localization of m¼llerian mimicry genes on a dense - genetics
TRANSCRIPT
Copyright � 2006 by the Genetics Society of AmericaDOI: 10.1534/genetics.106.057166
Localization of Mullerian Mimicry Genes on a Dense Linkage Map ofHeliconius erato
Durrell D. Kapan,*,1 Nicola S. Flanagan,*,2 Alex Tobler,*,3 Riccardo Papa,* Robert D. Reed,†
Jenny Acevedo Gonzalez,* Manuel Ramirez Restrepo,* Lournet Martinez,*Karla Maldonado,* Clare Ritschoff,‡ David G. Heckel§ and
W. Owen McMillan*
*Department of Biology, University of Puerto Rico, San Juan, PR 00931, †Department of Biology, Duke University, Durham, NC 27708-0338,‡Department of Biology, Cornell University, Ithaca, New York 14853 and §Max Planck Institute of Chemical Ecology, D-07745 Jena, Germany
Manuscript received September 9, 2005Accepted for publication February 15, 2006
ABSTRACT
We report a dense genetic linkage map of Heliconius erato, a neotropical butterfly that has undergone aremarkable adaptive radiation in warningly colored mimetic wing patterns. Our study exploited naturalvariation segregating in a cross between H. erato etylus and H. himera to localize wing color pattern loci on adense linkage map containing amplified fragment length polymorphisms (AFLP), microsatellites, andsingle-copy nuclear loci. We unambiguously identified all 20 autosomal linkage groups and the sexchromosome (Z). The map spanned a total of 1430 Haldane cM and linkage groups varied in size from26.3 to 97.8 cM. The average distance between markers was 5.1 cM. Within this framework, we localizedtwo major color pattern loci to narrow regions of the genome. The first gene, D, responsible for red/orange elements, had a most likely placement in a 6.7-cM region flanked by two AFLP markers on the endof a large 87.5-cM linkage group. The second locus, Sd, affects the melanic pattern on the forewing andwas found within a 6.3-cM interval between flanking AFLP loci. This study complements recent linkageanalysis of H. erato’s comimic, H. melpomene, and forms the basis for marker-assisted physical mapping andfor studies into the comparative genetic architecture of wing-pattern mimicry in Heliconius.
RECENT advances in molecular biology allow re-searchers to directly investigate the genetic basis
of adaptive variation in natural populations of widelydivergent organisms, ranging from Mimulus and crick-ets to sticklebacks and pocket mice (Bradshaw et al.1998; Peichel et al. 2001; Parsons and Shaw 2002;Albertson et al. 2003; Hoekstra et al. 2004; Nachman
et al. 2004; Colosimo et al. 2005). This research hasallowed novel insights into the mechanisms governingmorphological diversification by coupling traditionalstudies of ecological genetics and natural selection withmodern genomic approaches.
In this article, we localize two major color pattern lociunderlying adaptive variation in the wing patterns ofHeliconius erato to narrow regions of a high-resolutiongenetic map. This work replaces an initial linkage mapof H. erato (Tobler et al. 2005), and complements arecently published high-resolution linkage map of itscomimic H. melpomene (Jiggins et al. 2005). The nearly
identical wing color patterns of H. erato and H. melpomenenot only are adaptations that warn potential predators ofeach species’ unpalatability (Chai 1986; Langham 2005)and a textbook example of cooperative or Mullerianmimicry (Muller 1879; Nijhout 1991), but also play animportant role in speciation (McMillan et al. 1997;Jiggins et al. 2001). Although these species are fromdivergent clades within the genus and do not hybridize,they share nearly identical color pattern phenotypeswhere they co-occur and have undergone a paralleladaptive radiation into nearly 30 different geographicraces (Brown et al. 1974; Turner 1974, 1983; Sheppard
et al., 1985; Brower 1994, 1996; Turner and Mallet
1997; Beltran et al. 2002).Within each species interracial crossing experiments
have described up to 20 different loci responsible for thiswing color pattern diversification (Sheppard et al. 1985;Mallet 1989; Nijhout et al. 1990, 1991; Jiggins andMcMillan 1997; Gilbert 2003; Naisbit et al. 2003).However, one of the most striking aspects of both radi-ations is that the majority of phenotypic change is drivenby a small number of loci, or groups of very tightly linkedloci, of large effect (Sheppard et al. 1985; Mallet 1989;Nijhout et al. 1990; Nijhout 1991; Jiggins andMcMillan 1997; Gilbert 2003; Naisbit et al. 2003).These major genes cause discrete phenotypic shiftsacross large areas of the wing surface by changing the
1Corresponding author: Center for Conservation and Research Training,Pacific Biosciences Research Center, 3050 Maile Way, Gilmore 406,University of Hawaii, Manoa, Honolulu, HI 96822.E-mail: [email protected]
2Present address: School of Botany and Zoology, The Australian NationalUniversity, Canberra, ACT 0200 Australia.
3Present address: Department of Biology, Duke University, Durham,North Carolina 27708-0338
Genetics 173: 735–757 ( June 2006)
position, size, and shape of red/orange and melanicpatches on both the dorsal and ventral surfaces of thefore- and hindwings (Sheppard et al. 1985; Nijhout
1991). In both radiations, the genes underlying differ-ent pattern elements have a predictable epistatic effectdepending on scale and pigment type (Gilbert et al.1988; Gilbert 2003). These facts make it relatively easyto follow the segregation of specific alleles at a locusin genetic backgrounds chosen to minimize epistasis(Tobler et al. 2005).
Characterization of genomic regions directly respon-sible for adaptive differences in wing color patterns inboth these species will allow us to answer questionsabout the proximate and ultimate origins of this fasci-nating variation. As an important step toward this goal,we report the localization of two major H. erato colorpattern switch genes within a high-resolution linkagemap: the ‘‘Dennis’’ (or D) locus responsible for red andorange pattern elements and the ‘‘shortened’’ (Sd) locusthat affects forewing melanin patterns (Sheppard et al.1985). In contrast to our preliminary study (Tobler
et al. 2005), which utilized fewer markers and individualsin a backcross design, we trace inheritance patterns ofnumerous molecular markers and wing phenotypes in alarge outbred F2 cross (i.e., with two pairs of unrelatedoutbred grandparents). For this study, we crossed anorange and yellow race from southeast Ecuador H. eratoetylus with H. himera, a closely related sister species fromsouth Ecuador that displays yellow and bright red (seeFigure 1; Tobleret al. 2005). In H. erato races the D locuscontrols the presence of the orange and red elements,including (1) the orange patch at the base of theforewing in H. erato etylus known as ‘‘Dennis’’ [after anindividual H. melpomene with the mimetic phenotypenamed ‘‘Dennis the Menace’’ (Beebe 1955); see alsoSheppard et al. 1985, p. 457; J. Mallet, personal commu-nication], (2) the orange rays present on the hindwingof many Amazonian races (Sheppard et al. 1985;Mallet 1989), (3) the yellow forewing band of severalAmazonian races (Sheppard et al. 1985), and (4) thered hindwing bar in H. himera (Figure 1; Jiggins andMcMillan 1997; Tobler et al. 2005). Genetic control ofthe first three pattern elements was initially hypothe-sized to be controlled by three separate, albeit tightlylinked loci (D, R, and y, Sheppard et al. 1985), butrecombinant phenotypes are virtually unknown fromthousands of field-caught individuals along hybridzones (Sheppard et al. 1985; Mallet 1989) and virtuallyabsent from crossing experiments, suggesting that thephenotypic variation may be explained by allelic varia-tion at a single locus or possibly three very tightly linkedloci (see Sheppard et al. 1985; Mallet 1989; Jiggins
and McMillan 1997). In this article, in accordance withnomenclature established by Jiggins and McMillan
(1997), we utilize Det to describe the H. erato etylus allele(with the ‘‘DRy’’ pattern) of the D locus and demon-strate through Mendelian cosegregation and mapping
that the presence or absence of the red hindwing bar inH. himera is in fact due to D locus variation (Dhi) asinitially hypothesized by Jiggins and McMillan (1997)and subsequently mapped to one end of chromosomeHEC 3 (Tobler et al. 2005).
The Sd locus is one of several genes affecting mel-anization in the forewing (Sheppard et al. 1985) and wasoriginally described from a single cross between ahybrid east Ecuador H. erato (lativitta 3 notabilis) andan east Brazilian race of H. erato where Sd accounted forshortening of the forewing yellow band (Sheppard et al.,1985). In our cross, however, two complementary Sdpatterns were found: In H. erato etylus, individuals had ayellow patch present toward the distal tip of the fore-wing and absent more proximally (genotype Sdet Sdet,Figure 1a, individuals PR222 and PR217), and, con-versely, H. himera has a proximal yellow patch (genotypeSdhi Sdhi, Figure 1a, individuals PR246 and PR181).
To construct a linkage map we mapped molecularmarkers including amplified fragment length polymor-phisms (AFLP), microsatellites, and single copy nuclearloci (SCNL) in F2 offspring utilizing a three-step methodbased on Jiggins et al. (2005) and outlined in supplemen-tal Figure S1 (http://www.genetics.org/supplemental/).This method takes advantage of natural variation withinour source populations and achiasmatic oogenesis, whichleads to a complete absence of crossing over in femaleLepidoptera (Suomalainen et al. 1971, 1973; Turner
and Smiley 1975) to generate a single high-resolutiongenetic map for the cross. This map provides the basis forlocalizing color pattern loci in specific genomic regions,generating an integrated reference map of all the majorcolor pattern genes responsible for the H. erato diver-sification, and is an essential step toward our goal ofunderstanding the genetic architecture of convergentevolution between H. erato and its comimic, H. melpomene.
MATERIALS AND METHODS
Crossing design: To identify major wing pattern loci in theAmazonian rayed race under study, we collected individuals ofH. erato etylus from the Zamora River (3�559 S, 78�509 W) andindividuals of H. himera from the town of Vilcabamba (4�169 S,79�139 W), both in Loja Province in southern Ecuador, withpermits from the Ecuadorian government (Tobler et al.2005). Outbred stocks of both H. erato etylus and H. himerawere established in our insectaries at the University of PuertoRico. From these stocks we created an outbred F2 cross,designated ETF2-2, by mating two unrelated F1 hybridsderived from unrelated H. erato etylus female and H. himeramale grandparents (Figure 1, A and B). Butterflies were caredfor as outlined in McMillan et al. (1997). The 88 F2 offspringin the brood were raised until eclosion and killed by brieflyexposing them to �80�. The bodies were frozen at �80� untilneeded and the wings were digitally photographed and pre-served in glassine envelopes for later analysis.
AFLPs: Genomic DNA was extracted as outlined in Tobler
et al. (2005). AFLP analysis followed protocols described byVos et al. (1995) with modifications for fluorescent-labeledAFLP reactions run on an ABI PRISM 377 DNA sequencer
736 D. D. Kapan et al.
(Applied Biosystems) as outlined by Tobler et al. (2005). Wesurveyed 20 EcoRI/MseI primer combinations for fragmentsbetween �100 and 500 bp in length (Table 1, supplementalAppendix at http://www.genetics.org/supplemental/). Size-corrected data for all individuals and control lanes werecombined into a metagel for each primer combination usinga modified version of Genographer (Benham et al. 1999), anopen source java program for AFLP analysis (http://hordeum.oscs.montana.edu/genographer/). The modified version (2.1beta) is currently available for download (http://zephyr.hpcf.upr.edu/�mcmi-lab/). AFLP bands were scored as present if adistinct band appeared on the metagel (either in normalizedor in unnormalized view within Genographer) and accompa-nying peaked fluorescent signal was seen using the thumbnailview in Genographer and absent otherwise (weakly amplifyinglanes with shallow, indistinct peaks were scored as missingdata). All AFLP bands were scored as dominant alleles (i.e., noattempt was made to infer heterozygosity from reduced bandintensity). Fluorescent bands were named according to primercombination and band size in base pairs (see supplementalAppendix, http://www.genetics.org/supplemental/). Overall,�70,760 (610 AFLP loci 3 116 lanes ¼ 16 grandparental rep-licates, 12 parental replicates, and 88 offspring) binary geno-types were individually examined by one or more investigatorsbefore export from Genographer as a matrix of zeroes andones.
Codominant anchor loci: We also scored microsatellite andSCNL to anchor our maps relative to other mapping work inHeliconius. We surveyed variation at 19 microsatellites (Table2) and 21 SCNL (Table 3). Microsatellites were amplified andscored as described in Flanagan et al. (2002) and Tobleret al.(2005) and Table 2. SCNL for a variety of candidate andhousekeeping genes were utilized and segregating genotypeswere determined using allelic size variation or by variationat single nucleotide positions (Table 3) ( Jiggins et al. 2005;Tobler et al. 2005). Single nucleotide polymorphisms (SNPs)were scored by restriction fragment length polymorphism( Jiggins et al. 2005; Tobler et al. 2005) generated by restric-tion enzymes (under the conditions recommended by the
Figure 1.—Cross design and cosegregating wing pheno-types all shown with right dorsal (left) and left ventral (right)fore- and hindwings. (A) The outbred F2 cross shown with ma-ternal grandmother PR222 (H. erato etylus) 3 maternal grand-father PR246 (H. himera); paternal grandmother PR217(H. erato etylus) 3 paternal grandfather PR181 (H. himera).PR246 displays a cage code used to identify individuals onthe ventral side. (B) The F1 female PR269 (daughter ofPR222 and PR246) was mated to F1 male PR240 (the sonof PR217 and PR181). (C) Codominant phenotypic effectsof two major loci segregating in F2 offspring can be seen withcolumns representing the three Sd genotypes: SdetSdet, SdetSdhi,and SdhiSdhi, respectively. The rows represent the three D ge-notypes: DetDet, DetDhi, and DhiDhi. Thus the top left butterfly andthe bottom right butterfly resemble their respective grandpar-ents, H. erato etylus and H. himera, respectively. (D) Small spotphenotypes. PR269 dorsal and ventral showing inheritanceof the yellow discal cell spot (c-spot) from female H. erato etylusPR222, and PR240 dorsal and ventral showing the yellowR2-R1 spot (r-spot) inherited from female PR217. Both H. er-ato etylus-derived spot genotypes are visible only when presentin heterozygous form with the H. himera Sd allele; see, for in-stance, the middle individual PR921 in C showing the c-spotphenotype and the bottom row individual PR976 in C show-ing the r-spot (wing vein drawing modified from Emsley
1963).
Dense Linkage Map of Heliconius erato 737
vendors) or using the MegaBACE SNuPe Genotyping Kit(Amersham Biosciences). In the SNuPe technique, PCRprimers are used to amplify a fragment containing a previouslydetected SNP, followed by a single-base extension from anunlabeled primer adjacent to the SNP site. Each dNTP in theextension mix is labeled with a different dye allowing de-tection of heterozygotes and homozygotes by electrophoresison the MegaBACE system. A total of 5 ml of the PCR reactionwas incubated with 6 units of exonuclease and 0.45 units ofshrimp alkaline phosphatase in a final volume of 8.0 ml. Themix was heated to 37� for 30 min then denatured at 85� for10 min. Approximately 6 ng of PCR product was combinedwith 2 pmol of the SNP primer and 0.5 ml (one-eighth of thatrecommended by the vendor) of the SNuPe premix reagent ina final volume of 10 ml. The reaction was cycled 25 timesthrough the following temperature profile: 96� for 10 sec, 50�for 5 sec, and 60� for 10 sec. SNuPe reactions were cleanedusing AutoSeq96 plates (Amersham Biosciences). After wash-ing the columns twice with 100 ml of double-distilled water,samples were added to the top of the Sephadex and centri-fuged for 5 min at 910 3 g. Five microliters of the cleanedSNuPe reaction were added to a skirted 96-well plate andcombined with 5 ml of a master mix containing 498.75 ml ofMegaBACE loading solution and 1.25 ml of MegaBACE SNuPemultiple injection marker (Amersham Biosciences). Sampleswere injected into the MegaBACE using the protocols outlinedin the user manual.
Linkage analysis and map construction: Linkage analysisfollowed methods for Lepidoptera published by Shi et al.(1995), Yasukochi (1998), Tobler et al. (2005), and Jiggins
et al. (2005). By separately treating the three different geno-types corresponding to the dominant AFLP bands segregating
in this cross, a single map was constructed in a three-step pro-cess (Jiggins et al. 2005; see Figure S1 at http://www.genetics.org/supplemental/). Bands from female-informative (FI)AFLP markers are present in the F1 female, absent from the F1
male, and segregate 1:1 (e.g., band positive vs. band absent) inthe F2 progeny. Bands from male-informative (MI) markers arepresent in the F1 male, absent from the F1 female, and alsosegregate 1:1 in the F2 progeny. Bands present in both F1 maleand female parents (BI markers) segregate in a 3:1 ratio in theF2 progeny. The FI and MI marker types are rare or absent inF2 crosses formed by crossing highly inbred strains, but areabundant in outbred F2 crosses such as ours, where the fourgrandparents are polymorphic.
In step one, FI markers are used to identify linkage groupssegregating in the female F1 gametes. Because of achiasmaticoogenesis in Lepidoptera, FI markers on the same chromo-some are inherited as a unit and this association is not brokenup by crossing over ( Jiggins et al. 2005). We identified 21linkage groups by grouping these FI markers at LOD 8.0 usingJoinMap 3.0 (Van Ooijen and Voorrips 2001) applied to abackcross design. By observing which FI alleles originatedfrom H. himera and which from H. erato etylus, we predictedthe species origin of the maternally derived homolog ofevery chromosome in each of the F2 offspring ½known as thechromosome print, Yasukochi (1998)� using PERL script‘‘CHROMPRINT’’ (http://zephyr.hpcf.upr.edu/�mcmi-lab/),which compared the genotype of each offspring and linkagephase of each FI locus within a linkage group. Each chromo-some print was hand verified and numbered (LG-#) to uniquelyassociate FI linkage groups with BI and MI markers (see belowand supplemental Appendix at http://www.genetics.org/supplemental/).
TABLE 1
Summary of AFLP primer combinations and total number of polymorphic bands utilized in our analysis
MseI extension
EcoRI extension Data CAA CAC CAG CCA CCT CTA CTG CTT Total
CA FI 14 10 9 10 2 45BI 18 7 13 14 8 60MI 12 9 9 9 2 41
CC FI 10 7 0 10 27BI 2 4 0 7 13MI 8 7 0 6 21
CG FI 15 13 10 7 45BI 12 6 10 8 36MI 18 14 9 6 47
CT FI 3 22 10 10 13 8 8 74BI 9 30 11 7 10 3 3 73MI 4 23 16 11 9 10 8 81
Total FI 42 45 36 10 10 13 10 25 191BI 41 43 38 14 7 10 11 18 182MI 42 46 41 9 11 9 12 20 190
All 125 134 115 33 28 32 33 63 563
These totals do not include bands generated by the second metabulk scan for tightly linked markers gener-ated by primer combinations Eco-CC MseI-CTA/CTT, Eco-CT MseI-CTA/CTT, and new combinations Eco-CC/MseICTC, Eco-CC/MseI-CAC run utilizing Licor methodology (Tobler et al. 2005). An additional eight male in-formative bands were scored. For details see map and supplemental Appendix (http://www.genetics.org/supplemental/) for bands CC-CTA_Sd_120-145, CC-CTA*mbA, CT-CTT*mbA, CC- CTT_Sd_255, CT-CTA_DRY_2524, CT-CTT_DRY_.204, CC-CTC_Sd_145-175, and CC-CAC-491. Blanks indicate the primer combinationwas not assayed.
738 D. D. Kapan et al.
TA
BL
E2
Su
mm
ary
of
mic
rosa
tell
ite
anch
or
loci
use
din
this
stu
dy
Mar
ker
nam
eP
rim
ern
ame
Pri
mer
seq
uen
ceR
eact
ion
con
dit
ion
sN
o.
of
alle
les
Map
pin
ggr
ou
pN
ote
sSo
urc
e(G
enB
ank
no
.)
Hel
-03
GA
-7F
59-C
CA
AT
TA
TG
TC
AC
AT
GG
AT
CT
GT
T-39
54�;
3.0
mm
2a
bF
la
na
ga
net
al.
(200
2)G
A-7
R59
-CT
CT
GT
CG
TC
CT
CT
GC
AG
TC
-39
MgC
l 2H
el-0
4G
A-8
F59
-GG
AA
CG
GA
GT
GC
CC
TA
AA
AC
-39
55�;
3.0
mm
3H
EE
7F
la
na
ga
net
al.
(200
2)G
A-8
R59
-CG
TT
GC
CG
CT
TA
TA
CT
TT
CC
-39
MgC
l 2H
el-0
5G
A-1
1F59
-TG
CT
GT
CC
AT
AC
CC
AA
CT
CA
-39
55�;
3.0
mm
NA
ac
Fla
na
ga
net
al.
(200
2)G
A-1
1R59
-CG
AA
CT
CA
CA
AC
CA
TC
AG
TC
A-39
MgC
l 2H
el-0
6G
A-1
6F59
-TA
GC
CT
TC
AC
TT
TG
AA
CC
CG
-39
55�;
3.0
mm
2H
EC
ZSe
x-li
nke
dF
la
na
ga
net
al.
(200
2)G
A-1
6R59
-CC
CA
CT
CG
AA
GC
AA
TG
AA
AT
-39
MgC
l 2H
el-0
7G
A-1
9F59
-GC
GG
GG
AC
AA
CT
AC
AT
AA
GC
-39
55�;
3.0
mm
2H
EE
13A
lter
nat
ive
alle
leo
fH
el-0
1(B
6),
sex-
lin
ked
inT
ob
ler
etal
.(2
005)
Fla
na
ga
net
al.
(200
2)
GA
-19R
59-C
CC
GA
AC
CA
CT
AA
AG
TC
GA
A-39
MgC
l 2H
el-0
1B
6F59
-AG
GG
CG
TC
GT
TA
GT
TT
GT
GT
-39
54�;
3.0
mm
2H
EE
13A
lter
nat
ive
alle
leo
fH
el-0
7(G
A19
),se
x-li
nke
din
To
bler
etal
.(2
005)
Fla
na
ga
net
al.
(200
2)
B6R
59-T
CG
TA
GA
TA
TC
CA
TT
AC
TC
TG
GT
CT
G-39
MgC
l 2H
el-1
0C
A-9
F59
-TC
TC
AC
TT
TC
CC
AC
AC
AG
CA
-39
55�;
3.0
mm
2H
EC
10F
la
na
ga
net
al.
(200
2)C
A-9
R59
-TG
TG
AA
GA
GA
CA
CA
TG
GG
GA
-39
MgC
l 2H
el-1
1C
A-1
0F59
-TT
TC
TT
TT
GA
GT
CC
CG
AT
GG
-39
55�;
3.0
mm
3H
EC
12A
lter
nat
ive
alle
leo
fH
el-0
8(A
2)F
la
na
ga
net
al.
(200
2)
CA
-10R
59-A
TC
TC
AG
AA
CT
GG
TC
GG
CA
G-39
MgC
l 2H
el-0
8A
2F59
-AC
AT
CT
CA
GA
AC
TG
GT
CG
GC
-39
58�;
1.5
mm
3H
EC
12A
lter
nat
ive
alle
leo
fH
el-1
1(C
A10
)F
la
na
ga
net
al.
(200
2)
A2R
59-C
TC
GA
TC
AG
CC
GG
TG
AT
TA
T-39
MgC
l 2H
el-0
2B
32F
59-T
CA
AA
AT
GT
TG
CA
GA
CC
GA
G-39
55�;
3.0
mm
2H
EE
7F
la
na
ga
net
al.
(200
2)B
32R
59-T
GC
AC
TT
CA
TT
GT
AA
GG
CG
T-39
MgC
l 2H
el-1
4A
AT
4F59
-GC
AC
AT
TT
AC
TT
AC
AC
TA
AC
GC
-39
53�;
1.5
mm
3H
EC
2A
lter
nat
ive
alle
leo
fH
el-1
3(C
A13
),T
ob
ler
etal
.20
05
Fla
na
ga
net
al.
(200
2)
AA
T4R
59-A
TT
TG
TT
TC
GA
AC
GA
CT
GC
C-39
MgC
l 2H
e-C
A-0
01
A25
F59
-CG
GG
AC
AC
AA
AA
TG
TG
AC
TG
-39
58�;
3.0
mm
3H
EE
9dT
ob
ler
etal
.(2
005)
A25
R59
-AA
CT
GG
AC
GG
TG
GA
CG
TT
AG
-39
MgC
l 2H
e-C
A-0
02
CJ-
MS1
8F59
-AC
GT
CT
GC
GG
TC
TG
GC
TG
TC
-39
56�;
1.5
mm
3H
EC
ZT
his
stu
dy
(DQ
3802
25)
CJ-
MS1
8R59
-TG
GT
AG
GG
AA
GT
AA
TG
AA
AC
-39
MgC
l 2;
BSA
0.8
mg/
ml
He-
CA
-003
CJ-
MS3
2F59
-AG
CA
AG
AT
TG
TC
GC
AT
GT
CA
-39
54�;
1.5
mm
3a
eT
his
stu
dy
(DQ
3802
26)
CJ-
MS3
2R59
-AC
GA
AA
TG
GA
AC
CC
AA
AA
CA
-39
MgC
l 2;
BSA
0.8
mg/
ml
He-
CA
-004
CJ-
MS3
7F59
-TT
GT
AG
TC
GG
AT
TT
AT
GG
CA
AA
-39
56�;
1.5
mm
3H
EE
11T
his
stu
dy
(DQ
3802
27)
CJ-
MS3
7R59
-GG
TC
CT
TA
GT
CA
TG
TC
GC
AA
A-39
MgC
l 2;
BSA
0.8
mg/
ml
He-
CA
-005
CJ-
MS4
0F59
-TG
TC
CT
AC
AT
GG
AT
AC
GG
AG
AA
-39
56�;
1.5
mm
3H
EC
12T
his
stu
dy
(DQ
3802
28)
(con
tin
ued
)
Dense Linkage Map of Heliconius erato 739
In step two, we compared each chromosome print with BIAFLPs as well as codominant anchor loci with segregationratios of 1:1:1:1 or 1:2:1. Applying JoinMap 3.0 to an intercrossdesign, we identified loci that grouped at LOD 5.0 or higherwith each chromosome print (in a few cases codominantmarkers grouped at LOD 4.0 and were included in groups forsubsequent verification). Because the genetic model em-ployed by JoinMap assumes crossing over in both males andfemales, which was not so in our case, new groups wereindividually examined to verify linkage with a unique chro-mosome print (see results), check phase, and to identifyforbidden recombinants (Shi et al. 1995), defined as offspringgenotypes impossible without crossing over in females unlessthere were scoring errors. In our analysis we accept a smallamount of scoring error and retained BI loci with five or fewerforbidden recombinants that have a probability of P, 0.007 orlower of being unlinked under the hypothesis of no scoringerror (see Jiggins et al. 2005). The BI AFLP loci associatedwith a particular chromosome print were labeled by linkagegroup (LG-#). We then extracted the MI component fromthe BI AFLP scores (i.e., following only AFLP bands inheritedfrom the father) by censoring the scores that includedAFLP bands inherited from the mother ( Jiggins et al.2005). For the sex chromosome, Z, male offspring are alwayspositive for BI AFLPs and the recombinant analysis is limitedto female offspring segregating MI markers on Z ( Jiggins et al.2005).
In the third and final step, the censored BI markers (taggedwith LG-#) were combined with MI AFLPs as well as co-dominant anchor markers recoded to show only the MI allele(see the BI / MI allele designation). Linkage groups wereassembled at LOD 3.0 or greater and identified by one or moreshared LG-# tagged AFLPs (Jiggins et al. 2005). All LODs tothis point were calculated utilizing Joinmap 3.0. For final mapconstruction, the most likely order and spacing of markerson a chromosome was determined utilizing Mapmaker 3.0applied to a backcross design (Lincoln et al. 1987). First, allmarkers within a linkage group were converted to the samelinkage phase and compared using Mapmaker’s ‘‘join haplo-types’’ command. Haplotypes were consolidated over missingdata by grouping MI markers with identical genotypes (de-rived from either phase) or censored BI markers (in the sameoriginal linkage phase) as a haplotype in Mapmaker. CensoredBI markers originally in the opposite linkage phase may formspurious haplotype groups since the censoring step createsmarker sets of different complementary individuals for eachphase. Haplotypes including these latter markers were re-moved unless both joined to an MI locus of either linkagephase. Linkage group memberships of these haplotypes andthe remaining markers were reverified using Mapmaker’s‘‘assign’’ command.
To find the most likely order for linkage groups with eightor fewer unique loci (single markers and haplotypes), we usedMapmaker’s ‘‘compare’’ command to examine all possibleorders. For larger linkage groups we automated the construc-tion of a preliminary scaffold of five or fewer markers utilizingthe ‘‘order’’ command. Following standard Mapmaker meth-ods (Lincoln et al. 1987) additional markers were insertedinto the scaffold in order of their impact on the LOD score(utilizing cutoffs in descending order of LOD 3.0, LOD 2.0,LOD 1.0, and ,LOD 1.0) until the likelihood of the resultingmap was maximized. Although these LOD values are notnormally saved during the map-building process, we utilizedMapmaker scripts to keep track of markers mapping at eachLOD value (see Figure 2). In all cases we preferentiallyinitialized our scaffold with MI AFLP loci, which had thehighest number of scored individuals and the highest impacton the likelihood. All markers mapping to a unique interval
TA
BL
E2
(Co
nti
nu
ed)
Mar
ker
nam
eP
rim
ern
ame
Pri
mer
seq
uen
ceR
eact
ion
con
dit
ion
sN
o.
of
alle
les
Map
pin
ggr
ou
pN
ote
sSo
urc
e(G
enB
ank
no
.)
CJ-
MS4
0R59
-CC
AG
CA
TC
TT
TG
CC
CC
TT
A-39
MgC
l 2;
BSA
0.8
mg/
ml
Hm
_06
CJ-
MS4
4F59
-AA
AT
AG
TG
TG
CG
GC
GG
AA
TA
-39
54�;
1.5
mm
2H
EC
5fJi
gg
in
set
al.
(200
5)C
J-M
S44R
59-T
GG
AG
TA
GA
AA
TG
CG
GG
TT
TA
-39
MgC
l 2;
BSA
0.8
mg/
ml
He-
CA
-00
7C
J-M
S47F
59-G
CT
AC
GA
GC
CT
CA
CC
TG
AA
C-39
51�;
1.5
mm
4a
gT
his
stu
dy
(DQ
3802
29)
CJ-
MS4
7R59
-GA
CT
AA
CG
GC
AC
CT
CC
AC
AC
-39
MgC
l 2;
BSA
0.8
mg/
ml
He-
CA
-00
8C
J-M
S60F
59-C
GC
CA
CG
AC
GG
AT
AA
AA
AT
-39
56�;
1.5
mm
2H
EC
12T
his
stu
dy
(DQ
3802
30)
CJ-
MS6
0R59
-TT
AA
AC
CT
CA
AC
CG
CA
CT
TT
-39
MgC
l 2;
BSA
0.8
mg/
ml
NA
,n
ot
avai
lab
le.
aT
he
locu
sd
idn
ot
map
.Se
em
at
er
ia
ls
an
dm
et
ho
ds
for
det
ails
.bL
oo
sely
asso
ciat
edw
ith
LG
19b
ut
too
man
yo
bli
gate
reco
mb
inan
tsto
map
.cU
nab
leto
con
sist
entl
ysc
ore
.d
MI
gro
up
edo
nly
.eT
his
BI
locu
sw
asn
ot
con
vert
edto
MI
bec
ause
ther
ew
ere
too
man
yo
bli
gate
do
ub
lere
com
bin
ants
inth
eB
Ian
alys
is.
fF
Igr
ou
ped
on
ly.
gT
his
locu
so
nly
had
MI
alle
les
and
did
no
tgr
ou
pat
this
step
.
740 D. D. Kapan et al.
TA
BL
E3
Su
mm
ary
of
sin
gle
cop
yn
ucl
ear
loci
use
din
this
stu
dy
Nam
eP
rim
ern
ame
Pri
mer
seq
uen
ceIn
tro
na
PC
Rre
acti
on
con
dit
ion
sb
No
.o
fse
greg
atin
gal
lele
sSc
ori
ng
met
ho
dc
Map
ped
inH
.er
ato
Map
ped
inH
.m
elpo
men
eSo
urc
e
Apt
erou
s(a
p)A
p-F
(35)
59-T
GA
AT
CC
TG
AA
TA
CC
TG
GA
GA
-39
XIn
con
sist
ent
amp
lifi
cati
on
NA
NA
HM
EL
(Z)
Jig
gin
set
al.
(200
5)A
p-R
(357
)59
-CT
TT
TC
CG
CC
AT
TT
TG
TT
CT
C-39
Cu
bitu
sin
terr
upt
us
(ci)
Ci-F
(85)
59-A
AC
AC
AG
AT
CA
CA
TA
CA
CG
CA
-39
X53
�;2.
5m
mM
gCl;
0.8
mg/
ml
BSA
3SN
uP
EH
EC
3H
ME
L18
To
bler
etal
.(2
005)
Ci-R
(416
)59
-TG
TA
TC
TT
TT
AG
TG
CA
AC
CC
G-39
Dop
a-de
carb
oxyl
ase
(ddc
)D
dc-
F(5
0)59
-CA
GA
GG
GT
CA
AG
GA
AC
AG
CA
C-39
X60
�;2.
5m
mM
gCl;
0.8
mg/
ml
BSA
NA
NA
No
segr
egat
ing
vari
atio
nH
ME
L1
To
bler
etal
.(2
005)
Dd
c-R
(402
)59
-TC
AT
GA
GG
TA
GC
GG
TA
CT
CG
G-39
Dec
apen
tapl
egic
(dpp
)D
pp
-F(3
4)59
-AG
AG
AA
CG
TG
GC
GA
GA
CA
CT
G-39
52�;
2.5
mm
MgC
l;n
oB
SA2
SNu
PE
HE
C14
HM
EL
19T
ob
ler
etal
.(2
005)
Dp
p-R
(327
)59
-GA
GG
AA
AG
TT
GC
GT
AG
GA
AC
G-39
Win
gles
s(w
g)W
g-F
59-C
CC
AG
TT
TT
AG
AT
CT
GT
CG
-39
50�;
2.5
mm
MgC
l;n
oB
SA4
SNu
PE
HE
C14
Br
ow
er
(199
6)W
g-R
59-T
CT
CG
GT
CC
GG
TA
TC
CG
CG
-39
Tri
ose-
phos
phat
eis
omer
ase
(tpi
)T
pi-F
59-G
GT
CA
CT
CT
GA
AA
GG
AG
AA
CC
AT
CT
T-39
X50
�;2.
5m
mM
gCl;
no
BSA
3SN
uP
EH
EC
ZH
ME
LZ
Belt
ra
net
al.
(200
2)T
pi-R
59-C
AC
AA
CA
TT
TG
CC
CA
GT
TG
TT
GC
CA
A-39
Man
nos
eph
osph
ate
isom
eras
e(M
pi)
Mp
i-F59
-TT
TA
AG
GT
GC
TC
TA
TA
TA
AG
RA
AR
GC
-39
X50
�;2.
5m
mM
gCl;
no
BSA
3R
FL
P/
Alu
I(N
EB
)H
EC
5H
ME
L3
Belt
ra
net
al.
(200
2)M
pi-R
59-T
TC
TG
GT
TT
GT
GA
TT
TG
GA
TC
YTT
RT
A-39
Pat
ched
(ptc
)P
tc-F
(7)
59-C
TC
CG
AA
GA
AG
GT
CT
GC
CG
CA
AG
-39
X53
�;2.
5m
mM
gCl;
no
BSA
4SN
uP
EH
EE
6H
ME
L10
Jig
gin
set
al.
(200
5)P
tc-R
(364
)59
-AA
TT
CG
TG
CT
CG
TC
GT
AT
TT
TC
-39
kyu
ren
e3-h
ydro
xyla
se(c
inn
ibar
)
Cin
n-F
(23-
bo
b)
59-A
GG
AT
TG
TT
CG
TG
GA
CC
G-39
X50
�;2.
5m
mM
gCl;
no
BSA
4SN
uP
EH
EE
15T
his
stu
dy
(DQ
3802
23)
Cin
n-R
(376
-bo
b)
59-G
CA
TC
TC
TC
CA
TC
TT
TT
GC
C-39
Lon
g-w
avel
engt
hop
sin
(ops
in)
Op
sin
-59R
AC
ER
H1
59-T
GG
TT
AC
AA
TA
GG
GC
TG
A-39
Inco
nsi
sten
tam
pli
fica
tio
nN
AN
AH
ME
L11
Jig
gin
set
al.
(200
5)O
psi
n-R
LW
FD
459
-AC
AA
AA
AT
CA
AT
GG
GA
GA
GT
A-39
Elo
nga
tion
Fact
or1a
(EF1
a)E
F1H
-F59
-GA
GA
AG
GA
AG
CC
CA
GG
AA
AT
-39
52�;
2.5
mm
MgC
l;n
oB
SA3
SNu
PE
HE
E6
HM
EL
10Ji
gg
in
set
al.
(200
5)E
F1H
-R59
-CC
TT
GA
CR
GA
CA
CG
TT
CT
TT
-39
Hed
geho
g(h
h)H
h-F
(38)
59-A
AG
GA
AA
AA
CT
GA
AT
AC
GC
TG
GC
-39
52�;
2.5
mm
MgC
l;n
oB
SA3
RF
LP
/Sa
u96
Ian
dSa
u3A
Id
HE
C10
Th
isst
ud
y(D
Q38
0223
)H
h-R
(206
)59
-CG
AG
AC
GC
GC
CA
AC
TT
TC
C-39
Rib
osom
alpr
otei
nL
5(R
pL5)
Rp
L5-
F(4
4)59
-TC
CG
AC
TT
TC
AA
AC
AA
GG
AT
G-39
X52
�;2.
5m
mM
gCl;
no
BSA
4SN
uP
EH
EE
7H
ME
L11
Jig
gin
set
al.
(200
5)R
pL
5-R
(499
)59
-TG
AA
GA
CC
GA
AG
AT
GT
GT
GC
-39
Rib
osom
alpr
otei
nS5
(RpS
5)
Rp
S5-F
(31)
59-G
GT
TG
AG
GA
AA
AC
TG
GA
AC
G-39
52�;
2.5
mm
MgC
l;n
oB
SA3
SNu
PE
HE
E7
HM
EL
11Ji
gg
in
set
al.
(200
5)R
pS5
-R(4
90)
59-T
AC
GG
CT
TG
AC
GA
CG
TA
CT
G-39
(con
tin
ued
)
Dense Linkage Map of Heliconius erato 741
TA
BL
E3
(Co
nti
nu
ed)
Nam
eP
rim
ern
ame
Pri
mer
seq
uen
ceIn
tro
na
PC
Rre
acti
on
con
dit
ion
sb
No
.o
fse
greg
atin
gal
lele
sSc
ori
ng
met
ho
dc
Map
ped
inH
.er
ato
Map
ped
inH
.m
elpo
men
eSo
urc
e
Rib
osom
alpr
otei
nS8
(RpS
8)
Rp
S8-F
(56)
59-G
CC
CA
TT
CG
TA
AG
AA
GA
GG
AA
-39
X52
�;2.
5m
mM
gCl;
no
BSA
3R
FL
P/
Dpn
II(N
EB
)H
EE
7H
ME
L11
Jig
gin
set
al.
(200
5)R
pS8
-R(5
71)
59-C
AA
GG
AC
AT
AG
CC
AT
CA
GC
A-39
Rib
osom
alpr
otei
nS9
(RpS
9)
Rp
S9-F
(12)
59-G
CC
AT
CA
TG
GT
GA
AC
AA
CA
G-39
X52
�;2.
5m
mM
gCl;
no
BSA
2Si
zep
oly
mo
rph
ism
HE
E8
HM
EL
14Ji
gg
in
set
al.
(200
5)R
pS9
-R(4
93)
59-T
CA
AT
GT
GC
TT
GC
CA
GA
GT
C-39
Rib
osom
alpr
otei
nL
3(R
pL3)
Rp
L3-
F(5
9)59
-CC
GT
CA
TC
GT
GG
TA
AA
GT
GA
-39
X52
�;2.
5m
mM
gCl;
no
BSA
2SN
uP
EH
EC
4H
ME
L1
Jig
gin
set
al.
(200
5)R
pL
3-R
(537
)59
-TC
TC
CA
TG
AT
AT
GG
GC
CT
TC
-39
Rib
osom
alpr
otei
nL
10a
(RpL
10a)
Rp
L10
a-F
(192
)59
-GT
AC
AT
TC
CC
AG
GC
CA
AA
AA
-39
X52
�;2.
5m
mM
gCl;
no
BSA
3Si
zep
oly
mo
rph
ism
eH
EE
7H
ME
L11
Jig
gin
set
al.
(200
5)R
pL
10a-
R(5
74)
59-T
TC
TG
TG
CC
AA
CT
CA
TC
AG
G-39
Rib
osom
alpr
otei
nL
11
(RpL
11)
Rp
L11
-F(1
38)
59-C
TG
TG
TC
GG
TG
AA
TC
TG
GT
G-39
Inco
nsi
sten
tam
pli
fica
tio
nN
AN
AH
ME
L5
Jig
gin
set
al.
(200
5)R
pL
11-R
(571
)59
-CT
GT
TG
GA
AC
CA
CT
TC
AT
GG
-39
Rib
osom
alpr
otei
nL
19
(RpL
19)
Rp
L19
-F(1
31)
59-C
CC
GA
CA
GA
AC
AT
CC
GT
AA
G-39
Inco
nsi
sten
tam
pli
fica
tio
nN
AN
AH
ME
L10
Jig
gin
set
al.
(200
5)R
pL
19-R
(529
)59
-CG
GG
CT
TC
CT
TC
AC
TC
TG
T-39
Rib
osom
alpr
otei
nP
0(R
pP0)
Rp
P0-
F(1
42)
59-A
CC
CA
AA
AT
GT
TT
CA
TC
GT
G-39
X52
�;2.
5m
mM
gCl;
no
BSA
3Si
zep
oly
mo
rph
ism
HE
E7
HM
EL
11Ji
gg
in
set
al.
(200
5)R
pP
0-R
(565
)59
-TC
AC
CA
GG
CT
TC
AA
GA
TG
TG
-39
NA
,u
nab
leto
sco
reth
elo
cus
inth
em
app
ing
fam
ily.
aA
nX
ind
icat
esan
intr
on
was
pre
sen
t.bA
nn
eali
ng
tem
per
atu
re,
[MgC
l],
[BSA
].cIn
clu
din
gR
FL
Pen
zym
e.d
Fir
std
iges
tio
nw
ith
Sau
96I
allo
wed
dis
crim
inat
ion
of
two
of
fou
rp
oss
ible
off
spri
ng
gen
oty
pes
.T
od
isti
ngu
ish
ind
ivid
ual
sfo
rw
hic
hth
efi
rst
dig
esti
on
was
no
tin
form
ativ
e,th
eo
ther
two
gen
oty
pes
wer
ese
par
ated
by
ase
con
dd
iges
tio
nw
ith
Sau
3AI.
eSc
ore
dd
irec
tly
fro
mP
CR
amp
lifi
cati
on
(fat
her
240
had
an
ull
alle
leth
atn
ever
amp
lifi
ed)
lead
ing
toth
ree
dis
tin
guis
hab
lege
no
typ
es.
742 D. D. Kapan et al.
were considered part of the map’s framework. Censored BIloci have half or fewer genotypes than MI markers andoccasionally could not be mapped to a unique frameworkposition relative to one or more flanking markers. Thesemarkers were placed on the map using the ‘‘place’’ commandin Mapmaker in their most likely position to represent this
information visually (Figure 2). As a quality control measurewe also verified that the approximate method to handle largerlinkage groups, based on the order command, gave identicalresults to the compare command for the four autosomalgroups with eight or fewer loci as well as for all subsets of eightloci of larger linkage groups run for test purposes.
Figure 2.—Integrated genetic map for H. erato etylus and H. himera. Linkage groups with codominant anchor loci that allowbetween-study comparisons are named within a species by the race and linkage group they were first identified with, starting with Z(e.g., the sex chromosome in H. erato cyrbia is HEC Z; Tobler et al. 2005). (A) Groups homologous to H. erato cyrbia HEC 2, 3, 4, 5,10, 12, and 14. (B) In this study we recognize further linkage groups H. erato etylus (HEE 6, 7, 8, 9, 11, 13, and 15) in total iden-tifying 15 of 21 H. erato linkage groups. Tentative homology to H. melpomene is indicated by code HMEL and linkage group numberfrom Jiggins et al. (2005). (C) H. erato etylus groups a–f do not include any anchor loci and hence are not easily portable to otherstudies. A–C are drawn to different scales. On the map the L-# prefix signifies BI markers that segregated with a unique chromo-some print in the BI analysis and were subsequently mapped with the band-positive allele inherited from the father. AFLP markernames are read as follows: L prefix markers, such as LAAA374, indicate a BI censored marker LG1_Eco-CA_Mse-CAA_374 base pairswhere only boldface (non-italicized) letters are shown. Non-L prefix markers indicate MI markers and include the first letter of theEcoRI primer and all three bases of the MseI primer, such as: Eco-CT_Mse-CTT_245 for TCTT245. Codominant marker names in-clude segregation pattern in parentheses (FI, BI, BI / MI, MI, see supplemental Appendix at http://www.genetics.org/supplemental/for expanded marker names for all loci). Map position lines of variable thickness indicate framework markers mapping to uniqueintervals at .LOD 3.0 (thickest), .LOD 2.0 (medium), .LOD 1.0 (thin), ,LOD 1.0 (thinnest). Nonframework markers arefound to the right of framework markers. Markers joined with a 1 symbol form identical haplotypes with no recombination. Hal-dane cM values of 0.0–0.1 to the right of framework markers indicate the markers so joined are indistinguishable from frameworkmarkers at the 1% a priori error rate. Remaining markers to the right of framework markers are placed on the map but do notcontribute to its order, length, or likelihood (see materials and methods and results). Centimorgan values .0.1 indicate mostlikely placement below for one of these nonframework markers, no centimorgan values indicate marker could place on either sideof framework marker. Italics indicate placements with errors. A ‡ symbol indicates a nonframework marker that places in morethan one interval. A † symbol indicates a nonframework marker had $6 forbidden recombinants for a BI locus but may be part ofthe linkage group given other data. An * symbol indicates a nonframework marker with a segregation distortion of P , 0.02.Arrows (either [ or Y) indicate marker placed off respective end of chromosome at given distance. Microsatellites indicatedby a § symbol were mapped with alternative alleles of the same locus entering in different linkage phases (both shown, see Table2 and supplemental Appendix for details, http://www.genetics.org/supplemental/). Markers below each map are likely membersof the linkage group (on the basis of FI, BI, or LOD association data) but do not map in a 1:1 analysis. Dashed line at bottom ofHEC 10 represents the only segment whose maximum likelihood order was flipped with respect to remaining markers when theerror detection switch was turned off in Mapmaker. We also placed a small group of four ALFP loci at the bottom of HEE 7 thatdistantly associated with this group and were not used to calculate map length.
Dense Linkage Map of Heliconius erato 743
In summary, these methods allow integration of these threetypes of markers into a single unified map. Final linkagegroups were designated HEC # where homology between ourpreliminary mapping study of H. erato cyrbia and H. himera(Tobler et al. 2005) could be identified. Remaining linkagegroups containing one or more codominant anchor loci aregiven a number in the same series with an H. erato etylus (HEE#) designation to highlight the fact that although identical inthe present study, homology with previous HEC groups has notbeen ascertained. Finally, any remaining linkage groupswithout an anchor locus are given a letter, indicating thathomology for these linkage groups is not easily portable tofuture studies.
To explore the effect of any potential residual genotypingerrors we ran all of the final Mapmaker analyses both with andwithout error detection (Lincoln and Lander 1992). Finalgenetic distances are expressed in terms of uncorrectedHaldane centimorgans (cM) with error detection functionleft on, both recognizing unresolved genotyping errors andallowing a direct comparison with Jiggins et al. (2005).
Development of additional AFLP markers linked to colorpattern: We scanned for additional bands linked to colorpattern loci by running five primer combinations (Table 1legend) on the high-resolution LI-COR AFLP system withindividuals presorted by their color pattern genotype (seeTobler et al. 2005 for LI-COR methods). Specifically, weloaded groups of individuals with the same homozygousgenotype at a single color pattern locus on adjacent gel lanesallowing the quick visual identification of bands cis to one ofthe alternate alleles for a given color pattern. Perfectly linkedloci in these multilane metabulks will form a continuous band
corresponding to one or the other group of adjacent homo-zygous individuals. These instantly recognizable bands werereverified by repeated genotyping of the entire brood inoriginal plate order. The latter genotypes were used for map-ping (see Figure 2 and supplemental Appendix at http://www.genetics.org/supplemental/). This method differs frombulked segregant analysis (Michelmore et al. 1991) in that theindividuals in the genotypic groups are not added together orbulked into the same sample tube or well.
RESULTS
Phenotypic variation: Wing pattern variation acrossthe 88 individuals in this cross was easily understoodaccording to the classic methodology of Sheppard et al.(1985) and assuming allelic variation at both the D andSd locus (Figure 1). At the D locus, both parental F1’s(PR269 and PR240) had both the orange Dennis andrays characteristic of H. erato etylus with the rays broad-ened at the base to blend into the dorsal hindwing barof H. himera (Figure 1B, hindwings). At the Sd locus,parental F1’s had solid black dorsal forewings becausealternative alleles that encode for proximal Sdet vs. distalSdhi melanin together completely obscure the yellowbands (Figure 1B) also seen in heterozygous F2 off-spring (Figure 1C, middle column). The lack of strongepistasis and clear codominance of alleles at both D and
Figure 2.—Continued.
744 D. D. Kapan et al.
Sd was evident from the nine easily distinguishable phe-notypes among the F2 progeny (Figure 1C). The fore-wing Dennis and hindwing rays of H. erato etylus werealways inherited together in this and other F2 crossesof H. erato etylus and H. himera, consistent with a singlelocus hypothesis. Further, these pattern elements andthe hindwing bar of H. himera segregated in a clearMendelian fashion (21:46:21, G2 ¼ 0.18, P ¼ 0.913) andsuggested that the phenotypic effects are controlled byallelic variation at the same locus. Determining thegenotype of individuals at the Sd locus was similarlystraightforward. Although more distorted, the pheno-types segregated as expected for a single di-allelic co-dominant locus (16:54:18, G2 ¼ 4.70, P ¼ 0.095). Thesetwo codominant loci segregate independently (G8 ¼11.55, P ¼ 0.172).
In addition, individuals heterozygous at the Sd locusalso showed variation (presence/absence) of small yellowspots on the dorsal and ventral forewing (Figure 1).Some individuals possessed a small spot in the proximalwing cell (the discal cell spot, or c-spot, Figure 1D),whereas others showed a spot between wing veins R2and R1 sometimes extending to the SC vein (the r-spot
Figure 1D). The c-spot and r-spot never co-occur in thesame individual and segregate as alternate alleles ata single locus expressed in a 1:1 fashion (c-spot: present39:absent 49, G1 ¼ 1.14, P ¼ 0.29; r-spot: present45:absent 43, G1 ¼ 0.05, P ¼ 0.83). Further, the c-spothas MI phenotypic effects and shows a segregationpattern that is identical with Sd. These identical geno-types strongly suggest that the c-spot is controlled byan alternative allele of the H. erato etylus Sd locus. Ther-spot is also likely to be encoded by Sd (this FIphenotype grouped with the Sd locus by chromosomeprint), but in this cross we were unable to map r-spotsince it does not display MI phenotypic effects. InHeliconius crosses, alleles of loci coding for the pres-ence of melanic scales at a given wing position aregenerally dominant to alleles for yellow scales (Gilbert
et al. 1988). In this cross, individuals with all dark wingsreveal that, similar to the Sd locus, the two H. erato etylus-derived spot alleles have codominant effects (Figure 1).This interpretation leads to a hypothesized 1:1:1:1 ratiofor the phenotypes: no spots (all dark wings):r-spot:c-spot:no spots (all yellow H. himera forewing, G3 ¼ 1.18,P ¼ 0.76).
Figure 2.—Continued.
Dense Linkage Map of Heliconius erato 745
Markers: From the 20 EcoRI-CN/MseI-CNN (where Nrepresents variable nucleotides in the selective PCRamplification) AFLP primer combinations examined(Table 1), we scored 191 maternally inherited 1:1 loci(FI), 182 loci present in both the mother and father(BI), and 190 paternally inherited 1:1 loci (MI). Ofthese 564 loci, 29 of 382 1:1 loci (or �8%) had statis-tically significant segregation distortion (G test, P ,
0.05) while 24 of 182 (or �13%) of the BI loci showedsegregation distortion (G test, P , 0.05). In addition, wewere successful in scoring 18 microsatellite loci (out of19 attempted, Table 2) that produced as many as sevenalleles in the grandparents (range two to seven) andfour in the parents (Table 2). By following differentalleles inherited from either the mother or the father, orboth, we were able to score the same microsatellite lociwith different inheritance patterns (7 as 1:1 FI, 9 ascodominant loci with three genotypes, and 10 as 1:1 MIloci). In addition to codominant microsatellites, wesuccessfully genotyped 16 of the 21 SCNL screened,including 2 (Hh, Cinn) first reported here (Table 3).These SCNL had a range of 2–4 alleles in the parents,and depending on cross type, several could be scoredwith multiple segregation patterns (12 as 1:1 FI, 14 ascodominant, and 14 as 1:1 MI loci). Locus details,including primer sequence, the number of segregatingalleles, and comparison to previous mapping work inHeliconius, are given in Tables 1–3.
Sources of variation: To determine the proportion ofmarkers entering the cross from either the H. erato etylusor the H. himera side, we assessed whether a givendominant AFLP locus was absent, present as a poly-morphism, or likely arose as a fixed difference betweenH. erato etylus or H. himera grandparents by carefullyfollowing genotypes of the progenitors, linkage phasesof the loci, and segregation patterns of the offspring.For the 477 AFLP bands for which linkage phase of thegrandparents could be ascertained with certainty, 168were monomorphic negative (band always absent), 261were polymorphic, and 41 were probably fixed (bandpositive) in H erato etylus (due to the dominance of AFLPloci it is impossible to ascertain with certainty whether ornot a given locus was monomorphic in the grandparentsof either species). This contrasts with 244 monomor-phic negative, 130 polymorphic, and 96 probably fixed(band positive) loci for H. himera. A further 7 loci mayhave been fixed in either one but not both species. Thusmost of the variation derived from bands unique to H.erato etylus (�52%) vs. bands unique to H. himera (36%of the variation); however, in the latter a higher numberof these bands were likely monomorphic within the H.himera grandparents (20%) vs. the H. erato etylus grand-parents (9%).
The codominant anchor loci gave largely the samepicture and H. erato etylus was consistently more variablethan H. himera. For the 18 microsatellites and 1 SCNL(Ci) for which grandparental genotypes were known 17
were polymorphic in H. erato etylus. This contrasts withH. himera where 7 loci were monomorphic and 12 werepolymorphic. Although both species had a range of oneto four codominant alleles per locus, H. erato etylus hadan average of 2.5 alleles per locus, while H. himera had anaverage of 1.9 alleles per locus. Overall, 8 loci hadunique alleles in one genetic background, while 9shared one allele and 2 shared two alleles between eachspecies.
Chromosome prints and marker error estimation:We assessed the level of genotyping error using the 191FI polymorphic AFLP loci. All of these loci wereassigned to 1 of 21 linkage groups, matching the ex-pected number of chromosomes on the basis of earlycytogenetic work (Brown et al. 1992; Suomalainen et al.1971, 1973; Turner and Smiley 1975) and recentlyconfirmed for H. erato and H. himera (Tobleret al. 2005)and for H. melpomene ( Jiggins et al. 2005). Linkagegroups included as many as 15 and as few as 4 FI AFLPmarkers (Table 4). In addition to AFLP loci, we suc-cessfully grouped six of seven microsatellites, 11 of 12SCNL, and the one color pattern (r-spot) that could bescored in a 1:1 FI manner (see supplemental Appendixat http://www.genetics.org/supplemental/). Anchorloci He-CA-005 and EF1a did not group at this step be-cause they had many missing genotypes due to ampli-fication and scoring difficulties.
The nonrecombining female chromosomes permitthe unambiguous identification of forbidden recombi-nants providing an ideal standard against which to assessresidual genotyping error. Within the FI markers, thenumber of forbidden recombinants ranged from 0 to 21per locus (median 1, mean 3) and 0 to 16 per individual(median 6, mean 6.3) for a total of 550 of the 16,808genotypes (or 3%) showing a genotyping error from the88 offspring of the brood. However, unlike Jiggins et al.(2005), all FI AFLP loci without exception grouped atLOD 8.0; in addition, nearly 80% of the scored locishowed fewer than five forbidden recombinants, in-dicating that most loci were reliable.
Association of chromosome print with BI markers:Nearly all of the BI markers (182 AFLP, 9 microsatellites,and 14 SCNL) formed linkage at LOD 5.0 with eachother and a unique chromosome print and thus wereconsidered syntenic to the FI linkage groups. The LOD5.0 grouping left only 22 ungrouped BI AFLP mark-ers (9 were excluded prior to this step because of highsegregation distortion, x2 tests in Joinmap 3.0, P, 0.01).Of the 22 ungrouped loci, 7 grouped with a singlechromosome print at LOD 4.0 and 4 grouped with asingle chromosome print at LOD 3.0. In addition,one microsatellite, He-CA-005 (MS40), not linked inthe FI analysis, grouped with LG10 at LOD 4.0 (seesupplemental Appendix at http://www.genetics.org/supplemental/). In the absence of scoring errors, across88 offspring an unlinked locus would have�one-eighth,or 11, forbidden recombinants on average ( Jiggins et al.
746 D. D. Kapan et al.
TA
BL
E4
Su
mm
ary
of
mar
ker
dat
afo
rH.erato
etylus
lin
kag
egr
ou
ps
iden
tifi
edin
the
pre
sen
tst
ud
y
H.
erat
oli
nka
gegr
ou
p
Fin
aln
o.
of
AF
LP
mar
kers
SCN
LC
olo
rp
atte
rnM
icro
sate
llit
esF
inal
fram
ewo
rkm
arke
rn
o.
Size
(cM
)F
IB
IB
I/
MI
MIa
Z4
96
3T
pi(F
I,B
I,M
I)—
He-
CA
-00
3(M
S32-
BI)
,H
e-C
A-0
02
(MS1
8-M
I)9
114.
9
Z*b
——
15
——
Hel
-06
(GA
16-M
I)3
12.3
HE
C2
107
56
——
Hel
-14
(AA
T4-
FI,
BI,
MI)
1063
.0H
EC
39
1615
12ci
(FI,
BI,
MI)
Dlo
cus:
D(B
I),
D(B
I/M
I)—
2087
.5
HE
C4
146
513
wg
(FI,
BI,
MI)
,R
pL3
(MI)
——
1283
.4
HE
C5
133
28
Mpi
(FI,
BI,
MI)
—H
m0
6(M
S44-
FI)
841
.1H
EC
107
86
16hh
(FI,
BI,
MI)
—H
el-1
0(C
A9-
FI,
BI,
MI)
1686
.9H
EC
126
88
5—
—H
el-1
1(C
A10
-F
I,B
I,M
I),
He-
CA
-00
5(M
S40-
BI)
941
.0
HE
C14
127
710
dpp
(BI,
BI/
MI)
——
1192
.9H
EE
67
1210
11pt
c(F
I,B
I,M
I),
ef1
a(B
I)Sd
locu
s:r-
spo
t(F
I),
Sd(B
I),
c-sp
ot
(MI)
—17
94.6
HE
E7
54
310
RpL
5,
RpS
5,
RpL
10
a,R
pS8
,R
pPO
(FI,
BI,
MI)
—H
el-0
2(B
32-M
I),
Hel
-04
(GA
8-M
I)12
54.5
HE
E8
612
75
RpS
9(B
I)—
He-
CA
-00
8(M
S60-
BI,
BI/
MI)
1198
.7H
EE
913
86
15—
—H
e-C
A-0
01
(A25
-FI)
1384
.5H
EE
117
74
6—
—H
e-C
A-0
04
(MS3
7-M
I)9
85.0
HE
E13
410
105
——
Hel
-01
(B6-
BI,
MI)
,H
el-0
7(G
A19
-BI)
b
1157
.2
HE
E15
114
39
cin
n(F
I,B
I)—
—11
55.5
a9
99
16—
——
1672
.4b
65
510
——
—13
72.0
c11
67
5—
——
629
.5d
1512
128
——
—10
39.9
e13
53
6—
——
726
.3f
92
23
——
—4
37.2
To
tald
191
160
136
186
238
1430
.3
aM
app
edo
rp
lace
din
the
MI
dat
a.bT
wo
lin
kage
gro
up
sw
ith
Zm
arke
rsfo
rmin
the
MI
dat
a.cT
hes
etw
om
icro
sate
llit
em
arke
rso
nZ
infi
rst
H.
e.cy
rbia
map
(To
bler
etal
.20
05).
dT
his
excl
ud
es22
BI,
11co
nve
rted
BI
toM
Ian
d2
MI
mar
kers
no
tgr
ou
pin
gas
wel
las
16co
nve
rted
BI
toM
Ian
d2
MI
mar
kers
that
did
no
tm
ap(s
eesu
pp
lem
enta
lAp
pen
dix
for
det
ails
;h
ttp
://w
ww
.gen
etic
s.o
rg/
sup
ple
men
tal/
).
Dense Linkage Map of Heliconius erato 747
2005). As a conservative measure we retained loci with 5or fewer forbidden recombinants grouping at LOD 5.0with a unique chromosome print (see materials and
methods) and identified 16 loci with $6 forbiddenrecombinants. Unlike Jiggins et al. (2005) we did notcategorically eliminate these markers, but if they con-sistently associated with other LG-# markers from thesame group before and after censoring, they wereplaced on the framework map without affecting maporder, length, or likelihood (see markers with a daggersymbol denoting tentative membership, Figure 2). Weconverted the 167 remaining BI markers (163 BI AFLPmarkers, 2 microsatellites, and 2 SCNL) to 1:1 loci fol-lowing Jiggins et al. (2005). Subsequently, the predicted1:1 segregation ratio for the markers was tested.
Mapping of MI markers: Our final MI data setconsisted of 391 loci, including the 163 censored BIloci and 228 MI loci (198 AFLP markers, 14 micro-satellites, and 16 SCNL) segregating in a 1:1 MI fashion.Of these markers, 7% show some segregation distortionat a P , 0.05. Of the 391 loci 355 were associated with asingle linkage group (both D and Sd also associated atthis step). An additional 9 censored AFLP loci, althoughgrouping with other MI loci, were removed from themap because of inconsistent grouping between BI andMI scorings. A further 9 (5 censored AFLPs, 1 MI AFLP,and 1 microsatellite locus), although grouping withother MI loci, were removed from the map because of anexcess of recombinants (see supplemental Appendixat http://www.genetics.org/supplemental/). A final 18loci did not group at the final mapping step [3 MIAFLPs, 11 converted BI to MI AFLPs, as well as a mi-crosatellite anchor locus HE-CA-007 and a SCNL (EF1a)due to missing genotypes, amplification, and scoringdifficulties; see supplemental Appendix at http://www.genetics.org/supplemental/].
Despite some residual genotyping error, estimatedlinkage group size and map order was relatively stable.First, average linkage group length increased asymptot-ically 76, 83, and 96% with markers added to the map forLOD cutoffs 3.0, 2.0, and 1.0, respectively. To assess theeffect of genotyping error on marker order, we com-pared maps generated with the error detection algo-rithm in Mapmaker both enabled and disabled. As acompromise between the observed 3% overall error rate(per genotype) and low (0.5%) median error rate perlocus in the FI data we set the threshold for Mapmaker’serror detection function at 1%. Maps calculated withboth options were very similar; the largest difference wasthe inversion of a single group of four markers on HEC10 (Figure 2). Furthermore, the remaining differenceswere due to markers considered statistically indistin-guishable with error detection on (e.g., 0.0 or 0.1 cMdesignation in Figure 2 and supplemental Appendix,http://www.genetics.org/supplemental/) that mappeddirectly adjacent to markers that map to a uniqueposition with error detection off.
The final map contained a total of 238 (range 4–20)framework markers (i.e., those that map to a uniqueposition) spread across 21 linkage groups (Figure 2). Inaddition, 38 loci cosegregated completely and thusformed a haplotype with a framework marker, and 30markers were statistically indistinguishable from frame-work markers within the 1% error rate. We also placedan additional 49 nonframework markers on the frame-work map in their most likely position (Figure 2).Interestingly, although 21 chromosome prints includ-ing the sex chromosome (Z) were identifiable with theFI and the BI data, there were two separate linkagegroups associated with the sex chromosome Z in the MIanalysis (Figure 2, see below).
The overall size of the final map spanned 1430Haldane cM when calculated with the error detectionfunction in Mapmaker turned on. (With the errordetection switch turned off it was predictably larger, at1945 cM.) In general, the percentage of unique markersmapping at each LOD cutoff (�76–78%) remainedsteady, yet the number of centimorgans gained for eachnew unique marker decreased at each LOD cutoff asexpected for a nearly saturated map: 11.3 cM/markerfor the first 93 markers (.LOD 3.0), 6.3 cM/marker forthe next 20 (.LOD 2.0), 3.9 cM/marker for the next44 markers (.LOD 1.0), and 1.1 cM/marker for the last80 markers. This pattern suggests that additional map-ping effort on the same brood will not appreciablyincrease map resolution. Furthermore, given the re-sidual genotyping error with AFLPs and the other loci inthis study, increasing marker number would add littleinformation but it would expand map size.
The distribution of the number of markers perlinkage group and the intervals between markers acrosslinkage groups both for FI and MI loci fit a uniformdistribution suggesting that intermarker distances arerandomly spaced across the genome (Kolmogorov–Smirnov Test against a uniform distribution, KS ¼0.112, P ¼ 0.91 for MI mapping intervals). Nonetheless,there was a high variation in the size of linkage groupsindicated by both the number of markers in thechromosome prints and the map distance (Table 4).Autosomes had a minimum LG size of 26.3 cM andmaximum of 98.7 cM, and the average distance betweenmarkers was 5.1 cM (median 5.1). Relative to this mean,three linkage groups appeared to be outliers. Group fhad only 4 markers but had a high average of 12.4 cM/marker. In addition, the two linkage groups associatedwith the sex chromosome Z had apparently anoma-lous marker distributions. HEC Z (114.9 cM) had 9framework markers and the highest average inter-marker distance of 14.4 markers per cM. On the otherhand, the extra group associated with the sex chromo-some Z* (12.4 cM) has 6 markers and only 2.4 cM permarker.
All of the codominant anchor loci were mapped, withthe exception of EF1a (which grouped with HEE 6 in the
748 D. D. Kapan et al.
BI data) and He-CA-007. Eight of the 21 linkage groupshad two or more anchor loci predominantly found onHEE 7, including the majority of ribosomal proteingenes RpL5, RpS5, RpL10a, RpS8, RpP0, which mappedalong with two microsatellites Hel-02 and Hel-04.
Sex chromosomes: As mentioned above, Z-taggedmarkers associate with two separate MI-identifiedlinkage groups, Z and Z*. Each of these groups alsocontained an anchor locus associated with the sexchromosome. All but 2 of the loci in the larger groupZ (7 censored-BI AFLPs, 1 MI AFLP, Tpi, and He-CA-002)were inherited from the H. himera paternal grand-father’s Z chromosome (i.e., these loci were in the samelinkage phase). Furthermore, all but 1 of the small Z*loci were in the opposite linkage phase indicating theywere inherited from the H. erato etylus paternal grand-mother. These two linkage groups do not join until,LOD 4.0. This pattern may reflect a recombinationhotspot on Z or reduced pairing of the two species’Z chromosomes during meiosis. However, in the ab-sence of cytological evidence to the contrary, we choseto present the two clusters as a single linkage group,although there are insufficient data to orient the twoclusters with respect to each other (Figure 2). In-terestingly and somewhat surprisingly, despite scoringnearly 200 FI loci, we did not identify a single W-linked(female only) marker.
Recombination: In this study, we found an averageof 3.1 crossover events per marker per chromosomeaveraged over all linkage groups (assuming half thedouble recombinants are valid crossover events). HEE dhad the lowest estimate (1.7 per marker) and HEC Zhad the highest estimate (4.8 crossover events permarker). When estimated across autosomes the averagenumber of recombination events per cM was 0.59(range 0.35–0.94). The two linkage groups associatedwith Z were the lowest (0.32 crossovers per cM) and thehighest (1.38 crossovers per cM).
Mapping color pattern loci: The two color patternloci, Sd and D, mapped to linkage groups HEE 6 andHEC 3, respectively. When scored as a typical BI locusthe 95% confidence interval for Sd is 21.5 cM bracketedby markers LTAA169 and TCCT138 (Figure 3A). How-ever, the MI c-spot allowed a tighter placement of thislocus, decreasing the 95% confidence interval contain-ing the Sd locus to an 8.8-cM region defined by threeAFLP bands on linkage group HEE 6 where themaximum likelihood interval (�75% of the probabilitydensity) is a 6.3-cM window flanked by two MI markers,TCAC466 and ACCA167 (Figure 3A). Utilizing themetabulks to find additional loci linked to Sd producedtwo further loci: CCTAmbA �5.6 cM above the maxi-mum likelihood position and CCTTSd �12.9 cM belowthe maximum likelihood position.
On the basis of initial mapping the D locus was simi-larly localized, mapping�3 cM terminal to TCAC242 onHEC 3. To bracket D we searched for additional mark-
ers using metabulks organized around D described inmaterials and methods (see Table 1 legend for de-tails). These scans revealed two additional TCTADRYand CCAC491 loci linked to D. With the addition ofthese loci the 95% confidence limits showed this locusto be either 2–3 cM beyond the terminal marker or inthe 9.4 cM region between marker CCAC491 andTCTADRY (Figure 3B). Unlike the two spot alleles ofthe Sd locus, the D locus did not show 1:1 phenotypiceffects. However, by specifically following the H. eratoetylus D allele [D (MI)] of the male parent, predicted byassociation with the chromosome print (see solid barsin Figure 3B and supplemental Appendix at http://www.genetics.org/supplemental/), we are able to dou-ble the number of informative offspring and show themaximum likelihood position of D is the 6.7-cM intervalbetween CCAC491 and TCAC242 at the end of linkagegroup HEC 3 (�70% of the probability density, Figure3B, black bars).
DISCUSSION
Heliconius butterflies are renowned for their diversityof brilliant mimetic wing color patterns, making them amodel clade for studying adaptive radiation, mimicry,natural selection, and speciation (Emsley 1965; Benson
1972; Gilbert 1972, 2003; Turner 1984; Mallet et al.1990, 1998; Mallet 1993; Brower 1994, 1996;Srygley 1999; Jiggins et al. 2001; Kapan 2001; Beltran
et al. 2002; Naisbit et al. 2003; Flanagan et al. 2004;Langham 2005). Uniquely, the radiation in Heliconiuswarning color patterns couples both divergent evolu-tion within species and multiple independent cases ofconvergent evolution between distantly related species(Turner 1983, 1984; Brower 1996; Mallet et al. 1998;Flanagan et al. 2004). This pattern of divergent andconvergent evolution is best exemplified in H. erato andH. melpomene, two distantly related species that shareidentical wing pattern phenotypes due to Mullerianmimicry where they co-occur, as well as displayingparallel geographic patterns of divergence across Cen-tral and South America (Turner 1983). Warning colorvariation in both species is controlled by a handful ofloci with major phenotypic effects (Sheppard et al.1985), which raises questions about the exact nature ofthe loci involved and the extent that analogous colorpattern changes in the two species are caused bychanges in homologous loci. Are phenotypic changescaused by changes in regulatory regions, gene duplica-tions, or possibly adaptive recombination to form link-age among previously unlinked pattern elements? Arethe highly similar wing pattern elements derived fromstrictly homologous building blocks, or are differentloci used to create nearly identical phenotypic patterns?Answers to these questions will elucidate fundamentalprocesses of the evolution of adaptive phenotypes andwill advance our understanding of how evolutionary
Dense Linkage Map of Heliconius erato 749
Figure 3.—Placement of color pattern loci Sd (A) and D (B). Horizontal bars represent the probability that the gene locusresponsible for the phenotypic effects seen in the cross is located in the interval between the marker loci on a fixed map calculatedfrom the antilog of the base ten log-likelihoods of linkage for each interval returned with the Mapmaker ‘‘try’’ command, nor-malized by the sum of these transformed values for all intervals including unlinked (Veiland 1998). Absence of a bar indicates aneffectively zero probability that the locus occurs in the interval. The broad map backbones indicate the �95% confidence regionswhere color pattern loci occur (#LOD 2.0 units from maximum likelihood position) as gray bars for BI scoring and as black barsfor MI scoring. In A, the BI scoring of the codominant Sd phenotype is represented by gray bars, while the black bars represent the1:1 MI scoring of the c-spot, likely an alternative allele of Sd (see results and discussion). In B, the D locus is shown with the BIscoring as gray bars and the 1:1 MI scores following the H. erato etylus D allele of the male parent as black bars (see materials and
methods). Note, in B the 95% confidence limits are the same for both BI and MI scorings.
750 D. D. Kapan et al.
change is constrained by preexisting genetic variationand/or developmental processes.
As our second step toward the genetic and develop-mental dissection of convergent and divergent evolu-tion in Heliconius, we report the first high-resolutiongenetic map for H. erato. This new map is a considerableadvance over our preliminary work on H. erato (Tobler
et al. 2005) and similar in density to the genetic linkagemap recently published for H. melpomene ( Jiggins et al.2005). It is based on segregation of markers in a muchlarger brood and contains over three times the numberof loci. This combination reduces the average inter-marker distance nearly fourfold to 5.1 cM and decreasesthe coefficient of variation of linkage group size by.200%. The new map provides a stable backbone forlocalizing the color pattern genes responsible for H.
erato’s interracial diversification within the genome andrelative to potential candidate loci (Figure 2). Further-more, this map is an important reference for continuedlinkage analysis in Heliconius. In total, 13 shared SCNL(of 18 possible) and one microsatellite (Hm06) weremapped to eight linkage groups in both H. erato andH. melpomene ( Jiggins et al. 2005) allowing for the firstdirect comparisons of the genetic architecture underly-ing the replicate color pattern radiations (Figure 4).
Genetic basis of divergent evolution—mapping colorpattern genes in H. erato: H. erato has the advantage thatthe various patterning loci can be studied by crossingdifferent races to the same stock of its sister speciesH. himera. For example, in our H. erato etylus 3 H. himerabroods, both D and Sd phenotypes are codominant andhave complementary effects leading to expected single
Figure 4.—Comparison of linkage results in the present study with linkage groups from corresponding studies: H. melpomene(left) ( Jiggins et al. 2005), H. erato cyrbia (center) (Tobler et al. 2005), and H. erato cyrbia (right, this study) on the basis of sharedcodominant anchor loci. The density of AFLP markers is indicated by horizontal tick marks at each framework location. AFLP lociforming haplotypes are joined by 1 signs. (A) All linkage groups originally identified in Tobler et al. (2005) and also identified inJiggins et al. (2005). (B) Linkage groups originally identified in Tobler et al. (2005) not yet found in H. melpomene. (C) Linkagegroups with tentative homology between present study and H. melpomene ( Jiggins et al. 2005). A–C are drawn to different scales.
Dense Linkage Map of Heliconius erato 751
locus 1:2:1 and two locus 9:3:3:1 offspring ratios in F2
style crosses (Figure 1). In addition, scoring of both lociis simplified because of the absence of epistasis whencrossed with H. himera, which is normally seen in in-terracial crosses among many of the loci responsible forpattern variation in Heliconius (Mallet 1989; Jiggins
and McMillan 1997; Gilbert 2003; Naisbit et al. 2003;Tobler et al. 2005).
Between Tobleret al. (2005) and the present study wehave now mapped three color pattern loci in H. erato (D,Sd, and Cr). The D locus affects red and orange colorpattern elements across the fore- and hindwing; incontrast, both the Sd and the Cr loci affect the patternof melanization in the fore- and hindwing of H. erato andH. himera (Sheppard et al. 1985; Mallet 1989; Jiggins
and McMillan 1997). All three loci are linked tocodominant loci, which act as important anchors forintegrating different mapping projects within speciesand for comparing the patterns of synteny betweenspecies (see below). For example, we can now demon-strate homology of hypothesized D locus phenotypiceffects across H. himera, H. erato cyrbia, and H. erato etylus(as was hypothesized by Jiggins and McMillan 1997;Tobler et al. 2005). The D locus segregates in both ourH. himera 3 H. erato cyrbia and H. himera 3 H. erato etyluscrosses. In both crosses, the locus maps to the end of thesame linkage group containing the SCNL Cubitusinterruptus (Ci) (Tobler et al. 2005). Additional F2crosses between H. himera and H. erato cyrbia (threecrosses, 259 offspring) and H. erato etylus (five addi-
tional crosses, total of 598 offspring) performed in ourlab show identical results with Mendelian segregationof D and no recombinants observed between D locuspattern elements arising from H. himera and H. erato(D. D. Kapan and W. O. McMillan, unpublished data).
AFLPs are a valuable tool for investigating unex-plored genomes (Parsons and Shaw 2002) and serveseveral important functions in our ongoing mappingefforts in Heliconius. Foremost, they allow the tightlocalization of particular color pattern loci on a densemap. In this study, with only 20 primer combinations, wemapped over 350 segregating AFLP loci in our in-terspecific cross. This level of marker coverage allowedus to narrow the maximum likelihood interval contain-ing the Sd locus to 6.3-cM region flanked by two AFLPbands on linkage group HEE 6. With the addition ofseveral new primer combinations we were similarly ableto flank the D locus in a 6.7-cM interval near the end oflinkage group HEC 3. Given the estimated recombina-tion length of 1430 Haldane cM and a physical size of�395 Mb (Tobler et al. 2005) and assuming a one-to-one correspondence between physical and recombina-tion size suggests an interval size of �1.7 Mb (�2.4 Mb95% CI) for the Sd and 1.9 Mb (�3.5 Mb 95% CI) for theD locus. Of course the relationship between physicaland recombination size is likely to be complex (McVean
et al. 2004); however, our estimates are conservative andthere are several AFLP bands segregating in our screenof 88 offspring that were only one to two recombinantsdistant from the two major color pattern loci. This level
Figure 4.—Continued.
752 D. D. Kapan et al.
of recombination is within the estimated error thatwe encountered when mapping with dominant AFLPmarkers.
Second, AFLP loci identified in this and other geneticscreens of Heliconius broods provide a ready supplyof new loci tightly linked to color pattern loci andwith further development can be an important source ofcodominant loci for ongoing high-resolution mappingstudies. Using a combination of traditional mappingand color pattern metabulking we have found a numberof AFLP fragments tightly linked to Sd and D. We havenow isolated, cloned, and sequenced several of theselinked AFLP including the terminal marker on HEC 3that is tightly linked to D and have scored this markerin another large cross between H. himera and the verydistinctive east Ecuadorian race H. erato notabalis (K.Maldonado, H. A. Merchan Gonzalez, D. D. Kapan
and W. O. McMillan, unpublished results). In H.notabalis the D allele places a small patch of red scalesin the distal portion of the area of the wing betweenveins Cu1a and Cu1b in a background of otherwisewhite scales. Between these two broods this markershows a total of four potential recombinants across
.200 offspring, suggesting that it is #2 cM from D (K.Maldonado, H. A. Merchan Gonzalez, D. D. Kapan
and W. O. McMillan, unpublished results).AFLP markers will also be important for placing color
pattern loci that do not clearly segregate together on acommon reference map of the H. erato color patternradiation. For example, the Cr locus does not segregatein H. himera3H. erato etylus crosses, but is on group HEC2 on the basis of linkage to the microsatellite Hel 13/14(Tobler et al. 2005) (Figure 4B). However, the inflatedsize of HEC 2 estimated by Tobler et al. (2005) makesit impossible to more tightly localize Cr on the currentmap due to the extreme distances between Cr and thetwo segregating anchor loci (Hel 13/14, 87 cM, andallozyme ACO-S, 124 cM), both greater than the entirelength of HEC 2 in the present study (63.0 cM, Figure4B). As an alternative, we are currently using the colorpattern metabulking strategy to develop AFLP lociaround Cr that will work across different mappingexperiments by targeting flanking loci and convertingthese into anchor loci (see above). Given the high levelof epistasis among major color pattern loci in Helico-nius, the development of these genomic anchors will
Figure 4.—Continued.
Dense Linkage Map of Heliconius erato 753
further refine our understanding of the broad pheno-typic effects of color pattern loci in Heliconius.
Candidate loci: Our new high-resolution Heliconiuslinkage map allows us to test for the association betweencandidate loci, chosen on the basis of knowledge ofgene action in other organisms, and color patterngenes. This candidate gene approach has been verysuccessful in other organisms. For example, genesknown to be involved in wing development in Drosoph-ila were found to have novel but related roles in pat-tern specification in butterflies (Carroll et al. 1994;Brakefield et al. 1996; Koch et al. 1998, 2000; Keys et al.1999; Brunetti et al. 2001; Reed and Gilbert 2004;Reed and Serfas 2004) and variation at one of theseloci, distal-less, is speculated to cause variation in eyespotsize in selected lines of B. anayana (Beldade et al. 2002).In Heliconius, the candidate gene approach has allowedus to eliminate several potential candidates for colorpattern loci. Between this study and Tobler et al. (2005)we have excluded genes that play a role in eitherdevelopment or pigment synthesis as candidates forD, Cr, and Sd. The extracellular signaling ligand geneswingless, decapentaplegic, and hedgehog, the hedgehog
receptor gene patched, the eyespot-associated transcrip-tion factor gene cubitus interruptus, and the ommo-chrome pigment synthesis enzyme gene cinnabar donot show strong linkage with any of three mapped colorpattern genes (Figure 4). However, these loci may belinked to other color pattern loci within H. erato thathave not yet been mapped.
The comparative architecture of mimicry: The strongresemblance between H. erato and H. melpomene wingpatterns has led researchers to speculate that homolo-gous loci underlie the parallel radiations (Turner 1984;Nijhout 1991). This theory suggests that developmen-tal constraints might have played a significant role in theevolution of convergent wing patterns (see Naisbit et al.2003) and predicts that loci that have similar phenotypiceffects in the two comimics should reside on the samechromosome and show similar patterns of synteny withnearby markers.
Using mapping data to detect genetic homologyrequires that genomic rearrangements have been min-imal and that both gross- and fine-scale patterns ofsynteny are likely conserved between the two comimics.Although we have jointly mapped only a small number
Figure 4.—Continued.
754 D. D. Kapan et al.
of single copy nuclear genes in the two comimics thusfar, these shared markers indicate that gene order ishighly conserved. Indeed, there are no conflictinglinkage relationships between H. erato and H. melpomene.The strongest support of the general conservation forlinkage relationship within Heliconius comes from acluster of ribosomal proteins (RpL5, RpS5, RpL10a,RpS8, and RpPO), all of which map to the same linkagegroup (HEE 7 and HMEL 11) and show conserved geneorder (Figure 4). The conservation of linkage and theobservation that both species of Heliconius have thesame number of chromosomes is consistent with the hy-pothesis that there have not been many chromosomefusions or breakages since the two lineages diverged. Onthe basis of one or two shared codominant anchor lociper linkage group (Jiggins et al. 2005), we identifiedtentative homology between seven autosomal linkagegroups as well as the Z chromosome between the twocomimics (Figure 4). Although there was no correlationbetween the size of putatively homologous linkagegroups (r ¼ �0.097, t6 ¼ �0.24, P . 0.59), the overallspanning map size of the species was similar, 1430Haldane cM in H. erato vs. 1616 cM in H. melpomene( Jiggins et al. 2005). This similarity is out of accord withthe recent estimates of genome size, which suggest thatH. erato’s genome is �136% greater than that of H.melpomene ( Jiggins et al. 2005). The slightly largerrecombination length coupled with a smaller genomesize of H. melpomene suggests that H. melpomene has ahigher crossing over frequency than H. erato. This isintriguing in light of evidence that H. erato is almostalways more abundant and may commonly act as theMullerian model in this pair (Kapan 2001; Mallet
2001; Flanagan et al. 2004). If true, under these con-ditions H. melpomene may benefit from genomic flexibil-ity provided by an increased crossing-over rate to trackH. erato (see below).
To date, only one major color pattern locus, Yb, hasbeen mapped in H. melpomene (Jiggins et al. 2005). TheYb locus controls the yellow elements on the dorsal andventral hindwing surfaces in H. melpomene (Sheppard
et al. 1985; Mallet 1989; Jiggins et al. 2005) and mapsto H. melpomene group HMEL 15 along with two micro-satellite loci, Hm01 and Hm08. Yb is extremely tightlylinked to Sb, which controls the white fringe on thehindwing of H. m. cythera, and to N, which controls thepresence of melanin in the forewing some of Amazo-nian races (see Sheppard et al. 1985; Mallet 1989;Naisbit et al. 2003; Jiggins et al. 2005). Yb/Sb/N com-plex produces nearly identical phenotypes to the Crlocus in H. erato, which maps to HEC 2 based on thepresence of the microsatellite Hel-13/14 (Tobler et al.2005) (Figure 2). Unfortunately, there are no cosegre-gating color pattern loci in common between the H.erato and H. melpomene maps for these linkage groups,but the color pattern homology hypothesis predicts thatCr and Yb/Sb/N will map to the homologous linkage
group and will show similar linkage relationships toother loci in this group. An identical argument holds forthe D locus in H. melpomene and the D locus in H. erato.Alleles at these loci have similar (but not identical)phenotypic effects in the two comimics and have beenassumed to be homologous. In H. erato the D locus isloosely linked to Ci, suggesting that the D locus in H.melpomene should map to linkage group HMEL 18 inJiggins et al. (2005), a prediction recently borne out byadditional mapping work in H. melpomene (C. Jiggins,personal communication). There are several potentialH. melpomene homologs to H. erato’s Sd locus as a numberof loci affect melanin patterns (and therefore the shape,size, and position of yellow pattern elements) on theforewing of both species (Sheppard et al. 1985; Gilbert
2003); therefore, further mapping will be necessary toascertain color pattern homologs for Sd.
Codominant anchor loci and continued linkagemapping in Heliconius: Strong arguments for or againstcolor pattern homology will require comparing bothbroad- and fine-scale synteny. This will be especiallyimportant as many of the genes described in Heliconiushave major phenotypic effects and some appear to becomposed of complexes of tightly linked loci or super-genes (Sheppard et al. 1985; Mallet 1989; Nijhout
1991; Jiggins and McMillan 1997; Joron et al.2001; Naisbit et al. 2003). One explanation for thispattern is that mimicry-related selection has favoredstronger linkage between genes that work together toform specific adaptive color patterns. Adaptive evolu-tion in linkage seems unlikely in Heliconius (seeNaisbit et al. 2003 for discussion), but could be moreimportant than previously envisioned if Mullerianmimicry evolution typically proceeds in a one-sidedfashion where a rare species (in this case, H. melpomene)evolves similarity to (or adverges on) the pattern of amore common species (Mallet 2001; Kapan 2001,Flanagan et al. 2004). True integration of these mapsnecessitates the development of more anchor loci fordefining the broad-scale patterns of synteny between thetwo comimics. Of the two codominant marker systemsused in our mapping work, SCNL have proved to bemuch better anchors for comparative genomics inHeliconius. Only 2 (Hel 05, Hm06) of the .40 micro-satellites developed for Heliconius could be mapped inthe two species ( Jiggins et al. 2005, Tobler et al. 2005).In contrast, over half of the 21 SCNL developed fromcandidate and housekeeping genes have been mappedor placed in linkage groups in both species (Table 4,Figure 4) ( Jiggins et al. 2005; Tobler et al. 2005).Generating new gene-based markers in Heliconius isnow extremely efficient using a growing EST data-base (http://heliconius.cap.edu.ac.uk/butterfly/db/,see Papanicolaou et al. 2005). Furthermore, becausethere is no crossing over during oogenesis in Lepidop-tera (Suomalainen et al., 1973) only a small number ofindividuals need to be typed to assign particular markers
Dense Linkage Map of Heliconius erato 755
to particular linkage groups. However, a large number ofSCNL in both species will undoubtedly need to be typedto characterize the fine-scale patterns of synteny. In thisrespect, targeted AFLP loci can provide critical genomicanchors around color pattern loci. To date, primersdesigned around AFLP loci in one species have notworked in the other species (W. O. McMillan and C. D.Jiggins, unpublished data). Yet targeted AFLP loci aregood probes for BAC libraries, which are available forboth species, and gene-based markers tightly linked tocolor pattern genes are currently being developed frominitial BAC sequences.
Conclusions and future directions: Developing link-age maps for Heliconius is critical if we are to build uponthe long history of ecological and evolutionary workin this group to address questions concerning theinterplay between selection and morphological diversi-fication. The localization of two color pattern genes tonarrow regions of two linkage groups follows a similarsuccess for a single color pattern locus in H. erato’sdistantly related comimic, H. melpomene ( Jiggins et al.2005). Both projects utilized a three-step mappingstrategy that takes advantage of the diversity of segrega-tion markers in natural populations ( Jiggins et al. 2005;see supplemental Figure S1 at http://www.genetics.org/supplemental/). This technique is applicable to anyoutbred species with sex-restricted recombination, suchas nearly all Lepidoptera. It relies on FI alleles segregat-ing in an F2-style cross to generate a chromosomalcontig or print (Yasukochi 1998) that is utilized tounambiguously identify linkage groups from markersrecombining in the male parent. These Heliconiusmapping projects serve as a foundation for the compar-ative study of the evolutionary genetic basis of mimicry,represent an important step toward obtaining targetedsequence data from genomic regions linked to genesunderlying color pattern evolution, and promise toaccelerate the identification and molecular character-ization of the color pattern loci themselves. Ultimately,this work will help promote a wider understanding ofhow convergent morphologies arise in nature.
We thank Jim Mallet for his keen insights into color pattern geneticsin Heliconius and comments on early versions of this manuscript andChris Jiggins for assistance with Joinmap and early discussion of thethree-step method prior to publication. We also thank the Ministeriodel Ambiente in Ecuador for permission to collect butterflies andGermania Estevez for help and guidance during collecting trips. Wegive a special thanks to Ana Maria Quiles and many volunteers of thebutterfly rearing crew for help rearing larvae and maintaining theHeliconius insectaries at the University of Puerto Rico. We thankalso Wyndham Kapan for fully updating the program Genographerin JAVA (version 2.1 beta available at http://zephyr.hpcf.upr.edu/�mcmi-lab/). Finally, we thank Shannon N. Bennett and ValerieMcMillan for critical review of the manuscript and Daniel Lindstromfor his support and editing. Undergraduate students participating inthis project were supported in part by University of Puerto Rico’sResearch Initiative for Scientific Enhancement, Puerto Rico’s Alliancefor Minority Participation, and the University of Puerto Rico RioPiedras’ Center for Applied Tropical Ecology and Conservation.
Puerto Rico Experimental Program to Stimulate Competitive Re-search program and Biomedical Research Infrastructure Network inPuerto Rico and the National Institutes of the Health Support ofContinuous Research Excellence Program provided funding for theSequencing and Genotyping Center at University of Puerto Rico RioPiedras. Funding for this study was provided by National ScienceFoundation grants to W.O.M.
LITERATURE CITED
Albertson, R. C., J. T. Streelman and T. D. Kocher, 2003 Geneticbasis of adaptive shape differences in the cichlid head. J. Hered.94: 291–301.
Beebe, W., 1955 Polymorphism in reared broods of Heliconius but-terflies from Surinam and Trinidad. Zoologica 40: 139–143.
Beldade, P., P. M. Brakefield and A. D. Long, 2002 Contributionof Distal-less to quantitative variation in butterfly eyespots. Nature415: 315–318.
Beltran, M., C. D. Jiggins, V. Bull, M. Linares, W. O. McMillan
et al., 2002 Phylogenetic discordance at the species boundary:comparative gene genealogies among Heliconius butterflies.Mol. Biol. Evol. 19: 2176–2190.
Benham, J., J.-U. Jeung, M. Jasieniu, V. Kanazin and T. Blake,1999 Genographer: a graphical tool for automated AFLP andmicrosatellite analysis. Article 3, Journal of Agricultural Geno-mics. Published with permission from CAB International. Fulltext available from http://www.cabi-publishing.org/JAG
Benson, W. W., 1972 Natural selection for Mullerian mimicry inHeliconius erato in Costa Rica. Science 176: 936–939.
Bradshaw, H. D., Jr., K. G. Otto, B. E. Frewen, J. K. McKay andD. W. Schemske, 1998 Quantitative trait loci affecting differen-ces in floral morphology between two species of monkeyflower(Mimulus). Genetics 149: 367–382.
Brakefield, P. M., J. Gates, D. Keys, F. Kesbeke, P. J. Wijngaarden
et al., 1996 Development, plasticity, and evolution of butterflyeyespot patterns. Nature 384: 236–242.
Brower, A. V. Z., 1994 Rapid morphological radiation and conver-gence among races of the butterfly Heliconius erato inferred frompatterns of mitochondrial DNA evolution. Proc. Natl. Acad. Sci.USA 91: 6491–6495.
Brower, A. V. Z., 1996 Parallel race formation and the evolution ofmimicry in Heliconius butterflies: a phylogenetic hypothesis frommitochondrial DNA sequences. Evolution 50: 195–221.
Brown, K. S., P. M. Sheppard and J. R. G. Turner, 1974 Quaternaryrefugia in tropical America: evidence from race formation in Hel-iconius butterflies. Proc. R. Soc. Lond. Ser. B 187: 369–378.
Brown, K. S., T. C. Emmel, P. J. Eliazar and E. Soumalainen,1992 Evolutionary patterns in chromosome numbers of Neo-tropical Lepidoptera. I. Chromosomes of the Heliconiini (familyNymphalidae: subfamily Nymphalinae). Hereditas 117: 109–125.
Brunetti, C. R., J. E. Selegue, A. Monteiro, V. French, P. M.Brakefield et al., 2001 The generation and diversification ofbutterfly eyespot color patterns. Curr. Biol. 11: 1578–1585.
Carroll, S. B., J.Gates, D.N. Keys, S. W. Paddock, G. E.F. Panganiban
et al., 1994 Pattern formation and eyespot determination inbutterfly wings. Science 265: 109–114.
Chai, P., 1986 Field observations and feeding experiments on the re-sponses of rufous-tailed jacamars (Galbula ruficauda) to free-flyingbutterflies in a tropical rainforest. Biol. J. Linn. Soc. 29: 166–189.
Colosimo, P. F., K. E. Hosemann, S. Balabhadra, G. Villarreal, Jr.,M. Dickson et al., 2005 Widespread parallel evolution in stickle-backs by repeated fixation of ectodysplasin alleles. Science 307:1928–1933.
Emsley, M. G., 1963 A morphological study of imagine Heliconiinae(Lep.: Nymphalidae) with a consideration of the evolutionary re-lationships within the group. Zoologica 48: 85–130.
Emsley, M. G., 1965 Speciation in Heliconius (Lep. Nymphalidae):morphology and geographic distribution. Zoologica 50: 191–254.
Flanagan, N. S., M. Blum, A Davison, M Alamo, R Albarran et al.,2002 Characterization of microsatellite loci in neotropical Hel-iconius butterflies. Mol. Ecol. Notes 2: 398–401.
Flanagan, N., A. Tobler, D. Kapan, A. Davison, O. Pybus et al.,2004 Historical demography of Mullerian mimicry in the
756 D. D. Kapan et al.
Neo-tropical Heliconius butterflies. Proc. Natl. Acad. Sci. USA101: 9704–9709.
Gilbert, L. E., 1972 Pollen feeding and reproductive biology of Hel-iconius butterflies. Proc. Natl. Acad. Sci. USA 69: 1403–1407.
Gilbert, L. E., 2003 Adaptive novelty through introgression in Hel-iconius wing patterns: evidence for shared genetic ‘‘toolbox’’ fromsynthetic hybrid zones and a theory of diversification, pp. 281–318 in Ecology and Evolution Taking Flight: Butterflies as Model Sys-tems, edited by C. L. Boggs, W. B. Watt and P. R. Ehrlich.University of Chicago Press, Chicago.
Gilbert, L. E., H. S. Forrest, T. D. Schultz and D. J. Harvey,1988 Correlations of ultrastructure and pigmentation suggesthow genes control development of wing scales of Heliconius but-terflies. J. Res. Lepid. 26: 141–160.
Hoekstra, H. E., K. E. Drumm and M. W. Nachman, 2004 Ecologicalgenetics of adaptive color polymorphism in pocket mice: geo-graphic variation in selected and neutral genes. Evolution 58:1329–1341.
Jiggins, C. D., and W. O. McMillan, 1997 The genetic basis of anadaptive radiation: warning colour in two Heliconius species. Proc.R. Soc. Lond. Ser. B 246: 1167–1175.
Jiggins, C. D., R. E. Naisbit, R. L. Coe and J. Mallet, 2001 Re-productive isolation caused by colour pattern mimicry. Nature411: 302–305.
Jiggins, C. D., J. Mavarez, M. Beltran, W. O. McMillan, S.Johnston et al., 2005 A genetic linkage map of the mimeticbutterfly Heliconius melpomene. Genetics 171: 557–570.
Joron, M., I. R. Wynne, G. Lamas and J. Mallet, 2001 Variableselection and the coexistence of multiple mimetic forms of thebutterfly Heliconius numata. Evol. Ecol. 13: 721–754.
Kapan, D. D., 2001 Three-butterfly system provides a field test ofMullerian mimicry. Nature 409: 338–340.
Keys, D. N., D. L. Lewis, J. E. Selegue, B. J. Pearson, L. V. Goodrich
et al., 1999 Recruitment of a hedgehog regulatory circuit in but-terfly eyespot evolution. Science 283: 532–534.
Koch, P. B., D. N. Keys, T. Rocheleau, K. Aronstein, M. Blackburn
et al., 1998 Regulation of dopa decarboxylase expression duringcolour pattern formation in wild-type and melanic tiger swallow-tail butterflies. Development 125: 2303–2313.
Koch, P. B., U. Lorenz, P. M. Brakefield and R. H. Ffrench-Constant, 2000 Butterfly wing pattern mutants: developmen-tal heterochrony and co-ordinately regulated phenotypes. Dev.Genes Evol. 210: 536–544.
Langham, G. M., 2005 Specialized avian predators repeatedly attacknovel color morphs of Heliconius butterflies. Evolution 58: 2783–2787.
Lincoln, S. E., and E. S. Lander, 1992 Systematic detection of er-rors in genetic linkage data. Genomics. 14: 604–610.
Lincoln, S. E., M. J. Daly and E. S. Lander, 1987 MAPMAKER 3.0.Whitehead Institute for Biomedical Research/MIT Center forGenome Research, Boston.
Mallet, J., 1989 The genetics of warning colour in Peruvian hybridzones of Heliconius erato and H. melpomene. Proc. R. Soc. Lond. Ser.B 236: 163–185.
Mallet, J., 1993 Speciation, raciation, and color pattern evolu-tion in Heliconius butterflies: evidence from hybrid zones,pp. 226–260 in Hybrid Zones and the Evolutionary Process, editedby R. G. Harrison. Oxford University Press, New York.
Mallet, J., 2001 Causes and consequences of a lack of coevolutionin Mullerian mimicry. Evol. Ecol. 13: 777–806.
Mallet, J., N. H. Barton, G. Lamas, J. Santisteban, M. Muedas
et al., 1990 Estimates of selection and gene flow from measuresof cline width and linkage disequlibrium in Heliconius hybridzones. Genetics 124: 921–936.
Mallet, J., W. O. McMillan and C. D. Jiggins, 1998 Mate choice be-tween a pair of Heliconius species in the wild. Evolution 52: 503–510.
McMillan, W. O., C. D. Jiggins and J. Mallet, 1997 What initiatesspeciation in passion-vine butterflies? Proc. Natl. Acad. Sci. USA94: 8628–8633.
McVean, G. A. T., S. R. Myers, S. Hunt, P. Deloukas, D. R. Bentley
et al., 2004 The fine-scale structure of recombination rate vari-ation in the human genome. Science 304: 581–584.
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 genomic
regions by using segregating populations. Proc. Natl. Acad. Sci.USA 88: 9828–9832.
Muller, F., 1879 Ituna and Thyridia; a remarkable case of mimicryin butterflies. Trans. Entomol. Soc. 1879: xx–xxix.
Nachman, M. W., H. E. Hoekstra and S. L. D’Agostino, 2004 Thegenetic basis of adaptive melanism in pocket mice. Proc. Natl.Acad. Sci. USA 100: 5268–5273.
Naisbit, R. E., C. D. Jiggins and J. Mallet, 2003 Mimicry: develop-mental genes that contribute to speciation. Evol. Dev. 5: 269–280.
Nijhout, H. F., 1991 The Development and Evolution of Butterfly WingPatterns. Smithsonian Institution Press, Washington, DC.
Nijhout, H. F., G. A. Wray and L. E. Gilbert, 1990 An analysis ofthe phenotypic effects of certain colour pattern genes in Helico-nius (Lepidoptera: Nymphalidae). Biol. J. Linn. Soc. 40: 357–372.
Papanicolaou, A., M. Joron, W. O. McMillan, M. L. Blaxter andC. D. Jiggins, 2005 Genomic tools and cDNA derived markersfor butterflies. Mol. Ecol. 14: 2883–2897.
Parsons, Y. M., and K. L. Shaw, 2002 Mapping unexplored ge-nomes: a genetic linkage map of the Hawaiian cricket Laupala.Genetics 162: 1275–1282.
Peichel, C. L., K. S. Nereng, K. A. Ohgi, B. L. E. Cole, P. F. Colosimo
et al., 2001 The genetic architecture of divergence betweenthreespine stickleback species. Nature 414: 901.
Reed, R. D., and L. E. Gilbert, 2004 Wing venation and Distal-lessexpression in Heliconius butterfly wing pattern development. Dev.Genes Evol. 214: 628–634.
Reed, R. D., and M. S. Serfas, 2004 Butterfly wing pattern evolutionis associated with changes in a notch/distal-less temporal patternformation process. Curr. Biol. 14: 1159–1166.
Sheppard, P. M., J. R. G. Turner, K. S. Brown, W. W. Benson andM. C. Singer, 1985 Genetics and the evolution of Mullerianmimicry in Heliconius butterflies. Philos. Trans. R. Soc. Lond.Ser. B 308: 433–613.
Shi, J., D. G. Heckel and M. R. Goldsmith, 1995 A genetic linkagemap for the domesticated silkworm, Bombyx mori, based on re-striction fragment length polymorphisms. Genet. Res. 66: 109–126.
Srygley, R. B., 1999 Locomotor mimicry in Heliconius butterflies:contrast analyses of flight morphology and kinematics. Philos.Trans. R. Soc. Lond. B 354: 203–214.
Suomalainen, E., L. M. Cook and J. R. G. Turner, 1971 Chro-mosome numbers of heliconinne butterflies from Trinidad, WestIndies (Lepidoptera, Nymphalidae). Zoologica 56: 121–124.
Suomalainen, E., L. M. Cook and J. R. G. Turner, 1973 Achiasmaticoogenesis in the heliconiine butterflies. Hereditas 74: 302–304.
Tobler, A., D. Kapan, N. Flanagan, C. Gonzalez, E. Peterson et al.,2005 First generation linkage map of H. erato. Heredity 94:408–417.
Turner, J. R. G., 1974 Mimetic butterflies and punctuated equilibria:some old light on a new paradigm. Biol. J. Linn. Soc. 20: 277–300.
Turner, J. R. G., 1983 Mimetic butterflies and punctuated equilibria:some old light on a new paradigm. Biol. J. Linn. Soc. 20: 277–300.
Turner, J. R. G., 1984 Mimicry: the palatability spectrum and its con-sequences, pp. 141–161 in The Biology of Butterflies, edited by R. I.Vane-Wright and P. R. Ackery. Academic Press, London.
Turner, J. R. G., and J. Smiley, 1975 Absence of crossing-over infemale butterflies (Heliconius). Heredity 34: 265–269.
Turner, J. G. R., and J. L. B. Mallet, 1997 Did forest islands drivethe diversity of warningly coloured butterflies? Biotic drift andthe shifting balance. Philos. Trans. R. Soc. Lond. B 351: 835–845.
Van Ooijen, J. W., and R. E. Voorrips, 2001 JoinMap 3.0: softwarefor the calculation of genetic linkage maps. Plant Research Inter-national, Wageningen, The Netherlands.
Veiland, V., 1998 Bayesian linkage analysis, or: how I learned to stopworrying and love the posterior probability of linkage. Am. J.Hum. Genet.. 63: 947–954.
Vos, P., R. Hogers, M. Bleeker, M. Reijans, T. van de Lee et al.,1995 AFLP: a new technique for DNA fingerprinting. NucleicAcids Res. 23: 4407–4414.
Yasukochi, Y., 1998 A dense genetic map of the silkworm, Bombyxmori, covering all chromosomes based on 1018 molecularmarkers. Genetics 150: 1513–1525.
Communicating editor: D. A. Long
Dense Linkage Map of Heliconius erato 757