 25 mL with an Eppendorf Master Cycler epigradient S thermal cycler using an initial 5 min denaturation at 80°C; followed by 30 cycles of 95°C denaturation for 1 min, 1 min annealing at 50°C, and 4 min extension at 65°C; followed by 5 min of final extension at 65°C. PCR products were cleaned using PCR Purification Kit (Invitrogen K3100-01, Carlsbad, California, U.S.A.). Gel electrophoresis of PCR products was used to determine product size Version of Record (identical to print version). 981 Arias & Pires • Phylogeny of brassica crops and wild relatives TAXON 61 (5) • October 2012: 980–988 and concentration. Cycle sequencing reactions used the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Foster City, California, U.S.A.) and the thermocycler parameters 94°C for 5 min, 50 cycles of 94°C for 1 min, and final elongation at 60°C for 10 min. Samples were electrophoresed on a Beckman Coulter CEQ 8000 sequencer Applied Biosystems 3730xl automated DNA sequencing instrument, using 96 cm capillary arrays and POP-7 polymer. Data were analyzed using PE-Biosystems version 3.7 of the program Sequencing Analysis at the DNA Core facility of the University of Missouri, Columbia. DNA sequences were edited using SeqMan (DNAStar Inc., Madison, Wisconsin, U.S.A.) and aligned using MUSCLE v.3.8.31 (Edgar, 2004). All sequences were submitted to NCBI-GenBank (www.ncbi.nlm. nih.gov/genbank; JQ911043–JQ911509, JQ941446–JQ941493, JQ941495–JQ941598; Appendix) and gene alignments were submitted to TreeBASE (http://treebase.org, 13066). Phylogenetic inferences were optimized by the experimental removal of data (i.e., characters, ingroup taxa with incomplete data and/or long branches, outgroup taxa numbers; see Table 1) and manual alignment in some of the hypervariable regions. Phylogenetic analyses. — The four intergenic regions of the chloroplast were concatenated as one dataset given the uniparental inheritance and lack of recombination in the chloroplast genome. Phylogenetic relationships were inferred from the nucleotide data using maximum parsimony (MP), maximum likelihood (ML), and Bayesian inference (BI). Phylogenies were generated on the Cyberinfrastructure for Phylogenetic Research (CIPRES) portal 2 teragrid (http://www.phylo.org) (Miller & al., 2010). Phylogenetic analyses were rooted using 12 to 24 outgroups chosen based on recent phylogenies published for the family (Beilstein & al., 2008) (Appendix). For parsimony analysis, individual bases were considered multistate, unordered characters of equal weight; unknown nucleotides were treated as uncertainties. MP analyses were implemented in PAUPRat (Sikes & Lewis, 2001) based on Parsimony Ratchet (Nixon, 1999). Two hundred multiple independent searches were performed and each search involved 500 iterations. Tree-bisection-reconnection (TBR) branch swapping was used. Consistency indices (CI) and retention indices (RI) were calculated to evaluate the amount of homoplasy in the data using PAUP* v.4.0b10 (Swofford, 2003). Maximum likelihood and ML bootstrapping (MLB), with different models allowed for each gene partition, was performed using the program RAxML v.7.0.4 (Stamatakis, 2006; Stamatakis & al., 2008). The program automatically determines the number of bootstrap runs necessary to reach completion. For Bayesian methods (BI) the optimal model of sequence evolution was found using jModeltest and only the Akaike information criterion (AIC) was reported here (Posada & Crandall, 1998). Bayesian analyses were conducted using MrBayes v.3.1 (Ronquist & Huelsenbeck, 2003) allowing different models for each region and using default priors (Ronquist & Huelsenbeck, 2003; Alfaro & Holder, 2006). Two independent runs of 10,000,000 generations were completed with four chains each (three heated, one cold), using a chain temperature of 0.2 and uniform priors. Trees were sampled every 1000 generations. Likelihood-by-generation plots were created, and the first 25% of runs were discarded as burn-in. A majority-rule consensus of the remaining trees from the two runs was produced and used as the Bayesian inference tree with posterior probabilities (PP). Chain convergence was checked using the program AWTY (Nylander & al., 2008). Region/ Alignment Total aligned bp No. variable % variable No. invariable sites % invariable sites No. PIC % PIC Consistency index Retention index No. species ingroup No. species outgroup No. of accessions No. of MPTs Tree length Table 1. Summary of sequence information for four cpDNA loci (including three alternative alignments/taxon sets) and maximum parsimony tree information. The data matrix chosen for final analyses is indicated in bold. The number of accessions differs among loci as some accessions were sequenced multiple times; however the final concatenated matrix had 101 accessions per locus. MPTs: most parsimonious trees; PIC: phylogenetically informative characters. atpI-atpH 611 77 12.60% 388 63.50% 146 23.89% 0.69 0.88 86 24 154 362 445 psbD-trnT 1051 132 12.56% 740 70.41% 179 17.03% 0.74 0.92 89 22 146 523 502 rpl32-trnL 1471 343 23.32% 726 49.35% 402 27.33% 0.73 0.87 86 22 155 98 528 ycf6-psbM 566 138 24.38% 283 50.00% 145 26.62% 0.73 0.87 89 23 166 513 528 Four-cp concatenated and all accessions 3319 1197 36.06% 1170 35.25% 952 28.68% 0.60 0.69 89 24 142 83 5276 Four-cp concatenated, removing taxa, manual alignment 3110 621 19.96% 1788 57.49% 701 22.54% 0.59 0.77 89 24 125 332 2982 Four-cp concatenated, removing taxa, manual alignment* 3110 509 16.36% 2001 64.34% 600 19.29% 0.63 0.79 76 12 101 169 2285 982 Version of Record (identical to print version). TAXON 61 (5) • October 2012: 980–988 Arias & Pires • Phylogeny of brassica crops and wild relatives Brassica elongata Brassica gravinae Brassica desnottesii */* Brassica repanda Moricandi foetida */* Moricandia arvensis */* Diplotaxis harra */* Eruca vesicaria subsp. sativa */* Eruca pinnatifida Diplotaxis erucoides */* Diplotaxis erucoides Erucastrum gallicum 83/0.99Brassica rupestris 88/* 63/0.86 */* Brassica macrocarpa Brassica villosa Brassica napus */* Carrichtera annua 99/* Brassica juncea */* Brassica rapa subsp. chinensis Brassica rapa 84/0.99 Brassica cretica 96/1.0 B. oleracea var. gemmifera B. oleracea var. gongylodes B. oleracea var. alboglabra Brassica oleracea 92/0.87 99/* B. oleracea var. viridis B. oleracea var. botrytis B. oleracea var. italica Brassica incana 80/0.99 Brassica montana Enarthorcarpus lyratus */* Morisa monanthos Raphanus raphanistrum Erucastrum nasturtiifolium Brasica barrelieri Brasica deflexa 94/0.99 Fezia pterocarpa Savignya parviflora Coincya longirostra 96/0.94 */* Brassica nigra Brassica carinata Guiraoa arvensis Sinapis alba 87/0.95 */* Kremeriella cordylocarpus 96/0.94 Sinapis flexuosa Erucastrum littoreum */* Erucastrum rifanum */* Brassica spinecens Brassica maurorum Brassica fruticulosa */0.95 Cordylocarpus muricatus Cordylocarpus muricatus Ceratocnemum rapistroides Erucastrum ifniense */* Sinapidendron angustifolium Erucastrum canariense Rapistrum rugosum 82/0.82 95/0.99 Raffenaldia primuloides Brassica tournefortii */* */* 96/0.96 Sinapis pubescens Brassica procumbes Hemicrambe fruticulosa */* 99/* Erucastrum virgatum */* Hirschfeldia incana Erucastrum elatum Erucastrum elatum Erucaria erucariodes */* Eremophyton chevallieri */* Cakile maritima */* Didesmus bipinnatus 80/0.98 65/0.99 Cakile lanceolata 98/* Crambe hispanica */* Crambe juncea Crambe hispanica subsp. abyssinica */0.83 Crambe filiformis */* Pseuderucaria teretifolia Henophyton zygarrhenum 98/0.95 Trachystoma labasii */* Trachystoma labasii Ammosperma cinerea Ammosperma cinerea 83/0.98 Henophyton deserti Vella spinosa */* Vella bourgeana 73/* Vella anremerica 88/0.95 Schouwia purpurea 92/0.91 */* Foleyola billotii */* Zilla macroptera 0.02 Physorhynchus chamaerapistrum */* */* Oleracea Core Oleracea Core Brassiceae Fig. 1. Maximum likelihood (ML) phylogeny of tribe Brassiceae from the analysis of four concatenated plastid intergenic regions (rpl32-trnL, atpI-atpH, psbD-trnT, ycf6-psbM). The main eight lineages in the tribe Brassiceae are indicated on the right, including the novel lineage Henophyton. Arrows indicate newly defined clades: Core Oleracea and Core Brassiceae. The twelve outgroups are not shown. Numbers at branches are bootstrap support values for the ML analysis and Bayesian posterior probabilities shown for all clades where support was ≥ 50 or BPP ≥ 0.5. ML estimates of 100 or BPP estimates of 1.0 are indicated with an asterisk (*). If no numbers are indicated above branches, then the node is not supported (ML < 50 and BPP < 0.5). Savignya Nigra Cakile Crambe Henophyton Vella Zilla Version of Record (identical to print version). 983 Arias & Pires • Phylogeny of brassica crops and wild relatives RESULTS Initial preliminary analyses included 142 samples/ sequences representing 89 species of the tribe and 24 outgroups (see Table 1). Sequences were deposited in GenBank for 110 samples representing 89 of the 227 species (39.2%) and 39 of 47 genera (83%) of the tribe. In addition, 15 species belonging to different tribes in the Brassicaceae were used as outgroups (Appendix). In the final phylogenetic analyses, some accessions were excluded because of incomplete datasets for some of the regions sequenced and long-branch attraction problems; and a few characters were removed where homology could not be unambiguously assigned (taxon and character sampling summarized in Table 1). After experimenting with data removal (i.e., taxa and character removal and manual alignment in hypervariable regions) and analyzing several data matrices to optimize phylogenetic inference, the final combined matrix of cpDNA markers included 101 samples representing 76 species of the ingroup and 12 outgroups. The final aligned data matrix was 3110 bp long; of which 600 bp were parsimony-informative, and 509 bp were variable but parsimony-uninformative. The number of parsimony-informative characters varied among loci: rpl32-trnL had the highest number of informative characters in the alignment (402 bp of 1471 bp in the alignment), followed by psbD-trnT (179 bp of 1051 bp), atpI-atpH (146 bp of 611 bp) and ycf6-psbM (145 bp of 566 bp; Table 1). jModeltest generated the following nucleotide evolution models for the cp-DNA regions: rpl32-trnL fits to TPM1uf + I + G, atpI-atpH fits to TVM + G, psbD-trnT fits to GTR + G, ycf6-psbM fits to GTR + I + G. The BI consensus tree from MrBayes run with different models allowed for each partition had a best score of −lnL = 8727.53. Support values from MLB values were lower than BI posterior probability values (Figs. 1 & 2). MP analyses for the cp-DNA phylogeny recovered 169 trees of length 2285, with a CI of 0.63 and RI of 0.79 (Table 1). MP, ML and BI analyses yielded the same tree topology with the exception of few taxa within the Nigra and Oleracea clades. Bootstrap values and posterior probabilities both supported eight main clades in this phylogeny, and only a few clades were weakly supported within the Nigra and Oleracea clade (Figs. 1 & 2). The monophyly of the tribe is confirmed with the chloroplast markers (98% MLB). Eight well-supported lineages were identified in the tribe (Figs. 1 & 2): Cakile (100% MLB), Crambe (100% MLB), Nigra (96% MLB), Oleracea (100% MLB), Savignya (94% MLB), Vella (100% MLB), Zilla (100% MLB) and a new African clade Henophyton (98% MLB). The relationships among these eight clades are strongly supported. The earliest-divergent lineages in the tribe are the sister clades Vella + Zilla (92% MLB). Henophyton + Nigra + Crambe + Cakile + Savignya + Oleracea form a strongly supported clade (100% MLB). Henophyton is sister to the clade Nigra + Crambe + Cakile + Savignya + Oleracea (98% MLB). Core Brassiceae—a new clade defined here—is comprised of two subclades: (1) Nigra and Crambe + Cakile (82% 984 TAXON 61 (5) • October 2012: 980–988 MLB) and (2) Savignya + Oleracea (96% MLB) (Fig. 1). Within the Oleracea lineage, a new Core Oleracea clade is also defined (indicated in Fig. 1). Previously unsampled genera. — Four genera were sampled for the first time. Ammosperma, Eremophyton, and Fezia were found to fall into the tribe Brassiceae and Pseudofortuynia was found to fall outside the tribe. In agreement with previous studies, Calepina Adans., and Coringia C. Presl. were also found to fall outside the tribe (Warwick & al., 2010). Polyphyletic and paraphyletic genera. — Several genera were confirmed not to be monophyletic in the cp-phylogeny including Brassica, Erucastrum, Sinapis, Diplotaxis and Cakile. Other genera may also be polyphyletic with increased taxon sampling and increased phylogenetic resolution. For Brassica, 23 of 38 species were included here (61% of spp. in the genus), 16 of these belong to the Oleracea lineage, and seven to the Nigra lineage (Fig. 1). Eight of 23 Erucastrum species were included here; two are placed in the Oleracea lineage, and six in the Nigra lineage. Two of the six species of Cakile were included here, and the genus is confirmed as paraphyletic since C. lanceolata (Willd.) O.E. Schulz is more closely related to Didesmus bipinnatus DC. (65% MLB) than to C. maritime Scop. (Fig. 1) (Hall & al., 2011). Henophyton Coss & Durieu (2 spp.) is also paraphyletic as Henophyton deserti (Coss & Durieu) Coss & Durieu forms a clade with Ammosperma cinerea Baill. (83% MLB; Fig. 1) while Henophyton zygarrhenum (Maire) Gómez-Campo forms a polytomy with Trachystoma labasii Maire, and Henophyton deserti + Ammosperma cinerea (100% MLB; Fig. 1). */* Oleracea 96/* 94/0.99 Savignya 98/0.94 96/0.95 Nigra 82/0.82 */* Cakile 100/1.0 80/0.98 */* 98/* 98/0.95 */* 92/0.91 Crambe Henophyton Zilla */* Vella Fig. . Bayesian reconstruction using four chloroplast markers and highlighting the backbone phylogeny of the main eight lineages in the tribe Brassiceae. Numbers at branches are bootstrap support values for the ML analysis and Bayesian posterior probabilities shown for all clades where support was ≥ 50 or BPP ≥ 0.5. ML estimates of 100 or BPP estimates of 1.0 are indicated with an asterisk (*). Version of Record (identical to print version). TAXON 61 (5) • October 2012: 980–988 Arias & Pires • Phylogeny of brassica crops and wild relatives DISCUSSION Phylogenetic relationships. — The analyses presented here contain more sequence data per taxon and wider sampling of genera and species in the tribe Brassiceae than any other study published to date (Appendix). Only two genera it the tribe, Douepia and Quezeliantha, have not been sampled in molecular phylogenetic studies. This study supports the seven previously discovered lineages that were recovered with chloroplast markers (Warwick & Black, 1991, 1993, 1994, 1997) and adds a new lineage designated here as Henophyton—corresponding to the oldest genus name. This lineage contains previously unsampled and rare taxa (Ammosperma cinerea, Henophyton zygarrhenum) native to North Africa. This is also the first study to resolve the relationships among the eight main lineages in the tribe with high bootstrap support values ≥ 80% (Fig. 2). In addition to constructing a robust phylogeny among the main eight lineages of the Brassiceae, these results clarify many of the relationships among genera and species. Some of these polyphyletic genera are economically important and include domesticated species that occur primarily in two species-rich lineages (Oleracea, Nigra). Most of the Erucastrum species included in this study were placed within the Nigra lineage (5 of 7 spp.), while the two Diplotaxis species were placed within the Oleracea lineage, and all the Sinapis species belonged to the Nigra lineage. Before making taxonomic changes, the plastid phylogeny presented here should be tested with nuclear genes (Wendel & Doyle, 1998), especially given the high frequency of ancient and recent hybridization (Hauser & al., 1997, 1998) and polyploidization events (Lysak & Lexer, 2006; Schranz & al., 2006; Warwick & Hall, 2009) in the Brassicaceae; and because recent studies within the tribe show that nuclear and chloroplast markers recover markedly different topologies (Hall & al., 2011). MP, ML and BI analyses resulted in identical topologies for the backbone of the phylogeny and there were only minor disagreements in species relationships in the Oleracea and Nigra lineages. Three genera of the Brassicaceae were placed outside the tribe: Calepina, Coringia, and Pseudofortuynia; a result that agrees with Warwick & al. (2010) for the first two genera. Three genera previously included in the tribe based on morphological data were newly sampled here and were supported as belonging in the tribe: Ammosperma, Eremophyton, and Fezia. The new clade Henophyton contains the genera Henophyton (2 spp. included), Ammosperma (1 of 2 spp. included), and Pseuderucaria O.E. Schulz (1 of 2 spp. included). Previous studies have not placed these taxa in the main seven lineages (Warwick & Hall, 2009). Cole crops and wild species (n = 9) constitute the section Brassica of the polyphyletic genus Brassica. In the tribal phylogeny these species constitute the core of the Oleracea clade (Fig. 1). Crosses between B. oleracea and wild relatives are known to produce fertile or semi-fertile offspring, making the taxonomy and phylogenetic reconstructions in this clade challenging. Contrasting hypotheses about the origin of B. oleracea remain a heated topic of debate (Neutrofal, 1927; Song & al., 1990, Allender & al., 2007). Our phylogeny partially reconstructs the evolutionary relationships between wild relatives and B. oleracea varieties. Taxonomic implications. — These results suggest that there are eight lineages within the tribe Brassiceae (Fig. 1), assuming that future studies using multiple nuclear loci confirm the species tree inferred from the present plastid phylogeny. Four of the seven named subtribes, Cakilinae, Savignynae, Vellinae and Zillinae (Schulz, 1919, 1923, 1936; Gómez-Campo, 1980) can be retained as they correspond to the informally named monophyletic lineages Cakile, Savignya, Vella, and Zilla, respectively (Fig. 2). However, the traditional subtribes of Brassicinae, Moricandiinae, and Raphaninae should be phylogenetically and taxonomically recircumscribed as they do not correspond to the monophyletic lineages Oleracea, Nigra, and Crambe. Brassicinae includes genera like Brassica that are polyphyletic across the Oleracea and Nigra lineages. Moricandiinae is monophyletic but embedded within the Oleracea lineage. Raphaninae is problematic because it includes 18 genera that are in the Nigra, Cakile and Crambe lineages (Dixon, 2007). Finally, a new subtribe would need to be erected for the Henophyton lineage recognized in this study. In addition to defining lineages within the tribe, new generic-level classifications will be needed for polyphyletic genera such Brassica, Diplotaxis, Erucastrum and Sinapis (Warwick & Black, 1991, 1993, 1994, 1997; Warwick & Sauder 2005) and paraphyletic genera such as Cakile (Hall & al., 2011) and Henophyton. Economically important Brassica is a polyphyletic genus, with species found across three named subtribes (Brassicinae, Moricandiinae, Raphaninae) and two informal lineages (Nigra, Oleracea; Gómez-Campo & Prakash, 1999). The high taxon sampling and inclusion of specimens that represent the majority of type species for genera in the tribe provides a strong phylogenetic framework in which taxonomy can be re-evaluated. The use of type species in molecular studies can potentially be a reference to restructure the existing classification of the tribe. For example, Brassica may be restricted to just the species within the clade containing B. oleracea (the type species for the genus) and wild relatives (Core Oleracea; Fig. 1). However, Carrichtera annua L. (DC.) should not be transferred to Brassica since nuclear phylogenies and morphology suggested this species is more closely related to Vella L. than to B. oleracea (Gómez-Campo, 1980; Warwick & Sauder, 2005). No firm taxonomic decisions regarding the circumscription of genera in the tribe should be made until further studies (including multiple nuclear loci, chromosome number, morphology, and other data types) confirm these results. Concluding remarks. — A robust chloroplast phylogeny was constructed for members of the tribe Brassiceae and a new lineage was discovered. This lineage—informally recognized here as Henophyton—is composed of rare North African taxa. Relationships within and between the eight major clades were strongly supported and clade names are suggested for the Core Brassiceae and Core Oleracea. This phylogeny placed species belonging to the polyphyletic genera Brassica and Erucastrum in either of the two most species-rich lineages, Nigra or Oleracea, presenting a robust phylogenetic framework for future studies using multiple nuclear loci with which genuslevel taxonomy can be reevaluated. Version of Record (identical to print version). 985 Arias & Pires • Phylogeny of brassica crops and wild relatives ACKNOWLEDGEMENTS The authors thank P. McSteen, G. Conant, J. Birchler, M. Liscum, R. Steel, E. Wheeler, M. Kinney, C. Henriquez, M. Beilstein, J.C. Hall and A. Cardenas for valuable comments on the manuscript; P. Edger, M. Tang, C. Findley for helping with collections, lab work and data analyses; K. Hertweck for help with data analyses; I. Al-Shehbaz for identifying material included in this study, for useful conversations about the taxonomy of the group and for the review of the manuscript. M. Velayos, A.M. Benavides and J.A. Valencia for help in the field and in the MA herbarium. The authors acknowledge the following research grants and granting institutions: National Science Foundation (DEB 1146603, DGW 46), John Bies graduate student travel award from the University of Missouri and the International Association for Plant Taxonomy (IAPT). LITERATURE CITED Alfaro, M.E. & Holder, M.T. 2006. The posterior and the prior in Bayesian phylogenetics. Annual Rev. Entomol. 55: 189–206. Allender, C.J., Allainguillaume, J., Lynn, J. & King, G.J. 2007. Simple sequence repeats reveal uneven distribution of genetic diversity in chloroplast genomes of Brassica oleracea L. and (n = 9) wild relatives. Theor. Appl. Genet. 114: 609–618. Al-Shehbaz, I.A. 1985. The genera of Brassiceae (Cruciferae: Brassicaceae) in the southeastern United States. J. Arnold Arbor. 66: 279–351. Al-Shehbaz I.A. 2012. A generic and tribal synopsis of the Brassicaceae (Cruciferae). Taxon 61: 931–954. [This issue.] Anderson, J.K. & Warwick, S.I. 1999. Chromosome number evolution in the tribe Brassiceae (Brassicaceae): Evidence from isozyme number. Pl. Syst. Evol. 215: 255–285. Appel, O. 1999. The so-called ‘beak’, a character in the systematics of Brassicaceae? Bot. Jahrb. Syst. 121: 85–98. Appel, O. & Al-Shehbaz, I.A. 2003. Cruciferae. Pp. 75–174 in: Kubitzki, K. & Bayer, C. (eds.), The families and genera of vascular plants, vol. 5. Berlin: Springer. Bailey, C.D., Koch, M.A., Mayer, M., Mummenhoff, K., O’Kane, S.L., Jr., Warwick, S.I., Windham, M.D. & Al-Shehbaz, I.A. 2006. Toward a global phylogeny of the Brassicaceae. Molec. Biol. Evol. 23: 2142–2160. Beilstein, M.A., Al-Shehbaz, I.A. & Kellogg, E.A. 2006. Brassicaceae phylogeny and trichome evolution. Amer. J. Bot. 93: 607–619. Belstein, M.A., Al-Shehbaz, I.A., Mathews, S. & Kellogg, E.A. 2008. Brassicaceae phylogeny inferred from phytochrome A and ndhF sequence data: Tribe and trichomes revisited. Amer. J. Bot. 95: 1307–1327. Dixon, G.R. 2007. Vegetable brassicas and related crucifers. Cambridge, Massachusetts: CABI North America. Doyle, J.J. & Doyle, J.L. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19: 11–15. Duarte, J.M., Wall, P.K., Edger, P.P., Landherr, L.L., Ma, H., Pires, J.C., Leebens-Mack, J. & dePamphilis, C.W. 2010. Identification of shared single copy nuclear genes in Arabidopsis, Populus, Vitis and Oryza and their phylogenetic utility across various taxonomic levels. B. M. C. Evol. Biol. 10: 61, doi: 10.1186/1471-2148-10-61. Edgar, R.C. 2004. MUSCLE: Multiple sequence alignment with high accuracy and high throughput. Nucl. Acids. Res. 32: 591–597. Gómez-Campo, C. 1980. Morphology and morpho-taxonomy of the tribe Brassiceae. Pp. 3–31 in: Tsunoda, S., Hinata, K. & GomezCampo, C. (eds.), Brassica crops and the wild allies: Biology and breeding. Tokyo: Japan Scientific Societies Press. Gómez-Campo, C. & Prakash, S. 1999. Origin and domestication. 986 TAXON 61 (5) • October 2012: 980–988 Pp. 33–58 in: Gómez-Campo, C. (ed.), Developments in plant genetics and breeding, vol. 4, Biology of Brassica coenospecies. Amsterdam: Elsevier. Hall, J.C., Tisdale, T.E., Donohue, K., Wheeler, A., Al-Yahya, M.A. & Kramer, E.M. 2011. Convergent evolution of a complex fruit structure in the tribe Brassiceae (Brassicaceae). Amer. J. Bot. 98: 1989–2003. Hauser, T.P., Jørgensen, R.B. & Østergard, H.R. 1997. Preferential exclusion of hybrids in mixed pollinations between oilseed rape (Brassica napus) and weedy B. campestris (Brassicaceae). Amer. J. Bot. 84: 756–762. Hauser, T.P., Jørgensen, R.B. & Østergard, H.R. 1998. Fitness of backcross and F2 hybrids between weedy Brassica rapa and oilseed rape (Brassica napus). Heredity 81: 436–443. Hedge, I.C. 1976. A systematic and geographical survey of the Old World Cruciferae. Pp. 1–45 in: Vaughn, J.G., MacLeod, A.J. & Jones, B.M.G. (eds.), The biology and chemistry of the Cruciferae. London: Academic Press. Koch, M.A. 2003. Molecular phylogenetics, evolution and population biology in the Brassicaceae. Pp. 1–35 in: Sharma, A.S. & Enfield, N.H. (eds.), Plant genome: Biodiversity and evolution. New Hampshire: Science Publishers. Koch, M.[A.], Haubold, B. & Mitchell-Olds, T. 2001. Molecular systematics of the Brassicaceae: Evidence from coding plastidic matK and nuclear Chs sequences. Amer. J. Bot. 88: 534–544. Lysak, M.A. & Lexer, C. 2006. Towards the era of comparative evolutionary genomics in Brassicaceae. Pl. Syst. Evol. 259: 175–198. Lysak, M.A., Koch, M.A., Pecinka, A. & Schubert, I. 2005. Chromosome triplication found across the tribe Brassiceae. Genome Res. 15: 516–525. Miller, M.A., Pfeiffer, W. & Schwartz, T. 2010. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. Pp. 45–52 in: Gateway Computing Environments Workshop (GCE 2010), New Oreleans, Lousiana, U.S.A., 14 Nov. 2010. New Orleans: IEEE. Mizushima, U. & Tsunoda, S. 1967. An exploration in Brassica and allied genera. Tohoku J. Agric. Res. 17: 249–277. Neutrofal, F. 1927. Zytologische Studien über die Kulturrassen von Brassica oleracea. Oesterr. Bot. Z. 76: 105–115. Nixon, K.C. 1999. The parsimony ratchet, a new method for rapid parsimony analysis. Cladistics 15: 407–414. Nylander, J.C., Wilgenbusch, D.L., Warren & D.L. & Swofford, D.L. 2008. AWTY (Are we there yet?): A system for graphical exploration of MCMC convergence in Bayesian phylogenetics. Bioinformatics 24: 581–583. Posada, D. & Crandall, K.A. 1998. MODELTEST: Testing the model of DNA substitution. Bioinformatics 14: 817–818. Ronquist, F. & Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572–1574. Schranz, M.E., Lysak, M.A. & Mitchell-Olds, T. 2006. The ABC’s of comparative genomics in the Brassicaceae: Building blocks of crucifer genomes. Trends Pl. Sci. 11: 535–542. Schulz, O.E. 1919. Cruciferae-Brassiceae. Part 1. Subtribes Brassicinae and Raphaninae. Pp. 1–290 in: Engler, A. (ed.), Das Pflanzenreich, IV, 105 (Heft 68–70). Leipzig: Engelmann. Schulz, O.E. 1923. Cruciferae-Brassiceae. Part II. Subtribes Cakilinae, Zillinae, Vellinae, Savignyinae and Moricandiinae. Pp. 1–100 in: Engler, A. (ed.), Das Pflanzenreich, IV, 105 (Heft 82–85). Leipzig: Engelmann. Schulz, O.E. 1936. Cruciferae. Pp. 227–658 in: Engler, A. & Prantl, K. (eds.), Die naturlichen Pflanzenfamilien, 2nd edn., vol. 71. Leipzig: Engelmann. Shaw, J., Lickey, E.B., Beck, J.T., Farmer, S.B., Liu, W., Miller, J., Siripun, K.C., Winder, C.T., Schilling, E.E. & Small, R.L. 2005. The tortoise and the hare II: Relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. Amer. J. Bot. 92: 142–166. Version of Record (identical to print version). TAXON 61 (5) • October 2012: 980–988 Arias & Pires • Phylogeny of brassica crops and wild relatives Shaw, J., Lickey, E.B., Schilling, E.E. & Small, R.L. 2007. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: The tortoise and the hare III. Amer. J. Bot. 94: 275–288. Sikes, D.S. & Lewis P.O. 2001. Software manual for PAUPRat: A tool to implement Parsimony Ratchet searches using PAUP*. http:// www.ucalgary.ca/~dsikes/software2.htm. Song, K., Osborn, T.C. & Williams, P.H. 1990. Brassica taxonomy based on nuclear restriction fragment length polymorphisms (RFLPs) – 3. Genome relationships in Brassica and related genera and the origin of B. oleracea and B. rapa (syn. B. campestris). Theor. Appl. Genet. 79: 497–506. Stamatakis, A. 2006. RAxML-VI-HPC: Maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22: 2688–2690. Stamatakis, A., Hoover, P. & Rougemont, J. 2008. A rapid bootstrap algorithm for the RAxML web servers. Syst. Biol. 57: 758–771. Swofford, D.L. 2003. PAUP*: Phylogenetic analysis using parsimony (*and other methods), version 4.0. Sunderland, Massachusetts: Sinauer. Tatout, C., Warwick, S., Lenoir, A. & Deragon J.-M. 1999. SINE insertions as clade markers for wild crucifer species. Molec. Biol. Evol. 16: 1614–1621. Thompson, J.D. 2005. Plant evolution in the Mediterranean. New York: Oxford University Press. U Nagaharu 1935. Genome analysis in Brassica with special reference to the experimental formation of B. napus and peculiar mode of fertilization. Jap. J. Bot. 7: 389–452. Warwick, S.I. & Black L.D. 1991. Molecular systematics of Brassica and allied genera (Subtribe Brassicinae, Brassiceae): Chloroplast genome and cytodeme congruence. Theor. Appl. Genet. 82: 81–92. Warwick, S.I. & Black L.D. 1993. Molecular relationships in subtribe Brassicinae (Cruciferae, tribe Brassiceae). Canad. J. Bot. 71: 906–918. Warwick, S.I. & Black, L.D. 1994. Evaluation of the subtribes Moricandiinae, Savignyinae, Vellinae, and Zillinae (Brassicaceae, tribe Brassiceae) using chloroplast DNA restriction site variation. Canad. J. Bot. 72: 1692–1701. Warwick, S.I. & Black L.D. 1997. Phylogenetic implications of chloroplast DNA restriction site variation in subtribes Raphaninae and Cakilinae (Brassicaceae, tribe Brassiceae). Canad. J. Bot. 75: 960–973. Warwick, S.I. & Hall, J.C. 2009. Phylogeny of Brassica and wild relatives (tribe Brassiceae: Brassicaceae). Pp. 19–36 in: Gupta, S.K. (ed.), Biology and breeding of crucifers. Boca Raton: CRC Press, Taylor and Francis group. Warwick, S.I. & Sauder, C.A. 2005. Phylogeny of tribe Brassiceae (Brassicaceae) based on chloroplast restriction site polymorphisms and nuclear ribosomal internal transcribed spacer and chloroplast trnL intron sequences. Canad. J. Bot. 83: 467–483. Warwick, S.I., Francis, A. & Gugel, R.K. 2009a. Guide to the wild germplasm. Brassica and allied crops (tribe Brassiceae, Brassicaceae). http://www.brassica.info/. Warwick, S.I., Mummenhoff, K., Sauder, C.A., Koch, M.A. & AlShehbaz, I.A. 2010. Closing the gaps: phylogenetic relationships in the Brassicaceae based on DNA sequence data of nuclear ribosomal ITS region. Pl. Syst. Evol. 285: 209–232. Warwick, S.I., Sauder, C.A. & Al-Shehbaz, I.A. 2011. Systematic position of Ivania, Scoliaxon, and Phravenia (Brassicaceae). Taxon 60: 1156–1164. Warwick, S.I., Sauder, C.A., Mayer, M.S. & Al-Shehbaz, I.A. 2009b. Phylogenetic relationships in the tribes Schizopetaleae and Thelypodieae (Brassicaceae) based on nuclear ribosomal ITS region and plastid ndhF DNA sequences. Botany 87: 961–985. Wendel, J.F. & Doyle J.F. 1998. Phylogenetic incongruence: Window into genome history and molecular evolution. In: Soltis, D.E., Soltis, P.S. & Doyle, J.J. (eds.), Molecular systematics of plants, II, DNA sequencing. Boston: Kluwer Academic Publishers. Yanagino, T., Takahata, Y. & Hinata, K. 1987. Chloroplast DNA variations among diploid species in Brassica and allied genera. Jap. J. Genet. 62: 119–125. Appendix. Taxon; country, region; collector (TA = Tatiana Arias) with collection number and herbarium acronym (if not in seed bank: PM USDA PI = United States Department of Agriculture; BGV UPM = mustard seed bank at La Universidad Politécnica de Madrid, Spain); GenBank no. atpI-atpH; psbDtrnT; rpl32-trnL; ycf6-psbM. Ammosperma cinerea (Desf.) Hook. f.; Tunisia; Lambinon 99 MA; JQ941447; JQ911365; JQ911210; JQ911044. Ammosperma cinerea (Desf.) Hook. f., origin unknown; Pitard 656 MA; JQ941448; JQ911366; JQ911211; JQ911045. Brassica barrelieri (L.) Janka; Spain, Sierra de Guadarrama; TA 463 MO, UMO; JQ941449; JQ911367; JQ911212; JQ911046. Brassica carinata A. Braun; Ethiopia; USDA PI 331377; JQ941450, JQ941451; JQ911368; JQ911213; JQ911047, JQ911048, JQ911049. Brassica cretica Lam.; Turkey, Kusadasi, Millipark; BGV UPM 6346; JQ941453, JQ941454; JQ911369; JQ911215, JQ911216; JQ911051, JQ911052. Brassica deflexa Boiss.; Iran, Karadj; BGV UPM 371375; JQ941455; JQ911370; JQ911217; JQ911053. Brassica desnottesii Emb. & Maire; Morocco, Debdou; BGV UPM 440476; JQ941456; JQ911371; JQ911218; JQ911054. Brassica elongata Ehrh.; Iran, Kondor; BGV UPM 371875; JQ941457; JQ911372; JQ911219; JQ911055. Brassica fruticulosa Cirillo; Algeria, Tombeau dela Chretienne; BGV UPM 366875; JQ941458; JQ911373; JQ911220; JQ911056. Brassica fruticulosa subsp. rifana (Emb. & Maire) Maire; origin unknown; USDA PI 597841; JQ941527; JQ911509; JQ911287; JQ911129. Brassica gravinae Ten.; Algeria, El Kantara; BGV UPM 194772; JQ941459; JQ911503; JQ911221; JQ911057. Brassica incana Ten.; Italy; USDA PI 435898; JQ941460, JQ941461; JQ911374, JQ911375; JQ911222, JQ911223, JQ911224; JQ911058, JQ911059. Brassica insularis Moris; Italy, Cerdeña, Cabo Caccia; BGV UPM 381475; NA; NA; JQ911352; JQ911195. Brassica juncea (L.) Czern.; South Korea; USDA 537021; JQ941462; JQ911376; JQ911225; JQ911060. Brassica juncea (L.) Czern.; Turkey, Kayseri; USDA 175602; JQ941463; NA; JQ911226; JQ911061. Brassica macrocarpa Guss.; Italy, Sicilia, Islas Egadi; BGV UPM 381975; JQ941464, JQ941465; JQ911377; JQ911227; JQ911062, JQ911063. Brassica maurorum Durieu; Algeria, Sur de Oron; BGV UPM 196671; JQ941466; JQ911504; JQ911228; JQ911064. Brassica montana Pourr; origin unknown; TA 458 MO, UMO; JQ941467; JQ911378; JQ911229; JQ911065. Brassica napus L.: origin unknown; BGV UPM 399075; JQ941468; JQ911379; JQ911230; JQ911066. Brassica nigra (L.) W.D.J. Koch; India, Delhi; USDA 173861; JQ941469, JQ941470; JQ911380, JQ911381; JQ911231, JQ911232; JQ911067. Brassica nigra (L.) W.D.J. Koch; Ethiopia; USDA 331376; JQ941471; JQ911382; JQ911233; JQ911068. Brassica nigra (L.) W.D.J. Koch; India, Gujarat; USDA 183116; JQ941472; JQ911383; NA; JQ911069. Brassica oleracea L.; origin unknown; TA 432 MO, UMO; JQ941476; JQ911388; JQ911237; JQ911070, JQ911071; Brassica oleracea L.; origin unknown; BGV UPM 2862; JQ941475, JQ941478; JQ911384; JQ911234, JQ911239; JQ911072, JQ911199. Brassica oleracea var. sabauda L.; origin unknown; TA 451 MO, UMO; NA; NA; JQ911348; JQ911191. Brassica oleracea var. viridis L.; origin unknown; TA 457 MO, UMO; NA; NA; JQ911350; JQ911193. Brassica oleracea var. botrytis L.; United States; TA 453 MO, UMO; NA; NA; JQ911349; JQ911192. Brassica oleraceae var. gemmifera DC.; origin unknown; TA 449 MO, UMO; JQ941473, JQ941477; JQ911386; JQ911235, JQ911238; JQ911073, JQ911074. Brassica oleraceae var. gongylodes L.; origin unknown; TA 460 MO, UMO; JQ941474; JQ911387; JQ911236; JQ911075. Brassica oxyrrhina (Coss.) Willk.; Morocco, Norte de Kenitra; BGV UPM 2965; JQ941479; JQ911389; JQ911240; JQ911076. Brassica procumbens (Poir.) O.E. Schulz; Algeria, Jbel Ouasch, Casa Forestal; TA 439 MO, UMO; JQ941480; JQ911390; JQ911241; Q911077. Brassica rapa L.; origin unknown; IMB 218; JQ941481; JQ911391; JQ911242; JQ911078. Brassica rapa subsp. chinensis (L.) Hanelt; origin unknown; BGV UPM 126370; JQ941452; JQ911502; JQ911214; JQ911050. Brassica repanda (Willd.) DC.; Spain, Trevenque, Sierra Nevada; TA 431 MO, UMO; JQ941483; JQ911392; JQ911244; JQ911080. Brassica repanda (Willd.) DC.; Spain; BGV UPM 155472; JQ941482; NA; JQ911243; JQ911079. Brassica rupestris Raf.; Italy, Sicilia, Monte Pelegrino; BGV UPM 382275; JQ941484, JQ941485; Version of Record (identical to print version). 987 Arias & Pires • Phylogeny of brassica crops and wild relatives TAXON 61 (5) • October 2012: 980–988 Appendix. Continued. JQ911393, JQ911394; JQ911245, JQ911246; JQ911081, JQ911082. Brassica souliei subsp. amplexicaulis Greuter & Burdet; Morroco, Beni Snassen; BGV UPM 114167; NA; JQ911501; NA; NA. Brassica spinescens Pomel; Algeria, Oron, Cabo Falcon; BGV UPM 180070; JQ941486; JQ911395; JQ911247; JQ911083. Brassica tournefortii Gouan; Spain, Campo de Nijar; BGV UPM 091666; JQ941487; JQ911396; JQ911248; JQ911084. Brassica villosa Biv.; Italy, Sicilia, W. Castelmare di Golfo; BGV UPM 6581; NA; JQ911397; JQ911351; JQ911194. Cakile lanceolata (Willd.) O.E. Schulz; origin unknown; BGV UPM 2206; JQ941489; JQ911399; JQ911250; JQ911086. Cakile maritima Scop.; origin unknown; BGV UPM 6080; JQ941490; JQ911400; JQ911251; JQ911087, JQ911088. Carrichtera annua (L.) C. DC.; Spain, Jumilla; BGV UPM 086566; JQ941492; JQ911402; JQ911254; JQ911090. Ceratocnemum rapistroides Coss. & Balansa; Morocco, Marrakech; TA 448 MO, UMO; JQ941497, JQ941498; JQ911406; JQ911259; JQ911095, JQ911096. Coincya longirostra Greuter & Burdet; Spain, Despeñaperros; BGV UPM 1175; JQ941502; JQ911410; JQ911263; JQ911100. Cordylocarpus muricatus Desf.; Morocco, Imzouren; MS 1106 MA; JQ941504; NA; JQ911265; JQ911102. Cordylocarpus muricatus Desf.; Morocco, Nador; Forther 6924 MA; JQ941505; JQ911412; JQ911266; JQ911103. Crambe filiformis Jacq.; origin unknown; TA 441 MO, UMO; JQ941509; JQ911416; JQ911269; JQ911108, JQ911110. Crambe hispanica L.; origin unknown; USDA 388853; JQ941499, JQ941500; JQ911407, JQ911408; JQ911260, JQ911261; JQ911097, JQ911098. Crambe hispanica subsp. abyssinica (Hochst. ex R.E. Fr.) A. Prina; Poland; USDA PI 284861; JQ941488; JQ911398; JQ911249; JQ911085. Crambe juncea M. Bieb.; Former Soviet Union; USDA PI 325274; JQ941501, JQ941510, JQ941511; JQ911409, JQ911418; JQ911262; JQ911099, JQ911107, JQ911111. Crambella teretifolia (Batt. & Trab.) Maire; Morocco, Oujda; Podlech 5162 MA; JQ941512; JQ911417; NA; JQ911109. Didesmus aegypticus (L.) Desv.; Greece, Paros-Parikia; Raus 16349 MA; JQ941519; JQ911419; NA; JQ911112. Didesmus bipinnatus (Desf.) DC.; Algeria, Biskra Bou Saada; BGV UPM 185370; JQ941513; JQ911420; JQ911270; JQ911113. Diplotaxis erucoides (L.) DC.; Spain, Finestrat; TA 480 MA, MO; JQ941514; JQ911273; NA; JQ911115. Diplotaxis erucoides (L.) DC.; Spain, Teulada; TA 550 MA, MO; JQ941515; JQ911421; JQ911274; JQ911116. Diplotaxis sp.; Spain; TA 543 MA, MO; JQ941517; JQ911422; JQ911272; JQ911114. Diplotaxis harra (Forsssk.) Boiss.; Spain, Albuñol; TA 518 MA, MO; JQ941516; JQ911423; JQ911271; JQ911117. Diplotaxis virgata (Cav.) DC.; origin unknown; BGV UPM 210876; JQ941518; JQ911424; JQ911276, JQ911277; JQ911118. Enarthrocarpus lyratus (Forssk.) DC.; origin unknown; BGV UPM 120668; JQ941535; JQ911426; JQ911279; JQ911120. Eremophyton chevallieri (Barratte ex L. Chevall.) Bég.; Morocco, Tagmoute; Buira 85 MA; JQ941521; JQ911427; JQ911280; JQ911121. Eruca pinnatifida (Desf.) Pomel; origin unknown; BGV UPM 24931; JQ941537; JQ911430; JQ911281; JQ911122. Eruca vesicaria (L.) Cav.; Spain; USDA 633218; JQ941530; JQ911433; JQ911290; JQ911132. Eruca vesicaria (L.) Cav.; Iran, Persepolis; BGV UPM 375077; JQ941538; JQ911434; NA; JQ911133. Eruca vesicaria subsp. sativa (Mill.) Thell.; Spain; TA 440 MO, UMO; JQ941523; JQ911431; JQ911282, JQ911284; JQ911126, JQ911139. Eruca vesicaria subsp. sativa (Mill.) Thell.; CGN 6849; JQ941524; JQ911432; JQ911283; JQ911125. Erucaria erucarioides (Coss. & Durieu) Müll.Berol.; Algeria, near to Bechar; BGV UPM 194471; JQ941531; JQ911435; JQ911291; JQ911124. Erucaria ollivieri Maire; Morocco, Zreouila, Sur de Agadir; BGV UPM 298374; JQ941522; NA; NA JQ911123; Erucastrum canariense Webb & Berthel.; Spain, Femes; BGV UPM 530579; JQ941525; JQ911507; JQ911285; JQ911127. Erucastrum elatum (Ball) O.E. Schulz; origin unknown; USDA PI 597837; JQ941532; JQ911436, JQ911437; JQ911292, JQ911293; JQ911135. Erucastrum elatum (Ball) O.E. Schulz; Spain; TA 528; JQ941589; JQ911491; JQ911356; JQ911197. Erucastrum gallicum (Willd.) O.E. Schulz; origin unknown; BGV UPM 12069; JQ941526; JQ911508; JQ911286; JQ911128. Erucastrum ifniense Gómez-Campo; origin unknown; BGV UPM 21348; JQ941528; JQ911438, JQ911429; JQ911288; JQ911130. Erucastrum leucanthum Coss. & Durieu; origin unknown; USDA PI 597846; NA; NA; NA JQ911136. Erucastrum littoreum (Pau & Font Quer) Maire; origin unknown; USDA 21352; JQ941533, JQ941534; NA; JQ911294, JQ911295; JQ911137. Erucastrum nasturtiifolium (Poir.) O.E. Schulz; origin unknown; USDA PI 597840; JQ941536; JQ911439; JQ911296; JQ911138. Erucastrum virgatum C. Presl.; origin unknown; USDA PI 597846; JQ941539; JQ911440; JQ911297; NA. Fezia pterocarpa Pit.; Morocco, Oeste de Msoum; BGV UPM 145968; JQ941540; JQ911441; JQ911298; JQ911140. Foleyola billotii Maire; Morocco, Zagora; TA 443 MO, UMO; JQ941541, JQ941542, JQ941543; JQ911442, JQ911443, JQ911444; JQ911299, JQ911300; JQ911141, JQ911142, JQ911143. Guiraoa arvensis Coss.; Spain, El Campello; TA 442 MO, UMO; JQ941544; JQ911445; JQ911301; JQ911144. Hemicrambe fruticulosa Webb; Morocco, Ceuta; BGV UPM 2232; JQ941545; JQ911446; JQ911302; JQ911145. Henophyton deserti (Coss. & Durieu) Coss. & Durieu; Algeria, El Goléa. Chevalier MA; JQ941547; JQ911448; JQ911304; JQ911147. Henophyton zygarrhenum (Maire) Gómez-Campo; Tunisia; Benedi 40 MA; JQ941546; JQ911447; JQ911303; JQ911146. Hirschfeldia incana (L.) Lagr.-Fossat.; Spain, Bejar; BGV UPM 202471; JQ941548, JQ941549; JQ911449; JQ911305, JQ911306; JQ911148. Kremeriella cordylocarpus (Coss. & Durieu) Maire; Morocco, Beni Snassen; TA 445 MO, UMO; JQ941556; JQ911455; JQ911312; JQ911153. Moricandia arvensis (L.) DC.; Spain, Jumilla; BGV UPM 086366; JQ941557; JQ911456; JQ911313; JQ911154. Moricandia foetida Bourg. ex Coss.; Spain, Bejar; TA549 MA, MO; JQ941558; JQ911457; JQ911314; JQ911155. Morisia monanthos (Viv.) Asch.; Italy, Cerdeña, Sassari; BGV UPM 381675; JQ941559; JQ911458; JQ911315; JQ911156. Muricaria prostrata (Desf.) Desv.; Algeria, Djelfa; Dubuis 14094 MA; NA; JQ911459; JQ911316; JQ911157. Physorhynchus chamaerapistrum (Boiss.) Boiss; Iran, Oeste de Behbahan; BGV UPM 374975; JQ941561; JQ911461, JQ911462; JQ911318; JQ911160, JQ911161. Pseuderucaria teretifolia (Desf.) O.E. Schulz; Tunisia; Podlech 35288 MA; JQ941564; JQ911464; JQ911320; JQ911162. Raffenaldia primuloides Godr.; Morocco, Gran Atlas, Col de Tamrhemt; TA 438 MO, UMO; JQ941565; JQ911465; JQ911321; JQ911165. Raphanus raphanistrum L.; Algeria, El Kala; BGV UPM 195571; JQ941566; JQ911466; JQ911323, JQ911324; JQ911166, JQ911167. Raphanus sativus L.; origin unknown; BGV UPM 127967; JQ941568; JQ911468, JQ911469; JQ911325; JQ911169. Rapistrum rugosum (L.) All.; Greece; TA 430 MO, UMO; JQ941567; JQ911467; JQ911322; JQ911168. Rytidocarpus moricandioides Coss.; origin unknown; BGV UPM 070867; NA; JQ911470; JQ911326; JQ911170. Savignya parviflora (Delile) Webb; Morocco, Outat-Oulab-El Hadjt; BGV UPM 146968; JQ941569; JQ911471; JQ911327; JQ911171. Schouwia purpurea (Forssk.) Scheweinfurth; Egypt, Wadi Kharit (Boulos) ; TA 437 MO, UMO; JQ941570; JQ911472; JQ911328; JQ911172. Sinapidendron angustifolium (DC.) Lowe; Portugal, Camara de Lobos; TA 434 MO, UMO; JQ941574; JQ911475; JQ911329, JQ911333; JQ911176, JQ911177. Sinapis alba L.; Spain, Vianos de Alcaraz; TA 447 MO, UMO; JQ941571, JQ941572, JQ941573; JQ911473, JQ911474; JQ911330, JQ911331, JQ911332; JQ911173, JQ911174, JQ911175. Sinapis arvensis L.; origin unknown; TA 464 MO, UMO; NA; JQ911476, JQ911477; JQ911334; JQ911178. Sinapis flexuosa Poir.; Spain, Castillo de Lorca; BGV UPM 086766; JQ941575; JQ911478; JQ911335; JQ911179. Sinapis pubescens L.; Italy, Regio Calabria; BGV UPM 382665; JQ941576; JQ911479; JQ911336; JQ911180. Trachystoma aphanoneurum (Maire & Weiller) Maire & Weiller; Morocco, Zaer; Sauvage 12599 MA; JQ941592; JQ911494, JQ911495; JQ911354; JQ911201. Trachystoma labasii Maire; Morocco, Meknés; FC 6843 MA; JQ941591; JQ911496; JQ911353; JQ911200. Trachystoma labasii Maire; origin unknown; Sennen & Mauricio 79 MA; JQ941506; JQ911413; JQ911267; JQ911104. Vella anremerica (Litard. & Maire) Gómez-Campo; Morocco, Lago Tislit, Gran Atlas; BGV UPM 439776; JQ941593, JQ941594; JQ911497; JQ911359; JQ911202, JQ911203, JQ911204. Vella bourgeana (Cosson) Al-Shehbaz & Warwick; Spain, Tabernas; TA 578 MA, MO; JQ941595; JQ911498; JQ911358, JQ911360; JQ911205. Vella spinosa Boiss.; Spain, Sierra de Maria; BGV UPM 0924; JQ941596; JQ911499; JQ911361; JQ911206. Zilla macroptera Coss.; Morocco, Siroua, Cerca de Iriri; BGV UPM 1096; JQ941597, JQ941598; JQ911500; JQ911362, JQ911363; JQ911207, JQ911208. — OUTGROUPS: Alliaria petiolata (M. Bieb.) Cavara & Grande; Spain, Casa de Campo; BGV UPM 23; JQ941446; JQ911364; JQ911209; JQ911043. Calepina irregularis (Asso) Thell.; origin unknown; BGV UPM 2118; JQ941491; JQ911401; JQ911252, JQ911253; JQ911089. Caulanthus cooperi (S. Watson) Payson; United States, California, Julian; BGV UPM 1734; JQ941494; JQ911403; JQ911255; JQ911093. Caulanthus inflatus S. Watson; origin unknown; BGV UPM 1225; JQ941495; JQ911505; JQ911256; JQ911091. Caulanthus pilosus S. Watson; origin unknown; coll. unknown MO, UMO; JQ941496; JQ911404; JQ911258; JQ911094. Conringia orientalis (L.) Dumort.; Spain, Cañadas de Correpta; BGV UPM 093466; JQ941507; JQ911414, JQ911415; NA; JQ911105. Isatis tinctoria L.; Morocco, Midelt; BGV UPM 3011; JQ941555; JQ911453, JQ911454; JQ911310, JQ911311; JQ911151, JQ911152. Myagrum perfoliatum L.; Turkey, Iyiedere, Playa; BGV UPM 6264; JQ941560; JQ911460; JQ911317; JQ911158. Pseudofortuynia esfandiarii Hedge; Iran, Abadeh; TA 444 MO, UMO; JQ941562, JQ941563; JQ911463; JQ911319; JQ911163, JQ911164. Sisymbrium irio L.; Spain, Fuencarral; BGV UPM 1146; JQ941579; JQ911482; JQ911338; JQ911183. Sisymbrium officinale (L.) Scop; Spain; TA 503; JQ941580; JQ911483; JQ911341; JQ911186. Sisymbrium orientale L.; United States, California, Int. Mohave y Barstow; BGV UPM 7764; JQ941581, JQ941582; JQ911484, JQ911485; JQ911339, JQ911342, JQ911343; JQ911184. Stanleya bipinnata Greene; United States, California, E. de Mohave; BGV UPM 7763; NA; JQ911486; JQ911344; JQ911187. Streptanthus drepanoides Kruckeb. & J.L. Morrison; United States, California; JQ941585; JQ911489; JQ911346; JQ911190. Streptanthus heterophyllus Nutt.; United States, California; BGV UPM 1730; JQ941586, JQ941587; JQ911487; JQ911347; JQ911188. 988 Version of Record (identical to print version).