Genetic diversity of carotenoid-rich bananas evaluated by Diversity Arrays Technology (DArT).

1.9 MB·English

by  Amorim, Edson P.

GeneticsandMolecularBiology,32,1,96-103(2009)
Copyright©2009,SociedadeBrasileiradeGenética.PrintedinBrazil
www.sbg.org.br
ResearchArticle
Genetic diversity of carotenoid-rich bananas evaluated
by Diversity Arrays Technology (DArT)
EdsonP.Amorim1,AlbertoD.Vilarinhos1,KellyO.Cohen2,VanusiaB.O.Amorim1,
JanayA.dosSantos-Serejo1,SebastiãoOliveiraeSilva1,KátiaN.Pestana1,VâniaJ.dosSantos1,
NormaS.Paes2,DamaresC.Monte2andRonaldoV.dosReis1
1EmbrapaMandiocaeFruticulturaTropical,CruzdasAlmas,BA,Brazil.
2EmbrapaRecursosGenéticoseBiotecnologia,Brasília,DF,Brazil.
Abstract
Theaimofthisworkwastoevaluatethecarotenoidcontentandgeneticvariabilityofbananaaccessionsfromthe
MusagermplasmcollectionheldatEmbrapaCassavaandTropicalFruits,Brazil.Forty-twosampleswereanalyzed,
including21diploids,19triploidsandtwotetraploids.Thecarotenoidcontentwasanalyzedspectrophotometrically
andgeneticvariabilitywasestimatedusing653DArTmarkers.Theaveragecarotenoidcontentwas4.73(cid:2)g.g-1,and
rangedfrom1.06(cid:2)g.g-1forthetriploidNanica(Cavendishgroup)to19.24(cid:2)g.g-1forthetriploidSaney.Thediploids
ModokGierandNBA-14andthetriploidSaneyhadacarotenoidcontentthatwas,respectively,7-fold,6-foldand
9-foldgreaterthanthatofcultivarsfromtheCavendishgroup(2.19(cid:2)g.g-1).Themeansimilarityamongthe42acces-
sionswas0.63(range:0.24to1.00).DArTanalysisrevealedextensivegeneticvariabilityinaccessionsfromthe
EmbrapaMusagermplasm bank.
Keywords:diploids,germplasm,molecularmarkers,Musasp.,variability.
Received:March25,2008;Accepted:July21,2008.
Introduction sava and Tropical Fruits has an active banana germplasm
bank created through the introduction of local germplasm
Fruit consumption has increased markedly in recent
and from international collections. This germplasm bank
years,mainlybecauseofitsnutritionalvalueandtherapeu-
consists of more than 400 accessions maintained in the
ticeffects.Fruitscontaindistinctphytochemicals,manyof
field,includingwilddiploids,triploidsandtetraploids.
whichhaveantioxidantpropertiesthatdelayagingandpre-
vent diseases, including certain types of cancer. Com- Determinationofthecarotenoidcontentandgenetic
pounds such as (cid:3)-carotenes and vitamins C and E are variabilityofthesegenotypesusingDNAmolecularmark-
erscanprovideinformationthatisusefulinparentalselec-
important antioxidant components of fruits (Wang et al.,
tionforcrossesbetweendivergentgenotypesandfordevel-
1997).
oping novel cultivars with functional properties. Several
Bananasarewidelyconsumed,primarilybecauseof
molecular markers, especially those associated with poly-
their low cost and their potential as functional and nutra-
merasechainreaction(PCR)-basedmethods,includingre-
ceuticalfood.Brazilisthesecondlargestbananaproducer
striction fragment length polymorphism (AFLP), random
intheworld,withaproductionof7.1milliontonsin2006
amplified polymorphic DNA (RAPD) and microsatellites
from ~500,000 hectares of plantations (FAO, 2008). Cur-
orsimplesequencerepeats(SSR),havebeenwidelyusedto
rently,commercialcultivars,especiallythoseoftheCaven-
estimate genetic variability, choose genitors and investi-
dish group, do not contain significant amounts of sub-
gate phylogenetic relationships among bananas (Creste et
stances with nutritional or therapeutic potential, such as
al.,2003,2004;Wanetal.,2005;Ningetal.,2007;Wang
polyphenols,vitaminCandcarotenoids.Incontrast,some
etal.,2007;Jainetal.,2007;Ruangsuttaphaetal.,2007).
bananagenotypesidentifiedingermplasmbankshavehigh
contents of these substances (Setiawan et al., 2001; So- Diversityarraystechnology(DArT)isanovelgeno-
meyaetal.,2002;Englbergeretal.2003a,b,c,2005;Melo typing technique that was originally developed for rice
etal.,2006;Wall,2006;Daveyetal.,2007).EmbrapaCas- (Jaccoud et al., 2001) but has been applied to cassava,
wheat and Arabidopsis thaliana (Wittenberg et al., 2005;
Xia et al., 2005; Stodart et al., 2007; White et al., 2008).
SendcorrespondencetoEdsonP.Amorim.EmbrapaCassavaand
Tropical Fruits, 44380-000 Cruz das Almas, BA, Brazil. E-mail: DArTisaninexpensivetechniquethatisbasedonhybrid-
[email protected]. izationandamplification,requiresonlyasmallamountof
Amorimetal. 97
DNAandprovidedamplegenomecoverage(Stodartetal., Carotenoid analysis
2007).
The carotenoid content was determined spectropho-
The aims of this work were to quantify the total
tometrically, as described by Rodriguez-Amaya (1999).
carotenoid content of 42 accessions of the banana
Bananasamples(10g)weremaceratedinamortarcontain-
germplasm bank held at Embrapa Cassava and Tropical
ingceliteandcoldacetoneandfilteredthroughasintered
Fruits,includingdiploids(AA),triploids(AAA,AABand
glass funnel. The extract was transferred to a separation
ABB)andtetraploids(AAAB),andtoestimatethegenetic
funnel containing petroleum ether and ethylic ether (1:1,
variability by using DArT molecular markers to aid the
v/v)andtheacetone:petroleumether:ethylicethermixture
choiceofgenitorsforcrossesdoneinthebananabreeding
waswashedrepeatedlywithdistilledwatertheacetonewas
programatEmbrapa.Toourknowledge,thisisthefirstre-
completely removed. The extract was dried with sodium
portontheuseofDArTtoevaluatebananadiversity.
sulfate, transferred to a suitable volumetric flask, and the
volumewasadjustedwithpetroleumether.Thecarotenoid
Material and Methods
contentwasdeterminedspectrophotometricallyat450nm
ina1-cmcuvette.Thesampleswereassayedinduplicate.
Plant material
Forty-two banana accessions were used, including
Diversity analysis
diploids(AA),triploids(AAA,AABandABB)andtetra-
ploids(AAAB)maintainedintheactivebananagermplasm
DNA extraction
bank at Embrapa Cassava and Tropical Fruits (Cruz das
Almas,BA,Brazil).Theaccessionsinvestigatedwereorig- GenomicDNAwasextractedfromleavesofthedif-
inallyincludedinthegermplasmbankbasedontheirlevel ferentgenotypesbythecetyltrimethylammoniumbromide
offunctionalcompoundssuchastotalcarotenoidcontent. (CTAB)methodproposedbyDoyleandDoyle(1990).Af-
Thepresenceofnutraceuticalcompoundswasrecentlyin- terDNAextraction,thesampleswerestoredinmicrotiter
troducedasanadditionalcriterionforbananaselectionin platesandshippedtoDiversityArraysTechnology(DArT
breedingprogramsatEmbrapa.Table1showsthelevelof P/L, PO Box 7141, Yarralumba, ACT 2600, Australia),
ploidy,origin,colorpulp,andcarotenoidcontentoftheac- wheretheDArTanalysesweredone.Thepreparationofthe
cessionsusedinthiswork. PstI/TaqIgenomicrepresentationDNAsamplesforDArT
Table1-Carotenoidcontentof42accessionsfromthebanana(Musa)germplasmcollectionatEmbrapaCassavaandTropicalFruits.
Accession/ITC Ploidy Origin Pulpcolor CC Accession Ploidy Origin Pulpcolor CC
ModokGier AA NewGuinea Orange 16.05(cid:4)0.41* Ambei AAA NewGuinea Orangish-yellow 4.22(cid:4)0.29
NBA14/0267 AA NewGuinea Orange 13.04(cid:4)0.05 Bakar/1064 AAA Indonesia Yellow 3.98(cid:4)0.62
Khai/0532 AA Thailand Orangish-yellow 9.02(cid:4)0.22 FigueRoseNaine/1159 AAA France Yellow 3.50(cid:4)0.32
Jaran/0678 AA Indonesia Orangish-yellow 8.23(cid:4)0.79 Highgate/0263 AAA Honduras Yellow 3.34(cid:4)0.51
SN2 AA NewGuinea Yellow 7.23(cid:4)0.71 Lacatan/0768 AAA France Yellow 3.18(cid:4)0.16
JariBuaya/0312 AA Honduras Yellow 7.01(cid:4)0.78 Wasolay AAA NewGuinea Yellow 3.15(cid:4)0.17
Malbut AA NewGuinea Yellow 6.87(cid:4)0.24 Canela AAA Brazil Yellow 2.86(cid:4)0.44
Ouro AA Brazil Yellow 5.91(cid:4)0.38 GrosMichel/0484 AAA Brazil Cream 2.81(cid:4)0.67
PAPhatthalung AA Thailand Yellow 3.59(cid:4)0.34 Nam/1303 AAA Thailand Yellow 2.77(cid:4)0.14
2803-01 AA Brazil Yellow 3.53(cid:4)0.14 Pagatow AAA NewGuinea Yellow 2.64(cid:4)0.25
M-48 AA Equador Yellow 3.51(cid:4)0.19 Valery/0048 AAA Indonesia Yellow 2.32(cid:4)0.24
KhaiNaiOn/0532 AA Thailand Yellow 3.42(cid:4)0.23 Markatooa AAA NewGuinea White 2.29(cid:4)0.25
M-61 AA Equador Yellow 3.37(cid:4)0.52 Nanica AAA Brazil White 1.06(cid:4)0.04
KhiMaeo AA Thailand Yellow 3.32(cid:4)0.15 P.Nangka/0004 AAB Thailand Yellow 4.59(cid:4)0.01
Tongat/0063 AA Honduras Yellow 3.01(cid:4)0.31 Saney AAB NewGuinea Orange 19.24(cid:4)3.22
Pipit/0685 AA Indonesia Beige 2.95(cid:4)0.14 AABnoname AAB Honduras Orange 8.40(cid:4)0.34
F3P4 AA Equador Yellow 2.52(cid:4)0.03 Terra AAB Brazil Orangish-yellow 5.83(cid:4)1.16
Lidi/0395 AA Honduras Beige 2.45(cid:4)0.08 ThapMaeo/1301 AAB Honduras Cream 3.77(cid:4)0.04
Sowmuk/0266 AA NewGuinea Yellow 2.23(cid:4)0.07 FigoCinza ABB Brazil Yellow 2.15(cid:4)0.01
Jambi AA Indonesia Beige 1.47(cid:4)0.01 OurodaMata/0261 AAAB Brazil Orangish-yellow 4.70(cid:4)0.02
Tuugia AA Hawaii White 1.41(cid:4)0.04 Porp AAAB NewGuinea Beige 2.34(cid:4)0.06
ITC:InternationalTransitCentre.CC:carotenoidcontent((cid:2)g.g-1).*:means((cid:4)standarderror)oftworeplicatesperaccession.
98 Geneticdiversityofcarotenoid-richbananas
andthemethodologyforDArT,asprovidedbyDArTP/L, softwareautomaticallyrecognizesthearrayfeaturesusing
aredescribedbelow. aseeded-region-growthalgorithmand,foreachfluorescent
channel,reportstheaverageandstandarddeviation(SD)of
DArT procedure
pixel intensities within and around each array feature, the
Restrictionenzymedigestionandadapterligation fractionofsaturatedpixels,andthenumberofpixelswithin
Acombineddigestion/ligationreactionwasprepared eachfeature,amongotherparameters.Thelogarithmofthe
byadding1(cid:2)LofDNAsampleto9(cid:2)Lofdigestion/liga- ratio between the background-subtracted averages of fea-
tionmix.Thedigestion/ligationreactionmixturecontained ture pixels in the Cy3 and Cy5 channel
1 (cid:2)L of 10X RE buffer (100 mM Tris-acetate (OAc), (=log[Cy3target/Cy5reference])wasusedasameasureof
thedifferenceintheabundanceofthecorrespondingDNA
500mMKOAc,100mMMg(OAc) ,50mMDTT,pH7.8),
2
0.1(cid:2)LofBSA(NewEnglandBiolabs),0.2(cid:2)Lof50mM fragmentfortherepresentationshybridizedtoanarray.The
valuesoflog[Cy3/FAM]andlog[Cy5/FAM],approximate
ATP,0.1(cid:2)Lof5(cid:2)MPstIadapter,0.1(cid:2)LofPstI(20U/(cid:2)L
measuresoftheamountofhybridizationsignalperamount
NEB), 0.1 (cid:2)L TaqI (20 U/(cid:2)L NEB), 0.2 (cid:2)L of T4 DNA
ofDNAspottedonthearray,wereusedforqualitycontrol
ligase (30 Weiss units/(cid:2)L NEB) and 7.2 (cid:2)L of ultrapure
purposes.
water.Thesampleswereincubatedat37°Cfor90minand
Six hundred and fifty-three high-quality markers
at60°Cfor90min,heatinactivatedat80°Cfor20minand
were scored. The results of the DArT analysis were ex-
storedat-20°Cuntilused.
pressed as a matrix (0-1). The mean genetic similarity
PCR amplification of the genomic representation (target among the genotypes was estimated by using a simple
preparation) matchingcoefficient(SokalandMichener,1958).Cluster
analysis was done by using the unweighted pair-group
PCR was used to create a genomic representation
methodwitharithmeticaverage(UPGMA)asimplemented
containing PstI fragments that (i) showed no restriction
in NTSYS-pc v. 2.1 (Rohlf, 2000). Principal component
sites for the frequent cutter used for redigestion and
analysis (PCA)-based clustering with the sub-routine
(ii)wereshortenoughtobeefficientlyamplified.Thenum-
EIGENwasusedtoassessandvisualizethelevelofploidy
berofPCRcycleswaskepttoaminimumtolimitthebias
(diploids, triploids and tetraploids) in diversity space (as
towards fragments that were amplified more efficiently
opposed to visualizing only the genetic similarity among
than others. The starting parameters for optimization are
theaccessions).
describedbelowforthePstI/TaqImethod.
Forty-nine microliters of PCR mix were added to Thecopheneticcorrelationcoefficientwascalculated
1 (cid:2)L of PstI/TaqI digestion/ligation reaction used as the and the Mantel test (Mantel, 1967) was used to check the
fitness of a cluster analysis to the matrix on which it was
PCRtemplate.ThePCRmixconsistedof(volumesperre-
action):5(cid:2)Lof10xPCRbuffer(100mMTris-HClpH8.3, based.Bootstrapanalysiswasusedtoverifythenumberof
DArTmarkersneededtopreciselydeterminetheestimates
500mMKCl,15mMMgCl ,0.1%gelatin;Sigma),1(cid:2)L
2 ofgeneticsimilarityamongthe42bananaaccessions.This
of10mMdNTPs(Sigma),2(cid:2)Lof10(cid:2)MPstI+0,2(cid:2)Lof
analysis was done with the statistical software GQMol
REDTaq (1 U/(cid:2)L; Sigma) and 39 (cid:2)L of ultrapure water.
(CruzandSchuster,2004)whichestimatesthecorrelation
ThePCRamplificationprogramconsistedofaninitialstep
betweenvaluesfromtheoriginaldistancematrixwiththose
at94°Cfor1min,30cyclesat94°Cfor20s,58°Cfor40s,
of other matrices obtained by re-sampling from different
and 72 °C for 1 min, and a final step at 72 °C for 7 min.
samplesizes(Amorimetal.,2006).Thesoftwarealsocal-
Five microliters of PCR product were analyzed on 1.2%
culatesavalueofstress(S)thatindicatesanadjustmentbe-
agarosegelstoconfirmthatahomogeneoussmearoffrag-
tween the original matrix and considers the 653 markers
mentswasobtainedandtovisualisethesizedistribution.
andre-samplingmatrices(Kruskal,1964).
For hybridization and scanning, the targets were la-
belledwithfluorescentdye(Cy3)andhybridizedwiththe
Results
reference (Topo) labeled with FAM (Carboxyfluorescein)
andhybridizedtothearraycontaining6,000DArTclones. Genotypeswithapulpcolorrangingfromyellowto
Clone information cannot be made available because of orangehadahighercontentoftotalcarotenoidscompared
dataprotectionrecommendations.Afterovernighthybrid- tothosewithwhiteorbeigepulp(Table1andFigures1and
ization at 62 °C, the arrays were washed and scanned at 2).Pulpcolorwasfoundtobeasuitablephenotypiccrite-
20(cid:2)mresolutionat543nm(Cy3)and488nm(FAM)onna rion for making inferences about variability in the total
LS300confocallaserscanner(Tecan,Grödig,Austria). carotenoid content of segregating populations. The mean
content of total carotenoids among the 42 accessions was
DArT data analysis 4.73 (cid:2)g.g-1, and ranged from 1.06 (cid:2)g.g-1 for the triploid
ArrayimageswereanalyzedusingDArTsoft7.4(Di- Nanica (Cavendish group) to 19.24 (cid:2)g.g-1 for the triploid
versityArraysTechnologyP/L,Canberra,Australia).The Saney(Table1).
Amorimetal. 99
Figure1-Bananaaccessionswithahighcarotenoidcontentandpulp
color ranging from yellow to orange. Diploids: Modok Gier (a) and
NBA-14(b),triploids:Saney(c)andAABnoname(d),tetraploids:Ouro
daMata(e)andPorp(f).
Figure2-Bananaaccessionswithalowcarotenoidcontentandbeigeor
whitepulpcolor.(a)diploidJambiand(b)triploidMarkatooa.Detail:
presenceofseeds.
Within the diploid group, the mean carotenoid con-
tent was 5.24 (cid:2)g.g-1, with high levels for the diploids
Modok Gier (16.05 (cid:2)g.g-1), NBA-14 (13.04 (cid:2)g.g-1) and
Khai(9.02(cid:2)g.g-1).Themeancontentamongtriploidsfrom
thegenomicgroupsAAA,AABandABBwas2.93,8.37
and2.15(cid:2)g.g-1,respectively.Themeancarotenoidcontent
fortetraploidswas3.52(cid:2)g.g-1(Table1).Triploidswiththe
Figure3-Carotenoidcontentsof42banana(Musa)accessionsinthe
genomiccompositionAABhadhighercontentsoftotalca- germplasmcollectionheldatEmbrapaCassavaandTropicalFruits.
rotenoids compared to AAA triploids. In the Cavendish
group of cultivars, the total carotenoid content was 1.06, andSaney)hadthesamegeographicorigin(NewGuinea),
3.18and2.32(cid:2)g.g-1fortheNanica,LacatanandValeryge- indicatingthepossibilityofgeneflowamongthem.Ourre-
notypes, respectively, with a mean content of 2.19 (cid:2)g.g-1. sultsindicatethatitispossibletoobtainsegregatingdiploid
The carotenoid content of the diploids Modok Gier and populationsforthetrait‘carotenoidcontent’thatcouldbe
NBA-14 and the triploid Saney was, respectively, 7-fold, used to build genetic maps and identify quantitative trait
6-fold and 9-fold greater than the mean value for the loci(QTL)associatedwiththetrait.Inaddition,byselect-
Cavendishgroup(2.19(cid:2)g.g-1). ingsuitablediploidandtriploidgenitorsitispossibletode-
veloptetraploidcultivarswithhighcarotenoidcontents.
Figure3showsthattherewasextensivegeneticvari-
abilityinthecarotenoidcontentsofthebananaaccessions, Re-samplinganalysesshowedthat282DArTmark-
particularlyamongdiploids.Thethreegenotypeswiththe ers(43%ofthetotal)weresufficienttoobtainaprecisees-
highest total carotenoid content (Modok Gier, NBA-14, timate of the genetic divergence among the 42 banana
100 Geneticdiversityofcarotenoid-richbananas
accessions.Thecorrelationbetweenthematrixconsidering similarity among the diploid accessions was 0.67, and
all653markersandthematrixwith282markerswas0.96, rangedfrom0.24betweenNBA-14and2803-01to0.98be-
with a square sum of the deviations (SQ ) of 0.51 and a tweenModokGierand2803-01.Theaccessionswereclus-
d
stressvalue(E)of0.052(Figure4).Thesefindingsindicate tered into three groups (Figure 5), although there was no
thatthenumberofDArTmarkersanalyzedwassufficient clear separation between accessions with high and low
topreciselyestimatethegeneticdivergenceamongthe42 carotenoid contents. The cophenetic value was high
bananaaccessions.Moreover,~60%oftheDArTmarkers (r = 0.93, p < 0.0001, 10,000 permutations). As with the
wereinferredtohavesimilarmolecularinformation,proba- carotenoidcontent(Figure3),DArTanalysisalsodetected
blyasaresultofgeneticlinkageamongtheminthebanana significantgeneticvariabilityamongthe42bananaacces-
genome.Themeangeneticsimilarityamongthe42acces- sionsmaintainedintheEmbrapagermplasmbank(Figures
sionswas0.63,andrangedfrom0.24betweenthediploid 5and6).
NBA-14andthetriploidCanelato1.00betweenthediploid Principal component analysis (PCA) reinforced the
2803-01andtheaccessionsCanela(AAB)andModokGier resultsobtainedwithclusteringanalysisbyUPGMA(Fig-
(AA).
Figure5showsthedendrogramofgeneticsimilarity
obtainedbyUPGMAbasedonDArTamongthe42banana
accessions investigated. The cophenetic value was high
(r = 0.92, p < 0.0001, 10,000 permutations) and adequate
since values of r (cid:5) 0.56 are considered ideal and reflect
agreementwiththevaluesofgeneticsimilarity(VazPatto
etal.,2004).
Thecut-offvalueforthedendrogramwasassumedto
bethemeangeneticsimilarityamongallaccessionsgeno-
typedbyDArT(0.63).Basedonthisvalue,twomajorclus-
ters were formed (Figure 5). The carotenoid contents
showednotendencytoclustersinceaccessionswithhigher
carotenoidlevelsoccurredequallyinbothclustersand,in
some cases, with low genetic similarity, e.g., the diploids
2803-01andNBA-14(0.24).Similarly,DArTmarkersdid
not cluster the accessions based on their genomic group,
i.e., diploids (AA), triploids (AAA, AAB and ABB) and
tetraploids(AAAB).Anumberoftheaccessionsclustered
basedongeographicorigin,whereasothersshowednosuch
correlation(Figure5andTable1).
Figure6showsadendrogramofthe21diploidgeno-
typesinvestigatedbyDArTmarkers.Theaveragegenetic
Figure5-UPGMAdendrogramfor42banana(Musa)accessionsinthe
Figure4-Bootstrapanalysisusedtoobtainapreciseestimateofthege- germplasmcollectionheldatEmbrapaCassavaandTropicalFruits.The
neticsimilarityamong42banana(Musa)accessionsinthegermplasmcol- dendrogram was constructed with NTSYS version 2.0 software (Rohlf
lectionheldatEmbrapaCassavaandTropicalFruits. 2000)usingthesimplematchingcoefficient.
Amorimetal. 101
orange. Some genotypes with white and beige pulp were
alsoanalyzedtoconfirmtheassociationbetweenpulpcolor
and carotenoid content. Banana pulp color is directly re-
lated to carotenoid content and can be classified into five
categories: white, beige, yellow, orangish-yellow and or-
ange (Englberger et al., 2003a,c). Hence, an orangish-
yellowororangepulpmaybeusedasacriteriontoselect
suitablegenotypestodevelopcultivarswithahighercaro-
tenoidcontent.
The mean carotenoid contents obtained here agreed
withtheresultsreportedbyothers.Setiawanetal.(2001)
reportedthatthemean(cid:3)-carotenecontentofthreesamples
of Musa paradisiaca was 1.00 (cid:2)g.g-1 (range: 0.72 to
1.22(cid:2)g.g-1).Englbergeretal.(2003b)quantifiedthetotal
carotenoidcontentin21bananaaccessions,includingge-
notypesfromthesectionsAustralimusaandEumusa.The
meantotalcarotenoidcontentwas11.13(cid:2)g.g-1,withvalues
rangingfrom0.60to53.7(cid:2)g.g-1.Theseauthorsreporteda
similarvalueforanother17bananaaccessions(9.21(cid:2)g.g-1)
(Englberger et al., 2003c) and Melo et al. (2006) found a
value of 10.62 (cid:2)g.g-1 for the cultivar Comprida (Type
Terra,AAB).Daveyetal.(2007)reportedthatgenotypes
fromthegenomicgroupAABhadhighercontentsofcarot-
Figure6-UPGMAdendrogramforthe21diploidbanana(Musa)acces-
enoids compared to those from the AAA group, possible
sionsinthegermplasmcollectionheldatEmbrapaCassavaandTropical
because of the presence of the B genome. Similar results
Fruits.ThedendrogramwasconstructedasdescribedinFigure5.
werereportedbyEnglbergeretal.(2003a)andMeloetal.
(2006).
As shown here, the carotenoid content of some ba-
nana genotypes differed from that of Cavendish cultivars
byupto9-fold.Inagreementwiththis,otherstudiesofba-
nanaaccessionsingermplasmbankshavereportedcaroten-
oidcontentsupto25-foldhigherthanfoundincommercial
cultivars from the Cavendish group (Englberger et al.,
2003a,b,c;Wall,2006).
Thecarotenoidcontentmaybeinfluencedbyseveral
factors,includingthedegreeofmaturation,typeofsoil,cli-
matic conditions, storage conditions, geographic location
and,especially,genotype(SharrockandLutsy,2000;Setia-
wanetal.,2001;Souzaetal.,2004).Inthepresentwork,
themostimportantfactorwasthegenotypeandtheorigin
(geographic location) since all of the other factors were
Figure7-ClusteringobtainedbyPCAbasedonpooledDArTdataof42 constant for the accessions in the Embrapa germplasm
banana(Musa)accessionsinthegermplasmcollectionheldatEmbrapa bank,i.e.,theaccessionswereanalyzedatthesamestageof
CassavaandTropicalFruits.1-diploids,2-triploidsand3-tetraploids. maturation,storedunderthesameconditionsandkeptun-
ure7).Thefirstandthesecondcomponentswereresponsi- derthesameenvironmentalconditions.
ble for 74.7% and 10.8% of the genetic variation, respec- Re-sampling analyses indicated that the number of
tively.Thesetwocomponentswereresponsiblefor85.5% DArT markers used was more than sufficient to precisely
ofthevariationobservedamongthelevelsofploidy(dip- genotypethe42bananaaccessions.Thestressvalue(E)of
loids,triploidsandtetraploids). 0.052reinforcedthisconclusion,sinceEvalues(cid:6)0.05are
consideredidealforpreciseestimationofthegeneticvaria-
tioninagivenpopulation(Kruskal,1964).
Discussion
Extensivegeneticvariabilitywasobservedamongthe
In this work, 42 banana accessions were initially 42accessionsgenotypedusingDArT.Similarresultswere
screenedbasedonpulpcolor,whichrangedfromyellowto obtained by Creste et al. (2003, 2004) who used micro-
102 Geneticdiversityofcarotenoid-richbananas
satelliteorSSRmarkerstoevaluatediploidandtriploidac- CarreelF,De-LeonDG,LagodaPJL,LanaudC,JennyC,Horry
cessions from the Embrapa germplasm bank. The lack of JP and Du-Montcel HT (2002) Ascertaining maternal and
geneticcorrelationbetweenaccessionsofcommongenetic paternallineagewithinMusabychloroplastandmitochon-
drialDNARFLPanalyses.Genome45:679-692.
originisprobablyrelatedtorecentanthropicactionthathas
CresteS,Tulmann-NetoA,SilvaSOandFigueiraA(2003)Ge-
influencedgenotypedispersion(Jarretetal.,1993).
neticcharacterizationofbananacultivars(Musaspp.)from
Inthepresentstudy,severaltriploidcultivarsshared
Brazil using microsatellite markers. Euphytica 132:259-
high similarity values with diploid genotypes, e.g., the
268.
triploidNanica,whichshowed91%similaritywiththedip-
CresteS,Tulmann-NetoA,VencovskyR,SilvaSOandFigueira
loid M-48, 84% similarity with the diploids Khai Nai On A(2004)GeneticdiversityofMusadiploidandtriploidac-
andTuugiaand80%similaritywiththediploidsSowmuk, cessionsfromtheBrazilianbananabreedingprogramesti-
JaranandNBA-14.SinceNanicaisanexport-typetriploid mated by microsatellite markers. Gen Res Crop Evol
cultivar(Cavendish)andexhibitscompletefemalesterility, 51:723-733.
the high level of similarity with diploids provides novel Cruz CD and Schuster I (2004) GQMOL – Aplicativo Compu-
breedingpossibilities. tacionalparaAnálisedeDadosMolecularesedeSuasAsso-
ciaçõescomCaracteresQuantitativos.UFV,Viçosa.
Molecular analyses have revealed extensive genetic
Davey MW, Stals E, Ngoh-Newilah G, Tomekpe K, Lutsy C,
diversityinMusaacuminata(Lannaudetal.,1992;Carreel
MarkhamR,SwennenRandKeulemansJ(2007)Sampling
etal.,1994,2002;Udeetal.,2002;Ruangsuttaphaetal.,
strategiesandvariabilityinfruitpulpmicronutrientcontents
2007). In contrast, analyses using RAPD, SSR and AFLP
ofWestandCentralAfricanbananasandplantains(Musa
molecular markers have revealed lower levels of genetic sp.).JAgricFoodChem55:2633-2644.
variation among banana accessions (Nsabimana and Sta- DoyleJJandDoyleJL(1990)IsolationofplantDNAfromfresh
den,2007;Ningetal.,2007;Wangetal.,2007).Inthiscon- tissue.Focus12:13-15.
text,theuseofDArTmarkerstoclearlydefinetherelation- EnglbergerL,AalbersbergW,RaviP,BonninE,MarksGC,Fitz-
ships among and within wild material and to improve gerald MH and Elymore J (2003a) Further analyses on
diploid and triploid cultivars in germplasm collections Micronesian banana, taro, breadfruit and other foods for
couldfacilitatethepredictionofhowhybridswillperform, provitaminA,carotenoidsandminerals.JFoodCompAnal
16:219-236.
therebyincreasinggeneticgains.
EnglbergerL,Darnton-HillI,CoyneT,FitzgeraldMHandMarks
Ourresultsprovideadditionalinformationforbanana
GC (2003b) Carotenoid-rich bananas: A potential food
genetic breeding programs by allowing the selection of
sourceforalleviatingvitaminAdeficiency.FoodNutrBull
carotenoid-rich diploids for crosses with triploids to pro-
24:303-318.
vide novel tetraploid cultivars with improved functional
EnglbergerL,SchierleJ,MarksGCandFitzgeraldMH(2003c)
properties. Nutritional (biofortification) and functional
Micronesian banana, taro, and other foods: Newly recog-
breeding,alliedwithimprovedproductivityanddiseasere- nized sources of provitamin A and others carotenoids. J
sistance,shouldresultinbetternutritionalvalue.Sinceba- FoodCompAnal16:3-19.
nanas are the most consumed fruit in the world, the EnglbergerL,MarksGC,FitzgeraldMHandLippweK(2005)
ingestionofnutritionallyenhancedcultivarscouldsignifi- FoodcompositiondatafromFederatedStatesofMicronesia.
cantlyimproveconsumershealthbyreducingtheincidence Micronesica37:195-213.
of diseases and indirectly decrease spending on public JaccoudD,PengK,FeinsteinDandKilianA(2001)Diversityar-
health,whichiscurrentlysignificantlyelevatedworldwide. rays:Asolidstatetechnologyforsequenceinformationin-
dependentgenotyping.NucleicAcidsRes29:e25.
JainPK,SainiML,PathakHandGuptaVK(2007)Analysisof
Acknowledgments
geneticvariationindifferentbanana(Musaspecies)variety
This work was funded by the Generation Challenge usingrandomamplifiedpolymorphicDNAs(RAPDs).AfrJ
Programme/Genotyping Support Service (GSS). S. Silva Biotech6:1987-1989.
was supported by a fellowship from CNPq. The authors JarretRL,VuylstekeDR,GawelNJ,PimentelRBandDunbarLJ
(1993)Detectinggeneticdiversityindiploidsbananasusing
thank V. Quecini (Embrapa Grape and Wine) for help in
PCRandprimersfromahighlyrepetitiveDNAsequence.
draftingthemanuscript.
Euphytica68:69-76.
KruskalJB(1964)Multidimensionalscalingbyoptimizinggood-
References
nessoffittoanometrichypothesis.Psychometrika29:1-27.
AmorimEP,AmorimVBO,SantosJB,SouzaAPandSouzaJC LannaudC,Du-MontcelHT,JolivotMP,GlszmannJCandDe
(2006) Genetic distance based on SSR and grain yield of LeonDG(1992)Variationofribosomalgenespacerlength
inter and intrapopulational maize single cross hybrids. amongwildandcultivatedbanana.Heredity68:147-156.
Maydica51:507-513. MantelN(1967)Thedetectionofdiseaseclusteringandagener-
CarreelF,FauréS,De-LeonDG,LagodaPJL,PerrierX,BakryF, alizedregressionapproach.CancerRes37:209-220.
Du-MontcelHT,LanaudCandHorryJP(1994)Évaluation Melo EA, Lima VLAG and Maciel MIS (2006) Polyphenol,
deladiversitégénétiquechezlesbananiersdiploids(Musa ascorbicacidandtotalcarotenoidcontentsincommonfruits
sp.)GenSelEvol26:125s-136s. andvegetables.BrazJFoodTechnol9:89-94.
Amorimetal. 103
Ning SP, Xu LB, Lu Y, Huang BZ and Ge XJ (2007) Genome Vaz Patto MC, Satovic Z, Pêgo S and Fevereiro P (2004) As-
compositionandgeneticdiversityofMusagermplasmfrom sessing the genetic diversity of Portuguese maize germ-
China revealed by PCR-RFLP and SSR markers. Sci plasmusingmicrosatellitemarkers.Euphytica137:63-72.
Horticult114:281-288. WallM(2006)Ascorbicacid,vitaminA,andmineralcomposi-
NsabimanaAandStadenJV(2007)Assessmentofgeneticdiver- tion of banana (Musa sp.) and papaya (Carica papaya)
sity of Highland bananas from the National Banana cultivarsgrowninHawaii.JFoodCompAnal19:434-445.
Germplasm Collection at Rubona, Rwanda using RAPD Wan Y, Watanabe JA, Yi SS, Htaik T, Win K, Yamanaka S,
markers.SciHorticult113:293-299. Nakamura I and Watanabe KN (2005) Assessment of ge-
Rodriguez-AmayaDB(1999)AGuidetoCarotenoidAnalysisin neticdiversityamongthemajorMyanmarbananalandraces.
Foods,1stedition.ILSIPress,City,64pp. BreedSci55:365-369.
Rohlf FJ (2000) NTSys-pc: Numerical Taxonomy and Multi- WangH,CaoGandPriorRL(1997)Oxygenradicalabsorbing
variate Analysis System, v. 2.1. Exceter Software, New capacityofanthocyanins.JAgricFoodChem45:304-309.
York. WangXL,ChiangTY,RouxN,HaoGandGeXJ(2007)Genetic
Ruangsuttapha S, Eimert K, Schoder MB, Silayoi B, Den- diversityofwildbanana(MusabalbisianaColla)inChina
duangborinpat J and Kanchanapoom K (2007) Molecular as revealed by AFLP markers. Genet Res Crop Evol
phylogeny of banana cultivares on HAT-RAPD. Gen Res 54:1125-1132.
CropEvol54:1565-1572. WhiteJ,LawJR,MackayI,ChalmersKJ,SmithJSC,KilianA
andPowellW(2008)ThegeneticdiversityofUK,USand
Setiawan B, Sulaeman A, Giraud DW and Driskell JA (2001)
Australian cultivars of Triticum aestivum measured by
Carotenoid content of selected Indonesian fruits. J Food
DArT markers and considered by genome. Theor Appl
CompAnal14:169-176.
Genet116:439-453.
Sharrock S and Lutsy C (2000) Nutritive value of banana. In:
WittenbergAH,VanDerLeeT,CaylaC,KilianA,VisserRGand
INIPABAnnualReport.INIPAB,Montpellier,pp28-31.
Schouten HJ (2005) Validation of the high-throughput
SokalRRandMichenerCD(1958)Astatisticalmethodforevalu-
markertechnologyDArTusingthemodelplantArabidopsis
ating systematic relationships. Univ Kansas Sci Bull
thaliana.MolGenGenet274:30-39.
38:1409-1438.
XiaL,PengK,YangS,WenzlP,VicenteMC,FregeneMand
Someya S, Yoshiki Y and Okubo K (2002) Antioxidant com-
Kilian A (2005) DArT for high-throughput genotyping of
pounds form bananas (Musa cavendish). Food Chem
cassava (Manihot esculenta) and its wild relatives. Theor
79:351-354.
ApplGenet110:1092-1098.
Souza SL, Moreira APB, Santana HM and Alencar ER (2004)
Conteúdo de carotenos e provitamina A em frutas comer-
Internet Resources
cializadas em Viçosa, Estado de Minas Gerais. Acta Sci
Agron26:453-459. FoodandAgricultureOrganization(FAO)oftheUnitedNations,
StodartBJ,MackayMCandRamanH(2007)Assessmentofmo- www.faostat.fao.org/site/340/default.aspx. (March 19,
lecular diversity in landraces of bread wheat (Triticum 2008).
aestivumL.)heldinanexsitucollectionwithDiversityAr-
raysTechnology(DArT).AustJAgricRes58:1174-1182. AssociateEditor:EveraldoGonçalvesdeBarros
Ude G, Pillay M, Nwakanma DC and Tenkouano GUA (2002)
AnalysisofgeneticandsectionalrelationshipsinMusaus- Licenseinformation:Thisisanopen-accessarticledistributedunderthetermsofthe
CreativeCommonsAttributionLicense,whichpermitsunrestricteduse,distribution,and
ingAFLPmarkers.TheorApplGenet104:1239-1245. reproduction in any medium, provided the original work is properly cited.