Fissions, Fusions, and Translocations Shaped the Karyotype and Multiple Sex

1 Fissions, fusions, and translocations shaped the karyotype and multiple sex

2 chromosome constitution in the northeast-Asian wood white butterfly, Leptidea

3 amurensis

5 JINDRA ŠÍCHOVÁ1,2, MIZUKI OHNO3, VLAD DINCĂ4,5, MICHIHITO WATANABE6, KEN

6 SAHARA3 and FRANTIŠEK MAREC1,2*

8 1Institute of Entomology, Biology Centre CAS, 370 05 České Budějovice, Czech Republic

9 2Faculty of Science, University of South Bohemia, 370 05 České Budějovice, Czech Republic

10 3Laboratory of Applied Entomology, Faculty of Agriculture, Iwate University, Morioka 020-8550, Japan

11 4Biodiversity Institute of Ontario, University of Guelph, N1G 2W1 Guelph, Ontario, Canada

12 5Institut de Biologia Evolutiva (CSIC-Universitat Pompeu-Fabra), 08003 Barcelona, Spain

13 6NPO Mt Fuji Nature Conservation Center, 6603 Fujikawaguchiko-machi, Yamanashi, 401-0301, Japan

15 Running title: Multiple sex chromosomes in Leptidea amurensis

17 Word count: 6720

19 * Corresponding author: František Marec 20 Institute of Entomology 21 Biology Centre CAS 22 Branišovská 31 23 370 05 České Budějovice 24 Czech Republic 25 26 Phone: +420-387 775 212 27 Fax: +420-385 310 354 28 e-mail: [email protected] 29

30 ABSTRACT

32 Previous studies have shown a dynamic karyotype evolution and the presence of complex sex

33 chromosome systems in three cryptic Leptidea species from Western Palearctic. To further

34 explore the chromosomal particularities of Leptidea butterflies, we examined the karyotype of

35 an Eastern Palearctic species, L. amurensis. We found a high number of chromosomes that

36 differed between the sexes and slightly varied in females, i.e. 2n=118-119 in females and

37 2n=122 in males. The analysis of female meiotic chromosomes revealed multiple sex

38 chromosomes with three W and six Z chromosomes. The curious sex chromosome

39 constitution, i.e. W1-3/Z1-6 (females) and Z1-6/Z1-6 (males), together with the observed

40 heterozygotes for a chromosomal fusion is responsible for the sex-specific and intraspecific

41 variability in chromosome numbers. However, in contrast to the Western Palearctic Leptidea

42 species, the single chromosomal fusion and static distribution of cytogenetic markers (18S

43 rDNA and H3 histone genes) suggest that the karyotype of L. amurensis is stable. The data

44 obtained for four Leptidea species suggests that the multiple sex chromosome system, though

45 different among species, is a common feature of the genus Leptidea. Furthermore, inter- and

46 intraspecific variations in chromosome numbers along with complex meiotic pairing of these

47 multiple sex chromosomes point to the role of chromosomal fissions, fusions, and

48 translocations in the karyotype evolution of Leptidea butterflies.

50 ADDITIONAL KEYWORDS: Chromosome number variation – chromosome fusion –

51 fluorescence in situ hybridization – Leptidea amurensis – multiple sex chromosomes –

52 Pieridae.

53 INTRODUCTION

55 Moths and butterflies (Lepidoptera) exhibit several peculiar cytogenetic features by which they

56 differ greatly from other groups of insects except for their sister group, caddisflies

57 (Trichoptera). The most striking feature is the chromosomal mechanism of sex determination

58 with female heterogamety represented by Z0/ZZ and WZ/ZZ (female/male) sex chromosome

59 systems or their numerical variations. Female heterogamety is associated with achiasmatic

60 meiosis in females, i.e. the absence of crossing over and chiasmata in meiotic prophase I

61 oocytes (reviewed in Traut, Sahara & Marec, 2007). Furthermore, Lepidoptera are regarded as

62 organisms with holokinetic chromosomes due to the lack of a distinct primary constriction (the

63 centromere) and parallel disjunction of sister chromatids during mitotic metaphase. During cell

64 division the spindle microtubules attach to a large kinetochore plate covering most of the

65 chromosome surface (reviewed in Carpenter, Bloem & Marec, 2005). Lepidopteran

66 chromosomes are usually small and uniform in shape, and the majority of species are reported

67 to have haploid chromosome numbers ranging from n=28 to n=32 with the most common

68 number of n=31 (Suomalainen, 1969; Robinson, 1971; De Prins & Saitoh, 2003; Brown et al.,

69 2007). Based on the occurrence across the lepidopteran phylogenetic tree, the modal

70 chromosome number of n=31 has been proposed as the ancestral number for Lepidoptera

71 (Suomalainen, 1969; Lukhtanov, 2000). The putative ancestral karyotype is strongly supported

72 by recent results of comparative chromosome mapping (Baxter et al., 2011; Sahara et al.,

73 2013; Van’t Hof et al., 2013; Ahola et al., 2014; Yasukochi et al., 2015). Recent studies also

74 suggest extensive conservation at the chromosomal level and evolutionary stability of whole

75 lepidopteran genomic regions (Pringle et al., 2007; Sahara et al., 2007; 2013; Yasukochi et al.,

76 2006; 2009; 2015; Van’t Hof et al., 2013; Ahola et al., 2014).

77 The ancestral chromosome number is also common in butterflies, i.e. species of the

78 superfamily Papilionoidea (Robinson, 1971; Brown et al., 2007). However, some groups of

79 butterflies greatly deviate from this general pattern of karyotype stability. This applies

80 especially to butterflies of the genus Polyommatus (Lycaenidae), which display the greatest

81 interspecific variation in chromosome numbers known in the animal kingdom, ranging from

82 n=10 to n= ca224-226, where the latter comes from the recently reassessed number in the

83 Atlas blue, P. atlanticus, and represents the highest chromosome number not only of

84 Lepidoptera but of all non-polyploid eukaryotes (Kandul et al., 2004; Lukhtanov, 2015). The

85 karyotype variation is most likely caused by chromosomal rearrangements involving fusions

86 and fissions (Lukhtanov et al., 2005). In Lepidoptera, the smallest chromosome number of n=5

87 was also found in Papilionoidea, namely in a neotropical butterfly, Hypothyris thea

88 (Nymphalidae; Brown, von Schoultz & Suomalainen, 2004), and the Arizona giant-skipper,

89 Agathymus aryxna (Hesperiidae; De Prins & Saitoh, 2003). Butterflies of the family Pieridae

90 represent another group with dynamic karyotype evolution. Many species of this family have

91 reduced chromosome numbers apparently due to chromosome fusion, e.g. Eurema brigitta

92 (n=12) from the subfamily Coliadinae, Leptosia alcesta (n=12), Pinacopteryx eriphia (n=13),

93 and two well-known agricultural pests, the large white P. brassicae (n=15) and the small white

94 P. rapae (n=25), all from Pierinae (Robinson, 1971; Lukhtanov, 1991). The highest

95 chromosome number in Pieridae was described in wood white butterflies of the genus

96 Leptidea (Dismorphiinae), namely in L. duponcheli with n=102-104 (Lorković 1941; de Lesse

97 1960). Moreover, exceptional variation in chromosome numbers not only between but also

98 within species was found in this genus (Dincă et al., 2011; Lukhtanov et al., 2011; Šíchová et

99 al., 2015). For two species with predominantly Eastern Palearctic distribution, L. morsei and L.

100 amurensis, a high but constant number of chromosomes was reported, n=54 and n=61,

101 respectively (Maeki, 1958). However, three recently recognized cryptic species from the

102 Western Palearctic, L. juvernica, L. reali, and L. sinapis, have a variable number of

103 chromosomes (Dincă et al., 2011, 2013). Their diploid chromosome numbers range from

104 2n=51-55 in L. reali and 2n=80-91 in L. juvernica to 2n=56-106 in L. sinapis, the latter

105 representing the widest known intraspecific chromosome number variability, excluding cases

106 of polyploidy (Dincă et al., 2011; Lukhtanov et al., 2011; Šíchová et al., 2015). Interestingly,

107 detailed analyses of their karyotypes revealed different chromosome numbers even in the

108 progenies of individual females and variable number and location of two cytogenetic markers,

109 major rDNA and H3 histone genes. The results obtained suggested a dynamic karyotype

110 evolution through multiple chromosome fusions and fissions resulting in a frequent occurrence

111 of multivalents during meiotic divisions. Hence, the uneven chromosome segregation of the

112 multivalents is apparently being responsible for the intraspecific karyotype variation in the

113 Leptidea butterflies (Šíchová et al., 2015). In addition to the variable number of chromosomes,

114 each of the three cryptic species has a unique set of multiple sex chromosomes with

115 W1W2W3Z1Z2Z3Z4 in L. juvernica, W1W2W3W4Z1Z2Z3Z4 in L. reali, and W1W2W3Z1Z2Z3 in L.

116 sinapis (Šíchová et al., 2015).

117 Chromosomal rearrangements that give rise to complex multiple sex chromosomes can

118 result in unbalanced segregation, which could have serious consequences for the fertility

119 and/or viability of individuals. Yet, multiple sex chromosomes, composed of more than four

120 elements, evolved several times independently in different plant and animal lineages. In plants,

121 the presence of well-established sex chromosomes is generally rare (Vyskot & Hobza, 2004),

122 and meiotic sex chromosome multivalents were reported only in a few cases, such as a

123 translocation chain composed of four X and five Y in Viscum fischeri (Wiens & Barlow, 1975).

124 In animals, sex chromosome trivalents or quadrivalents are common in vertebrates except for

125 birds (reviewed in Gruetzner et al., 2006; Pokorná, Altmanová & Kratochvíl, 2014) and can

126 also be found in a number of invertebrate species (e.g. del Cerro, Cuñado & Santos, 1998;

127 Bardella et al., 2012; Palacios-Gimenez et al., 2013) including Lepidoptera (reviewed in Marec,

128 Sahara & Traut, 2010). Neo-sex chromosomes of this type usually arise by sex-chromosome

129 autosome fusion or translocation (Bertollo et al. 1997; Bertolotto, Rodrigues & Yonenaga-

130 Yassuda, 2001; Yoshido et al., 2011). However, two-four multiple X chromosomes of spiders

131 represent a special case. They form univalents that associate with each other during meiotic

132 prophase and it is believed that they originated by non-disjunction of an ancestral X

133 chromosome (Král et al., 2011). Similarly, three X chromosomes were recently reported in a

134 heteropteran insect, but in this case they probably originated by fragmentation (Kaur & Gaba,

135 2015). Nevertheless, known meiotic multiples of more than four sex chromosomes are

136 confined to invertebrates and monotremes. In the latter, the duck-billed platypus is an

137 extraordinary case with a chain of ten sex chromosomes that arose by sex-chromosome

138 autosome translocations (Grützner et al., 2004; Rens et al., 2004). In invertebrates, the most

139 complicated sex chromosome systems were described for some termites, in which males are

140 permanent translocation heterozygotes and form sex-linked chains or rings of up to 19

141 chromosomes in meiosis (Syren & Luykx, 1981). A variable number of X chromosomes (2-4)

142 plus a single Y chromosome are the hallmark of the North American tiger beetles

143 (Cicindelidae). However, changes in the number of X chromosomes are probably caused by

144 fusion or fission of the X chromosomes, i.e. without the participation of autosomes (Galián,

145 Proença & Vogler, 2007). On the contrary, a neo-X1X2X3X4X5Y system in the spider Malthonica

146 ferruginea probably evolved from an ancestral X1X2X30 system, which included an additional

147 pair of homomorphic proto-sex chromosomes, by Robertsonian fusion between the proto-Y

148 chromosome and an autosome (Král, 2007). Moreover, some mygalomorph spiders exhibit up

149 to 13 X chromosomes in males; the X-multiples originated by different chromosomal

150 rearrangements including duplications, fissions, X-X and X-autosome fusions (Král et al.,

151 2013). The evolutionary significance of the above mentioned complex multiple sex

152 chromosome systems, including those recently described in three Leptidea butterfly species

153 (Šíchová et al., 2015), is poorly understood and deserves our full attention.

154 All three Leptidea species (L. juvernica, L. reali, and L. sinapis), showing an

155 exceptional karyotype variation and unique system of multiple sex chromosomes are mainly

156 distributed in the Western Palearctic. To extend our knowledge and verify the results of the

157 earlier study of Maeki (1958), we performed a detailed karyotype analysis in one species from

158 the Eastern Palearctic, the northeast-Asian wood white L. amurensis. A comparison of male

159 and female mitotic chromosomes allowed us to determine more accurately the range of diploid

160 chromosome numbers in this species. We also mapped major rDNA and H3 histone genes by

161 fluorescence in situ hybridization (FISH) and analysed sex chromosome constitution using

162 genomic in situ hybridization (GISH). Results obtained help us to better understand the

163 karyotype and sex chromosome evolution in Leptidea butterflies.

164

165 MATERIALS AND METHODS

166

167 Insect collecting and dissection

168 Leptidea amurensis larvae and adults were collected in one area around Mt. Takazasu,

169 Yamanashi (Honshu island), Japan. In this locality, the confusion of L. amurensis with other

170 species is unlikely, as the closely related L. morsei occurs on the island of Hokkaido (Maeki,

171 1958). In the laboratory, adult females were kept in plastic containers to lay eggs and all

172 collected newly hatched larvae were reared on the host plant, Vicia amoena. Once the larvae

173 reached the fifth instar, two types of spread chromosome preparations were made following

174 the procedure described in Mediouni et al. (2004) and Yoshido, Sahara & Yasukochi (2014).

175 Briefly, both gonads for meiotic chromosomes and wing imaginal discs for mitotic

176 chromosomes were dissected, swollen for 15 min in a hypotonic solution (0.075 M KCl), fixed

177 in Carnoy fixative (ethanol/chloroform/acetic acid, 6:3:1) for 15 min, macerated with tungsten

178 needles in a drop of 60% acetic acid and spread on the slide using a heating plate at 45°C.

179 Ovaries were directly fixed without hypotonic treatment to preserve the pattern of W-

180 chromosome heterochromatin. The bodies of all dissected larvae were frozen in liquid nitrogen

181 and stored at -20°C until DNA extraction. The preparations were passed through graded

182 ethanol series (70%, 80%, and 100%, 1 min each) and stored at -20°C until further use.

183

184 Specimen sequencing and sequence analysis

185 To confirm species determination, DNA was extracted from two male and eight female larvae

186 using standard phenol-chloroform procedure (Blin & Stafford, 1976). We did not analyse all the

187 larvae, since the probability of confusion with another Leptidea species was negligible. The

188 species level identification was confirmed based on 658 bp sequences of the mitochondrial

189 gene cytochrome c oxidase subunit 1 (COI) that has been reported as reliable for the

190 identification of Leptidea species (Dincă et al., 2011, 2013; Solovyev, Ilinsky & Kosterin, 2015;

191 Šíchová et al., 2015). The COI marker was amplified using pairs of primers LepF1 (5′-

192 ATTCAACCAATCATAAAGATATTGG-3′) and LepR1 (5′-

193 TAAACTTCTGGATGTCCAAAAAATCA-3′) (Dincă et al., 2011). PCR was carried out in 25-μL

194 reaction volumes containing 1× Ex Taq buffer (TaKaRa, Otsu, Japan), 0.2 mM dNTP mix, 5

195 μmol of each primer, 0.25 U Ex Taq Hot Start DNA polymerase (TaKaRa), and about 100 ng

196 of template DNA. The PCR profile for COI consisted of denaturation for 5 min at 95°C followed

197 by 30 cycles of 30 s at 95°C, 1 min at 44°C, and 1 min at 72°C, and by a final extension of 7

198 min at 72°C. PCR products were purified using a Wizard SV Gel and PCR Clean-Up System

199 (Promega, Madison, WI, USA) and sequenced.

200 Sequences were edited and aligned using GENEIOUS PRO 4.7 created by Biomatters

201 (http://www.geneious.com/). Our sequences were combined with Leptidea data available in

202 GenBank, as follows: representatives of all unique COI haplotypes of Leptidea juvernica, L.

203 reali, and L. sinapis identified by Dincă et al. (2013), as well as all COI sequences of these

204 taxa published by Solovyev et al. (2015); all published COI sequences of other Leptidea taxa

205 (L. morsei, L. amurensis, L. lactea, and L. duponcheli) that overlapped with our sequenced

206 mtDNA fragment by at least 600 bp (Supplementary Table S1). Thus, the final COI alignment

207 included nucleotide sequences of 114 specimens and was 658 bp long.

208 To confirm the identification of the examined specimens and to place them into a

209 broader phylogenetic context, a maximum likelihood (ML) analysis was performed using

210 PhyML 2.4 (Guindon & Gascuel, 2003) implemented in GENEIOUS PRO 4.7. The substitution

211 model used was GTR+I+G and was chosen according to AIC values obtained in jModeltest 2

212 (Darriba et al., 2012). Node supports were assessed through 100 bootstrap replicates

213 (Felsenstein, 1985).

214

215 FISH with fluorochrome-labelled probes

216 Unlabelled (TTAGG)n telomeric probes, used for the identification of chromosome ends, were

217 prepared by non-template PCR according to the protocol of Sahara, Marec & Traut, (1999).

218 The probes were labelled with Cy3-dUTP (GE Healthcare, Milwaukee, WI, USA) or Orange-

219 dUTP (Abbott Molecular Inc., Des Plaines, IL, USA) using a Nick Translation Kit (Abbott

220 Molecular Inc.) with 1 h incubation at 15°C. For chromosome counts we used fluorescence in

221 situ hybridization (FISH) with (TTAGG)n telomeric probes (tel-FISH) on spread chromosome

222 preparations from wing imaginal discs (Šíchová et al., 2015). The hybridization mixture

223 contained 100 ng of telomeric probe and 25 μg of sonicated salmon sperm DNA (Sigma-

224 Aldrich, St. Louis, MO, USA) in 10 μl of 50% formamide and 10% dextran sulphate in 2× SSC.

225 To identify sex chromosomes we used genomic in situ hybridization (GISH) combined

226 with tel-FISH as described in Yoshido, Marec & Sahara (2005) and Šíchová et al. (2015). Male

227 and female gDNA was extracted from larvae by standard phenol-chloroform procedure. A part

228 of male gDNA was sonicated using a Sonopuls HD 2070 (Bandelin Electric, Berlin, Germany)

229 and used as a competitor DNA (Šíchová et al., 2013). The extracted female gDNA was

230 labelled with fluorescein-12-dUTP (Invitrogen, Carlsbad, CA, USA) or Green-dUTP (Abbott

231 Molecular Inc.) using the Nick Translation Kit with 6 h incubation at 15°C. The hybridization

232 mixture contained female gDNA (300 ng), Cy3-labelled telomeric probe (100 ng), unlabelled

233 sonicated male gDNA (3 µg), and sonicated salmon sperm DNA (25 µg). The preparations

234 were counterstained with 0.5 mg/mL DAPI and mounted in antifade based on DABCO (Sigma-

235 Aldrich).

236

237 FISH with biotin-labelled probes

238 Fragments of 18S rDNA were generated from the codling moth (Cydia pomonella) gDNA using

239 PCR as described in Fuková, Nguyen & Marec (2005). The probe was labelled with biotin-16-

240 dUTP (Roche Diagnostics GmbH, Mannheim, Germany) using the Nick Translation Kit (Abbott

241 Molecular Inc.) with 1 h and 45 min incubation at 15°C. A biotin-labelled H3 histone probe was

242 prepared from Leptidea sinapis gDNA using PCR according to the protocol described in

243 Šíchová et al. (2013; 2015).

244 FISH experiments with 18S rDNA and H3 histone probes were performed as described

245 in Fuková et al. (2005). Briefly, chromosome preparations were treated with 100 µg/mL RNase

246 A for 1 h to remove the excess of rRNAs, and were denatured and hybridized with 15 ng of

247 biotinylated probe and 25 µg of sonicated salmon sperm DNA (Sigma-Aldrich) per slide.

248 Hybridization signals were detected with Cy3-conjugated streptavidin (Jackson ImmunoRes.

249 Labs. Inc., West Grove, PA, USA), amplified with biotinylated anti-streptavidin (Vector Labs.

250 Inc., Burlingame, CA, USA) and again detected with Cy3-conjugated streptavidin. The

251 preparations were counterstained with 0.5 µg/mL DAPI and mounted in antifade based on

252 DABCO.

253

254 Microscopy and image processing

255 Preparations from FISH experiments were observed under a Zeiss Axioplan 2 microscope

256 (Carl Zeiss Jena, Germany) or under a DM 6000B microscope (Leica Microsystems Japan,

257 Tokyo, Japan). Black-and-white images were recorded with a cooled F-View CCD camera

258 using the AnalySIS software, version 3.2 (Soft Imaging System GmbH, Münster, Germany),

259 installed on the Zeiss Axioplan 2 microscope or with a DFC350FX CCD camera installed on

260 the DM 6000B microscope. In all preparations, images were captured separately for each

261 fluorescent dye, pseudocoloured (light blue for DAPI, green for fluorescein and Green, red for

262 Cy3, and yellow for Orange), and superimposed with Adobe Photoshop, version 7.0.

263

264 RESULTS

265

266 Molecular identification of Leptidea specimens

267 The ML tree based on mitochondrial COI sequences confirmed the taxonomical identity of the

268 material used in this study. All analysed larvae clustered with the L. amurensis sequences

269 downloaded from GenBank, but formed a well-supported and slightly differentiated clade with

270 respect to all mainland specimens available (Figure 1), likely reflecting the geographical

271 isolation of the Japanese population of L. amurensis (minimum p-distance to the nearest

272 mainland conspecific was 0.5%). Moreover, two samples, NC_022686 and JX274648,

273 corresponding to mitochondrial genomes published as belonging to L. morsei (Hao et al.,

274 2014), were clearly recovered by our analysis as L. amurensis (as indicated by the COI

275 fragment analysed here).

276

277 Chromosome number and structure

278 Chromosome numbers of L. amurensis were determined from repeated counts of mitotic

279 metaphase chromosomes prepared from wing imaginal discs. To facilitate the identification of

280 individual chromosomes we used FISH with (TTAGG)n telomeric probes. Mitotic metaphase

281 complements showed similar characteristics in all studied larvae. The karyotypes consisted of

282 a high number of various-sized chromosomes with DAPI-positive heterochromatin blocks

283 evenly distributed throughout the whole genome (Figure 2A, B). Based on repeated counts, we

284 found differences in chromosome numbers between sexes and among individual females.

285 While seven examined male larvae had an identical diploid chromosome number of 2n=122

286 (Figure 2B), three out of six female larvae had 2n=118 and the other three female larvae had

287 2n=119 (Figure 2A). No intraindividual variability in chromosome number was found. In male

288 meiotic metaphase I (MI) we observed 61 elements, most probably bivalents (Figure 2C),

289 which corresponded to the diploid chromosome number found in male mitotic complements.

290 We also observed two MI bivalents that were slightly larger than the other bivalents (Figure

291 2C).

292 In female mitotic complements, four chromosomes stood out due to their size.

293 Moreover, two of these large chromosomes were partially differentiated with heterochromatin

294 (Figure 2A, 3B). Similarly, large chromosomes were observed in male mitotic nuclei (Figure

295 2B). However, in contrast with the two largest chromosomes in females, they were not

296 heterochromatinized and the size difference was not as pronounced. Thus, based on the

297 comparison between male and female mitotic complements we concluded that the two largest

298 heterochromatinized chromosomes in females represent two W chromosomes, W1 and W2.

299 In male and female meiotic pachytene complements we observed conspicuous

300 heterochromatin blocks highlighted with DAPI that were predominantly present at the

301 chromosomal ends in the majority of bivalents (Figure 2D; asterisks). These DAPI-positive

302 blocks were also found in other studied Leptidea species (Šíchová et al., 2015). However, they

303 were distributed throughout the whole chromosomes. Compared to other studied Leptidea

304 species, we found a low number of chromosomal rearrangements in female pachytene nuclei.

305 In total, we analysed meiotic complements from six female larvae. In all of them we observed

306 a sex chromosome multivalent (see later; Figure 2D; arrow) and in four of them we also

307 observed one extra trivalent (Figure 2D; arrowhead).

308

309 Chromosomal location of major rDNA and H3 histone genes

310 FISH with the 18S rDNA probe performed in three Leptidea amurensis larvae did not reveal

311 any differences in the number and location of rDNA sites. The probe localized one rDNA

312 cluster at the end of one small-sized pachytene bivalent (Figure 2E). In mitotic complements,

313 the probe mapped correspondingly to two small mitotic metaphase chromosomes (not shown).

314 These results clearly indicate the presence of a single bivalent bearing the nucleolar organizer

315 region (NOR). The NOR was associated with a large block of DAPI-positive heterochromatin,

316 covering about two-fifths of the NOR-bivalent length (inset of Figure 2E).

317 FISH with the H3 histone probe was performed in three L. amurensis larvae. In all

318 examined larvae we found two clusters in middle-sized pachytene bivalents, one terminal and

319 one interstitial (Figure 2F). Only the latter cluster was associated with a block of DAPI-positive

320 heterochromatin (inset of Figure 2F). In accordance with this hybridization pattern, the H3

321 probe identified four mitotic chromosomes (not shown).

322

323 Sex chromosome identification

324 To identify the sex chromosomes we used GISH combined with tel-FISH on mitotic and

325 meiotic chromosomes of Leptidea amurensis females. While GISH was expected to

326 differentiate the W chromosome(s) by strong binding of the female genomic probe, the

327 telomeric probe helped us to determine the ends of individual chromosomes. In total, the

328 analysis was performed in chromosome preparations from six female larvae. In all female

329 mitotic metaphase complements, we observed four large chromosomes (Figure 3A, B; arrows).

330 Two of these large chromosomes were preferentially labelled with the female gDNA probe

331 (Figure 3A-C; asterisks), indicating that these are the W sex chromosomes. They were the

332 largest chromosomes in the female karyotype of L. amurensis and were designated W1 and

333 W2. The W1 chromosome was homogeneously labelled over most of its length except for a

334 small terminal gap (Figure 3C). However, the W2 chromosome was strongly highlighted with

335 the probe only in a terminal segment (Figure 3C). The female-derived genomic probe also

336 hybridized to heterochromatin blocks on autosomes (Figure 3C).

337 The analysis of female pachytene complements helped us to identify a complex sex

338 chromosome multivalent with the following constitution: W1W2W3Z1Z2Z3Z4Z5Z6 (Figure 3D-H).

339 In the multivalent, each W chromosome paired with 2-4 Z chromosomes (see a scheme in

340 Figure 4). Three Z chromosomes paired each with two W chromosomes (Z1 with W1 and W2;

341 Z2 and Z3 with W2 and W3). On the contrary, three other Z chromosomes paired exclusively

342 with one W chromosome each (Z4 and Z5 with W1; Z6 with W2). These results were confirmed

343 in four out of six female larvae. In two larvae we were not able to resolve the multivalent due to

344 the lack of well-spread pachytene nuclei in chromosome preparations. In accordance with

345 results from mitotic metaphases, the female gDNA probe highlighted two large W

346 chromosomes, W1 and W2 (Figure 3D, F). These W chromosomes were also partially

347 differentiated by DAPI-positive heterochromatin (Figure 3E). Along with the strong binding of

348 the gDNA probe, this pattern indicated the accumulation of repetitive sequences and

349 transposable elements in both W chromosomes (Sahara et al., 2003). However, we found only

350 a small heterochromatin block at the terminal part of the third W chromosome (Figure 3E, the

351 lower arrowhead). The fact that the W3 sex chromosome was not largely differentiated by the

352 female genomic probe suggests its recent origin (Figure 3F). Two largest Z chromosomes

353 (Figure 3H) were designated Z1 and Z2. These Z chromosomes can represent two large

354 bivalents in male meiotic metaphase I (Figure 2C) and also two large chromosomes that were

355 not heterochromatinized in female mitotic nuclei (Figure 2A, 3A-B).

356

357 DISCUSSION

358

359 Previous studies on three cryptic Leptidea species from Western Palearctic (L. juvernica, L.

360 reali, and L. sinapis) showed exceptional inter- and intra-specific variation in chromosome

361 numbers and location of cytogenetic markers (major rDNA and H3 histone genes) and a

362 curious sex determining system with 3–4 W and 3–4 Z chromosomes (Dincă et al., 2011;

363 2013; Lukhtanov et al., 2011; Šíchová et al., 2015). These results suggested a dynamic

364 karyotype evolution and stressed the role of chromosomal rearrangements in the speciation of

365 Leptidea butterflies. The present study enabled us to extend the research and examine the

366 karyotype of L. amurensis, a wood white butterfly with Eastern Palearctic distribution,

367 previously reported as a species with a constant haploid chromosome number of n=61 (Maeki,

368 1958).

369 Wood white butterflies are generally similar to each other in external morphology, and,

370 especially for some of the species in this genus, it is often very difficult to reliably assign

371 specimens to species. In such cases, it is necessary to use several additional characters,

372 including molecular (mitochondrial and/or nuclear DNA markers), cytological (chromosome

373 number), and morphological (genitalia morphometry) data, for the precise identification of

374 closely related species (Dincă et al., 2011; Lukhtanov et al., 2011). In the Takazasu region of

375 Japan, the probability of misidentifying specimens of L. amurensis is small because no other

376 Leptidea species has been reported from this area. Nevertheless, to eliminate any potential

377 misidentification due to incomplete faunistic data or cryptic diversity we used mitochondrial

378 COI sequences that accurately proved the taxonomical identity of the larvae used in this study.

379

380 Chromosome number variation

381

382 The first note about the chromosome number of L. amurensis dates back to 1958,

383 when Maeki analysed meiotic chromosomes in the stage of metaphase I (MI) spermatocytes.

384 He observed 61 elements in meiotic nuclei and, most importantly, described two elements that

385 stood out due to their size. Unfortunately, meiotic spermatocytes do not allow the analysis of

386 complex meiotic figures such as multivalents. Taking into account the complicated structures

387 of meiotic chromosomes found in three closely related Leptidea species (Šíchová et al., 2015)

388 and the absence of data from L. amurensis females, we carried out a comparative analysis of

389 male and female mitotic metaphase complements, which allowed us to determine more

390 accurately the range of diploid chromosome numbers in L. amurensis. Our results in males

391 confirmed the findings of Maeki (1958). All studied males showed a diploid chromosome

392 number of 2n=122 in mitotic metaphase nuclei and the haploid number of n=61 elements

393 including two larger elements in metaphase I of meiotic spermatocytes. However, females

394 showed a lower number of chromosomes in mitotic metaphase, either 2n=118 or 2n=119. The

395 sex-specific difference in chromosome numbers between L. amurensis males and females

396 resulted from the unique constitution of multiple sex chromosomes: females had a total of nine

397 sex chromosomes (three W and six Z) and males had twelve sex chromosomes (six Z-

398 chromosome pairs), while the difference in L. amurensis females depended on the presence

399 or absence of a trivalent in pachytene oocytes. The trivalent indicated the occurrence of

400 chromosome fusion that reduced the chromosome number by one in heterozygous females.

401 No male heterozygotes and no homozygotes of either sex were found for the fusion.

402 Previous studies showed that fusion polymorphism involving autosomes and/or sex

403 chromosomes leads almost exclusively to the variation in chromosome numbers (Papeschi,

404 1994; Poggio et al., 2013; Yoshido et al., 2013). In species with holokinetic chromosomes,

405 where the kinetic activity is distributed along most of the chromosome surface, a fusion does

406 not dramatically alter meiotic segregation, as in the case of a monocentric chromosome that

407 may become dicentric after fusion. Especially in lepidopteran females, which have the

408 achiasmatic meiosis, meiotic trivalents often exhibit regular segregation of chromosomes and

409 generate genetically balanced gametes (Marec et al., 2001; Melters et al., 2012). Moreover, it

410 was shown that chromosomal fusions affect the frequency of recombination in both

411 monocentric and holokinetic chromosomes (Basset et al., 2006; Hipp, Rothrock & Roalson,

412 2009; Bureš & Zedek, 2014). The reduced recombination enables the accumulation of genetic

413 incompatibilities and can ultimately lead to divergence and speciation (Noor et al., 2001;

414 Rieseberg, 2001; Faria & Navarro 2010).

415 To further explore the variability of L. amurensis karyotypes, we mapped the

416 chromosomal location of two cytogenetic markers, clusters of major rDNA and H3 histone

417 genes that were also mapped by FISH in three Western Paleartic species, L. juvernica, L. reali,

418 and L. sinapis. In L. reali, the species with the lowest chromosome number of the three

419 (2n=51-55), all analysed larvae showed consistent results with a single terminal rDNA cluster

420 and an interstitial cluster of H3 genes per haploid genome. In L. juvernica and L. sinapis,

421 species with higher chromosome numbers (2n=80-91 in L. juvernica and 2n=56-106 in L.

422 sinapis), significant differences in the number and location of both cytogenetic markers were

423 observed, even among the offspring of individual females (Šíchová et al., 2015). Such inter-

424 and intra-population variation in rDNA distribution is consistent with the evolutionary mobility of

425 rDNA observed in Lepidoptera and other groups of organisms (Nguyen et al., 2010, and

426 references therein; Pucci et al., 2014). However, the variability in otherwise conserved H3

427 histone gene clusters is rather surprising and highlights the ongoing explosive karyotype

428 evolution in Leptidea species (Šíchová et al., 2015). Interestingly, all examined larvae of L.

429 amurensis exhibited a single terminal rDNA cluster and two clusters of H3 genes, one terminal

430 and one interstitial, per haploid genome. These findings, together with the low number of

431 chromosomal multivalents when compared to other Leptidea species, point to the stability of

432 the L. amurensis karyotype despite the high chromosome number. However, the analysis of

433 other L. amurensis populations should be conducted to confirm these results.

434

435 Multiple sex chromosomes

436

437 In pachytene oocytes of L. amurensis females, the multiple sex chromosomes formed a

438 complex W1W2W3Z1Z2Z3Z4Z5Z6 multivalent (Figures 3 and 4), which is even more complex than

439 those found in Western Palearctic Leptidea species (cf. Šichová et al., 2015). Three Z

440 chromosomes of the multivalent paired each with two W chromosomes, suggesting that the Z

441 chromosomes and corresponding segments of the W chromosomes originated by autosome-

442 sex-chromosome translocations. However, three other Z chromosomes paired only with one W

443 chromosome, which indicates that these components of the multivalent evolved either via

444 chromosome fission (Z chromosomes) or fusion (corresponding parts of the W chromosomes).

445 We assume that the W1 chromosome, which was the only W almost entirely highlighted by

446 GISH (see Figure 3) indicating a high level of molecular differentiation (cf. Fuková et al., 2005),

447 is largely composed of an ancestral W chromosome. This would favour the origin of Z4 and Z5

448 by fission of an ancestral Z chromosome. A very weak hybridization pattern of the W3

449 chromosome suggests a recent autosomal origin of this element, which has not yet

450 accumulated a sufficient amount of repetitive sequences to be differentiated by GISH.

451 Similarly, the multiple sex chromosome systems in three Western Palearctic Leptidea species

452 most likely originated by complex translocations between the ancestral WZ pair and several

453 autosomes (Šíchová et al., 2015). These complex changes could be facilitated by the

454 presence of transposable elements, as in the case of gene insertions in Drosophila

455 melanogaster (Jakubczak, Xiong & Eickbush, 1990), Apis mellifera and other Hymenoptera

456 (Bigot et al., 1992), or the preponderance of other repetitive sequences, which is supported by

457 the presence of evenly distributed heterochromatin blocks in the three previously studied

458 Leptidea species (Šíchová et al., 2015) and also in the genome of L. amurensis (this study).

459 By contrast to the highly complex constitution of L. amurensis multiple sex

460 chromosomes, the majority of moths and butterflies have a WZ/ZZ (female/male) chromosome

461 system of sex determination. Systems without the W chromosome (Z0/ZZ) also occur, though

462 much less frequently (Traut et al., 2007). In addition, two types of multiple sex chromosome

463 systems with three elements, W1W2Z/ZZ, and WZ1Z2/Z1Z1Z2Z2, occur sporadically. They have

464 been described only in seven genera from different lineages of the lepidopteran phylogenetic

465 tree (reviewed in Marec et al., 2010). Their origin can be ascribed either to sex-chromosome

466 fission or to sex-chromosome-autosome fusion, where the remaining autosome becomes a W2

467 or Z2 chromosome (Marec et al., 2010). However, only the latter mechanism has been clearly

468 demonstrated in Lepidoptera (Yoshido et al., 2011; Sahara, Yoshido & Traut, 2012). The so-

469 called neo-sex chromosomes, resulting from the fusion with an autosome, can evolve in both

470 the Z and W chromosomes of a particular species, if each member of an autosome pair fuses

471 with one sex chromosome. These neo-sex chromosomes can be exceptionally large in

472 comparison with the other chromosomes of the respective genome and may play an important

473 role in the evolution of large groups of Lepidoptera. Such a neo-Z chromosome originating

474 through a fusion of the ancestral Z chromosome with an autosome has been recently

475 demonstrated in the codling moth, Cydia pomonella (Tortricidae: Olethreutinae). Available data

476 suggest that this fusion happened in a common ancestor of the main tortricid subfamilies,

477 Olethreutinae and Tortricinae, and that it increased the adaptive potential of tortricids

478 contributing to their spectacular radiation (Nguyen et al., 2013). Furthermore, studies on the

479 neo-sex chromosomes in populations of Samia cynthia suggest that repeated autosome-sex

480 chromosome fusions that gave rise to neo-sex chromosomes, may accelerate the

481 accumulation of genetically based incompatibilities and ultimately contribute to the formation of

482 reproductive barriers between populations (Yoshido et al., 2011; 2013). Similarly, sex

483 chromosome multiples of L. amurensis represent a highly derived neo-sex chromosome

484 system that originated through complex chromosomal rearrangements. These rearrangements

485 increased the number of sex-linked genes and thus could have played a major role in the

486 divergence and speciation of Leptidea butterflies as in the case of above-mentioned leaf-

487 rollers of the family Tortricidae and geographic subspecies of S. cynthia.

488

489 Conclusions

490

491 Based on our data, the emerging picture of chromosome evolution in L. amurensis is that: (i)

492 the sex chromosome number is constant in both sexes, i.e. twelve sex chromosomes (Z1-6/Z1-6)

493 in males and nine sex chromosomes (W1-3/Z1-6) in females; (ii) based on the most frequent

494 diploid chromosome number of 2n=122 found in males, the modal autosome number is 110,

495 which means a total diploid chromosome number of 2n=119 in females; (iii) the autosomal

496 fusion in heterozygotes reduces the number of autosomes to 109, resulting in a total diploid

497 chromosome number of 2n=121 in males and 2n=118 in females (Figure 5). The fact that we

498 did not find heterozygous males with 2n=121 and homozygotes for the fusion in any sex can

499 be due to the low number of individuals examined or recent origin of the fusion which is not yet

500 widespread in the L. amurensis population.

501 Our study confirmed the high number of small chromosomes in the karyotype of L.

502 amurensis, one of the highest in Leptidea species and twice the ancestral number of n=31 in

503 Lepidoptera. These findings point to chromosome fission as the main force in the karyotype

504 evolution of L. amurensis. In contrast to the previously studied Western Palearctic species

505 (Šíchová et al., 2015), the karyotype of L. amurensis is relatively stable, but shows a striking

506 difference in chromosome numbers between sexes. We clearly showed that this difference

507 results from the unique constitution of multiple sex chromosomes, i.e. W1-3Z1-6 in females and

508 Z1-6Z1-6 in males. Meiotic configurations in females suggest that this system originated by

509 complex chromosomal rearrangements between ancestral sex chromosomes and autosomes,

510 including fusion, fission, and translocation events. The presence of sex chromosome multiples

511 in the karyotypes of four Leptidea species examined so far suggests that a multiple sex

512 chromosome system is an ancestral trait for all Leptidea. Our findings also support a

513 hypothesis according to which the complex rearrangements of sex chromosomes contribute to

514 the formation of reproductive barriers between the closely related Leptidea species.

515

516 Acknowledgements

517 We thank V.A. Lukhtanov and two anonymous reviewers for their helpful comments which

518 greatly improved the manuscript. This research was funded by Grant 14-22765S of the Czech

519 Science Foundation (JŠ and FM) and by Grant 23380030 of the Japan Society for the

520 Promotion of Science (KS). JŠ acknowledges the additional support by Grant 052/2013/P of

521 the Grant Agency of the University of South Bohemia (USB) and from the Development

522 Project of SBU for Internationalization. VD was supported by a Marie Curie International

523 Outgoing Fellowship within the 7th European Community Framework Programme (project no.

524 625997).

525

526 Conflict of Interest

527 Authors declare no conflict of interest.

528

529 Data archiving

530 Sequence data have been submitted to GenBank: accession numbers KR363156 - KR363165.

531

532 References 533 534 Ahola V, Lehtonen R, Somervuo P, Salmela L, Koskinen P, Rastas P, Välimäki N, Paulin 535 L, Kvist J, Wahlberg N, Tanskanen J, Hornett EA, Ferguson LC, Luo S, Cao Z, de 536 Jong MA, Duplouy A, Smolander O-P, Vogel H, McCoy RC, Qian K, Chong WS, 537 Zhang Q, Ahmad F, Haukka JK, Joshi A, Salojärvi J, Wheat CW, Grosse-Wilde E, 538 Hughes D, Katainen R, Pitkänen E, Ylinen J, Waterhouse RM, Turunen M, 539 Vähärautio A, Ojanen SP, Schulman AH, Taipale M, Lawson D, Ukkonen E, 540 Mäkinen V, Goldsmith MR, Holm LH, Auvinen P, Frilander MJ, Hanski I. 2014. The 541 Glanville fritillary genome retains an ancient karyotype and reveals selective 542 chromosomal fusions in Lepidoptera. Nature Communications 5: 4737. 543 Bardella VB, Dias A.L., Giuliano-Caetano L, Ribeiro JRI, Da Rosa R. 2012. Sex 544 chromosome differentiation in Belostoma (Insecta: Heteroptera: Belostomatidae). 545 Genetics and Molecular Research 11: 2476-2486. 546 Basset P, Yannic G, Brünner H, Hausser J. 2006. Restricted gene flow at specific parts of 547 the shrew genome in chromosomal hybrid zones. Evolution 60: 1718-1730. 548 Baxter SW, Davey JW, Johnston JS, Shelton AM, Heckel DG, Jiggins CD, Blaxter ML. 549 2011. Linkage mapping and comparative genomics using next-generation RAD 550 sequencing of a non-model organism. PLoS ONE 6: e19315.

551 Bertollo LAC, Fontes MS, Fenocchio AS, Cano J. 1997. The X1X2Y sex chromosome 552 system in the fish Hoplias malabaricus. I. G-, C- and chromosome replication banding. 553 Chromosome Research 5: 493-499. 554 Bertolotto CEV, Rodrigues MT, Yonenaga-Yassuda Y. 2001. Banding patterns, multiple sex

555 chromosome system and localization of telomeric (TTAGGG)n sequences by FISH on 556 two species of Polychrus (Squamata, Polychrotidae). Caryologia 54: 217-226. 557 Bigot Y, Lutcher F, Hamelin MH, Periquet G. 1992. The 28S ribosomal RNA-encoding gene 558 of Hymenoptera - Inserted sequences in the retrotransposon-rich regions. Gene 121: 559 347-352. 560 Blin N, Stafford DW. 1976. A general method for isolation of high molecular weight DNA from 561 eukaryotes. Nucleic Acids Research 3: 2303-2308. 562 Bureš P, Zedek F. 2014. Holokinetic drive: centromere drive in chromosomes without 563 centromeres. Evolution 68: 2412-2420. 564 Brown Jr. KS, Freitas AVL, Wahlberg N, von Schoultz B, Saura AO, Saura A. 2007. 565 Chromosomal evolution in the South American Nymphalidae. Hereditas 144: 137-148. 566 Brown Jr. KS, von Schoultz B, Suomalainen E. 2004. Chromosome evolution in Neotropical 567 Danainae and Ithomiinae (Lepidoptera). Hereditas 141: 216–36.

568 Carpenter JE, Bloem S, Marec F. 2005. Inherited sterility in insects. In: Dyck VA, Hendrichs J, 569 Robinson AS, eds. Sterile insect technique. Principles and practice in area-wide 570 integrated pest management. Dordrecht: Springer, 115-146. 571 Darriba D, Taboada GL, Doallo R, Posada D. 2012. jModelTest 2: more models, new 572 heuristics and parallel computing. Nature Methods 9: 772. 573 de Lesse H. 1960. Spéciation et variation chromosomique chez les Lépidoptères 574 Rhopalocères. Annales des sciences naturelles-zoologie et biologie animale 12: 1-223. 575 De Prins J, Saitoh K. 2003. Karyology and sex determination. In: Kristensen NP, ed. 576 Lepidoptera, moths and butterflies: morphology, physiology, and development. Berlin: 577 Walter de Gruyter, 449-468.

578 del Cerro AL, Cuñado N, Santos JL. 1998. Synaptonemal complex analysis of the X1X2Y 579 trivalent in Mantis religiosa L. males: inferences on the origin and maintenance of the 580 sex-determining mechanism. Chromosome Research 6: 5-11. 581 Dincă V, Lukhtanov VA, Talavera G, Vila R. 2011. Unexpected layers of cryptic diversity in 582 wood white Leptidea butterflies. Nature Communication 2: 324. 583 Dincă V, Wiklund C, Lukhtanov VA, Kodandaramaiah U, Norén K, Dapporto L, Wahlberg 584 N, Vila R, Friberg M. 2013. Reproductive isolation and patterns of genetic differentiation 585 in a cryptic butterfly species complex. Journal of Evolutionary Biology 26: 2095-2106. 586 Faria R, Navarro A. 2010. Chromosomal speciation revisited: rearranging theory with pieces 587 of evidence. Trends in Ecology & Evolution 25: 660-669. 588 Felsenstein J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. 589 Evolution 39: 783-791. 590 Fuková I, Nguyen P, Marec F. 2005. Codling moth cytogenetics: karyotype, chromosomal 591 location of rDNA, and molecular differentiation of sex chromosomes. Genome 48: 1083- 592 1092. 593 Galián J, Proença SJR, Vogler AP. 2007. Evolutionary dynamics of autosomal-heterosomal 594 rearrangements in a multiple-X chromosome system of tiger beetles (Cicindelidae). BMC 595 Evolutionary Biology 7: 158. 596 Grützner F, Rens W, Tsend-Ayush E, El-Mogharbel N, O'Brien PC, Jones RC, Ferguson- 597 Smith MA, Marshall Graves JA. 2004. In the platypus a meiotic chain of ten sex 598 chromosomes shares genes with the bird Z and mammal X chromosomes. Nature 432: 599 913-917. 600 Gruetzner F, Ashley T, Rowell DM, Graves JAM. 2006. How did the platypus get its sex 601 chromosome chain? A comparison of meiotic multiples and sex chromosomes in plants 602 and animals. Chromosoma 115: 75-88. 603 Guindon S, Gascuel O. 2003. A simple, fast and accurate method to estimate large 604 phylogenies by maximum-likelihood. Systematic Biology 52: 696-704.

605 Hao JJ, Hao JS, Sun XY, Zhang LL, Yang Q. 2014. The complete mitochondrial genomes of 606 the Fenton's wood white, Leptidea morsei, and the lemon emigrant, Catopsilia pomona. 607 Journal of Insect Science 14: 130. 608 Hipp AL, Rothrock PE, Roalson EH. 2009. The evolution of chromosome arrangements in 609 Carex (Cyperaceae). Botanical Review 75: 96-109. 610 Jakubczak JL, Xiong Y, Eickbush TH. 1990. Type-I (R1) and Type-II (R2) ribosomal DNA 611 insertions of Drosophila melanogaster are retrotransposable elements closely related to 612 those of Bombyx mori. Journal of Molecular Biology 212: 37-52. 613 Kandul NP, Lukhtanov VA, Dantchenko AV, Coleman JW, Sekercioglu CH, Haig D, 614 Pierce NE. 2004. Phylogeny of Agrodiaetus Hübner 1822 (Lepidoptera: Lycaenidae) 615 inferred from mtDNA sequences of COI and COII and nuclear sequences of EF1-α: 616 karyotype diversification and species radiation. Systematic Biology 53: 278-298.

617 Kaur H, Gaba K. 2015. First report on a multiple sex chromosome system (X1X2X30) and 618 population variations in the frequency of ring bivalents in Pyrrhocoridae (Hemiptera: 619 Heteroptera). European Journal of Entomology 112: 419-425. 620 Král J. 2007. Evolution of multiple sex chromosomes in the spider genus Malthonica 621 (Araneae: Agelenidae) indicates unique structure of the spider sex chromosome systems. 622 Chromosome Research 15: 863-879. 623 Král J, Kořínková T, Forman M, Krkavcová L. 2011. Insights into the meiotic behavior and 624 evolution of multiple sex chromosome systems in spiders. Cytogenetic and Genome 625 Research 133: 43-66. 626 Král J, Kořínková T, Krkavcová L, Musilová J, Forman M, Herrera IMÁ, Haddad CR, 627 Vítková M, Henriques S, Vargas JGP, Hedin M. (2013). Evolution of karyotype, sex 628 chromosomes, and meiosis in mygalomorph spiders (Araneae: Mygalomorphae). 629 Biological Journal of Linnean Society 109: 377-408. 630 Lorković Z. 1941. Die Chromosomenzahlen in der Spermatogenese der Tagfalter. 631 Chromosoma 2: 155-191. 632 Lukhtanov VA. 1991. Evolution of the karyotype and system of higher taxa of the Pieridae 633 (Lepidoptera) of the world fauna. Entomologicheskoe Obozrenie 703: 619-641. 634 Lukhtanov VA. 2000. Sex chromatin and sex chromosome systems in nonditrysian 635 Lepidoptera (Insecta). Journal of Zoological Systematics and Evolutionary Research 38: 636 73-79. 637 Lukhtanov VA. 2015. The blue butterfly Polyommatus (Plebicula) atlanticus (Lepidoptera, 638 Lycaenidae) holds the record of the highest number of chromosomes in the non- 639 polyploid eukaryotic organisms. Comparative Cytogenetics 9: 683-690.

640 Lukhtanov VA, Dincă V, Talavera G, Vila R. 2011. Unprecedented within-species 641 chromosome number cline in the Wood White butterfly Leptidea sinapis and its 642 significance for karyotype evolution and speciation. BMC Evolutionary Biology 11: 109. 643 Lukhtanov VA, Kandul NP, Plotkin JB, Dantchenko AV, Haig D, Pierce NE. 2005. 644 Reinforcement of pre-zygotic isolation and karyotype evolution in Agrodiaetus butterflies. 645 Nature 436: 385-389. 646 Maeki K. 1958. On the cytotaxonomical relationship in Leptidea (Lepidoptera-Rhopalocera). 647 Japanese Journal of Genetics 33: 283-285. 648 Marec F, Sahara K, Traut W. 2010. Rise and fall of the W chromosome in Lepidoptera. In: 649 Goldsmith MR, Marec F, eds. Molecular biology and genetics of the Lepidoptera. Boca 650 Raton: CRC Press, 49-63. 651 Marec F, Tothová A, Sahara K, Traut W. 2001. Meiotic pairing of sex chromosome 652 fragments and its relation to atypical transmission of a sex-linked marker in Ephestia 653 kuehniella (Insecta: Lepidoptera). Heredity 87: 659-671. 654 Mediouni J, Fuková I, Frydrychová R, Dhouibi MH, Marec F. 2004. Karyotype, sex 655 chromatin and sex chromosome differentiation in the carob moth, Ectomyelois 656 ceratoniae (Lepidoptera: Pyralidae). Caryologia 57: 184-194. 657 Melters DP, Paliulis LV, Korf IF, Chan SWL. 2012. Holocentric chromosomes: convergent 658 evolution, meiotic adaptations and genomic analysis. Chromosome Research 20: 579- 659 93. 660 Nguyen P, Sahara K, Yoshido A, Marec F. 2010. Evolutionary dynamics of rDNA clusters on 661 chromosomes of moths and butterflies (Lepidoptera). Genetica 138: 343-354. 662 Nguyen P, Sýkorová M, Šíchová J, Kůta V, Dalíková M, Čapková Frydrychová R, Neven 663 LG, Sahara K, Marec F. 2013. Neo-sex chromosomes and adaptive potential in tortricid 664 pests. Proceedings of the National Academy of Sciences of the United States of America 665 110: 6931-6936. 666 Noor MA, Grams KL, Bertucci LA, Reiland J. 2001. Chromosomal inversions and the 667 reproductive isolation of species. Proceedings of the National Academy of Sciences of 668 the United States of America 98: 12084-12088. 669 Papeschi AG. 1994. Chromosome rearrangements in Belostoma plebejum (Stål) 670 (Belostomatidae, Heteroptera). Caryologia 47: 223-231. 671 Palacios-Gimenez OM, Castillo ER, Martí DA, Cabral-de-Mello DC. 2013. Tracking the 672 evolution of sex chromosome systems in Melanoplinae grasshoppers through 673 chromosomal mapping of repetitive DNA sequences. BMC Evolutionary Biology 13: 167. 674 Poggio MG, Gaspe MS, Papeschi AG, Bressa MJ. 2013. Cytogenetic study in a mutant of 675 Triatoma infestans (Hemiptera: Reduviidae) carrying a spontaneous autosomal fusion 676 and an extra chromosome. Cytogenetic and Genome Research 139: 44-51.

677 Pokorná M, Altmanová M, Kratochvíl L. 2014. Multiple sex chromosomes in the light of 678 female meiotic drive in amniote vertebrates. Chromosome Research 22: 35-44. 679 Pucci MB, Barbosa P, Nogaroto V, Almeida MC, Artoni RF, Pansonato-Alves JC, Foresti 680 F, Moreira-Filho O, Vicari MR. 2014. Population differentiation and speciation in the 681 genus Characidium (Characiformes: Crenuchidae): effects of reproductive and 682 chromosomal barriers. Biological Journal of the Linnean Society 111: 541-553. 683 Pringle EG, Baxter SW, Webster CL, Papanicolaou A, Lee SF, Jiggins CD. 2007. Synteny 684 and chromosome evolution in the Lepidoptera: Evidence from mapping in Heliconius 685 melpomene. Genetics 177: 417-426. 686 Rens W, Grützner F, O’Brien PC, Fairclough H, Graves JA, Ferguson-Smith MA. 2004. 687 Resolution and evolution of the duck-billed platypus karyotype with an

688 X1Y1X2Y2X3Y3X4Y4X5Y5 male sex chromosome constitution. Proceedings of the National 689 Academy of Sciences of the United States of America 101: 16257-16261. 690 Rieseberg LH. 2001. Chromosomal rearrangements and speciation. Trends in Ecology & 691 Evolution 16: 351-358. 692 Robinson R. 1971. Lepidoptera Genetics. Oxford: Pergamon Press. 693 Sahara K, Marec F, Eickhoff U, Traut W. 2003. Moth sex chromatin probed by comparative 694 genomic hybridization (CGH). Genome 46: 339-342. 695 Sahara K, Marec F, Traut W. 1999. TTAGG telomeric repeats in chromosomes of some 696 insects and other arthropods. Chromosome Research 7: 449-460. 697 Sahara K, Yoshido A, Marec F, Fuková I, Zhang H-B, Wu C-C, Goldsmith MR, Yasukochi 698 Y. 2007. Conserved synteny of genes between chromosome 15 of Bombyx mori and a 699 chromosome of Manduca sexta shown by five-color BAC-FISH. Genome 50: 1061-1065. 700 Sahara K, Yoshido A, Shibata F, Fujikawa-Kojima N, Okabe T, Tanaka-Okuyama M, 701 Yasukochi Y. 2013. FISH identification of Helicoverpa armigera and Mamestra 702 brassicae chromosomes by BAC and fosmid probes. Insect Biochemistry and Molecular 703 Biology 43: 644-653. 704 Sahara K, Yoshido A, Traut W. 2012. Sex chromosome evolution in moths and butterflies. 705 Chromosome Research 20: 83-94. 706 Solovyev VI, Ilinsky Y, Kosterin OE. 2015. Genetic integrity of four species of Leptidea 707 (Pieridae, Lepidoptera) as sampled in sympatry in West Siberia. Comparative 708 Cytogenetics 9: 299-324. 709 Suomalainen E. 1969. Chromosome evolution in the Lepidoptera. Chromosomes Today 2: 710 132-138. 711 Syren RM, Luykx P. 1981. Geographic variation of sex-linked translocation heterozygosity in 712 the termite Kalotermes approximatus snyder (Insecta: Isoptera). Chromosoma 82: 65-88.

713 Šíchová J, Nguyen P, Dalíková M, Marec F. 2013. Chromosomal evolution in tortricid moths: 714 conserved karyotypes with diverged features. PLoS ONE 8: e64520. 715 Šíchová J, Voleníková A, Dincă V, Nguyen P, Vila R, Sahara K, Marec F. 2015. Dynamic 716 karyotype evolution and unique sex determination systems in Leptidea wood white 717 butterflies. BMC Evolutionary Biology 15: 89. 718 Traut W, Sahara K, Marec F. 2007. Sex chromosomes and sex determination in Lepidoptera. 719 Sexual Development 1: 332-346. 720 Van’t Hof AE, Nguyen P, Dalíková M, Edmonds N, Marec F, Saccheri IJ. 2013. Linkage 721 map of the peppered moth, Biston betularia (Lepidoptera, Geometridae): a model of 722 industrial melanism. Heredity 110: 283-295. 723 Vyskot B, Hobza R 2004. Gender in plants: sex chromosomes are emerging from the fog. 724 Trends in Genetics 20: 432-438. 725 Wiens D, Barlow BA. 1975. Permanent translocation heterozygosity and sex determination in 726 East African mistletoes. Science 187: 1208-1209. 727 Yasukochi Y, Ashakumary LA, Baba K, Yoshido A, Sahara K. 2006. A second-generation 728 integrated map of the silkworm reveals synteny and conserved gene order between 729 lepidopteran insects. Genetics 173: 1319-1328. 730 Yasukochi Y, Ohno M, Shibata F, Jouraku A, Nakano R, Ishikawa Y, Sahara K. 2015. A 731 FISH-based chromosome map for the European corn borer yields insights into ancient 732 chromosomal fusions in the silkworm. Heredity 110: 283-295. 733 Yasukochi Y, Tanaka-Okuyama M, Shibata F, Yoshido A, Marec F, Wu C, Zhang H, 734 Goldsmith MR, Sahara K. 2009. Extensive conserved synteny of genes between the 735 karyotypes of Manduca sexta and Bombyx mori revealed by BAC-FISH mapping. PLoS 736 ONE 4: e7465. 737 Yoshido A, Marec F, Sahara K. 2005. Resolution of sex chromosome constitution by

738 genomic in situ hybridization and fluorescence in situ hybridization with (TTAGG)n 739 telomeric probe in some species of Lepidoptera. Chromosoma 114: 193-202. 740 Yoshido A, Sahara K, Marec F, Matsuda Y. 2011. Step-by-step evolution of neo-sex 741 chromosomes in geographical populations of wild silkmoths, Samia cynthia ssp. Heredity 742 106: 614-24. 743 Yoshido A, Sahara K, Yasukochi Y. 2014. Silkmoths (Lepidoptera). In: Sharakhov IV, ed. 744 Protocols for cytogenetic mapping of arthropod genomes. Boca Raton: CRC Press, 219- 745 256. 746 Yoshido A, Šíchová J, Kubíčková S, Marec F, Sahara K. 2013. Rapid turnover of the W 747 chromosome in geographical populations of wild silkmoths, Samia cynthia ssp. 748 Chromosome Research 21: 149-64.

749 FIGURE LEGENDS 750 751 Figure 1. Maximum likelihood tree inferred from mitochondrial COI sequences of Leptidea

752 taxa. Specimens sequenced and analyzed in this study are indicated by an asterisk and were

753 combined with available sequences of other Leptidea taxa from published studies. Samples

754 with two asterisks correspond to COI sequences from mitochondrial genomes that were

755 published as belonging to L. morsei, but were recovered as L. amurensis by our analysis. For

756 L. juvernica, L. reali, and L. sinapis representatives of all haplotypes identified by Dincă et al.

757 (2013) were used. For the origin of all specimens and GenBank accession numbers, see

758 Supplementary Table S1. The scale represents 0.02 substitutions per site. Bootstrap supports

759 (≥ 50, 100 replicates) are shown next to the recovered nodes.

760

761 Figure 2. Karyotype analysis of mitotic and meiotic chromosomes (A-D) and FISH localization

762 of rDNA (E) and H3 histone gene (F) clusters in spread chromosome preparations of Leptidea

763 amurensis larvae. Hybridization signals of the Cy3-dUTP- (red) and Orange-labelled (yellow)

764 (TTAGG)n telomeric probe indicate chromosome ends in (A, B, D), and hybridization signals of

765 the 18S rDNA and H3 histone probe (red) are marked with arrows in (E, F). Chromosomes

766 were counterstained with DAPI (blue). Asterisks indicate DAPI-positive blocks of

767 heterochromatin. (A) Female mitotic metaphase (2n=119); arrows indicate four largest

768 chromosomes in complement. (B) Male mitotic metaphase (2n=122); arrows indicate four

769 large chromosomes. (C) Male metaphase I (MI) complement (n=61); arrows indicate two large

770 chromosomal elements. (D) Female pachytene complement; the arrow indicates a sex

771 chromosome multivalent; the arrowhead indicates a trivalent. (E) Female pachytene

772 complement with a large terminal rDNA cluster (arrow); the inset in the upper right corner

773 shows the pachytene NOR-bivalent with a large terminal DAPI-positive block of

774 heterochromatin. (F) Female pachytene complement with one terminal and one interstitial H3

775 histone gene cluster (arrows); the inset in the upper right corner shows a block of DAPI-

776 positive heterochromatin in one H3 histone gene-bearing bivalent. Scale bar (A-C) = 5 µm; (D-

777 F) = 10 µm.

778

779 Figure 3. Genomic in situ hybridization (GISH) combined with (TTAGG)n telomeric probe in

780 pachytene oocytes of Leptidea amurensis females. Female-derived genomic probes were

781 labelled with fluorescein-12-dUTP or Green-dUTP (green), and the telomeric probe with Cy3-

782 dUTP (red). Chromosomes were counterstained with DAPI (blue). The arrows indicate four

783 large chromosomes from the complement. Asterisks show two W chromosomes, W1 and W2,

784 which are more intensely highlighted with the female genomic probe. Arrowheads indicate

785 DAPI-positive heterochromatin segments of the W sex chromosomes. Figures (A-C) show

786 detailed analysis of female mitotic metaphase: (A) merged images of the female-derived

787 genomic probe and DAPI staining; (B) DAPI image; (C) hybridization pattern of the female

788 genomic probe. Figures (D-H) show detailed analysis of the sex chromosome multivalent W1-

789 3Z1-6: (D) merged images of the female genomic probe, (TTAGG)n telomeric probe, and DAPI

790 staining; (E) DAPI image; (F) hybridization pattern of the female genomic probe; (G)

791 hybridization pattern of the (TTAGG)n telomeric probe; (H) schematic drawing of the sex

792 chromosome multivalent; red dots indicate the ends of individual elements in the multivalent.

793 Scale bar = 10 µm.

794

795 Figure 4. Simplified schematic drawing of the Leptidea amurensis sex chromosome

796 multivalent in pachytene oocytes with the constitution W1W2W3Z1Z2Z3Z4Z5Z6. In the multivalent,

797 the W1, W2, W3, Z1 and Z2 chromosomes are designated according to their morphological and

798 labelling characteristics (i.e. with GISH W1 was homogeneously stained over most of its length,

799 W2 was preferentially labelled at one chromosomal end, and W3 was poorly differentiated; Z1

800 and Z2 were the largest Z chromosomes from the multivalent). The Z3, Z4, Z5, and Z6

801 chromosomes are named arbitrarily.

802

803 Figure 5. Schematic drawing of a chromosomal fusion reducing the number of autosomes in

804 Leptidea amurensis. The sex chromosome number appears to be constant in each sex and

805 consists of 12 sex chromosomes (Z1-6/ Z1-6) in males and 9 sex chromosomes (W1-3/Z1-6) in

806 females. The original autosome number is 110, resulting in a total diploid chromosome number

807 of 2n=122 in males and 2n=119 in females. In fusion heterozygotes, the autosome number is

808 reduced to 109, resulting in a total diploid chromosome number of 2n=121 in males and

809 2n=118 in females. Individuals that are homozygous for the fusion with 108 autosomes should

810 also occur in wild populations (although not found in the present study).

811

812 Supplementary Table S1. List of Leptidea specimens included in the DNA analyses.

813 Sequences obtained in this study are in blue, while the other sequences were downloaded

814 from GenBank. Because a large number of sequences of L. juvernica, L. reali, and L. sinapis

815 are available in GenBank, we used only representatives for all unique COI haplotypes of these

816 three species identified by Dincă et al. (2013), as well as the specimens from Solovyev et al.

817 (2015). The haplotype numbers for these species correspond to those in Dincă et al. (2013).

818 Samples NC_022686 and JX274648 appear as mitogenomes of L. morsei in GenBank, but

819 our COI analyses recovered them as L. amurensis.