Taxonomy and Biogeography of Apomixis in Angiosperms and Associated Biodiversity ...

14 Pages·2014·0.83 MB·English

CriticalReviewsinPlantSciences,33:414–427,2014
PublishedwithlicensebyTaylor&Francis
ISSN:0735-2689print/1549-7836online
DOI:10.1080/07352689.2014.898488
Taxonomy and Biogeography of Apomixis in Angiosperms
and Associated Biodiversity Characteristics
DiegoHojsgaard,1 SimoneKlatt,1 RolandBaier,2JohnG.Carman,3
andElviraHo¨randl1
1GeorgAugustUniversityGo¨ttingen,Albrecht-von-HallerInstituteforPlantSciences,Department
ofSystematicBotany,Go¨ttingen,Germany
2Gesellschaftfu¨rwissenschaftlicheDatenverarbeitungmbHGo¨ttingen(GWDG),Arbeitsgruppe
Anwendungs-undInformationssysteme,Go¨ttingen,Germany
3Plants,SoilsandClimateDepartment,UtahStateUniversity,Logan,UT,USA
TableofContents
I. INTRODUCTION .................................................................................................................................................................................................415
II. ASSESSMENTOFOCCURRENCES.........................................................................................................................................................417
A. DataCollectionandDatabaseConstruction ..........................................................................................................................................417
B. DataAnalyses ...................................................................................................................................................................................................417
III. PHYLOGENETICDISTRIBUTIONS ........................................................................................................................................................419
A. NovelRecords ..................................................................................................................................................................................................419
B. ApomixisinGenera,FamiliesandOrders .............................................................................................................................................420
C. FamilyDiversityandApomixis .................................................................................................................................................................421
D. BiogeographyandApomixis .......................................................................................................................................................................421
IV. APOMIXISANDEVOLUTION ....................................................................................................................................................................422
A. PatternsatHigherTaxonomicLevels.......................................................................................................................................................422
B. Apomixis-sexSystems ..................................................................................................................................................................................422
C. ApomixisBiodiversityParallelsGeneralBiodiversity ......................................................................................................................424
D. TheExpandedTransitionTheory ..............................................................................................................................................................424
V. PERSPECTIVESFORFUTURERESEARCH......................................................................................................................................425
ACKNOWLEDGMENTS ...............................................................................................................................................................................................425
FUNDING ..............................................................................................................................................................................................................................425
REFERENCES ....................................................................................................................................................................................................................425
©DiegoHojsgaard,SimoneKlatt,RolandBaier,JohnG.Carman,andElviraHo¨randl
AddresscorrespondencetoElviraHo¨randl,GeorgAugustUniversityGo¨ttingen,Albrecht-von-HallerInstituteforPlantSciences,Department
ofSystematicBotany,UntereKarspu¨le2,D37073Go¨ttingen,Germany.E-mail:[email protected]
Colorversionsofoneormoreofthefiguresinthearticlecanbefoundonlineatwww.tandfonline.com/bpts.
414
APOMIXIS,TAXONOMICDIVERSITY,ANDBIOGEOGRAPHY 415
sis or mitotic-like division, resulting in an unreduced func-
Apomixis in angiosperms is asexual reproduction from seed. tionalspore(diplospory).Ingametophyticapomicts,endosperm
Its importance to angiospermous evolution and biodiversity has formation may require fertilization of polar nuclei (pseudoga-
been difficult to assess mainly because of insufficient taxonomic mousapomixis)oritmaydevelopindependently(autonomous
documentation.Thus,weassembledliteraturereportingapomixis
apomixis).Adventitiousembryosofsporophyticapomictsform
occurrencesamongangiospermsandtransferredtheinformation
from nucellar or integumentary cells, and their formation usu-
to an internet database (http://www.apomixis.uni-goettingen.de).
Wethensearchedforcorrelationsbetweenapomixisoccurrences ally occurs in parallel with the formation of sexual embryos.
andwell-establishedmeasuresoftaxonomicdiversityandbiogeog- Bothsexualandadventitiousembryonyrequireendosperm,the
raphy.Apomixiswasfoundtobetaxonomicallywidespreadwith formation of which usually involves fertilization of the polar
no clear tendency to specific groups and to occur with sexuality
nuclei.Speciesexhibitingadventitiousembryonyoftenproduce
atalltaxonomiclevels.Adventitiousembryonywasthemostfre-
multiple embryos per seed. Hence, it is considered a form of
quentform(148genera)followedbyapospory(110)anddiplospory
(68).Allthreeformsarephylogeneticallyscattered,butthisscat- polyembryony,anditiswidespreadamongtropicalplants(Nau-
teringisstronglyassociatedwithmeasuresofbiodiversity.Across mova,1992).Polyembryony,however,canalsoarisefromsex-
apomictic-containing orders and families, numbers of apomict- ualpathways,e.g.fromrarecasesoffunctionaltwinmegasporo-
containing genera were positively correlated with total numbers
cytes(MMC)orbyfertilizationofasynergidthathasassumed
of genera. In general, apomict-containing orders, families, and
egg-likeproperties.
subfamiliesofAsteraceae,Poaceae,andOrchidaceaewerelarger,
i.e., they possessed more families or genera, than non-apomict- Gametophyticapomicts,exceptinafewcases,arepolyploid,
containing orders, families or subfamilies. Furthermore, many andtheyareoftenrelatedtospeciesthatexhibitotherdevelop-
apomict-containinggenerawerefoundtobehighlycosmopolitan. mental abnormalities such as polyspory (Carman, 1997). Ga-
Inthisrespect,62%occupymultiplegeographiczones.Numbers
metophytic and sporophytic apomixis are heritable, but they
of genera containing sporophytic or gametophytic apomicts de-
aregenerallyexpressedfacultatively,inindividualplants,along
creasedfromthetropicstothearctic,atrendthatparallelsgeneral
biodiversity.Whileangiospermsappeartobepredisposedtoshift with sex (Ozias-Akins and van Dijk, 2007). Confusion aris-
fromsextoapomixis,thereisalsoevidenceofreversionstosexu- ingfromfacultativenessandtheexistenceofmultipleapomixis
ality.Suchreversionsmayresultfromgeneticorepigeneticdesta- pathways has hampered the gathering of reliable quantitative
bilizationeventsaccompanyinghybridization,polyploidy,orother
data for relating apomixis to various evolutionary and biogeo-
cytogeneticalterations.Becauseofincreasedwithin-plantgenetic
graphicalphenomena.
andgenomicheterogeneity,rangeexpansionsanddiversifications
atthespeciesandgenuslevelsmayoccurmorerapidlyuponre- Apomixis and sexuality are not exclusive traits, as almost
versiontosexuality.Thesignificantly-enrichedrepresentationsof allapomicticplantsexhibitfacultativesexuality.Inaposporous
apomictsamonghighlydiverseandgeographically-extensivetaxa, plants, the sexual process in ovules generally aborts during
fromgeneratoorders,supportthisconclusion.
meiosis or early embryo sac formation, and it is replaced by
aposporous embryo sac formation, which occurs more or less
Keywords angiosperms,biodiversity,biogeography,onlinedatabase,
simultaneously. In some ovules, a reduced embryo sac forms,
evolution,polyploidy,phylogeny,geographicalpartheno-
andthesemayproducerecombinant1nembryosbyhaploidor
genesis
polyhaploid parthenogenesis or recombinant 2n embryos fol-
lowing fertilization. In other ovules, a reduced embryo sac
and one or more aposporous embryo sacs may form in par-
I. INTRODUCTION allel,buttheapomicticpathwayisoftenmoresuccessful(e.g.,
Apomixis in angiosperms is reproduction via asexually Hojsgaard et al., 2013). Nevertheless, polyembryony is com-
formed seed (Asker and Jerling, 1992). By excluding vege- moninaposporousplants,andoccasionallyanindividualovule
tative propagation per definition, apomixis is development of will contain sexually and aposporously-produced embryos. In
an embryo from an unreduced and unfertilized egg cell (ga- the case of diplospory, facultative sexuality occurs at lower or
metophytic apomixis) or from a somatic cell of the ovule higherfrequenciesduetocompletionofanormalsexualmeiosis
(sporophytic apomixis). Apomixis thus combines the benefits and embryo sac formation followed either by parthenogenesis
of seed dispersal with those of asexual reproduction (Mogie, andtheformationofarecombinant1nhaploid(orpolyhaploid)
1992).Thecellularmechanismsofapomixishavebeenstudied embryo or by fertilization and the formation of a recombinant
for over 100 years and have recently been reviewed (Ozias- 2nembryo(Aliyuetal.,2010).
Akins,2006;TuckerandKoltunow,2009;Rodriguez-Lealand In natural populations of apomicts, genetic studies using
Vielle-Calzada,2012).Inbrief,gametophyticapomixisinvolves charactercompatibilityanalysishaveconfirmedtheregularap-
development of an unreduced embryo sac (apomeiosis) and pearanceofrecombinantgenotypesandploidyvariants(Bashaw
formation of an embryo from the unreduced egg, within the etal.,1992;VanderHulstetal.,2000;Paunetal.,2006;Aliyu
embryo sac, without fertilization (parthenogenesis). The unre- et al., 2010; Paule et al., 2011; Cosendai et al., 2013). Such
duced embryo sac can develop from a somatic cell of the nu- variations suggest that the reproductive elements of sex and
cellus,whichreplacesthemegaspore(apospory).Alternatively, apomixis,i.e.,meiosis,syngamy,apomeiosisandparthenogene-
the megaspore mother cell may undergo a restitutional meio- sis,involvedifferentgeneticandepigeneticcontrolmechanisms
416 D.HOJSGAARDETAL.
that can be uncoupled. This phenomenon is well documented Facultative apomixis in perennials provides long-term re-
experimentally(Ogawaetal.,2013).Thus,apomeiosisfollowed productive stability for the colonization of large areas. Such
bysyngamyleadstoanincreaseinploidy(2n+n;B offspring), long-term stability may also provide time for multiple genetic
III
whilemeiosisfollowedbyparthenogenesisleadstoareduction andgenomicvariantstoformandbetested,e.g.,progenywith
inploidy(1n+0;haploidorpolyhaploidoffspring).Suchcases unique genetic recombinations and/or genomic translocations,
of partial apomixis and partial sex occur regularly in progeny inversions, aneuploidy, etc. Some of these could be sexual re-
testpopulations(BicknellandKoltunow,2004)andnaturalpop- vertants with normal Polygonum-type embryo sac formation
ulations(CosendaiandHo¨randl,2010;Pauleetal.,2011).The or with developmentally-unusual embryo sac formation, e.g.,
pointofinteresthereisthatsexualprocessesandvariouscom- polyspory.Afterreversiontoobligatesexuality,suchgenetically
binationsofsexualandapomicticprocessesoccurinfrequently enriched and recombined lineages may diversify more rapidly
toregularlyinmostapomicticspecies. thantheirobligately-sexualancestors(Carman,1997;Ho¨randl
ApomixiswasfirstdescribedinAntennariabyJuelin1898 and Hojsgaard, 2012). Hence, the breadth of distribution and
(Nogler, 2006). By 1941, apomixis had been reported in 44 richnessofgeneraandspeciesinfamiliescontainingapomicts
generafrom23families(Stebbins,1941).Gustafsson’sseminal couldreasonablyincreaseasadirectconsequenceofapomixis.
treatmentofapomixisinangiosperms(Gustafsson,1946;1947a; For this review, we have updated the apomixis occurrence lit-
b)listed73agamospermousgenerafrom25families(excluding eratureandhaverelatedittobiodiversityatvarioustaxonomic
viviparousspecies),anditprovidedcomprehensivedescriptions levels.Suchrelationshipshavenotpreviouslybeenextensively
of developmental pathways, evolutionary origins and ecologi- characterized.
calpreferences.AskerandJerling(1992)recognized108gen- Adeeperinterestinbiogeographicalandecologicalaspects
era in which apomixis occurs in at least one species (referred of apomixis emerged in the last decades of the twentieth cen-
tohereinasapomict-containinggenera),andtheylistedanad- tury. Consequently, it was found that some apomicts have
ditional 26 questionable records. Naumova (1992) recognized larger distributions in higher latitudes than their sexual rela-
116 genera with sporophytic apomixis, and Carman (1997) tivesandhaveabundantlypopulatedpreviouslyglaciatedareas
listed 222 genera that contain sporophytic or gametophytic (Bierzychudek, 1985; Kearney, 2005; Ho¨randl, 2006; Ho¨randl
apomicts. etal.,2008).Theterm“geographicalparthenogenesis”isused
The phylogenetic distributions of apomicts and their evo- todescribethisphenomenon,butithasbeenstudiedextensively
lutionary and biogeographical relevance remain unclear. The inonlyafewgenera(Ho¨randletal.,2008).Formostapomic-
doomed view of Darlington (Darlington, 1939), Stebbins tictaxa,informationondistributionpatternshasnotpreviously
(Stebbins, 1950) and Grant (Grant, 1981), that apomixis is an been assembled or remains too scarce to allow for generaliza-
evolutionary dead end, was based on the assumption that loss tions.Intropicalapomicts,biogeographicalpatternsofapomic-
ofgenotypeheterogeneityinpopulationswouldresultinlossof ticplantsarelargelyunexplored,andrecordsofpolyembryony
potential to adapt to environmental change. Thus, agamic lin- withoutdetailedinformationonapomicticbiotypeshaveblurred
eageswerethoughttobedoomedforextinction.Thisview,ofa the patterns (Carman, 1997). Superior colonizing ability and
static,closedsystem,hasbeenrevisedbymorerecenttheoret- polyploidy can make apomicts formidable invaders (Richards,
ical papers (Carman, 1997; Whitton et al., 2008; Ho¨randl and 2003),ashasoccurredonallcontinents(Chapmanetal.,2003;
Hojsgaard, 2012), which are based on empirical observations Brock, 2004; Rambuda and Johnson, 2004; Hao et al., 2011).
that agamic complexes harbor considerable genetic variability Butgeneralizingtheseobservationstothemajorityofapomicts
(Grant,1981;Ho¨randlandPaun,2007)andrepresentdynamic is premature. The necessary biogeographical surveys simply
andflexiblesystems. havenotbeenconducted.
Three large families, Poaceae, Asteraceae, and Rosaceae, Hereinweprovideacomprehensivesurveyoftheliterature
contain a majority of the known apomict-containing genera, that describes the occurrence and biogeography of apomict-
and this has led to speculations that a developmental predis- containing genera, and we discuss hypotheses relating phylo-
position may exist in these families or that apomixis is more geniesandbiogeographicalpatternstobiodiversityatmultiple
likelytoemergeinlargefamilies(Richards,1997;Ozias-Akins taxonomiclevels.Specifically,wetestedwhetherangiosperms
and van Dijk, 2007). It has also been postulated that a rapid in general are predisposed to apomixis or whether a predispo-
spread of apomixis in major clades could account for the ob- sitionstatisticallyfavorscertainfamilies.Further,weexamined
servedpattern(vanDijkandVijverberg,2005).Recentphyloge- frequencies of the three major forms of apomixis (apospory,
neticreconstructionsofthemajorclades(orders)ofangiosperms diplospory,adventitiousembryony)andrelatedthesetoputative
(Ho¨randl and Hojsgaard, 2012) support the view of a broadly taxonomicandbiogeographicalpreferences.Wealsotestedfor
scattered distribution of apomixis over the entire phylogeny positive correlations between apomixis frequencies and biodi-
of angiosperms coupled with rapid spread in large families. versityacrossapomictandnon-apomict-containingsisterclades
However, correlation analyses have not previously been con- among orders and families. Such correlations are expected if
ductedbetweenapomixisoccurrencesandgeneralmeasuresof highly-successful apomixis-to-sex reversals occur. Finally, we
biodiversity. report frequencies of apomict-containing genera that occupy
APOMIXIS,TAXONOMICDIVERSITY,ANDBIOGEOGRAPHY 417
multiple geographic zones and discuss hypotheses for inter-
preting the observed patterns. The database constructed for
this review is publically available (http://www.apomixis.uni-
goettingen.de)andincludeslistsofapomict-containinggenera,
typesofapomixisfound,andoriginalreferences.
II. ASSESSMENTOFOCCURRENCES
A. DataCollectionandDatabaseConstruction
We updated the occurrence of apomixis in angiosperms as
representedinCarman(1997)bysearchingtheWebofKnowl-
edge (Thomson Reuters (http://www.webofknowledge.com/),
ScientificCommons(St.Gallen,InstituteforMediaandCom-
munications Management, http://en.scientificcommons.org/),
Redalyc (Universidad Auto´noma del Estado de Me´xico,
http://www.redalyc.org/home.oa, version 2.0© 2012), Ingenta
Connect(http://www.ingentaconnect.com/),andTheReference
in Scientific Document Supply (French National Center for
ScientificResearch,CNRS,http://www.refdoc.fr/).Articlesnot
inthepublicdomainwereaccessedthroughtheLibraryofthe
University of Goettingen, the Library of the University of Vi-
enna, or the Merrill Library, Utah State University. Literature
was searched through December, 2012. Records were main-
tained at the genus level. For records originating before 1997,
onlythefirstdocumentationofapomixispergenusisreported,
andonlyrecordswithsufficientreliabilitywereaccepted.Cri-
teriaforacceptanceincludedverificationbymicroscopicinves-
tigationsofembryosacand/orembryodevelopment,flowcyto-
metricseedscreening(Matzketal.,2000),molecularprogeny FIG.1. Apomixisamongordersofangiosperms.Ovalsindicatesisterclades
tests, or well documented isolation and emasculation experi- containingandnotcontainingapomixis;filledovalsindicatepolytomies.(A)
apospory.(D)diplospory.(E)adventitiousembryony.ExpandedfromHo¨randl
ments.Preliminaryrecordsbasedonindirectobservation(e.g.,
andHojsgaard(2012).
clonality in wild populations, lack of pollen tubes on stig-
mas,etc.)wereincludedasuncertain,butthesewereexcluded
fromstatisticalinvestigations.Recordsofpolyembryonywith- B. DataAnalyses
outproofofadventitiousembryonyarereportedseparately.We Frequencies of apomict-containing families and genera
didnotincludereportsofoccasionalhaploidparthenogenesisin were compared across orders. In modern systematics, these
sexualplants.RecordsforAsteraceaerejectedbyNoyes(2007) taxonomic units represent almost exclusively clades, i.e.,
were included in the uncertain category to encourage detailed they are phylogenetic groups reflecting common diversifi-
re-investigation.Weconsiderallangiospermousapomictstobe cation processes and representing hierarchically-nested enti-
facultative,withthepossibleexceptionofAlchemilla(Askerand ties. We refrained from species-level taxonomy because of
Jerling, 1992), which means both sexually and apomictically- (i) scarcity of information concerning apomixis at the species
produced seeds can form on the same plant. Since apomixis level, and (ii) difficulties in defining species in agamic com-
as a reproductive character state is not exclusive, we refrained plexes where reticulate evolution and other speciation pro-
from standard phylogenetic tests of diversification, e.g., John- cesses are occurring (Grant, 1981; Ho¨randl, 1998). To pro-
sonetal.(2011).Allrecordsweretransferredtothetaxonomic vide an overview, we mapped records onto the phylogenetic
system of families and orders as in APG III (Bremer et al., treeoftheAPGIIIclassificationasinHo¨randlandHojsgaard
2009), and records published since 1997 were plotted by year (2012),andweaddedinformationconcerningtypesofapomixis
ofpublication. (Figure1).Moredetailedphylogeneticanalysesondiversifica-
ThereferencesreportedinCarman(1997)werere-examined tion rates (as in Johnson et al., 2011) are difficult to conduct
andtracedbacktotheoriginalliterature,andonlyoriginaldocu- because of facultativeness, which exists at the population and
mentsareincludedinourdatabase.Typesofapomixisexpressed individual levels. Sex and facultative apomixis simply are not
arealsoreported.RecordswereimportedintoOracle(cid:2)R,andthe exclusive binary character states. Hence, they cannot be cor-
online query application was generated using Oracle Applica- related with speciation and extinction rates on a phylogenetic
tionExpress(APEX). tree.
418 D.HOJSGAARDETAL.
To test the hypothesis that apomixis is causal to diver-
sity (Ho¨randl and Hojsgaard, 2012), we selected apomict and
non-apomict-containing sister clades in the phylogenetic tree
(Figure1).Incasesofpolytomies(seeFigure1),everycombi-
nationamongbrancheswasevaluated.AShapiro-Wilktestfor
normality was applied to numbers of genera and families, and
similaritiesinmediansweretestedusinganon-parametricone-
tailedWilcoxonMatched-PairsSigned-RankTestwherediffer-
ences in scores were assumed to be distributed symmetrically
aroundtheirmedian.Numbersofgeneraperfamilywereasin
Heywood(2007)exceptforminoradjustmentstoaccommodate
theAPGIIIsystem.AQ–Qplotcomparingnumbersofapomict
tonon-apomict-containinggeneraperordersuggestedthedata
arenotdistributednormally(X∼N(0,1)).Thehypothesisthat
FIG.2. Discoveriesofapomixisingenerawherepreviouslyithadnotbeen
apomixis occurs preferentially in certain orders was tested by
reported(1997–2012).
non-parametric Spearman’s rank correlation (r ). Numbers of
s
apomict-containinggeneraperorderwereplottedagainsttotal
numbers of genera per order for all 63 angiospermous orders. ofPoaceae(http://archive.is/4yN5E),andgenerainAsteraceae
Regressionmodelswereevaluated,andthreeresidualstatistics, andOrchidaceaewereaccordingtotheAngiospermPhylogeny
Mahalanobisdistance,Cook’sdistanceandStandarizedDFFit, Website (http://www.mobot.org/MOBOT/research/APweb/).
wereusedtodetectthepresenceofoutliersandtoexplaintheir PhylogenetictreesweremodifiedafterPaneroandFunk(2008)
influencesonmodelpredictability. for Asteraceae, the Grass Phylogeny Working Group (GPWG
Familiesweregroupedinto21categoriesbasedonnumbers II, 2012) for Poaceae, and Freudenstein et al. (2004) for
of genera per family (family richness). The percentage of Orchidaceae.
familiesineachcategorythatcontainapomictswasthencalcu- Biogeography information for apomict-containing genera
lated (apomixis presence), and family richness and apomixis was taken from Mabberley (2008) and Tropicos.org (Missouri
presencewerecomparedbySpearman’srankcorrelation.Three Botanical Garden, http://www.tropicos.org) with minor taxo-
families, Asteraceae (eudicot with gametophytic apomixis), nomic adjustments as described above. Distributions of these
Poaceae (monocot with gametophytic apomixis), and Or- genera were then classified according to floral zone (Meusel,
chidaceae (monocot with sporophytic apomixis), were then 1978)withthegoalofdeterminingwhethernumbersofapomict-
analyzed for intra-familial patterns of apomixis by regressing containing genera increase or decrease when moving from
numbers of apomict-containing genera per subfamily against warmertocolderclimates.Thisanalysiswasbasedonapomict-
total numbers of genera per subfamily. Genera in Poaceae containing genera per climatic zone only, not frequencies of
were according to A World-wide Phylogenetic Classification apomict-containinggenerarelativetototalnumbersofgenerain
TABLE 1
Newrecordsofapomict-containingfamiliesaddedsinceCarman(1997)
Family Order Genera Type Reference
Bignoniaceae Lamiales Tabebuia AE Saloma˜oandAllem,2001
Cannabaceae Rosales Humulus AorD Wettstein,1925
Erythroxylaceae Malpighiales Erythroxylum A Berryetal.,1991
Juglandaceae Fabales Juglans A SanandDumanoglu,2006;Guoliangetal.,2010
Lauraceae Laurales Lindera A Dupont,2002
Lecythidaceae Ericales Cariniana AE Saloma˜oandAllem,2001
Magnoliaceae Magnoliales Woonyoungia A Zengetal.,2003
Orobanchaceae Lamiales Orobanche,Cistanche D Jensen,1951;Pazy,1998
Phyllanthaceae Malpighiales Uapaca AE MaliroandKwapata,2000
Potamogetonaceae Alismatales Potamogeton D Teryokhinetal.,2002
Rubiaceae Gentianales Coprosma AorD Heenanetal.2002,2003
Sapindaceae Sapindales Magonia AE Saloma˜oandAllem,2001
Note.A=Apospory;D=Diplospory;AE=Adventitiousembryony.Forfullreferences,seetheapomixisdatabase(http://www.apomixis.uni-
goettingen.de).
APOMIXIS,TAXONOMICDIVERSITY,ANDBIOGEOGRAPHY 419
FIG.3. Percentagesofapomict-containingfamiliesandgenerabyorder.
therespectivezones.Thesixclimaticzonesincludedwere:arc- (http://www.apomixis.uni-goettingen.de). Total numbers of
tic(includingtheantarcticregion),boreal,temperate(including generavariedfrom12,758to13,150dependingoncircumscrip-
submeridional,warmtemperate,andaustralzones),meridional tion.Fifty-onegenerawereaddedtothelistofCarman(1997)
(includingtheMediterraneanregion),subtropical(includingN- asnewlyidentifiedapomicts(Figure2;67papers),andanother
orboreosubtropicalandS-oraustrosubtropicalzone),andtrop- 35 were identified that had been reported between 1918 and
ical. 1997butwerenotincludedinCarman(1997).Thus,86genera
Statistical analyses were performed using IBM(cid:2)R SPSS(cid:2)R have been added. These represent 12 new apomict-containing
Statistics, Version 20 for Windows. For all evaluated cases, families from among 10 orders (Table 1). According to our
significance was tested at the p ≤ 0.05 (significant) and updatedlist,adventitiousembryony,apospory,diplospory,and
p≤0.01(highlysignificant)levelsofprobability. apospory and diplospory together occur in 148, 110, 68, and
17 genera, respectively. Adventitious embryony and gameto-
phytic apomixis also occur together in some genera, i.e., they
III. PHYLOGENETICDISTRIBUTIONS
arenotexclusivetraits.Thetypeofapomixisisnotknownfor
A. NovelRecords afewgenerawhereexperimentalevidenceconfirmsapomixis,
Thirty-two orders (52%), 78 families (19%) and 293 andanadditionalfivegenera,resultsofwhichwerepublished
genera (≈2.2%) were found to contain apomictic species from1998-2012,expressdevelopmentalfeaturessuggestiveof
420 D.HOJSGAARDETAL.
allsevenmajorcladesofangiospermsandinmostlargeorders
(Figure1).Thesethreeformsofapomixisoccurin24,17,and28
orders, respectively, which parallels distributions at the family
andgenuslevels.Insomeclades,apomixisisreconstructedas
ancestral. But here, the trait is often lost on certain terminal
branches, e.g., Canellales, Arecales, Oxalidales, Cornales and
Garryales(Figure1).
B. ApomixisinGenera,FamiliesandOrders
Among the 32 apomict-containing orders, apomict-
containing families ranged from 6.25–100% with a mean of
28.3%(Figure3).Interestingly,thethreeorderscontainingthe
three most apomixis-populated families, Poaceae, Asteraceae,
andRosaceae,alsocontainmultiplefamiliesinwhichapomixis
has not been reported. To date, Poaceae and Asteraceae are
the only apomict-containing families in the Poales (16 fam-
ilies) and Asterales (11 families). Four of nine families in the
Rosalescontainapomicts.Orderswiththehighestfrequenciesof
apomict-containingfamiliesweresmall,e.g.,Dipsacales(both
families),Dioscoreales(2of3families),andSapindales(5of9
families).
FIG.4. Apomict-containinggeneraversustotalgenerabyorder.Twooutliers, Numbers of apomict-containing genera in apomict-
AandP,wereremovedfromthedatasetpriortopowerregression.(A)Asterales.
containing orders were positively correlated with total num-
(P)Poales.
bers of genera in the respective orders. This occurred regard-
less as to whether the total number of genera was considered
apomixisbutverificationstudieshavenotyetbeenconducted. to be 13,150 (Figure 4, r = 0.66, p < 0.01) or 12,758 (r
s s
Werecordedtheseinthedatabaseas“uncertain.” = 0.68, p < 0.01). However, frequency values (numbers of
Apomixis occurs scattered over the whole phylogeny and apomict-containinggeneradividedbytotalnumbersofgenera
appearsbothinbasalstemgroupsandinthederivedcladesof in the respective orders) were negatively correlated with to-
monocotsanddicots(Figure1).Adventitiousembryonyisthe tal numbers of genera (r = −0.513, p < 0.11). Two orders
s
predominantmodeofapomixisinfabids,malvids,andlamiids. werelargelyresponsibleforthisnegativecorrelation:Asterales,
Otherthanthis,clearphylogenetictendenciesarenotapparent, with 1829 genera but with only one of 11 families containing
i.e., apospory, diplospory and adventitious embryony occur in apomicts (1.5% of genera), and Poales, with 930 genera but
FIG.5. Familybiodiversityandapomixis.(grayline)Familycategoriesbasedonnumbersofgeneraperfamily.(blackline)Percentageoffamiliesineachgroup
thatcontainapomicticspecies.(dashedline)Percentageofangiospermousfamiliesthatcontainapomicts.
APOMIXIS,TAXONOMICDIVERSITY,ANDBIOGEOGRAPHY 421
with only one of 16 families containing apomicts (6.25% of
genera). In contrast, high frequencies of apomict-containing
genera per order were observed in small orders, e.g., Buxales
(16.7%) where six genera and 100 species are grouped into
twofamilies.Frequenciesofapomict-containinggenerainmost
orderswerelow,whichfitsapatternofabroadtaxonomicdis-
tributionforapomixisamongmanyordersratherthanaconcen-
trationofapomixisinfeworders.Spearman’srankcorrelations
betweennumbersofapomict-containinggeneraandtotalnum-
bersofgeneraamongapomict-containingordersrevealedsim-
ilarhighlysignificantcorrelationcoefficientsforgametophytic
(r =0.477,p<0.05)andsporophytic(r =0.475,p<0.05)
s s
apomixis.
AsshownbyQ-Qplotanalysisofthecompletedatasetand
Shapiro-Wilktestsofsisterorders,normalityatthe1–α=0.99
confidencelevel(SW=0.757;P<0.01)couldnotbeassumed.
Wilcoxon tests revealed that median values for families and
generaofsisterclades(W =3,Z=−2.8241,p=0.0024,N=
12)aresignificantlyhigherinapomicticclades,andtheydepict
anexceptionallylowprobabilityofrandomnessforthesevalues.
Hence,apomixisisstronglyassociatedwithbiodiversity.
C. FamilyDiversityandApomixis
Mostfamiliesofangiospermslackdiversity,e.g.,252fami-
lies(61%)containonly1–5genera.Inthisrespect,presenceof
apomixisinfamilieswaspositivelycorrelatedwithnumbersof
genera per family (Figure 5). Fifty percent of families with >
40generacontainedatleastoneapomict.Thisincreasedto63%
forfamilieswith>80genera,andallfamilieswith>250gen-
era contained apomicts. Likewise, case studies of three highly
diversefamiliesrevealedstrongcorrelationsbetweenpresence
of apomixis in intra-familial taxa and total numbers of genera
inthesetaxa(Figure6).However,thesecorrelationsdonottell
thewholestory.Averagelevelsofbiodiversityamongfamilies
are also important. For example, a genus selected at random
fromalargegroupofgeneracomposedofacombinationoflow
diversity families, e.g., 1–5 genera each, was less likely to be
apomicticthanifitwereselectedfromasimilar-sizedgroupof
generatakenfromamongfamiliesorsub-familiesthatcontain
manygenera.
D. BiogeographyandApomixis
Apomict-containinggeneraweredistributedworldwidewith
38, 13, and 11% occurring in 2, 3, and > 3 climatic zones,
respectively. They also tended to occur mostly in tropical to
temperate zones. Few apomict-containing genera were found
in boreal to arctic zones compared to lower latitude zones
(Figure 7A, B). Sporophytic and gametophytic apomixis ex-
hibited similar patterns of distribution across climatic zones
(parametric Spearman correlation r = 0.95, p < 0.05) with
FIG. 6. Apomixis in subfamilies. (A) Asteraceae. (B) Poaceae. (C)
sporophyticapomixisbeingslightlymoreabundantinthetrop-
Orchidaceae.
ics and gametophytic apomixis being slightly more abundant
elsewhere(Figure8).
422 D.HOJSGAARDETAL.
FIG.7. Geographicdistributionsofapomict-containinggenera.(A)Byclimaticzone.(B)Byoccupationofonetomultipleclimaticzones.
IV. APOMIXISANDEVOLUTION and it also involves the differential expression of gene fami-
liesthatregulatesexualdevelopment(Grimanelli,2012).Thus,
A. PatternsatHigherTaxonomicLevels
the totipotency and epigenetic flexibility of plant cells makes
Apomixisiswidespreadandscatteredtaxonomicallyamong
various routes toapomixis possible.Adirectintervention dur-
orders, and we found strong evidence that it has arisen multi-
ing meiosis, as occurs in diplospory, is perhaps more difficult
pletimeswithnocleartiestospecificplantfamilies.However,
toachieveinnaturalplantpopulations.Thus,diplosporyisthe
the phylogenetic tree (Figure 1) does suggest that apomixis is
leastfrequentpathway.TheartificialMiMesystemofproduc-
potentiallyancientinsomemajorcladesespeciallythoselead-
ing clonal embryos in seeds of Arabidopsis requires a com-
ing to the commelinids, fabids and lamiids. While sexuality is
bination of mutations in three meiosis genes (d’Erfurth et al.,
ancestralinangiosperms,howandwhenapomixisaroseinthis
2009). As pointed out earlier, such combinations of mutations
group is less certain. What we can say is that stable forms of
are unlikely to occur in natural populations because selec-
apomixis either repeatedly evolved de novo during angiosper-
tion would act against each intermediate step (Van Dijk and
mous evolution (hypothesis of multiple de novo origins) or its
Vijverberg, 2005). Meiosis itself is highly conserved in eu-
expressionintypicallysexuallineageswasrepeatedlytriggered
karyotes(Maliketal.,2007),andinfacultativelydiplosporous
during evolution (conserved pathways hypothesis). The latter
plants, steps of meiosis tend to be skipped rather than altered
suggests unreduced gamete formation coupled with partheno-
(Carman,1997).
genesis(basicapomixis)isanancientepigeneticcompetencyin-
Likeapomixisingeneral,apospory,diplosporyandadventi-
heritedfromangiospermousancestorswithinnovationsintypes
tiousembryonyareeachtaxonomicallywidespread(Figure1),
ofapomixisarisingthereafter(Carmanetal.,2011).Inthisre-
and their genetic or epigenetic regulatory mechanisms differ
spect, it is remarkable that adventitious embryony is the most
(Ozias-Akins and van Dijk, 2007). The hypothesis of a tax-
frequenttypeofapomixis.Thedirectdevelopmentofembryos
onomic predisposition for the evolution of apomixis within
has been reported from other plant tissues, e.g., from anthers
certain clades with apomixis-inducing genetic factors moving
(Sorianoetal.,2013),leaves,andotherorgans(Tisseratetal.,
rapidlywithintheseclades(VanDijkandVijverberg,2005)is
1979).Ingeneral,embryonycanbeactivatedinmanyplantcells
notsupportedbyourcompilation.Incontrast,ourdatasuggest
independentoffertilization(Carman,1990).
thatangiospermsingeneralhaveacapacitytoswitchfromsex
Thesecondaspectofgametophyticapomixis,i.e.,partheno-
to apomixis. But this capacity, particularly with regard to ga-
geneticdevelopmentofaneggcell,alsooccursspontaneously
metophyticapomixis,ismorepronouncedinlarge,diverseand
in otherwise sexual plants (Asker and Jerling, 1992). In this
genomically-complexclades(Figures5and6).
respect, methylation patterns and other epigenetic marks are
knowntoparticipateinregulatingembryoandendospermdevel- B. Apomixis-sexSystems
opment(LawandJacobsen,2010).Incontrastwithmammals, Thelowfrequencyofapomixisweobservedatthegenuslevel
whereaparentalcontributionisessentialforembryony,genomic (about 2%) is likely due to (i) under-sampling, as suggested
imprintinginplantsappearstoberequiredforendospermfor- by the linear increase in newly discovered apomicts per year
mationonly,andnotforembryoformation(Engelsta¨dter,2008). between1997and2012(Figure2)coupledwithmethodological
Independencefromapaternalepigenomiccontributionmayal- problemsofrecognizingapomixis(Leblanc,2000;Matzketal.,
low for more flexibility in the evolution of alternative asexual 2000;Ho¨randletal.,2008),(ii)naturalconstraintsthatinhibit
reproductivepathways. shifts from sexuality to apomixis (van Dijk and Vijverberg,
Strikingly, our data indicate that apospory is more frequent 2005;Engelsta¨dter,2008;Ho¨randletal.,2008;Ho¨randl,2009),
than diplospory. Apospory involves activation of gametophyte and (iii) spread of apomictic lineages followed by reversions
development in a nucellar cell rather than in the megaspore, to sexuality and further diversification. According to the latter
APOMIXIS,TAXONOMICDIVERSITY,ANDBIOGEOGRAPHY 423
hypothesized that recombination among facultative apomicts
coupled with wide hybridization increases genetic diversity
andadaptivepotential(Ho¨randlandPaun,2007).Furthermore,
apomictic taxa may diversify rapidly by accumulating muta-
tions that could lead to advantageous (e.g., Mark Welch and
Meselson, 2001; Pellino et al., 2013) as well as deleterious
(i.e.,Muller’sratchet)allelicdivergences.Partialapomixisadds
a further level of complexity, as combinations of apomeiosis
withsyngamyormeiosiswithparthenogenesismayrapidlyin-
creaseploidydiversity(BicknellandKoltunow,2004;Cosendai
and Ho¨randl, 2010; Paule et al., 2011). Such flexibility of de-
velopmental pathways begs the discussion of novel hypothe-
ses to explain putative relationships between apomixis and
biodiversity.
A model for apomixis as a facilitator of diversification
through cycles of polyploidy and reversions to sex was pro-
posed by Carman (1997). Ho¨randl and Hojsgaard (2012) ex-
panded on this idea by suggesting that apomictic polyploids
FIG.8. Numbersofgeneraexhibitingsporophyticorgametophyticapomixis serveaspioneerexplorersofnewnicheswherebytheyrapidly
byclimaticzone.Notethatsomegeneraspanmultipleclimaticzones,and32 expand the distribution areas of their progenitor sexual popu-
generaexhibitbothsporophythicplusgametophyticapomixis.
lations by occupying new ecological and geographical niches
(geographical parthenogenesis). Thereafter, reversals to com-
possibility, high taxa diversity coupled with the presence of plete sexuality, accompanied by and possibly caused by sub-
apomixismaybeanormalandexpecteddiversificationsignature stantialcytogeneticandgenomicperturbations(Carman,1997),
that results from cycles of neopolyploidy-originated apomixis allow for the establishment of new sexual populations in new
followedbyreversionstosexuality(Carman,1997;Ho¨randland habitatswithoutthelong-termdisadvantagesofasexuality.The
Hojsgaard,2012).Sincepoints(i)and(ii)havebeendiscussed newsexualrecombinants,potentiallynewspeciesorevennew
elsewhere,wediscussherethenovelhypothesisthatapomixis genera, are genetically isolated from the original sexual pop-
enhancesdiversification. ulations and predisposed to a divergent evolutionary destiny.
Diversification results from speciation minus extinction. If Empirical examples of reversals from apomixis to sex have
apomixisleadstohighratesofextinction,allelsebeingequal, been reported for Pennisetum (Taliaferro and Bashaw, 1966),
then taxa with apomixis should be less diverse than taxa Pilosella (Chapman et al., 2003) and Paspalum (Ortiz et al.,
without it. However, the apomixis occurrence data reviewed 2013), which supports the notion that sexuality is the default
hereinsuggesttheopposite.Familieswithapomixismusthave modeinfacultativeapomicts(Koltunowetal.,2011).Hence,re-
either extraordinarily high sexual speciation rates, to compen- versalsfromfacultativeapomixistoobligatesex,withaccompa-
sate for extinction rates, or apomixis somehow plays a pos- nyingexpansionsofspeciesdiversity,seemlikely.Thispossibil-
itive and previously-undetected role in evolution. The preva- ityhasbeenlargelyoverlookedbyevolutionarybiologists(Car-
lent distribution of apomixis among large angiospermous man,1997).Towhatextentitmayexplainancientdivergences
familiessuggestsapositiverole.Correlationsbetweennumbers withinlargeclades,whichtodaycontinuetomaintainhighfre-
ofapomict-containinggeneraandtotalnumbersofgeneraacross quenciesofapomixis,e.g.,asterids,rosidsandmonocots(Fig-
apomict-containing families (or sub-families) were high (Fig- ure1),isadifficultbutimportantquestionforfutureresearchto
ures5and6).Interestingly,asimilarcorrelationwasobserved address.
amongpolystichoidferns,i.e.,betweenspeciesrichnessandthe Many species of Oenothera mimic asexuality through sys-
occurrenceofapogamy(aposporyfollowedbyadventitiousem- temsofpermanent translocationheterozygosity, whichrestrict
bryonyfromanon-gameticcelloftheunreducedgametophyte). recombination to telomeres. Balanced lethals in these systems
However, apomixis in this group is not associated with higher enforceinbreedingthroughtheformationofunique,genetically
diversificationratesbutratherwiththereticulateevolutionand compatible male and female gametes. Recent studies suggest
establishmentofpolyploidhybrids(Liuetal.,2012). that functional asexuality in Oenothera coupled with rever-
The high frequencies of apomict-containing genera found sions to sexual reproduction accelerate species diversification
in genus-rich subfamilies (Figure 6) may occur simply be- (Johnson et al., 2011). Williams (1975) proposed a general
causeofincreasedopportunitiestoshifttoapomixis.Thus,we model explaining the effects of shifts between modes of re-
cannot infer that apomixis alone causes diversification. How- production. If shifts from asexuality to sex are more frequent
ever,acombinationofbothtraitscouldbesuperiortoobligate thanshiftsfromsextoasexuality,thenthefrequencyofsexat
sexuality (Cosendai et al., 2013). In this respect, it has been the tips of the phylogeny will be higher than the frequency of