convergent evolution in the genetic basis of mu¨llerian ... · from a genomic bac library and...

Copyright � 2008 by the Genetics Society of AmericaDOI: 10.1534/genetics.107.082982

Convergent Evolution in the Genetic Basis of Mullerian Mimicry inHeliconius Butterflies

Simon W. Baxter,*,1 Riccardo Papa,† Nicola Chamberlain,‡ Sean J. Humphray,§

Mathieu Joron,** Clay Morrison,†† Richard H. ffrench-Constant,‡

W. Owen McMillan†† and Chris D. Jiggins*

*Department of Zoology, University of Cambridge, Cambridge CB2 3EJ, United Kingdom, †University of California, Irvine, California92697, ‡University of Exeter, Cornwall TR10 9EZ, United Kingdom, §The Wellcome Trust Sanger Institute, Cambridge

CB10 1SA, United Kingdom, **Institute for Evolutionary Biology, University of Edinburgh, Edinburgh EH9 3JT,United Kingdom and ††Department of Genetics, North Carolina State University,

Raleigh, North Carolina 27695

Manuscript received October 5, 2007Accepted for publication August 31, 2008

ABSTRACT

The neotropical butterflies Heliconius melpomene and H. erato are Mullerian mimics that display the samewarningly colored wing patterns in local populations, yet pattern diversity between geographic regions.Linkage mapping has previously shown convergent red wing phenotypes in these species are controlled byloci on homologous chromosomes. Here, AFLP bulk segregant analysis using H. melpomene crosses identifiedgenetic markers tightly linked to two red wing-patterning loci. These markers were used to screen a H.melpomene BAC library and a tile path was assembled spanning one locus completely and part of the second.Concurrently, a similar strategy was used to identify a BAC clone tightly linked to the locus controlling themimetic red wing phenotypes in H. erato. A methionine rich storage protein (MRSP) gene was identified withinthis BAC clone, and comparative genetic mapping shows red wing color loci are in homologous regionsof the genome of H. erato and H. melpomene. Subtle differences in these convergent phenotypes implythey evolved independently using somewhat different developmental routes, but are nonetheless regulatedby the same switch locus. Genetic mapping of MRSP in a third related species, the ‘‘tiger’’ patterned H.numata, has no association with wing patterning and shows no evidence for genomic translocation of wing-patterning loci.

ONLY recently has it become feasible to study the mo-lecular basis of phenotypic adaptation in higher

organisms, which has generated considerable interest inidentifying genes controlling adaptive traits (Nachman

et al. 2003; Mundy 2005; Protas et al. 2006). Commonly,positional cloning of qualitative traits involves perform-ing crosses for trait segregation, creating linkage mapsusing the progeny, identification of the genomic regionresponsible for variation, and then genetic mapping ofcandidate genes to test for association by linkage (Heckel

et al. 1998). This approach has been successful in asso-ciating candidate genes to the causal locus in severalcases, including insecticide resistance in Heliothis vir-escens (Gahan et al. 2001), pelvic reduction in threespinesticklebacks (Shapiro et al. 2004), wing color patterningin butterflies (Kronforst et al. 2006b), albinism in cavepopulation Astyanax fish (Protas et al. 2006), and

adaptive variation in coat color of mice (Steiner et al.2007). Identifying the adaptive trait becomes moredifficult where no suitable candidate genes exist or allknown candidates can be ruled out by linkage map-ping. In this case, sequencing of genomic bacterialartificial chromosome (BAC) clones spanning the ge-nome region is required. Recently, (Colosimo et al.2005) positionally mapped and sequenced the locuscontrolling armor plating in threespine sticklebacksfrom a genomic BAC library and identified a strongcandidate for determination of lateral plate pheno-types. Identifying genetic markers tightly linked to theplate morph locus was achieved by screening stickle-back crosses with amplified fragment length poly-morphism (AFLP) markers, enabling markers linked tothe region of interest to be identified without any priorknowledge of the genome sequence. Markers devel-oped in this way can also facilitate comparative studiesto determine whether the same regions of the genomeare involved in adaptation in different species.

A comparative mapping approach has already beenapplied to the three butterfly species, Heliconius melpom-ene, H. erato, and H. numata ( Joron et al. 2006). The firsttwo species are comimics that share the same warningly

Sequence data from this article have been deposited with GenBankunder accession nos. FI109935–FI109945, FI107283–FI107295, FI136209–FI136223, EU711403–EU711404, and AC216670.

1Corresponding author: Department of Zoology, University of Cambridge,Downing St., Cambridge CB2 3EJ, United Kingdom.E-mail: [email protected]

Genetics 180: 1567–1577 (November 2008)

colored wing patterns in local populations and differentphenotypes between geographic regions of Central andSouth America, while H. numata is a species with poly-morphic populations in which a single genetic locuscontrols multiple morphs. A positional cloning ap-proach in H. melpomene led to identification of markerslinked to the Yb patterning locus, which controls ayellow hind-wing bar. These were subsequently mappedin the other two species and found to be tightly linkedto the H. erato locus Cr, which has similar phenotypiceffects to Yb, and the H. numata locus P, which controlswhole-wing phenotypic polymorphism. This surprisingresult indicates that a single genetic locus is controllingboth convergent and divergent phenotypes in differentlineages ( Joron et al. 2006).

Wing color and pattern variation among the �25mimetic races of H. melpomene and H. erato is controlledby several unlinked loci with major phenotypic effect(Sheppard et al. 1985). Extensive crossing experimentshave shown that in H. melpomene, genetic factors on atleast 4 of the 21 chromosomes are involved in wing colorpatterning (Sheppard et al. 1985; Joron et al. 2006).H. erato has a similar architecture, with the involvementof at least 4 different chromosomes. Recent work hasshown preliminary evidence for further homology be-tween the two species in addition to that of Yb and Cr.The locus Ac, which controls a dumbbell-shaped ele-ment in H. melpomene, is on the homologous chromo-some to Sd, which controls forewing band shape in H.erato (Kronforst et al. 2006a). Loci controlling red wingpattern elements genetically map to the same chromo-some as Cubitus interruptus (Ci) in H. erato (D locus), H.melpomene (B and D loci), and sister species H. cydno (Glocus) ( Joron et al. 2006; Kronforst et al. 2006a).

While H. melpomene B/D (hereafter HmB/HmD) andH. erato D (HeD) control broadly similar, and adaptivelyconvergent mimetic pattern elements, there are none-theless some fundamental differences both in the actionof the switch loci in the two species and in the pheno-typic expression of their pattern elements. First, HeDcontrols both red/orange and yellow color in the me-dial forewing band. In contrast the H. melpomene yellowforewing is controlled by a completely unlinked locus,N, and is possibly influenced by an additional locus M(Mallet 1989). Second, the hind-wing ray elementsin H. erato are determined by vein position, with all pat-tern elements lying in intervein compartments. The H.melpomene rays involve an orange band that bisects veinboundaries and is apparently homologous in positionto the yellow band controlled by the locus Yb on LG15.Thus, these genes control adaptively convergent pat-terns but their resultant phenotypes and mode of actionare sufficiently distinct, leading to speculation whetherthe same genes are homologous (Nijhout 1991) or not(Mallet 1989). We here carry out fine-scale mapping todetermine whether red patterning elements of HmB/HmD and HeD are indeed homologous.

Genetic analysis using AFLPs provides a method forrapidly generating large numbers of markers that can beused to analyze poorly characterized genomes (Heckel

et al. 1999; Brugmans et al. 2002; Kazachkova et al.2004; Jiggins et al. 2005). Previous genetic work on H.melpomene wing patterning relied on the discovery of anAFLP marker tightly linked to the Yb locus on linkagegroup 15 ( Jiggins et al. 2005). Here we apply an AFLPmethod for targeted identification of molecular markerslinked to phenotypic traits of interest in Lepidoptera.Markers flanking H. melpomene HmB and HmD weredeveloped and, unexpectedly, show these loci to betightly linked. BAC library screening using AFLP mark-ers linked to red patterning loci identified candidateclones for sequencing that will ultimately enable posi-tional cloning of these phenotypes. Finally, comparativemapping shows a gene linked to HeD in H. erato is in ahomologous chromosomal region to HmB/HmD of H.melpomene, however is unlinked to wing patterning in therelated species H. numata.

MATERIALS AND METHODS

Wing pattern phenotypes: Here, we use the terminologyHmB for the locus controlling presence or absence (Hmb) ofthe H. melpomene red medial forewing band and HmD for thelocus controlling the presence or absence (Hmd) of the redforewing base, hind-wing bar, and hind-wing rayed phenotype.A single locus, D, controls this convergent pattern in H. erato,referred to here as HeD. Note that HeD was originally termedDRy by Sheppard et al. (1985) and subsequently simplified to Dby ( Jiggins and McMillan 1997) (Figure 1).

Insect crosses: All H. melpomene crosses were performedwithin sealed, outdoor cages at the Smithsonian Institute,Panama. Single-pair matings were set up between H. m.melpomene collected from French Guiana (4�549 N, 52�219W), homozygous for the HmB phenotype and H. m. malleticollected from Ecuador (0�109 S, 77�419 W) homozygous forthe HmD phenotype. Crosses were then performed betweenunrelated F1 heterozygotes, to generate three large F2 broodsthat were reared to adulthood (brood 44 with 164 progeny;brood 48 with 109 progeny; and brood 52 with 99 progeny).The combined 372 F2 progeny segregated for the presence orabsence of HmB and HmD phenotypes in Mendelian ratios of1:2:1 (100 bD/b-:197 B-/-D:75 Bd/-d), not differing significantlyfrom expected x2¼ 4.66, P . 0.1 [2 degrees of freedom (d.f.)].A similar crossing strategy was preformed on H. erato at theUniversity of Puerto Rico. For this cross, outbred stocks of H.erato etylus, collected from the Zamora River (3�559 S, 78�509W) and H. himera, collected from Vilcabamba (4�169 S, 79�139W), both in Loja Province in southern Ecuador were crossed toproduce F1 hybrid lines. Unrelated F1 individuals were crossedto produce a family containing 92 progeny that segregated forthe H.e. etylus HeDet phenotype and H. himera HeDhi pheno-types in the expected 1:2:1 Mendelian ratio (23 HeDetHeDet:44HeDetHeDhi:25 HeDhiHeDhi) not differing significantly fromexpected x2 ¼ 0.26, P . 0.8 (2 d.f.) (Kapan et al. 2006). Forall crosses, wings were removed from butterflies and storedin paper envelopes, and bodies preserved in a salt-saturatedsolution of 10% DMSO and 0.2 m EDTA at –20�.

H. numata crosses have been described previously ( Joron

et al. 2006). Here, two polymorphic F2 broods were analyzedthat segregated for three alleles of the supergene P from the

1568 S. W. Baxter et al.

morphs H. numata f. elegans (Pele), H. n. f. aurora (Paur), and H. n.f. silvana (Psil) from eastern Peru. Four phenotypes segregatedfor brood 502 among 88 F2 progeny (21 PelePaur:18 PeleP sil:32PsilPaur:17 PsilP sil) (also see Figure 2C in Joron et al. 2006) andtwo phenotypes segregated for brood 472 among 80 progeny(45 PeleP sil:35 PsilP sil).

Nucleotide extraction and processing: Genomic DNA wasextracted from small sections of abdomen or thorax with theDNeasy tissue kit (QIAGEN) and resuspended in TE buffer(10:0.1, pH 8.0). AFLP templates were prepared for 35 H.melpomene brood 44 progeny, F1 parents, and grandparentsusing half reactions of AFLP analysis system II (Invitrogen LifeTechnologies) according to manufacturer’s instructions, withEcoRI and MseI restriction enzymes. AFLP PCR reactions wereperformed using fluorescently labeled Eco primers with two orthree selective bases and paired with an Mse primer containingthree selective bases. AFLP PCR reactions were diluted 1/60in water, and 1 ml added to 9 ml of HiDi formamide solutioncontaining 0.01 ml of GeneScal-500 LIZ size standard (AppliedBiosystems). Products were separated on a 3730 DNA Analyzer(Applied Biosystems) using GeneMapper. Data files wereconverted to GeneScan format and AFLP peaks sized usingGeneScan Analysis software (version 3.1.2). Analyzed fileswere then imported into Genographer (version 2.0), whichdigitally reconstructed chromatogram peaks to form imagessimilar to electrophoretic gels.

Sequencing AFLP markers: AFLP PCR reactions containingbands linked to the color pattern loci were separated usingpolyacrylamide gel electrophoresis (acrylamide 6%, 13 TBE,7 m urea) at 1750 V, 80 W for 1.5–3 hr. Acrylamide gels werestained with Silver Sequence (Promega) to visualize DNAbands, which were then excised and resuspended in water. Allproducts were reamplified and sequenced using BigDye ter-minator v3.1 (Applied Biosystems) using an ABI3730. TenAFLP markers were sequenced (FI109935–FI109944) and usedto design genotyping assays for genetic mapping in H. melpomene

brood 44, and where possible, broods 48 and 52. Genotypingwas performed using (i) size variation in PCR products whenseparated using agarose gel electrophoresis or (ii) sequencingF1 parents from crosses to identify restriction enzymes thatrecognize single nucleotide polymorphisms segregating in F2

progeny. CodonCode Aligner V1.5.2 software was used toanalyze sequence chromatograms.

Genetic markers and linkage map assembly: As crossingover between sister chromatids does not occur during oogen-esis in Lepidoptera, all markers on the same chromosomeinherited from the mother are in complete linkage. Three H.melpomene F2 broods (44, 48, and 52) segregating for HmB andHmD were genotyped for two linkage group 18 (LG18) mo-lecular markers, nuclear gene Cubitus interruptus and micro-satellite Hm14. Alleles inherited from the F1 mother were usedto assign chromosome prints to progeny. This provided amolecular diagnostic to determine whether HmB/D heterozy-gous progeny inherited a Bd or bD chromosome from the F1

mother. To calculate recombinational distances between HmBand HmD and known markers on LG18, chromosomes in-herited from F1 fathers were analyzed (crossing over doesoccur during spermatogenesis in Lepidoptera). Direct esti-mation of recombination between HmB and HmD was notpossible as these dominant cosegregating phenotypes are in-herited in repulsion (Figure 2).

Linkage map assembly: The H. erato linkage map wasadapted from Kapan et al. (2006). MapMaker Macintosh2.0 was used to assemble maps for H. melpomene and H. numata.H. numata brood 472 contained 80 F2 individuals and wasgenotyped for RpS30 and MRSP (there was no variation at Ci),and brood 502 contained 88 individuals and was genotyped forRpS30 and Ci (there was no variation at MRSP).

H. melpomene BAC library preparation, screening, and endsequencing: A H. melpomene BAC library was constructed usingvector pECBAC1 and contains 18,816 clones with an averageinsert size of 123 kb, providing a 7.93 genome coverage(Amplicon Express) ( Joron et al. 2006). A bacterial clonephysical map of the genome was then constructed usingHindIII restriction enzyme fingerprinting (Humphray et al.2001; Schein et al. 2004). BAC library plates were directlycultured into 170 ml 23 TY in 384-well plates. After overnightgrowth the BAC DNA was extracted by alkaline lysis on aPackard MiniTrack and digested with HindIII in the 384-wellplates. Following electrophoresis on 121 lane 1% agarose gels,the data were collected using a Typhoon 8600 fluorimagerand raw images were entered into the fingerprint databaseusing the software IMAGE (http://www.sanger.ac.uk/Software/Image). The output of normalized band values, sizes, and geltraces were analyzed in FPC (Soderlund et al. 2000), whichbins and orders clones on the basis of shared bands, takingmarker data into account when available. Fingerprinting wassuccessful for 17,494 clones, of which 13,010 assembled into2086 contigs (the remainder were singletons). Each contigaveraged 6.2 clones and 791 contigs contained 5 or moreclones. AWebFPC database displaying all contigs is available athttp://www.sanger.ac.uk//Projects/H_melpomene/WebFPC/helim/small.shtml.

H. melpomene BAC library clones were printed onto nylonmembranes and hybridization performed using eight probeslinked to the HmB and HmD loci including three AFLP mark-ers, Hm_eCCmCTA (primers CC-CTA_F2 X CC-CTA_R2),Hm_eCGmATG213, and Hm_eACTmGTA195; four BAC endsequences, bHM40A21_sp6, bHM36K2_sp6, bHM28L23_sp6,and bHM40C14_sp6; and gene MRSP (primers MRSP_F1 3MRSP_R1). PCR products were generated under the followingconditions: 50 ng genomic DNA of H. melpomene 13 reactionbuffer, 2.5 mm MgCl2, 0.2 m dNTP, 50 pmol each primer, Taqpolymerase (Bio-Line). PCR products were then labeled with

Figure 1.—Heliconius melpomene and H. erato comimics. Dphenotype: Dorsal and ventral wings from H. melpomene ecua-dorensis and H. erato etylus, displaying the D phenotype.Specimens were collected near Zamora, Ecuador. The yellowforewing patch is controlled by HeD in H. erato and by asecond locus (called N) in H. melpomene. B phenotype:H. melpomene melpomene and H. erato hydara were collected nearCayenne, French Guiana (4�919 N, 52�369 W). All individualsare males.

Mimicry in Heliconius Butterflies 1569

a-32P using Prime-It II random primer labeling kit (Stratagene)and hybridization performed as outlined in ( Joron et al. 2006)(Table 1). BAC clones identified by the eight probes are listed insupplemental Table 1.

H. erato BAC library preparation and screening: Two H.erato BAC libraries, one partially restricted with EcoRI and onewith BamHI, were created from a line of H. erato petiveranacollected in Gamboa, Panama, and inbred for several gener-ations (C. Wu, D. Proestou, D. Carter, E. Nicholson, F. A.Santos, S. Zhao, H.-B. Zhang and M. R. Goldsmith,unpublished results). Both libraries contain 19,200 clonesarrayed in 384-well plates and the average insert sizes for theEcoRI and BamHI libraries were estimated to be 153 kb and 175kb, respectively. Libraries were printed to nylon membranes ina high-density array. Under this array strategy, each clone wasspotted twice in an effort to minimize false positives inscreening experiments. An AFLP marker tightly linked tothe HeD locus, He_CCAC491 (FI109945) (Kapan et al. 2006),was isolated, sequenced, and used to probe the H. erato BAClibraries. Probing both BAC libraries identified 10 positiveclones, the largest of which, BBAM-25K4 (AC216670), wassequenced at 83 coverage by the Baylor College of Medicine’sGenome Sequencing Center (Houston).

RESULTS

Molecular markers for red wing pattern loci: Geneticcrosses have previously shown HmB and HmD are dom-inant loci on the same linkage group (Turner 1972),designated linkage group 18 (LG18) in H. melpomene( Jiggins et al. 2005; Joron et al. 2006). To generatemolecular markers linked to these loci, AFLP bulk seg-regant analysis was performed on the F1 parents ofbrood 44 and two F2 progeny bulks. The first bulkcontained 12 F2 individuals, homozygous Hmb (no redforewing band, HmbD/b-), and the second bulk con-tained 12 F2 individuals homozygous for Hmd (no redhind-wing rays, bar, and red forewing base, HmBd/-d). Asthe F2 progeny were siblings, the two bulks were ex-pected to share many AFLP bands, but should bestrongly differentiated in genome areas tightly linkedto Hmb or Hmd. In Lepidoptera, chromosomal crossingover occurs during spermatogenesis and not oogenesis(Maeda 1939; Turner and Sheppard 1975; Traut

1977). Therefore, AFLP analysis focused on bands in-herited from the F1 father for recombinational linkagemapping. Two hundred fifty-six AFLP primer combina-tions produced�5100 bands in both F2 bulks, inheritedfrom the F1 father. In addition, 19 bands were presentin only one bulk and represented candidate molecularmarkers for red pattern elements. AFLP primer combi-nations for these 19 bands were reanalyzed in F1 parentsand 35 individual progeny (including the 24 individualsused to assemble the bulks). Ten AFLP markers wereconfirmed linked to Hmb or Hmd and 9 of the bandswere excluded as false positives due to banding patternsnot consistent with linkage group 18.

Five AFLP markers (Hm_eACAmCAT133, Hm_eCTmGCG148, Hm_eACTmGTA195, Hm_eCGmGAA139, andHm_eCCmACA177) were present in 16/16 of the Hmbindividuals tested and absent from most of the Hmdindividuals, suggesting these markers were geneticallylinked to the Hmb phenotype. The remaining five markers(Hm_eCCmCTA386, Hm_eCGmATG213, Hm_eACCmCAG224, Hm_eCTmCTC156, and Hm_eCAmAGC355)

Figure 2.—HmB and HmD phenotypes are inherited in re-pulsion. Schematic of LG18 chromosomes from an F1 crossbetween the heterozygous Bb/Dd female (dark gray) and male(light gray). Crossing over of sister chromatids does not occurduring oogenesis, so F2 progeny inherit either bD or Bd gen-otypes from the F1 female. As all LG18 chromosomal markersare completely linked when inherited from the F1 mother,molecular markers are used to assign chromosome prints. Re-combination between HmD and HmB or Hmd and Hmb mayoccur in the F1 male during spermatogenesis. Consequently,phenotypes are only informative for one of the B/b allelesor one of the D/d alleles and are therefore in repulsion.For example in the case of a bD/b- F2 individual, the F1 female-derived chromosome contributed b/D and F1 male-derivedchromosome b/-. Here, it is unclear whether one or twoHmD alleles are present, as the F1 female HmD phenotypemasks any contribution by the F1 male. Male informative gen-otypes from heterozygous B-/D- progeny were deduced usingchromosome prints. A B-/D- F2 heterozygote that inherited ab/D chromosome from the F1 female must have inherited adominant HmB allele from the F1 father to produce the ob-served phenotype. A hyphen ‘‘-’’ signifies unknown male-derived alleles.


TA

BL

E1

Pri

mer

seq

uen

ces

for

lin

kag

em

app

ing

and

H.

mel

pom

ene

BA

Cli

bra

ryp

rob

es

Map

pin

gm

eth

od

inH

.m

elpo

men

e

Pri

mer

Seq

uen

ceA

nn

eali

ng

tem

p.

Bro

od

44B

roo

d48

Bro

od

52B

AC

scre

enre

sult

Acc

essi

on

no

.

AF

LP

mar

kers

AC

C_C

AG

_FT

TC

TC

AG

CC

CT

AA

CA

GT

TT

CA

58L

PA

——

NA

FI1

0994

4A

CC

_CA

G_R

AT

TC

AC

CG

CC

GC

TA

AT

AC

AA

CC

-CT

A_F

1T

CC

CG

TC

TT

AG

CG

AG

CA

C68

LP

AL

PA

LP

AN

on

spec

ific

FI1

0993

9C

C-C

TA

_R1

GT

TC

TC

TA

CT

TT

CC

GC

CC

AT

TC

C-C

TA

_F2

TA

TC

TC

GC

GA

AG

CT

CT

CG

TA

60—

——

Co

nti

g19

6F

I109

939

CC

-CT

A_R

2G

GC

AG

AG

GG

AA

CC

TT

GT

GT

CG

-AT

GF

1A

AT

TC

CG

CC

AC

TA

AA

CA

GG

A58

LP

A—

—C

on

tig

446

Co

nti

g93

6F

I109

941

CG

-AT

GR

1G

GT

AG

CT

GA

CG

CT

GG

TG

AC

AC

A-C

AT

F1

TC

GC

CC

GA

TG

AT

AA

TA

TA

A55

LP

A—

—N

AF

I109

937

AC

A-C

AT

R1

GA

CA

TC

TA

CG

TC

AC

TA

TT

CC

CA

CT

-GT

AF

1T

AA

GG

CT

TA

TT

TT

TA

CT

TG

GC

55L

PA

LP

Ase

qC

on

tig

1729

FI1

0993

6A

CT

-GT

AR

1T

CT

GG

TG

TA

GA

GG

CA

AT

GT

CT

-CT

C_F

1G

AT

GC

TA

TT

CG

CT

GA

A55

LP

A—

—N

AF

I109

935

CT

-CT

C_R

1A

CT

CC

AG

TG

AA

TA

AT

AA

GT

TT

MR

SPM

RSP

_F1

GA

AC

AC

GC

TG

TG

CT

TT

TC

AA

59—

Ban

IISt

yIC

on

tig

16E

U71

1404

MR

SP_R

1A

AT

GG

TT

GG

CA

GA

AC

CA

GA

CM

RSP

_F2

GG

GT

CG

TT

GA

AG

AG

AA

CA

TG

G60

Mse

I—

—N

AE

U71

1403

MR

SP_R

2T

AC

GC

CA

GT

CA

AT

GA

AG

AC

G

BA

Cen

dp

rim

ers

bH

M5K

10_t

7FA

AA

CG

AC

CT

GG

GG

AA

TT

TT

T58

LP

AL

PA

LP

AN

AF

I107

292

bH

M5K

10_t

7RC

GT

GT

CA

CG

TA

GA

GT

CA

CG

AA

bH

M40

C14

_sp

6FT

CC

CA

TT

CA

AG

CC

TG

TA

GG

T58

Nla

III

SspI

Nla

III

Co

nti

g17

29F

I107

289

bH

M40

C14

_sp

6RT

GT

GA

AA

GT

TG

GT

CA

CG

TT

TG

bH

M7G

5_sp

6FA

GT

CT

GT

GG

TT

GG

TG

CC

AT

Ta

58L

PF

—L

PF

NA

FI1

0729

3b

HM

7G5_

sp6R

GA

CC

AG

CG

AT

TG

AC

AA

GA

CA

bH

M35

J3_t

7FT

TG

GT

GT

TA

CA

AA

GC

GG

TG

A58

—L

PA

LP

AN

AF

I107

286

bH

M35

J3_t

7RA

CG

AA

CG

CT

TC

GT

GC

TC

TA

Tb

HM

7G5_

t7F

AG

CC

GC

AG

GA

AA

AA

CC

TA

GT

a58

LP

FL

PF

LP

FN

AF

I107

294

bH

M7G

5_t7

RA

TG

CC

AT

GC

CT

TT

TT

AC

CA

Gb

HM

40A

21_s

p6F

GA

GG

CG

TT

GA

GT

GA

GG

AA

AC

58—

——

Co

nti

g93

6C

on

tig

1080

Co

nti

g13

05F

I107

288

bH

M40

A21

_sp

6RT

AA

CA

AG

GC

GT

TG

AG

TG

CA

G

bH

M47

N1_

t7F

TC

GA

CG

GC

TA

AT

GC

AT

AA

AA

a58

LP

FL

PF

LP

FN

AF

I107

290

bH

M47

N1_

t7R

AA

GG

CG

AC

TT

CG

AA

TG

AC

AT

Ab

HM

5D10

_sp

6FG

GT

AC

CA

CT

AC

TC

GG

GG

AG

A58

LP

FL

PF

LP

FN

AF

I107

291

bH

M5D

10_s

p6R

AC

AG

CG

GC

TA

AC

GA

AA

TG

AT

bH

M36

K2_

sp6F

bH

M36

K2_

sp6R

GC

TG

TA

TT

TC

TA

TC

TC

GC

AC

TG

GG

CA

CT

CT

TT

CG

AG

GG

TG

GT

A58

Alu

I—

Hae

III

Co

nti

g10

80C

on

tig

1352

Co

nti

g122

4F

I107

287

(con

tin

ued

)


were absent in all HmbD/b- individuals and present in mostof the HmBd/-d, suggesting these DNA fragments werelinked to the Hmd phenotype. Recombination eventsbetween AFLP markers and Hmd were observed in twoindividuals that were not used in the bulked segregantanalysis (Table 2).

HmB and HmD are tightly linked loci: The 10 Hmb- orHmd-linked AFLP bands were sequenced, followingseparation with polyacrylamide gel electrophoresis,and ranged in size from 133 bp to 386 bp. Oligonucleo-tides were designed for each AFLP marker (Table 1) forrecombinational linkage mapping in brood 44 to de-termine distances to HmB and HmD. A linkage map ofLG18 was constructed using 164 progeny from brood 44using microsatellite Hm14, recently identified LG18nuclear genes Bm44, Ribosomal protein gene S30 (RpS30)(Pringle et al. 2007) as well as Ci and 6 of the 10 AFLPsequence-tagged sites (Hm_eCGmATG213, Hm_eACCmCAG224, Hm_eACAmCAT133, Hm_eACTmGTA195,Hm_eCTmCTC156, and Hm_eCCmCTA386). The re-maining four AFLP HmB/D-linked markers either con-tained no sequence variation, which is required forlinkage mapping, or amplified .2 alleles (Figure 3). AllAFLPs that were successfully mapped were within 5.9 cMof Hmb and Hmd and closely linked to nuclear geneBm44. There were no recombinants between AFLPmarker Hm_eCGmATG213 and the HmB/D locus.

Although HmD and HmB were inherited in repulsion,molecular markers show both phenotypes geneticallymap to the same chromosomal region in brood 44. Totest the robustness of the association, two additionalbroods were genotyped for two HmB/D-linked AFLPmarkers. From the combined 372 progeny of broods 44,48, and 52, 199 inherited the B or b from the F1 male and173 inherited D or d from the F1 male. The first marker,Hm_eCCmCTA386, showed 1/173 recombinants to theHmD/d locus and 1/199 recombinants to HmB/b. Amarker flanking the two wing-patterning loci, Hm_eACTmGTA195, showed 3/173 recombinants to HmD/d and 4/199 recombinants to HmB/b. The AFLP markerclosest to HmB/D in brood 44 (Hm_eCGmATG213) wasnot polymorphic in broods 48 or 52.

BAC tile path surrounding HmB and HmD loci: Radio-labeled probes designed from three HmB/D-linked AFLPmarkers (Hm_eACTmGTA195, Hm_eCCmCTA386, andHm_eCGmATG213) were hybridized to H. melpomene BAClibrary filters and positive clones referenced against theBAC fingerprint database. Hm_eACTmGTA195 hybrid-ized to a single clone, in contig 1729 (Figure 4). Geneticmarkers designed from the sequenced BAC end of contig1729 were mapped in 372 progeny from H. melpomenebroods. Three recombination events were detected be-tween contig ends, enabling orientation and an estimationof recombination frequency in physical distance (3/372recombinants in �140 kb).

Probe Hm_eCCmCTA386 hybridized to 19 of 32 BACclones within the same fingerprint contig 196. Again,

TA

BL

E1

(Co

nti

nu

ed)

Map

pin

gm

eth

od

inH

.m

elpo

men

e

Pri

mer

Seq

uen

ceA

nn

eali

ng

tem

p.

Bro

od

44B

roo

d48

Bro

od

52B

AC

scre

enre

sult

Acc

essi

on

no

.

bH

M49

H2_

sp6F

AG

AG

CG

GC

TT

CG

TT

CA

GA

T58

—A

luI

—N

AF

I107

283

bH

M49

H2_

sp6R

AG

GG

AC

CC

AA

AT

CC

GT

AC

AT

bH

M28

F19

_sp

6FG

AC

GT

CG

CT

TC

GT

AC

CA

AA

T58

LP

AL

PA

LP

AN

AF

I107

284

bH

M28

F19

_sp

6RC

AT

GG

CA

CT

TC

GG

TT

AG

AC

Ab

HM

7N20

_sp

6FG

CG

GC

GT

GC

AA

AT

TA

TA

CT

Ca

58L

PF

seq

seq

NA

FI1

0729

5b

HM

7N20

_sp

6RA

TC

CA

TT

CA

TT

CT

GG

CT

TC

Gb

HM

28L

23_s

p6F

TG

AG

CA

GA

TA

TT

TT

CG

CG

TT

A58

——

—C

on

tig

1080

Co

nti

g13

52C

on

tig

1224

FI1

0728

5b

HM

28L

23_s

p6R

AC

GA

AT

TT

AG

CG

TT

GG

CA

AT

bH

M3O

8_t7

FT

GC

CA

AT

TA

AG

GA

TG

CC

AT

T58

——

—C

on

tig

446

FI1

3621

9b

HM

3O8_

t7R

TT

TC

AA

CA

CT

CT

TG

CC

CT

CA

LP

A,

len

gth

po

lym

orp

his

mid

enti

fied

usi

ng

agar

ose

gel

elec

tro

ph

ore

sis;

LP

F,le

ngt

hp

oly

mo

rph

ism

iden

tifi

edu

sin

gfl

uo

resc

entl

yla

bel

edP

CR

pro

du

cts

reso

lved

on

anA

BI3

730

seq

uen

cer.

Res

tric

tio

nen

do

nu

clea

ses

use

dfo

rid

enti

fica

tio

no

fp

oly

mo

rph

icsi

tes

inP

CR

pro

du

cts

are

list

ed.S

eq,p

oly

mo

rph

ism

iden

tifi

edu

sin

gd

irec

tse

qu

enci

ng.

No

po

lym

orp

his

mw

asd

etec

ted

inb

roo

ds

mar

ked

wit

ha

‘‘—’’.

aP

rim

ers

flu

ore

scen

tly

lab

eled

.


recombination was used to orientate the clone, with oneend containing no recombinants to the HmB or HmDloci. Probe Hm_eCGmATG213 hybridized to all 6clones from contig 446 and 5 of the 10 clones fromcontig 936. The BAC ends from contigs identified using

AFLPs were used to rescreen the library and close thegap between contigs 446 and 196. A BAC tile pathspanned the color-pattern locus HmB, flanked byrecombinant individuals (Figure 4). The HmD locus islocated in a larger region, overlapping with HmB.Primers designed for 22 BAC end sequences were usedto confirm clone overlap between the distinct contigs(supplemental Figure 1, supplemental Table 2).

Convergent patterns map to homologous loci: Theloci controlling red wing pattern elements in H.melpomene and H. erato are known to be on homologouschromosomes ( Joron et al. 2006; Kapan et al. 2006;Kronforst et al. 2006a). A previously identified H. eratoAFLP marker (He_CCAC491), positioned 1.5 cM fromthe HeD locus (Kapan et al. 2006), was sequenced andused to screen a H. erato BAC library. The probe hybrid-ized to clone BBAM-25K4 (GenBank AC216670); how-ever, recombinational linkage mapping ruled out thepossibility that the HeD locus was contained within thisclone. To identify nuclear genes close to HeD, BBAM-25K4was sequenced and contained a gene encoding a Methi-onine Rich Storage Protein (MRSP) (EU711403). Compara-tive mapping of MRSP in H. melpomene broods identified2/173 recombinants to the HmD locus and 2/199 re-combinants to the HmB locus and is positioned fartherfrom contig 196 (Figure 4). H. melpomene BAC libraryscreening was performed with MRSP and identifiedcontig 16, containing 63 clones that did not overlap withany previously detected contigs. For H. erato, we observedone recombinant between the HeD locus and MRSP from88 F2 individuals (in the H. erato etylus 3 H. himera cross).

Color diversity in a third Heliconius species, H.numata, is controlled by a single locus called P onlinkage group 15 of this species. Theoretical modelsconverge on the view that supergenes cannot form bybringing together previously unlinked loci through agraduation tightening of linkage or through transloca-tion of genetic elements from different regions of thegenome; instead, supergenes are believed to form via the

TABLE 2

AFLP markers identified using bulked segregant analysis that are linked to HmB/b and HmD/dwing-patterning loci

Band present in F2 progeny

H. melpomene AFLP marker Accession no. Size (bp) bb/D- progeny B-/dd progeny

Hm_eCCmCTA386 FI109939 386 0/16 19/19Hm_eCGmATG213 FI109941 213 0/16 19/19Hm_eACCmCAG224 FI109944 224 0/16 19/19Hm_eCAmAGC355 FI109943 355 0/16 15/16Hm_eCTmCTC156 FI109935 156 0/16 17/19Hm_eACAmCAT133 FI109937 133 16/16 1/18Hm_eCTmGCG148 FI109938 148 16/16 1/19Hm_eACTmGTA195 FI109936 195 16/16 1/18Hm_eCGmGAA139 FI109940 139 16/16 1/18Hm_eCCmACA177 FI109942 177 16/16 2/19

Figure 3.—Linkage map of H. melpomene LG18 (77.3 cM,log likelihood ¼ �92.61). Three nuclear genes (Ci, RpS30,and Bm44), one microsatellite marker (Hm14), and six AFLPsequence-tagged sites were genetically mapped in 164 F2 indi-viduals from brood 44. The linkage map was constructed us-ing MapMaker 2.0. Four AFLP markers identified by anasterisk were positioned manually on the basis of AFLP band-ing patterns for up to 35 individuals. All AFLP markers arewithin a 6.5-cM region.


recruitment of linked genetic elements (Charlesworth

and Charlesworth 1975; Turner 1977; Le Thierry

d’Ennequin et al. 1999). To empirically evaluate thosetheoretical results, we wanted to determine whetherthere was evidence for translocation from the HmB/HmD region to the P locus in H. numata. Three genes,MRSP, RpS30, and Ci were genetically mapped in crossessegregating for P phenotypes. Both MRSP and RpS30mapped to LG18 in H. numata, separated by 2.9 cM,while RpS30 and Ci were distantly linked at 59 cM,demonstrating the same gene order and similar dis-tances between markers as in H. melpomene and H. erato.This provides no evidence for large-scale rearrange-ments between LG18 and LG15 in H. numata. Thedistances between Ci and MRSP in these three speciesare H. erato, 74.4 cM; H. melpomene, 68.7 cM; and H.numata, 61.9 cM (Figure 5).

DISCUSSION

The mimetic color patterns of Heliconius butterfliesare striking phenotypic adaptations controlled by few

loci of major effect. Through adapting traditional AFLPmethods, we have identified molecular markers tightlylinked to the HmB and HmD loci, controlling thepresence or absence of red and orange pattern elementsin H. melpomene. Here we used the process of DNA-basedbulked segregant analysis, which has been widely ap-plied in plant studies to identify markers linked to aphenotypic locus (Michelmore et al. 1991). The prin-ciple involves dividing siblings of a cross into differentpools corresponding to a segregating phenotype. Thepools (or bulks) are largely heterozygous, except aroundthe locus controlling the phenotype. When analyzed withPCR-based methods such as AFLPs, markers that aredifferentially amplified between the bulks should genet-ically map near the locus of interest. Although we have yetto identify the actual gene(s) controlling red wing patternphenotypes, this work nonetheless demonstrates thatpositional cloning for phenotypic traits is a feasible goalfor nonmodel organisms outside plant research fields.

Our comparative genetic mapping data in H. melpom-ene and H. erato have shown that homologous chromo-somal regions are responsible for similar (but not

Figure 4.—Analysis of recombinants spanning the HmB/D region on LG18. A H. melpomene BAC library was probed with threeAFLP markers (open circles) linked to the HmB and HmD loci. Hm_eACTmGTA195 hybridized with 1 BAC clone, bHM40C14,within contig 1729. Hm_eCCmCTA386 hybridized to 19 clones from contig 196, (containing 32 clones). Hm_eCGmATG213 hy-bridized to 6 clones from contig 446 (containing 6 clones) plus 5 clones from contig 936 (containing 10 clones). PCR amplifi-cation using BAC end primers designed from contig 446 successfully amplified in BAC clones from contig 936, and vice versa,demonstrating these two contigs overlap (supplemental Figure 1). This implies contigs 446 and 936 may represent very divergenthaplotypes for the B/D region, as fingerprinting did not group these clones into a single contig. The BAC ends of some cloneswere sequenced and used to design genotyping assays for the 372 progeny (solid circles). As HmB and HmD are inherited in re-pulsion, recombinants from BAC ends to these loci are listed independently. Recombinants between contig ends were observed,enabling orientation of the HmB and HmD loci. The H. melpomene BAC library was screened using additional BAC end probes(solid squares), bHM40A21_sp6 (contig 936), bHM28L23_sp6 (contig 1305), and bHM36K2_sp6 (contig 196), which createda single tile path spanning the HmB locus and part of the HmD locus. Finally, a gene near the H. erato HeD locus (MRSP) wasused as a probe. For full details of BAC screening results see supplemental Table 1. Accession nos. for BAC end sequencesare FI107283–FI107295 and FI136209–FI136223. The figure scale is based on relative sizes of fingerprinted contigs.


identical) red wing phenotypes in these two species.This provides further evidence of homologous genomicregions shared between H. melpomene and H. erato, whichcontrol similar pattern elements ( Joron et al. 2006).There are however notable interspecific differencesbetween the D locus of these species. First, the H. eratoHeD locus controls not only the switch for presence orabsence of red elements, but the switch between yellowand red in the forewing, while the yellow forewing bandin H. melpomene is controlled by a separate unlinkedlocus, N. Second, the rayed phenotype of H. melpomeneconsists of a nail-shaped ray and a hind-wing bar thatcuts across several wing veins and appears to be veinindependent. In contrast the hind-wing rayed pattern ofH. erato appears to be entirely vein dependent with allpattern elements lying in intervein regions. Differencessuch as these between the species have led some tospeculate that the loci controlling pattern in the twospecies are unlikely to be homologous (Mallet 1989).

The results presented here and in a previous study( Joron et al. 2006) imply that common genetic switchmechanisms are responsible for convergent patterns inH. melpomene and H. erato, and these evolved indepen-dently in the two lineages. Despite findings consistentwith this hypothesis, it may not prove true for all races ofH. melpomene. Sheppard et al. (1985) reported recombi-nation between HmB and HmD loci, on the basis of aseries of H. melpomene progeny tests involving parentalindividuals collected from Suriname or Belem do Para,Brazil, (HmD homozygotes) and Apure in southwesternVenezuela or Trinidad (HmB homozygotes) (Turner

and Crane 1962). Parental genotypes were calculatedon the basis of phenotypes of progeny, and in 3 of11 cases, parents must have undergone recombinationevents to produce the phenotypes observed in the prog-eny. From this, Sheppard et al. (1985) calculated acrossover rate of 3/11 ¼ 27% (SE 6 13%). In contrast,here we show HmB and HmD to be tightly linked, moresimilar to that of the HeD locus in H. erato that controlsthe parallel red wing phenotype. Individuals used togenetically map the HmB/D loci in this study werecollected from Ecuador (HmD) and French Guiana(HmB), and therefore it is possible that the geneticlinkage relationships of these loci vary across the geo-graphic range of H. melpomene, perhaps due to chro-mosomal rearrangement, which would explain theapparent discrepancy in our results. Alternatively, ithas been suggested that two genes producing similarphenotypes to HmB may be present on LG18 ( J. R. G.Turner, personal communication). There is evidenceto support two independent loci controlling a differentphenotype in H. melpomene: the yellow forewing band(Mallet 1989; Mallet et al. 1990). Generally, individ-uals with genotypes NNNN display a strong yellow fore-wing band and NBNB do not. In some cases however,field-caught individuals expected to be NBNB present ayellow band that may be due to a second locus mm(Mallet et al. 1990). Determining whether multipleloci control similar phenotypes in H. melpomene requiresfurther crossing experiments and linkage mappingusing markers linked to known color pattern loci.

There are several cases of convergent phenotypes thatarose independently, either within the same gene orthrough different adaptive mechanisms. Protas et al.(2006) for example reported that albinism in twoisolated populations of the Mexican cave fish Astyanaxarose via independent mutations in the same pigmen-tation gene, Oca2. Hoekstra et al. (2006) investigatedvariable coat color between subpopulations of beachmice along the Florida Gulf Coast and identified a singlenucleotide polymorphism (SNP) in the melanocortin-1receptor pigmentation gene (Mc1r) with a major phe-notypic effect. When compared to beach mice withsimilar coat variability along the Atlantic coast, thecausal-derived SNP was absent, suggesting phenotypicadaptation evolved independently in this case. The

Figure 5.—Linkage maps comparing homologous chro-mosomes from H. erato, H. melpomene, and H. numata. H. eratolinkage map adapted from Kapan et al. (2006), showing HeDin the most likely position (70% probability density). BAC li-brary screening using H. erato AFLP marker He_CCAC491identified clone BBAM-25K4. Sequence analysis identified aMRSP gene, which was genetically mapped along with Ci inall three species. Two H. melpomene AFLP markers (Hm_eACTmGTA195 and Hm_eCTmCTC156) and one microsatellite(Hm14) cross amplified in H. numata and were geneticallymapped to LG18. Numbers shown are centimorgans.


major-effect loci that control color pattern across theHeliconius species investigated so far demonstrate acommon ‘‘toolbox’’ of genes controlling pattern poly-morphism. Between the comimetic species, homolo-gous loci have broadly similar effects on color pattern,suggesting functional similarities. However, the com-mon switch loci may affect downstream pathways indifferent ways, which would explain differences in modeof action of the major loci in different lineages.

The H. numata wing-patterning supergene is con-trolled at a single locus ‘‘P’’ on LG15, which enables thisspecies to mimic specific Melinaea butterflies. Accord-ing to theory, mimicry supergenes are unlikely to beformed by bringing together unlinked patterning genesto a single chromosomal region, e.g., via a translocation;instead, they are more likely to originate from a major genemutation, followed by a series of linked modifiers toperfect the mimicry (Charlesworth and Charlesworth

1975) or by the selection of gene clusters through a sieveeffect from redundant gene networks (Turner 1977;LeThierry d’Ennequin et al. 1999). Our data representthe first explicit empirical evaluation of those predic-tions: in accordance with theory, we have found nogenomic evidence, at this level of resolution, for thetranslocation of H. numata’s HmB/HmD orthologousregion from LG18 to LG15, to form part of the P su-pergene. Although the translocation hypothesis may beentirely ruled out only following definitive identifica-tion and cloning of the HmB/D genetic elements, theformation of the P supergene in H. numata does notappear to have involved a rearrangement of chromo-somal regions unlinked in related species. Nevertheless,it is interesting to note the differences in recombinationrates between markers RpS30 and MRSP flanking HmBand HmD loci. In H. melpomene, 7.7 cM separates thesetwo genes, yet in H. numata, the region is only 2.9 cM.The centimorgan differences may simply be an artifactof brood size, reduced recombination, or physicalgenome size differences. This lends support to thehypothesis that the genes involved in switching redpatterns in various Heliconius species are positionallyconserved in H. numata, but play little or no role in thegene networks underlying pattern polymorphism in thisspecies. Whatever the explanation, these observationsclearly imply that there is not a simple mapping ofphenotype to genotype. Instead, a conserved develop-mental switch has been adopted to generate bothdivergent and convergent patterns in different ecolog-ical and genetic contexts.

In H. melpomene, we have demonstrated the feasibilityof cloning color pattern genes using AFLP bulk segre-gant analysis, coupled with a BAC screening walk.Recombinants either side of the HmB locus were de-termined, within a physical distance of �700 kb, and arecombinant bordering one side of the HmD locus wasidentified, producing a larger region to search forcandidate genes. It is striking that multiple overlapping

fingerprinted contigs, each containing many clones,were identified in the proximity of HmD/HmB. Thatthese clones clustered into different contigs whenfingerprint data were analyzed implies that this regioneither must contain highly divergent alleles sampled inour BAC library or possibly contains duplicated geno-mic segments. The H. melpomene BAC library was con-structed using six pupae of unknown wing pattern froma polymorphic laboratory population segregating fordifferent alleles of HmB and HmD. Consequently, it isnot unexpected that allelic variability has been sampled,which might reflect different color patterns. Sequenc-ing multiple, overlapping clones will therefore be auseful tool for identifying allelic variants for futurestudies of field populations.

In our crosses, the components forming HmD andHeD (proximal red patch at the base of the forewing, thered hind-wing bar plus red hind-wing rays) are alwaysobserved together, in both species studied (Figure 1).Nonetheless, these phenotypes are not always associ-ated. Rare recombinants (or mutants) are known inboth species, in which the forewing patch and hind-wingbar are separated from the hind-wing rays (Mallet

1989). Furthermore, the races H. melpomene meriana andH. erato amalfreda display the red forewing proximalpatch and hind-wing bar, but not the ray phenotype.These races and occasional recombinants suggest theHeD and HmD loci are controlling these red patternelements by separate, tightly linked sites that may or maynot be part of the same gene.

The data presented here represent an important steptoward cloning the HmB/D gene(s) in H. melpomene andHeD in H. erato. For HmB, we have now assembled acontinuous BAC tile path encompassing the locus andare sequencing the BAC clones to identify candidategenes (supplemental Figure 1). The recombinationfraction in H. melpomene is estimated to be 180 kb/1cM, and just outside the HmB/D tile path (contig 1729,Figure 4) we have found 3/372 recombinants within�140 kb, which does not differ from expected (x2 ¼0.25, P . 0.6). Of course, factors such as inversions,duplications, or recombinational ‘‘deserts’’ could po-tentially complicate cloning of the locus, but we arenonetheless optimistic that this study represents asignificant step toward identification of the HmB/Dgene(s). This will offer novel insights into the molecularevolution underlying these dramatic color patterns andthe role of single loci in regulating developmentalpathways in a recent evolutionary radiation. The studyalso demonstrates the feasibility of this method forpositional cloning that can be applied to other non-model organisms with major-effect genes of economicor evolutionary interest.

We are grateful for the helpful comments and discussion providedby two anonymous reviewers. Technical support was received fromSequencing and Genotyping Facility at Universidad de Puerto Ricoand The Wellcome Trust Sanger Institute, United Kingdom. The


Ministerio del Ambiente in Ecuador granted permission to collectbutterflies and Autoridad Nacional del Medio Ambiente grantedpermission to work in Panama. This work was supported by theBiotechnology and Biological Sciences Research Council, the NationalScience Foundation (DEB-0715096 and IBN-0344705), the RoyalSociety, and the Smithsonian Tropical Research Institute.

LITERATURE CITED

Brugmans, B., A. Fernandez del Carmen, C. W. B. Bachem, H. van

Os, H. van Eck et al., 2002 A novel method for the constructionof genome wide transcriptome maps. Plant J. 31: 211–222.

Charlesworth, D., and B. Charlesworth, 1975 Theoretical ge-netics of Batesian mimicry. 2. Evolution of supergenes. J. Theor.Biol. 55: 305–324.

Colosimo, P., K. Hosemann, S. Balabhadra, G. J. Villarreal, M.Dickson et al., 2005 Widespread parallel evolution in stickle-backs by repeated fixation of Ectodysplasin alleles. Science307: 1928–1933.

Gahan, L. J., F. Gould and D. G. Heckel, 2001 Identification of agene associated with Bt resistance in Heliothis virescens. Science293: 857–860.

Heckel, D. G., L. J. Gahan, J. C. Daly and S. Trowell, 1998 A ge-nomic approach to understanding Heliothis and Helicoverparesistance to chemical and biological insecticides. Philos. Trans.R. Soc. Lond. B, Biol. Sci. 353: 1713–1722.

Heckel, D. G., L. J. Gahan, Y. B. Liu and B. E. Tabashnik,1999 Genetic mapping of resistance to Bacillus thuringiensistoxins in diamondback moth using biphasic linkage analysis.Proc. Natl. Acad. Sci. USA 96: 8373–8377.

Hoekstra, H. E., R. J. Hirschmann, R. A. Bundey, P. A. Insel and J.P. Crossland, 2006 A single amino acid mutation contributesto adaptive beach mouse color pattern. Science 313: 101–104.

Humphray, S., S. Knaggs and I. Ragoussis, 2001 Contiguation ofbacterial clones. Methods Mol. Biol. 175: 69–108.

Jiggins, C. D., J. Mavarez, M. Beltran, O. McMillan, J. S. Johnston

et al., 2005 A genetic linkage map of the mimetic butterfly, Hel-iconius melpomene. Genetics 171: 557–570.

Jiggins, C. D., and W. O. McMillan, 1997 The genetic basis of anadaptive radiation; mimicry in two Heliconius sibling species.Proc. R. Soc. Lond. B 264: 1167–1175.

Joron, M., R. Papa, M. Beltran, N. Chamberlain, J. Mavarez et al.,2006 A conserved supergene locus controls colour pattern di-versity in Heliconius butterflies. PLoS Biol. 4: 1831–1840.

Kapan, D. D., N. S. Flanagan, A. Tobler, R. Papa, R. D. Reed et al.,2006 Localization of Mullerian mimicry genes on a dense link-age map of Heliconius erato. Genetics 173: 735–757.

Kazachkova, N., J. Fahleson and J. Meijer, 2004 Establishment ofthe amplified fragment length polymorphism (AFLP) techniqueforgenotypingofpollenbeetle(Meligethesaeneus):anoxious insectpest on oilseed rape (Brassica napus). Mol. Biol. Rep. 31: 37–42.

Kronforst, M., D. Kapan and L. Gilbert, 2006a Parallel geneticarchitecture of parallel adaptive radiations in mimetic Helico-nius butterflies. Genetics 174: 535–539.

Kronforst, M., L. Young, D. Kapan, C. McNeely, R. O’Neill et al.,2006b Linkage of butterfly mate preference and wing colorpreference cue at the genomic location of wingless. Proc. Natl.Acad. Sci. USA 103: 6575–6580.

Le Thierry d’Ennequin, M., B. Toupance, T. Robert, B. Godelle

and P. H. Gouyon, 1999 Plant domestication: a model forstudying the evolution of linkage. J. Evol. Biol. 12: 1138–1147.

Maeda, T., 1939 Chiasma studies in the silkworm Bombyx mori. Jpn.J. Genet. 15: 118–127.

Mallet, J., 1989 The genetics of warning colour in Peruvian hybridzones of Heliconius erato and H. melpomene. Proc. R. Soc. Lond. B236: 163–185.

Mallet, J., N. Barton, G. Lamas, J. Santisteban, M. Muedas et al.,1990 Estimates of selection and gene flow from measures ofcline width and linkage disequilibrium in Heliconius hybridzones. Genetics 124: 921–936.

Michelmore, R. W., I. Paran and R. V. Kesseli, 1991 Identificationof markers linked to disease-resistance genes by bulked segregantanalysis: a rapid method to detect markers in specific genomicregions by using segregating populations. Proc. Natl. Acad. Sci.USA 88: 9828–9832.

Mundy, N. I., 2005 A window on the genetics of evolution: MC1R andplumage colouration in birds. Proc. Biol. Sci. 272: 1633–1640.

Nachman, M. W., H. E. Hoekstra and S. L. D’Agostino, 2003 Thegenetic basis of adaptive melanism in pocket mice. Proc. Natl.Acad. Sci. USA 100: 5268–5273.

Nijhout, H. F., 1991 The Development and Evolution of Butterfly WingPatterns. Smithsonian Institution Press, Washington, DC.

Pringle, E. G., S. W. Baxter, C. L. Webster, A. Papanicolaou, S. F.Lee et al., 2007 Synteny and chromosome evolution in the Lep-idoptera: evidence from mapping in Heliconius melpomene. Genet-ics 177: 417–426.

Protas, M. E., C. Hersey, D. Kochanek, Y. Zhou, H. Wilkens et al.,2006 Genetic analysis of cavefish reveals molecular conver-gence in the evolution of albinism. Nat. Genet. 38: 107–111.

Schein, J., T. Kucaba, M. Sekhon, D. Smailus, R. Waterston et al.,2004 High-throughput BAC fingerprinting. Methods Mol. Biol.255: 143–156.

Shapiro, M., M. Marks, C. Peichel, B. Blackman, K. Nereng

et al., 2004 Genetic and developmental basis of evolutionary pel-vic reduction in threespine sticklebacks. Nature 428: 717–723.

Sheppard, P. M., J. R. G. Turner, K. S. Brown, W. W. Benson and M.C. Singer, 1985 Genetics and the evolution of Mullerian mim-icry in Heliconius butterflies. Philos. Trans. R. Soc. Lond. B 308:433–613.

Soderlund,C.,S.Humphray,A.DunhamandL.French,2000 Contigsbuilt with fingerprints, markers, and FPC V4.7. Genome Res. 10:1772–1787.

Steiner, C. C., J. N. Weber and H. E. Hoekstra, 2007 Adaptive var-iation in beach mice produced by two interacting pigmentationgenes. PLoS Biol. 5: 1880–1889.

Traut, W., 1977 A study of recombination, formation of chiasmataand synaptonemal complexes in female and male meiosis ofEphestia kuehniella (Lepidoptera). Genetica 47: 135–142.

Turner, J. R. G., 1972 The genetics of some polymorphic forms ofthe butterflies Heliconius melpomene (Linnaeus) and Heliconius er-ato (Linnaeus). II. The hybridization of subspecies of H. melpom-ene from Surinam and Trinidad. Zoologica 56: 125–157.

Turner, J. R. G., 1977 Butterfly mimicry: the genetical evolution ofan adaptation. Evol. Biol. 10: 163–206.

Turner, J. R. G., and J. Crane, 1962 The genetics of some poly-morphic forms of the butterflies Heliconius melpomene Linnaeusand Heliconius erato Linnaeus. I. Major genes. Zoologica 47:141–152.

Turner, J. R. G., and Sheppard, 1975 Absence of crossing-over infemale butterflies (Heliconius). Heredity 34: 265–269.

Communicating editor: D. Rand


convergent evolution in the genetic basis of mu¨llerian ... · from a genomic bac library and...

Documents