sign in sign up
<<
Home , Angaria (gastropod), Angaria delphinus, Areneidae, Bolma rugosa, Calliostomatidae, Calliostoma zizyphinum

1 Revised version: November 21st, 2016

2

3

4 Denser mitogenomic sampling improves resolution of the phylogeny of the

5 superfamily (: )

6

7

8 Juan E. Uribe1, Suzanne T. Williams2, José Templado1, Samuel

9 Abalde1, and Rafael Zardoya1*

10

11 1Museo Nacional de Ciencias Naturales (MNCN-CSIC), José Gutiérrez

12 Abascal 2, 28006, Madrid, Spain

13 2Department of Life Sciences, Natural History Museum, Cromwell Rd,

14 London SW7 5BD, UK

15

16

17

18

19

20

21 *Correspondence: R. Zardoya; email: [email protected]

22 23 ABSTRACT

24 The great morphological and ecological diversity within the superfamily Trochoidea s.l.

25 (Gastropoda: Vetigastropoda) has in the past hindered the reconstruction of a robust

26 phylogeny for the group based on morphology. Moreover, previous molecular

27 phylogenies disagreed on the monophyly and internal relationships of Trochoidea s.l.,

28 as well as on its relative phylogenetic position within Vetigastropoda. In order to further

29 resolve the trochoidean and vetigastropod phylogenetic trees, we considerably increased

30 the representation of trochoidean families for which no previous mitochondrial (mt)

31 genomes were available: the complete mt genome of pica () and the

32 nearly complete mt genomes of virgatus (Tegulidae), umbilicaris

33 (), and vorticiferus () were sequenced. In addition,

34 the nucleotide sequences of all protein coding and rRNA genes of

35 margaritarius (Trochidae) and of zizyphinum () were

36 derived from transcriptomic sequence data. The reconstructed phylogenetic trees using

37 probabilistic methods and Neomphalina as outgroup recovered with maximal support a

38 Trochoidea sensu Hickman & McLean, 1990 clade that included superfamilies

39 Angarioidea and Phasianelloidea deeply nested within superfamily Trochoidea sensu

40 Williams (2012). The families Trochidae and Calliostomatidae were the sister group to

41 the remaining trochoidean lineages. Of these, the family Margaritidae was sister to a

42 clade including Phasianelloidea + Angarioidea and + Tegulidae, this latter

43 family being paraphyletic (Cittarium and Tectus need to be assigned to a new family).

44 Gene order within newly determined mt genomes was very stable (with only few

45 rearrangements restricted to tRNA genes) and conformed to the vetigastropod and

46 gastropod consensus genome organizations.

47

48 Keys words: Mitogenomic phylogeny, rearrangement, Vetigastropoda, Trochoidea,

49 Trochidae, Calliostomatidae, Margaritidae, Cittarium, Tectus.

50 51 INTRODUCTION

52 Trochoidea s.l. Rafinesque, 1815 (top shells, turban shells, and allies) is one of the

53 most ecologically and morphologically diverse lineage of marine gastropods and by far

54 the largest superfamily belonging to the subclass Vetigastropoda, with more than 2,000

55 living grouped into about 500 recognized genera (Hickman, 1996; Geiger,

56 Nützel & Sasaki, 2008). The clade is distributed worldwide and is present throughout

57 all seas and , at all latitudes and bathymetric ranges (Hickman & McLean, 1990;

58 Williams, Karube & Ozawa, 2008). Trochoideans play an important ecological role as a

59 predominant element in different marine communities such as intertidal rocky shores,

60 seagrass beds, or coral reefs, and they are also found in many other marine habitats

61 (Williams et al., 2008). They have a long fossil record that goes back to the Middle

62 , 228-245 million years ago, but the time of the origin of the group is certainly

63 much older (Hickman & McLean, 1990; Williams et al., 2008).

64 The taxonomic internal classification of Trochoidea has a long history of

65 controversy and instability. In their comprehensive morphological monograph on

66 trochacean gastropods, Hickman & McLean (1990) maintained the three families

67 traditionally recognized within the superfamily i.e., Trochidae, Turbinidae and

68 , and organized the different genera into various subfamilies and tribes based

69 on suites of shared morphological characters. Later, in the taxonomic classification of

70 gastropods proposed by Bouchet et al. (2005), the family Turbinidae (including the

71 subfamily Skeneinae) was classified within the superfamily . However,

72 major changes to the systematics of Trochoidea were based on recent molecular

73 phylogenies (Geiger & Thacker, 2005; Williams & Ozawa, 2006; Kano, 2008; Williams

74 et al., 2008; Williams, 2012), which challenged the monophyly of the superfamily as

75 well as of several of the internal groups as defined by Hickman & McLean (1990), and 76 prompted for important changes to the taxon composition and arrangement of families

77 (Williams, 2012). For instance, some taxa were transferred to the superfamily

78 Seguenzioidea (Verrill, 1884), newly redefined by Kano (2008), and a number of

79 minute skeneimorph genera were variously relocated either to Seguenzoidea (Kano,

80 Chikyu & Warén, 2009; Haszprunar et al., 2016), Neomphalina (Kunze et al., 2008), or

81 to the new family Crosseolidae of uncertain taxonomic position (Hickman, 2013).

82 Furthermore, several molecular studies redefined the family Turbinidae (Williams &

83 Ozawa, 2006), reinterpreted the superfamilies Angarioidea and Phasianelloidea

84 (Williams et al., 2008), and restricted Trochoidea to the families Trochidae, Turbinidae,

85 , Calliostomatidae, , Skeneidae, Margaritidae and Tegulidae

86 (Williams, 2012).

87 None of these taxonomic changes was definitive and the debates over the final

88 composition and internal phylogenetic relationships of Trochoidea remain more alive

89 than ever. Moreover, this question is directly related to resolving phylogenetic

90 relationships among the different superfamilies of Vetigastropoda. In this regard, some

91 studies recovered Phasianelloidea and/ or Angarioidea in early-branching positions of

92 the Vetigastropoda tree after the divergence of Pleurotomarioidea ((Williams & Ozawa,

93 2006; Kano, 2008; Williams et al., 2008; Aktipis & Giribet, 2012) whereas several

94 recent phylogenies grouped Phasianelloidea and/ or Angarioidea with Trochoidea

95 (Zapata et al., 2014; Uribe et al., 2016; Lee et al. 2017; Wort, Fenberg & Williams,

96 2017). While earlier studies were based on few partial mitochondrial and nuclear genes

97 and a rather extensive lineage representation, later ones were based on phylogenomic

98 data but with reduced taxon sampling.

99 Phylogenetic analysis of complete mitochondrial (mt) genomes resulted in good

100 resolution among vetigastropod superfamilies (e.g. Uribe et al., 2016) and therefore, 101 they are good candidates to resolve phylogenetic relationships within Trochoidea. Until

102 recently, there were available 22 complete or near-complete mt genomes of

103 Vetigastropoda, which represent the living superfamilies Fissurelloidea,

104 Lepetodriloidea, Seguenzioidea, Haliotoidea, Angarioidea, Phasianelloidea, and

105 Trochoidea (no mt genome has been sequenced for Pleurotomarioidea and

106 Lepetelloidea). However, the great diversity of Trochoidea was clearly

107 underrepresented, as mt genomes for only 12 species belonging to families Turbinidae,

108 Trochidae, and Tegulidae had been published (Uribe et al., 2016; Lee et al. 2017; Wort

109 et al. 2017). Here, we increased the number of complete mt genomes representing

110 different families within Trochoidea to test the monophyly and address internal

111 phylogenetic relationships of the superfamily (in particular the relative positions of

112 families Trochidae, Calliostomatidae, and Margaritidae plus of the genera Tectus and

113 Cittarium , of which onlyTrochidae was previously included in phylogenetic analyses),

114 as well as to resolve its relative phylogenetic position within Vetigastropoda. In

115 addition, the reconstructed phylogeny was used to determine whether trochoidean mt

116 genomes show rearrangements in their genes orders. During the review process of this

117 paper, Lee et al. (2016) published a related mitogenomic phylogenetic study, which

118 complemented our taxon sampling and enriched it at the family level. Therefore, the mt

119 genomes reported by Lee et al. (2016) were incorporated into our phylogenetic

120 analyses.

121

122 MATERIALS AND METHODS

123 Samples and DNA/ RNA extraction

124 One specimen each of (Tegulidae), Tectus virgatus (Tegulidae),

125 Gibbula umbilicaris (Trochidae), Clanculus margaritarius (Trochidae), Calliostoma 126 zizyphinum (Calliostomatidae), and Margarites vorticiferus (Margaritidae) was used for

127 this study (See Table 1, for details on the locality, collector, and voucher ID of each

128 sample; family assignment was based on WoRMS: accessed October 2016, Gofas,

129 2009). Samples of C. pica, T. virgatus, G. umbilicaris, and M. vorticiferus, were stored

130 in 100% ethanol at -20 ºC, and total genomic DNA was isolated from up to 30 mg of

131 foot tissue following a standard phenol chloroform extraction.

132 Samples of C. margaritarius and C. zizyphinum were stored in RNALater at -80

133 ºC, and total RNA was isolated from mantle tissue using the RNeasy Fibrous Tissue

134 Mini Kit (Qiagen) according to the manufacturer’s instructions. Total RNA was

135 quantified and its integrity assessed using a Qubit® 2.0 Fluorometer RNA assay kit and

136 an Agilent 2200 Tapestation with a high sensitivity R6K Screen Tape, respectively.

137 Dynabeads® mRNA DIRECT™ Micro Kit (Ambion, Life Technologies) were used to

138 isolate mRNA using the 100ng-1µg total RNA protocol.

139

140 PCR amplification and sequencing

141 Two alternative strategies were carried out to obtain mitogenomic sequence data.

142 For C. pica, T. virgatus, G. umbilicaris, and M. vorticiferus, complete or near-complete

143 mt genomes were PCR amplified and sequenced, whereas for C. margaritarius and C.

144 zizyphinum, transcriptomic sequence data was generated and mt protein-coding and

145 rRNA genes were identified.

146 For obtaining complete or near-complete mt genomes from genomic DNA a

147 three-step strategy was used. First, fragments of cox1, rrnL, and cox3 genes were

148 amplified using the primers respectively detailed in Folmer et al. (1994), Palumbi et al.

149 (1991), and Boore & Brown (2000). The standard PCR reactions contained 2.5 µl of

150 10x buffer, 1.5 µl of MgCL2 (25 mM), 0.5 µl of dNTPs (2.5 mM each), 0.5 µl of each 151 primer (10 mM), 0.5-1 µl (20-100 ng) of template DNA, 0.2 µl of Taq DNA polymerase

152 5PRIME (Hamburg, Germany), and sterilized distilled water up to 25 µl. The PCR

153 temperature and cycle conditions used were: a denaturalization step at 94° C for 60 s;

154 45 cycles of denaturalization at 94° C 30 s, annealing at 44° C (cox1) or 52° C (rrnL

155 and cox3) for 60 s and extension at 72° C for 90 s; a final extension step at 72° C for 5

156 min. Second, the amplified PCR fragments were sequenced using Sanger sequencing,

157 and new primers were designed (see Suppl. Mat. for primer sequences) for amplifying

158 outwards from the short fragments in the next step. Third, the remaining mtDNA was

159 amplified in two-three overlapping fragments by long PCR using the newly designed

160 primers. The long PCR reaction contained 2.5 µl of 10x LA Buffer II (Mg2+ plus), 3 µl

161 of dNTPs (2.5 mM each), 0.5 µl of each primer (10 mM), 0,5-1 µl (20-100 ng) of

162 template DNA, 0.2 µl TaKaRa LA Taq DNA polymerase (5 units/µl), and sterilized

163 distilled water up to 25 µl. The following PCR conditions were used: a denaturing step

164 at 94° C for 60 s; 45 cycles of denaturation at 98° C for 10 s, annealing at 53° C for 30 s

165 and extension at 68° C for 60 s per kb; and a final extension step at 68 °C for 12 min.

166 Long-PCR products were purified by ethanol precipitation. Overlapping

167 fragments from the same mt genome were pooled together in equimolar concentrations

168 and subjected to massive parallel sequencing. For each mt genome, a separate indexed

169 library was constructed using the NEXTERA XT DNA library prep Kit (Illumina, San

170 Diego, CA, USA) and sequenced in a single lane of Illumina MiSeq V2 500 at Sistemas

171 Genómicos (Valencia, Spain).

172 Transcriptomes of C. margaritarius and C. zizyphinum were sequenced using the

173 following procedure: Illumina libraries were prepared for each transcriptome using the

174 ScriptSeq™ v2 RNA-Seq Library Preparation Kit from Epicentre (Epicentre

175 Biotechnologies, Madison, WI, USA), size checked with an Agilent 2200 Tapestation 176 and quantified using qPCR. The libraries were loaded onto one fifth of a MiSeq V2 500

177 cycle sequencing run. Each library was run twice.

178

179 Assembly and annotation

180 The reads corresponding to the different PCR amplified mt genomes were sorted

181 using the corresponding library indices. Adapter sequences were removed using

182 SeqPrep (St John, 2011). Assembly was performed using the TRUFA webserver

183 (Kornobis et al., 2015). The quality (randomness) of the sequencing was checked using

184 FastQC v.0.10.1 (Andrews, 2010). Reads were trimmed and filtered out according to

185 their quality scores using PRINSEQ v.0.20.3 (Schmieder & Edwards, 2011). Filtered

186 reads were used for de novo assembly of mt genomes, searching for contigs with a

187 minimum length of 3 kb. The complete or nearly complete sequence of each mt genome

188 was finally assembled by overlapping the various contigs in Sequencher 5.0.1. The

189 assembled sequence was used as reference to map the original (raw) reads with a

190 minimum identity of 99% using Geneious® 8.0.3 to estimate coverage.

191 Genome annotation was performed by setting a limit of nucleotide identity of

192 75% to previously reported vetigatropod mt genomes (Uribe et al., 2016) using

193 Geneious® 8.0.3. The annotated 13 mt protein-coding genes were further corroborated

194 by identifying the corresponding open reading frames using the invertebrate

195 mitochondrial code. The transfer RNA (tRNA) genes were further identified with

196 tRNAscan-SE 1.21 (Schattner, Brooks & Lowe, 2005), which infers cloverleaf

197 secondary structures. The ribosomal RNA (rRNA) genes were identified by sequence

198 comparison with previously reported vetigastropod mt genomes, and assumed to extend

199 to the boundaries of adjacent genes (Boore, Macey & Medina, 2005). GenBank

200 accession numbers of each newly sequenced mt genome are provided in Table 1. 201 Transcriptomes of C. margaritarius, and C. zizyphinum were assembled with

202 Galaxy (Giardine et al., 2005; Blankenberg et al., 2010; Goecks, Nekrutenko & Taylor,

203 2010) as outlined in Williams et al. (in review). Reads for the two separate sequencing

204 runs were concatenated and filtered for reads that contained all but three identical bases.

205 Trimmomatic (Lohse et al., 2012) was then used with the initial ILLUMINACLIP step

206 and a sliding window to trim reads (averaging across four bases and requiring an

207 average quality score of 24), and to remove all reads with a length of less than 30 bases.

208 Transcriptome assembly was performed using Trinity (Grabherr et al., 2011), with

209 default settings and a minimum contig length of 200 bases. Open reading frames were

210 identified using the program TransDecoder (Haas et al., 2013). The mt protein coding

211 and rRNA genes of C. margaritarius, and C. zizyphinum were extracted from the

212 corresponding transcriptomes in Geneious by using published amino acid sequences for

213 each mitochondrial gene from rugosa (GenBank KT207824; Uribe et al., 2016)

214 to identify matching sequences in the dataset of assembled contigs using the tBLASTx

215 option, then the new contig was used as a reference sequence against the original reads

216 to obtain full length genes. Gene boundaries for rRNA genes were determined by

217 comparison with other vetigastropod sequences.

218

219 Sequence alignment

220 The nucleotide sequences of the 13 protein coding and two rRNA genes encoded

221 in the newly determined complete or nearly complete mt genomes were aligned each

222 separately with the corresponding orthologous sequences of all vetigastropod complete

223 or nearly complete mt genomes available at NCBI (www.ncbi.nlm.nih.gov/; see Table

224 1). The complete mt genome of Chrysomallon squamiferum (Neomphalina) was used as

225 outgroup following Uribe et al. (2016). Each protein-coding gene was aligned with 226 Translator X (Abascal, Zardoya & Telford, 2010) using the deduced amino acid

227 sequence as guide whereas rRNA genes were aligned separately using MAFFT v7

228 (Katoh & Standley, 2013) with default parameters. Ambiguously aligned positions were

229 removed using Gblocks v.0.91b (Castresana, 2000) with the following settings:

230 minimum sequence for flanking positions: 85%; maximum contiguous non-conserved

231 positions: 8; minimum block length: 10; gaps in final blocks: no. The generated single

232 alignments were concatenated using Geneious® 8.0.3.

233

234 Phylogenetic analyses

235 Phylogenetic relationships were reconstructed using Bayesian inference (BI;

236 Huelsenbeck & Ronquist, 2001) and maximum likelihood (ML; Felsenstein, 1981). BI

237 analyses were conducted using MrBayes v3.1.2 (Ronquist & Huelsenbeck, 2003) and

238 running four simultaneous Monte Carlo Markov chains (MCMC) for 10 million

239 generations, sampling every 1,000 generations, and discarding the first 25% generations

240 as burn-in (as judged by plots of ML scores and low SD of split frequencies) to prevent

241 sampling before reaching stationarity. Two independent BI runs were performed to

242 increase the chance of adequate mixing by the MCMC and to increase the chance of

243 detecting failure to converge, as determined using Tracer v1.6 (Rambaut & Drummond,

244 2007). ML analyses were conducted with RAxML v7.3.1 (Stamatakis, 2006) and

245 default parameters using the rapid hill-climbing algorithm and 10,000 bootstrap

246 pseudoreplicates.

247 The program Partition Finder (Lanfear et al., 2012) was used to select best

248 partition schemes and best-fit models of substitution according to the Bayesian

249 information criterion (BIC; Schwarz, 1978). For protein-coding genes, the partitions

250 tested were: all genes combined, all genes separated except atp6-atp8 and nad4-nad4L, 251 and genes grouped by subunits (atp, cox, cob and nad). In addition, the three above

252 partition schemes were tested considering first, second, and third codon positions

253 separated. For the mt rRNA genes, the two genes combined or separated were tested.

254

255 RESULTS AND DISCUSSION

256 Sequencing and assembly

257 The nucleotide sequences and gene arrangement of the complete mt genome of C.

258 pica and the nearly complete mt genomes of T. virgatus, G. umbilicaris, and M.

259 vorticiferus were determined (see annotation and main features in Suppl. Mat.). In

260 Tectus and Gibbula a fragment of about 3kb between rrnL and cox3 genes could not be

261 PCR amplified. In the case of Margarites, a shorter fragment of about 2kb between rrnS

262 and cox3 genes was missing (Fig. 1). In addition, the nucleotide sequences of all protein

263 coding and rRNA genes of C. margaritarius and of C. zizyphinum were derived from

264 transcriptomic sequence data. The number of reads, mean coverage, and sequence

265 length (bp) of each complete or nearly complete mt genome are: C. pica (165,292,

266 1,390x and 17,949); T. virgatus (205,498, 2,218x and 13,891); G. umbilicaris (142,074,

267 1,666x and 12,885); and M. vorticiferus (290,484, 2,858x and 15,254). The GenBank

268 accession number and coverage for each of the mitochondrial genes of C. margaritarius

269 and C. zizyphinum are shown in Suppl. Mat.

270

271 Mitochondrial genome organization

272 Genome organization could only be determined for those mt genomes that were

273 amplified by long PCR (all but C. margaritarius and C. zizyphinum). These mt

274 genomes share the same gene order with regards to the relative position of protein-

275 coding genes, and only minor changes affecting individual tRNA genes were observed 276 (Fig. 1). The consensus gene order for Trochoidea s.l. (including Phasianelloidea and

277 Angarioidea) is the same observed in Haliotoidea and Seguenzoidea but not in

278 Fissurelloidea and Lepetodriloidea (see Lee et al., 2016 and Uribe et al., 2016 for

279 further information and discussion on vetigastropod mt gene arrangements). Moreover,

280 this consensus gene order conforms to the genome arrangement of the hypothetical

281 ancestor of gastropods (Fig. 1; Uribe et al., 2016). With respect to this gastropod

282 ancestral gene order, the mt genome of T. virgatus showed a translocation of the trnQ to

283 a new relative position between cob and nad6 genes in the minor strand (Fig. 1). The mt

284 genome of G. umbilicaris had an inversion of the trnT gene from major to minor strand

285 (Fig. 1). The mt genome of M. vorticiferus showed a translocation of the trnM to a new

286 relative position between nad6 and trnP genes in the minor strand (Fig. 1). Finally, the

287 mt genomes of , Bolma, , Cittarium, , and showed

288 rearrangements affecting trnG and trnE genes, and in some instances, one or both genes

289 were missing (Fig. 1; Lee et al., 2016; Uribe et al., 2016). It is not possible to infer the

290 exact evolution of these rearrangements given that this part of the mt genome could not

291 be sequenced in Tectus, Margarites, and Gibbula, and is not available for Clanculus and

292 Calliostoma (Fig. 1). However, it is important to note that these two genes are located at

293 the end of the ancestral MCYWQGE tRNA gene cluster, and just before the

294 hypothesized control region of gastropod mt genomes, which is known to act as hotspot

295 of gene order rearrangements (Duarte, De Azeredo-Espin & Junqueira, 2008).

296

297 Phylogenetic relationships among vetigastropod superfamilies and within Trochoidea

298 s.l.

299 A molecular phylogeny of Vetigastropoda was reconstructed using probabilistic

300 methods. The final alignment was 11,475 positions long. The best partition scheme was 301 the one having all protein-coding genes combined (but with each codon position

302 analyzed separately) and the two rRNA genes combined. The best-fit model for the

303 different partitions was GTR+I+G. The ML (-lnL = 17,998.18) and BI phylogenetic

304 analyses arrived at the same topology using Neomphalina as outgroup (Fig. 2). The

305 superfamily Lepetodriloidea was recovered as sister group of the remaining

306 vetigastropds, although only with moderate statistical support (61% BP, 0.94 BPP).

307 This lineage was recovered as sister group of Seguenzoidea + Haliotoidea in previous

308 mt genome phylogenies (Lee et al., 2016; Uribe et al., 2016; Wort et al., 2017). The

309 next lineage that branched off was Fissurelloidea, whose members exhibited relatively

310 long branches (Fig. 2). Fissurelloidea has been normally recovered as sister group of the

311 remaining vetigastropod lineages in previous mt genome phylogenies, and the

312 possibility of a long-branch attraction effect by the outgroup cannot be dismissed (Lee

313 et al., 2016; Uribe et al., 2016; Wort et al., 2017). The superfamilies Seguenzoidea and

314 Haliotoidea formed a well-supported clade (84% BPP, 1 BPP), which was the sister

315 group of Trochoidea s.l. (including Phasianelloidea and Angarioidea), and this

316 relationship received relatively high support (75% BP, 1 BPP). The mt genome

317 phylogenies clearly differed from a recent phylogeny based on nuclear transcriptomic

318 data in which phylogenetic relationships within Vetigastropoda were fully resolved (all

319 nodes received maximal statistical support; Zapata et al., 2014). In Zapata et al. study,

320 Seguenzoidea were recovered as the sister group of a clade in which Lepetodriloidea

321 was sister to Lepetelloidea and Haliotoidea was sister to Trochoidea s.l. (including

322 Phasianelloidea), however no representatives of Fissurelloidea were included (Zapata et

323 al., 2014). Here, we did not incorporate a representative of the superfamily

324 Pleurotomarioidea, which in other phylogenies is placed as sister group of the remaining

325 vetigastropod lineages (Kano, 2008; Williams et al., 2008; Zapata et al., 2014) or even 326 unrelated to Vetigastropoda (Aktipis & Giribet, 2012). Other missing superfamilies

327 were Lepetelloidea and Scissurelloidea. Several molecular phylogenies based on partial

328 gene sequences recovered a close relationship between Scissurelloidea and

329 Lepetodriloidea (Yoon & Kim 2005; Williams & Ozawa 2006; Kano, 2008), whereas

330 the relative phylogenetic position of Lepetelloidea remains controversial (Aktipis &

331 Giribet, 2012)

332 The main focus of the present phylogenetic analysis was Trochoidea s.l. This

333 clade received maximal support and included Trochoidea, Phasianelloidea and

334 Angarioidea sensu Williams et al 2008 (Fig. 2). The recognition of Phasianelloidea and

335 Angarioidea as valid superfamilies different from Trochoidea sensu Hickman &

336 McLean (1991) was based on phylogenetic analyses of partial mt and nuclear genes that

337 placed these two lineages in early diverging positions in the vetigastropod tree

338 (Williams et al., 2008; Aktipis & Giribet, 2012; see also the position of Phasianelloidea

339 in Kano, 2008). However, our results are in agreement with more recent phylogenies

340 based on mt (Lee et al., 2016; Uribe et al., 2016; Wort et al., 2017) and nuclear (Zapata

341 et al., 2014) genomic data sets, which also recovered a clade grouping Trochoidea

342 together with Phasianelloidea and Angarioidea (the latter was missing in Zapata et al.,

343 2014). Interestingly, Phasianelloidea and Angarioidea show relatively long branches in

344 the mt genome phylogenies (but shorter than non-trochoidean taxa; this study; Lee et

345 al., 2016; Uribe et al., 2016), which in previous studies with different taxon sampling

346 and molecular markers (18S and 28S also showed long branches for Phasianelloidea

347 and in particular for the family of Angarioidea in Williams et al., 2008) may

348 have produced a long-branch attraction effect and the pulling of these two lineages to

349 more basal positions. The recovery of a monophyletic Trochoidea sensu Hickman &

350 McLean (1991), which is supported here, has been shown to be particular sensitive to 351 the choice of mt and nuclear genes used in the phylogenetic analyses and the use of

352 amino acids versus nucleotides (Wort et al., 2017).

353 The representation of Trochoidea in recent phylogenomic analyses has been rather

354 limited, a situation that has been addressed in the present phylogenetic analysis with the

355 inclusion of represetatives of the families Margaritidae, Trochidae and Calliostomatidae

356 as well as exemplars from the genera Tectus and Cittarium (see also Lee et al., 2016).

357 The reconstructed phylogenetic tree recovered the family Trochidae as sister group of

358 the family Calliostomatidae with maximal statistical support, and this clade was the

359 sister group of the remaining trochoidean lineages, which formed a monophyletic group

360 with high support (70% BP; 1 BPP). The family Trochidae sensu Williams et al. (2008)

361 is the largest and most diverse in terms of diet and habitat and comprise up to 10

362 subfamilies, more than 600 known species and more than 60 genera. Therefore, the

363 representation in our study is still quite incomplete with only representatives of the

364 subfamilies (Gibbula), () and Trochinae

365 (Clanculus). While Trochidae species are mostly or detritivores (Hickman &

366 McLean, 1990), the members of the family Calliostomatidae constitute an uniform

367 group of carnivorous snails that can be distinguished from Trochidae by their distinct

368 feeding adaptations, which resulted in differences in their alimentary tracts and radular

369 morphology (Hickman & McLean, 1990; Marshall, 1995). The family Margaritidae was

370 sister of a maximally supported clade including Phasianelloidea + Angarioidea and

371 Turbinidae + (paraphyletic) Tegulidae (Fig. 2). The family Margaritidae, historically

372 included as a subfamily within Trochidae, was recognized for the first time at familial

373 rank by Williams (2012). It could represent an early radiation that diverged from the

374 tropical and subtropical groups by adaptation to cold waters (high latitudes and deep

375 waters). The family Tegulidae has a long controversial taxonomic history due to the 376 unusual distribution of character states of its members. Hickman & McLean (1990)

377 retained the group as a subfamily within Trochidae, emphasizing the evolutionary

378 conservativeness of conchological characters, such as the oblique and

379 interrupted peristome of the shell and the short growing edge of the . Later,

380 Hickman (1996) suggested that Tegula and allies represented an enigmatic group

381 located somewhere between Trochidae and Turbinidae, and Williams (2012) finally

382 raised it to familial rank. Our results support Tegulidae as a distinct lineage closely

383 related to Turbinidae (Fig. 2). The reconstructed phylogeny also suggests that the

384 genera of this family should be redefined since the speciose Tegula has proved to

385 be non-monophyletic (or alternatively genera and need to be

386 assigned to Tegula). Near thirty species are grouped today within this genus (Bouchet,

387 2011) based on similarity of shell characters.

388 The recovered internal phylogenetic relationships of Trochoidea s.l. are fully

389 congruent with the five-gene tree of Williams (2012), who did not include

390 Phasianelloidea and Angarioidea in her phylogenetic analysis. In particular, it is worth

391 noting that Cittarium and Tectus (and possibly Rochia; Williams, 2012) need to be

392 assigned to a new family. Hickman & Maclean (1990) included Tectus within Trochinae

393 and Cittarium within Gibbulinae. More recently, Bouchet et al. (2005) assigned both

394 genera to Tegulidae. It was Williams (2012) who first recovered Tectus, Cittarium, and

395 Rochia as a distinct clade, although with low support. Although not formally described,

396 she suggested a familial rank for this clade pending further studies. Our results support

397 her suggestion and highlight the need to study the morphological and anatomical

398 peculiarities of these genera with respect to other trochoidean families.

399 To summarize, the recovered phylogeny prompts for a redefinition of Trochoidea

400 sensu Williams et al. (2008), supporting instead the hypothesis of Hickman & Maclean 401 (1990). However, this redefinition in order to be complete should await further

402 mitogenomic studies including missing families such as Skeneidae, Solariellidae, and

403 Liotiidae.

404

405

406 SUPPLEMENTARY MATERIAL

407 Supplementary material is available at Journal of Molluscan Studies online.

408

409 ACKNOWLEDGEMENTS

410 We thank Yasunori Kano and one anonymous reviewer for their insightful comments on

411 a previous version of the paper. We are grateful to Jesus Marco and Luis Cabellos who

412 provided access to the supercomputer Altamira at the Institute of Physics of Cantabria

413 (IFCA-CSIC), member of the Spanish Supercomputing Network, for performing

414 phylogenetic analyses. This work was supported by the Spanish Ministry of Science and

415 Innovation (CGL2010-18216 and CGL2013-45211-C2-2-P to RZ; BES-2011-051469 to

416 JEU; BES‐2014‐069575 to SA) and by funding from the NHM Department of Life

417 Sciences to STW.

418 419

420 REFERENCES

421 ABASCAL, F., ZARDOYA, R. & TELFORD, M.J. 2010. TranslatorX: multiple alignment

422 of nucleotide sequences guided by amino acid translations. Nucleic Acids Research,

423 38: W7-13.

424 AKTIPIS, S.W. & GIRIBET, G. 2012. Testing relationships among the vetigastropod

425 taxa: a molecular approach. Journal of Molluscan Studies, 78: 12-27.

426 ANDREWS, S. 2010. FastQC.

427 http://www.bioinformatics.babraham.ac.uk/projects/fastqc/.

428 BLANKENBERG, D., VON KUSTER, G., CORAOR, N., ANANDA, G., LAZARUS, R.,

429 MANGAN, M., NEKRUTENKO, A. & J., T. 2010. Galaxy: a web-based genome

430 analysis tool for experimentalists. Current Protocols in Molecular Biology, 19.10: 1-

431 21.

432 BOORE, J.L. & BROWN, W.M. 2000. Mitochondrial genomes of Galathealinum,

433 Helobdella, and Platynereis: sequence and gene arrangement comparisons Indicate

434 that Pogonophora is not a phylum and Annelida and Arthropoda are not sister taxa.

435 Molecular Biology and Evolution, 17: 87-106.

436 BOORE, J.L., MACEY, J.R. & MEDINA, M. 2005. Sequencing and comparing whole

437 mitochondrial genomes of . Methods in Enzymology, 395: 311-348.

438 BOUCHET, P., ROCROI, J.P., FRÝDA, J., HAUSDORF, B., PONDER , W.F.,

439 VALDÉS, Á. & WÁREN, A. 2005. Classification and nomenclator of gastropod

440 families. Malacologia, 47: 1-397.

441 BOUCHET, P. (2011). Tegula Lesson, 1832. In: MolluscaBase (2016). Accessed

442 through: World Register of Marine Species at

443 http://www.marinespecies.org/aphia.php?p=taxdetails&id=413467 on 2016-11-18

444 CASTRESANA, J. 2000. Selection of conserved blocks from multiple alignments for

445 their use in phylogenetic analysis. Molecular Biology and Evolution, 17: 540-552. 446 DUARTE, G.T., DE AZEREDO-ESPIN, A.M.L. & JUNQUEIRA, A.C.M. 2008. The

447 mitochondrial control region of blowflies (Diptera: Calliphoridae): a hot spot for

448 mitochondrial genome rearrangements. Journal of Medical Entomology, 45: 667.

449 FELSENSTEIN, J. 1981. Evolutionary trees from DNA sequences: A maximum

450 likelihood approach. Journal of Molecular Evolution, 17: 368-376.

451 FOLMER, O., BLACK, M., HOEH, W., LUTZ, R. & VRIJENHOEK, R. 1994. DNA

452 primers for amplification of mitochondrial cytochrome c oxidase subunit I from

453 diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology, 3:

454 294-299.

455 GEIGER, D.L., NÜTZEL, A. & SASAKI, T. 2008. Vetigastropoda. In: Phylogeny and

456 Evolution of the : (Ponder, W.F. and Lindberg, D.R., eds), pp. 297-330.

457 University of California Press, Berkeley.

458 GEIGER, D.L. & THACKER, C.E. 2005. Molecular phylogeny of Vetigastropoda

459 reveals non-monophyletic Scissurellidae, Trochoidea, and Fissurelloidea. Molluscan

460 Research, 25: 47-55.

461 GIARDINE, B., RIEMER, C., HARDISON, R.C., BURHANS, R., ELNITSKI, L., SHAH,

462 P., ZHANG, Y., BLANKENBERG, D., ALBERT, I., TAYLOR, J., MILLER, W., KENT,

463 W.J. & NEKRUTENKO, A. 2005. Galaxy: A platform for interactive large-scale

464 genome analysis. Genome Research, 15: 1451-1455.

465 GOECKS, J., NEKRUTENKO, A. & TAYLOR, J. 2010. Galaxy: a comprehensive

466 approach for supporting accessible, reproducible, and transparent computational

467 research in the life sciences. Genome Biology, 11: R86.

468 GOFAS, S. 2009. Trochoidea Rafinesque, 1815. In: MolluscaBase (2016). Accessed

469 through: World Register of Marine Species at

470 http://www.marinespecies.org/aphia.php?p=taxdetails&id=156489 on 2016-11-18

471 GRABHERR, M.G., HAAS, B.J., YASSOUR, M., LEVIN, J.Z., THOMPSON, D.A.,

472 AMIT, I., ADICONIS, X., FAN, L., RAYCHOWDHURY, R., ZENG, Q., CHEN, Z.,

473 MAUCELI, E., HACOHEN, N., GNIRKE, A., RHIND, N., DI PALMA, F., BIRREN, 474 B.W., NUSBAUM, C., LINDBLAD-TOH, K., FRIEDMAN, N. & REGEV, A. 2011. Full-

475 length transcriptome assembly from RNA-Seq data without a reference genome.

476 Nature Biotechnology, 29: 644-652.

477 HAAS, B.J., PAPANICOLAOU, A., YASSOUR, M., GRABHERR, M., BLOOD, P.D.,

478 BOWDEN, J., COUGER, M.B., ECCLES, D., LI, B., LIEBER, M., MACMANES,

479 M.D., OTT, M., ORVIS, J., POCHET, N., STROZZI, F., WEEKS, N., WESTERMAN,

480 R., WILLIAM, T., DEWEY, C.N., HENSCHEL, R., LEDUC, R.D., FRIEDMAN, N. &

481 REGEV, A. 2013. De novo transcript sequence reconstruction from RNA-seq using

482 the Trinity platform for reference generation and analysis. Nature Protocols, 8: 1494-

483 1512.

484 HASZPRUNAR, G., KUNZE, T., BRÜCKNER, M. & HES, M. 2016. Towards a sound

485 definition of Skeneidae (Mollusca, Vetigastropoda): 3D interactive anatomy of the

486 type species, serpuloides (Montagu, 1808) and comments on related taxa.

487 Organisms Diversity & Evolution: 1-19.

488 HICKMAN, C.S. 1996. Phylogeny and patterns of evolutionary radiation in trochoidean

489 gastropods. In: Origin and Evolutionary Radiation of the Mollusca: (Taylor, J.D.,

490 ed), pp. 177–198. Oxford University Press, Oxford.

491 HICKMAN, C.S. 2013. Crosseolidae, a new family of skeneiform microgastropods and

492 progress toward definition of monophyletic Skeneidae. American Malacological

493 Bulletin, 31: 1-16.

494 HICKMAN, C.S. & MCLEAN, J.H. 1990. Systematic revision and suprageneric

495 classification of trochacean gastropods. Natural History Museum of Los Angeles

496 County, Science Series, 35: 1-169.

497 HUELSENBECK, J. & RONQUIST, F. 2001. MrBayes: Bayesian inference of

498 phylogenetic trees. Bioinformatics, 17: 754-755.

499 KANO, Y. 2008. Vetigastropod phylogeny and a new concept of Seguenzioidea:

500 independent evolution of copulatory organs in the deep-sea habitats. Zoologica

501 Scripta, 37: 1-21. 502 KANO, Y., CHIKYU, E. & WARÉN, A. 2009. Morphological, ecological and molecular

503 characterization of the enigmatic planispiral snail genus Adeuomphalus

504 (Vetigastropoda: Seguenzioidea). Journal of Molluscan Studies, 75: 397-418.

505 KATOH, K. & STANDLEY, D.M. 2013. MAFFT Multiple Sequence Alignment Software

506 Version 7: Improvements in Performance and Usability. Molecular Biology and

507 Evolution, 30: 772-780.

508 KORNOBIS, E., CABELLOS, L., AGUILAR, F., FRÍAS-LÓPEZ, C., ROZAS, J.,

509 MARCO, J. & ZARDOYA, R. 2015. TRUFA: A User-Friendly Web Server for de

510 novo RNA-seq Analysis Using Cluster Computing. Evolutionary Bioinformatics, 11:

511 97-104.

512 KUNZE, T., BECK, F., BRUÅNCKNER, M., HES, M. & HASZPRUNAR, G. 2008.

513 Skeneimorph gastropods in Neomphalina and Vetigastropoda - A preliminary report.

514 Zoosymposia, 1: 119-131.

515 LANFEAR, R., CALCOTT, B., HO, S.Y.W. & GUINDON, S. 2012. PartitionFinder:

516 combined selection of partitioning schemes and substitution models for phylogenetic

517 analyses. Molecular Biology and Evolution, 29: 1695-1701.

518 LEE, H., SAMADI, S., PUILLANDRE, N., TSAI, M.-H., DAI, C.-F. & CHEN, W.-J. 2016.

519 Eight new mitogenomes for exploring the phylogeny and classification of

520 Vetigastropoda. Journal of Molluscan Studies, 82: 534-541.

521 LOHSE, M., BOLGER, A.M., NAGEL, A., FERNIE, A.R., LUNN, J.E., STITT, M. &

522 USADEL, B. 2012. RobiNA: a user-friendly, integrated software solution for RNA-

523 Seq-based transcriptomics. Nucleic Acids Research, 40: W622-W627.

524 MARSHALL, B. A. 1995. Calliostomatidae (Gastropoda: Trochoidea) from New

525 Caledonia, the Loyalty Islands, and the northern Lord Howe Rise. In: (Bouchet P.,

526 ed.) Résultats des Campagnes Musorstom. Mémoires du Muséum National

527 d’Histoire Naturelle, 14: 382–458. 528 PALUMBI, S., MARTIN A, MCMILLAN, W., STICE, L. & GRABOWSKI, G. 1991. The

529 simple fool’s guide to PCR. http://palumbi.stanford.edu/SimpleFoolsMaster.pdf: 1-

530 45.

531 RAMBAUT, A. & DRUMMOND, A.J. 2007. Tracer v1.4, Available from

532 http://beast.bio.ed.ac.uk/Tracer.

533 RONQUIST, F. & HUELSENBECK, J.P. 2003. MrBayes 3: Bayesian phylogenetic

534 inference under mixed models. Bioinformatics, 19: 1572-1574.

535 SCHATTNER, P., BROOKS, A.N. & LOWE, T.M. 2005. The tRNAscan-SE, snoscan

536 and snoGPS web servers for the detection of tRNAs and snoRNAs. Nucleic Acids

537 Research, 33: W686-689.

538 SCHMIEDER, R. & EDWARDS, R. 2011. Quality control and preprocessing of

539 metagenomic datasets. Bioinformatics, 27: 863-864.

540 SCHWARZ, G. 1978. Estimating the dimension of a model. The Annals of Statistics, 6:

541 461-464.

542 ST JOHN, J. 2011. SeqPrep. Available from https://github.com/jstjohn/SeqPrep.

543 STAMATAKIS, A. 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic

544 analyses with thousands of taxa and mixed models. Bioinformatics, 22: 2688-2690.

545 URIBE, J.E., KANO, Y., TEMPLADO, J. & ZARDOYA, R. 2016. Mitogenomics of

546 Vetigastropoda: insights into the evolution of pallial symmetry. Zoologica Scripta, 45:

547 145-159.

548 VERRILL, A.E. 1884. Second catalogue of Mollusca recently added to the fauna of the

549 New England coast and the adjacent parts of the Atlantic, consisting mostly of deep-

550 sea species, with notes on others previously recorded. Transactions of the

551 Connecticut Academy of Science, 6: 139- 294.

552 WILLIAMS, S.T. 2012. Advances in molecular systematics of the vetigastropod

553 superfamily Trochoidea. Zoologica Scripta, 41: 571-595. 554 WILLIAMS, S.T., KARUBE, S. & OZAWA, T. 2008. Molecular systematics of

555 Vetigastropoda: Trochidae, turbinidae and trochoidea redefined. Zoologica Scripta,

556 37: 483-506.

557 WILLIAMS, S.T. & OZAWA, T. 2006 Molecular phylogeny suggests polyphyly of both

558 the turban shells (family Turbinidae) and the superfamily Trochoidea (Mollusca:

559 Vetigastropoda). Molecular Phylogenetics and Evolution, 39: 33-51.

560 WORT, E., FENBERG, P. & WILLIAMS, S.T. 2017. Testing the contribution of

561 individual genes in mitochondrial genomes for assessing phylogenetic relationships

562 in Vetigastropoda. Journal of Molluscan Studies, In press.

563 YOON, S. H. & KIM, W. 2005. Phylogenetic relationships among six vetigastropod

564 subgroups (Mollusca, Gastropoda) based on 18S rDNA sequences. Molecules and

565 Cells, 19: 283–288.

566 ZAPATA, F., WILSON, N.G., HOWISON, M., ANDRADE, S.C.S., JÖRGER, K.M.,

567 SCHRÖDL, M., GOETZ, F.E., GIRIBET, G. & DUNN, C.W. 2014. Phylogenomic

568 analyses of deep gastropod relationships reject . Proceedings of

569 the Royal Society of London, Biological Sciences, 281: 20141739.

570 571 Legend to Figures

572

573 Figure 1. Gene orders of selected Trochoidea s.l. mitochondrial genomes. The

574 consensus genome organization is shown for each lineage as well as the

575 ancestral gene order for Gastropoda. The genes encoded in the major and

576 minor strands are shown in the top and bottom lines, respectively. Gene

577 rearrangements (restricted to tRNA genes) are highlighted by colours.

578 Translocated genes are in green. Inverted genes are in blue. Genes for which

579 the exact rearrangement could not be inferred are in orange. Striped boxes

580 indicate regions not sequenced. The gene order of Clanculus and Calliostoma

581 could not be determined because sequence data was derived from RNASeq.

582

583 Figure 2. Phylogenetic relationships among vetigastropod superfamilies and within

584 Trochoidea s.l. based on 29 mitochondrial (protein-coding + rRNA) genes.

585 The reconstructed ML phylogram using Neomphalina as outgroup is shown.

586 Numbers at nodes are statistical support values for ML (BP)/ BI (BPP). An

587 asterisk indicates maximal support in ML (100% BP) and BI (1 BPP).

588 Vetigastropod superfamilies and trochoidean families are indicated.

589 cox1 cox2 D atp8 atp6 T cox3 K A R N I nad3 S nad2 T. brunnea F nad5 H nad4 nad4L S cob nad6 P nad1 L L rrnL V rrnS M Y C W Q G Tegulidae cox1 cox2 D atp8 atp6 T cox3 K A R N I nad3 S nad2 T. lividomaculata F nad5 H nad4 nad4L S cob nad6 P nad1 L L rrnL V rrnS M Y C W Q

cox1 E cox2 D atp8 atp6 G cox3 K A R N T I nad3 S nad2 F nad5 H nad4 nad4L S cob nad6 P nad1 L L rrnL V rrnS M Y C W Q Turbinidae Bolma cox1 cox2 D atp8 atp6 E G cox3 K A R N T I nad3 S nad2 Lunella F nad5 H nad4 nad4L S cob nad6 P nad1 L L rrnL V rrnS M Y C W Q

cox1 cox2 D atp8 atp6 T G cox3 K A R N I nad3 S nad2 Unassigned Cittarium F nad5 H nad4 nad4L S cob nad6 P nad1 L L rrnL V rrnS M Y C W Q

cox1 cox2 D atp8 atp6 T cox3 K A R N I nad3 S nad2 Tectus F nad5 H nad4 nad4L S cob Q nad6 P nad1 L L rrnL V rrnS

PHASIANELLOIDEA cox1 cox2 D atp8 atp6 T E G cox3 K A R N I nad3 S nad2 Phasianella F nad5 H nad4 nad4L S cob nad6 P nad1 L L rrnL V rrnS M Y C W Q

ANGARIOIDEA cox1 cox2 D atp8 atp6 T G cox3 K A R N I nad3 S nad2 Angaria F nad5 H nad4 nad4L S cob nad6 P nad1 L L rrnL V rrnS M Y C W Q E

Margaritidae cox1 cox2 D atp8 atp6 T cox3 K A R N I nad3 S nad2 Margarites F nad5 H nad4 nad4L S cob nad6 M P nad1 L L rrnL V rrnS

Clanculus undetermined order Trochidae cox1 cox2 D atp8 atp6 cox3 K A R N I nad3 S nad2 Gibbula F nad5 H nad4 nad4L T S cob nad6 P nad1 L L rrnL

Calliostomatidae Calliostoma undetermined order

cox1 cox2 D atp8 atp6 T cox3 K A R N I nad3 S nad2 Ancestral Gastropoda F nad5 H nad4 nad4L S cob nad6 P nad1 L L rrnL V rrnS M C Y W Q G E * Omphalius nigerrimus * Chlorostoma argyrostomum TROCHOIDEA: Tegulidae * 93/1 Lunella aff. cinerea 89/1 * Lunella granulata TROCHOIDEA: Turbinidae 93/1 * haematragus Tectus virgatus 99/1 TROCHOIDEA: Unassigned * Cittarium pica * PHASIANELLOIDEA 70/1 Phasianella solida Angaria neglecta 99/1 * ANGARIOIDEA Margarites vorticiferus * TROCHOIDEA: Margaritidae * Gibbula umbilicalis * Gibbula umbilicaris TROCHOIDEA: Trochidae * Stomatella planulata 75/1 Clanculus margaritarius * TROCHOIDEA: Calliostomatidae Haliotis rubra * HALIOTOIDEA 61/0.94 Haliotis tuberculata 84/1 Granata lyrata SEGUENZIOIDEA * Diodora graeca * Fissurella volcano FISSURELLOIDEA Variegemarginula punctata nux * LEPETODRILOIDEA Lepetodrilus schrolli Chrysomallon squamiferum NEOMPHALINA (outgroup) 5.0 0.5 5.0 Table 1. Mitochondrial (mt) DNA data analyzed in this study. Length in bp, Genbank accesion number, museum voucher, sampling location, and name of collector are provided

New mt data Species Family Superfamily bp Acc. No. Voucher Location Collector Cittarium pica Unassigned Trochoidea 17,949 KY212109 MNCN/ADN: 91331 Guanahacabibes, Bolivar, José Templado Tectus virgatus* Unassigned Trochoidea 15,956 KY205709 MNCN/ADN: 91332 Aqaba, Jordan José Templado Margarites vorticiferus* Margaritidae Trochoidea 15,253 KY205708 NHMUK: 20110451 Amchitka I., Cannikin, USA Piotr Kuklinski Gibbula umbilicaris* Trochidae Trochoidea 13,269 KY205707 MNCN/ADN: 86692 El Mohon, Murcia, SE Spain José Templado § Clanculus margaritarius† Trochidae Trochoidea — — NHMUK 20150502 Kitahama, Shirahama, Nishimuro-gun, Wakayama Pref., Japan Tomo Nakano Calliostoma zizyphinum† Calliostomatidae Trochoidea — — NHMUK 20160315 Shetland Islands, 60° 14.9'N, 01° 5.1'W, UK Piotr Kuklinski

GenBank mt data Species Family Superfamily bp GenBank Acc. No. Reference Chlorostoma argyrostomum Tegulidae Trochoidea 17,780 KX298892 Lee et al. 2016 Omphalius nigerrimus Tegulidae Trochoidea 17,755 KX298895 Lee et al. 2016 Tegula brunnea Tegulidae Trochoidea 17,690 NC_016954 Simison, 2011 (unpublished) Tegula lividomaculata Tegulidae Trochoidea 17,375 NC_029367 Uribe et al., 2016 Astralium haematragum Tegulidae Trochoidea 16,310 KX298891 Lee et al. 2016 Gibbula umbilicalis* Trochidae Trochoidea 16,277 KX646541 Wort et al. 2017 Stomatella planulata Trochidae Trochoidea 17,151 KX298894 Lee et al. 2016 Bolma rugosa Turbinidae Trochoidea 17,432 NC_029366 Uribe et al., 2016 Lunella aff. cinerea Turbinidae Trochoidea 17,670 KF700096 Williams et al., 2014 Lunella granulata Turbinidae Trochoidea 17,190 KX298890 Lee et al. 2016 Phasianella australis* Phasianelloidea 18,397 KX298888 Lee et al. 2016 Phasianella solida Phasianellidae Phasianelloidea 16,698 KR297251 Uribe et al., 2016 Angaria delphinus Angariidae Angarioidea 19,554 KX298893 Lee et al. 2016 Angaria neglecta Angariidae Angarioidea 19,470 KR297248 Uribe et al., 2016 Haliotis rubra Haliotidae Haliotoidea 16,907 NC_005940 Maynard et al., 2005 Haliotis tuberculata Haliotidae Haliotoidea 16,521 NC_013708 VanWormhoudt et al., 2009 Granata lyrata Chilofdontidae Seguenzioidea 17,632 NC_028708 Uribe et al., 2016 Fissurella volcano Fissurellidae Fissurelloidea 17,575 NC_016953 Simison, 2011 (unpublished) Diodora graeca Fissurellidae Fissurelloidea 17,209 KT207825 Uribe et al., 2016 Variegemarginula punctata* Fissurellidae Fissurelloidea 14,440 KX298889 Lee et al. 2016 * Lepetodriloidea 16,353 LC107880 Nakajima et al., 2016 Lepetodrilus schrolli* Lepetodrilidae Lepetodriloidea 15,579 KR297250 Uribe et al., 2016 Chrysomallon squamiferum Peltospiridae Neomphaloidea 15,388 AP013032 Nakagawa et al., 2014 *Nearly complete mt genomes † The GenBank Acc. No. of each mt gene is shown in Supplementary Data 3 along with coverage data. § voucher is a different specimen from another locality. Supplementary Data 1. Amplification strategy. Long PCR and primer walking primers Cittarium pica Long PCR Primer Sequence 5'-3' Fragment (bp) Citt-cox1-F TGGTTAATTCCTCTGATATTGGGAGCTCC cox1-rrnL (11436) TROmt16sF GATAACAGCGTAATCTTTCTGGAGAGATC TROmt16sR AAGCTCAACAGGGTCTTCTTGTCCC rrnL-cox3 (3439) 85CPcox3R CATAGACACCATCTGAGATAGTTAACGG 85CPcox3F GAGCTTATTTTCATAGAAGTCTCGCTTC cox3-cox1 (3580) Citt-cox1-R GCAGGATCAAAGAAGGATGTGTTAAAATTTC

Margarites vorticiferus Long PCR Primer Sequence 5'-3' Fragment (bp) Alecox3F_UJ CTGAGCATATTTCCATAGAAGCCTGGC cox3-cox1 (3074) Alecox1R CTGATCAAGTGAATAGTGGTAGGCGTTC Alecox1F CTTAGTTTTCGGGATTTGAGCAGGCC cox1-rrnS (12686) Ale12SF TTTAAATCCTTCCAGGGGAACCTGTCC

Gibbula umbilicaris Long PCR Primer Sequence 5'-3' Fragment (bp) GVcox3F TTTCCACAGAAGACTTGCTCCTACTCC cox3-cox1 (2867) G2-cox1-r AATAGAAGAAACACCYGCTAAGTGAAGGGA G2-cox1-F CCGGTGCTATTACTATGCTGCTCACTGA cox1-rrnL (9870) TROmt16sF GATAACAGCGTAATCTTTCTGGAGAGATC

Tectus virgatus Long PCR Primer Sequence 5'-3' Fragment (bp) TVcox3F GTATTTCCACAGAAGGTTGGCTTCTGC cox3-cox1 (3567) TVcox1R GAAGAGATAGCAGCAACAAAATGGCCGT TVcox1F GCATTTCCGCGACTTAATAACATGAGATT cox1-rrnL (10646) TROmt16sF GATAACAGCGTAATCTTTCTGGAGAGATC Supplementary Data 2. Mitochondrial genome features

Cittarium pica Gene Codon Gene Type Start Stop Length Start Stop Strand cox1 CDS 1 1539 1539 ATG TAA forward cox2 CDS 1669 2364 696 ATG TAG forward trnD tRNA 2535 2605 71 forward atp8 CDS 2608 2785 178 –– TAA forward atp6 CDS 2975 3670 696 ATG TAA forward trnF tRNA 3730 3799 70 reverse nad5 CDS 3937 5683 1747 ATG TAA reverse trnH tRNA 5684 5754 71 reverse nad4 CDS 5805 7202 1398 ATG TAA reverse nad4L CDS 7196 7495 300 ATG TAG reverse trnT tRNA 7600 7669 70 forward trnS (tga) trna 7803 7870 68 reverse cob CDS 7902 9041 114 ATG TAA reverse nad6 CDS 9231 9737 507 ATG TAA reverse trnP tRNA 9741 9810 70 reverse nad1 CDS 9953 10903 951 ATG TAA reverse trnL (taa) tRNA 10905 10972 68 reverse trnL (tag) tRNA 11281 11348 68 reverse rrnL rRNA 11349 13012 1664 reverse trnV tRNA 13013 13086 74 reverse rrnS rRNA 13087 14153 1067 reverse trnM tRNA 14154 14222 69 reverse trnY tRNA 14312 14379 68 reverse trnC tRNA 14381 14447 67 reverse trnW tRNA 14455 14521 67 reverse trnQ tRNA 14539 14607 69 reverse trnG tRNA 14618 14686 69 forward cox3 CDS 14740 15519 780 ATG TAA forward trnK tRNA 15718 15777 60 forward trnA tRNA 15778 15845 68 forward trnR tRNA 15906 15974 69 forward trnN tRNA 15989 16061 73 forward trnI tRNA 16109 16179 71 forward nad3 CDS 16184 16537 354 ATG TAA forward trnS (cgt) tRNA 16626 16692 67 forward nad2 CDS 16696 17949 1254 ATG T–– forward Margarites vorticiferus Gene Codon Gene Type Start Stop Length Start Stop Strand cox1 CDS 2522 4066 1545 ATG TAA forward cox2 CDS 4117 4844 728 ATA TAA forward trnD tRNA 4875 4941 67 forward atp8 CDS 4943 5122 180 ATG TAG forward atp6 CDS 5200 5895 696 ATG TAA forward trnF tRNA 5928 5993 66 reverse nad5 CDS 6012 7765 1754 ATG TAG reverse trnH tRNA 7766 7831 66 reverse nad4 CDS 7894 9282 1389 GTG TAA reverse nad4L CDS 9276 9575 300 ATG TAG reverse trnT tRNA 9635 9705 71 forward trnS (tga) tRNA 9710 9776 67 reverse cob CDS 9786 10925 114 ATG TAA reverse nad6 CDS 10997 11500 504 ATG TAA reverse trnM tRNA 11500 11563 64 reverse trnP tRNA 11641 11709 69 reverse nad1 CDS 11769 12716 948 ATG TAA reverse trnL (taa) tRNA 12718 12785 68 reverse trnL (tag) tRNA 12811 12878 68 reverse rrnL rRNA 12879 14458 158 reverse trnV tRNA 14459 14527 69 reverse rrnS rRNA 14528 15253 727 reverse

cox3 CDS 1 483 483 –– TAA forward trnK tRNA 520 578 59 forward trnA tRNA 579 650 72 forward trnR tRNA 690 751 62 forward trnN tRNA 752 821 70 forward trnI tRNA 828 894 67 forward nad3 CDS 897 125 354 ATG TAG forward trnS (cgt) tRNA 1271 1338 68 forward nad2 CDS 1343 2499 1157 TGT TAA forward Gibbula umbilicaris Gene Codon Gene Type Start Stop Length Start Stop Strand cox1 CDS 2419 3954 1536 ATG TAA forward cox2 CDS 3983 4675 693 ATG TAA forward trnD tRNA 4706 4770 65 forward atp8 CDS 4772 4936 165 ATG TAA forward atp6 CDS 4981 5679 699 ATG TAG forward trnF tRNA 5712 5776 65 reverse nad5 CDS 5801 7555 1755 ATG TAA reverse trnH tRNA 7556 7623 68 reverse nad4 CDS 7700 9088 1389 ATG TAG reverse nad4L CDS 9082 9381 300 ATG TAG reverse trnT tRNA 9406 9474 69 reverse trnS (tga) tRNA 9503 9569 67 reverse cob CDS 9601 1074 114 ATG TAA reverse nad6 CDS 10856 11362 507 ATG TAA reverse trnP tRNA 11366 11431 66 reverse nad1 CDS 11494 12435 942 ATG TAG reverse trnL (taa) tRNA 12437 12504 68 reverse trnL (tag) tRNA 12530 12597 68 reverse rrnL rRNA 12598 13269 671 reverse

cox3 CDS 1 479 479 –– TAA forward trnK tRNA 513 570 58 forward trnA tRNA 571 638 68 forward trnR tRNA 641 709 69 forward trnN tRNA 716 782 67 forward trnI tRNA 784 850 67 forward nad3 CDS 855 1208 354 ATG TAA forward trnS (cgt) tRNA 1218 1285 68 forward nad2 CDS 1289 2418 113 ATG T–– forward Tectus virgatus Gene Codon Gene Type Start Stop Length Start Stop Strand cox1 CDS 2979 4514 1536 ATG TAA forward cox2 CDS 4586 5276 691 ATG TAA forward trnD tRNA 5453 5520 68 forward atp8 CDS 5521 5706 186 ATG TAG forward atp6 CDS 5866 6561 696 ATG TAA forward trnF tRNA 6596 6665 70 reverse nad5 CDS 6782 8515 1734 ATG TAA reverse trnH tRNA 8516 8584 69 reverse nad4 CDS 8682 10079 1398 ATG TAA reverse nad4L CDS 10073 10372 300 ATG TAA reverse trnT tRNA 10429 10497 69 forward trnS (tga) tRNA 10553 10618 66 reverse cob CDS 10629 11768 114 ATG TAA reverse trnQ tRNA 11767 11831 65 reverse nad6 CDS 11834 12340 507 ATG TAA reverse trnP tRNA 12345 12415 71 reverse nad1 CDS 12497 13447 951 ATG TAG reverse trnL (taa) tRNA 13449 13516 68 reverse trnL (tag) tRNA 13561 13628 68 reverse rrnL rRNA 13629 15163 1534 reverse trnV tRNA 15164 15231 67 reverse rrnS rRNA 15232 15956 724 reverse

cox3 CDS 1 546 546 –– TAA forward trnK tRNA 734 791 58 forward trnA tRNA 792 859 68 forward trnR tRNA 916 985 70 forward trnN tRNA 103 1098 69 forward trnI tRNA 1146 1215 70 forward nad3 CDS 1219 1572 354 ATG TAG forward trnS (cgt) tRNA 1703 1769 67 forward nad2 CDS 1774 2978 1205 ATG T–– forward Supplementary Data 3. Coverage based on unpaired reads for mitochondrial genes obtained from NGS shotgun sequencing and their respectively Genbank accesion number Calliostoma zizyphinum Clanculus margaritarius Mean Minimum Maximum Acc. No. Mean Minimum Maximum Acc. No. COX1 7291.8 269 28585 KY200866 12286.6 530 32082 KY200867 COX2 9657.6 2342 26004 KY200869 48166.1 7545 112145 KY200868 COX3 5034.7 195 11993 KY200871 6472.2 256 10505 KY200870 CYTB 2744.7 127 5458 KY200872 13055.9 671 52237 KY200873 ND1 953.8 16 1757 KY200875 3740.5 1521 6799 KY200874 ND2 162 60 6810 KY200876 10042.5 652 25779 KY200877 ND3 5823.2 176 9325 KY200879 4870.9 283 7895 KY200878 ND4 1130.2 76 2404 KY200880 4858 1107 8186 KY200881 ND4L 600.6 19 941 KY200883 3175.6 508 4804 KY200882 ND5 460 131 1076 KY200885 3018.9 525 11168 KY200884 ND6 972.4 105 1754 KY200887 15338.5 1433 31569 KY200886 ATP6 2486.5 122 3949 KY200888 9141.6 1037 20583 KY200889 ATP8 27.9 18 34 KY200891 20252.5 8564 24420 KY200890 12S 6258 20 14620 KY200892 8161.1 408 15373 KY200893 16S 6608 1296 13496 KY200894 21247.6 1492 65018 KY200895