UC Berkeley UC Berkeley Previously Published Works

Title Cryptic host-specific diversity among western hemisphere broomrapes (Orobanche s.l., Orobanchaceae).

Permalink https://escholarship.org/uc/item/3q57c2cm

Journal Annals of botany, 118(6)

ISSN 0305-7364

Authors Schneider, Adam C Colwell, Alison EL Schneeweiss, Gerald M et al.

Publication Date 2016-11-01

DOI 10.1093/aob/mcw158

Peer reviewed

eScholarship.org Powered by the California Digital Library University of California Annals of Botany 118: 1101–1111, 2016 doi:10.1093/aob/mcw158, available online at www.aob.oxfordjournals.org

Cryptic host-specific diversity among western hemisphere broomrapes (Orobanche s.l., Orobanchaceae)

Adam C. Schneider1,2,*, Alison E. L. Colwell2, Gerald M. Schneeweiss3 and Bruce G. Baldwin1,2 1Department of Integrative Biology, 1005 Valley Life Sciences Building, University of California, Berkeley, CA 94720, USA, 2Jepson Herbarium, 1001 Valley Life Sciences Building, University of California, Berkeley, CA 94720, USA and 3Department of Botany and Biodiversity Research, University of Vienna, 1030 Vienna, Austria *For correspondence. E-mail [email protected]

Received: 30 April 2016 Returned for revision: 6 June 2016 Accepted: 24 June 2016 Published electronically: 18 August 2016

Background and Aims The broomrapes, Orobanche sensu lato (Orobanchaceae), are common root parasites found across Eurasia, Africa and the Americas. All species native to the western hemisphere, recognized as Orobanche sections Gymnocaulis and Nothaphyllon, form a clade that has a centre of diversity in western North America, but also includes four disjunct species in central and southern South America. The wide ecological distri- bution coupled with moderate taxonomic diversity make this clade a valuable model system for studying the role, if any, of host-switching in driving the diversification of plant parasites. Methods Two spacer regions of ribosomal nuclear DNA (ITS þ ETS), three plastid regions and one low-copy nuclear gene were sampled from 163 exemplars of Orobanche from across the native geographic range in order to infer a detailed phylogeny. Together with comprehensive data on the parasites’ native host ranges, associations be- tween phylogenetic lineages and host specificity are tested. Key Results Within the two currently recognized species of O. sect. Gymnocaulis, seven strongly supported clades were found. While commonly sympatric, members of these clades each had unique host associations. Strong support for cryptic host-specific diversity was also found in sect. Nothaphyllon, while other taxonomic species were well supported. We also find strong evidence for multiple amphitropical dispersals from central North America into South America. Conclusions Host-switching is an important driver of diversification in western hemisphere broomrapes, where host specificity has been grossly underestimated. More broadly, host specificity and host-switching probably play fundamental roles in the speciation of parasitic plants.

Key words: Amphitropical disjunction, cryptic speciation, holoparasite, host-switching, Orobanche, Orobanchaceae, parasite, phylogeny.

INTRODUCTION clades (Schneeweiss et al., 2004a; Park et al., 2008; McNeal et al.,2013; Schneeweiss, 2013), and the significant economic Parasitism is a highly successful life strategy that has evolved damage caused by several Eurasian species to major agricul- independently >60 times among animals, at least 12 times tural systems worldwide (Joel et al.,2013). among angiosperms, and repeatedly in protozoans and prokary- Despite the interest in this group, relatively little is known otes (Poulin and Morand, 2000; Westwood et al.,2010). While about the role of host specificity in broomrape diversification. the evolutionary significance of host–parasite associations has Understanding host specificity of parasites is predicated on a long been recognized (Kellogg, 1913), the main evolutionary comprehensive understanding of lineage boundaries in the host mechanisms involved in the generation and maintenance of (e.g. Labrousse et al.,2001; Timko et al.,2012)and,moreim- such ecological and phylogenetic diversity are still poorly un- portantly for Orobanche, the parasite. That is, failure to recog- derstood, especially among parasitic flowering plants (de nize evolutionary diversity in the parasite results in an Vienne et al., 2013; Joel et al.,2013). overestimation of host breadth and may limit the ability to un- The parasitic broomrapes, Orobanche sensu lato (s.l.)[alter- derstand the evolutionary processes responsible for speciation natively circumscribed as the genera Aphyllon and Myzorrhiza in plant parasites (Refre´gier et al., 2008). Therefore, it is impor- in the New World, and Boulardia, Orobanche sensu stricto tant to distinguish true host generalists from taxa that comprise (s.s.)andPhelipanche in the Old World: Schneeweiss (2013)], several cryptic lineages artificially united on the basis of super- have attracted significant attention as an important system for ficial similarity but distinguished genetically and ecologically. understanding the evolutionary consequences of parasitism. Host specificity to the family or genus level has been cited as a This attention is in part a result of their extensive worldwide di- key factor in the differentiation and genetic isolation of three versity (at least 170 species; Ulrich et al., 1995), a detailed and subspecies of the European O. minor (Thorogood et al.,2008, well-supported understanding of their placement within the 2009), but this has not been broadly tested across other family Orobanchaceae as well as the relationships among major Orobanche lineages. Several recently described species of

VC The Author 2016. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved. For Permissions, please email: [email protected] 1102 Schneider et al. — Evolution of New World Orobanche

Marie-Victorin 27906 WIS282511 nrDNA Talbot 160 ALAv122774 Lipkin 04-324 ALAv156078 (ITS + ETS) Bouchard 86-94 MT26938 1 Wood s.n. MO6010393 O. uniflora subsp. uniflora Sheviak 7072 UC2046154 Melburn s.n. V101353 Lomer 7268 UBC234660 (Astereae + Rudbeckia) Kucyniak 76 WIS282505 Henson 1499 WIS282510 dePamphilis 94-15 PENN 1 Cochrane 255 WIS282508 Anderson 2267 MIN Thorne 4794 NY 0·99 Oswald 5615 UC1609174 Polster s.n. V93858 Antennaria 0·98 Ahart 10765 CHSC87344 Cowell 07-58 YM216905 + Senecio 1 1 Lackschewitz 4307 NY 1 Heckard 3286 JEPS70862 Saxifragales 1 Colwell 04-63 YM118104 O. uniflora subsp. 1 Colwell 96-WA-LC WTU344845 Fiely 91 WS285758 occidentalis Duthie s.n. WIS282512 1 Ahart 9846 CHSC82361 1 Ahart 7286 CHSC63258 Ahart 1984 CAS853689 Apioideae Colwell 14-11 JEPS 126167 Chisaki 661 WIS282509 Hitchcock 18630 WS185460

0·96 Artemisia 1 Halse 908 ARIZ187532 Lackschewitz 6619 WTU272369 Bohrer1684 ARIZ189922 1 Merner s.n. WIS282500 0·98 Keith s.n. WIS282501 0·99 Batten 78-279 ALAv85763 1 0·99 Colwell 95-CO-TFS WTU344763 Orobanche sect. Bell 159 SD136414 1 Banks 689 SD137701 Gymnocaulis 1 Harvey s.n. UC 0.94 Cox 188 CHSC12931 Galium 1 Colwell 07-53 YM 0·97 Colwell 10-107 YM 1 0·98 Colwell 04-83 YM118109 Yatskievych 82-196 ARIZ236136 O. fasciculata Boyd 9673 RSA599655 Egger 1295 WTU Ahart 3393 CHSC44286 Eriogonum, 1 Colwell04-54 YM117945 Colwell 04-03 YM118118 Eriodictyon, Howell 51680 CAS641193 1 Colwell 02-44 YELLO Eriophyllum, Colwell 02-09 WTU351387 Schneider 920 JEPS122893 + Phacelia Schneider 468 JEPS 0.98 Colwell 01-95 WTU351381 Colwell 99-CA-RCRS WTU344759 0·93 Mistretta 1867 RSA641370 Mistretta 1866 RSA641369 subsp. valida 1 Colwell 99-73 WTU344847 O. valida (Garrya) Rugyt 1823 JEPS110567 1 Colwell 02-04 WTU351386 subsp. howellii 0·93 Santana 5977 WIS282502 O. dugesii Vandevender 95-1215 ARIZ321887 Collins 2027 (subsp.palmeri) O. cooperi 0·95 1 1 Colwell 02-06 WTU351385 Colwell CA-SN WTU344843 (Heliantheae) 1 Colwell 01-01 WTU344743 Schneider 936 JEPS122909 Hammond 10858 ASU60715 0·99 1 1 Collins 1541 WIS282497 Colwell 03-08 JEPS126150 O. bulbosa (Adenostoma) Orobanche 1 Colwell 99-01 WTU344744 Colwell (5-July-2009) 1 Hamilton s.n. JEPS126153 sect. Nothaphyllon Colwell 05-413 JEPS126144 O. parishii subsp. parishii (Artemisia) 1 1 Colwell 95-CO-DD WTU344764 Colwell 01-114 WTU351399 O. ludoviciana (Artemisia) 1 Collins 2035 MO6012141 (subsp. mutabilis) 0·97 Leidolf 2385 WTU344762 O. corymbosa (Artemisia) Colwell 99-08 WTU344753 Taylor 18332 WTU351384 Taylor 18333 WTU351383 O. parishii subsp. parishii Colwell 03-53 JEPS126149 1 Colwell 07-72 JEPS126137 (Asteroideae) Colwell 05-414 JEPS126143 0·93 Colwell 14-26 JEPS126165 0·98 Colwell 06-421 JEPS126163 Colwell 05-212 JEPS126139 O. corymbosa (Artemisia) 0·99 Holmgren 1402 WTU224934 0·97 Colwell 14-25 YM subsp. grayana (Astereae) Colwell 05-259 JEPS Colwell 03-57 WTU O. californica 1 AC 97-CA-BLO WTU344755 1 AC 99-77 WTU344846 subsp. californica (Grindelia)

AC 96-WA-DPSP O. californica AC 99-76 WTU344749 0·93 Stoughton 583 RSA767875 subsp. parishii 1 Colwell (7-May-2006) JEPS O. parishii 1 Sproul & Wolf s.n. JEPS subsp. brachyloba 0·98 Colwell JEPS126159 0·95 Colwell 07-25 JEPS126141 subsp. parishii Colwell 01-104 WTU351396 subsp. jepsonii 1 Colwell 99-75 WTU344745 Colwell JEPS126158 1 Colwell 99-02 WTU344758 subsp. feudgei (Artemisia) O. californica Colwell 05-33 JEPS126147 1 Colwell JEPS126157 complex 1 Colwell 02-57 WTU351404 1 Colwell 01-116 WTU351394 Colwell 01-106 WTU351393 O. vallicola (Sambucus) 1 Prothero s.n. JEPS126154 Colwell JEPS 126162 JEPS 1 Colwell 02-52 WTU351391 Colwell 00-44 WTU344754 subsp. grandis (Lessingia) Colwell JEPS126164 Colwell 05-101 JEPS126146 Gowen s.n. JEPS126142 subsp. jepsonii Colwell 08-599 JEPS126151 subsp. grandis (Lessingia) Taylor 18568 JEPS100408 Taylor 18566 JEPS100409 subsp. condensa (Heterotheca) Taylor 18565 JEPS100410 Colwell 04-314 0·98 Colwell 02-51 WTU351389 O. californica 1 Colwell 02-49 WTU351400 Grindelia Colwell 02-48 WTU351388 Colwell 04-313 JEPS126160 subsp. californica 1 Schneider 293 JEPS Colwell 15-01 JEPS126152 Eriophyllum 0.98 Colwell 04-196 JEPS126148 Colwell 02-54 1 Egger 804 WTU332502 Collins 1528 WIS282503 Colwell 96-WA-IC WTU344760 O. pinorum (Holodiscus) Colwell 01-107 WTU352202 1 Ricardi 3326 CONC19273 O. tarapacana (Trixis)

Bobadilla s.n. CONC164753 O. ludoviciana Yatskyevich 85-215 ARIZ264197 Collins 2031 MO6012144 (M) O. sp. nov (Zaluzania) 1 Collins 2030 MO6012145 (M) Collins 2026 UC2046172 (M) Collins 2025 MO5876462 (M) Colwell 01-93 WTU344844 (L) (subsp. arenosa) Collins 2034 MO (M) O. ludoviciana + multiflora Collins 2024 MO5990497 (M) 0·99 Dueholm 1164 WIS10644 (L) (Astereae + Varilla) Collins 2033 UC2046173 (L) Colwell 01-113 WTU351402 (L) Collins 1533 WIS (L) 1 Smith 2903 MO3646410 (M) complex Alfaro 3461 MO2637419 O. tacnaensis Long 2240 UC2046156 1 1 Vilagran 8616 SGO142749 Rosas 3327 CONC169912 O. chilensis (Grindelia) Garcia 3877 SGO154435 Harrington 10081 CS46066 CS42888 1 Collins 2032 MO6012143 Collins 1622 MO5933323 O. riparia (Ambrosiinae) Collins 1620 MO4903216 Yatskievych 91-195 MO3907149 Halse897 ARIZ187291 O. arizonica (Gutierrezia) Schneider et al. — Evolution of New World Orobanche 1103

Orobanche in North America also have unique host preferences information is provided in Supplementary Data Table S1). This in the Asteraceae: Orobanche riparia parasitizes Helianthieae data set includes at least one exemplar of all taxa of Orobanche sub-tribe Ambrosiinae and O. arizonica parasitizes Gutierrezia recognized within the last 75 years except for O. weberbaueri,a spp. However, neither these species concepts nor those of the poorly known taxon from southern coastal Peru, perhaps known other American Orobanche species has ever been tested only from the type. Denser population sampling across sect. phylogenetically. Gymnocaulis enabled more comprehensive geographic and host Inclusion of western hemisphere Orobanche (sections range sampling in the two currently recognized species of this Gymnocaulis and Nothaphyllon) in phylogenetic studies has section, O. fasciculata and O. uniflora (Fig. 1). Identifying the been limited to several exemplars included in larger genus- or host breadth for each taxon was challenging, as many collectors family-level analyses. These studies, supported by karyological note the nearest living plant as the host species without confirm- and morphological evidence, have shown that these two sections ing a haustorial connection, resulting in a proliferation of dubi- are sister groups and together are sister to an Old World clade ous records. Our criteria for accepting a host was that a host corresponding to Orobanche sect. Trionychon (Schneeweiss taxon must have been independently reported at several popula- et al.,2004a; Park et al.,2008), more recently treated as the ge- tions by more than one collector, or a haustorial connection to an nus Phelipanche (Schneeweiss, 2013). This larger clade is sup- identifiable fragment of host must be present on the herbarium ported by a shared base chromosome base number of x ¼ 12 voucher. Host associations for sampled populations are listed in (Heckard and Chuang, 1973; Schneeweiss et al.,2004b). Table S1. For molecular phylogenetic analyses, one individual Ecologically, Orobanche sections Gymnocaulis and each of O. gracilis and O. hederae were used as the outgroup Nothaphyllon parasitize a wide range of eudicot hosts, but most (Park et al.,2008; McNeal et al., 2013). Sequence data for the commonly perennial Asteraceae. Taxonomic diversity is con- waxy locus were not available for these outgroups, so instead centrated in the California Floristic Province; however, species two more distantly related outgroup taxa were used, Castilleja can be found across the Americas, as far north as the Alaska ambigua and Triphysaria versicolor. Peninsula and the Yukon Territory, east to Newfoundland, and south to central Mexico. Four poorly known species are found in South America. Affinities between South American DNA extraction, amplification and sequencing Orobanche chilensis and North American O. ludoviciana have DNA was extracted from dried floral tissue using a long been recognized (Beck, 1890), but explicit biogeographic DNeasy Plant Mini Kit (Qiagen, Valencia, CA, USA), or us- hypotheses for this or other such relationships within the clade ing a CTAB (cetyltrimethylammonium bromide) protocol have yet to be proposed. (Doyle and Doyle, 1987). A total of six regions from the nu- The wide ecological and host diversity among western hemi- clear and plastid genomes were used to estimate the phylog- sphere Orobanche, as well as its tractable taxonomic diversity, eny: internal and external transcribed spacers of nuclear make it a valuable model system for understanding the main ribosomal DNA (ITS and ETS, respectively), introns 9–11 of ecological and evolutionary processes affecting parasite diversi- the nuclear low-copy gene waxy, as well as the plastid trnL- fication and speciation. Such investigations, however, are requi- trnF region (comprising the trnL intron and the trnL - site for a robust understanding of evolutionary lineages, their UAA UAA trnF intergenic spacer) and the matK and rps2 genes. host breadths and their relationships. Specifically our goals GAA ITS, matK,andrps2 were selected based on their prior use in were to (1) reconstruct a well-resolved phylogeny of western genus- and family-level phylogenetic studies of Orobanche hemisphere that could be used to develop a revised, Orobanche (Schneeweiss et al., 2004a; McNeal et al., 2013), and waxy natural classification for the group; (2) evaluate the evolution- for its use in the related (hemi-)parasitic genus Castilleja ary significance of host-switching in Orobanche sect. (Tank and Olmstead, 2008). The remaining two regions, ETS Gymnocaulis by comprehensively sampling across the geo- and trnL-trnF, were selected to provide additional rapidly graphic and host ranges of each taxon; (3) test the monophyly evolving characters from the nuclear and plastid compart- of longstanding taxa as well as recently described segregates; ments, respectively. Due to difficulty assessing homology and (4) infer biogeographical relationships between North within some species of sect. Nothaphyllon,thewaxy locus American and South American spp. Orobanche was mainly used to assess monophyly of sect. Nothaphyllon and to infer relationships within sect. Gymnocaulis. Polymerase chain reaction (PCR) amplifications were per- MATERIALS AND METHODS formed using AccuPower PCR PreMix kits (Bioneer, Alameda, CA, USA) or by generating a master mix of 10 lLof5 Taxon and population sampling Promega buffer, 4 lLof25mM MgCl2,125 lLof10mM 163 Orobanche populations were sampled either from fresh col- dNTPs, 1 lLof20lM of each primer and 025 lLofGo-Taq lected tissue or from herbarium collections: 57 from sect. DNA Polymerase (Promega, Madison, WI, USA) diluted to Gymnocaulis and 106 from sect. Nothaphyllon (voucher and host 50 lL. Complete information about primers, cycling parameters

FIG. 1. Bayesian inference majority-rule consensus tree of 162 Orobanche populations inferred from nrDNA (ITS þ ETS). Tip labels include the collection number followed by the herbarium accession number, if available. Posterior probabilities >09areshowninboldfornodeswith>70 % maximum likelihood bootstrap (BS) support and in italics if BS support is < 70 %. The internal branches leading to section Gymnocaulis and section Nothaphyllon have been shortened by a factor of 1/2. Host associations are indicated in blue (Asteraceae) or green (other) to the genus or higher taxonomic level. Informally named clades are in purple. Outgroup taxa are not shown. Photographs, from top to bottom: O. fasciculata parasitizing Eriodicyton sp. (Schneider 606); O. cooperi parasitizing Hymenoclea salsola (Schneider 415); O. vallicola parasitizing Sambucus mexicana (Schneider 316); O. corymbosa parasitizing Artemisia tridentata (Colwell 14–26). 1104 Schneider et al. — Evolution of New World Orobanche and amplicon sizes are provided in Table 1. PCR products were corresponding to subsp. uniflora. Members of this clade parasit- purified using ExoSAP (USB Products, Cleveland, OH, USA), ize Rudbeckia and several genera of Astereae in the Asteroideae. and both DNA strands were sequenced using an ABI 3730 Populations of the remaining American Orobanche species, DNA analyzer (Applied Biosystems, Foster City, CA, USA). representing sect. Nothaphyllon were generally resolved in one GenBank accession numbers can be found in Table S1. of eight major clades (PP > 095, BS > 90): (1) a clade of popu- lations from the western USA parasitic on Artemisia previously determined as one of three taxa: O. parishii subsp. parishii, Sequence alignment and phylogenetic reconstruction O. ludoviciana or O. corymbosa; (2) a taxonomically and eco- logically diverse clade, the O. californica complex, which in- Sequences were checked for base-calling errors and assem- cluded O. californica and O. vallicola, as well as the remainder bled into contigs using Geneious v.6.1.7 (Biomatters, Auckland, of O. parishii and O. corymbosa populations; (3) O. pinorum; New Zealand). Sequence alignments were generated using the (4) O. tarapacana; (5) the O. ludoviciana complex, including MUSCLE plug-in with default settings. Maximum likelihood O. multiflora, O riparia, O. chilensis, O. tacnaensis, O arizon- (ML) and Bayesian inference (BI) analyses were conducted sep- ica, the remainder of O. ludoviciana and a collection from arately on the concatenated chloroplast DNA matrix (cpDNA), Hidalgo, Mexico (Yatskievych 85-215) that does not match the the concatenated ribosomal spacers (nrDNA) and the waxy lo- morphology of any described species; (6) O. valida;(7)O. cus using the CIPRES Science Gateway (Miller et al., 2010). cooperi and O. dugesii; and (8) O. bulbosa. Clades 6–8, found The ML analyses were performed with RAxML-HPC2 v.8.2.6 predominantly in south-western North America, constituted a (Stamatakis, 2014) using the GTRCAT model with 25 rate cate- monophyletic group (PP ¼ 095, BS ¼ 77) that was sister to the gories and 1000 rapid bootstrap (BS) replicates. The BI analyses rest of the section (clades 1–5). Resolution at the subspecific were performed using MrBayes v.3.2.6 (Ronquist et al., 2012). level of the paraphyletic O. californica was variable. For exam- We used an AIC (Akaike information criterion) comparison im- ple, populations of subsp. californica along the central plemented in jmodeltest2 (Darriba et al., 2012)toselectaGTR California coast parasitizing Grindelia stricta and those in far þ C substitution model (approximated using four rate catego- northern California and Washington parasitizing Grindelia ries). The estimated substitution rates for the nrDNA, cpDNA integrifolia were resolved in separate strongly supported sub- and waxy alignments were then used as priors in the MrBayes clades within the O. californica complex (clade 2, above). analysis. Default settings were used for other priors. Three inde- Other subspecies, such as subsp. grandis and subsp. condensa, pendent runs of four chains each (one cold, three heated) were formed a polytomy. The polyploid O. parishii subsp. brachy- sampled every 1000 generations for 2 500 000 generations. The loba was nested within one of three separate clades of O. par- first 20 % of samples were discarded as burn-in. Convergence ishii subsp. parishii. was assessed in several ways: the average standard deviation of split frequencies was <001, the potential scale reduction factor was close to 100 for all parameters, and the effective sample sizes (ESS) were >800. cpDNA Strongly supported clades from the Bayesian analysis of three plastid regions (Fig. 2) were consistent with those identi- RESULTS fied by ML (data not shown). Orobanche sect. Gymnocaulis was resolved as monophyletic (PP 1 0, BS 100). Within nrDNA ¼ ¼ sect. Gymnocaulis, six host-specific clades were resolved, con- Strongly supported clades in the Bayesian ITS/ETS analysis gruent with the nrDNA results. Three of these were sub-clades (Fig. 1) were consistent with those identified by ML (data not of the monophyletic O. uniflora (PP ¼ 099, BS ¼ 97): a clade shown). Orobanche sect. Gymnocaulis and sect. Nothaphyllon of plants parasitizing Antennaria and Senecio (PP ¼ 10, were both resolved as monophyletic [posterior probability (PP) BS ¼ 93) and a less supported clade of plants parasitizing ¼ 10, BS ¼ 100] and sister to each other. Within sect. Apioideae (Apiaceae), Saxifragaceae and Crassulaceae Gymnocaulis, seven major clades were resolved (PP ¼ 10, (PP ¼ 071, BS ¼ 88), together corresponding to subsp. occi- BS 80). Under the current classification, three of these to- dentalis (PP ¼ 10, BS ¼ 100) and sister to a clade of plants that gether correspond to a paraphyletic O. fasciculata. Plants from parasitize several genera of Asteroideae corresponding to subsp. each of these clades showed unique host preferences: plants in uniflora (PP ¼ 1, BS ¼ 100). Orobanche fasciculata was found two of these groups parasitize hosts of single genera, Artemisia to be paraphyletic: a strongly supported clade parasitizing (Asteraceae) and Galium (Rubiaceae). The third group of plants Artemisia (PP ¼ 10, BS ¼ 100) was resolved sister to O. uni- form a clade of generalists that parasitize numerous species flora. The remaining two clades of O. fasciculata were resolved within Eriogonum (Polygonaceae), Eriophyllum (Asteraceae), as sister groups, one strongly supported and parasitizing and Eriodictyon and Phacelia (Hydrophyllaceae). The remain- Galium spp. in California and Oregon (PP ¼ 10, BS ¼ 100), ing four clades constituted a monophyletic O. uniflora and the other weakly supported and parasitizing a variety of dis- (PP ¼ 10, BS ¼ 100). Three of these clades include parasites tantly related core eudicot genera (PP ¼ 065, BS < 50). specific to hosts in the genera Antennaria and Senecio Deep relationships within Orobanche sect. Nothaphyllon (Asteraceae), on members of Saxifragaceae and Crassulaceae were generally well resolved, albeit with variable support at (Saxifragales s.s.) and on Apioideae (Apiaceae), respectively. the species and subspecies level. Populations of O. bulbosa These clades together are currently recognized as O. uniflora formed a clade (PP ¼ 10, BS ¼ 96) that was sister to the re- subsp. occidentalis and were resolved sister to the fourth clade mainder of the section, which in turn comprised two well- Schneider et al. — Evolution of New World Orobanche 1105

TABLE 1. Molecular regions used in the phylogenetic analyses of Orobanche sections Gymnocaulis and Nothaphyllon, approximate lengths of complete ingroup sequences, PCR primers (50–30) and thermocycling parameters

Gene region Approximate Primer sequences Reference Thermocycling parameters amplicon length

ITS 590 bp AB_101: TGG TCC CGT GAA GTG TTC G Schneeweiss et al. (2004a)94C, 4 min; 35 (95 C, 1 min; 48 C, 1 min; 72 C, 1 min); 72 C, 10 min. AB_102: CCG GTT CGC TG CCG TAA C Schneeweiss et al. (2004a) ETS 430 bp ETS_B: ATA GAG CGC GTG AGT GGT G Beardsley and Olmstead, 96 C, 2 min; 35 (94 C, 30 s; 2002) 56 C, 30 s; 72 C, 45 s); 72 C, 3 min. ETS_seq: (C) TGG CAG GAT CAA CCA GGT A This study waxy 585–630 bp waxy_9F-ORO: GAT GCT AAG CCW TTG TTG A This study 92 C, 5 min; 40 (94 C, 45 s; (introns 9–11) 535 C, 45 s; 72 C, 1 min); 72 C, 5 min. waxy_11R: CCA TRT GGA ASC CAG TRT A Tank and Olmstead (2009) matK 30 intron 680–760 bp matK 8: CTT CGA CTT TCT TGT GCT Steele and Vilgalys (1994) 94 C, 5 min; 40 (92 C, 1 min; 51 C, 40 s; 72 C, 1 min); 72 C, 10 min. matK_psbA50R: AAC CAT CCA ATG TAA Shaw et al. (2005) AGA CGG TTT rps2 675 bp rps2_2F: AAA TGG AAT CCT AAA ATG GC This study 94 C, 2 min 30 s; 35 (94 C, 1 min; 50 C, 1 min; 72 C, 1 min); 72 C, 7 min. rps2_18F: GGR KAR AAA TGA CAA GAA GAT dePamphilis et al. (1997) ATT GG rps2_661R: ACC CTC ACA AAT GCG AAT ACC AA dePamphilis et al. (1997) trnL-trnF spacer 710–810 bp trL ‘c’: CGA AAT CGG TAG ACG CTA CG Taberlet et al. (1991) 94 C, 5 min; 40 (92 C, 1 min; 515 C, 1 min; 72 C, 1 min); 72 C, 5 min. trnF ‘f’: ATT TGA ACT GGT GAC ACG AG Taberlet et al. (1991)

Two different forward primers for rps2 were used. supported sub-clades (PP ¼ 10, BS > 95). The first included another that included a sub-clade of parasites on Antennaria strongly supported clades corresponding to single taxa that di- and Senecio (Asteraceae) united in a moderately supported verged from the remainder of the sub-clade in succession: O. polytomy with several populations that parasitize Apioideae valida (PP ¼ 10, BS ¼ 100), O. parishii (PP ¼ 10, BS ¼ 100) (PP ¼ 089, BS ¼ 074). The remaining two strongly supported and finally O. tarapacana (PP ¼ 094, BS ¼ 72), which was clades include plants currently recognized as O. fasciculata: sister to a clade of O. cooperi, O. dugesii and one accession one was sister to O. uniflora and parasitizes Artemisia; the of O. corymbosa (PP ¼ 098, BS ¼ 68). The second well- other parasitizes Galium and was sister to the remaining popu- supported sub-clade included the only sampled population of lations of O. fasciculata, which formed a third, weakly sup- O. pinorum sister to the O. californica and O. ludoviciana ported clade (PP ¼ 074, BS ¼ 67) including parasites on a complexes. Relationships within this sub-clade were poorly variety of core eudicot hosts. In contrast to Orobanche sect. resolved, except for strong support of O. riparia þ O. arizon- Gymnocaulis, infraspecific sampling density and phylogenetic ica, O. vallicola, a clade of O. californica subsp. californica resolution within O. sect. Nothaphyllon was limited, although parasitic on Eriophyllum staechadifolium, and O. chilensis þ conspecific populations of O. valida, O. californica subsp. several populations from central North America (PP ¼ 10, californica and O. cooperi, as well as O. chilensis þ O. multi- BS > 97). flora were each resolved as monophyletic (PP > 094, BS > 90). Tree files were uploaded to Open Tree of Life (http://www.opentreeoflife.org), study ID ot_732. waxy Orobanche sect. Gymnocaulis and sect. Nothaphyllon were each resolved as monophyletic (PP ¼ 099, BS > 75). Within DISCUSSION sect. Gymnocaulis, five host-specific clades were resolved Host specificity and speciation with strong support (PP > 092, BS > 73), congruent with both nrDNA and cpDNA results. These included a clade of Among extant western hemisphere Orobanche,wereportmany plants parasitizing several genera in the Asteroideae corre- previously unrecognized, host-specific lineages in both sect. sponding to O. uniflora subsp. uniflora, as well as two clades Gymnocaulis and sect. Nothaphyllon that are strongly supported together corresponding to O. uniflora subsp. occidentalis – by both plastid and nuclear DNA sequences (Figs 1–3). This the first, which was comprised of plants parasitizing cryptic diversity has two complementary implications – one Saxifragaceae and Crassulaceae (Saxifragales s.s), and evolutionary, the other ecological. First, biodiversity within Wood s.n. MO6010393 Talbot 160 ALAv122774 1106 1 Lipkin 04-324 ALAv156078 O. uniflora Lomer 7268 UBC234660 Astereae + cpDNA Kucyniak76 WIS282505 Rudbeckia subsp. uniflora Cochrane 255 WIS282508 (matK + rps2 + trnL-trnF) Bouchard & Hay 86-94 MT26938 0·99 Anderson 2267 MIN 1 Ahart 10765 CHSC87344 Antennaria Polster s.n. V93858 Oswald 5615 UC1609174 + Senecio 1 0·96 Colwell 07-58 YM216905 Chisaki 661 WIS282509 1 Ahart 1984 CAS853689 O. uniflora subsp. Colwell 14-11 JEPS126167 Apioideae + Colwell 96-WA-IC WTU344845 occidentalis Fiely 91 WS85758 Saxifragales Ahart 9846 CHSC82361 0·98 Colwell 04-63 YM118104 1 Bohrer 1684 ARIZ189922 1 Batten 78-279 ALAv85763 Artemisia Colwell 95-CO-TFS WTU344763 1 Bell 159 SD136414 1 Banks689 SD137701 1 Harvey s.n. UC Galium Cox 188 CHSC12931 O. Schneider 0.97 Colwell 10-107 YM 1 Colwell 07-53 YM fasciculata Yatskievych 82-196 ARIZ236136 1 Schneider 920 JEPS122893 Eriogonum, Ahart 3393 CHSC44286 Eriodictyon 1 Schneider 468 JEPS 1 Schneider 936 JEPS122909 Egger 1295 WTU Eriophyllum, al. et Colwell 03-08 JEPS126150 O. bulbosa Colwell 02-09 WTU351387 0·94 Colwell (5-July-2009) Colwell 01-95 WTU351381 Phacelia 0.98 Mistretta 1867 RSA641370 Colwell 99-CA-RCRS WTU344759

subsp. valida World New of Evolution — 1 Colwell 99-73 WTU344847 Rugyt1823 JEPS110567 O. valida subsp. howellii 1 Colwell 02-04 WTU351386 1 1 Colwell (7-May-2006) 1 Sproul & Wolf s.n. O. parishii subsp. brachyloba 0·94 Ricardi 3326 CONC19273 1 Bobadilla s.n. CONC164753 O. tarapacana O. dugesii Santana 5977 WIS282502 O. cooperi Vandevender 95-1215 ARIZ321887 0·98 O. corymbosa Colwell 99-08 WTU344753 O. cooperi + O. dugesii + O. corymbosa 1 O. cooperi Colwell 02-06 WTU351385 1 O. cooperi Colwell 01-01 WTU344743 Colwell 96-WA-IC WTU344760 O. pinorum Colwell 01-93 WTU344844 Colwell 01-114 WTU351399 O. ludoviciana 1 Colwell 01-116 WTU351394 0·91 0·98 1 Colwell JEPS126162 O. vallicola 1 Schneider 293 JEPS Colwell 15-01 JEPS126152 O. californica (Eriophyllum) Orobanche 1 Colwell 02-54 Colwell 03-53 JEPS126149 O. parishii subsp. parishii 1 O. californica subsp. jepsonii Colwell 01-104 WTU351396 0·95 O. californica subsp. feudgei Colwell JEPS126157 O. californica subsp. feudgei Colwell 99-02 WTU344758 O. multiflora Collins 2031 MO6012144 O. multiflora Collins 2030 MO6012145 O. californica subsp. grayana Colwell 14-25c YM O. californica subsp. grayana Colwell 03-57 WTU O. californica + 1 O. californica subsp. condensa Taylor18565 JEPS100410 O. californica subsp. californica Colwell 97-CA-BLO WTU344755 O. ludoviciana O. californica subsp. californica Colwell 99-77 WTU344846 O. californica subsp. californica Colwell 02-49 WTU351400 complexes O. californica subsp. grandis Colwell 02-52A WTU351391 O. californica subsp. californica Colwell 02-48 WTU351388 O. corymbosa Colwell 14-26 JEPS126165 O. multiflora Collins 2024 MO5990497 1 O. sp. nov. Yatskyevich 85-215 ARIZ264197 O. ludoviciana Collins 2033 UC2046173 0·97 Long 2240 UC2046156 1 0·98 Vilagran 8616 SGO142749 0·99 Rosas 3327 CONC169912 O. chilensis 1 0.02 Garcia 3877 SGO154435 CS42888 O. riparia 1 Collins 1620 MO4903216 Holmgren 7066 NY Halse 897 ARIZ187291 O. arizonica

FIG. 2. Bayesian inference majority-rule consensus tree of 86 Orobanche populations inferred from three concatenated cpDNA regions (matK, rps2 and trnL-trnF). Tip labels consist of the taxon name (if not in- cluded in a sidebar), the collection number and the herbarium accession numbers if available. Posterior probabilities >09 are shown in bold for nodes with >70 % maximum likelihood bootstrap (BS) support and in italics if BS support is < 70 %. Some host associations are indicated in blue (Asteraceae) or green (other) to the genus or higher taxonomic level; for others, see Fig. 1. Informally named groups are given in purple. Outgroups are not shown. Schneider et al. — Evolution of New World Orobanche 1107

Wood s.n. MO6010393 subsp. waxy Henson 1499 WIS282510 Cochrane 255 WIS282508 (introns 9–11) 1 Astereae + Kucyniak 76 WIS282505 Rudbeckia uniflora Lomer 7268 UBC234660 Bouchard 86-94 MT26938 O. uniflora Anderson 2267 MIN 1 Lipkin 04-324 ALAv156078 subsp. 0·94 Fiely 91 WS285758 Antennaria Ahart 9846 CHSC82361 + Senecio 0·89 Ahart 7286 CHSC63258 occidentalis Oswald 5615 UC1609174 Apiodieae 1 1 Polster s.n. V93858 Ahart 10765 CHSC87344 Colwell 04-63 YM118104 Saxifragales

0·94 Colwell 96-WA-IC WTU344845 Lackschewitz 6619 WTU272369 1 Keith s.n. WIS 282501 Halse 908 ARIZ187532 Artemisia 0·99 Batten 78-279 ALAv85763 Bohrer 1684 ARIZ189922 Cox 188 CHSC12931 O. fasciculata Bell 159 SD136414 0·93 Banks 689 SD137701 Galium Colwell 07-53 YM Harvey s.n. JEPS Colwell 10-107 YM Howell 51680 CAS641193 Egger 1295 WTU Eriogonum, 1 1 Schneider 920 JEPS122893 Eriodictyon Colwell 02-09 WTU351387 Eriophyllum, Yatskievych 82-196 ARIZ236136 Phacelia Colwell 01-95 WTU351381 Colwell 99-CA-RCRS WTU344759 0·95 O. valida subsp. valida Colwell 99-73 WTU344847 O. valida subsp. howellii Colwell 02-04 WTU351356 O. tarapacana Ricardi 3326 CONC19273 O. tacnaensis Alfaro 3461 MO2637419 0·99 1 O. californica Schneider 293 JEPS O. californica Colwell 02-54 Orobanche 1 O. cooperi Colwell 02-06 WTU351385 sect. O. cooperi Colwell 01-01 WTU344743 Nothaphyllon O. multiflora Collins 2024 MO5990497 1 O. chilensis Rosas 3327 CONC169912 0·02 O. chilensis Long 2240 UC2046156 O. chilensis Garcia 3877 SGO154435 O. bulbosa Schneider 936 JEPS122909

FIG. 3. Bayesian inference majority-rule consensus tree of 47 Orobanche populations inferred from the waxy locus (introns 9–11). Tip labels consist of the taxon name (if not included in a sidebar), the collection number and the herbarium accession numbers if available. Posterior probabilities >08 are shown. All labelled nodes have maximum likelihood bootstrap scores 74 %. Host associations for sect. Gymnocaulis are indicated in blue (Asteraceae) or green (other) to the genus or higher taxonomic level; for sect. Nothaphyllon see Fig. 1 or Table S1. Outgroups are not shown. western hemisphere Orobanche is substantially richer than rec- Although some Orobanche taxa are specific to a single ognized by current taxonomy, perhaps because extensive reduc- host species, most parasitize several closely related species tion of structural characters in these parasites has limited the that are unique and sometimes phylogenetically distant from potential for morphological diagnosis of recently diverged evo- the hosts of their nearest relatives. In many ways, Orobanche lutionary lineages. Secondly, the host breadth of each evolu- spp. occupy an ecological middle-ground between species tionary lineage is narrower than previously assumed, although such as Epifagus virginiana (Orobanchaceae), which can some lineages with wide host ranges are still present (e.g. O. only grow on Fagus grandifolia, and true generalists such as fasciculata p.p.). Host specificity in plant parasites has been dodders (Cuscuta spp., Convolvulaceae) in which a single in- correlated to various life history and other host traits such as dividual may parasitize numerous distantly related hosts weediness or perenniality (Schneweeis, 2007). Host-switching (Press and Graves, 1995). Therefore, it is unlikely that host– has been cited as a driver of speciation of numerous parasites parasite co-speciation plays an appreciable role in driving di- across the tree of life (Ricklefs et al.,2004; de Vienne et al., versification in western hemisphere Orobanche in contrast to 2013), including other lineages of parasitic plants (Norton and some plant–animal, animal–animal or prokaryote–animal Carpenter, 1998; Norton and Lange, 1999; Bolin et al., 2011), host–parasite systems (de Vienne et al., 2013). Instead, we as well as within the genus Orobanche (Thorogood et al., argue that the more common mode – host-switching followed 2009). Our evidence strongly supports this hypothesis. The by physiological specialization and divergence – is dominant abundance of host-specific clades found here suggests that in this system. host-switching may be an even more important driver of evolu- Specialization and evolutionary divergence (cladogenesis) fol- tionary divergence in parasitic plants than previously lowing host-switching is an expected outcome given the complex recognized. challenges of host detection, host invasion and evasion or 1108 Schneider et al. — Evolution of New World Orobanche neutralization of host defences, which may occur pre- or post- a regional or even a local scale. This is particularly well pro- attachment. Pre-attachment host defences may include reduced nounced in sect. Gymnocaulis, discussed in detail below. germination stimulants (i.e. strigolactones, Cameron et al., 2006; Xie et al., 2010), increased germination inhibitory compounds (Fernandez-Aparicio et al., 2011), chemical inhibition of hausto- Cryptic diversity in section Gymnocaulis rial development (Pe´rez-de-Luque et al., 2005a, b)orstructural Cryptic lineages are found in both sections of New World fortifications to serve as a mechanical barrier to invasion. Orobanche [e.g. a polyphyletic O. parishii subsp. parishii Potential hosts can repel parasitic plants following attachment (Fig. 1)], but most extensively in O. sect. Gymnocaulis,in using a variety of mechanisms that disrupt the flow of nutrients which we identified over twice as many host-specific clades as or block vessel elements (Goldwasser et al., 1999, 2000; Pe´rez- exist commonly recognized taxa. Moreover, these clades are of- de-Luque et al., 2005a), initiate programmed cell death (Gurney ten subtended by long stem branches relative to clades that rep- et al., 2006), increase lignification and suberization of cell walls resent different recognized species in sect. Nothaphyllon. This (Labrousse et al., 2001; Pe´rez-de-Luque et al., 2008) or elicit disparity, which is robust to the gene region(s) used (Figs 1–3), chemical defence through increased peroxidases or the transfer may be due to more extensive reduction of morphological and of toxins from the host to the parasite (Gurney et al., 2003). thus diagnostic features in sect. Gymnocaulis,aswellasmore These multiple layers of incompatibility must be overcome for a limited systematic and taxonomic study of this section (Achey, successful invasion of the host, and provide the physiological ba- 1933; Watson, 1975)relativetosect.Nothaphyllon (Munz, sis for host specificity in parasitic Orobanchaceae (Yoder, 1997; 1930; Collins, 1973; Heckard, 1973; Heckard and Chuang, Yoshida and Shirasu, 2009; Thorogood and Hiscock, 2010). 1975; Collins and Yatskievich, 2015). Similar levels of cryptic Consequently, distantly related hosts with more divergent physi- diversity may be found in other holoparasitic lineages, particu- ologies probably require different invasion strategies. Various lar endoparasites such as Cytinus (Cytinaceae) that show even suites of host-specific traits may therefore represent different more extensive morphological reduction than Orobanche and a adaptive peaks for an Orobanche lineage. more intimate host–parasite relationship (De Vega et al., 2008). Dre`s and Mallet (2002) citeanumberofinsect–plantsystems Each clade of Orobanche sect. Gymnocaulis shows at least to show how the formation of host-specific races may eventually partial range overlap with its sister group, with generally in- lead to sympatric speciation of parasites through outbreeding de- creasing overlap with decreasing phylogenetic distance (Fig. 4). pression, even in the presence of gene flow. The generalist clade The clade of O. fasciculata parasitic on Galium is entirely in- of O. fasciculata shows poorly supported phylogenetic sub- cluded within the range of its sister group, which is a generalist structure and may provide the opportunity to explore this hy- clade parasitic on various eudicot hosts. The clade of O. fasci- pothesis in a plant–parasite system. Among the other host- culata parasitic on Artemisia grows coarsely sympatrically (i.e. specific clades of O. sect. Gymnocaulis, sympatric speciation sympatric at regional scales) with both subspecies of its sister following this model may already have occurred. The strong group, O. uniflora. These subspecies, O. subsp. uniflora and O. support for these clades by all three loci (nrDNA, cpDNA and subsp. occidentalis, once thought to be allopatric, are now waxy) suggest minimal, if any, continued gene flow among known to co-occur based on a recent floristic discovery in these lineages, even between geographically neighbouring popu- southern British Columbia and subsequent reinterpretation of lations. Isolation by host may also be reinforced by autogamy or historic herbarium records (A. C. Schneider, unpubl. data). apomixis, which is common in New World Orobanche species Most strikingly, the three closely related clades resolved within in contrast to more variable mating systems among species of O. uniflora subsp. occidentalis, which parasitize species in the Eurasian Orobanche and predominance of outcrossing among Asteraceae, Apiaceae, and Saxifragaceae plus Crassulaceae, re- other lineages of parasitic angiosperms (Musselman et al., 1982; spectively, share nearly entirely overlapping ranges at both Jones, 1989; Bellot and Renner, 2013). Autogamy has been coarse continental and local scales. For example, populations of identified as the predominant mating system in O. pinorum, all three clades can be found in Yosemite National Park and the with occasional outcrossing by bees (Ellis et al.,1999), is com- adjacent Sierra National Forest. mon among O. fasciculata parasitizing Artemisia (Reuter, 1986), and has been anecdotally reported in O. uniflora subsp. occidentalis and Orobanche bulbosa (K. L. Chambers 2952, Relationships in section Nothaphyllon OSC198410; Butterwick 5434 & Parfitt, ASU, JEPS; Schneider 1032, JEPS (Parfitt and Butterwick, 1981)). Some populations In sect. Nothaphyllon, we also find strong support for host- of Orobanche uniflora subsp. uniflora are obligatorily partheno- specific species, including the recently described O. arizonica, genic, while other populations show a ‘wholly different...repro- O. riparia and a clade currently recognized as O. californica ductive process’ (Jenson, 1951). As discussed previously, gene subsp. californica that parasitzes Eriophyllum stachaedifolium flow between different host races is expected to be detrimental on the central California coast, which is currently being de- if parent taxa are adapted to separate hosts, since a hybrid may scribed by the second author and George Yatskievych. Most be adapted to neither of them. other clades have distinct host associations, generally with pe- Geographic differentiation may play a subordinate role in lin- rennial Asteraceae, but usually not specific to the species level eage diversification, and may be restricted to cases where sister (Fig. 1). clades parasitize closely related hosts, such as between the sub- Most of the taxonomic diversity in O. sect. Nothaphyllon is species of O. valida, which both parasitize Garrya. Much more concentrated in a large clade supported by nrDNA and commonly, ranges are at least partially overlapping, and closely cpDNA, which is comprised of two sub-clades supported by related parasite lineages differing in their hosts can co-occur on nrDNA (Fig. 1) and morphological analysis (Collins 1973; Schneider et al. — Evolution of New World Orobanche 1109

180° 160° 140° 120° 100° 60° 50° 40°

60°

50°

40°

Orobanche clades (by host) O. fasciculata Artemisia Eriogonum, Eriophyllum or Hydrophyllaceae Galium O. uniflora subsp. occidentalis Apioidiae 30° Asteraceae Saxifragales O. uniflora subsp. uniflora Asteraceae

FIG. 4. Range map of host-specific clades of Orobanche sect. Gymnocaulis. Coloured circles represent individuals sampled in the phylogeny (Figs 1–3). Coloured lines show the approximate range of each clade. Further study is needed to determine the range of each of the three host-specific lineages of O. uniflora subsp. occi- dentalis, which in this figure are treated as one unit. Range maps should be considered tentative, particularly in northern Canada and west-central USA, pending a thorough taxonomic and phytogeographical study.

Heckard, 1973). The first sub-clade corresponds to the O. cali- and host affinities (Artemisia, especially A. tridentata), O. cor- fornica complex, which includes O. californica and its sub- ymbosa may represent a hybrid between O. californica and O. species, O. parishii, O. corymbosa and O. vallicola. The ludoviciana, both of which in part also parasitize Artemisia. In second clade represents the O. ludoviciana complex, which certain other cases, polyploidy may be a driver of speciation. includes O. ludoviciana (except for populations parasitizing Heckard and Chuang (1975) published detailed chromosome Artemisia), O. multiflora, the recently described O. arizonica, counts for most species of North American Orobanche.Theoc- O. riparia, the disjunct South American species O. chilensis toploid O. parishii subsp. brachyloba forms a clade nested and O. tacnaensis, and a collection from Hidalgo, Mexico that within O. parishii subsp. parishii (Fig. 1), its likely tetraploid does not fit the description of any described taxon progenitor (ploidy assignment based on a chromosome base (Yatskievych 85-215; ARIZ). number of x ¼ 12; for a more detailed discussion, see Several earlier diverging lineages native to western North Schneeweiss et al., 2004b), or, if an allopolyploid, one of two America are also strongly supported as monophyletic by both parental lineages. Octoploid lineages have also been reported in nrDNA and cpDNA, including O. valida, O. bulbosa and the re- O. cooperi and O. corymbosa subsp. corymbosa (but not O. cently revised O. cooperi þ O. dugesii complex (Figs 1 and 2; ludoviciana). A full discussion of the systematics and taxonomy Collins and Yatskievych, 2015). However, relationships among of these and other individual species is needed, but is beyond these lineages are unclear: O. bulbosa is resolved either as sister the scope of this paper. to the rest of the section (nrDNA, Fig. 1) or as a grade with O. bulbosa diverging earliest (cpDNA, Fig. 2). The conflict among gene partitions is in most cases probably explained by incom- Repeated dispersal to South America plete lineage sorting, but in other cases may be a result of retic- ulate evolution. For example, based on its phylogenetic Based on the nrDNA phylogeny, we find support for the placement in two separate clades (Fig. 1), and morphological longstanding hypothesis that O. chilensis is closely related to O. 1110 Schneider et al. — Evolution of New World Orobanche ludoviciana and O. multiflora (Beck, 1890), thereby contribu- Beardsley PM, Olmstead RG. 2002. Redefining Phrymaceae: the placement of ting to the broadly recognized pattern of amphitropical disjunc- Mimulus, tribe Mimuleae, and Phryma. American Journal of Botany 87: 1093–1102. tion between the Great Plains of North America and northern Beck G. 1890. Monographe der Gattung Orobanche. Biblitoheca Botanica 19. Chile/southern Peru (Wen and Ickert-Bond, 2009). Of the two Cassel: Theoder Fischer. other sampled Orobanche species from South America, O. tac- Bellot S, Renner SS. 2013. Pollination and mating systems of Apodanthaceae naensis, was resolved with O. chilensis, but the two samples of and the distribution of reproductive traits in parasitic angiosperms. Phil. formed a separate, earlier diverging lineage American Journal of Botany 100: 1083–1094. O. tarapacana Bolin JF, Maass E, Mussellman LJ. 2011. A new species of Hydnora resulting from north to south dispersal. Phylogenetic placement (Hydnoraceae) from South Africa. Systematic Botany 36: 255–260. of O. tarapacana is uncertain due to conflict between the Cameron DC, Coats AM, Seel WE. 2006. Differential resistance among host nrDNA and cpDNA trees; O. tarapacana is sister to either the and non-host species underlies the variable success of the hemiparasitic O. ludoviciana complex, the O. cooperi complex or perhaps a plant Rhinanthus minor. Annals of Botany 98: 1289–1299. Collins LT. 1973. Systematics of Orobanche sect. Myzorrhiza. PhD thesis, hybrid between the two (Figs 1 and 2). University of Wisconsin–Milwaukee. Collins, L.T. and Yatskievych G. 2015. Orobanche arizonica sp. nov. and nomenclatural changes in Orobanche cooperi (Orobanchaceae). Conclusions and future directions Phytoneuron 2015-48: 1–19. Darriba D, Taboada GL, Doallo R, Posada D. 2012. jModelTest 2: more mod- Parasitic Orobanchaceae is becoming a model system for un- els, new heuristics and parallel computer. Nature Methods 9:772. derstanding plant parasitism at various levels of biological or- De Vega C, Berjano R, Arista M, Ortiz PL, Talavera S, Stuessy TF. 2008. ganization and scale (Joel et al.,2013; McNeal et al.,2013; Genetic races associated with the genera and sections of host species in the holoparasitic plant Cytinus (Cytinaceae) in the Western Mediterranean Wicke et al., 2013; Yang et al., 2015). Our results emphasize basin. New Phytologist 178: 875–887. the importance of host specificity and host-switching as a driver dePamphilis CW, Young ND, Wolfe AD. 1997. Evolution of plastid gene rps2 of evolutionary divergence in obligate plant parasites. We find in a lineage of hemiparasitic and holoparasitic plants: many losses of photo- evidence for twice as many host-specific lineages in O. sect. synthesis and complex patterns of rate variation. Proceedings of the Gymnocaulis as recognized taxa, and denser sampling in other National Academy of Sciences, USA 94: 7367–7372. Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities clades such as O. sect. Nothaphyllon is likely to uncover more. of fresh leaf tissue. Phytochemical Bulletin of the Botanical Society of This robust understanding of fine-scale evolutionary relation- America 19: 11–15. ships provides the necessary phylogenetic framework to de- Dre`s M, Mallet J. 2002. Host races in plant-feeding insects and their importance velop a more natural classification for this group, and in sympatric speciation. Proceedings of the Royal Society B: Biological Sciences 357: 471–492. understand genetic, ecological, functional and life-history con- Ellis MW, Taylor RJ, Harrod RJ. 1999. The reproductive biology and host sequences of host–parasite associations more broadly. specificity of Orobanche pinorum Geyer (Orobanchaceae). Madrono~ 46: 7–12. Fernandez-Aparicio M, Yoneyama K, Rubiales D. 2011. The role of strigolac- SUPPLEMENTARY DATA tones in host specificity of Orobanche and Phelipanche seed germination. Seed Science Research 21:55–61. Supplementary data are available online at www.aob.oxfordjour Goldwasser Y, Hershenhorn J, Plakhine D, Kleifeld Y, Rubin B. 1999. nals.org and consist of Table S1: vouchers, host associations Biochemical factors involved in vetch resistance to Orobanche aegyptiaca. and GenBank accession numbers for 163 Orobanche popula- Physiolgical and Molecular Plant Pathology 54: 87–96. tions and outgroup taxa. Goldwasser Y, Plakhine D, Kleifeld Y, Zamski E, Rubin B. 2000. The differ- ential susceptibility of vetch (Vicia spp.) to Orobanche aegyptiaca: anatom- ical studies. Annals of Botany 85:257–262. ACKNOWLEDGEMENTS Gurney AL, Grimanelli D, Kanampiu F, Hoisington D, Scholes JD, Press MC. 2003. Novel sources of resistance to Striga hermonthica in Tripsacum We thank the curatorial staff at the following herbaria for pro- dactyloides, a wild relative of maize. New Phytologist 160: 557–568. viding access to specimens and/or providing tissue for DNA Gurney AL, Slate J, Press MC, Scholes JD. 2006. A novel form of resistance in rice to the angiosperm parasite Striga hermonthica. New Phytologist 169: sequencing. ACAD, ALA, ARIZ, CAS, CHSC, CONC, JEPS, 199–208. MO, MT, NY, OSC, RSA, SD, SGO, UBC, UC, V, WIS, WS Heckard LR. 1973. A taxonomic re-interpretation of the Orobanche californica and WTU. We also thank Rafael Casanova and Derek complex. Madrono~ 22: 41–70. Anderson for assistance in procuring collections, Dave Tank Heckard LR, Chuang TI. 1975. Chromosome numbers and polyploidy in Orobanche (Orobanchaceae). Brittonia 27: 179–186. for advice on using the waxy locus, Harold Chun and Amanda Jenson HW. 1951. The normal and parthenogenic forms of Orobanche uniflora Seng for laboratory assistance, Turner Collins and George in the eastern United States. Cellule 54: 135–142. Yatskievych for fruitful discussion, and Jakub Tesitel and one Joel DM, Gressel J, Musselman LJ, eds. 2013. Parasitic Orobanchaceae. anonymous reviewer for their comments on an earlier version Heidelberg: Springer. of the manuscript. Funding was provided by grants to A.C.S. Jones M. 1989. Studies on the pollination of Orobanche species in the British Isles. Progress in Orobanche Research Proceedings. Tubingen, from the Lawrence R. Heckard Endowment Fund of the 6–17. Jepson Herbarium, the California Native Plant Society Kellogg VL. 1913. Distribution and species-forming of ecto-parasites. American (CNPS) Doc Burr Grant, and the CNPS Santa Clara Valley Naturalist 47: 129–158. Chapter. A.C.S. was also supported by an internal Berkeley Labrousse P, Arnaud MC, Serieys H, Berville´ A, Thalouarn P. 2001. Several Graduate Fellowship. mechanisms are involved in resistance of Helianthus to Orobanche cunana Wallr. Annals of Botany 88: 859–868. McNeal JR, Bennett JR, Wolfe AD, Mathews S. 2013. Phylogeny and origins LITERATURE CITED of holoparasitism in Orobanchaceae. American Journal of Botany 100: 971–983. Achey DM. 1933. A revision of the section Gymnocaulis of the genus MillerMA,PfeifferW,SchwartzT.2010.Creating the CIPRES Science Orobanche. Bulletin of the Torrey Botanical Club 60: 441–451. Gateway for inference of large phylogenetic trees. In: Proceedings of the Schneider et al. — Evolution of New World Orobanche 1111

Gateway Computing Environments Workshop (GCE).14November2010, Shaw J, Lickey EB, Beck JT, et al.2005.The tortoise and the hare II: relative New Orleans, Louisiana, USA. utility of 21 noncoding chloroplast DNA sequences for phylogenetic analy- Munz PA. 1930. The North American species of Orobanche section Myzorrhiza. sis. American Journal of Botany 92: 142–166. Bulletin of the Torrey Botanical Club 57: 611–624. Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and Musselman LJ, Parker C, Dixon N. 1982. Notes on autogamy and flower struc- post-analysis of large phylogeneies. Bioinformatics 30: 1312–1313. ture in agronomically important species of Striga (Scrophulariaceae) and Steele KP, Vilgalys R. 1994. Phylogenetic analyses of Polemoniaceae using nu- Orobanche (Orobanchaceae). Beitrage€ zur Biologie der Pflanzen 5: cleotide sequences of the plastid gene matK. Systematic Botany 19: 329–343 126–142. Norton DA, Carpenter MA. 1998. Mistletoes as parasites: host specificity and Taberlet P, Gielly L, Pautou G, Bouvet J. 1991. Universal primers for amplifi- speciation. Trends in Ecology and Evolution 13: 101–104. cation of three non-coding regions of chloroplast DNA. Plant Molecular Norton DA, De Lange PJ. 1999. Host specificity in parasitic mistletoes Biology 17: 1105–1109. (Loranthaceae) in New Zealand. Functional Ecology 13: 552–559. Tank DC, Olmstead, RC. 2008. From annuals to perennials: phylogeny of subtribe Parfitt BD, Butterwick M. 1981. Noteworthy collections: Orobanche uniflora Castillejinae (Orobanchaceae). American Journal of Botany 95: 608–625. L. subsp. occidentalis. Madrono~ 28:37–38. Tank DC, Olmstead, RG. 2009. The evolutionary origin of a second radiation of Park J-M, Manen J-F, Colwell AE, Schneeweiss GM. 2008. A plastid gene annual Castilleja (Orobanchaceae) speceis in South America: the role of long phylogeny of the non-photosynthetic parasitic Orobanche (Orobanchaceae) distance dispersal and allopolyploidy. American Journal of Botany 96: and related genera. Journal of Plant Research 121: 365–376. 1907–1921. Pe´rez-de-Luque A, Jorrın J, Cubero JI, Rubiales D. 2005a. Orobanche cren- Thorogood CJ, Hiscock SJ. 2010. Compatibility interactions at the cellular ata resistance and avoidance in pea (Pisum spp.) operate at different devel- level provide the basis for host specificity in the parasitic plant Orobanche. opmental stages of the parasite. Weed Research 45: 379–387. New Phytologist 186: 571–575. Pe´rez-de-Luque A, Rubiales D, Cubero JI, et al.2005b. Interaction between Thorogood CJ, Rumsey FJ, Harris SA, Hiscock J. 2008. Host-driven diver- Orobanche crenata and its host legumes: unsuccessful haustorial penetra- gence in the parasitic plant O. minor. Molecular Ecology 17: 4289–4303. tion and necrosis of the developing parasite. Annals of Botany 95: 935–942. Thorogood CJ, Rumsey FJ, Hiscock SJ. 2009. Host-specific races in the holo- Pe´rez-de-Luque A, Moreno MT, Rubiales D. 2008. Host plant resitance parasitic angiosperm Orobanche minor: implications for speciation in para- against broomrapes (Orobanche spp.): defence reactions and mechanisms of sitic plants. Annals of Botany 103: 1005–1014. resistance. Annals of Applied Biology 152: 131–141. TimkoMP,HuangK,LisKE.2012.Host resistance and parasite virulence in Poulin R, Morand S. 2000. The diversity of parasites. Quarterly Review of Striga–host plant interactions: a shifting balance of power. Weed Science Biology 75: 277–293. 60: 307–315. Press M, Graves J, eds. 1995. Parasitic plants. London: Chapman & Hall. Uhlich H, Pusch J, Barthel KJ. 1995. Die Sommerwurzarten Europas. Refre´gier G, Le Gac M, Jabbour F, et al. 2008. Cophylogeny of the anther Magdeburg: Westarp Wissenschaften. smut fungi and their caryophyllaceous hosts: prevalence of host shifts and de Vienne DM, Refre´gier G, Lopez-Villavicencio M, Tellier A, Hood ME, importance of delimiting parasite species. BMC Evolutionary Biology 8: Giraud T. 2013. Cospeciation vs host-shift speciation: methods for testing, 100. evidence from natural associations and relation to coevolution.New Reuter BC. 1986. The habitat, reproductive ecology, and host relations of Phytologist 198: 347–385. Orobanche fasciculata Nutt. (Orobanchaceae) in Wisconsin. Bulletin of the Watson KC. 1975. Systematics of Orobanche section Gymnocaulis Torrey Botanical Club 113: 110–117. (Orobanchaceae). MS thesis, California State University Chico. Ricklefs RE, Fallon SM, Bermingham E. 2004. Evolutionary relationships, Wen J, Ickert-Bond SM. 2009. Evolution of the Madrean–Tethyan disjunctions cospeciation, and host switching in avian malaria parasites. Systematic and the North American amphitropical disjunctions in plants. Journal of Biology 53: 111–119. Systematics and Evolution 47: 331–348. Ronquist FM, Teslenko P, van der Mark DL, et al.2012.MrBayes 3.2: effi- Westwood JH, Yoder JI, Timko MP, dePamphilis CW. 2010. The evolution cient Bayesian phylogenetic inference and model choice across a large of parasitism in plants. Trends in Plant Science 15: 227–235. model space. Systematic Biology 61: 539–542. Wicke S, Mu¨ller KF, de Pamphilis CW, et al.2013.Mechanisms of functional Schneeweiss GM. 2007. Correlated evolution of life history and host range in and physical genome reduction in photosynthetic and nonphotosynthetic the nonphotosynthetic parasitic flowering plants Orobanche and parasitic plants in the broomrape family. The Plant Cell 25: 3711–3725. Phelipanche (Orobanchaceae). Journal of Evolutionary Biology 20: Xie X, Yoneyama K, Kisugi T. 2010. The strigolactone story. Annual Review of 471–478. Phytopathology 48:93–117. Schneeweiss, GM. 2013. Phylogenetic relationships and evolutionary trends in Yang Z, Wafula EK, Honaas LA, et al.2015. Comparative transcriptome anal- Orobanchaceae. In: Joel DM, Gressel J, Mussleman LJ, eds. Parasitic yses reveal core parasitism genes and suggest gene duplication and repur- Orobanchaceae. Berlin: Springer, 243–265. posing as sources of structural novelty. Molecular Biology and Evolution Schneeweiss GM, Colwell AEL, Park J-M, Jang C-G, Stuessy TF. 2004a. 32: 767–790. Phylogeny of holoparasistic Orobanche (Orobanchaceae) inferred from nu- Yoder J. 1997. A species-specific recognition system directs haustorium devel- clear ITS sequences. Molecular Phylogenetics and Evolution 30: 465–478. opment in the parasitic plant Tryphysaria (Scrophulariaceae). Planta 202: Schneeweiss, GM, Palomeque T, Colwell AE, Weiss-Schneeweiss H. 2004b. 407–413. Chromosome numbers and karyotype evolution in holoparasitic Orobanche Yoshida S, Shirasu K. 2009. Multiple layers of incompatibility to the parasitic (Orobanchaceae) and related genera. American Journal of Botany 91: witchweed, Striga hermonthica. New Phytologist 183: 180–189. 439–448.