No evidence of sex-linked heteroplasmy or doubly-uniparental inheritance of mtDNA in five gastropod species

Mitochondria are organelles that contain their own genetic material (mitochondrial DNA or mtDNA), which is usually strictly maternally inherited in animals (strict maternal inheritance or SMI) (Birky, 2001). One animal group diverges from the SMI rule, i.e. bivalve molluscs with their doubly-uniparental inheritance (DUI) system (see Breton et al., 2007; Passamonti & Ghiselli, 2009; Zouros, 2013 for reviews). DUI is an mtDNA inheritance system in which females transmit their ‘F mtDNA’ to all offspring, and males transmit their highly divergent ‘M mtDNA’ only to their sons. Usually, females contain and express the F mtDNA in all their tissues, whereas males contain and express the F mtDNA in their soma, and the M mtDNA in their gametes (Breton et al., 2007; Passamonti & Ghiselli, 2009; Zouros, 2013). The levels of DNA divergence between F and M mtDNAs in male bivalves vary among species, e.g. from 8% in the veneroid Arctica islandica (uncorrected nucleotide p-distance for cytb; Dégletagne et al., 2016) to >40% in freshwater mussels (uncorrected nucleotide p-distance for all protein-coding genes except atp8; Doucet-Beaupré et al., 2010), and they are often greater than the amount of divergence which, when experimentally induced, impairs the mitochondrial function in model systems for the study of mito-nuclear coevolution (Breton et al., 2007). Bivalves are thus unique in that they possess two very different sets of mt-encoded proteins ‘forced’ to work in concert with proteins coded by a single nuclear genome. One of the leading hypotheses to explain the origin and maintenance of this unusual mtDNA transmission system in bivalves is that the F and M mtDNAs are key elements of the sex determination system, as heteromorphic sex chromosomes are absent in this taxon (e.g. Breton et al., 2011, 2014; Breton & Stewart, 2015). However, the link between DUI and sex determination still remains to be elucidated. So far, DUI has been found in 101 bivalve species belonging to four orders, viz. Mytiloida, Nuculanoida, Unionoida and Veneroida, and ten families (out of c. 105; www.bivatol.org), viz. Arcticidae, Donacidae, Hyriidae, Mactridae, Margaritiferidae, Mytilidae, Nuculanidae, Solenidae, Unionidae and Veneridae (Breton et al., 2007; Passamonti & Ghiselli, 2009; Doucet-Beaupré et al., 2010; Zouros, 2013; Boyle & Etter, 2013; Plazzi, 2015). DUI could also be present in other molluscan taxa (e.g. Parakatselaki, Saavedra & Ladoukakis, 2015), but this remains to be demonstrated. Specifically, DUI has been reported in gonochoristic bivalve species (hermaphroditic species do not possess DUI; Breton et al., 2011), many of which show strong sex biases in offspring following parental crosses, e.g. Mytilus spp. (Saavedra, Reyero & Zouros, 1997; Kenchington et al., 2002), Ruditapes philippinarum (Ghiselli et al., 2012) and Unio delphinus (Machordom et al., 2015). Recently, Parakatselaki et al. (2015) used these two main criteria, i.e. the presence of gonochorism and sex bias, to investigate whether DUI exists outside bivalves, specifically in gastropods, the phylogenetically closest molluscan class to bivalves. Although their results suggested that the mtDNA in Pomacea diffusa is maternally transmitted, the authors proposed that their pipeline could be adopted for the search for DUI in other animals (Parakatselaki et al., 2015). Following this proposed pipeline, we searched for DUI in five gonochoristic gastropod species for which biases in sex ratio have been reported: Littorina littorea (Linnaeus, 1758) (Caenogastropoda: Littorinidae) (Lambert et al., 2012); Nucella lapillus (Linnaeus, 1758) (Caenogastropoda: Muricidae) (Crothers, 1985); Viviparus ater (De Cristofori & Jan, 1832) (Caenogastropoda: Viviparidae) (Brown et al., 1989); Lunatia heros (Say, 1822) (Caenogastropoda: Naticidae; biased sex ratios have been reported in the related species Neverita lewisiiBernard, 1967) and Tectura testudinalis (O.F. Müller, 1776) (Patellogastropoda: Lottiidae; biased sex ratios have been reported in other patellogastropod species, Branch, 1981). Littorina littorea, N. lapillus and T. testudinalis were collected in Rivière Madeleine (49.3028°N, 65.3812°W, Québec, Canada) and L. heros in Carleton-sur-mer (48.1781°N, 66.1666°W, Québec, Canada) in 2012, 2013 and 2014. Viviparus ater was collected in Lac Saint-Pierre (46.2044°N, 72.8284°W, Québec, Canada) in 2013. Specimens were either kept alive (L. littorea, L. heros, N. lapillu, T. testudinalis) or preserved in 95% ethanol (all species) until use. Fresh or preserved specimens were sexed under the microscope and four males and four females were used per species, except for N. lapillus and L. heros for which only three males and four females, and three males and two females were unambiguously sexed, respectively. For each individual, total genomic DNA was isolated from the foot and the gonad with the Qiagen DNeasy tissue kit (Qiagen Inc., Valencia, CA) using the animal-tissue protocol. Two mitochondrial regions were amplified: a 637-nt fragment from the protein-coding gene cytochrome c oxidase subunit 1 (cox1) using the primers LCO1490 and HCO2198 (Folmer et al., 1994) and a c. 580-nt fragment from the rRNA gene 16 S using the primers 16Sar and 16Sbr (Palumbi et al., 1991). PCR amplifications were performed in 50 µl volume containing ~20 ng of total DNA, 0.4 µM of each primer, 1× Qiagen PCR buffer, 2.5 mM MgCl2, 200 µM of each dNTP and 2.5 U Qiagen Taq DNA polymerase. The thermal cycling parameters were as follows: 5 min at 94 °C then 35 cycles of (94 °C for 15 s; 45 °C for 30 s; 72 °C for 60 s), with a final extension step of 10 min at 72 °C. The PCR products were purified using a QIAquick PCR purification kit according to supplier's (Qiagen Inc.) instructions. The purified PCR products were sequenced at the sequencing platform of McGill University (Montréal, Canada). Sequences were aligned with ClustalW (Thompson, Higgins & Gibson, 1994). BLAST analyses (Altschul et al., 1997) confirmed the identifications at the species level. As stated by Parakatselaki et al. (2015), two main features are expected from mtDNA sequences of species with DUI: (1) the divergence between sequences from male and female gonads should be higher than the divergence among sequences from gonads of the same sex and (2) distinct sequences should be found in somatic vs gonad tissues only in males. The authors did not observe these features in P. diffusa after having sequenced three females and four males; the foot and the gonad provided identical sequences for each individual (Parakatselaki et al., 2015). Using the same approach, i.e. with two sequences for each individual, one from the foot and one from the gonad, a total of 128 sequences were obtained for this study (Table 1). Number of DNA sequences for each species and each mitochondrial gene. Two sequences were obtained for each individual (from foot and gonad tissues). Number of individuals is indicated in parentheses. For V. ater, it was only possible to sequence the 16 S fragment since cox1 could not be amplified. Male and female haplotypes are counted together. Number of DNA sequences for each species and each mitochondrial gene. Two sequences were obtained for each individual (from foot and gonad tissues). Number of individuals is indicated in parentheses. For V. ater, it was only possible to sequence the 16 S fragment since cox1 could not be amplified. Male and female haplotypes are counted together. As in P. diffusa (Parakatselaki et al., 2015), we recovered the same haplotype from somatic and gonadal tissues for both mitochondrial regions in each gastropod species (Table 1). Therefore we must conclude that we do not have any evidence of sex-linked heteroplasmy in these five species. In some cases, however, different haplotypes were found in the population sampled: for cox1 in T. testudinalis two haplotypes differed by 2 nt; for 16 S in L. littorea two haplotypes differed by 1 nt and for 16 S in N. lapillus two haplotypes differed by 3 nt; however, none of our individuals was heteroplasmic, meaning that the sequences obtained for each gene/tissue were identical, and different haplotypes were also found in females and males. Although further work trying to reveal rare heteroplasmic variants by molecular cloning could strengthen our results, all the sequences obtained for each gene/tissue/sex/species were identical and no double peaks were observed in the chromatograms, suggesting the absence of heteroplasmy in the tissues tested (Table 1). Theoretically, the absence of DUI in a species can be unambiguously demonstrated (1) when the mtDNAs from uncontaminated collection of eggs and sperm from several individuals are shown to be the same, and (2) when newly fertilized eggs from several females are all shown to have a dispersed pattern of sperm mitochondria in blastomeres. In species with DUI, sperm mitochondria are dispersed in blastomeres only in female embryos whereas, in male embryos, sperm mitochondria remain grouped together in a single blastomere that will give rise to the germline (Cao, Kenchington & Zouros, 2004; Milani, Ghiselli & Passamonti, 2012). In practice, most studies have used PCR-based analyses to prove or refute the presence of DUI by, respectively, showing the consistent coexistence of distinct mtDNA forms only in males (and only in male gonads vs male somatic and female tissues) or not (e.g. Parakatselaki et al., 2015; Plazzi, 2015; Plazzi, Cassano & Passamonti, 2015; Vargas et al., 2015; Dégletagne, Abele & Held, 2016). The presence of DUI is relatively easy to prove using this approach and, contrary to DUI species, occasional heteroplasmy in species with maternal inheritance of mtDNA is typically present in one or several tissues in a non-consistent way and in both sexes. However, the absence of DUI is much more difficult to establish for at least two reasons: (1) the possible preferential annealing of the PCR primers with only the maternal mtDNA (e.g. Saavedra et al., 1997; Theologidis, Saavedra & Zouros, 2007) and (2) the possibility that the paternal mtDNA could be indistinguishable from the maternal mtDNA because of an invasion of the paternal transmission route by the latter, which would therefore be inherited through the sperm (Zouros, 2013). In the first case, the problem can be overcome by targeting more than one mitochondrial region/gene and by using several pairs of primers designed in conserved mtDNA regions to be able to amplify these regions from both closely- and distantly-related organisms, thus decreasing considerably the probability that the paternal genome will be missed (Parakatselaki et al., 2015). This approach has been used in the present study. Regarding the second situation, which is called a masculinization event and has been occasionally observed only in Mytilus spp., we agree with Parakatselaki et al. (2015) that this possibility is extremely unlikely, because the phenomenon is very rare (Ladoukakis et al., 2002) and the probability that it happened in our five gastropod species is almost nil. Additionally, masculinized mt genomes in sperm are usually quite different (3% DNA divergence, evolutionary distances by Tamura-Nei model for cox1) from the mtDNA present in somatic tissues (Hoeh et al., 1996, 1997). In our study, the sequences recovered from the gonadal and somatic tissues were all identical, suggesting an absence of DUI. DUI has been hypothesized to be involved in sex determination, explaining its long-term persistence in bivalves, as heteromorphic sex chromosomes are absent in this taxon (e.g. Breton et al., 2011, 2014; Breton & Stewart, 2015). If this hypothesis is true, then the occurrence of sex chromosomes in gastropods (reviewed by Thiriot-Quiévreux, 2003) could explain the absence of DUI in this group. To our knowledge, the presence of sex chromosomes has been confirmed in Littorina species and V. ater (Thiriot-Quiévreux, 2003). Otherwise, all DUI species for which the complete F and M mitochondrial genomes have been sequenced possess at least one of the following features: (1) additional sex-specific, functional open reading frames (ORFs) in the F and M mtDNAs (e.g. Breton et al., 2011; Milani et al., 2013, 2014; Minoiu, Burzynski & Breton, 2016); (2) a duplication or an extension of the cox2 gene (Doucet-Beaupré et al., 2010; Passamonti et al., 2011; Bettinazzi et al., 2016) and/or (3) a highly modified version of the atp8 gene (e.g. Breton, Stewart & Hoeh, 2010), which does not possess the MPQL amino-acid signature conserved at the N-terminus of metazoan ATP8 (Gissi, Iannelli & Pesole, 2008). To our knowledge, mitochondrial genomes in gastropods are all of relatively reduced size (13–15 kb) with the exception of Lottia digitalis, which possesses two large non-coding regions and a mtDNA of 26 kb (Grande, Templado & Zardoya, 2008; White et al., 2011), and they all consist of 13 protein-coding genes, without additional ORFs, duplicated genes or gene extension, and with a ‘typical’ ATP8 protein (Grande et al., 2008; White et al., 2011). If for some reasons the deviations observed in bivalve mtDNAs are a consequence of DUI, then maybe these features should be taken into account when searching for DUI in other animals. Such studies are underway in our laboratory. Given the fact that sex-linked heteroplasmy has not been demonstrated in any gastropod species so far, there is currently no evidence for DUI in this class and we therefore must continue to consider this unusual phenomenon as restricted to bivalves. We would like to thank Amélie Genovese, Antonia Cattaneo and Christiane Hudon for providing us with samples. We would also like to thank three anonymous reviewers for insightful comments on the manuscript. This work was supported by the National Sciences and Engineering Research Council (NSERC) (grant no. RGPIN/435656-2013 to S.B.). A.G. was financially supported by the Groupe de Recherche Interuniversitaire en Limnologie et en Environnement Aquatique (GRIL).

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.