See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/267929008

Molecular phylogenetics of tribe Eudemeae (Brassicaceae) and implications

for its morphology and distribution

Article  in  Molecular Phylogenetics and Evolution · January 2015

DOI: 10.1016/j.ympev.2014.09.030

CITATIONS READS

9 803

4 authors, including:

Diego L. Salariato F. O. Zuloaga

National Scientific and Technical Research Council Instituto de Botánica Darwinion
87 PUBLICATIONS   290 CITATIONS    374 PUBLICATIONS   4,654 CITATIONS   

SEE PROFILE SEE PROFILE

Asunción Cano Echeverría

National University of San Marcos
120 PUBLICATIONS   853 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Phylogeny of Panicum View project

Global Caryophyllales Synthesis View project

All content following this page was uploaded by Diego L. Salariato on 11 September 2018.

The user has requested enhancement of the downloaded file.

 Molecular Phylogenetics and Evolution 82 (2015) 43–59

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

Molecular phylogenetics of tribe Eudemeae (Brassicaceae)

and implications for its morphology and distribution
Diego L. Salariato a,⇑, Fernando O. Zuloaga a, Asunción Cano b,c, Ihsan A. Al-Shehbaz d
a
Instituto de Botánica Darwinion (CONICET – ANCEFN), Labardén 200, Casilla de Correo 22, B1642HYD San Isidro, Buenos Aires, Argentina
b
Museo de Historia Natural, Universidad Nacional Mayor de San Marcos (UNMSM), Av. Arenales 1256, Lima 11, Peru
c
Instituto de Investigación de Ciencias Biológicas, Facultad de Ciencias Biológicas (UNMSM), Av. Venezuela s/n, Lima 1, Peru
d
Missouri Botanical Garden, P.O. Box 299, St. Louis, MO 63166-0299, USA

a r t i c l e i n f o a b s t r a c t

Article history: Tribe Eudemeae comprises a morphologically heterogeneous group of genera distributed along the Andes
Received 29 April 2014 of South America from Colombia southward into southern Chile and Argentina. The tribe currently
Revised 21 August 2014 includes seven genera: Aschersoniodoxa, Brayopsis, Dactylocardamum, Delpinophytum, Eudema, Onuris,
Accepted 19 September 2014
and Xerodraba, and exhibits a wide morphological diversification in growth habit, inflorescences, and
Available online 17 October 2014
fruits. However, little is known about the phylogenetic relationships and evolution of the tribe. We pres-
ent here a molecular phylogeny of representative sampling of all genera, utilizing sequence data from the
Keywords:
nuclear ribosomal ITS region and chloroplast regions trnL-F, trnH-psbA, and rps16. Additionally, climatic
Andes
Brassicaceae
niches of the tribe and its main lineages, along with the evolution of diagnostic morphological characters,
Eudemeae were studied. All analyses confirmed the monophyly of Eudemeae, with the exception of Delpinophytum
Morphology that was included with genera of the lineage I of Brassicaceae. Eudemeae is divided into two main lin-
South America eages differentiated by their geographical distribution and climatic niche: the primarily north-central
Phylogeny Andean lineage included Aschersoniodoxa, Brayopsis, Dactylocardamum, and Eudema, and the Patagonian
and southern Andean lineage included Onuris and Xerodraba. Finally, ancestral-state reconstructions in
the tribe generally reveal multiple and independent gains or losses of diagnostic morphological charac-
ters, such as growth form, inflorescence reduction, and fruit type. Relevant taxonomic implications stem-
ming from the results are also discussed.
Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction
The Brassicaceae (Cruciferae), a family of approximately 320
genera and 3660 species (Al-Shehbaz, 2012a), is most abundant
in the temperate regions of the northern hemisphere, especially
the Irano-Turanian region, Mediterranean area, and western North
America (Appel and Al-Shehbaz, 2003). The family is distributed
worldwide (except Antarctica) and is well represented in the
Southern Hemisphere, with 374 native species in South America
(Al-Shehbaz, unpubl.). Based on the extensive molecular phyloge-
nies published over the past eight years (e.g., Al-Shehbaz and
Warwick, 2007; Alexander et al., 2010, 2013; Bailey et al., 2006,
2007; Beilstein et al., 2006, 2008; Couvreur et al., 2010; Franzke
Abbreviations: BI, Bayesian inference; BS, bootstrap support; JK, jackknife
support; ML, maximum likelihood; MP, maximum parsimony; NCA, north-central
Andes; SA, southern Andes.
⇑ Corresponding author. Fax: +54 (11) 4747 4748.
E-mail address: dsalariato@darwin.edu.ar (D.L. Salariato).
et al., 2009, 2011; German et al., 2009, 2011; Goodson et al.,
2011; Hall et al., 2011; Jordon-Thaden et al., 2010; Karl and
Koch, 2013; Khosravi et al., 2007, 2008, 2009a,b; Koch, 2012;
Koch et al., 2007, 2008, 2010; Rešetnik et al., 2013; Salariato
et al., 2013a; Urban and Bailey, 2013; Warwick and Hall, 2009;
Warwick et al., 2007, 2008, 2009, 2010, 2011; Yue et al., 2009;
Zhao et al., 2010), Al-Shehbaz (2012a) recognized 49 tribes in the
Brassicaceae. Of these, the Cremolobeae R.Br., Schizopetaleae
R.Br. ex Barnéoud, and Eudemeae Al-Shehbaz et al. are endemic
to South America. Other tribes (e.g., Cardamineae Dumort., Desc-
urainieae Al-Shehbaz et al., Lepidieae DC., Thelypodieae Prantl)
are well represented in South America and include a high number
of endemic taxa, though they are well diversified elsewhere (Al-
Shehbaz, 2008, 2012b). The Cremolobeae, Eudemeae, and Schizo-
petaleae are distributed along the Andes range and were shown
to be monophyletic within linage II of the family (Salariato et al.,
2013b; Warwick et al., 2010). For a comprehensive phylogenetic
tree including all tribes and lineages see the ‘‘BrassiBase’’ http://
brassibase.cos.uni-heidelberg.de/ (Kiefer et al., 2014; Koch et al.,

http://dx.doi.org/10.1016/j.ympev.2014.09.030
1055-7903/Ó 2014 Elsevier Inc. All rights reserved.
44 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59

Fig. 1. Distribution map of tribe Eudemeae and its genera. (A) Tribe Eudemeae; (B) Aschersoniodoxa; (C) Brayopsis; (D) Dactylocardamum; (E) Delpinophytum; (F) Eudema; (G)
Onuris; (H) Xerodraba. Dots represent herbarium records for the different genera. Elevation is depicted in gray scale from sea level (white) to over 4000 m above sea level
(black).

2012). The Eudemeae includes seven genera: Aschersoniodoxa Gilg
& Muschl. (4 spp.), Brayopsis Gilg & Muschl. (7 spp.), Dactylocarda-
mum Al-Shehbaz (1 sp.), Delpinophytum Speg. (1 sp.), Eudema Bon-
pl. (4 spp.), Onuris Phil. (6 spp.), and Xerodraba Skottsb. (7 spp.). By
contrast, Cremolobeae includes Cremolobus DC. (7 spp.), Menonvil-
lea DC. (24 spp.), and Aimara Salariato & Al-Shehbaz (1 sp.),
whereas Schizopetaleae comprises Mathewsia Hook. & Arn. (6
spp.), and Schizopetalon Sims (10 spp.). Although comprehensive
molecular phylogenies of the last two tribes have recently been
published (Salariato et al., 2013b; Toro-Nuñez et al., 2013), little
is known about the phylogenetic relationships and evolution of
Eudemeae.
 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59 45

Fig. 2. Habit of Eudemeae species. (A–E) Rosette habit. (A) Aschersoniodoxa cachensis ; (B) Aschersoniodoxa peruviana; (C) Brayopsis monimocalyx; (D) Eudema nubigena; (E)
Onuris spegazziniana. (F–N) Cushion habit. (F–H) Dactylocardamum imbricatifolium; (I–K) Delpinophytum patagonicum; (L–N) Xerodraba pectinata. Photographs by: A, Trinidad
H.; B, Navarro E.; C, Zanotti C.; D, Ulloa C.; F–G, Cano A.; I–N, Zuloaga. F.

Tribe Eudemeae was established on the basis of ITS phylogeny
that showed it as a well-supported clade including Brayopsis
colombiana Al-Shehbaz, Eudema nubigena Bonpl., and Xerodraba
pectinata Skottsb. (Warwick et al., 2010). Based mainly on mor-
phology, Al-Shehbaz (2012a,b) and Al-Shehbaz et al. (2012)
included in this tribe the remaining four genera listed above.
The Eudemeae is a very interesting tribe in several ways. It is
distributed along the Andes from Colombia in the north southward
into Patagonian Chile and Argentina (Fig. 1). Its species grow
mainly at high altitudes (to 5300 m) and high southern latitudes
(to 55°S), and many of its species are poorly represented in the
major herbaria of the world. The tribe exhibits a wide variation
in vegetative and reproductive characters that are otherwise rare
in the family. These include the formation of dense cushions with
imbricate leaves that characterize all species of Dactylocardamum,
Delpinophytum, and Xerodraba (Al-Shehbaz, 1989a, 2012b)
(Fig. 2). The tribe exhibits the basic inflorescence structure in fam-
ily of an indeterminate raceme (Benlloch et al., 2007; Bradley et al.,
1997; Ratcliffe et al., 1999; Yoon and Baum, 2004), but it is highly
variable in the presence vs. absence of bracts (e.g., Onuris vs. Asch-
ersoniodoxa; Fig. 3), or the production of single vs. several terminal
rosette flowers (e.g., Dactylocardamum and Xerdraba vs. Brayopsis).
Fruits in Eudemeae also exhibit substantial diversity in size and
shape (silicles or siliques), compression (latiseptate, terete, or
angustiseptate), and opening (indehiscent or acropetally or basip-
etally dehiscent). As a result of its tremendous morphological var-
46 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59
Fig. 3. Diagrams of inflorescences present in Eudemeae. (A), Ebracteate inflores-
cence flowering; (B) Bracteate inflorescence flowering; (C) Ebracteate rosette
flowering; (D) Bracteate rosette flowering; (E) solitary pseudoterminal flower.
iation, the Eudemeae do not have robust morphological synapo-
morphies. The few characters shared by species of this tribe
include the pulvinate habit without cauline leaves, presence of
caudices, and incumbent cotyledons. However, these characters
are also present in many other South American Brassicaceae of
other tribes (Appel and Al-Shehbaz, 2003; Al-Shehbaz, 2012b).
Although all genera of Eudemeae were studied taxonomically
(Al-Shehbaz, 1989a,b, 1990a,b, 2008, 2012b; Al-Shehbaz and
Cano, 2011; Al-Shehbaz et al., 2012; Boelcke, 1984; Boelcke and
Romanczuk, 1984), the morphological characters used to support
them (e.g., habit, fruit type) are subject to considerable conver-
gence in the rest of the family (Appel and Al-Shehbaz, 2003;
Bowman, 2006; Franzke et al., 2011; Hall et al., 2011) and can
therefore be taxonomically unreliable (Al-Shehbaz et al., 2006).
Aims of the present study on the Eudemeae are to: (1) assess
the monophyly of the tribe, (2) establish the phylogenetic relation-
ships among its genera, and (3) infer geographical distribution and
morphological variation observed on these phylogenies. To achieve
that, we generated comprehensive molecular phylogenies of the
tribe using nuclear (ITS) and plastid (trnL-F, trnH-psbA, and rps16)
sequences, and a representative sampling of all genera.
2. Materials and methods
2.1. Taxon sampling
Thirty-nine accessions representing ca. 75% of taxa of all genera
of the tribe Eudemeae [Aschersoniodoxa (3 spp. out of 4 total), Bray-
opsis (5/7), Dactylocardamum (1/1), Delpinophytum (1/1), Eudema (3
spp. and 2 subsp./4), Onuris (6/6), and Xerodraba (3/7)] were sam-
pled from all major geographical areas and covering all morpholog-
ical variation. The taxonomic circumscription of Eudemeae and its
genera follows Al-Shehbaz (2012a,b). Phylogenetic studies were
initially conducted to determine the monophyly of Eudemeae
within the Brassicaceae, and later to establish the phylogenetic
relationships within the tribe. For analyses at the family level,
143 (ITS) and 122 (trnL-F) sequences were used as outgroup repre-
senting 45 and 34 tribes, respectively. Based on these family-wide
analyses, only sequences of Smelowskia C.A. Mey., Lepidium L.,
Schizopetalon, and Menonvillea were included as outgroups at the
tribal-level analyses using ITS, trnL-F, trnH-psbA, and rps16
sequences. All DNA sequences of the outgroups were downloaded
from GenBank (https://www.ncbi.nlm.nih.gov/genbank/) (Gen-
Bank accession numbers are provided in the Appendix A of the sup-
plementary material).
2.2. Extraction, amplification and DNA sequencing
Total DNA was isolated from leaves (collected in the field and
dried in silica gel) using a modified (CTAB) protocol by Doyle and
Doyle (1987), or from herbarium material using a DNeasy Plant
Mini kit (Qiagen, Hilden, Germany). Sequence data from the
nuclear ribosomal ITS region and chloroplast regions trnL-F,
trnH-psbA, and rps16 were obtained because they were shown
to have phylogenetic utility for studies at lower taxonomic levels
(Shaw et al., 2005; Small et al., 1998) and useful in evaluating
relationships among Brassicaceae, particularly at the generic level
(e.g., Alexander et al., 2010; Bailey et al., 2006; Carlsen et al.,
2010; German et al., 2011; Goodson et al., 2011; Koch and
Mummenhoff, 2001; Moazzeni et al., 2010; O’Kane and Al-
Shehbaz, 2003; Salariato et al., 2013b; Toro-Nuñez et al., 2013;
Warwick et al., 2002, 2006, 2007, 2008, 2009, 2010, 2011). The
nuclear ribosomal ITS region (ITS1-5.8S-ITS2) was amplified by
PCR in one or two fragments using the ITS2, ITS3, ITS4 and ITS5
primers of Baldwin (1992); the chloroplast trnL-F region (trnL
intron/trnL-F spacer) was amplified in one or two fragments using
primers C, D, and E of Taberlet et al. (1991) and Fdw (Salariato
et al., 2013b). Sequences for trnH-psbA spacer and rps16 intron
were amplified in one fragment using primers trnH(GUG)/psbA
(Hamilton, 1999) and rps16F/rps16R (Shaw et al., 2005), respec-
tively. PCR reactions were performed in 25 lL final volumes with
50–100 ng of template DNA, 0.2 lM of each primer, 25 lM dNTP,
5 mM MgCl2, 1  buffer and 1.5 units of Taq polymerase provided
by Invitrogen Life Technologies. PCR amplifications were set at
the following conditions for most species: (ITS) a first period of
denaturation at 94 °C for 5 min, followed by 35 cycles of denatur-
ation at 94 °C for 30 s, annealing at 50 °C for 60 s, and extension
at 72 °C for 90 s, with a final extension at 72 °C for 7 min; (trnL-F)
a first period of denaturation at 94 °C for 3 min, followed by 35
cycles of denaturation at 94 °C for 30 s, annealing at 48 °C for
60 s, and extension at 72 °C for 90 s, with a final extension at
72 °C for 10 min; (trnH-psbA) and (rps16) a first period of dena-
turation at 94 °C for 3 min, followed by 35 cycles of denaturation
at 94 °C for 30 s, annealing at 55 °C for 60 s, and extension at
72 °C for 90 s, with a final extension at 72 °C for 10 min. In addi-
tion, a variety of PCR additives and enhancing agents (e.g., bovine
serum albumin, dimethyl sulfoxide, formamide) were used to
 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59 47

Table 1
Sequences generated in this study: Species, specimen voucher, and Genbank accession numbers. For sequences download from the Genbank see Appendix A.

Species Voucher Country ITS trnL-F trnH-psbA rps16

Aschersoniodoxa cachensis (Speg.) Al-Shehbaz Smith 14394 (LPB) Bolivia: La Paz KM376250 KM376288 KM376360 KM376325
Aschersoniodoxa cachensis (Speg.) Al-Shehbaz Menhofer 1460 (LPB) Bolivia: La Paz KM376251 KM376289 KM376361 KM376326
Aschersoniodoxa mandoniana (Wedd.) Gilg & Muschl. Beck 14033 (LPB) Bolivia: La Paz KM376252 KM376290 KM376362 KM376327
Aschersoniodoxa peruviana Al-Shehbaz, Navarro & A. Navarro 771 (USM) Peru: Lima KM376253 KM376291 KM376363 KM376328
Cano
Brayopsis alpaminae Gilg & Muschl. Smith 10288 (MO) Peru: Áncash KM376242 – – –
Brayopsis calycina (Desv.) Gilg & Muschl. Cano 20104 (MO) Peru: Junín KM376246 KM376284 – –
Brayopsis calycina (Desv.) Gilg & Muschl. Fuentes 12433 (LPB) Bolivia: La Paz KM376247 KM376285 KM376357 KM376322
Brayopsis calycina (Desv.) Gilg & Muschl. Zuloaga 14302 (SI) Argentina: Salta KM376248 KM376286 KM376358 KM376323
Brayopsis calycina (Desv.) Gilg & Muschl. Zuloaga 14327 (SI) Argentina: Jujuy KM376249 KM376287 KM376359 KM376324
Brayopsis colombiana Al-Shehbaz Sklenar 2450 (MO) Ecuador KM376244 – – –
Brayopsis diapensioides (Wedd.) Gilg & Muschl. Fuentes 14089 (MO) Bolivia: La Paz KM376243 KM376282 KM376356 KM376320
Brayopsis gamosepala Al-Shehbaz Krach 7640 (MO) Bolivia: La Paz KM376245 KM376283 – KM376321
Brayopsis monimocalyx O.E. Schulz Salomon 172 (SI) Argentina: La Rioja KM376240 KM376280 KM376354 KM376318
Brayopsis monimocalyx O.E. Schulz Zuloaga 14304 (SI) Argentina: Salta KM376241 KM376281 KM376355 KM376319
Dactylocardamum imbricatifolium Al-Shehbaz Cano 21546 (USM) Peru: Áncash KM376257 KM376295 KM376367 KM376331
Delpinophytum patagonicum (Speg.) Speg. Zuloaga 14115 (SI) Argentina: Santa Curz KM376225 KM376265 KM376339 KM376303
Delpinophytum patagonicum (Speg.) Speg. Zuloaga 14654 (SI) Argentina: Santa Cruz KM376258 KM376296 KM376368 KM376332
Eudema nubigena Bonpl. subsp. nubigena Holm-Nielsen 25009 Ecuador: Pichincha KM376255 KM376292 KM376364 KM376329
(MO)
Eudema nubigena Bonpl. subsp. nubigena Chimbolema 668 Ecuador: Tungurahua – KM376294 KM376366 –
(QCNE)
Eudema nubigena Bonpl. subsp. remyana (Wedd.) Al- Sklenar 2284 (MO) Ecuador: Chimborazo KM376256 KM376293 KM376365 KM376330
Shehbaz
Eudema rupestris Bonpl. Jörgensen 1736 (QCNE) Ecuador: Azuay KM376254 – – –
Onuris alismatifolia Gilg ex Skottsb. Pisano 8242 (CONC) Argentina: Tierra del KM376232 KM376272 KM376346 KM376310
Fuego
Onuris graminifolia Phil. Zuloaga 12540 (SI) Argentina: Neuquén KM376226 KM376266 KM376340 KM376304
Onuris graminifolia Phil. Zuloaga 13866 (SI) Argentina: Río Negro KM376227 KM376267 KM376341 KM376305
Onuris hatcheriana (Gilg ex Macloskie) Gilg & Muschl. Landero 611B (CONC) Chile: Región XII KM376237 KM376277 KM376351 KM376315
Onuris hatcheriana (Gilg ex Macloskie) Gilg & Muschl. Arroyo 85060 (CONC) Chile: Región XII KM376238 KM376278 KM376352 KM376316
Onuris hatcheriana (Gilg ex Macloskie) Gilg & Muschl. Landero 735 (CONC) Chile: Región XII KM376239 KM376279 KM376353 KM376317
Onuris hauthalii (Gilg & Muschl.) Al-Shehbaz Arroyo 850953 (CONC) Chile: Región XII KM376233 KM376273 KM376347 KM376311
Onuris hauthalii (Gilg & Muschl.) Al-Shehbaz Arroyo 850848 (CONC) Chile: Región XII KM376234 KM376274 KM376348 KM376312
Onuris hauthalii (Gilg & Muschl.) Al-Shehbaz Arroyo 92191A (CONC) Chile: Región XII KM376235 KM376275 KM376349 KM376313
Onuris hauthalii (Gilg & Muschl.) Al-Shehbaz Arroyo 92293 (CONC) Chile: Región XII KM376236 KM376276 KM376350 KM376314
Onuris papillosa O.E. Schulz Zuloaga 14040 (SI) Argentina: Santa Cruz KM376228 KM376268 KM376342 KM376306
Onuris papillosa O.E. Schulz Zuloaga 14063 (SI) Argentina: Santa Cruz KM376229 KM376269 KM376343 KM376307
Onuris spegazziniana Gilg & Muschler Zuloaga 14069 (SI) Argentina: Santa Cruz KM376230 KM376270 KM376344 KM376308
Onuris spegazziniana Gilg & Muschler Arroyo 870215 (CONC) Chile: Región XII KM376231 KM376271 KM376345 KM376309
Xerodraba patagonica (Speg.) Skottsb. Zuloaga 14035 (SI) Argentina: Santa Cruz KM376220 KM376260 KM376334 KM376298
Xerodraba patagonica (Speg.) Skottsb. Zuloaga 14104 (SI) Argentina: Santa Cruz KM376224 KM376264 KM376338 KM376302
Xerodraba pectinata Skottsb. Zuloaga 14030 (SI) Argentina: Santa Cruz KM376219 KM376259 KM376333 KM376297
Xerodraba pectinata Skottsb. Zuloaga 14066 (SI) Argentina: Santa Cruz KM376221 KM376261 KM376335 KM376299
Xerodraba pectinata Skottsb. Zuloaga 13986 (SI) Argentina: Santa Cruz KM376223 KM376263 KM376337 KM376301
Xerodraba pycnophylloides (Speg.) Skottsb. Zuloaga 14111 (SI) Argentina: Santa Cruz KM376222 KM376262 KM376336 KM376300

increase the yield, specificity, and consistency of PCRs. Cleaning
of PCR products was done by Macrogen, Inc. (Seoul, Korea) using
the Montage PCR purification kit from Millipore and following the
manufacturer’s protocol. Sequencing reactions were also per-
formed by Macrogen using the ABI PRISM BigDye Terminator
Cycle Sequencing Kits with AmpliTaq DNA polymerase (Applied
Biosystems, Seoul, Korea) following the protocols supplied by
the manufacturer. Sequences were assembled and edited using
the program Chromas Pro v.1.41 (Technelysium Pty, Ltd.), which
was also used for checking the presence of single peaks in the
chromatograms, especially in the ITS sequences. One hundred
and forty-nine new sequences were obtained and submitted to
GenBank. Although only ITS sequences were obtained for Brayop-
sis colombiana and Eudema rupestris Bonpl., they were also
included in the combined dataset. Voucher information and Gen-
Bank accession numbers are provided in the Table 1. Alignments
were generated with Muscle v.3.6 (Edgar, 2004) using a first
round of multiple alignments and posterior rounds of refinement
under the default settings. The alignments obtained were then
checked and improved manually where necessary by visual
refinement using the program Bioedit v.7.0.9.0 (Hall, 1999). All
aligned matrices were submitted to TreeBase (http://purl.org/
phylo/treebase/phylows/study/TB2:S15731).
2.3. Monophyly of Eudemeae
To evaluate the monophyly of Eudemeae, the Brassicaceae data-
set from ITS and trnL-F were analyzed separately using maximum
parsimony (MP), maximum likelihood (ML), and Bayesian inference
(BI). In these analyses gaps were treated as missing data. For MP
analysis, tree searches were generated in TNT v1.1 (Goloboff
et al., 2008) using heuristic searches with 1000 random addition
sequences, tree bisection and reconnection (TBR) branch swapping,
and holding 10 trees per replicate. Generated trees were then sub-
mitted to a new round of TBR branch swapping to completion.
Branch support was estimated by jackknife (JK) analysis (Farris
et al., 1996) with 2000 replicates of 10 random addition sequences,
holding four trees per replicate and using the default removal prob-
ability (0.36). The ML analysis was conducted in RAxML v7.2.6
(Stamatakis, 2006). The models of nucleotide substitution were
selected by using the Akaike information criterion (AIC) imple-
mented in jModeltest v2.1.4 (Darriba et al., 2012): GTR + I + G
(ITS) and TVM + G (trnL-F). RAxML was used to conduct nonpara-
metric bootstrap (BS) analysis and searches for the best-scoring
ML tree in a single run (Stamatakis et al., 2008). We executed
1000 rapid bootstrap inferences and, thereafter, a thorough ML
search under the GTR + I + G (ITS) and the GTR + G (trnL-F) models.
48 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59
Bayesian analyses were conducted using MrBayes v3.2.2 (Ronquist
et al., 2012) using GTR + I + G (ITS) or GTR + G (trnL-F) models, and
the priors on state frequencies, rates and shape of the gamma dis-
tribution were estimated automatically from the data assuming
no prior knowledge about their values (uniform Dirichlet prior).
Two simultaneous analyses, starting from different random trees
and with four Markov Monte Carlo chains run for eight million gen-
erations, sampling every 1000 generations to ensure independence
of the successive samples. The first 2000 trees (25% of total trees)
were discarded as burn-in. The convergence and effective sample
size (ESS) of each replicate were checked using Tracer v. 1.6.0
(Rambaut et al., 2013), verifying that the effective sample size for
all parameters was over 400. The remaining samples of each run
were combined, and a 50% majority consensus tree was calculated.
Because in the analyses conducted the genus Delpinophytum was
excluded from the rest of Eudemeae, the hypothesis of monophyly
of Eudemeae including Delpinophytum was tested using the SH test
(Shimodaira and Hasegawa, 1999) implemented in PAUP⁄ v.4.0b10
(Swofford, 2003). Searches of constrained topologies, where Delpin-
ophytum was forced to be included into a monophyletic group,
together with the remaining taxa of Eudemeae, were conducted
in RAxML with 1000 replicates and the models used above, and
the significance of differences between the best ML unconstrained
and constrained trees (Eudemeae including Delpinophytum) was
determined in PAUP⁄ v.4.0b10 using 1000 BS replicates and reject-
ing the hypothesis when p < 0.05.
2.4. Phylogenetic relationships within Eudemeae
Datasets from ITS, trnL-F, trnH-psbA, and rps16, representing the
tribe Eudemeae, were analyzed separately and combined using
maximum parsimony, maximum likelihood, and Bayesian infer-
ence, following the same methodology described above. However,
for MP analyses gaps were coded as present or absent in accordance
with the ‘‘simple indel coding’’ method implemented by Simmons
and Ochoterena (2000) using the program FastGap v.1.2
(Borchsenius, 2009). Gaps derived from ambiguous alignment
regions of mononucleotide repeat units (poly-N’s) were discarded
following recommendations of Kelchner (2000). Models selected
by jModeltest v2.1.4 for each region were: SYM + I + G (ITS),
TPM1uf + G (trnL-F and rps16), and TIM1 + G (trnH-psbA). To
address levels of incongruence among data partitions (ITS and
cpDNA) and their influence on combined datasets, congruence
among datasets were analyzed by the incongruence length differ-
ence test of Farris et al. (1995) in TNT v1.1 using the ‘‘ILD.run’’ script
with 1000 replications. Datasets were combined because they were
not significantly incongruent (p > 0.05) [p(ITS vs trnL-F) = 0.145,
p(ITS vs trnH-psbA) = 0.075, p(ITS vs rps16) = 0.761, p(trnL-F vs
trnH-psbA) = 0.724, p(trnL-F vs rps16) = 0.638, p(trnH-psbA vs
rps16) = 0.21, p(ITS vs cpDNA) = 0.081]. The combined datasets
(cpDNA and ITS + cpDNA) were subjected to a ML search using RAx-
ML under three (cpDNA) or four (ITS + cpDNA) partitions, 1000 rep-
licates and the GTR + GAMMA (cpDNA) or GTR + GAMMA + I
(ITS + cpDNA) models. Combined datasets were set in MrBayes as
nst = 6 and rates = invgamma (ITS) or = gamma (trnL-F, trnH-psbA,
rps16) with rate-matrix parameters, state frequencies, gamma-
shape parameter, and proportion of invariable sites unlinked across
partitions. All analyses were conducted using two run for eight mil-
lion generations, sampling every 1000 generations and discarding
the first 2000 trees (25% of total trees). Convergence and effective
sample size (ESS) of each replicate were checked in Tracer v.
1.6.0, verifying the ESS for all parameters >400.
In addition to analyses of the combined data set, incongruences
between ITS and cpDNA data were visualized in a filtered super-
network calculated with SplitsTree 4.8 (Huson and Bryant, 2006)
using 1000 Bayesian posterior trees (see below) per each nuclear
and plastid data set, and filtering the splits to show only those
present in a minimum of 35% input trees.
2.5. Climatic niche of main lineages
We implemented maximum entropy modeling to geographic
distribution using the program Maxent v.3.3.3k (Phillips and
Dudík, 2008) to investigate which environmental variables con-
tribute the most to Eudemeae distribution.
Our distribution data included latitude and longitude informa-
tion from 362 herbarium specimens deposited in B, BAA, BAB,
CONC, GH, HIP, LIL, LPB, MO, SGO, SI, US and USM. Geographic coor-
dinates were obtained from GPS coordinates in the specimen labels
or georeferenced localities. The dataset represented all species
included in Eudemeae except Delphinophytum (because this genus
resulted segregated from the tribe in our molecular phylogenies,
see results) and their entire geographical range. As input data for
the present climatic conditions, we used 19 eco-climatological vari-
ables in 2.5 arc-minutes developed by Hijmans et al. (2005) and
downloaded from the WorldClim dataset (version 1.4 release 3).
We performed initial analyses on all 19 variables and then chose cli-
matic variables that contribute most in the Maxent models (using
jackknife test) and with a low coefficient of correlation (r < 0.7).
All Maxent analyses were performed using 10 replicates, maximum
iterations were set to 1000, and all other options were left as default
(logistic output, convergence threshold of 0.00001, 10,000 back-
ground points, regularization multiplier of 1, default prevalence of
0.5 and autofeatures). Additionally, variation of the selected cli-
matic variables for the main lineages was studied using principal
component analysis (PCA) in Past v2.17 (Hammer et al., 2001) with
the correlation matrix. Significance differences between variables
of these lineages were assessed using the ANOSIM test (Clarke,
1993) implemented in PAST v2.17 with 10,000 permutations.
Finally, univariate significative differences of these variables and
altitude between the main lineages were tested using the non-para-
metric Mann–Whitney U-test in PAST v2.17.
2.6. Reconstruction of character-state evolution
To study implications of the obtained phylogenies on the mor-
phological variation in Eudemeae, ancestral-state reconstructions
were conducted on six diagnostic morphological characters related
to habit, inflorescences, and fruits. Traits and their states codified
were: (1) growth form (rosette/cushion) (Fig. 2), (2) raceme rachis
(elongated/reduced/absent) (Fig. 3), (3) bracts (absent/present)
(Fig. 3), (4) number of flowers (1/>1), (5) fruit type (silicle:
<3  longer than broad/silique: >3  longer than broad) (Fig. 4),
and (6) fruit compression (angustiseptate/terete/latiseptate). The
morphological data matrix is included in Appendix B (Supplemen-
tary material). These characters were selected because they are con-
stant and diagnostic within each genus but variable among genera.
For the reconstructions we used the trees obtained with the com-
bined ITS + cpDNA dataset, but pruning them to include only the
members of Eudemeae and the closely related genera Schizopetalon
and Menonvillea. Ancestral states were reconstructed on the ML tree
using maximum parsimony (MP) and Bayesian inference (BI). MP
reconstructions were estimated using TNT v1.1, and the resulting
ancestral-state reconstructions were visually displayed by color-
coding the branches of the ML tree. BI reconstructions were con-
ducted employing a continuous-time Markov model of trait evolu-
tion (Pagel, 1994, 1997, 1999), implemented in BayesTraits v2
(Pagel and Meade, 2013), with two (characters 1, 3–5) or six (char-
acters 2, 6) instantaneous rates representing all possible state
changes. Ancestral state reconstructions were executed using the
MCMC method implemented in Bayestraits (Pagel et al., 2004) and
the 12,002 trees obtained in the Bayesian analyses using the com-
 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59
Fig. 4. Fruit diversity in Eudemeae. (A–E) silicles. (A) Xerodraba patagonica (Zuloaga
14104); (B) X. pycnophylloides (Zuloaga 14111); (C) Onuris alismatifolia (Donat 436);
(D) O. papillosa (Zuloaga 14725); (E) Delpinophytum patagonicum (Zuloaga 14654).
(F–I) siliques. (F) Brayopsis calycina (Zuloaga 14327); (G) B. monymocalix (Zuloaga
10803); (H) Aschersoniodoxa cachensis (Sleumer 5456); A. mandoniana (Mandon 895).
Scale bars, A–G, 1 mm; H–I, 2 mm.
bined ITS + cpDNA dataset. An exponential prior for rate coefficients
was used, but because there is little information about the mean of
exponential prior, this parameter was seeded from a uniform hyper-
prior, which allows values of the prior to be estimated from the data
(as recommended by Pagel et al., 2004; Pagel and Meade, 2013). The
ranges for the uniform hyperprior were obtained using an empirical
approach: the rate coefficient on each of the Bayesian trees was esti-
mated under maximum likelihood and then these values were used
to set the range of the hyperprior. Two independent analyses were
run for ten million generation and sampled every 1000 generations
to ensure independence. The first million generations were dis-
carded as burn-in (convergence and ESS > 400 were checked with
Tracer) and the rest of the samples from the two replicates were
combined (18,000 samples). In all cases character-state transitions
were assumed as unordered.
3. Results
3.1. Monophyly of Eudemeae
The ITS alignment for the Brassicaceae dataset included 182
sequences (45 from ingroup) and was 743 bp long, of which 372
(50%) were parsimony informative. The MP analysis of this region
resulted in 7200 most parsimonious trees (MPT). The trnL-F align-
ment included 160 sequences (43 from ingroup) and was 1339 bp
49
long, of which 330 (24%) were parsimony-informative. The MP
analyses of this region recovered 10,000 MPT. All MP, ML and BI
analyses of both regions included members of the tribe Eudemeae,
except Delpinophytum, in a well-supported clade [ITS, jackknife
support (JK): 91%, bootstrap support (BS): 77%, posterior probabil-
ity (PP): 100%; trnL-F JK: 82%, BS: 82%, PP: 100%] (Fig. 5). Further-
more, all analyses included Eudemeae in a clade with tribes
Cremolobeae and Schizopetaleae (ITS JK and BS: <50%, PP: 83%;
trnL-F JK and BS: <50%, PP: 82%); within this group the sister group
of Eudemeae was not resolved. Delpinophytum was excluded from
Eudemeae in all analyses conducted and recovered in a clade with
genera of the Brassicaceae lineage I. MP analyses with ITS data and
MP/ML/BI analyses with trnL-F sequences placed Delpinophytum as
the sister group of tribe Lepidieae (ITS JK: <50%; trnL-F JK: 51%, BS:
58%, PP: 73%]. The hypothesis of monophyly of Eudemeae includ-
ing Delpinophytum was rejected under the SH test for both ITS
and trnL-F datasets (ITS: lnL 17629.06 vs. 17651.43, DD-
lnL = 22.37, p < 0.001; trnL-F: lnL 9808.99 vs. 9908.68, DD-
lnL = 99.69, p < 0.001, respectively).
3.2. Phylogenetic relationships within Eudemeae
Characteristics of ITS, trnL-F, trnH-psbA, and rps16 sequences
and alignments are summarized in Table 2. The MP, ML and BI
analyses of the ITS and cpDNA data recovered similar topologies,
showing the same strongly supported clades (Fig. 6A and B)
(results from analyses of the individual cpDNA regions are shown
in Supplementary Figs. S1 and S2). Both datasets recover Eude-
meae, excluding Delpinophytum, as monophyletic. Within Eude-
meae there are two main clades: one, the ‘‘Northern-Central
Andes’’ clade (NCA), comprises Aschersoniodoxa, Brayopsis, Dactylo-
cardamum, and Eudema; and the ‘‘Southern Andes clade’’ (SA),
which includes Onuris and Xerodraba (Fig. 6). The NCA clade had,
in all results, higher support than the SA clade, and it was also
recovered with ITS indel data (Supplementary Fig. S2). Within
the NCA clade, the relationships among genera were poorly sup-
ported, and they appeared as non-monophyletic (Fig. 6A and B).
On the other hand, in the SA clade the same three strongly sup-
ported clades were present, the first one including species of
Xerodraba (hereafter the ‘‘xerodraba’’ subclade), the second speci-
mens of Onuris hauthalli, (hereafter the ‘‘hauthalli’’ subclade), and
the third clade including the remaining species of Onuris (hereafter
‘‘onuris s.s.’’ subclade). Phylogenetic relationships among these
three subclades appeared ambiguous and poorly supported in the
different analyses. Results from MP, ML and BI using the
ITS + cpDNA combined dataset (3 kb) recovered the same main
clades presented above: Eudemeae (excluding Delpinophytum)
(JK: 99%, BS and PP: 100%) (Fig. 6C), NCA (JK, BS and PP: 100%),
and SA (JK: 72%, BS: 94%, PP: 100%). Within the NCA clade, Ascher-
soniodoxa peruviana Al-Shehbaz, Navarro & A. Cano was placed as
the most basal taxa, while the remaining species of the genus were
recovered in a highly supported clade (JK, BS and PP: 100%) (here-
after the aschersoniodoxa s.s subclade). Neither Brayopsis nor
Eudema appear monophyletic. The SA clade also included the
strongly supported xerodraba, onuris s.s., and hauthalli subclades
(all JK, BS and PP: 100%). The xerodraba and hauthalli are shown
as sister subclades, although this relationship was unsupported
(MP) or only weakly supported (ML and BI).
Despite non-significantly incongruence between chloroplast
and nuclear data, differences between the cpDNA and ITS trees
were represented graphically by a filtered supernetwork (Supple-
mentary Fig. S3), where cycles in the network represent conflicting
phylogenetic signals. Areas of incongruence were mainly located
within the NCA clade, but supported clades obtained in the phylo-
genetic analyses were also present in the supernetwork.
50 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59

Fig. 5. Bayesian 50% majority-rule consensus tree from 12,002 trees generated by Bayesian inference with MrBayes in the analyses at the family level within Brassicaceae
using: (A) the ITS dataset; (B) the trnL-F dataset. Tribes are indicated to the right of the boxes. Thick branches indicate >0.90 Bayesian posterior probability. The black arrow
indicates the phylogenetic position of Delpinophytum.
 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59
Table 2
Features of the DNA regions included in the phylogenetic analyses within Eudemeae.
ITS (ITS1-5.8S-ITS2) trnL-F (trnL intron/trnL-
F spacer)
Number of terminals 45 42
Length of the alignment (bp) 662 941
Length of sequences (ingroup) 614 (several species of 706 (O.hatcheriana)–
Brayopsis, Dactylocardamum and 914 (D.patagonicum)
Eudema)–645 D.patagonicum
No. of parsimony-informative 181 (27%) 66 (7%)
characters
Maximum sequence divergence 21% 6%
within ingroup (%)*
No. of gaps in the alignment 25 16
No. of parsimony-informative 11 7 21
gaps in the alignment
Model selected by AIC SYM + I + G TPM1uf + G
Number of MPT 3 54
*
Calculated as p-distance (dissimilarity distance).
3.3. Climatic niche of main lineages
Four variables according to their importance in the Maxent
models and with a low coefficient of correlation were selected:
mean temperature of coldest quarter (BIO11) (Fig. 7A), min tem-
perature of coldest month (BIO6) (Fig. 7B), isothermality (BIO3)
(Fig. 7C), and mean temperature of driest quarter (BIO9) (Fig. 7D)
(Pearson correlation coefficient of these variables are shown in
supplementary Table S1). The average AUC (Area Under the recei-
ver operating curve) test values of the four-variables models for
the NCA and SA lineages were above 0.9. The ecologic niche of both
lineages resulted similar to their empirical distribution and mainly
spread along the north-central Andes for the NCA lineage (Fig. 7E),
and southern Andes and Patagonia for the SA lineage (Fig. 7F).
However predicted distributions for SA lineage also include,
although no specimens of this lineage were found so far, the Malv-
inas islands in the south and the Argentinean provinces Mendoza
and San Juan in the North. The PCA analysis using the four variables
capture ca. 89% of the variance in the first two components (PC1:
52.42%, PC2: 38.38%) and the variable loadings (Table 3) showed
that the first component was most highly influenced by the mean
temperature of coldest quarter (BIO11) and min temperature of
coldest month (BIO6), while variables that most influence the sec-
ond component were isothermality (BIO3) and mean temperature
of driest quarter (BIO9). Specimens of both lineages were separated
into two, well-defined groups (Fig. 7G) that resulted in significant
differentiation under the ANOSIM test (p < 0.01). Isothermality
(BIO3) (72 ± 10 NCA vs. 49 ± 2 SA) and mean temperature of the
driest quarter (BIO9) (2.9 °C ± 1.85 NCA vs. 6.5 °C ± 2.72 SA) were
the variables that most contribute to this segregation (Fig. 7C, D)
and were significantly differentiated, together with the elevation
range (4143 ± 565 m in NCA vs 830 ± 502 m in SA), in the habitats
of the two lineages.
3.4. Reconstruction of character-state evolution
Reconstructions based on MP and BI showed similar results
(Figs. 8 and 9). Growth form. All analyses revealed that the presence
of dense cushions with imbricate leaves in tribe Eudemeae appears
to have arisen independently twice: in Xerodraba within the SA
clade, and in Dactylocardamum within the NCA clade (Fig. 8A). On
the other hand,the rosette morphology proved to be the plesiomor-
phic state of the tribe. Inflorescence architecture. These appeared
highly variable in the tribe, and each of character studied has
undergone at least two state transitions characterizing several
clades within the Eudemeae. Both NCA and SA clades included all
51
trnH-psbA (spacer) rps16 (intron) cpDNA ITS + cpDNA
41 40 43 46
510 905 2356 3018
259 (O.hauthall)– 761 (Xerodraba)–854 – –
384(X.pectinata) (O.hatcheriana)
93 (18%) 75 (8%) – –
14% 5% – –
38 30 – –
15 – –
TPM1uf + G TPM1uf + G – –
32 9 48 168
character states reviewed. Regardingthe raceme development
(Fig. 8B), inflorescence flowering characterized the onuris s.s. and
aschersoniodoxa s.s. subclades (p > 0.99), while the rosette flower-
ing was assigned to the NCA clade and the hautalli subclade
(p = 0.83 and 0.99, respectively). The rosette flowering was slightly
favored by the BI reconstructions as the ancestral state of Eude-
meae and was ambiguous in the MP assignation, while assigna-
tions of this character to the SA clade resulted ambiguous in both
cases. According to MP assignations, bracts in the inflorescences
of Eudemeae occurred twice (Fig. 8C), one in the NCA clade (in A.
peruviana) and again in the SA clade (in the onuris s.s. subclade),
while in the BI reconstructions of the ancestral state of this charac-
ter was ambiguous (p 6 0.6). Finally, the reduction of the inflores-
cences to solitary flowers (Figs. 8D) occurred independently in the
cushion genera, showing the same assignation of the growth form.
Fruit morphology. The evolution of fruit type in Eudemeae showed
two main different patterns (Fig. 9A and B). The SA clade was
defined by the presence of latiseptate silicles; this fruit type
remains unchanged in the clade. On the other hand, the NCA clade,
characterized by having siliques as the ancestral fruit, registered
several transitions from siliques to silicles in species of Eudema
and Dactylocardamum. Furthermore, fruit compression was highly
variable within the NCA clade, including angustiseptate, terete,
and latiseptate fruits, being the ancestral state ambiguous both
in the MP and BI reconstructions. Ancestral fruit type for the tribe
resulted ambiguous in the MP reconstruction, and was a latisepte
silicle in the BI reconstruction; these state assignations were recov-
ered with p < 0.9.
4. Discussion
4.1. Phylogenetic relationships of Eudemeae and characteristics of its
main lineages
The present nuclear and chloroplast phylogenetic analyses
showed that the tribe Eudemeae, minus Delpinophytum, is mono-
phyletic. The tribal assignment of Delpinophytum varied from
including it by Schulz (1936) in tribe Lepidieae, based on having
subdidymous angustiseptate fruits with two, 1-seeded segments
as in Coronopus Zinn (= Lepidium), to tentatively including it in
the Eudemeae by Al-Shehbaz (2012a,b) based on forming dense
cushions with imbricate leaves, as in Dactylocardamum and Xerodr-
aba. Delpinophytum is indistinguishable from Xerodraba in almost
every aspect except for having angustiseptate vs. latiseptate fruits.
Our results support the inclusion of Delpinophythum in Lepidieae
and show that the similar cushion-type habitat and imbricate
52 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59

Fig. 6. Bayesian 50% majority-rule consensus trees from 12,002 trees generated by Bayesian inference with MrBayes in the analyses at the tribe level within Eudemeae and
using: (A) the cpDNA dataset (trnL-F/trnH-psbA/rps16 intron); (B) the ribosomal ITS dataset; (C) the combined cpDNA + nrITS dataset. Values on branches correspond to
maximum likelihood Bootstrap P50%, thick branches indicate >0.90 Bayesian posterior probability. (D) Geographical distribution of specimens included in the main NCA and
SA clades, red dots represent herbarium records for specimens of the NCA clade, blue dots represent specimens of the SA clade. Elevation is depicted in gray scale from sea
level (white) to over 4000 m above sea level (black). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59 53

Fig. 7. Eco-climatological niche in Eudemeae and its main lineages. (A–D) values of variables that most contribute in the MaxEnt models along the distribution area of
Eudemeae, (A) mean temperature of coldest quarter [BIO11]; (B) min temperature of coldest month [BIO6]; (C) Isothermality [BIO3]; (D) Mean temperature of driest quarter
[BIO9]; (E, F) Probability of presence (logistic output) predicted from the MaxEnt ecological niche models for the NCA (E) and SA (F) lineages; (G) Biplot of the two principal
components extracted in the principal component analysis (variance PC1: 52.42%, PC2: 38.38%) of the four variable data matrix (BIO11, BIO6, BIO3, BIO9) for the specimens of
Eudemeae, blue dots correspond to specimens of the NCA lineage and red dots to the SA lineage. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)

leaves evolved independently in the above-mentioned three cush- (Al-Shehbaz, 2012b), also forms dense cushions and belongs to the
ion-forming genera. Lithodraba Boelcke, a monospecific genus Lepidieae based on molecular data (Warwick et al., 2010; this work
endemic to Mendoza and Neuquén provinces of Andean Argentina Fig. 5A). The latter genus also has angustiseptate, 2-seeded silicles
54 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59
Table 3
Factor loadings on the first three components for the four BIO-variables used in the
PCA.
BIO-variables PC1 PC2
BIO3 0.24 0.93
BIO6 0.91 -0.13
BIO9 0.68 -0.65
BIO11 0.86 0.39
Variance (%) 52.43 36.38
BIO-variables: BIO11 = mean temperature of coldest quarter; BIO6 = min tempera-
ture of coldest month; BIO3 = Isothermality; BIO9 = Mean temperature of driest
quarter.
(Boelcke, 1951) but differs by having non-imbricate long leaves
and four stamens. The ITS topologies did not recover Delpinophy-
tum and Lithodraba in a monophyletic group, suggesting that their
cushion-forming habit also evolved independently in the Lepidieae
(see below).
The tribes Eudemeae, Cremolobeae, and Schizopetaleae are
exclusively South American groups first shown to be monophyletic
by Warwick et al. (2010). Tribe Cremolobeae is distributed from
Colombia into southern Argentina and Chile, and its genera grow
in the Andean highlands and from the dry coastal deserts of north-
ern Chile to the semi-arid regions (matorral) of central Chile
(Khanna and Rollins, 1965; Rollins, 1955; Salariato et al., 2014).
The Schizopetaleae are primarily distributed across the southern
Atacama Desert and the matorral of central Chile, with a few species
in southern Peru and neighboring Andean highlands of Chile and
Argentina (Al-Shehbaz, 1989c; Toro-Nuñez et al., 2013). Although
the Eudemeae also has an Andean distribution from Colombia into
Patagonia, it differs from the other two tribes by having no taxa in
the arid regions of north-central Chile and southern Peru (Al-
Shehbaz, 1989a,b, 1990a,b, 2008, 2012b; Al-Shehbaz and Cano,
2011; Al-Shehbaz et al., 2012, 2013; Boelcke 1984; Boelcke and
Romanczuk, 1984). A detailed analysis of the historical biogeogra-
phy and diversification of the three tribes is beyond the scope of
this paper, but will be conducted in the near future. Our results
showed that tribe Eudemeae diverged into two, geographically sep-
arated lineages. Genera of the NCA clade, which includes Ascherso-
niodoxa, Brayopsis, Dactylocardamum, and Eudema, are distributed
along the the Andean highlands from Colombia into northern
Argentina. Their species grow mainly between 3500 and 5300 m
in the humid mountain areas of Colombia, Ecuador, and Peru,
though some reach the semiarid to arid areas of southwestern Boli-
via and northwestern Argentina but never the deserts of western
Peru and northern Chile.
Within the NCA clade, Aschersoniodoxa is clearly distinguished
by having leaflike, strongly latipseptate, basipetally dehiscent sili-
ques with valves remaining adnate to the replum base (Al-
Shehbaz, 1990b). However, A. peruviana differs from its congeners
by having crenate-dentate (vs. entire) basal leaves, complete fruit
septum (vs. reduced to a narrow rim), and a rosette flowering, in
which each flower arises from axils of basal leaves (vs. ebracteate
racemes) (Al-Shehbaz et al., 2012).
Dactylocardamum is a monospecific genus the type of which, D.
imbricatifolium, is known from two collections made in 1959 and
2013 (Fig. 2F–H). It is not closely related to Xerodraba, mainly dif-
fering by its fruit type (indehiscent and terete silicles in Dactylocar-
damum vs. dehiscent and latiseptate silicles in Xerodraba). As
suggested by Al-Shehbaz (1989a), imbricate foliage and cushion
habit seem to have evolved independently in both genera.
Brayopsis and Eudema, the other two genera of the NCA lineage,
are morphologically similar in having distinctive rosette flowering
with ebracteate flowers arising from the center of the rosette. They
differ mainly by fruit type (terete siliques in Brayopsis vs.
angustiseptate silicles in Eudema). Al-Shehbaz (1990a) critically
evaluated the boundaries between these genera, and added addi-
tional differences as the funicle length (long and filiform vs. short
PC3 and thick) and the seed-coat sculpture (colliculate-reticulate vs.
0.05 coarsely reticulate). Our results show that both genera are not
-0.38 monophyletic, with B. gamosepala Al-Shehbaz and B. colombiana
0.26 Al-Shehbaz related to species of Eudema, while E. rupestris Bonpl.
0.18
6.33
related to species of Brayopsis. Since the fruit type, the main diag-
nostic character distinguishing Brayopsis from Eudema, is highly
variable within the Eudemeae (see below), new multilocus phylog-
enies including all species of both genera, as well as a species-level
phylogenetic approach, should be conducted in order to clarify the
boundaries of Brayopsis and Eudema within the NCA lineage.
The SA clade, which includes Xerodraba and Onuris, is distrib-
uted exclusively in the Andean–Patagonian region of southern
Argentina and Chile. Their species grow primarily at elevations of
the Altoandean biogeographic province (Cabrera and Willink,
1973) up to 2000 m, although they are also present in lowlands
of the Patagonian province, reaching the Argentinean eastern coast
(Figs. 6D and 8F).
Species of Xerodraba form dense cushions with imbricate leaves
and solitary flowers (Fig. 2L–N.). They grow on rocky slopes of the
Andes and Patagonian deserts and can tolerate lower mean tem-
peratures, strong winds, frost throughout the year, and receive
an annual rainfall below 150 mm (Ancibor, 1969; Cabrera and
Willink, 1973).
Onuris s.s. (excluding O. hauthalii) has a distribution similar to
Xerodraba, but it reaches Neuquén province in northern Patagonia
(Fig. 1G), whereas the northern limits of Xerodraba is Chubut prov-
ince in central Patagonia. Onuris s.s. is strongly supported as a
monophyletic group, and is easily distinguished from the rest of
Eudemeae by its many-flowered racemes that are bracteate
throughout and elongated in fruit (Boelcke, 1984). On the other
hand, Onuris hauthalii, a species restricted to Santa Cruz Province
in Argentina and Magallanes and the Antarctic Region in Chile,
clearly differs from the remaining species of Onuris by having a
rosette flowering similar to those of Brayopsis and Eudema.
The generic placement of Onuris hauthalii fluctuated from
Eudema (Gilg and Muschler, 1909), Brayopsis (Skottsberg, 1916),
or Onuris (Al-Shehbaz, 2012a,b). The latter author presented several
differences between O. hauthalii and Eudema, as the presence of rhi-
zomes (vs. caudex), nearly smooth (vs. conspicuously reticulate)
seeds, and latiseptate (vs. angustiseptate) fruits. Although the pres-
ent study confirms that this species is more closely related to Onuris
s.s. than to Eudema, its sister-group relationship was ambiguous
and low supported within the SA clade. The molecular data, coupled
with the morphological differences between O. hauthalii and Onuris
s.s. and Xerodraba, suggest that its inclusion in a new monotypic
genus would be the best taxonomic solution of the problem.
4.2. Distribution and climatic niches of Eudemeae and main lineages
Tribe Eudemeae, like most other South American Brassicaceae,
presents a geographic distribution closely associated with the
Andean range. Climatic variables, which contribute most to its dis-
tribution, are linked to the alpine climate (e.g., lowest temperatures,
mean temperature of coldest quarter, minimum temperature of the
coldest month). Habitat and climatic niches of the North-Central
(NCA) and Southern Andes (SA) lineages are well differentiated.
Species of the NCA lineage grow at high elevations (about 3500–
5300 m) while species of the SA lineage occupy the southern portion
of the Andes and the Patagonian steppe, at much lower elevations
principally from sea level to 2000 m. Regions inhabited by species
of the NCA lineage are exposed to higher day-to-night temperature
oscillation than the species of the SA lineage. Furthermore, during
 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59 55

Fig. 8. Ancestral-state reconstruction in Eudemeae for (A) growth form; (B) raceme elongation; (C) presence of bracts; and (D) number of flowers. Ancestral states were
reconstructed on the maximum likelihood tree obtained with the ITS + cpDNA dataset using maximum parsimony in TNT, and bayesian inference in BayesTraits. Colors of
branches reflect reconstruction of ancestral character-states using maximum parsimony; pie-diagrams represent the mean probabilities of different states at each node from
18,000 asignations under Bayesian inference. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
56 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59

Fig. 9. Ancestral-state reconstruction in Eudemeae for (A) fruit shape; and (B) fruit compression. Ancestral states were reconstructed on the maximum likelihood tree
obtained with the ITS + cpDNA dataset using maximum parsimony in TNT, and bayesian inference in BayesTraits. Colors of branches reflect reconstruction of ancestral
character-states using maximum parsimony; pie-diagrams represent the mean probabilities of different states at each node from 18,000 asignations under Bayesian
inference. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

the driest season, the temperature is particularly low in the distri-
bution area of the NCA lineage compared to that of the SA lineage.
4.3. Evolution of diagnostic morphological traits
Growth form. While the hemicryptophyte rosette (Raunkiær,
1934) (Fig. 2A–E) is the ancestral growth form in Eudemae, the
cushion habit (Fig. 2F–N) evolved independently twice in the tribe
(Dactylocardamum and Xerodraba). Delpinophytum, the remaining
cushion-former originally included in Eudemeae, was placed within
the lineage I, probably within tribe Lepidieae, showing that this
character evolved twice also in South American members of this
tribe (Lithodraba and Delpinophytum). Therefore, cushion formation
in South American Brassicaceae evolved independently four times.
Cushion formation is rather uncommon elsewhere in the family
and is known in Draba L. (Jordon-Thaden et al., 2010), Baimashania
Al-Shehbaz (Al-Shehbaz, 2000), and Lepidium (Rollins, 1948). Cush-
ion formation is one of the most conspicuous growth forms in highly
exposed alpine habitats and is especially abundant in temperate
and subpolar regions. This life-form represents an efficient trap
for heat and water, with a maximum reduction of losses due to its
lowest surface-volume ratio (Aubert et al., 2014). It is common in
various parts of South America, such as Patagonia, the subantarctic
mountains, and the tropical alpine environments where ‘‘winter’’
comes nearly every night (Bliss, 1971; Körner, 1995; Went, 1971).
Indeed, the revised worldwide catalogue of cushion plants (Aubert
et al., 2014), shows the Andes and Patagonia as the richest zones
worldwide in the number of cushion-former genera. Therefore, con-
vergence in cushion formation is not rare and represents an adapta-
tion to extreme sites (Böcher, 1977; Went, 1971).
Inflorescence architecture. The inflorescence structure, with
indeterminate racemes, is quite uniform throughout most of the
Brassicaceae (Appel and Al-Shehbaz, 2003). However, Eudemeae
is highly variatiable in having inflorescence and rosette flowering,
ebracteate or bracteate racemes, and solitary pseudo terminal
flowers (Fig. 3). Rosette flowering in Brassicaceae, cited as an adap-
tation to growing in low-competition enviroments with short and
unpredictable growing seasons (Bosch et al., 2008), has arisen
independently in multiple lineages (Appel and Al-Shehbaz, 2003;
Benlloch et al., 2007). Liu et al. (2011) indicated that changes in
genes such as TERMINAL FLOWER 1 (TFL1) and LEAFY (LFY) con-
tribute to the evolution of several aspects of the derived architec-
ture exhibited in rosette flowering, most notably in internode
compression and pedicel elongation. However, evolution of rosette
flowering can be achieved in two ways involving (1) suppression of
internode elongation in the inflorescence meristem, or (2) conver-
sion of axillary meristems to floral identity (Yoon and Baum, 2004).
Therefore, models must be confirmed in each group using studies
of gene regulation and expression and structural development.
Regarding the presence/absence of bracts in the Eudemeae, our
findings suggest the independent development of bracts once each
in the NCA (Aschersoniodoxa peruviana) and SA (Onuris s.s.) lin-
 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59
eages. Flowers in Brassicaceae are usually not subtended by bracts,
though numerous genera have bracteate species and others with
ebracteate racemes, and sometimes the bracts are present along
the entire raceme or restricted to only its lowermost flowers. In
many cases, the early ontogeny shows primordia of subtending
bracts but these generally fail to develop further (Appel and Al-
Shehbaz, 2003; Penin, 2008). Several genes in Brassicaceae influ-
ence bract development [e.g., LFY (Weigel et al., 1992), TFL1
(Schultz and Haughn, 1993), APETALA 1 (AP1) (Bowman et al.,
1993), BLADE-ON-PETIOLE 1 and 2 (BOP1 and BOP2) (Norberg
et al., 2005), and JAGGED (JAG) (Dinneny et al., 2004)], generally
as the result of an intensification of vegetative shoot traits
(Penin, 2008). The existence of several genetic pathways, lending
to the bract emergence/reduction, may account for the indepen-
dent events of bract emergence/reduction in Eudemeae. Finally,
the single pseudoterminal flowers in Dactylocardamum, Delpino-
phytum, and Xerodraba appear to be associated with the cushion
habit, but the cushion-forming Lithodraba produces 1–5-flowered
short racemes (Ancibor, 1971;Boelcke, 1951). Therefore, the corre-
lation between growth form and the production of pseudoterminal
flowers is unclear. Due to the high diversity in the inflorescences of
Eudemeae, the tribe can be an ideal group to study the genetic con-
trol and the evolutionary development of the inflorescence.
Fruit morphology. Fruits of Brassicaceae exhibit an enormous
diversity in size, shape, and structure (Appel and Al-Shehbaz,
2003), though this variation has repeatedly been shown to be
highly homoplastic (Al-Shehbaz et al., 2002, 2006; Bowman,
2006; Franzke et al., 2011; Hall et al., 2011; Koch et al., 2003). Fruit
evolution in the Eudemeae exhibits two different trends: the SA
lineage (Onuris s.l. and Xerodraba) has latisepte silicles, whereas
the NCA lineage (Aschersoniodoxa, Brayopsis, Dactylocardamum
and Eudema) is highly variable and homoplastic among and within
genera. Therefore, fruit morphology in the NCA lineage seems to be
unreliable, especially in the delimitation of Brayopsis and Eudema.
Fruits in the related South American tribes Cremolobeae and
Schizopetaleae are more conserved than the SA clade of Eudemeae
(Salariato et al., 2014; Al-Shehbaz, 1989c).
4.4. Concluding remarks
Eudemeae s.s. represents a monophyletic group with a high
morphological diversity, that divide it into two main lineages geo-
graphically and climatically differentiated. These are the north-
central Andean (NCA) lineage that includes Aschersoniodoxa, Bray-
opsis, Dactylocardamum, and Eudema, and the southern Andean
(SA) lineage that includes Onuris and Xerodraba. Ancestral-state
reconstruction generally reveals multiple independent evolution
of diagnostic morphological characters such as cushion forming,
solitary pseudoterminal flowers, presence of bracts, and fruit type.
Future studies on the historical biogeography and niche evolution
in the monophyletic clade of the tribes Cremolobeae, Eudemeae,
and Schizopetaleae should enhance our understanding about the
evolution of the South American Brassicaceae along the Andes.
Acknowledgments
Funding of this work was provided by ANPCyT (Agencia Nacional
de Promoción Científica y Tecnológica) Grant PICT-2013-1042,
CONICET (Consejo Nacional de Investigaciones Científicas y Técni-
cas) Grants PID 11220100100207, PIP 11220100100155, and
D4541-12, and the National Geographic Society Grants 8365-07
and 8862-10, for which we are profoundly grateful. Fieldwork and
visits to herbaria were supported by the Myndel Botanical Founda-
tion grants in 2010, 2011, and 2012. Asunción Cano is also grateful
for funding from the National Science Foundation grant to Kenneth
Young (CNH # 1010381). Funding to Al-Shehbaz to visit herbaria
57
was supported by the US NSF Grant DEB-1252905, for which he is
prfoundly grateful. We especially appreciate the help of the Associ-
ate Editor Jocelyn Hall and two anonymous reviewers who provided
useful suggestions to improve an early version of this paper. We are
also profoundly grateful to the directors, curators, and collection
managers of the visited herbaria. We also thank E. Navarro, H. Trin-
idad, C. Ulloa, and C. Zannoti for the photographs used in this work.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ympev.
2014.09.030.
References
Al-Shehbaz, I.A., 1989a. Dactylocardamum (Brassicaceae), a remarkable new genus
from Peru. J. Arnold Arbor. 70, 515–521.
Al-Shehbaz, I.A., 1989b. The South American genera Brayopsis and Englerocharis
(Brassicaceae). Nordic J. Bot. 8, 619–625. http://dx.doi.org/10.1111/j.1756-
1051.1989.tb01736.x.
Al-Shehbaz, I.A., 1989c. Systematics and phylogeny of Schizopetalon (Brassicaceae).
Harvard Pap. Bot. 1, 10–46.
Al-Shehbaz, I.A., 1990a. Generic limits and taxonomy of Brayopsis and Eudema
(Brassicaceae). J. Arnold Arbor. 71, 93–109.
Al-Shehbaz, I.A., 1990b. The genus Aschersoniodoxa (Brassicaceae). Syst. Bot. 15,
387–393. http://dx.doi.org/10.2307/2419352.
Al-Shehbaz, I.A., 2000. Baimashania (Brassicaceae), a new genus from China. Novon
10, 320–322.
Al-Shehbaz, I.A., 2008. Brassicaceae. In: Zuloaga, F.O., Morrone, O., Belgrano, M.J.
(Eds.), Catalogue of the vascular plants of the southern cone (Argentina,
southern Brazil, Chile, Paraguay and Uruguay), vol. 2, Dicotyledoneae:
Acanthaceae-Fabaceae (Abarema–Schizolobium). Monogr. Syst. Bot. Missouri
Bot. Gard. 107, Missouri Botanical Garden Press, St. Louis, pp. 1663–1709.
Al-Shehbaz, I.A., 2012a. A generic and tribal synopsis of the Brassicaceae
(Cruciferae). Taxon 61, 931–954.
Al-Shehbaz, I.A., 2012b. Eudemeae. In: Anton, M.A., Zuloaga, F.O. (Eds.),
Brassicaceae, Flora Argentina, vol. 8. Editorial Sigma, Buenos Aires, pp. 135–150.
Al-Shehbaz, I.A., Warwick, S.I., 2007. Two new tribes (Dontostemoneae and
Malcolmieae) in the Brassicaceae (Cruciferae). Harvard Pap. Bot. 12, 429–433,
http://dx.doi.org/10.3100/1043-4534(2007)12[429:TNTDAM]2.0.CO;2.
Al-Shehbaz, I.A., Cano, A., 2011. Englerocharis dentata and Eudema peruviana
(Brassicaceae), two new species from Peru. Harvard Pap. Bot. 16, 275–278.
http://dx.doi.org/10.3100/0.25.016.0203.
Al-Shehbaz, I.A., Mummenhoff, K., Appel, O., 2002. Cardaria, Coronopus, and
Stroganowia are united with Lepidium (Brassicaceae). Novon, 5–11.
Al-Shehbaz, I.A., Beilstein, M.A., Kellogg, E.A., 2006. Systematics and phylogeny of
the Brassicaceae (Cruciferae): an overview. Pl. Syst. Evol. 2, 89–120. http://
dx.doi.org/10.1007/s00606-006-0415-z.
Al-Shehbaz, I.A., Navarro, E., Cano, A., 2012. Aschersoniodoxa peruviana
(Brassicaceae), a remarkable new species from Peru and a synopsis of the
genus. Kew. Bull. 67, 483–486. http://dx.doi.org/10.1007/s12225-012-9392-2.
Al-Shehbaz, I.A., Cano, A., Trinidad, H., Navarro, E., 2013. New species of Brayopsis,
Descurainia, Draba, Neuontobotrys and Weberbauera (Brassicaceae) from Peru.
Kew. Bull. 68, 219–231. http://dx.doi.org/10.1007/s12225-013-9447-z.
Alexander, P.J., Windham, M.D., Govindarajulu, R., Al-Shehbaz, I.A., Bailey, C.D.,
2010. Molecular phylogenetics and taxonomy of the genus Thysanocarpus
(Brassicaceae). Syst. Bot. 35, 559–577. http://dx.doi.org/10.1600/
036364410792495926.
Alexander, P.J., Windham, M.D., Beck, J.B., Al-Shehbaz, I.A., Allphin, L., Bailey, C.D.,
2013. Molecular phylogenetics and taxonomy of the genus Boechera and related
genera (Brassicaceae: Boechereae). Syst. Bot. 38, 192–209. http://dx.doi.org/
10.1600/036364413X661917.
Ancibor, E., 1969. Notas sobre la anatomía de Xerodraba. I – Anatomía foliar Bol. Soc.
Argent. Bot. 11 (4), 227–234.
Ancibor, E., 1971. Estudio anatómico y morfológico de una Crucifera andina en
cojin: Lithodraba mendocinensis. Darwiniana 16 (3–4), 519–561.
Appel, O., Al-Shehbaz, I.A., 2003. Cruciferae. In: Kubitzki, K., Bayer, C. (Eds.), The
Families and Genera of Vascular Plants. Springer-Verlag, Berlin, pp. 75–174.
http://dx.doi.org/10.1007/978-3-662-07255-4_17.
Aubert, S., Boucher, F., Lavergne, S., Renaud, J., Choler, P., 2014. 1914-2014: a revised
worldwide catalogue of cushion plants 100 years after Hauri and Schröter. Alp.
Bot. 124, 59–70. http://dx.doi.org/10.1007/s00035-014-0127-x.
Bailey, C.D., Koch, M.A., Mummenhoff, K., Mayer, M., O’Kane, S.L., Warwick, S.I.,
Windham, M.D., Al-Shehbaz, I.A., 2006. Toward a global phylogeny of the
Brassicaceae. Mol. Biol. Evol. 23, 2142–2160. http://dx.doi.org/10.1093/molbev/
msl087.
Bailey, C.D., Al-Shehbaz, I.A., Rajanikanth, G., 2007. Generic limits in tribe
Halimolobeae and the description of the new genus Exhalimolobos
(Brassicaceae). Syst. Bot. 32, 140–156. http://dx.doi.org/10.1600/
036364407780360166.
58 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59
Baldwin, B.G., 1992. Phylogenetic utility of the internal transcribed spacers of
nuclear ribosomal DNA in plants: An example from the Compositae. Mol.
Phylogenet. Evol. 1, 3–16. http://dx.doi.org/10.1016/1055-7903(92)90030-K.
Beilstein, M.A, Al-Shehbaz, I.A., Kellogg, E.A., 2006. Brassicaceae phylogeny and
trichome evolution. Am. J. Bot. 93, 607–619. http://dx.doi.org/10.3732/
ajb.93.4.607.
Beilstein, M.A., Al-Shehbaz, I.A., Mathews, S., Kellogg, E.A., 2008. Brassicaceae
phylogeny inferred from phytochrome A and ndhF sequence data: Tribes and
trichomes revisited. Am. J. Bot. 95, 1307–1327. http://dx.doi.org/10.3732/
ajb.0800065.
Benlloch, R., Berbel, A., Serrano-Mislata, A., Madueño, F., 2007. Floral initiation and
inflorescence architecture: a comparative view. Ann. Bot. 100, 659–676. http://
dx.doi.org/10.1093/aob/mcm146.
Bliss, L.C., 1971. Arctic and alpine plant life cycles. Annu. Rev. Ecol. Evol. Syst. 2,
405–438.
Böcher, T.W, 1977. Convergence as an evolutionary process. Bot. J. Linn. Soc. 75 (1),
1–19. http://dx.doi.org/10.1111/j.1095-8339.1977.tb01476.x.
Boelcke, O., 1951. Lithodraba, nuevo género de Crucíferas de la Argentina.
Darwiniana 9, 348–357.
Boelcke, O., 1984. El genero Onuris (Cruciferae), endemico de la Patagonia.
Parodiana 3 (1), 53–65.
Boelcke, O., Romanczuk, M.C., 1984. Cruciferae. In: Correa, M.N. (Ed.), Flora
Patagónica, Colección Científica del Instituto Nacional de Tecnología
Agropecuaria 8(4a). Instituto Nacional de Tecnología Agropecuaria, Buenos
Aires, pp. 373–544.
Borchsenius, F., 2009. FastGap, version 1.2. Department of Biosciences, Aarhus
University, Denmark. <http://www.aubot.dk/FastGap_home.htm>.
Bosch, J.A., Heo, K., Sliwinski, M.K., Baum, D.A., 2008. An exploration of LEAFY
expression in independent evolutionary origins of rosette flowering in
Brassicaceae. Am. J. Bot. 95, 286–293. http://dx.doi.org/10.3732/ajb.95.3.286.
Bowman, J.L., 2006. Molecules and morphology: comparative developmental
genetics of the Brassicaceae. Pl. Syst. Evol. 259, 199–215. http://dx.doi.org/
10.1007/s00606-006-0419-8.
Bowman, J.L., Alvarez, J., Weigel, D., Meyerowitz, E.M., Smyth, D.R., 1993. Control of
flower development in Arabidopsis thaliana by APETALA 1 and interacting genes.
Development 119, 721.
Bradley, D., Ratcliffe, O., Vincent, C., Carpenter, R., Coen, E., 1997. Inflorescence
commitment and architecture in Arabidopsis. Science 275, 80–83. http://
dx.doi.org/10.1126/science.275.5296.80.
Cabrera, A.L., Willink, A., 1973. Biogeografía de América Latina, Serie Biológica,
Monografía 13. Organización de los Estados Americanos, Washington, D.C.
Carlsen, T., Elven, R., Brochmann, C., 2010. The evolutionary history of Beringian
Smelowskia (Brassicaceae) inferred from combined microsatellite and DNA
sequence data. Taxon 59, 427–438.
Clarke, K.R., 1993. Non-parametric multivariate analyses of changes in community
structure. Aust. J. Ecol. 18, 117–143. http://dx.doi.org/10.1111/j.1442-
9993.1993.tb00438.x.
Couvreur, T.L.P., Franzke, A., Al-Shehbaz, I.A., Bakker, F., Koch, M.A., Mummenhoff,
K., 2010. Molecular phylogenetics, temporal diversification and principles of
evolution in the mustard family (Brassicaceae). Molec. Biol. Evol. 27, 55–71.
http://dx.doi.org/10.1093/molbev/msp202.
Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. jModelTest 2: more models,
new heuristics and parallel computing. Nat. Methods 9. http://dx.doi.org/
10.1038/nmeth.2109, 772–772.
Dinneny, J.R., Yadegari, R., Fischer, R.L., Yanofsky, M.F., Weigel, D., 2004. The role of
JAGGED in shaping lateral organs. Development 131, 1101–1110. http://
dx.doi.org/10.1242/dev.00949.
Doyle, J.J., Doyle, J.L., 1987. A rapid DNA isolation procedure for small quantities of
fresh leaf tissue. Phytochem. Bull. Bot. Soc. Am. 19, 11–15.
Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and
high throughput. Nucl. Acids Res. 32, 1792–1797. http://dx.doi.org/10.1093/
nar/gkh340.
Farris, J.S., Källersjö, M., Kluge, A.G., Bult, C., 1995. Testing significance of
incongruence. Cladistics 10, 315–319. http://dx.doi.org/10.1111/j.1096-
0031.1994.tb00181.x.
Farris, J.S., Albert, V.A., Källersjö, M., Lipscomb, D., Kluge, A.G., 1996. Parsimony
jackknifing outperforms neighbor-joining. Cladistics 12, 99–124. http://
dx.doi.org/10.1111/j.1096-0031.1996.tb00196.x.
Franzke, A., German, D.A., Al-Shehbaz, I.A., Mummenhoff, K., 2009. Arabidopsis
family ties: molecular phylogeny and age estimates in Brassicaceae. Taxon 58,
425–437.
Franzke, A., Lysák, M.A., Al-Shehbaz, I.A., Koch, M.A., Mummenhoff, K., 2011.
Cabbage family affairs: the evolutionary history of Brassicaceae. Trend Pl. Sci.
16, 108–116. http://dx.doi.org/10.1016/j.tplants.2010.11.005.
German, D.A., Friesen, N., Neuffer, B., Al-Shehbaz, I.A., Hurka, H., 2009.
Contribution to ITS phylogeny of the Brassicaceae, withs pecial reference to
some Asian Taxa. Pl. Syst. Evol. 283, 33–56. http://dx.doi.org/10.1007/s00606-
009-0213-5.
German, D.A., Grant, J.R., Lysák, M.A., Al-Shehbaz, I.A., 2011. Molecular phylogeny
and systematics of the tribe Chorisporeae (Brassicaceae). Pl. Syst. Evol. 294, 65–
86. http://dx.doi.org/10.1007/s00606-011-0452-0.
Gilg, E., Muschler, R., 1909. Aufzählung aller zur Zeit bekannten südamerikanischen
Cruciferen. Bot. Jahrb. Syst. 42, 437–487.
Goloboff, P.A., Farris, J.S., Nixon, K., 2008. TNT, a free program for phylogenetics
analysis. Cladistics 24, 774–786. http://dx.doi.org/10.1111/j.1096-0031.2008.
00217.x.
Goodson, B.E., Rehman, S.K., Jansen, R.K., 2011. Molecular systematics and
biogeography of Descurainia (Brassicaceae) based on nuclear ITS and non-
coding chloroplast DNA. Syst. Bot. 36, 957–980. http://dx.doi.org/10.1600/
036364411X604976.
Hall, T.A., 1999. Bioedit: a user-friendly biological sequence alignment editor and
analysis program for Windows 95/98/NT. Nucl. Acids Symp. Ser. 41, 95–98.
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). Am. J. Bot. 98, 1989–2003. http://dx.doi.org/10.3732/
ajb.1100203.
Hamilton, M.B., 1999. Four primer pairs for the amplification of chloroplast
intergenic regions with intraspecific variation. Mol. Ecol. 8, 521–523. http://
dx.doi.org/10.1046/j.1365-294x.1999.00510.x.
Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: paleontological statistics
software package for education and data analysis. Palaeontol. Electron. 4 (1), 1–
9.
Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G., Jarvis, A., 2005. Very high
resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25,
1965–1978. http://dx.doi.org/10.1002/joc.1276.
Huson, D.H., Bryant, D., 2006. Application of phylogenetic networks in evolutionary
studies. Mol. Biol. Evol. 23, 254–267. http://dx.doi.org/10.1093/molbev/msj030.
Jordon-Thaden, I., Hase, I., Al-Shehbaz, I.A., Koch, M.A., 2010. Molecular phylogeny
and systematics of the genus Draba s.l. (Brassicaceae) and identification of its
closest related genera. Mol. Phylogenet. Evol. 55, 524–540. http://dx.doi.org/
10.1016/j.ympev.2010.02.012.
Karl, R., Koch, M.A., 2013. A world-wide perspective on crucifer speciation and
evolution: phylogenetics, biogeography and trait evolution in tribe Arabideae.
Ann. Bot. 112, 983–1001. http://dx.doi.org/10.1093/aob/mct165.
Kelchner, S.A., 2000. The evolution of non-coding chloroplast DNA and its
application in plant systematics. Ann. Mo. Bot. Gard. 87, 482–498. http://
dx.doi.org/10.2307/2666142.
Khanna, K.R., Rollins, R.C., 1965. A taxonomic revision of Cremolobus (Cruciferae).
Contrib. Gray Herb. Harvard Univ. 195, 135–157.
Khosravi, A.R., Mohsenzadeh, S., Mummenhoff, K., 2007. Phylogenetic position of
Brossardia papyracea (Brassicaceae) based on sequences of nuclear ribosomal
DNA. Feddes Repert. 119, 13–23. http://dx.doi.org/10.1002/fedr.200811146.
Khosravi, A.R., Mohsenzadeh, S., Mummenhoff, K., 2008. Phylogenetic position of
Acanthocardamum erinaceum (Boiss) Thell. (Brassicaceae) based on sequences of
nuclear ribosomal DNA. Nordic J. Bot. 26, 25–30. http://dx.doi.org/10.1111/
j.0107-055X.2008.00205.x.
Khosravi, A.R., Jacquemoud, F., Mohsenzadeh, S., Menke, M., Mummenhoff, K.,
2009a. Phylogenetic position and taxonomic classification of Aethionema
trinervium (Brassicaceae): a morphologically variable subshrub from
southwestern Asia. Ann. Missouri Bot. Gard. 96, 564–574. http://dx.doi.org/
10.3417/200.
Khosravi, A.R., Mohsenzadeh, S., Mummenhoff, K., 2009b. Phylogenetic
relationships of Old World Brassicaceae from Iran based on nuclear ribosomal
DNA sequences. Biochem. Syst. Ecol. 37, 106–115. http://dx.doi.org/10.1016/
j.bse.2009.01.010.
Kiefer, M., Schmickl, R., German, D.A., Mandáková, T., Lysak, M.A., Al-Shehbaz, I.A.,
Franzke, A., Mummenhoff, K., Stamatakis, A., Koch, M.A, 2014. BrassiBase:
introduction to a novel knowledge database on brassicaceae evolution. Plant
Cell Physiol. 55, e3–e3 http://dx.doi.org/10.1093/pcp/pct158.
Koch, M.A., 2012. Mid-Miocene divergence of Ionopsidium and Cochlearia and its
impact on the systematics and biogeography of the tribe Cochlearieae
(Brassicaceae). Taxon 61, 76–92.
Koch, M., Mummenhoff, K., 2001. Thlaspi s.str. (Brassicaceae) versus Thlaspi s.l.:
Morphological and anatomical characters in the light of ITS nrDNA sequence
data. Pl. Syst. Evol. 227, 209–225. http://dx.doi.org/10.1007/s006060170049.
Koch, M., Al-Shehbaz, I.A., Mummenhoff, K., 2003. Molecular systematics, evolution,
and population biology in the mustard family (Brassicaceae). Ann. Missouri Bot.
Gard. 90, 151–171. http://dx.doi.org/10.2307/3298580.
Koch, M.A., Dobeš, C., Kiefer, C., Schmickl, R., Klimeš, L., Lysák, M.A., 2007.
Supernetwork identifies multiple events of plastid trnF pseudogene evolution
in the Brassicaceae. Mol. Biol. Evol. 24, 63–73. http://dx.doi.org/10.1093/
molbev/msl130.
Koch, M.A., Karl, R., Kiefer, C., Al-Shehbaz, I.A., 2010. Colonizing the American
continent: systematics of the genus Arabis in North America (Brassicaceae). Am.
J. Bot. 97, 1040–1057. http://dx.doi.org/10.3732/ajb.0900366.
Koch, M.A., Wernisch, M., Schmickl, R., 2008. Arabidopsis thaliana wild relatives: an
updated overview on systematics, taxonomy, and evolution. Taxon 57, 943–
993.
Koch, M., Kiefer, M., German, D.A., Al-Shehbaz, I.A., Franzke, A., Mummenhoff, K.,
Schmickl, R., 2012. BrassiBase: tools and biological resources to study
characters and traits in the Brassicaceae – version 1.1. Taxon 61, 1001–1009.
Körner, C., 1995. Alpine plant diversity. In: Chapin, F.S., Körner, C. (Eds.), Arctic and
Alpine Biodiversity Ecological Studies, vol. 113. Springer-Verlag, Berlin, pp. 45–
62. http://dx.doi.org/10.1007/978-3-642-78966-3_4.
Liu, N., Sliwinski, M.K., Correa, R., Baum, D.A., 2011. Possible contributions of
TERMINAL FLOWER 1 to the evolution of rosette flowering in Leavenworthia
(Brassicaceae). New Phytol. 189, 616–628. http://dx.doi.org/10.1111/j.1469-
8137.2010.03511.x.
Moazzeni, H., Zarre, S., Al-Shehbaz, I.A., Mummenhoff, K., 2010. Phylogeny of Isatis
(Brassicaceae) and allied genera based on ITS sequences of nuclear ribosomal
DNA and morphological characters. Flora 205, 337–343. http://dx.doi.org/
10.1016/j.flora.2009.12.028.
 D.L. Salariato et al. / Molecular Phylogenetics and Evolution 82 (2015) 43–59
Norberg, M., Holmlund, M., Nilsson, O., 2005. The BLADE ON PETIOLE genes act
redundantly to control the growth and development of lateral organs.
Development 132, 2203–2213. http://dx.doi.org/10.1242/dev.01815.
O’Kane, S.L., Al-Shehbaz, I.A., 2003. Phylogenetic position and generic limits of
Arabidopsis (Brassicaceae) based on sequences of nuclear ribosomal DNA. Ann.
Missouri Bot. Gard. 90, 603–612. http://dx.doi.org/10.2307/3298545.
Pagel, M., 1994. Detecting correlated evolution on phylogenies: a general method
for comparative analysis of discrete characters. Proc. R. Soc. Lond. B Biol. Sci.
225, 37–45. http://dx.doi.org/10.1098/rspb.1994.0006.
Pagel, M., 1997. Inferring evolutionary processes from phylogenies. Zool. Scr. 26,
331–348. http://dx.doi.org/10.1111/j.1463-6409.1997.tb00423.x.
Pagel, M., 1999. The maximum likelihood approach to reconstructing ancestral
character states of discrete characters on phylogenies. Syst. Biol. 48, 612–622.
http://dx.doi.org/10.1080/106351599260184.
Pagel, M., Meade, A., Barker, D., 2004. Bayesian estimation of ancestral character
states on phylogenies. Syst. Biol. 53, 673–684. http://dx.doi.org/10.1080/
10635150490522232.
Pagel, M., Meade, A., 2013. BayesTraits version 2. Software <http://
www.evolution.rdg.ac.uk/BayesTraits.html>.
Penin, A.A., 2008. Bract reduction in Cruciferae: possible genetic mechanisms and
evolution. Wulfenia 15, 63–73.
Phillips, S.J., Dudík, M., 2008. Modeling of species distributions with Maxent: new
extensions and a comprehensive evaluation. Ecography 31, 161–175. http://
dx.doi.org/10.1111/j.0906-7590.2008.5203.x.
Rambaut A, Suchard M.A., Xie, D., Drummond, A.J., 2013. Tracer v1.6.0 <http://
beast.bio.ed.ac.uk/>.
Raunkiær, C., 1934. The Life Forms of Plants and Statistical Plant Geography; Being
the Collected Papers of C. Clarendon Press, Oxford, Raunkiaer.
Ratcliffe, O.J., Bradley, D.J., Coen, E.S., 1999. Separation of shoot and floral identity in
Arabidopsis. Development 126, 1109–1120.
Rešetnik, I., Satovic, Z., Schneeweiss, G.M., Liber, Z., 2013. Phylogenetic relationships
in Brassicaceae tribe Alysseae inferred from nuclear ribosomal and chloroplast
DNA sequence data. Mol. Phylogenet. Evol., 69: 772–786. http://dx.doi.org/
10.1016/j.ympev.2013.06.026.
Rollins, R.C., 1948. On two perennial caespitose Lepidiums of western North
America. Madroño 9, 162–165.
Rollins, R.C., 1955. A revisionary study of the genus Menonvillea (Cruciferae).
Contrib. Gray Herb. Harvard Univ. 177, 3–57.
Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., Larget,
B., Liu, L., Suchard, M.A., Huelsenbeck, J.P., 2012. MrBayes 3.2: efficient Bayesian
phylogenetic inference and model choice across a Large Model Space. Syst. Biol.
61, 539 542. http://dx.doi.org/10.1093/sysbio/sys029.
Salariato, D.L., Zuloaga, F.O., Al-Shehbaz, I.A., 2013a. Revision and tribal placement
of the Argentinean genus Parodiodoxa (Brassicaceae). Pl. Syst. Evol. 299, 305–
316. http://dx.doi.org/10.1007/s00606-012-0722-5.
Salariato, D.L., Zuloaga, F.O., Al-Shehbaz, I.A., 2013b. Molecular phylogeny of
Menonvillea and recognition of the new genus Aimara (Brassicaceae:
Cremolobeae). Taxon 62, 1220–1234. http://dx.doi.org/10.12705/626.6.
Salariato, D.L., Zuloaga, F.O., Al-Shehbaz, I.A., 2014. A revision of the genus
Menonvillea (Cremolobeae, Brassicaceae). Phytotaxa 162 (5), 241–298. http://
dx.doi.org/10.11646/phytotaxa.162.5.1.
Schulz, O.E. 1936. Cruciferae. In: Harms, H. (Ed.), Nat. Pflanzenfam, Ed. 2, 17B.
Wilhelm En-gelmann, Leipzig, pp. 227–658.
Schultz, E.A., Haughn, G.W., 1993. Genetic analysis of the floral initiation process
(FLIP) in Arabidopsis. Development 119 (3), 745–765.
Shaw, J., Lickey, E.B., Beck, J.T., Farmer, S.B., Liu, W., Miller, J., Small, R.L., 2005. The
tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA
sequences for phylogenetic analysis. Am. J.Bot. 92, 142–166. http://dx.doi.org/
10.3732/ajb.92.1.142.
Shimodaira, H., Hasegawa, M., 1999. Multiple comparisons of log-likelihoods with
applications to phylogenetic inference. Mol. Biol. Evol. 16, 1114–1116.
Simmons, M.P., Ochoterena, H., 2000. Gaps as characters in sequence-based
phylogenetic analyses. Syst. Biol. 49, 369–381. http://dx.doi.org/10.1093/
sysbio/49.2.369.
Skottsberg, C.J.F., 1916. Botanische Ergebnisse der Schwedischen Expedition Nach
Patagonie und dem Feuerlande 1907–1909, V. Die Vegetationsverhältnisse
View publication stats
59
Längs der Cordillera de los Andes S. von 41° S Br. Ein Beitrag Zur Kenntnis der
Vegetation on Chiloé, West-Patagonien, dem Andinen Patagonien und Feurland.
Kungl. Svenska vetenskapsakademiens handlingar 56(5), 1–366.
Small, R.L., Ryburn, J.A., Cronn, R.C., Seelanan, T., Wendel, J.F., 1998. The tortoise and
the hare: choosing between noncoding plastome and nuclear Adh sequences for
phylogeny reconstruction in a recently diverged plant group. Am. J. Bot. 85,
1301–1315.
Stamatakis, A., 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic
analyses with thousands of taxa and mixed models. Bioinformatics 22, 2688–
2690. http://dx.doi.org/10.1093/bioinformatics/btl446.
Stamatakis, A., Hoover, P., Rougemont, J., 2008. A rapid bootstrap algorithm for the
RAxML web-servers. Syst. Biol. 57, 758–771. http://dx.doi.org/10.1080/
10635150802429642.
Swofford, D.L., 2003. PAUP⁄: Phylogenetic Analysis using Parsimony (⁄and Other
Methods), Version 4. Sinauer Associates, Sunderland, MA.
Taberlet, P., Gielly, L., Pautou, G., Bouvet, J., 1991. Universal primers for
amplification of three non-coding regions of chloroplast DNA. Pl. Mol. Biol.
17, 1105–1109. http://dx.doi.org/10.1007/BF00037152.
Toro-Nuñez, O., Mort, M.E., Ruiz-Ponce, E., Al-Shehbaz, I.A., 2013. Phylogenetic
relationships of Mathewsia and Schizopetalon (Brassicaceae) inferred from
nrDNA and cpDNA regions: taxonomic and evolutionary insights from an
Atacama Desert endemic lineage. Taxon 62, 343–356. http://dx.doi.org/
10.12705/622.4.
Urban, L., Bailey, C.D., 2013. Phylogeny of Leavenworthia and Selenia (Brassicaceae).
Syst. Bot. 38, 723–736. http://dx.doi.org/10.1600/036364413X670322.
Warwick, S.I., Hall, J.C., 2009. Phylogeny of Brassica and wild relatives. In: Gupta, S.
(Ed.), Biology and Breeding of Crucifers. CRC Press, New York, pp. 19–36.
Warwick, S.I., Al-Shehbaz, I.A., Price, R.A., Sauder, C.A., 2002. Phylogeny of
Sisymbrium (Brassicaceae) based on ITS sequences of nuclear ribosomal DNA.
Canad. J. Bot. 80, 1002–1017. http://dx.doi.org/10.1139/b02-089.
Warwick, S.I., Al-Shehbaz, I.A., Sauder, C.A., 2006. Phylogenetic position of Arabis
arenicola and generic limits of Eutrema and Aphragmus (Brassicaceae) based on
sequences of nuclear ribosomal DNA. Canad. J. Bot. 84, 269–281. http://
dx.doi.org/10.1139/b05-161.
Warwick, S.I., Sauder, C.A., Al-Shehbaz, I.A., Jacquemoud, F., 2007. Phylogenetic
relationships in the tribes Anchonieae, Chorisporeae, Euclidieae, and
Hesperideae (Brassicaceae) based on nuclear ribosomal ITS DNA sequences.
Ann. Missouri Bot. Gard. 94, 56–78. http://dx.doi.org/10.3417/0026-
6493(2007)94[56:PRITTA]2.0.CO;2.
Warwick, S.I., Sauder, C.A., Al-Shehbaz, I.A., 2008. Phylogenetic relationship in the
tribe Alysseae (Brassicaceae) based on nuclear ribosomal ITS DNA sequences.
Canad. J. Bot. 86, 315–336. http://dx.doi.org/10.1139/B08-013.
Warwick, S.I., Sauder, C.A., Mayer, M.S., Al-Shehbaz, I.A., 2009. Phylogenetic
relationships in the tribes Schizopetaleae and Thelypodieae (Brassicaceae)
based on nuclear ribosomal ITS region and plastid ndhF DNA sequences. Botany
87, 961–985. http://dx.doi.org/10.1139/B09-051.
Warwick, S.I., Mummenhoff, K., Sauder, C.A., Koch, M.A., Al-Shehbaz, 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.
http://dx.doi.org/10.1007/s00606-010-0271-8.
Warwick, S.I., Sauder, C.A., Al-Shehbaz, I.A., 2011. Systematic position of Ivania,
Scoliaxon, and Phravenia (Brassicaceae). Taxon 60, 1156–1164.
Weigel, D., Alvarez, J., Smyth, D.R., Yanofsky, M.F., Meyerowitz, E.M., 1992. LEAFY
controls floral meristem identity in Arabidopsis. Cell 69 (5), 843–859.
Went, F.W., 1971. Parallel evolution. Taxon 20, 197–226. http://dx.doi.org/10.1016/
0092-8674(92)90295-N.
Yoon, H.S., Baum, D.A., 2004. Transgenic study of parallelism in plant morphological
evolution. Proc. Natl. Acad. Sci. U.S.A. 101, 6524–6529. http://dx.doi.org/
10.1073/pnas.0401824101.
Yue, J.P., Sun, H., Baum, D.A., Li, J.H., Al-Shehbaz, I.A., Ree, R., 2009. Molecular
phylogeny of Solms-laubachia (Brassicaceae) s.l., based on multiple nuclear and
plastid DNA sequences, and its biogeographic implications. J. Syst. Evol. 47,
402–415. http://dx.doi.org/10.1111/j.1759-6831.2009.00041.x.
Zhao, B., Liu, L., Tan, D., Wang, J., 2010. Analysis of phlogenetic relationships of
Brassicaceae species based on Chs sequences. Biochem. Syst. Ecol. 38, 731–739.
http://dx.doi.org/10.1016/j.bse.2010.06.003.