Human paleogenetics of Europe - The known knowns and the known unknowns

20 Pages·3.251 MB·English

by  Brandt G. et al.

JournalofHumanEvolution79(2015)73e92
ContentslistsavailableatScienceDirect
Journal of Human Evolution
journal homepage: www.elsevier.com/locate/jhevol
Human paleogenetics of Europe e The known knowns and the known
unknowns
Guido Brandt a, Anna Sze(cid:1)cse(cid:1)nyi-Nagya,b, Christina Rotha, Kurt Werner Alt c,d,e,
Wolfgang Haak f,*
aInstituteofAnthropology,JohannesGutenbergUniversityMainz,Colonel-Kleinmannweg2,D-55099Mainz,Germany
bArchaeologicalInstitute,ResearchCentrefortheHumanities,HungarianAcademyofSciences,H-1014Budapest,Hungary
cStateOfficeforHeritageManagementandArchaeologySaxony-AnhaltandStateHeritageMuseum,Richard-Wagner-Straße9,D-06114Halle,Germany
dInstituteforPrehistoryandArchaeologicalScience,BaselUniversity,Petersplatz1,4003Basel,Switzerland
eDanubePrivateUniversity,FacultyofMedicineandDentistry,Doktor-Karl-Dorrek-Straße23,3500KremsanderDonau,Austria
fAustralianCentreforAncientDNA,TheUniversityofAdelaide,NorthTerraceCampus,SA-5005Adelaide,Australia
a r t i c l e i n f o a b s t r a c t
Articlehistory: ThenumberofancienthumanDNAstudieshasdrasticallyincreasedinrecentyears.Thisresultsina
Received4February2014 substantial record of mitochondrial sequences available from many prehistoric sites across Western
Accepted19June2014 Eurasia,butalsogrowingY-chromosomeandautosomalsequencedata.Wereviewthecurrentstateof
Availableonline13November2014 researchwithspecificemphasisontheHolocenepopulationeventsthatlikelyhaveshapedthepresent-
daygeneticvariationinEurope.Wereconcileobservationsfromthegeneticdatawithhypothesesabout
Keywords: thepeoplingandsettlementhistoryfromanthropologyandarchaeologyforvariouskeyregions,andalso
AncientDNA
discussthedatainlightofevidencefromrelateddisciplines,suchasmodernhumangenetics,clima-
Paleolithic
tologyandlinguistics.
Mesolithic
Neolithic ©2014ElsevierLtd.Allrightsreserved.
Archaeology
Introduction eventsfrompresent-daydistributionpatterns,aDNAstudieshave
theadvantageofprovidingdirectgeneticevidenceatagivenpoint
WithinthefieldofancientDNA(aDNA),considerablefocushas intime(eitherfromaradiocarbondateofthesampleitselforits
beenplacedonexploringthegeneticsofancienthomininsandthe stratigraphic layer and archaeological context). This allows us to
population history of anatomically modern humans. Despite the directly test hypotheses about the genetic affinity of ancient in-
controversialearlydaysofaDNAworkandinherentpitfalls,espe- dividualsand/orpastpopulationsasawhole.
cially when working with ancient human samples (Cooper and The temperate and cooler climate zones of the Americas and
Poinar, 2000; Pa€€abo et al., 2004; Gilbert et al., 2005), recent Europe are the two main areas best studied at this point, due to
studiesonprehistoricindividualshavenowadvancedsufficiently theirnaturalconditionsfavorableforDNApreservation.Thelarge
toproducereliableinsightsintospecimensfrompastpopulations number of aDNA laboratories, close and regional collaborations
by applying strict authentication criteria, including ‘DNA-free’ witharchaeologistsandanthropologists,andsignificantresources
samplinginthefield,dedicatedaDNAfacilities,aseriesofdecon- availabletostudythepeoplinghistoryoftheseregionscanalsobe
tamination procedures and experimental controls, and a hierar- listedasfactors.Thesedays,thenumberofaDNArecordsamounts
chical framework of independent replication of results (see e.g., toseveralhundredsofindividualsforbothAmericaandEuropeor
Gilbert et al., 2005, 2006; Malmstro€m et al., 2005; Deguilloux Eurasia. The paleogenetic recordofEuropehasbeen summarised
et al., 2011a, 2012; Brandt et al., 2013 for more details on this recently (e.g., Deguilloux et al., 2012; Pinhasi et al., 2012; Lacan
topic).IncontrasttomodernDNAstudies,whichcanonlyinferpast etal.,2013),andimplicationsforneighboringfieldssuchasphys-
icalanthropologyandzooarchaeology(consideringdomestication
events accompanying major cultural transitions) have been dis-
Abbreviations:aDNA,ancientDNA;calBC,calibratedyearsBeforeChrist;LBK, cussed(e.g.,Vigne,2011;vonCramon-TaubadelandPinhasi,2011;
LinearPotterycultureorLinearbandkeramischeKultur;mtDNA,mitochondrialDNA; PinhasiandvonCramon-Taubadel,2012;LarsonandBurger,2013).
PCA,principalcomponentanalysis. However,twomorerecentstudieshavesubstantiallysolidifiedthe
* Correspondingauthor.
observed population events forming the European genetic land-
E-mailaddress:[email protected](W.Haak).
URL:http://www.adelaide.edu.au/acad scape,resultinginafour-foldincreaseofaDNAdataavailablefrom
http://dx.doi.org/10.1016/j.jhevol.2014.06.017
0047-2484/©2014ElsevierLtd.Allrightsreserved.
74 G.Brandtetal./JournalofHumanEvolution79(2015)73e92
Europe, which warrants a fresh synthesis of the work to date Todate,83mtDNAsequencesfrom32hunter-gatherersitesof
(Fig. 1A). The first study has provided more data on Mesolithic Central, East, North, and South Europe have been published
hunter-gatherers and the complexity surrounding the Meso- (Caramelli et al., 2003; Chandler, 2003; Chandler et al., 2005;
Neolithic transition (Bollongino et al., 2013). The second one, a Bramantietal.,2009;Malmstro€metal.,2009;Krauseetal.,2010;
studybyourgroup,hasestablishedthefirstcomprehensivegenetic Hervellaetal.,2012;Sanchez-Quintoetal.,2012;Skoglundetal.,
transectthroughtimeinCentralGermany'sMittelelbe-Saaleregion 2012;Bollonginoetal.,2013;DerSarkissianetal.,2013;Fuetal.,
(Brandtetal., 2013), which canserveas ageneticreference stra- 2013;Lazaridisetal.,2014)(Fig.1A,sites1e30,63,80).However,
tigraphytoanchorandassesssinglesite/culturestudiesinadjacent someoftheseforagerpopulations,suchasthePittedWareculture
regions.Inshort,thisdetailedchronologicalstudyhasrevealeda (Malmstro€metal.,2009;Skoglundetal.,2012)ortheOstorfsitein
seriesofpopulationeventsinCentralEurope,whichinsumcould northernGermany(Bramantietal.,2009)werecontemporaneous
explainmostifnotallofthegeneticdiversityobservedtoday: with Neolithic cultures of the relevant regions and therefore
potentiallyinfluencedbyfarmingcommunities.Inthelattercase,
(cid:1) TheinitialpeoplingeventprovidingaPaleolithicandMesolithic archaeologicalresearchhassuggestedthattheOstorfpeoplewere
geneticsubstratum formerlyfarmerswhobecamefishermen(Lübkeetal.,2007).The
(cid:1) The impact of incoming farmers during the Early Neolithic 57 remaining sequences from confirmed Mesolithic contexts
(eventA) clearly cement the Upper Paleolithic and post-glacial Mesolithic
(cid:1) A period of genetic continuity followed by an assimilation of female substratum as predominantly members of mtDNA hap-
surrounding hunter-gatherers in peripheral contact zones logroup2U,e.g.,U2,U4,U5a,U5b,andU8(Bramantietal.,2009;
(eventsB1andB2) Krause et al., 2010; Hervella et al., 2012; Sanchez-Quinto et al.,
(cid:1) A period of renewed genetic influx during the Late Neolithic 2012;Bollonginoetal.,2013;DerSarkissianetal.,2013;Fuetal.,
withvariableregionalrepercussions(eventsCandD). 2013;Lazaridisetal.,2014).Insharpcontrast,thegeneticlineages
ofEarlyNeolithicfarmersfromCentralandSouthwestEurope(238
Inthisreviewwewillusethischronologicalreferencedataset, individualsfrom30sites(Fig.1A,sites31e58,81))arecomprisedof
updated with most recent studies, and discuss parallels and dif- awiderarrayofmtDNAlineageswithagenerallyhigherdiversity
ferences to various regions in Europe with specific focus on the (e.g.,H,HV,V,K,J,T2,X,W,N1a),inwhichUlineagesaregenerally
morerecent‘Holocene’populationhistory.Wewillsummarisethe less frequent, or in the case of Central Europe very rare or even
current status of uni-parentally inherited markers (i.e., mostly absent(Fig.1Band Fig.2AeC) (Chandleretal., 2005; Haaket al.,
mitochondrialsequencedatabutalsoY-chromosomedata)andthe 2005, 2010; Bramanti, 2008; Lacan et al., 2011a; Gamba et al.,
firstexcitingresultsfromgenomesequencingefforts,butwillalso 2012; Hervella et al., 2012; Lee et al., 2012; Brandt et al., 2013;
discussfurtherimplicationsofgeneticresultsinlightoflinguistic Lazaridis et al., 2014), supporting a genetic discontinuity of
andclimaticrecords. maternallineagesattheadventoffarming(Bramantietal.,2009;
Haaketal.,2010;Brandtetal.,2013).
ThewidegeographicdistributionofhaplogroupUintheUpper
PaleolithicandMesolithicrangingfromsitesinPortugal(Chandler
SettingthesceneeThegeneticsofpre-agriculturalEurope et al., 2005) and the two La Bran~a individuals (possibly closely
related;Sanchez-Quintoetal.,2012)inthefarWesttotheYuzhny
InordertoshedlightonthepeoplingofEurope,andespecially
Oleni Ostrov site in Karelia's North (Der Sarkissian et al., 2013),
the process of Neolithisation1 with associated genetic changes in
Kostenki in Russia (Krause etal., 2010)andthe Mal'taboyinthe
variousregionsof Europe, it isof utmostimportancetofullyun-
Lake Baikal region (Raghavan et al., 2014) is very interesting and
derstand the genetic substratum, i.e., the prevailing genetic di-
argues for a relatively small population size of hunter-gatherer
versityinEuropeatthetimeoftheLateMesolithic.Thegeneticdata
bands, stretching across vast longitudinal space. Associated cul-
availableatpresentconsistmostlyofmitochondrialDNA(mtDNA) turalattributes,suchasornamentsandfigurinesseemtoindicate
studies, but nuclear genomic data will undoubtedly widen our cultural affinities across large geographic distances (Sanchez-
knowledge about past populations. This will include information
Quintoetal.,2012;Olaldeetal.,2014).
abouttiming,numberandmodeofadmixtureevents,population ThefindingofUtypesinpre-agriculturalEuropeansisconsis-
structureaswellassex-biasedmigrationpatterns,whichcannotbe
tentwithastudyofRichardsandcolleagues(Richardsetal.,2000),
inferred from uni-parentally inherited mtDNA. However, the
who, based on coalescence age calculation of modern-day Euro-
chances of retrieving nuclear data at a sufficiently high level of
peanpopulationdata,suggestedthathaplogroupUwasoneofthe
complexity decrease as we gradually move from cooler and
oldest Eurasian branches in the mitochondrial phylogeny and
temperateclimatezonestowarmerandmorechallengingregions
therefore a prime candidate for an expanding lineage during the
(Burger et al., 1999; Bollongino and Vigne, 2008; Pruvost et al.,
initialpeoplingofEurasiabyanatomicallymodernhumans~42kya
2008), such as the Mediterranean coastlines including North Af-
(thousandyearsago)(Mellars,2006).Anotherinterestingaspectis
ricaandtheNearandMiddleEast,whichpromisetoholdtheul- thatsomeoftheoldestsamples(e.g.,Mal'taandDolníVĕstonice)
timate answer to many of the questions tackled here. In all
show basal U haplotypes (Fu et al., 2013; Raghavan et al., 2014),
likelihoodanddespitegreattechnologicaladvances(Meyeretal.,
which cannot be attributed to anyof the known sub-clades. This
2012; Gansaugeand Meyer, 2013), aDNA studies in these regions
not only supports the authenticity of individual results but also
willrequireintensivescreeningofmanysamplesinorderto find
matchesourexpectationsaboutsequenceevolution(withtheage
ones with suitable DNA preservation, and will often remain
restricted to mtDNA evidence for the majority of specimens
(Salamonetal.,2005;Hugheyetal.,2013).
2 Human mitochondrial sequence polymorphisms are commonly described as
‘haplotypes’, whereas characteristic polymorphisms, which are shared between
1 TheprocessofNeolithisationdescribestheshiftfromanomadicorseasonally individuals,arecombinedin‘haplogroups’(orlineages),followingahierarchical
changinghunter-gatherer-fisherlifestyletomoresedentarysubsistencestrategies phylogeneticstructureandnomenclature,includingmacro-andsub-haplogroups.
involvingagriculturalpractices,animalandplanthusbandry,andaccompanying Haplogroups commonly found in Europe today include
changesinmaterialcultureandsocialorganisation. U2eU5,K,R0,HV,V,H,J,T,W,I,X,andN1a(Richardsetal.,1998).
G.Brandtetal./JournalofHumanEvolution79(2015)73e92 75
Figure1. Locationandprincipalcomponentanalysis(PCA)ofpaleogeneticdatapublishedtodate.(A)MapshowingthelocationofprehistoricsitesfromwhichaDNAdatahave
beenpublished.SymbolsindicatepopulationsfromCentralEurope(squaresanddiamonds),outhernScandinavia(circles),South/SouthwestEurope(triangles),andEastEurope/Asia
(stars).Colourshadingsdenotehunter-gatherer(grey),EarlyNeolithic(brown),MiddleNeolithic(orange),andLateNeolithic/BronzeAge(yellow)samples.(B)Themajorityofthese
ancientDNAsequenceswerepooledinto22groupsaccordingtoculturalaffiliationorspatialandtemporalrelations.ThemtDNAhaplogroupfrequenciesofthesegroupswereused
toperformPCAasdescribedinBrandtetal.(2014).Eachhaplogroupissuperimposedascomponentloadingvector(grey),proportionallytoitscontribution.Stripedcoloursonthe
mapindicatedatathatwerenotincludedinthePCAduetosmallsamplessizes.Abbreviations:BAC¼Baalbergeculture,BAK¼BronzeAgeKazakhstan,BAS¼BronzeAgeSiberia,
BBC¼BellBeakerculture,BEC¼Bernburgculture,BLA¼Bl€atterho€hle,CAR¼(Epi)Cardialculture,CWC¼CordedWareculture,FBC¼FunnelBeakerculture,HGC¼hunter-
gatherersCentralandNorthEurope,HGE¼hunter-gatherersEastEurope,HGS¼hunter-gatherersSouthwestEurope,LBK¼LinearPotteryculture,NBQ¼NeolithicBasqueCountry
andNavarre,NCA¼NeolithicCatalonia,NPO¼NeolithicPortugal,PWC¼PittedWareculture,RSC¼Ro€ssenculture,SCG¼Scho€ningengroup,SMC¼Salzmündeculture,
TRE¼Treillesculture,UC¼Uneticeculture.(Forinterpretationofthereferencestocolourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle).
76 G.Brandtetal./JournalofHumanEvolution79(2015)73e92
A Central Germany
6,000 cal BC 5,000 cal BC 4,000 cal BC 3,000 cal BC 2,000 cal BC 2,000 cal AD
100
shared lineages
Hunter-gatherer
80 Early/Middle Neolithic
Late Neolithic
other
60
%
40
20
0
HGC LBK RSC SCGBAC SMC BEC CWC BBC UC CEM
B Southwest Europe
6,000 cal BC 5,000 cal BC 4,000 cal BC 3,000 cal BC 2,000 cal BC 2,000 cal AD
100
80
60
%
40
20
0
HGS CAR NBQ NCA TRE SEM
NPO
C Southern Scandinavia
6,000 cal BC 5,000 cal BC 4,000 cal BC 3,000 cal BC 2,000 cal BC 2,000 cal AD
100
80
60
%
40
20
0
HGC FBC PWC NEM
G.Brandtetal./JournalofHumanEvolution79(2015)73e92 77
of the older samples being closer to the coalescence date and describedfromtheCardialcultureinNortheasternSpain(Gamba
thereforeclosertotherootofthebranch). etal.,2012),whichrepresentsthefirstfarmingcultureoftheIbe-
Despitetheapparentuniformityatthehaplogrouplevel,thereis rianPeninsula.IfweacceptthatN*wasalreadypresentinMeso-
increasingevidenceforaMesolithicpopulationstructureinEura- lithic Southwestern Europe, this can be likely interpreted as the
sia,justifyingacloserlookattheavailabledata.Malyarchuketal. resultofadmixturebetweenimmigrantfarmersandlocalhunter-
(2010)studiedthemodern-daydistributionofsubhaplogroupU5 gatherers. As such, based on the current state of research, the
in particular and proposed different places of origin during the presence of haplogroups N*, H, U5b, and U4 (albeit at low fre-
UpperPaleolithicforeachofthesub-clades,withU5bintheWest quency)andtheabsenceofU2andU5a,couldbeinterpretedasthe
andU5aintheEast.Thisisinaccordancewithstudiesthatlinked geneticMesolithicsubstratumofSouthwesternEurope.Incontrast,
sub-haplogroups of U5b, such as U5b1b1 and U5b3, with an East European hunter-gatherers show a different haplogroup
expansion after the Last Glacial Maximum from southwestern composition,whichisbasedonhigherfrequenciesofU2,U4,and
refugia in Franco-Cantabria and Italy, respectively (Achilli et al., U5athaninCentralandSouthEurope,verylowfrequenciesofH,
2005;Palaetal.,2009).Sofar,finalconclusionscannotbedrawn andtheabsenceofU5bandN*.HaplogroupU2hasbeenidentified
fromthesmallnumberofancientindividualsfromeachseparate inPaleolithicindividualsfromKostenki(Krauseetal.,2010),Motala
aDNAstudy.However,whentakentogetheritappearsthatancient in Sweden (Lazaridis et al., 2014) and Mesolithic data from the
hunter-gatherersamplesconfirmthehypothesisofdifferentplaces RepublicofKareliainNorthwestRussia(DerSarkissianetal.,2013).
oforiginforeachoftheU5sub-clades,showinghigherfrequencies In contrast, U2 has been detected only rarely in Central Europe
of U5b in the West (Chandler et al., 2005; Hervella et al., 2012; (Bollonginoetal.,2013),andhasnotyetbeenobservedinSouth-
Sanchez-Quinto et al., 2012), with U5a more common in the western Europe. Therefore, it appears that haplogroup U2 was
(North)east (Bramanti et al., 2009; Krause et al., 2010; Der morefrequentintheMesolithicofNorthandEastEuropethanin
Sarkissianetal.,2013;Lazaridisetal.,2014),andamixedcompo- CentralorSouthwestEurope.Furthermore,aconsiderableamount
sition in Central Europe (Bramanti et al., 2009; Bollongino et al., ofhaplogroupC,commoninAsiaandtheAmericastoday,hasbeen
2013; Fu et al., 2013). This is also indicated by the separation of detected in forager populations of Northwest Russia, which was
distanthunter-gatherergroupsfromCentral,East,andSouthwest interpretedasgenetic influxfromWest SiberiaduringtheMeso-
Europeintheprincipalcomponentanalysis(PCA)(Fig.1B).Inter- lithic(DerSarkissianetal.,2013).
estingly,sub-cladeU4seemstobemostcommonaroundtheBaltic The emerging genome data from a few Mesolithic and early
regionandSouthScandinaviatodayandhasbeenproposedtohave farmingindividuals(Kelleretal.,2012;Sanchez-Quintoetal.,2012;
originated in a Southeast European glacial refugium (Malyarchuk Skoglund et al., 2012; Lazaridis et al., 2014; Olalde et al., 2014;
et al., 2008) and is indeed more frequent in central and north- Raghavan et al., 2014) confirms the discontinuity between
easternhunter-gatherers(Bramantietal.,2009;Malmstro€metal., hunter-gatherers and early farmers by identifying at least two
2009; Der Sarkissian et al., 2013; Lazaridis et al., 2014) (Fig.1B), distinct genetic components. Skoglund and colleagues attributed
whichalsoarguesforstructuredancientpopulationsduringUpper thefirstcomponenttoSouth/SoutheastEuropeanfarmers,whichis
PaleolithicandMesolithictimes.Thispicturebecomesmorecom- best represented by modern-day Sardinians, but also matches
plexwiththeobservationofnon-Ulineagesinperipheralregionsin genomicdatafromthefamousTyroleanIceman(Kelleretal.,2012;
the far West (haplogroup H and N*) and Northern Russia (hap- Skoglund et al., 2012). The second component characterises
logroupHandC),whiletheseareabsentinCentralEuropeanfor- NorthwestEuropeanhunter-gatherersandisstillpresentinmany
agers.HaplogroupHhasbeenreportedwithhighfrequencies(38%) modern-day North European populations. Low coverage genome
from hunter-gatherers in Portugal and the Basque country sequencingoftwoMesolithicindividualsfromIberiaconfirmsthis
(Chandleretal.,2005;Hervellaetal.,2012)andoneindividualfrom discontinuity in Western Europe (Sanchez-Quinto et al., 2012).
theYuzhnyyOleniOstrovsiteinNorthernKarelia(DerSarkissian Importantly,thefirstautosomaldatanotonlysupportthediscon-
etal.,2013),indicatingawidespreadandpossiblyclinaldistribu- tinuity seen in the mitochondrial results, and as such critically
tionofhaplogroupHinMesolithicEurope.Underthishypothesisit undermine the idea of an exclusively cultural diffusion of the
seems to be a question of time when the first H lineage will be Neolithicwayoflife,butalsosuggestamorecomplexpopulation
reported from Central European hunter-gatherers. In addition, structureinpre-agriculturalEurope,whichin turnwillaffectour
haplogroupN*hasbeendescribedinMesolithicindividualsfrom interpretation of the events following the initial Meso-Neolithic
Portugal(Chandleretal.,2005)andonePaleolithicindividualfrom transition. In particular, the most recent study by Lazaridis et al.
SouthItaly(Caramellietal.,2003),suggestingadistributionacross (2014),includinghighcoveragegenomedatafromoneMesolithic
SouthernEuropeduringPaleolithicandMesolithictimes.Whilethe individual from Loschbour/Luxembourg plus additional low
data from Portugal have not been published in a peer-reviewed coverage data from the Swedish hunter-gatherer site Motala and
journal and therefore are best viewed cautiously, the reported one early farmer of the Linear Pottery culture or Linearbandker-
presenceofU5binthisMesolithicdatasetisabsolutelyplausible, amische Kultur (LBK) from Stuttgart-Mühlhausen, is most eluci-
while N* and H have seemingly been confirmed by independent dating,suggestingatleastthreeancientpopulations,contributing
studies (Caramelli et al., 2003; Hervella et al., 2012). N* has also invariousproportionstoallmodern-dayEuropeans.Thefirstisan
Figure 2. Shared haplotype analyses. Shared haplotype analyses were performed in three key regions of Europe including Central Germany's Mittelelbe-Saale region (A),
SouthwesternEurope(B)andSouthernScandinavia(C),followingtheconceptdescribedin(Sze(cid:1)cse(cid:1)nyi-Nagyetal.,2014).AvailableNeolithicdatafromeachregionwerecombined
withrespectivehunter-gathererandmodern-daymetapopulations(CEM¼CentralEuropeanmetapopulation,SEM¼SouthEuropeanmetapopulation,NEM¼NorthEuropean
metapopulation)andplacedinchronologicalorder.Themodern-daymetapopulationsareeachrepresentedby500randomlyselectedindividuals.Toestimatetheamountof
lineagesineachregion,whichwerepresentduringthePaleo/MesolithicandlikelyhavebeenbroughtinbyincomingfarmersoremergedinlaterNeolithicperiods,wetracedeach
lineageofatemporalgroupbacktoitsearliestmatchinthechronologyandregardeditasanancestrallineagethatappearedinthisgroupforthefirsttime.Coloredlinesindicate
summarisedmatcheswithlineagesfromhunter-gatherers(grey),Early/MiddleNeolithic(brown)andLateNeolithic/EarlyBronzeAge(yellow)cultures.Haplotypeswithoutdirect
matchesinpreviouscultures/populationswerenotresolvedanddefinedasothers(blackdottedline).SincetheavailabledatafromMittelelbe-Saaleprovidesagaplessgenetic
recordfromtheintroductionoffarmingtotheEarlyBronzeAge,weusedthisgeneticreferencestratigraphytogetherwithrespectiveMesolithicdatatoalsoevaluatetheamountof
hunter-gatherer,earlyfarmerandLateNeolithiccomponentsinSouthwesternEuropeandSouthernScandinaviaviadirectcomparisonwithMittelelbe-Saaledata.Abbreviationsare
consistentwithFigure1.(Forinterpretationofthereferencestocolourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle).
78 G.Brandtetal./JournalofHumanEvolution79(2015)73e92
ancient North Eurasian population, exemplified by the Upper complexes, namely the Bell Beaker complex in Western Europe
Paleolithic Mal'ta boy from Siberia (Raghavan et al., 2014), the (R1b;Leeetal.,2012)andtheCordedWarecultureinEastEurope
second is a Western hunter-gatherercomponent (represented by (R1a; Haak et al., 2008; see also below). Interestingly, the afore-
LoschbourandalsothetwoLaBran~aindividuals;Sanchez-Quinto mentionedMal'taboyfromSiberiacarriesabasalRhaplogroupand
et al., 2012), and the third is an early European farmer compo- likelypredatesthesplitintosubgroupsR1a,R1bandR2(frequent
nent, whichis represented bythe LBK farmer fromStuttgart, the inIndia).SinceRisasister-cladeofQ,whichiscommoninNative
Iceman (Kelleret al., 2012)and a Scandinavian farmer (Skoglund Americans,thisarguesforacommonoriginofRQinSiberia(Wells
et al., 2012), closely related tomodern-daySardinians. The Scan- etal.,2001;Raghavanetal.,2014).Howandwhenthenowfrequent
dinavian hunter-gatherers from Motala form their own cluster R1 subtypes reached Western Eurasia is an important question,
together with data from Skoglund et al. (2012), explained by the whichwillneedtobeaddressedinfutureaDNAstudies(Underhill
proportionofsharedalleleswiththeancientNorthEurasianpop- et al., 2010; Myres et al., 2011). In light of the population events
ulation,whichisotherwiseabsentintheWesternhunter-gatherer AeDmentionedabove,whichcontributedsubstantiallytomodern-
andearlyEuropeanfarmerclusters. dayCentralEuropeanmtDNAdiversity(Brandtetal.,2013),itcould
TherecentstudybyLazaridisandcolleagues(2014)alsoreports bespeculatedthatsimilardemographicprocesseshadalsoshaped
thefirstY-chromosomeresultsforfiveEuropeanhunter-gatherers theEuropeanY-chromosomevariability.However,apartfromthe‘I
(LoschbourandfourmalesfromMotala).Mostinterestingly,allfive versus G2a pattern’, which now has been detected in sufficient
samplescanbeattributedtoYhaplogroupI,whichappearstobe numbersandfromvarioussites,thepresence/absenceofotherY-
thepredominanthaplogroupinEuropeanhunter-gatherersandso chromosomehaplogroupsinprehistoriccontextsisclearlyofsto-
farparallelstherelativeantiquityandconstrainttoonesingle(but chastic nature and can only serve as a working hypothesis until
major)branchasseeninmtDNAhaplogroupU.Thisisconsistent moreinformationbecomesavailable.
with population genetic studies that described the highest fre-
quencies of Y haplogroup I in present-day Scandinavian and TheimportanceoftheMeso-Neolithictransition
Southeast European populations, and proposed that I subgroups
had been present in Europe since the late Upper Paleolithic and The transition to farming (or the process of Neolithisation)
expanded after the Last Glacial Maximum from the Franco- emergedintheNearEast'sFertileCrescentabout12,000yearsago
Cantabrian and a Southeast European glacial refugium (Rootsi and is described as one of if not the most fundamental cultural
etal.,2004;Pericicetal.,2005;Karlssonetal.,2006;Lappalainen changeinhumanhistory(Whittle,1996;Price,2000;Whittleand
etal.,2009).However,theavailableYdatafromearlyfarmersdo Cummings, 2007). It is deeply linked to the transition from
notparallelthediversityindicatedbythemitochondrialdataand foragingtoagricultureandanimaldomesticationalongsideamore
appear,atthemomentatleast,restrictedtoonlyafewhaplogroups sedentarylifestyle,whichwasaccompaniedbyradicallyneweco-
among whichG2aispredominant. G2aisratherrarein modern- nomicstrategiesandnumerouschangesinmaterialculture,social
day Europeans but was found in one out of three samples from life,customsandideologies(Price,2000;WhittleandCummings,
LBK farmers in Central Europe (Haak et al., 2010), five out of six 2007).Overthecourseofthefollowingmillenniafarmingspread
malesatanEpicardialsiteinSpain(Lacanetal.,2011a),20outof22 across Europe and replaced the foraging lifestyle of indigenous
males at the Mid/Late Neolithic site Treilles in Southern France hunter-gathererpopulations.
(Lacanetal.,2011b),andtheTyroleanIceman(Kelleretal.,2012).A The nature and mode of the transition from a foraging to a
recent study of late hunter-gatherers and early farmers from the farminglifestylearethematteroflongstandingdebateinarchae-
Carpathian Basin by our group confirms this pattern further, ology and adjacent fields (Rowley-Conwy, 2011). Archaeologists
observingpredominantlyG2aandtoasmallerproportionIinnine havedescribedanddiscussedtwomajorandlargelycontempora-
maleindividuals(Sze(cid:1)cse(cid:1)nyi-Nagyetal.,2014). neousroutesoftheexpansionoffarming,namelytheContinental
WhilethepresenceofhaplogroupIinNeolithiccontextscould andtheMediterraneanroute(Price,2000;WhittleandCummings,
be interpreted as a signal of hunter-gatherer introgression in 2007; Rowley-Conwy, 2011). The emergence and dispersal of
farmingcommunities,andthereforerepresentsaPaleolithiclegacy, Neolithiccultures,aswellasthespeedoftheNeolithisation,varied
the precise way in which modern-day European Y-chromosome across different regions in Europe in a series of complex in-
diversity was formed remains elusive. To date, the only other Y- teractionsandtemporalfluctuations(Bocquet-Appeletal.,2009).
haplogroups observed in early farming sites are haplogroup F in Traditionally,scholarsexplainedtheNeolithictransitioneitherby
Germany and Hungary (Haak et al., 2010; Sze(cid:1)cse(cid:1)nyi-Nagy et al., expansionofearlyfarmersfromtheNearEast,whobroughtnew
2014), and E1b in one individual in Spain (Lacan et al., 2011b). ideas as well as new genes (demic diffusion) (Childe, 1925;
ThepresenceofhaplogroupFisverysurprising,asitisveryrarein Ammerman and Cavalli-Sforza, 1984; Cavalli-Sforza et al., 1994;
modern-dayEuropeanpopulationsandthereforenotwellstudied. Chikhietal.,1998,2002)orbyadoptionofinnovativesubsistence
IthasbeenreportedatalowfrequencyinSoutheastEuropeandthe strategiesofindigenoushunter-gathererpopulationsbyatransfer
Near East (Underhill and Kivisild, 2007), whereas subgroups of F of ideas with little or no genetic influence (cultural diffusion)
havebeenprimarilyfoundinIndia(Kivisildetal.,2003).E1bandJ2 (Whittle,1996; Richards et al.,1996, 2000; Semino et al., 2000).
are common in extant South and Southeast Europeans and were Consequently, both the existence and the extent of a Neolithic
thoughttohavespreadwiththeNeolithicexpansion(Seminoetal., contributiontothegenepoolofpresent-dayEuropeanshavebeen
2004), and while E1b has been observed, J2 and other common intensely debated until recently (Achilli et al., 2004; Dupanloup
modern haplogroups, such as N1c (frequent in Siberian and et al., 2004; Currat and Excoffier, 2005; Balaresque et al., 2010;
Northeast European populations; Rootsi et al., 2007; Balanovsky Soares et al., 2010; Pala et al., 2012; Pinhasi et al., 2012; Olivieri
etal.,2008)haveyettobeobservedinprehistoriccontexts.Simi- etal.,2013).
larly,themostpredominanthaplogroupsinmodern-dayEuropean Uni-parentallyinherited markers, such as mtDNA and Y-chro-
populations (R1b in West Europe and R1a in East Europe; mosome,haveearnedtheirmeritsinpopulationgeneticsasthey
Balaresque et al., 2010; Underhill et al., 2010; Myres et al., 2011) are relatively easy to track through generations, but the true
havenotbeenfoundinearlyfarmingsites,butinsteadhavebeen advantage is that the respective modes of inheritance provide us
reportedfromLateNeolithiccontextswithasurprisinglymatching with a uniquely male and female perspective of human evolu-
geographic distribution of the associated archaeological cultural tionary history (Pakendorf and Stoneking, 2005). It is crucial to
G.Brandtetal./JournalofHumanEvolution79(2015)73e92 79
investigatethesetwolinesofevidenceinparalleltodatafromthe package’ that reached CentralEurope in the sixth millennium BC
recombining autosomes, as they provide insights into social and (Brandtetal.,2013).Theremaininglineagesbelongtohaplogroup
demographic parameters, such as sex-biased introgression or H(16.5%),U3(0.9%),andU5a(0.9%).Ifwefollowedthelogicabove,
mobility,whichareimportantforthereconstructionofpastsoci- thentheUhaplotypescouldlikelybeinterpretedasaMesolithic
eties(Haaketal.,2008). legacy.However,thesituationismoredifficulttodecipherinthe
Ethnographic and demographic studies have shown that eco- caseofhaplogroupH,whichhasnotbeenobservedinMesolithic
nomic factors (status and wealth) and social differentiation drive Central Europe and Scandinavia but is present in Iberian hunter-
variationinsex-biasedmobility(e.g.,Bentleyetal.,2009,2012,and gatherers (as shown above). A recent aDNA study by our team
referencestherein).Itisgenerallyassumedthatthetransitionfrom hasrevealedthatthedistributionofHsubgroupsinCentralEurope
aforagingtoafarminglifestylebroughtaboutdrasticchangesto changes over the course of the Neolithic, arguing for a more
thesocialstructureandresidencepatterns,shiftingfrombilocality complex evolutionary history of this branch rather than being
to patrilocality in farming societies, which guaranteed the inter- attributed to a single major event (Brotherton et al., 2013). Hap-
generational transfer of status, wealth and territory (Marlowe, logroup H is the most common maternal lineage in extant Euro-
2004;Wilkins,2006;WilkinsandMarlowe,2006;Rasteiroetal., pean populations with frequency peaks of over 45% in Western
2012). Ethnographic research in contact zones between forager Europe. The origin of H has been attributed to the Near Eastern
andfarmersocietieshasalsohighlightedthesocialandeconomic UpperPaleolithic,fromwhereitbranchedofffromitssister-clades
imbalancebetweenthetwosocieties,oftenresultingina‘culture R0 and HV0 upon entering Europe. Most of the subgroups in
clash’, in which the farmers directly compete with hunter- present-day Europe show late glacial or post-glacial coalescence
gatherers,andthesustainablepracticesoffarmingseemtobean dates,arguingforare-expansionduringthemajorwarmingpha-
important economic drawing card. As a consequence, mobility ses after 15 kya (Achilli et al., 2004; Pereira et al., 2005; Soares
undersuchademographicscenarioismostoftenuni-directional, et al., 2010). Subgroups H1, H3, and H5 are believed to have
withhunter-gathererwomenmarryingintofarmercommunities, spread from a Western refugium in Franco-Cantabria based on
butneverorveryrarelytheopposite.Atthegeneticlevel,wewould largelyoverlappingdates(11.1kya,11.5kya,13.9kya,respectively)
therefore expect to see a consolidation of farmer Y lineages (Soaresetal.,2010)andthefactthatthesearethemostcommon
replacing hunter-gatherer lines, whereas the female hunter- typesinWesternEurope.WhilehaplogroupHhasbeenreported
gatherer lineages would contribute to the diversity of farmers. from Iberian Mesolithic individuals, the typing resolution unfor-
Suchascenarioishighlylikelyincontactzonesatthewave-frontof tunately does not provide an unambiguous assignment to any of
an expanding farming culture and would provide a plausible these subgroups (Chandler et al., 2005; Hervella et al., 2012).
explanation for a slow but steady increase (or comeback) of e.g., However,allMiddleNeolithicindividualsfromTreillesinSouthern
mtDNAUlineagesinculturalhorizons,followingthefirstassimi- France could be assigned to subgroups H1 and H3 via coding re-
lationandconsolidationphases.Atthesametime,thiswouldalso gion SNP typing (Lacan et al., 2011b). In contrast, near complete
explaintherelativelylimiteddiversityofYlineagesinearlyfarming mitochondrialgenomesequencesfromLBKindividualsshowthat
communities. onlythreeoutofninefallwithinsubgroupH1.Theremainingsix
individuals carry basal and ancestral H lineages, which are rare
TheearlyfarmersoftheLBKinCentralEurope todayand showaffinities tothe Caucasus, Anatolia and the Near
East,in supportofanoriginofbasalHinthisregion(Brotherton
Arecentgeneticstudybyourgrouphassubstantiallyextended et al., 2013). Based on these preliminary results, we feel that
thedatasetforLBKfarmers(Brandtetal.,2013),nowcomprising there is considerable structure within this large branch, which
109 individuals from 13 sites in Germany, Austria and the Czech necessitates a higher resolution (complete mtDNA genomes)
Republic (Fig.1A, sites 31e43). This represents to ourknowledge beforewecanascribecertainsubgroupstoaspecifictimeperiodin
the largest dataset from a single cultural horizon, and can be prehistory.
consideredasamplesizelargeenoughtobetrulyrepresentative, Aside from the haplogroup H conundrum, the overall mtDNA
allowingforminimalstatisticalerror.Theresultinggeneticprofile compositionamongandbetweenextentLBKcommunities,ranging
issurprisinglyconsistentwithourpreviousresults fromthiscul- fromthewellknownarchaeologicalsitesFlomborn,Vaihingenand
ture (Haak et al., 2005, 2010) and fully confirms the conclusions Schwetzingen in Southwest Germany (Haak et al., 2005) to
drawnfromtheanalysis:thehaplogroupcompositionsofthelast VedroviceintheCzechRepublic(Bramanti,2008)(Fig.1A),issur-
hunter-gatherersandearlyLBKfarmersarevirtuallyexclusiveand prisingly uniform with non-significant differences both on the
a genetic discontinuity during the Meso-Neolithic transition ap- haplotypeandhaplogrouplevel(Brandtetal.,2014).Thissuggests
pearstobe‘setinstone’,asshownbytestsforpopulationconti- that the Neolithic transition in Central Europe driven by the LBK
nuity (Brandt et al., 2013) and coalescent simulations (Bramanti affectedlargeterritoriesandreplacedmostoftheknownhunter-
et al., 2009; Haak et al., 2010). Several lines of analytical evi- gatherer variability with a relatively homogenous and coherent
dencesupportaffinitiesoftheLBKtomodern-daypopulationsof mitochondrial signature. This finding is in accordance with the
the Near East and reveal that the transition from a foraging to a archaeologicalhypothesis,whichproposedaspreadoffarmersinto
farminglifestyleinCentralEuropeinthemid-sixthmillenniumBC CentralEurope alongaContinentalroute following theDanubian
was clearly associated with genetic influx via expanding LBK corridor (Gronenborn, 1999, 2007a; B(cid:1)anffy, 2000; Price, 2000;
communities(eventA,~5500calBC,Fig.3A),whichbroughtnew Bogucki and Crabtree, 2004; Whittle and Cummings, 2007;
genetic elements fromregionsin Southeast Europe,Anatolia, the Bocquet-Appel et al., 2009; Rowley-Conwy, 2011). Detailed
CaucasusandtheNearEast(Haaketal.,2010;Brandtetal.,2013). archaeologicalresearchhasshownthattheculturalcharacteristics
A shared haplotype analysis shows that only 1.8% of the LBK oftheLBKemergedaround5600calBCfromEarlyNeolithiccul-
haplotypesmatchhunter-gathererslineages,whiletheremaining tures in the Carpathian Basin (Star(cid:3)cevo-Ko€ro€seCris¸ complex,
98.2% havenotyet been reported fromthe Central and Northern ~6000e5400calBC)(Gronenborn,1999;B(cid:1)anffy,2000).Ourgroup
European Mesolithic (Fig. 2A). The majority of these lineages is currently investigating Neolithic samples from the Carpathian
(79.8%)canbeassignedtohaplogroupsN1a,T2,K,J,HV,V,W,and Basinandthefirstresultsindeedhintataclosegeneticaffinityof
X, which appear tobe the most characteristic for the LBK period LBK farmers with its preceding neighbors (Sze(cid:1)cse(cid:1)nyi-Nagy et al.,
(Fig.1B) and can thus be defined as the ‘mitochondrial Neolithic 2014).
80 G.Brandtetal./JournalofHumanEvolution79(2015)73e92
Starčevo/Körös/Criş complex (~6,000-5,400 cal BC)
Impresso/Impressa culture (~5,800-5,300 cal BC)
Cardial culture (~5,600-4,700 cal BC)
Linear Pottery culture (~5,600-4,750 cal BC)
??
A
Ertebølle culture (~5,400-4,100 cal BC)
Stroke Ornamented Pottery culture (4,925-4,650 BC)
Lengyel culture (~4,900-4,200 cal BC)
Rössen culture (~4,700-4,250 cal BC)
Schöningen group (4,100-3,950 cal BC)
????
B
Chasséen culture (~4,500-3,500 cal BC)
Michelsberg culture (~4,300-3,500 cal BC)
Funnel Beaker culture (~4,100-2,650 cal BC)
Baalberge culture (~4,100-3,400 cal BC)
Salzmünde culture (3,400-3,100/3,025 cal BC)
Bernburg culture (3,100-2,650 cal BC)
Globular Amphora culture (~3,100-2,650 cal BC)
C
Yamnaja culture (~3,300-2,300 cal BC)
Corded Ware culture (~2,900-2,000 cal BC)
Bell Beaker culture (~2,800-1,800 cal BC)
Unetice culture (~2,300-1,550 cal BC)
???
D
G.Brandtetal./JournalofHumanEvolution79(2015)73e92 81
PreviousstudieshavesingledouthaplogroupN1aasapotential colonisation model, mosaic model, dual model, and capillary
mtDNAmarkerassociatedwiththespreadofthefirstfarmersinto model)(Vicent-García,1997;Zilha~o,1997,2001;Schuhmacherand
Central Europe (Haak et al., 2005, 2010). The enlarged dataset of SanzGonza(cid:1)lezdeLema,2013,andcitationstherein).Itseemsthat
109LBKindividualshasrevealedmoreN1aindividuals,confirming different areas of the Iberian Peninsula present themselves with
a stable frequency of N1a (11.9%) during the Early Neolithic. The varyingdegreesanddensitiesofNeolithicversusMesolithiccom-
transect through time in Mittelelbe-Saale showed that the fre- ponents. Moreover, older radiocarbon dates and questionable
quencyofhaplogroupN1adecreasesduringtheEarlyandMiddle contexts of many multiphase archaeological sites complicate the
Neolithic and is absent from the Bernburg culture and the Late interpretationofarchaeologicaldata.However,ithasbeenwidely
Neolithic/Early Bronze Age samples (Brandt et al., 2013). Other acceptedthatthecomplexityoftheshiftfromforagingtofarming
associated early farming lineages (e.g., T2, K, J, HV, and V) also cannot be covered by oversimplifying models and should be
decreasedinfrequencyduringtheMiddleandLateNeolithic,which interpretedinthespecificcontextofeachregion(Fe(cid:1)rnandezLo(cid:1)pez
can be explained by the impacts of subsequent migration and dePabloandGo(cid:1)mezPuche,2009).
expansionevents(e.g.,B,CandDinFig.3AeD;Brandtetal.,2013, Recent aDNA studies have provided direct insights into the
see also below). However, the reasons for the almost complete mitochondrial and Y-chromosome diversity of Southwestern
disappearance of N1a (<1% in modern-day Eurasianpopulations) Europe during the Late Mesolithic and the Neolithic periods
arenotfullyunderstood.Interestingly,N1ahasalsobeenfoundin (Chandleretal.,2005;Sampietroetal.,2006;Lacanetal.,2011a,b;
one individual of the Megalith culture (4340e4076 cal BC) from Gamba et al., 2012; Hervella et al., 2012; Sanchez-Quinto et al.,
SouthwestFrance(Deguillouxetal.,2011b)andoneindividualof 2012) (Fig. 1A, sites 17e25, 48e58, 65e66). In particular, the
theAlfo€ldLinearPotteryculture(Ecsegfalva23A,5250e5000BC; growing number of aDNA studies of early farmers from different
Haaketal.,2005),whichcanberegardedasaparalleldevelopment regions in Iberia, namely Northeast Spain, Northern Spain, and
totheLBKinEastHungary(Alfo€ldregion).Inaddition,datafrom CentralPortugal,allowustodrawamoredetailedpictureofpop-
precedingculturesoftheCarpathianBasinconfirmthepresenceof ulation dynamic processes at the onset of farming in the Iberian
N1aduringtheNeolithicinthisregion(Sze(cid:1)cse(cid:1)nyi-Nagyetal.,2014). Peninsula.
Future aDNA studies targeting archaeological cultures south and Four Early Neolithic sites of the Cardial and Epicardial culture
eastoftheCarpathianBasinandfromEasternAnatoliaaredesired havebeenpublishedcoveringthecoreregionofNeolithisationin
tofurthertracetheexpansionofearlyfarmersthroughspaceand Iberia (Lacan et al., 2011a; Gamba et al., 2012) (Fig. 1A, sites
time. 48e51). The composition of mitochondrial lineages comprises
bothsouthwesternhunter-gathererhaplogroups(U5b,H,andN*)
GoingWestetheNeolithictransitioninSouthwestEuropevia and haplogroups that are common among Early and Middle
theMediterraneanroute NeolithicfarmersfromCentralEurope(T2,K,andX),resultingin
an intermediate positioning between these two in the PCA
InparalleltothespreadoffarmingalongtheContinentalroute, (Fig.1B).Onthehaplotypelevel,11.1%oftheCardiallineagescan
theNeolithicwayoflifealsoexpandedfromtheNearEastFertile be traced back to the pre-Neolithic gene pool of Southwestern
Crescent to the Iberian Peninsula on a route that followed the Iberiawhile55.6%directlymatchthoseofearlyCentralEuropean
coastlinesofthenorthernMediterraneanSeafromtheAegeanover farmers(Fig.2B).Thissharpincreaseoffarmerlineagessuggestsa
the Tyrrhenian and Ligurian Sea to Southern France and North- fastgeneticturnoverduringtheNeolithictransitioninSouthwest
easternSpain.Fromthere,farmingspreadfurtherinlandviamajor Europe, similar to the scenario in Central Europe, and is also
river systems, such as the Rho^ne and Ebro, and along the Medi- supported by 14C dates ranging in a narrow temporal window
terraneancoastlinesintoPortugal(Price,2000;Gronenborn,2003; (Zilha~o, 2001). In contrast, Early Neolithic samples from Central
Rowley-Conwy, 2011). Additional ways for the distribution of Portugal (Chandler et al., 2005) and Northern Spain (Hervella
Neolithic elements have been put forward, such as an expansion et al., 2012) harbor a larger number of forager lineages (41.2%
along the Pyrenees rather than the Mediterranean coasts and27.9%,respectively)andfewersharedfarminglineages(11.8%
(SchuhmacherandSanzGonz(cid:1)alezdeLema,2013)oraparallelroute and 23.3%, respectively; Fig. 2B), resulting in greater affinities to
alongthesouthernMediterraneancoastlineinNorthAfrica,which southwesternhunter-gatherersasshownbythePCAplot(Fig.1B).
is thought to have influenced the southern parts of the Iberian Taken together, this suggests that genetic elements of the
Peninsula(Lewthwaite,1989;Linsta€dter,2008). ‘Neolithic package’, which had reached Central Europe via the
The earliest Neolithic of the Iberian Peninsula is mostly con- Continental route (event A), also arrived in Southwest Europe
nectedtotheCardialculture,whichwasestablishedinthefirsthalf through the Mediterranean route (Fig. 3A). However, the hunter-
ofthesixthmillenniumBCprimarilyontheMediterraneancoastof gatherer legacy is more dominant in the Iberian Neolithic
SpainandSouthernFrance,concurrentwiththeappearanceofthe compared with Central Europe, indicating a Neolithic transition
LBK in Central Europe (Zilha~o, 1997, 2001; Price, 2000; Rowley- withalargercontributionoftheindigenouspopulationcombined
Conwy,2011).EarlyNeolithic sitesare knownfromtheEbroVal- withareducedimpactofearlyfarmers.Thisapparenttrendseems
ley and Portugal from the mid-sixth millennium BC onwards, to be accentuated with growing distance from Northeast Spain,
arguing for a rapid spread along the Iberian Peninsula (Zilha~o, bothinlandandalongthecoast,andisconsistentwithaproposed
2001; Zapata et al., 2004; Rojo Guerra et al., 2006). Archaeolo- pioneer colonisation model for Southwest Europe, according to
gists have discussed numerous models for the transition from which small farmer groups from the Near East spread along the
foragingtofarming,eachsuggestingvaryingimpactsofNeolithic Mediterranean coastlines (Chandler et al., 2005; Gamba et al.,
encounters or Mesolithic residents (e.g., maritime pioneer 2012;Hervellaetal.,2012).
Figure3. SummaryofpopulationdynamiceventsduringtheNeolithicperiodinEurope.Differentshadingsandpatternsdenotethegeographicdistributionsofculturesduringthe
NeolithicperioduntiltheBronzeAge:EarlyNeolithic(AandB),MiddleNeolithic(C),andLateNeolithic/EarlyBronzeAge(D).Stripedareasindicatearchaeologicalculturesfor
whichnoaDNAdataisavailablesofar.GreenarrowsdisplaypotentialgeographicexpansionsandtheirassociatedmtDNA(mt)andYchromosomal(Y)haplogroups.ThemtDNA
variabilityofindigenoushunter-gathererpopulationsfromCentral/North,Southwest,andEastEuropeisgivenintherespectiveregions(A).(Forinterpretationofthereferencesto
colourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle).
82 G.Brandtetal./JournalofHumanEvolution79(2015)73e92
Todate,onlytwostudieshavefocusedonlaterNeolithicperiods contemporaneous cultures like the Pitted Ware culture still
inSouthwesternEurope.Thefirststudyanalyzedindividualsfrom retained their Mesolithic lifestyle (Bogucki and Crabtree, 2004;
the Middle Neolithic period in Catalonia, Spain (~3500e3000 cal Rowley-Conwy,2011).
BC)(Sampietroetal.,2006),andthesecondoneinvestigatedthe Recent studies focusing on Mesolithic/Neolithic hunter-
Treillesculture(~3030e2890calBC)inSoutheasternFrance(Lacan gatherer and farmer communities of Northern Europe have pro-
etal.,2011b).WhiletheTreillescultureshowsageneticcomposi- duced useful insights into the population history of the region
tion comparable with the Cardial culture with 20.7% hunter- (Bramanti et al., 2009; Malmstro€m et al., 2009; Skoglund et al.,
gatherer and 58.6% early farmer lineages, the samples from the 2012; Lazaridis et al., 2014) (Fig.1A, sites 13e16, 63e64). Seven
MiddleNeolithicinNortheasternSpainharborahigheramountof individualsfromthesiteofMotalainSouthernSwedenrepresent
forager(27.3%)andalowerfrequencyoffarminglineages(18.2%),a genuine Mesolithic data from this region (Lazaridis et al., 2014),
composition very similar to the Early Neolithic samples from andshowexclusivelyhaplogroupU(U2,U4,U5a),consistentwith
Portugal, Northern Spain, and present-day populations of South- other data from Central/Northern Europe (as outlined above). In
west Europe (Fig. 2B). As a result, the high abundance of charac- contrast, samples attributed to the Funnel Beaker culture
teristicfarmerhaplogroupsobservedintheTreillescultureleadsto (Bramanti et al., 2009; Malmstro€m et al., 2009) can be charac-
apositioningclosertotheEarlyandMiddleNeolithicculturesfrom terisedbyamixtureoftypicalhunter-gathererlineages(U5aand
Central Europe in the PCA plot (Fig.1B). Despite the fact that all U5b)andhaplogroups fromthe‘Neolithicpackage’,suchasT2,J,
individualsfromEarlyandMiddleNeolithicNortheastSpainaswell and K. This results in an intermediate position of the Funnel
astheTreillescultureweresampledinregionswherefarmingwas Beaker culture between hunter-gatherer populations and the
firstestablishedinSouthwestEurope,thevaryingamountofearly Early/Middle Neolithic cultures on the PCA plot (Fig. 1B). On a
farminglineagesbetweenthesepopulationsarguesforacomplex moredetailedhaplotypelevel,20%and40%oftheFunnelBeaker
populationhistorythroughtimeandacrossspace. samples match hunter-gatherer and early farmer haplotypes,
Coalescentsimulationssuggestedthattheeffectofgeneticdrift respectively (Fig. 2C), suggesting genetic influence from Central
instructuredpopulationsmusthaveplayedanimportantrolebe- EuropeintoFunnelBeakercommunitiesinSouthernScandinavia.
tweentheEarlyandMiddleNeolithic,resultinginadilutionofearly We therefore proposeda dynamic population or migration event
farmerlineagesfromtheNearEast,andanmtDNAcompositionof B2 at ~4100 cal BC (Brandt et al., 2013) (Fig. 3B), which can be
Middle Neolithic individuals similar to present-day Iberians considered as delayed extension of the initial farming expansion
(Gambaetal.,2012).ThemitochondrialcompositionoftheTreilles during the LBK. Genomic sequence data from one Funnel Beaker
group on the other hand points to more stable conditions in individual has shown the greatest similarity to present day
Southern France, in which farming lineages from the Near East South(east) European populations, Sardinians in particular, and
persistedforalongerperiodoftimethaninneighboringregionsof has supported genetic influx fromthe South into SouthernScan-
NortheasternSpain.However,EarlyNeolithicdata fromSouthern dinaviaattheadventoffarming(Skoglundetal.,2012).Interest-
andCentralSpainandotherpartsofFrancewouldbedesirableto ingly, contemporaneous Neolithic foragers of the Pitted Ware
testthishypothesis.Inconclusion,thespreadoffarmingintoand culture from the Island of Gotland (Malmstro€m et al., 2009) also
overSouthwestEuropeandthetransitiontolaterNeolithicperiods showamixedmtDNAcompositionofhunter-gathererandfarmer
werecharacterisedbyseveralandvaryingprocessesataregional elements comparable to the Funnel Beaker culture, albeit with
level(Hervellaetal.,2012).Despitetheostensiblewealthofavail- higher proportions of Mesolithic lineages (26.3%), predominantly
able aDNA data, we feel that the emerging regional complexity consistingofhaplogroupU4,andloweramountsoffarmingline-
necessitates further detailed investigations both at larger agesofhaplogroupsT2,K,HV,andV(21.1%)(Fig.2C).ThehighU4
geographic scale and chronological resolution to increase our frequency in the Pitted Ware culture is in accordance with a
knowledge about (regional) demographic processes during the proposed eastern Baltic refugium for some of the European
NeolithisationandsubsequentperiodsinSouthwestEurope. hunter-gatherer populations during the Last Glacial Maximum
(Malmstro€m et al., 2009), which has been further supported by
PopulationshiftsinSouthernScandinavia genomic affinities of three Pitted Ware culture individuals to
modern-day Northeast European populations (Skoglund et al.,
TheMiddleNeolithicperiodinCentralEuropeischaracterised 2012). The presence of farmer lineages in the Pitted Ware cul-
by the Neolithic expansion into regions outside the former LBK turecanbeinterpretedeitherasdirectgeneflowfromthesouth,
territory,especiallyNorthandNortheastEurope.Inadditiontothe i.e., Central Europe, or indirectly via admixture between Funnel
abundance of natural resourcesalong the coastlines of the North Beaker and the Pitted Ware groups. However, the amount of
andBalticSea,whichdidnotnecessitateachangeinsubsistence hunter-gatherer lineages in both Scandinavian cultures is
strategy,harsherclimaticconditionsandlessfertilesoilsnorthof remarkablyhigherthanin anyEarly/MiddleNeolithicMittelelbe-
the preferred loess plains also prevented a rapid spread of agri- Saale culture indicating a stronger contribution of Mesolithic fe-
culture in these regions (Whittle and Cummings, 2007). As a males during the Neolithic transition in Southern Scandinavia. It
consequence, local hunter-gatherer communities of North Ger- seemshighlyplausibletousthattheprocessofNeolithisationwas
many,Denmark,andSouthSweden(Ertebølleculture,~5400e4100 muchslowerinnorthernfrontierzonesandhadatleastpartially
calBC)stillretainedaforaginglifestyleformorethan1500years relied on Mesolithic subsistence or good relations to hunter-
aftertheemergenceoftheLBK,despitecontactswithCentralEu- gatherer-fisher neighbors compared with Central Europe. Pre-
ropean farmers (Bogucki and Crabtree, 2004; Whittle and sumably, subsistence was introduced by small pioneer groups
Cummings, 2007; Rowley-Conwy, 2011). There is a considerable from the former LBK area, using what Rowley-Conwy (2011) de-
debateinthearchaeologicalliteratureexaminingwhetherfarming scribes as ‘lurches of advance’, which encouraged local hunter-
wasintroducedbutfailed,afterwhichpeoplerevertedbacktothe gatherer groups of the Ertebølle culture to slowlyadopt the new
original/alternative subsistence strategies (Rowley-Conwy, 2011). lifestyle. Archaeologists have described contacts of the Ertebølle
However,farmingwaseventuallyestablishedinSouthernScandi- culturewithsouthernfarmingcommunities,suchastheLBKand
naviawiththeemergenceoftheFunnelBeakerculturecomplexor the Ro€ssen culture, which likely introduced new farming tech-
Trichterbecher Kultur (~4100e2650 cal BC), which had a clear niquesandsubsistencestrategiesinSouthernScandinaviaresult-
emphasis on animal husbandry over cultivation of crops, while inginthemanifestationoftheFunnelBeakercomplexabout1500