1 An ancestral miR-1304 allele present in Neanderthals regulates genes involved in enamel ...

38 Pages·2012·0.66 MB·English

by  Yolanda Espinosa

An Ancestral miR-1304 Allele Present in Neanderthals
Regulates Genes Involved in Enamel Formation and Could
Explain Dental Differences with Modern Humans
Maria Lopez-Valenzuela,1 Oscar Ram´ırez,1 Antonio Rosas,2 Samuel Garc´ıa-Vargas,2
Marco de la Rasilla,3 Carles Lalueza-Fox,1 and Yolanda Espinosa-Parrilla*,1
1Institut de Biologia Evolutiva, Universitat Pompeu Fabra—Consejo Superior de Investigaciones Cient´ıficas, Barcelona, Catalonia,
Spain
2Paleoanthropology Group, Department of Paleobiology, Museo Nacional de Ciencias Naturales, Consejo Superior de
Investigaciones Cient´ıficas, Madrid, Spain
3A´rea de Prehistoria, Departamento de Historia, Universidad de Oviedo, Oviedo, Spain
*Corresponding author: E-mail: [email protected]. D
Associate editor: Anne Stone Row
enlo
sa
Abstract earded fro
Genetic changes inregulatory elements are likely to resultin phenotypiceffects that might explain population-specific as cm
well as species-specific traits. MicroRNAs (miRNAs) are posttranscriptional repressors involved in the control of almost h http
s
ethveemry brieomloagiincahlipgrholycescso.nTsheervseedsmatallthneonsceoqduienngcReNlAevseal.reMpicreroseRnNtAin-mveadriioautesdphreyglouglaetnieotnicdgerpouenpds,sanond apelarrfegcetnmumatbcehrinogf ar://ac
ta
betweenthesevennucleotidesofitsseedregionandthetargetsequenceusuallylocatedatthe3#untranslatedregionof ide
cm
the regulated gene. Hence, even single changes in seed regions are predicted to be deleterious as they may affect miRNA lic
e.o
target specificity. In accordance to this, purifying selection has strongly acted on these regions. Comparison between the u
p
genomes of present-day humans from various populations, Neanderthal, and other nonhuman primates showed an .c
o
m
miRNA,miR-1304,thatcarriesapolymorphismonitsseedregion.TheancestralalleleisfoundinNeanderthal,nonhuman /m
primates,atlowfrequency(;5%)inmodernAsianpopulationsandrarelyinAfricans.UsingmiRNAtargetsiteprediction b
e
athlgaonrittehnm-fso,ldw,einfdoiucnadtintghaatntihmepdoerrtiavnetdfuanlleclteioinnaclreeavsoelsuttihoennfourmmbieRr-1o3f0p4u.tAantiavleystisarogfetthgeepnreesdfioctretdhteadrgeertivsefdormdeiRriNveAdmmoiRre- /article
1304 indicates an association with behavior and nervous system development and function. Two of the predicted target -a
b
s
genes for the ancestral miR-1304 allele are important genes for teeth formation, enamelin, and amelotin. MicroRNA tra
c
overexpression experiments using a luciferase-based assay showed that the ancestral version of miR-1304 reduces the t/2
enamelin- and amelotin-associated reporter gene expression by 50%, whereas the derived miR-1304 does not have any 9/7
effect.DeletionofthecorrespondingtargetsitesformiR-1304inthesedentalgenesavoidedtheirrepression,whichfurther /1
7
9
supportstheirregulationbytheancestralmiR-1304.Morphologicalstudiesdescribedseveraldifferencesinthedentitionof 7
/1
Neanderthals and present-day humans like slower dentition timing and thicker enamel for present-day humans. The 0
6
9
observed miR-1304-mediated regulation of enamelin and amelotin could at least partially underlie these differences 1
4
betweenthetwoHomospeciesaswellasotherstill-unraveledphenotypicdifferencesamongmodernhumanpopulations. 8 b
y
Key words: Neanderthal, gene regulation, miR-SNP, microRNA, miR-1304, tooth development. g
u
e
s
t o
n
Introduction eithermicroRNAs(miRNAs)ortheirregulatorymachinery 1
3
are major contributors to phenotypic diversity in human A
Investigating the genetic differences associated with phe- p
notypicdiversityamonghomininsisacrucialsteptoward populationsandmaythusbeimportantsourcesofpheno- ril 2
typic variation and have a role in the pathophysiology of 01
the understanding of human adaptation and evolution. 9
several disorders (Borel and Antonarakis 2008).
Genetic and genomic alterations in regulatory regions
miRNAsaresmallnoncodingRNAsof19–25nucleotides
are a significant source of phenotypic diversity underlying
in length in their mature form, processed from a longer
important interindividual and interspecies differences
hairpinstructure,thatactasposttranscriptionalregulators
(Carroll 2008; Hindorff et al. 2009). This record has been
recently confirmed in primates by the discovery that the of gene expression by either mRNA degradation or trans-
human loss of specific regulatory DNA, in particular the lational repression (Krol et al. 2010). It is estimated that
lossofnoncodingRNAwithenhancerfunction,associates miRNAs regulate more than 30% of all protein-coding
with the appearance of specific human traits such as the genes, building complex regulatory networks that control
expansion of specific brain regions (McLean et al. 2011). almost every cellular process (Filipowicz et al. 2008).
Increasingevidencesupportsthatallelicchangesinvolving MicroRNAs act by means of partial complementarity to
©TheAuthor2012.PublishedbyOxfordUniversityPressonbehalfoftheSocietyforMolecularBiologyandEvolution.Allrightsreserved.Forpermissions,please
e-mail:[email protected]
Mol.Biol.Evol.29(7):1797–1806.2012 doi:10.1093/molbev/mss023 AdvanceAccesspublicationJanuary27,2012 1797
MBE
Lopez-Valenzuela et al. · doi:10.1093/molbev/mss023
miRNAbindingsitesusuallylocatedinthe3#untranslated andthespectrumoftargetgenespredictedforthederived
regions (UTRs) of their target genes (Bartel 2004). Perfect and ancestral versions of this miRNA. By means of func-
complementaritybetweenthetargetsequenceofthereg- tional studies, we show repression of a cluster of dental
ulated messenger RNA and the so-called ‘‘seed’’ region— genes by the ancestral version of miR-1304 illustrating
nucleotides2through7or8fromthe5#endofthemature how a single nucleotide change in a regulatory element
miRNA—isthoughttodeterminesuccessfulbindingand,to- may underlie particular phenotypic differences.
getherwiththestabilityoftheRNAhybrids,arethebasisof
many miRNA target site prediction algorithms (Brennecke Materials and Methods
etal.2005;Lewisetal.2005).Strongpurifyingselectionacts
Sequencing of Chimpanzee miR-1304 and
on the mature miRNA and particularly on nucleotides cor-
Neanderthal AMTN 3# UTR
responding to the seed region, where no mutation is toler-
miR-1304 was polymerase chain reaction (PCR)-amplified
atedasitwouldmostlikelyproduceachangeinthetarget
from genomic DNA of three chimpanzee individuals using
spectrum that could give rise to the emergence of a novel D
primers listed in supplementary table 1 (Supplementary o
miRNA (Chen and Rajewsky 2006; Liu et al. 2008; Quach w
et al. 2009). Conservation of miRNAs through evolution is Materialonline).Eightclonesperindividualweresequenced nlo
welldocumentedandhasbeenusedforthediscoveryofho- using standard procedures. ade
massoulomgeodusthmaitRNcoAnssearcvreodssmdiifRfeNreAntmpahyylhoagveenehtiigchgrfouunpctsi.oIntaisl andTehrethtaalrAgeMtTsiNtewfoasrPmCiRR--a1m30p4lifioendtihnea3N#eUanTdReortfhtahlespNeec-- d from
imen(SD-1253)fromElSidro´nsite(Asturias,Spain),and55 h
relevanceandhencemanypreviousresearcheffortsfocused ttp
clonesweresequenced (supplementaryfig.1,Supplemen- s
on finding and characterizing those miRNAs (Grad et al. tary Material online) following a previously described ://a
2003;Berezikovetal.2005;ChenandRajewsky2006).None- c
a
methodology (Lalueza-Fox et al. 2007) using specific pri- d
theless, the identification of species-specific miRNAs may e
mers(supplementarytable1,SupplementaryMaterialon- m
help to understand evolutionary novelties among different ic
line).Thisbonesamplehasbeenusedtoretrieve;14,000 .o
phylogenetic groups. Several studies reach the conclusion u
that miRNAs are strongly conserved among primates, but protein-coding positions (Krause et al. 2007; Lalueza-Fox p.c
o
et al. 2008, 2009; Burbano et al. 2010), as well as 0.1% of m
still there is a set of miRNAs that are found only in thenucleargenome(Greenetal.2010),andthecomplete /m
present-day humans and thus are good candidates to con- b
e
tBreibreuztiekotvoehtuaml.a2n0-0s5p;eLciiuficetphale.n2o0t0y8p;eBsra(Bmeenietwri2c0h10e;tLailn.2e0t0a5l.; mcointotacmhoinnadtriioanl gheansombeeen(Berisgtgims aetteadl.t2o00b9e).0T.2h7e–0d.e2g9r%eefoorf /article
2010).The first published draft of theNeanderthal genome the mitochondrial DNA and 2% upper bound for autoso- -ab
mal DNA capture (Krause et al. 2007; Lalueza-Fox et al. s
revealedthat present-day humans differ from Neanderthals tra
2008); these low figures render this specimen one of the c
bmyiRa-1n3u0c4leo(Gtidreeesnubesttiatul.ti2o0n10in)tthheatsetehderreefgoioreniotfims ilcikroelRyNtAo most suitable Neanderthal samples for targeted paleoge- t/29/7
change the spectrum of target genes for miR-1304. netic analysis. /1
7
9
Neanderthalsaretheclosestknownevolutionaryrelatives 7
Firefly Luciferase Constructs and Mutagenesis /1
of modern humans. They inhabited parts of Europe and 0
The3#UTRsofENAM,AMTN,andGAD1werePCR-amplified 69
WesternAsiaduringasuccessionofclimaticcycles,exhibit- 1
ingbothbehavioralandmorphologicaladaptationsprobably from genomic DNA with Platinum Taq DNA polymerase 48 b
relatedtocoldclimates,untiltheirextinctionaround30,000 (reInsvtriticrtoigoenns,iCtearaltsbthade,5C#Ae)n,dus(isnugppplreimmeernstacorynttaabinlein1g,aSnupXpblea-I y gu
yearsago(Mellars2004;Finlaysonetal.2006).Themorpho- e
s
logicalfeaturesthatdistinguishNeanderthalsfromotherhu- mentaryMaterialonline).PCRfragmentswerelaterpurified, t o
mans, including specific craniofacial and dental traits, first XbaI-digested, and cloned into pGL4.13 vector (Promega, n 1
3
appear in the European fossil record around 4,00,000 years Madison, WI) downstream of the firefly luciferase reporter A
p
gagenoo(mStricingdeartaa,ntdhHesuebHlinom19o99p;oHpuublalitnio2n0s09d)i.vEesrtgiemdatbeedtwfreoemn gdeenscer.iMbeudt(aMnturien˜poos-rGteirmpelnasomeitdasl.w2e0r0e9g)ewnietrhattehdeaQsupirkeCvhioaunsgley ril 201
site-directedmutagenesiskit(Stratagene,LaJolla,CA),using 9
440,000 and 270,000 years ago (Endicott et al. 2010; Green
either the AMTN or the ENAM pGL4.13 construct as
etal.2010).TheanalysisoftheNeanderthalgenomehasre-
a template and primers carrying the desired deletions
vealedthatabout80protein-codinggenesshowfixedamino
(supplementary table 1, Supplementary Material online).
acid changes between Neanderthals and modern humans
(Green etal. 2010). Phenotypes evolve by functional differ-
ences in proteins but also do largely through mutations in Cell Culture and Transfection
regulatory regions (Carroll 2008); thus, it seems clear that HeLacellsweremaintainedinDulbecco’smodifiedEagle’s
genetic differences related to the distinctive Neanderthal medium supplemented with 10% fetal bovine serum, 100
phenotype should not be restricted to a set of protein- units/ml penicillin, and 100 lg/ml Streptomycin (GIBCO;
coding genes and that the analysis shall be broadened to Invitrogen). Cotransfection optimization has been previ-
incorporate gene regulation as well. ously described (Guidi et al. 2010). Briefly, HeLa cells were
Here, we analyze the primate-specific miR-1304 by seededat1.3(cid:1)104cellsperwellin96-wellplatesandco-
studyinggeneticvariationofthehumanmiR-1304‘‘locus’’ transfected 24 h later with the Firefly reporter constructs
1798
MBE
An Ancestral miR-1304 Allele Regulates Dental Genes · doi:10.1093/molbev/mss023
FIG. 1. Genomic sequence conservation of miR-1304 among primates. First row of the figure shows the RNA sequence of mature reference
humanmiR-1304.TheseedregionandthenucleotidechangeinitarehighlightedingrayonthegenomicDNAalignment.hsa,Homosapiens;
nea,Homo neanderthalensis; ptr, Pantroglodytes; ggo,Gorillagorilla;ppy, Papiopygmaeus; mmu, Macaca mulatta.
describedaboveortheemptypGL4.13vector(24ng),the Results
RenillareporterplasmidpGL4.75(3ng),and10nMmiRNA
Conservation of miR-1304 among Primates
mimicforderivedmiR-1304,ancestralmiR-1304,andneg- D
ativecontrols#2and#4(miRIDIAN;Dharmacon,Lafayette, As noted by Green et al. (2010), the Neanderthal draft ow
CO) using Lipofectamine 2000 (Invitrogen). genome sequence differs from the reference genome of nlo
a
present-day humans in one base in the seed region of d
e
LTuhceifaecrtaisveityAoctfivFiitreyflAysasnayd Renilla luciferases was deter- stpheonpdriinmgagteen-sopmeciicfisceqmuieRnNcAefomriRN-e1a3n0d4er(tfihga.ls1)is. GThCeCTcoCrGreA- d from
andGCCTCAAforthereferencehumansequence.Wetook h
mined24haftertransfectionusingtheDual-GloLuciferase ttp
advantageoftherecentlycompletedgenomesequencingof s
AssaySystem(Promega).Relativereporteractivitywasob- four primates—M. mulatta, P. troglodytes, G. gorilla, and P. ://a
tainedbynormalizationtotheRenillaluciferaseactivity.In c
a
pygmaeus—to compare primate sequences orthologous d
order to correct for vector-dependent unspecific effects, e
m
withhumanmiR-1304(fig.1).Thesequencealignmentcon-
eachrelativereporteractivitywasnormalizedtotheempty ic
firmedthatthe referencehumangenomebear thederived .o
vector cotransfected with the corresponding miRNA u
p
(Guidi et al. 2010). Results were then compared with state at this nucleotide position, whereas Neanderthal and .c
o
other nonhuman primates share the ancestral state. The m
the mean of the two negative controls. Each experiment /m
restofthehairpinsequencewasidenticalbetweentheNe-
was done in triplicate, and at least three independent ex- b
e
perimentswereperformedforeachmiRNA.Statisticalsig- anderthals and the reference human genome and showed /a
nificancewasdeterminedusingStudent’st-test(P,0.05). fewchangesintheotherfourprimates.Inthecaseofchim- rticle
Bonferroni correction for multiple comparisons was ap- panzee, the published shotgun assembly (March 2006 -a
b
Pan_troglodytes-2.16x)showedadeletionof26nucleotides s
pliedtakingintoaccounttheanalysisoftwoindependent tra
rreiap,otrhteercogerrneecsteadndletvweloomf siRigNniAficmanimceicws.aUsssientgutpheesqeucarlitteo- tpoagrtesthoefrthweitmhiaRnNAin.sTerotivoenriftyhathteseptosaspibairltitythtehattwcohirmesptainng- ct/29/7
0.0125 (four comparisons). zeehadlostthismiRNA,wesequencedthreechimpanzees’ /1
7
DNAsand,afteranalysisofeightclonesperindividual,iden- 9
7
Computational Methods tifiedaconsensussequenceidenticaltothegorillamiR-1304 /10
6
sequence that differs in only one nucleotide position from 9
Targets were predicted using the web-based prediction 1
methodsTargetScan(www.targetscan.org,release5.1,Fried- theancestralmiR-1304 (fig. 1) indicating thatchimpanzees 48 b
manetal.2009)andTargetRank(www.hollywood.mit.edu/ likely also have a functional copy of miR-1304. y g
u
targetrank, Nielsen et al. 2007) on the human genome as- e
s
sembly (NCBI36/hg18, March 2006). Genomic coordinates Analysis of the Genetic Variation of miR-1304 in t o
n
are according to the following assemblies: Homo sapiens Human Populations 1
3
GRCh37/hg19, Pan troglodytes CHIMP2, Gorilla gorilla gor- We checked for genetic variation at the miR-1304 locus A
p
Gor3, Pongo pygmaeus PPYG2, Macaca mulatta MMUL_1. among human populations using the single nucleotide ril 2
Sequences, unless sequenced by ourselves, were obtained polymorphism (SNP) database. We found one SNP, 01
9
from the University of California Santa Cruz (UCSC) Ge- rs79759099, at this particular position in the seed region
nomeBrowser(http://www.genome.ucsc.edu)andEnsembl indicating that this change is not fixed in human popula-
Genome Browser release 61 (www.ensembl.org). Pathway tions. Using the new release of the 1000 Genomes Project
analysiswasperformedwiththeIngenuityPathwayAnalysis (June2011datarelease)with1,094individualsrepresenting
software(IPA)version6.3(www.ingenuity.com).Humange- 14populationsworldwide,wefoundthatalthoughallthe
netic variation on miR-1304 was assessed using genotypes individuals in the European (GBR, FIN, IBS, CEU, and TSI),
for 1,094 individuals in the June 2011 data release of the Colombian (CLM), Mexican (MXL), and Kenyan (LWK)
1000GenomesProject(www.1000genomes.org).Exploration populationsonlypresentedthederivedallele,theancestral
for natural selection signatures in the human genome was allele of miR-1304 was present as the minor allele in the
performedby the analysisofdatafrom HapMap, usingthe Asian Japanese (JPT, MAF 5 0.067) and Chinese (CHB,
UCSCbrowser,andfromthe53populationsoftheHuman MAF50.072;CHS,MAF50.05)populationsand,atvery
GenomeDiversityPanel(HGDP),usingtheHGDPselection lowfrequency,intheYoruban(YRI,MAF50.028);Puerto
browser (http://hgdp.uchicago.edu/) (Pickrell et al. 2009). Rican (PUR, MAF 5 0.009), and African-American (ASW,
1799
MBE
Lopez-Valenzuela et al. · doi:10.1093/molbev/mss023
Table 1.Target GenePrediction fortheAncestral VersionofmiR-1304 byTargetRank.
NonconservedSeed
ConservedSeedMatches Matches
M8 A1 M8 A1
Conserved
Gene TargetGene Score Total 8mer 7mer 7mer 6mer 8mer 7mer 7mer 6mer inPrimates
PTGFRN ProstaglandinF2receptornegativeregulator 0.337 1 0 0 0 0 1 0 0 0 H,N,C
DEK DEKoncogene 0.309 1 0 0 0 0 1 0 0 0 H,N,C
AMTN Amelotin 0.302 1 0 0 0 0 1 0 0 0 H,N,C
CD24 CD24antigenprecursor 0.281 1 0 0 0 0 1 0 0 0 H,N
PIK3AP1 Phosphoinositide-3-kinaseadaptorprotein1 0.281 1 0 0 0 0 1 0 0 0 H,N,C
C1orf85 KidneypredominantproteinNCU-G1 0.263 2 0 0 0 0 1 1 0 0 H,N,C
CAMKK2 Calcium/calmodulin-dependentproteinkinase 0.261 2 0 0 0 0 1 1 0 0 H,N,C
KATNAL1 Kataninp60subunitA-like1 0.259 1 0 0 0 0 1 0 0 0 H,N,C
D
MGC13017 HypotheticalproteinLOC91368 0.259 1 0 0 0 0 1 0 0 0 H,N,C o
w
FOXH1 ForkheadboxH1 0.257 1 0 0 0 0 1 0 0 0 H,N,C n
DLST DihydrolipoamideS-succinyltransferase 0.253 1 0 0 0 0 1 0 0 0 nr lo
a
ENAM Enamelin 0.253 1 0 0 0 0 1 0 0 0 H,N,C de
d
RCI1Co8rAf21 RCehsriostmanocsoemtoei1nhoipbeitnorrseaodfinchgoflrianmesete2r1ase8 00..225531 12 00 00 00 00 10 01 01 00 HH,,NN,,CC from
TNNI1 TroponinI 0.240 3 0 0 0 0 1 0 0 2 H,N,C h
RBPMS2 RNA-bindingproteinwithmultiplesplicing2 0.237 2 0 0 0 0 0 0 1 1 nr ttp
s
ITGCFFB4Pa3 TInrsaunlsincr-ilpikteiognrofawctthorfa4ctorbindingprotein3 00..223377 11 00 00 00 00 00 11 00 00 HH,,NN,,CC ://ac
a
FIZ1 FLT3-interactingzincfinger1 0.236 2 0 0 0 0 0 1 1 0 H,N,C d
e
CABP7 Calcium-bindingprotein7 0.210 1 0 0 0 0 1 0 0 0 H,N,C m
ic
CLCF1 Cardiotrophin-likecytokinefactor1 0.210 1 0 0 0 0 1 0 0 0 H,N,C .o
u
GRAP GRB2-relatedadaptorprotein 0.210 1 0 0 0 0 1 0 0 0 H,N,C p
KLF15 Kruppel-likefactor15 0.210 1 0 0 0 0 1 0 0 0 H,N,C .co
m
LRRN2 Leucine-richrepeatneuronal2 0.210 1 0 0 0 0 1 0 0 0 H,N,C
/m
PI4K2A Phosphatidylinositol4-kinasetype2alpha 0.210 1 0 0 0 0 1 0 0 0 H,N,C b
e
SLAMF8 SLAMfamilymember8 0.210 1 0 0 0 0 1 0 0 0 H,N,C /a
TNIP2 A20-bindinginhibitorofNF-kappaBactivation2 0.210 1 0 0 0 0 1 0 0 0 H,N,C rtic
ATP1B3 Na1/K1ATPasebeta3subunit 0.209 1 0 0 0 0 0 1 0 0 H,N,C le
L3MBTL3 Lethal(3)malignantbraintumor-likeprotein3 0.209 1 0 0 0 0 0 1 0 0 H,N,C -ab
s
SOX17 SRY-box17 0.209 2 0 0 0 0 0 0 1 1 nr tra
TMED4 Transmembraneemp24proteintransportdomain 0.209 1 0 0 0 0 0 1 0 0 H,N c
DAPL1 Death-associatedprotein-like1 0.203 1 0 0 0 0 1 0 0 0 H,N,C t/29
EDF1a Endothelialdifferentiation-relatedfactor1 0.203 1 1 0 0 0 0 0 0 0 H,N,C /7
/1
FAM90A1 HypotheticalproteinLOC55138 0.203 1 0 0 0 0 1 0 0 0 H,N,C 7
9
TADA3L Transcriptionaladaptor3-like 0.203 1 0 0 0 0 1 0 0 0 H,N 7
/1
0
NOTE.—Underlinedarethetwogenesinvolvedinteethformation.A1indicatespresenceofanadenosineoppositetomiRNAbase1;M8indicatesaWatson–Crickmatchto 69
miRNAbase8.ConservedinPrimatesindicatesconservationofthebindingtargetsiteamongdifferentprimatespecies:H,human;N,Neanderthal;C,chimpanzee;nr,no 1
4
Neanderthalreadsavailable. 8
aTargetgenepredictioncommontoTargetScanandTargetRank. by
g
u
MAF50.008)populations(supplementarytable2,Supple- only four genes (LCORL, RIMBP2, EDF1, and TCF4, the last e
s
mentaryMaterialonline).Furthermore,sincethisdifferent two being common to both prediction programs, table 1). t o
n
allelic distribution could be the result of selective sweeps Alimitationofthesepredictionsisthattheywereperformed 1
3
withinrecenthumanpopulations,welookedforsignatures onthemodernhumangenome,thuswecheckedifpredicted A
p
of selection for the derived miR-1304 by the study and targetgeneshadidenticalbindingsitesfortheancestralmiR- ril 2
comparison of linkage disequilibrium (integrated haplo- 1304 in the chimpanzee and Neanderthal genomes. Among 01
9
type score), population differentiation (F ), and the fre- the37differentpredictedtargetgenesforthisvariant,there
st
quency of rare variants (Tajima’s D) along the genomic were 29 genes with an identical binding site in chimpanzee,
region using public data from HapMap and HGDP. How- whichweconsideredavalidproxyforconservationinNean-
ever,neitherasignificantexcessofrarevariantsnorsignif- derthalsaswell.Asfortheothereightgenes,theNeanderthal
icant population differentiation indexes compatibles with genomeassemblyindicatedthatinfourcases(CD24,TMED4,
a selective sweep were found in the region. TAD3L,andRIMBP2),thetargetsiteisidenticalbetweenthe
two hominins, whereas for the other four genes (DLST,
Analysis of Target Gene Predictions for miR-1304 RBPMS2, LCORL, and SOX17) complete Neanderthal reads
To assess the variation in the target gene spectrum for the werenotavailable(table1).Interestingly,forthederivedver-
ancestralandderivedmiR-1304,weuseddifferentprediction sion of miR-1304, we found a large increase in the putative
algorithms based on seed sequence matching, namely Tar- targetsincomparisonwiththeancestralversion.TargetRank
getScan and TargetRank. For the ancestral miR-1304, Targe- predictionsgeneratedalistof515targetgenes,whereasTar-
tRank predicted 35 target genes and TargetScan predicted getScan generated 140, with an intersection of 79
1800
MBE
An Ancestral miR-1304 Allele Regulates Dental Genes · doi:10.1093/molbev/mss023
Table 2.Top Associations ofthe Predicted TargetGenes fortheDerived VersionofmiR-1304 withBiologicalProcesses.
TopNetworks Score
Cellmorphology,cellularfunctionandmaintenance,cellularmovement 51
Neurologicaldisease,behavior,geneticdisorder 51
Lipidmetabolism,moleculartransport,smallmoleculebiochemistry 30
Cellcycle,cellulardevelopment,cellulargrowthandproliferation 5
Auditorydisease,geneticdisorder,neurologicaldisease 5
DiseasesandDisorders Pvalue Molecules
Geneticdisorder 5.93310205to4.04310202 24
Neurologicaldisease 5.93310205to4.04310202 14
Cancer 1.27310203to3.10310202 5
Infectionmechanism 2.12310203to1.81310202 6
Cardiovasculardisease 4.49310203to4.40310202 4
D
MolecularandCellularFunctions Pvalue Molecules o
w
Cellularassemblyandorganization 3.21310204to4.40310202 16 n
Cellcompromise 3.89310204to3.10310202 6 loa
d
Cellulargrowthandproliferation 1.13310203to4.83310202 11 e
d
Cellulardevelopment 2.00310203to4.68310202 9 fro
Geneexpression 2.12310203to4.83310202 19 m
h
PhysiologicalSystemDevelopmentandFunction Pvalue Molecules ttp
s
NTiesrsvuoeumsosyrsptheomlodgeyvelopmentandfunction 55..4455331100220044ttoo44..4803331100220022 183 ://ac
a
Tumormorphology 1.13310203to3.97310202 5 de
Connectivetissuedevelopmentandfunction 3.48310203to4.83310202 8 m
Behavior 4.49310203to4.49310203 1 ic.o
u
p
NOTE.—Thefivemostsignificantassociationswithmolecularnetworksandwithdifferentcategoriesofbiologicalfunctionsareshown(IPAsoftware). .c
o
m
(supplementary table 3, Supplementary Material online). in theregulation ofENAM and AMTN, it is necessary that /m
b
e
Next,weanalyzedtheassociationofthese79potentialtarget Neanderthals match humans in the target site sequences /a
genesforthederivedmiR-1304withbiologicalprocessesus- forthesegenes.InthecaseoftheENAMNeanderthalgene, rtic
le
ing the IPA software. The program was interrogated for en- the sequence corresponding to the target site for this -a
b
richmentinbiologicalfunctionsandmolecularnetworks,and miRNA is identical to the modern human sequence, and s
the statistical significance of the associations was calculated the 3# UTR exhibits only two changes in its vicinity trac
with the right-tailed Fisher’s exact test. As shown in table 2 (fig. 2A). The first is a C to G substitution corresponding t/29
/7
and supplementary fig. 2 (Supplementary Material online), to a common SNP also present in modern humans /1
7
some of the most relevant associations for the derived (rs7664896)located7bpfromtheseedregionbindingsite. 9
7
miR-1304 target genes were found with genetic disorders ThesecondisaGtoAnucleotidechange,whichisamong /10
6
(24genes)andneurologicaldiseases(14genes)aswellasner- themostcommonformsofpostmortemDNAdamageand 9
1
voussystemdevelopmentandfunction(13genes).Asfaras may be considered an artifact induced by cytosine deam- 48
top networks, the highest significance was found in one re- ination occurring in the complementary strand (Hofreiter by
g
lated to neurological disease and behavior and another one et al. 2001; Briggs et al. 2007). Neither of these changes ue
regardingcellularfunctionandmaintenance(supplementary is predicted to interfere with the correct binding of the st o
fig.2,SupplementaryMaterialonline).Despitethesmallnum- miRNA. In the case of the AMTN gene, the sole available n 1
ber of predicted target genes, we observed that 2 of the 35 readfromVindijaNeanderthalendsupatthebeginningof 3 A
gmeinRe-s13p0r4edbicyteTdartgoetbReanrke—guElaNteAdMbyantdheAaMnTceNsttrhalatvecrosdioenfoorf tnhueclteaortgiedtescehqaunegnecsetfhoartm, aiRs-s1t3a0te4danbedfodriesp,lcaoyusltdwboeGantoarA- pril 20
1
the proteins enamelin and amelotin, respectively—are in- tifactduetopostmortemDNAdamage(fig.2A).Toascer- 9
volvedinteethformation.Thefindingoftwogenesbothin- tainiftheextinctHomohadacompletebindingsiteforthe
volvedinaprocessofsuchhighspecificityasodontogenesis ancestralmiR-1304atAMTN3#UTR,wesequencedpartof
withinarelativelysmallsetofputativetargetswasinteresting,
this region in a Neanderthal specimen (SD-1253) from El
particularlygiventhatthebest-describeddifferencesbetween Sidro´n site (Asturias, Spain) dated to about 49,000 years
Neanderthalsandmodernhumansarerelatedtocranialand
ago and extracted under controlled conditions (Rosas
dentaltraits.Thus,amongthepredictedtargetgenes,wefo-
etal.2006;Lalueza-Foxetal.2011).Theanalysisof55clones
cused on the study of the ENAM and AMTN genes.
showedthattherecognitionsiteformiR-1304intheAMTN
gene is identical in both hominin groups (supplementary
Analysis of the 3# UTR of the ENAM and AMTN fig. 1, Supplementary Material online). The 3# UTRs of
Genes in Neanderthal ENAM and AMTN are also highly conserved among other
Because the human genome was used as a reference for primatesstudied(chimpanzee,gorilla,orangutan,andrhe-
predicting the role of the ancestral variant of miR-1304 sus monkey), and all but the rhesus monkey carries the
1801
MBE
Lopez-Valenzuela et al. · doi:10.1093/molbev/mss023
A
D
o
w
n
B lo
a
d
e
d
fro
m
h
ttp
s
://a
c
FIG.2.GenomicsequenceconservationofENAMandAMTN3#UTRsamongprimates.(A)3#UTRoftheENAMandAMTNgenes,thetarget ad
e
sequencethatbindstheseedregionofmiR-1304ishighlightedingray.hsa,Homosapiens;nea,Homoneanderthalensis;ptr,Pantroglodytes; m
ggo,Gorillagorilla;ppy,Papiopygmaeus;mmu,Macacamulatta.(B)SequencealignmentsfortheancestralmiR-1304(Anc.miR-1304)showing ic.o
the predicted bindingwiththe 3#UTRs ofthe Neanderthal ENAMand AMTNmRNAs. up
.c
o
m
/m
exact matching sequence for the ancestral miR-1304 seed miR-1304 in the ENAM and AMTN pGL4.13 constructs b
e
region(fig.2).Accordingly,theseedregionofthisversionof and cotransfected the plasmids with the respective /a
miR-1304 presents perfect complementarity with its pre- miRNAs.Astatisticallysignificantrecoveryoftheluciferase rtic
le
dictedtargetsiteregionatthe3#UTRofENAMandAMTN activitywasobservedwhentheancestralmiR-1304wasco- -a
b
s
(fig. 2B) and might be regulating both dental genes in transfectedwiththedeletedENAMandAMTNplasmidsin tra
Neanderthal as well as in other nonhuman primates and comparison with the wild-type constructs, reaching levels ct/2
humans carrying the ancestral allele of miR-1304. ofrecoveryofabout100%(fig.3B).Theobservedrescueof 9
/7
thequantitativephenotypefurthersupportsthepredicted /1
7
Functional Screening of miR-1304 Target Sites in repression of these two dental genes by this version of 97
/1
ENAM and AMTN miR-1304. 0
6
9
To investigate the interaction between ENAM and AMTN 1
4
mRNAsandthedifferentversionsofmiR-1304,functional Analysis of the Genetic Variation of ENAM and 8 b
validation of the predicted ancestral miR-1304 target site AMTN in Human Populations y g
u
was performed using a dual-luciferase assay in HeLa cells. To further characterize the gene-miRNA interaction, we e
s
A luciferase reporter pGL4.13 construct carrying the 3# analyzed genetic variation on the ENAM and AMTN t o
n
UTRofeithertheENAMortheAMTNgeneswascotrans- miR-1304targetsitesequencesinpresent-daypopulations 1
3
fected with the corresponding miRNA mimic: ancestral usingthe1000GenomesProjectdata.Althoughwedidnot A
p
miR-1304, derived miR-1304, or two different control find genetic variation in the target site of AMTN, the ril 2
miRNAs. As shown in figure 3A, a statistically significant ENAMtargetsiteharborstheSNPrs117342040(G/Awith 01
9
reductionoftheluciferaseactivitywasobservedforENAM Abeingtheminorallele).Interestingly,theaffectednucle-
and AMTN when cotransfected with the ancestral otideinthetargetsiteofENAMislocatedexactlymatching
miR-1304 as compared with the modern miR-1304 (P , thedescribedchange inthemiR-1304 seedregioninsuch
0.0125, Bonferroni corrected Student’s t-test). In the case awaythattheminoralleleofrs117342040wouldinterfere
of the ENAM construct, the associated luciferase activity withtheproperbindingoftheancestralalleleofmiR-1304.
descended to 51% and for the AMTN construct to 39% Moreover, the allele frequencies for this SNP are differen-
when cotransfected with the ancestral miR-1304, which tially distributed among human populations being absent
is compatible with a strong repression of the ENAM and inEuropeanpopulations,veryrareinAfricans(onecarrier
AMTNgenesbythis regulator.Finally,inordertodemon- ofAfricanoriginfromKenya),andpresentinabout2.5%of
strate the specificity of the binding of the ancestral Asian individuals (supplementary table 2, Supplementary
miR-1304 to the ENAM and AMTN 3# UTRs, we used Material online). As this variant is found in populations
site-directedmutagenesistodeletethecorrespondingtar- where the ancestral version of miR-1304 is also detected,
get sequences that bind the seed region of the ancestral westudiediftheywerecorrelatedusingahypergeometric
1802
MBE
An Ancestral miR-1304 Allele Regulates Dental Genes · doi:10.1093/molbev/mss023
A phenotypic implications contributing sometimes to hu-
man disease susceptibility (Borel and Antonarakis 2008).
Since one miRNA can have multiple targets, SNPs located
on the miRNA regions important for target recognition
would be expected to exhibit a broader biological effect
than SNPs on the target sequences. So far, very few func-
tional variants in miRNA genes have been described, and
only a small number of studies have been devoted to de-
scribetheirfunctionalconsequences(Jazdzewskietal.2008;
Sun et al. 2009; Vinci et al. 2011).
AnalysisoftheNeanderthalgenomesequenceledtothe
detectionofamiRNA,miR-1304,thatdiffersfromtheref-
erencehumanmiR-1304sequenceatonenucleotideposi- D
o
tion in the seed region (Green et al. 2010). Interestingly, w
n
analysis of 1000 Genomes Project data revealed that this lo
a
B nucleotidechangeisnotfixedinpresent-dayhumanpop- de
d
ulations,anditturnedouttobeanSNP(rs79759099)pres- fro
entinAsianpopulations(allelefrequencyofabout0.06).In m
h
EuropeanandEast-Africanpopulations,theancestralallele ttp
s
was not found, and populations with West-African origin ://a
carrytheancestralallelebutatverylowfrequencies.Since c
a
d
this distribution schema could be the result of selective e
m
sweepswithinrecenthumanpopulationscompatiblewith ic
.o
a beneficial role for the new derived miR-1304 allele, we u
p
looked for signatures of selection along the region using .c
o
m
HapMapand HGDPdata, but no evidences for a selective /m
sweep, such as an excess rare allele variants or a block of b
e
epxlatnenatsiiovenlfinokratgheedpisreeqseunilciberioufmth,ecoaunldcebsterafoluanlldel.eOitnhAersieaxn- /article
3F#IG.U3T.RTsh.eRaensucletsstroaflvtheersilouncifoeframseiRr-e1p30o4rtetarrgaesstasyENuAseMd taondteAstMtThNe populationscouldbethatitwasreintroducedthroughad- -abs
interaction between ancestral (Anc.) and derived (Der.) miR-1304 mixturewithNeanderthals;however,becauseNeanderthal trac
withtheENAMandAMTNgenes.(A)HeLacellswerecotransfected admixtureisrestrictedtonon-Africanpopulations(Green t/2
with the miRNA mimic of interest, the Renilla reporter plasmid et al. 2010) and given the presence of the ancestral miR- 9/7
pGL4.75,andeithertheemptypGL4.13plasmidorpGL14.3carrying 1304 in some Africans and its low frequency in Asia, the /17
the 3# UTR of GAD1, ENAM, or AMTN followed by the firefly distribution of the ancestral miR-1304 could not solely 97
/1
luciferasereportergene.ThepreviouslyprovenregulationofGAD1 be explained in terms of introgression. Furthermore, 0
6
by miR-7 was used as a positive control. The ratios of Firefly to miR-1304isnotincludedintothe13regionsthathavebeen 91
Rtchoeenntilreloamlslpu.cty(ifBe)rpaGTseLh4elu.1m3biinnadensidncegntocseitteahreefoprmreetshaenenteaodnfcatefwstteorranldoimfrfmeiRrae-ln1iz3ta0t4mioinwmtaiocs iddeernthtiafilesdanadsmcaonddeidrnatheugmeannesfltohwatrweegrioendsesbcertibweedenthNroeuagnh- 48 by gu
removed from the 3# UTRs in the constructs by mutagenesis, and theanalysisonextendedNeanderthalhaplotypeblocksin es
the luciferase assay was repeated. Each experiment was done in modern human genomes (Green et al. 2010). In this sce- t o
n
triplicate, and at least three independent experiments were nario, a combination of admixture in Asia such as the de- 1
3
performed.Datareportedherearethemeans±standarddeviation scribed by Skoglund (Skoglund and Jakobsson 2011) with A
p
orofniincdoerpreecntdioennt(Pex,pe0r.i0m1e2n5,tsc.oSrrigenctifiedcaSnttudaessnotc’siatt-itoensts)aafrteeirndBiocnafteerd- rcsa79c7o5u9ld09b9ebtehinegmaorsettapinlaeudsiabnlecieenxptlpaonlaytmioonr.phisminAfri- ril 201
withan asterisk.Anc. miR-1304, ancestral miR-1304 mimic. 9
miR-1304 is a relatively novel primate-specific miRNA
that was recently discovered at very low levels in human
test and observed that, for the total of 1,094 individuals
embryonicstemcellsandafterdifferentiationofthesecells
analyzed, our finding of 3 individuals having both minor
intoembryonicbodies(Morinetal.2008).Sincethendeep
alleles was significant (P 5 0.0162) meaning that there
RNA sequencing studies have found miR-1304, also lowly
aremorecarriersofbothminorENAMandmiR-1304alleles
expressed,indiversetissuessuchasperipheralblood,brain
than expected by chance.
cortex,andmelanoma(Mart´ıetal.2010;Starketal.2010;
Vazetal.2010),butitstargetsandfunctionremainlargely
Discussion
unknown.Theinsilicoapproachbasedontargetsitepre-
The main basis of miRNA action is its sequence comple- dictionsperformedinthisworkallowedustogaininsights
mentarity with the target regions of the genes that they into the putative functional differences between the two
regulate. Functional SNPs that create or disrupt these versionsofmiR-1304.Oneofthemainrestrictions tofind
miRNA target sites have been shown to have diverse target genes for the ancestral miR-1304 was the dearth of
1803
MBE
Lopez-Valenzuela et al. · doi:10.1093/molbev/mss023
good quality Neanderthal sequence data. The use of the enamel formation, including amelogenin, ameloblastin,
modern humans genome for target gene predictions and enamelin, in the organic enamel matrix. Enamelin is
may have lead us to disregard Neanderthal target genes the largest protein in the enamel matrix of developing
that diverged from human genes in their 3# UTRs; never- teeth.Itisinvolvedinthemineralizationandstructuralor-
theless,wedidnotfindfalsepositiveasalltargetgenesfor ganization of enamel. Amelotin was discovered in mouse,
theancestralmiR-1304thatcouldbetested(33of37,89%) whereitisspecificallyexpressedinmaturation-stageame-
have an identical and conserved binding site in the loblasts.Ithasbeenhypothesizedthatitfunctionsasapro-
Neanderthal and human genomes. tease helping the processing of the enamel matrix at this
Remarkably, we observed a significant difference in the stage(Iwasakietal.2005).Wecanonlyspeculateaboutthe
number of predicted target genes between both miRNAs: function of the ancestral miR-1304 in Neanderthals but,
the derived version of miR-1304 has more than 14 times giventheinvolvementofENAMandAMTNinodontogen-
more predicted targets than the ancestral one, which is esis, their observed repression by the ancestral miR-1304
indicative of a critical functional evolution for miR-1304. couldimplicatethismiRNA,actingtogetherwithotherdi- D
o
Interestingly, the analysis performed using the Ingenuity vergentgenes,inarangeofdifferencesrelatedtoNeander- w
n
software associated the predicted targets of derived thal and modern humans dentition (Macchiarelli et al. lo
a
miR-1304 with biological processes and disorders related 2006;Olejniczaketal.2008).Forexample,theunionsurface de
d
to central nervous system development and function, betweendentineandenamelwasmorecomplexinNean- fro
which suggests that the evolutionary change of miR- derthals than it is in Homo sapiens. Additionally, the vol- m
h
1304 may affect aspects of human brain functioning and umeofcoronaldentinewaslargerintheextinctHomoand, ttp
s
cognition. Previous work showed that SNP variants in sinceenamelvolumeissimilarinbothspecies,thisresults ://a
the vicinity of miR-22, miR-148a, and miR-488 associate in significantly thinner cuspal enamel in Neanderthals c
a
d
with panic disorder and that these miRNAs effectively re- (Ramirez Rozzi and Bermudez De Castro 2004; Olejniczak e
m
press candidate genes for anxiety (Muin˜os-Gimeno et al. et al. 2008). Moreover, it is believed that enamel cusps ic
.o
2011). In this context, involvement of miR-1304 in brain formed faster in Neanderthals as evidenced by a signifi- u
p
functionregulationraisestheinterestingpossibilityofpar- cantlylowerperiodicityoftheirenamelgrowthmarks(Re- .c
o
m
ticularneurologicalphenotypesbeingassociatedwithone tzius marks or perikymata) (Aiello and Dean 1990), a fact /m
miR-1304 allele or the other.On theother hand,compar- thathasoftenleadtooverestimationofNeanderthal’sage b
e
insoomneosftidheenNtiefiaenddaertnhuamlgbeenroomfegeannodmmicodreegrinonhsumthaantsm’gaey- Tathidsewatohul(dOilmejnpilcyzaakdeiftfearle.n2t0r0a8t;eSomfitahmeetloabll.a2st0i0c9a,c2t0iv1i0ty),. /article
havebeenaffectedbypositiveselectioninancestralmod- a process that may be influenced by miR-1304. This hy- -ab
s
ern humans, including regions involved in cognitive abili- pothesis is in agreement with recent studies performed tra
c
ties like NRG3, a gene involved in schizophrenia (Green inrodentsthatdenotethatmiRNAshaveaprominentrole t/2
etal.2010).Takentogetherthesedatasuggestthathuman inteethdevelopmentbeingrequiredfornormalameloblast 9/7
cognition has been an important target of recent human differentiation and enamel matrix formation (Cao et al. /1
7
9
evolutionthatcouldhavebeenshapedinpartbymiRNAs. 2010; Michon et al. 2010). Since the ancestral allele of 7
/1
Amongthefewpredictedtargetgenesfortheancestral miR-1304 still segregates in some modern humans, this 0
6
9
miR-1304,CD24andthetranscriptionfactor4(TCF4)areof miR-1304variantcouldbeinvolvedindentaldevelopment 1
4
interest for their involvement in neurological disorders in andbeassociated,forexample,withthedegreeofenamel 8 b
humans. CD24 has been associated with multiple sclerosis thickness.Althoughlittleisknownaboutvariationsofden- y g
u
(Zhouetal.2003),andTCF4(predictedbyTargetRankand talmorphologyamonghumanpopulations,ithasbeenre- e
s
TargetScan) has been recently associated with a range of ported that average enamel thickness is similar among t o
n
neuropsychiatric phenotypes including schizophrenia, im- Asian, European, and African dentitions (Feeney et al. 1
3
pairedverballearning,andPitt–Hopkinssyndrome(Amiel 2010). Nevertheless, given the fact that only about 0.5% A
p
etal.2007;Stefanssonetal.2009;Lennertzetal.2011).Two ofAsianindividualswouldbehomozygousfortheancestral ril 2
othernoteworthytargetgenesareENAMandAMTN,both miR-1304, we do not expect this allele to be involved in 01
9
involvedinodontogenesis.Wefocusedonthesegenesdue phenotypictraitsthatdifferamongpopulationsbutrather
totheknowndentaldifferencesbetweenNeanderthalsand inlesscommontraitssomehowrelatedwithdisease.Inthis
modern humans and demonstrated the conservation of regard, defects in enamel formation create the condition
their 3# UTRs among primates as well as their differential knownasamelogenesisimperfecta(AI),adiseaseinwhich
response after transfection with either the derived or the the enamel does not fully form or forms in insufficient
ancestralmiR-1304.Thesetwogenesareinvolvedinteeth amounts and teeth affected may be discolored, sensitive,
formationandmapneareachotherinaninterestingclus- or prone to disintegration. AI is commonly inherited as
ter on chromosome 4 together with other genes also in- anautosomictrait,andENAMmutationsappeartobere-
volved in dental formation as well as salivary proteins sponsibleforabigpartoftheautosomallyinheritedcases.
such as ameloblastin (AMBN) or mucin 7 (MUC7). However,geneticstudiesprovideevidencefortheexistence
Tooth enamel, the hardest substance in vertebrates, is of at least one further autosomal AI locus (Ka¨rrman et al.
formedbyepithelium-derivedameloblasts.Asameloblasts 1997;Dongetal.2000).TheancestralmiR-1304,asrepres-
differentiate, they deposit specific proteins necessary for sor of ENAM and AMTN, is a strong candidate to be
1804
MBE
An Ancestral miR-1304 Allele Regulates Dental Genes · doi:10.1093/molbev/mss023
involved in the susceptibility to AI and other forms of AmielJ,RioM,dePontualL,etal.(11co-authors).2007.Mutations
enamel hypoplasia. It would be of interest to test this hy- in TCF4, encoding a class I basic helix-loop-helix transcription
pothesis in familial cases of the disorder for which the un- factor, are responsible for Pitt-Hopkins syndrome, a severe
epileptic encephalopathy associated with autonomic dysfunc-
derlying genetic defect has not been identified yet and
tion. Am JHum Genet.80:988–993.
furthermore study if Asian populations show higher AI in-
BartelDP.2004.MicroRNAs:genomics,biogenesis,mechanism,and
cidencethanotherpopulations(theexactincidenceofAIis
function. Cell 116:281–297.
uncertainandestimatesvaryfrom1:700peopleto1:14,000 Bentwich I, Avniel A, Karov Y, et al. (13 co-authors). 2005.
accordingtothepopulationsstudied;Seow1993).Thefind- Identification of hundreds of conserved and nonconserved
ingofcorrelationbetweenthepresenceoftheancestralal- human microRNAs. NatGenet.37:766–770.
lele of miR-1304 and the minor allele of rs117342040 in Berezikov E, Guryev V, van de Belt J, Wienholds E, Plasterk R,
Cuppen E. 2005. Phylogenetic shadowing and computational
ENAM, a variant that would interfere with the binding of
identification ofhumanmicroRNA genes.Cell 120:21–24.
the ancestral miR-1304 to the mRNA and would avoid its
Borel C, Antonarakis SE. 2008. Functional genetic variation of
repressionrescuingahypotheticalphenotype,isveryappeal- humanmiRNAsandphenotypicconsequences.MammGenome. D
o
ing.Itstrengthenstheideathatthepresenceoftheancestral 19:503–509. w
n
miR-1304 in modern humans could have some adverse ef- Brameier M. 2010. Genome-wide comparative analysis of micro- lo
a
fect, and thus, removal of downregulation of ENAM could RNAs inthree non-humanprimates. BMC ResNotes.3:64. de
d
hwaevceoubledennobtefinnedfiecviaildetnocemsofodrearnselheucmtivaenssw. eUenpfohratpupneanteinlyg, BrenmnieccrkoeRNJ,AS-ttaarrgketAr,eRcougssneitllioRnB.,PLCooShBeniol.SM3:e.8250.05. Principles of from
Briggs AW, Good JM, Green RE, et al. (18 co-authors). 2009. h
inthemiR-1304gene,whichwouldhavereinforcedthishy- Targeted retrieval and analysis of five Neandertal mtDNA ttps
pothesis.Finally,weshouldnotforgetaboutotherpredicted genomes. Science 325:318–321. ://a
targetgenesfortheancestralmiR-1304thatarealsorelated Briggs AW, Stenzel U, Johnson PL, et al. (11 co-authors). 2007. ca
d
to disease such as the above stated CD24 and TCF4 genes Patterns of damage in genomic DNA sequences from e
m
that are involved in neurological disorders in humans. a Neandertal.Proc NatlAcad SciU SA.104:14616–14621. ic
Futureanalysisonthedifferentialgenerepressionbythe Burbano HA, Hodges E, Green RE, et al. (20 co-authors). 2010. .ou
p
two alleles of miR-1304 and deeper studies on the func- TargetedinvestigationoftheNeandertalgenomebyarray-based .c
o
sequence capture.Science 328:723–725. m
tional significance of this regulation would be of great in- Cao H, Wang J, Li X, et al. (11 co-authors). 2010. MicroRNAs play /m
terestnotonlytogaininsightsintoprimateevolutionbut b
a criticalrole in toothdevelopment.J DentRes.89:779–784. e
atylspoicinditfhfeerepnocsessibalemiomnpglipcaretisoennto-dfatyhihsummiRanNApoinpuplahteionnos- CarraolgleSnBe.t2ic00t8h.eEovryo-odfevmooarnpdhoalnogeixcpalanevdoinlugtieovno.luCteiolln1a3r4y(s1y)n:2t5h–e3si6s.: /article
and its relationship with complex human disorders. Chen K, Rajewsky N. 2006. Deep conservation of microRNA-target -ab
relationships and 3#UTR motifs in vertebrates, flies, and stra
Supplementary Material DonngeJm,GatuodTeTs,.SCimolmdoSnprsinDg,MHaarcbDoSyumgapllQMu.a2n0t00B.ioEln.a7m1:1e4li9n–m15a6p.sto ct/29
Supplementary figures 1 and 2 and supplementary tables human chromosome 4q21 within the autosomal dominant /7/1
1–3areavailableatMolecularBiologyandEvolutiononline amelogenesis imperfecta locus. EurJ OralSci. 108:353–358. 79
(http://www.mbe.oxfordjournals.org/). Endicott P, Ho SY, Stringer C. 2010. Using genetic evidence to 7/1
evaluatefourpalaeoanthropologicalhypothesesforthetimingof 06
9
Neanderthalandmodernhumanorigins.JHumEvol.59:87–95. 1
Acknowledgments 4
Feeney RN, Zermeno JP, Reid DJ, Nakashima S, Sano H, Bahar A, 8
b
WewouldliketothankM.Valle´s,B.Kagerbauer,andF.Sa´n- Hublin JJ, Smith TM. 2010. Enamel thickness in Asian human y g
chez-Quinto for their technical support and advice and to canines andpremolars. AnthropolSci.118(3):191–198. ue
the anonymous referees whose comments and suggestions FilipowiczW,BhattacharyyaSN,SonenbergN.2008.Mechanismsof st o
post-transcriptional regulation by microRNAs: are the answers n
greatlyimprovedthismanuscript.Supportwasprovidedby 1
in sight?NatRev Genet.9:102–114. 3
theSpanishNationalInstituteforBioinformatics(www.inab. Finlayson C, Pacheco FG, Rodr´ıguez-Vidal J, et al. (26 co-authors). Ap
org).ThisworkwassupportedbytheMinisteriodeCienciae 2006. Late survival of Neanderthals at the southernmost ril 2
Innovacio´n, Espan˜a (BFU2010-18477, BFU2009-06974, and extreme ofEurope.Nature 443:850–853. 0
1
CGL2009-09013) and the European Union Seventh Frame- Friedman RC, Farh KK, Burge CB, Bartel DP. 2009. Most mammalian 9
workProgramme(PIOF-GA-2009-236836).Thispublication mRNAsareconservedtargetsofmicroRNAs.GenomeRes.19:92–105.
GradY,AachJ,HayesGD,ReinhartBJ,ChurchGM,RuvkunG,KimJ.
has been cofinanced by FEDER—European Regional Devel-
2003. Computational and experimental identification of C.
opment Fund ‘‘A way to build Europe.’’ Fieldwork at El Si-
elegans microRNAs. Mol Cell.11:1253–1263.
dro´n site has been funded by the Consejer´ıa de Cultura,
GreenRE,KrauseJ,BriggsAW,etal.(56co-authors).2010.Adraft
Principado de Asturias. M.L.V. is funded by a Beca per sequence oftheNeandertal genome. Science 328:710–722.
a la Formacio´ de PersonalInvestigador(FI) fellowshipfrom Guidi M, Muin˜os-Gimeno M, Kagerbauer B, Mart´ı E, Estivill X,
the Age`ncia de Gestio´ d’Ajuts Universitaris i de Recerca, Espinosa-ParrillaY.2010.OverexpressionofmiR-128specifically
Generalitat de Catalunya. inhibitsthetruncatedisoformofNTRK3andupregulatesBCL2
in SH-SY5Yneuroblastoma cells.BMC MolBiol.11:95.
Hindorff LA, Sethupathy P, Junkins HA, Ramos EM, Mehta JP,
References
CollinsFS,ManolioTA.2009.Potentialetiologicandfunctional
AielloLC,DeanMC.1990.Anintroductiontohumanevolutionary implicationsofgenome-wideassociationlociforhumandiseases
anatomy. London: Academic Press Limited.p.106–132. and traits.Proc NatlAcad SciU SA. 106:9362–9367.
1805
MBE
Lopez-Valenzuela et al. · doi:10.1093/molbev/mss023
HofreiterM,JaenickeV,SerreD,HaeselerAvA,Pa¨a¨boS.2001.DNA profilinganddiscoveryinhumanembryonicstemcells.Genome
sequencesfrommultipleamplificationsrevealartifacts induced Res.18:610–621.
by cytosine deamination in ancient DNA. Nucleic Acids Res. Muin˜os-Gimeno M, Espinosa-Parrilla Y, Guidi M, et al. (14 co-
29:4793–4799. authors). 2011. Human microRNAs miR-22, miR-138-2, miR-
HublinJJ.2009.TheoriginofNeandertals.ProcNatlAcadSciUSA. 148a, and miR-488 are associated with panic disorder and
106:16022–16027. regulate several anxiety candidate genes and related pathways.
Iwasaki K, Bajenova E, Somogyi-Ganss E, Miller M, Nguyen V, BiolPsychiatry. 69:526–533.
Nourkeyhani H, Gao Y, Wendel M, Ganss B. 2005. Amelotin—a Muin˜os-Gimeno M, Guidi M, Kagerbauer B, Mart´ın-Santos R,
novelsecreted,ameloblast-specificprotein.JDentRes.84:1127–1132. Navine´s R, Alonso P, Mencho´n JM, Grataco`s M, Estivill X,
JazdzewskiK,MurrayEL,FranssilaK,JarzabB,SchoenbergDR,dela Espinosa-ParrillaY.2009.AllelevariantsinfunctionalMicroRNA
Chapelle A. 2008. Common SNP in pre-miR-146a decreases target sites of the neurotrophin-3 receptor gene (NTRK3) as
mature miR expression and predisposes to papillary thyroid susceptibility factors for anxiety disorders. Hum Mutat.
carcinoma. ProcNatlAcad Sci USA. 105(20):7269–7274. 30:1062–1071.
Ka¨rrman C, Ba¨ckman B, Dixon M, Holmgren G, Forsman K. 1997. Nielsen CB, Shomron N, Sandberg R, Hornstein E, Kitzman J,
Mapping of the locus for autosomal dominant amelogenesis Burge CB. 2007. Determinants of targeting by endogenous and D
o
imperfecta(AIH2)toa4-MbYACcontigonchromosome4q11- exogenous microRNAs andsiRNAs. RNA 13:1894–1910. w
n
q21.Genomics 39:164–170. Olejniczak AJ, Smith TM, Feeney RN, et al. (14 co-authors). 2008. lo
a
KrauseJ,Lalueza-FoxC,OrlandoL,etal.(13co-authors).2007.The Dental tissue proportions and enamel thickness in Neandertal de
d
dNeeraivneddertFaOlsX.PC2urrvaBriioaln.t17o:1f9m08o–d1e9r1n2.humans was shared with PickarneldlJmK,oCdoeronphGu,mNaonvemmoblarers.J,JeHtuaml.(8Evcool.-a5u5t:1h2o–rs2)3..2009.Signals from
Krol J, Loedige I, Filipowicz W. 2010. The widespread regulation of of recent positive selection in a worldwide sample of human h
microRNA biogenesis, function and decay. Nat Rev Genet. populations. Genome Res.19(5):826–837. ttp
s
11:597–610. Quach H, Barreiro LB, Laval G, et al. (11 co-authors). 2009. ://a
Lalueza-FoxC,GigliE,delaRasillaM,ForteaJ,RosasA.2009.Bitter Signatures of purifying and local positive selection in human ca
d
taste perception in Neanderthals through the analysis of the miRNAs. Am JHum Genet.84:316–327. e
m
TAS2R38gene.Biol.Lett. 5:809–811. RamirezRozziFV,BermudezDeCastroJM.2004.Surprisinglyrapid ic
Lalueza-Fox C, Gigli E, de la Rasilla M, Fortea J, Rosas A, growth inNeanderthals. Nature 428:936–939. .ou
Bertranpetit J, Krause J. 2008. Genetic characterization of the Rosas A, Mart´ınez-Maza C, Bastir M, et al. (18 co-authors). 2006. p.c
ABObloodgroup in Neandertals. BMC EvolBiol.8:342. PaleobiologyandcomparativemorphologyofalateNeandertal om
Lalueza-FoxC,Ro¨mplerH,CaramelliD,etal.(17co-authors).2007. samplefromElSidron,Asturias,Spain.ProcNatlAcadSciUSA. /m
b
Amelanocortin1receptorallelesuggestsvaryingpigmentation 103:19266–19271. e
/a
Laluaemzao-FnogxNCe,anRdoesratshaAls,.ESsctiaelnrrciech31A8:,1e4t53a–l.14(1565.co-authors). 2011. SeowamWeloKg.e1n9e9s3is. iCmlipneicrafelcdtaiavganroiasinstsa.nPdedmiaatnraDgeenmt.e1n5t:3s8tr4a–t3eg9i3e.s of rticle
Genetic evidence for patrilocal mating behavior among Skoglund P, Jakobsson M. 2011. Archaic human ancestry in East -a
b
Neandertalgroups. ProcNatl AcadSci US A.108:250–253. Asia.ProcNatl AcadSci US A.108:18301–18306. stra
LennertzL,RujescuD,WagnerM,etal.(14co-authors).2011.Novel Smith TM, Harvati K, Olejniczak AJ, Reid DJ, Hublin JJ, c
schizophrenia risk gene TCF4 influences verbal learning and Panagopoulou E. 2009. Brief communication: dental develop- t/2
9
memoryfunctioninginschizophreniapatients.Neuropsychobiology ment and enamel thickness in the Lakonis Neanderthal molar. /7
/1
63:131–136. AmJ PhysAnthropol.138:112–118. 7
9
Lewis BP, Burge CB, Bartel DP. 2005. Conserved seed pairing, often Smith TM, Tafforeau P, Reid DJ, et al. (14 co-authors). 2010. 7
/1
flankedbyadenosines,indicatesthatthousandsofhumangenes Dental evidence for ontogenetic differences between modern 0
6
are microRNAtargets. Cell 120:15–20. humans and Neanderthals. Proc Natl Acad Sci U S A. 91
4
Lin S, Cheung WK, Chen S, Lu G, Wang Z, Xie D, Li K, Lin MC, 107:20923–20928. 8
Kung HF. 2010. Computational identification and characteriza- Stark MS, Tyagi S, Nancarrow DJ, Boyle GM, Cook AL, by
tionofprimate-specificmicroRNAsinhumangenome.Comput Whiteman DC, Parsons PG, Schmidt C, Sturm RA, gu
e
BiolChem. 34:232–241. HaywardNK.2010.CharacterizationofthemelanomamiRNAome s
LiuN,OkamuraK,TylerDM,PhillipsMD,ChungWJ,LaiEC.2008. bydeep sequencing. PLoSOne. 5:e9685. t on
The evolution and functional diversification of animal micro- Stefansson H, Ophoff RA, Steinberg S, et al. (87 co-authors). 2009. 1
3
RNAgenes.Cell Res.18:985–996. Common variants conferring risk of schizophrenia. Nature A
p
MacchiarelliR,BondioliL,Debe´nathA,MazurierA,TournepicheJF, 460:744–747. ril 2
Birch W, Dean MC. 2006. How Neanderthal molar teeth grew. Stringer CB, Hublin JJ. 1999. New age estimates for the Swanscombe 0
1
Nature444:748–751. hominid,andtheirsignificanceforhuman.JHumEvol.37:873–877. 9
Mart´ıE,PantanoL, Ban˜ez-CoronelM, LlorensF,Min˜ones-Moyano E, SunG,YanJ,NoltnerK,FengJ,LiH,SarkisDA,SommerSS,RossiJJ.
Porta S, Sumoy L, Ferrer I, Estivill X. 2010. A myriad of miRNA 2009. SNPs in human miRNA genes affect biogenesis and
variantsincontrolandHuntington’sdiseasebrainregionsdetected function.RNA 15:1640–1651.
bymassivelyparallelsequencing.NucleicAcidsRes.38:7219–7235. Vaz C, Ahmad HM, Sharma P, Gupta R, Kumar L, Kulshreshtha R,
McLean CY, Reno PL, Pollen AA, et al. (13 co-authors). 2011. Bhattacharya A. 2010. Analysis of microRNA transcriptome by
Human-specific loss of regulatory DNA and the evolution of deepsequencingofsmallRNAlibrariesofperipheralblood.BMC
human-specifictraits. Nature 471:216–219. Genomics 11:288.
MellarsP.2004.Neanderthalsandthemodernhumancolonization VinciS,GelminiS,PratesiN,ContiS,MalentacchiF,SimiL,PazzagliM,
ofEurope.Nature 432:461–465. Orlando C. 2011. Genetic variants in miR-146a, miR-149, miR-
MichonF,TummersM,Kyyro¨nenM,FrilanderMJ,ThesleffI.2010. 196a2,miR-499andtheirinfluenceonrelativeexpressioninlung
Tooth morphogenesis and ameloblast differentiation are cancers.ClinChemLabMed.doi:10.1515/CCLM.
regulatedby micro-RNAs.Dev Biol.340:355–368. Zhou Q, Rammohan K, Lin S, et al. (14 co-authors). 2003. CD24 is
Morin R, O’Connor M, Griffith M, et al. (12 co-authors). 2008. ageneticmodifierforriskandprogressionofmultiplesclerosis.
Application of massively parallel sequencing to microRNA ProcNatl AcadSci US A.100:15041–15046.
1806