Mitochondrial genome sequencing of marine leukaemias reveals cancer contagion between clam species in the Seas of Southern Europe

  1. Daniel Garcia-Souto  Is a corresponding author
  2. Alicia L Bruzos
  3. Seila Diaz
  4. Sara Rocha
  5. Ana Pequeño-Valtierra
  6. Camila F Roman-Lewis
  7. Juana Alonso
  8. Rosana Rodriguez
  9. Damian Costas
  10. Jorge Rodriguez-Castro
  11. Antonio Villanueva
  12. Luis Silva
  13. Jose Maria Valencia
  14. Giovanni Annona
  15. Andrea Tarallo
  16. Fernando Ricardo
  17. Ana Bratoš Cetinić
  18. David Posada
  19. Juan Jose Pasantes
  20. Jose MC Tubio  Is a corresponding author
  1. Genomes and Disease, Centre for Research in Molecular Medicine and Chronic Diseases (CIMUS), Universidade de Santiago de Compostela, Spain
  2. Department of Zoology, Genetics and Physical Anthropology, Universidade de Santiago de Compostela, Spain
  3. Cancer Ageing and Somatic Mutation Programme, Wellcome Sanger Institute, United Kingdom
  4. Phylogenomics Lab, Universidade de Vigo, Spain
  5. CINBIO, Universidade de Vigo, Spain
  6. Galicia Sur Health Research Institute (IIS Galicia Sur), SERGAS-UVIGO, Spain
  7. Centro de Investigación Mariña, Universidade de Vigo, ECIMAT, Spain
  8. Instituto Español de Oceanografía (IEO), Centro Oceanográfico de Cádiz, Spain
  9. Laboratori d’Investigacions Marines i Aqüicultura, (LIMIA) - Govern de les Illes Balears, Spain
  10. Instituto de Investigaciones Agroambientales y de Economía del Agua (INAGEA) (INIA-CAIB-UIB), Spain
  11. Stazione Zoologica Anton Dohrn, Italy
  12. ECOMARE, Centre for Environmental and Marine Studies (CESAM), Department of Biology, University of Aveiro, Santiago University Campus, Portugal
  13. Department of Aquaculture, University of Dubrovnik, Croatia
  14. Department of Biochemistry, Genetics and Immunology, Universidade de Vigo, Spain
  15. Centro de Investigación Mariña, Universidade de Vigo, Spain

Abstract

Clonally transmissible cancers are tumour lineages that are transmitted between individuals via the transfer of living cancer cells. In marine bivalves, leukaemia-like transmissible cancers, called hemic neoplasia (HN), have demonstrated the ability to infect individuals from different species. We performed whole-genome sequencing in eight warty venus clams that were diagnosed with HN, from two sampling points located more than 1000 nautical miles away in the Atlantic Ocean and the Mediterranean Sea Coasts of Spain. Mitochondrial genome sequencing analysis from neoplastic animals revealed the coexistence of haplotypes from two different clam species. Phylogenies estimated from mitochondrial and nuclear markers confirmed this leukaemia originated in striped venus clams and later transmitted to clams of the species warty venus, in which it survives as a contagious cancer. The analysis of mitochondrial and nuclear gene sequences supports all studied tumours belong to a single neoplastic lineage that spreads in the Seas of Southern Europe.

Editor's evaluation

This paper describes a previously unknown lineage of transmissible cancer in a clam, in which the cancer arose from a different, but related, species. The data are clear and overall, the conclusions well-supported, and this finding increases our understanding of transmissible cancers in nature and will be of broad interest.

https://doi.org/10.7554/eLife.66946.sa0

eLife digest

In humans and other animals, cancer cells divide excessively, forming tumours or flooding the blood, but they rarely spread to other individuals. However, some animals, including dogs, Tasmanian devils and bivalve molluscs like clams, cockles and mussels, can develop cancers that are transmitted from one individual to another. Despite these cancers being contagious, each one originates in a single animal, meaning that even when the cancer has spread to many individuals, its origins can be traced through its DNA.

Cancer contagion is rare, but transmissible cancers seem to be particularly common in the oceans. In fact, 7 types of contagious cancer have been described in bivalve species so far. These cancers are known as ‘hemic neoplasias’, and are characterized by the uncontrolled division of blood-like cells, which can be released by the host they developed in, and survive in ocean water. When these cells encounter individuals from the same species, they can infect them, causing them to develop hemic neoplasia too

There are still many unanswered questions about contagious cancers in bivalves. For example, how many species do the cancers affect, and which species do the cancers originate in? To address these questions, Garcia-Souto, Bruzos, Díaz et al. gathered over 400 specimens of a species of clam called the warty venus clam from the coastlines of Europe and examined them for signs of cancer. Clams collected in two regions of Spain showed signs of hemic neoplasia: one of the populations was from the Balearic Islands in the Mediterranean Sea, while the other came from the Atlantic coast of northwestern Spain.

Analyzing the genomes of the tumours from each population showed that the cancer cells from both regions had likely originated in the same animal, indicating that the cancer is contagious and had spread through different populations. The analysis also revealed that the cancer did not originally develop in warty venus clams: the cancer cells contained DNA from both warty venus clams and another species called striped venus clams. These two species live close together in the Mediterranean Sea, suggesting that the cancer started in a striped venus clam and then spread to a warty venus clam. To determine whether the cancer still affected both species, Garcia-Souto, Bruzos, Díaz et al. screened 200 striped venus clams from the same areas, but no signs of cancer were found in these clams. This suggests that currently the cancer only affects the warty venus clam.

These findings confirm that contagious cancers can jump between clam species, which could be threat to the marine environment. The fact that the cancer was so similar in clams from the Atlantic coast and from the Mediterranean Sea, however, suggests that it may have emerged very recently, or that human activity helped it to spread from one place to another. If the latter is the case, it may be possible to prevent further spread of these sea-borne cancers through human intervention.

Introduction

Cancers are clonal cell lineages that arise due to somatic changes that promote cell proliferation and survival (Stratton et al., 2009). Although natural selection operating on cancers favours the outgrowth of malignant clones with replicative immortality, the continued survival of a cancer is generally restricted by the lifespan of its host. However, clonally transmissible cancers – from now on, transmissible cancers – are somatic cell lineages that are transmitted between individuals via the transfer of living cancer cells, meaning that they can survive beyond the death of their hosts (Murchison, 2008). Naturally occurring transmissible cancers have been identified in dogs (Murgia et al., 2006; Murchison et al., 2014; Baez-Ortega et al., 2019), Tasmanian devils (Murchison et al., 2012; Pye et al., 2016) and, more recently, in marine bivalves (Metzger et al., 2015; Metzger et al., 2016; Yonemitsu et al., 2019).

Hemic neoplasia (HN), also called disseminated neoplasia, is a type of leukaemia cancer found in multiple species of bivalves, including oysters, mussels, cockles, and clams (Carballal et al., 2015). Although these leukaemias represent different diseases across bivalve species, they have been classically grouped under the same term because neoplastic cells share morphological features (Carballal et al., 2015). Some HNs have been proven to have a clonal transmissible behaviour (Metzger et al., 2015), in which neoplastic cells, most likely haemocytes (i.e. the cells that populate the haemolymph and play a role in the immune response), are likely to be transmitted through marine water. In late stages of the disease, leukaemic cells invade the surrounding tissues and, generally, animals die because of the infection (Carballal et al., 2015), although remissions have also been described (Burioli et al., 2019). Despite the observation that leukaemic cells are typically transmitted between individuals from the same species, on occasion they can infect and propagate across populations from a second, different bivalve species (Metzger et al., 2016; Yonemitsu et al., 2019). Hence, these cancers represent a potential threat for the ecology of the marine environment, which argues for the necessity of their identification and characterization for their monitoring and prevention.

Here, we use multiplatform next-generation genome sequencing technologies, including Illumina short reads and Oxford Nanopore long reads, together with cytogenetics, electron microscopy, and cytohistological approaches to identify, characterize, and decipher the evolutionary origin of a new marine leukaemia that is transmitted between two different clam species that inhabit the Seas of Southern Europe, namely warty venus (Venus verrucosa) and striped venus (Chamelea gallina) (Video 1).

Video 1
Mitochondrial genome sequencing of marine leukaemias reveals cancer contagion between clam species in the Seas of Southern Europe.

Infographic video outlining the main findings of the research carried out.

Results and discussion

We investigated the prevalence of HN in the warty venus clam (V. verrucosa), a saltwater bivalve found in the Atlantic Coast of Europe and the Mediterranean Sea. We collected 345 clam specimens from six sampling regions in the Atlantic and the Mediterranean coasts of Europe across five different countries, including Spain, Portugal, France, Ireland, and Croatia (Figure 1a; Supplementary file 1). Cytohistological examination identified HN-like tumours in eight specimens from two sampling points in Spain (Figure 1b–e; Figure 1—figure supplement 1). Three HN-positive specimens (ERVV17-2995, ERVV17-2997, and ERVV17-3193) were collected in Galicia, northwest of the Iberian Peninsula in the Atlantic Ocean, and another five specimens (EMVV18-373, EMVV18-376, EMVV18-391, EMVV18-395, and EMVV18-400) were collected in the Balearic Islands, bathed by the Mediterranean Sea (Figure 1a; Supplementary file 1). Four of these specimens (ERVV17-2995, ERVV17-3193, EMVV18-391, and EMVV18-395) showed a severe form of the disease – classified as N3 stage – which is characterized by high levels of neoplastic cells infiltrating the gills, different levels of infiltration of the digestive gland and gonad, and low/very low infiltration of the mantle and foot (Figure 1d,e; Figure 1—figure supplement 1); one specimen (EMVV18-400) was found that was affected with an intermediate form of the disease – N2 stage – characterized by low levels of neoplastic cells infiltrating the gill vessels, digestive gland, and gonad, but not the foot (Figure 1—figure supplement 1); and three specimens (ERVV17-2997, EMVV18-373, and EMVV18-376) were diagnosed with a light form of the disease – N1 stage – characterized by low levels of neoplastic cells infiltrating the gills vessels only, and no infiltration in the remaining tissues (Figure 1—figure supplement 1). Electron microscopy analysis through gill’s ultrathin sections from two neoplastic warty venus specimens (ERVV17-2995 and ERVV17-3193) revealed tumour cells with a round shape and a pleomorphic nucleus, which are morphological features that generally characterize bivalves’ HN (Figure 1f; Figure 1—figure supplement 2). Finally, one additional neoplastic warty venus specimen (EVVV11-02) was included in the study. The animal, which was sampled in 2011 in Galicia and came from a private collection, showed abnormal metaphases in the gills that were suggestive of HN. Although the species typically shows a 2n = 38 karyotype with metacentric chromosomes that are homogeneous in size (García-Souto et al., 2015), the tumoural metaphases from this individual showed around 100 chromosomes that were variable in size and shape (Figure 1g).

Figure 1 with 2 supplements see all
Geographical location of warty venus (V. verrucosa) specimens and diagnosis of hemic neoplasia.

(a) Locations of V. verrucosa clams collected for this study and specimens diagnosed with hemic neoplasia. Size of the pie charts correlates with the number of samples collected (number of samples ‘n’ is shown together with each pie chart). Pie charts show the proportion of samples with hemic neoplasia (black, no neoplastic specimens; red, neoplastic specimens). Codes of neoplastic samples are shown. Top-right corner shows a representative specimen of the species V. verrucosa. (b) Cytological examination of haemolymph smear (Hemacolor stain) from a healthy (N0) specimen, ERVV17-2963, shows normal haemocytes. (c) Haemolymph smear of a V. verrucosa specimen with high-grade (N3 stage) hemic neoplasia, ERVV17-3193, shows neoplastic cells that replaced normal haemocytes. (d) Detail of haematoxylin and eosin-stained of histological section from the gills of the healthy (N0) specimen ERVV17-2990. (e) Same for ERVV17-2995, a specimen infected with a high-grade (N3 stage) hemic neoplasia, showing neoplastic cells infiltrating the gills. (f) Transmission electron microscopy analysis of a V. verrucosa hemic neoplasia tumour cell shows a round shape, pseudopodia ‘p’, pleomorphic nucleus ‘n’ with scattered heterochromatin, and mitochondria ‘m’. (g) Metaphase chromosomes from a neoplastic cell found in the gills of the V. verrucosa specimen EVVV11-02, showing abnormal chromosome number (>19 pairs) and abnormal chromosome morphology. Chromosomes stained with 4′,6-DiAmidino-2-PhenylIndole (DAPI) and Propidium Iodide (PI).

To obtain some biological insights into the clonal dynamics of this cancer, we carried out whole-genome sequencing with Illumina paired-ends in DNA samples isolated from the tumoural haemolymph from eight out of nine neoplastic specimens mentioned above (Table 1). Their feet were also sequenced, as foot typically represents the tissue with lower infiltration of neoplastic cells, making it a good candidate tissue to act as ‘matched-normal’ (i.e. host tissue). As for the animal with an abnormal karyotype (EVVV11-02) that was compatible with HN, we sequenced the only tissue available, which were gills (Table 1). Only one neoplastic specimen (EMVV18-373) that had a very low proportion of tumour cells in its haemolymph was excluded from the sequencing. Then, we mapped the paired-end reads onto a dataset containing non-redundant mitochondrial Cytochrome C Oxidase subunit 1 (Cox1) gene references from 118 Venerid clam species. In six out of eight sequenced neoplastic specimens, the results revealed an overrepresentation (>99%) of reads in the sequenced tissues mapping to Cox1 DNA sequences that exclusively identified two different clam species (Figure 2a): the expected one, warty venus clam (V. verrucosa), and a second, unexpected one, the striped venus (C. gallina), a clam that inhabits the Mediterranean Sea (Figure 2b). Preliminary analysis by PCR and capillary sequencing of Cox1 in the haemolymph of two neoplastic specimens, EMVV18-373 and EVVV11-02, revealed an electropherogram with overlapping peaks apparently containing two different haplotypes that match the reference Cox1 sequences for warty and striped venus (Figure 2c).

Figure 2 with 2 supplements see all
Mitochondrial DNA sequencing and phylogenetic analyses reveal cancer contagion between warty venus (V. verrucosa) and striped venus (C. gallina) clam species.

(a) In eight warty venus specimens sequenced with Illumina paired-ends, the pie charts show the proportion of reads mapping Cox1 reference sequences from 137 different Verenidae species, including V. verrucosa (red), C. gallina (blue), and the remaining species (grey). Two different tissues were sequenced: the tumour tissue (left pie chart), typically haemolymph, and the host/matched-normal tissue (right pie chart), typically foot. Note that for specimen EVVV11-02 only the host/matched-normal tissue (gills) was available. ‘n’ denotes the total number of reads mapping the Cox1 reference for the tumour tissue (left), and the host tissue (right). (b) Representative specimen of the species C. gallina. (c) Capillary sequencing electropherograms of mitochondrial Cox1 gene fragments from two neoplastic V. verrucosa specimens (EMVV18-373 and EVVV11-02) and two healthy reference specimens from V. verrucosa and C. gallina. The results show overlapping peaks (arrows) in the sequenced tissues from the neoplastic animals, which suggest coexistence of mitochondrial DNA (mtDNA) haplotypes from two clam species. (d) In V. verrucosa neoplastic (N2-stage) specimen EMVV18-400, mtDNA read depth shows different proportion of warty venus and striped venus mtDNA haplotypes in the tumour tissue (haemolymph) and the matched-normal tissue (foot). (e) Molecular phylogeny using Bayesian inference inferred on the alignment of all mitochondria coding genes and rRNA gene sequences (15 loci) that includes six neoplastic V. verrucosa specimens with evidence of cancer contagion from C. gallina. Bootstrap values are shown above the branches.

Table 1
Clam specimens and tissues sequenced with Illumina paired-ends.

Sixteen specimens (eight neoplastic and eight non-neoplastic) from three different clam species (V. verrucosa, C. gallina, and C. striatula) were sequenced with Illumina paired-ends. Columns 5 and 6 show the number of reads generated for the host tissue (when neoplastic, matched-normal tissue was foot) and the tumoural haemolymph, respectively. (*) denotes the only available tissue from this neoplastic animal, collected in 2011, were gills. (#) denotes hemic neoplasia stage was not determined because cytohistological examination was not possible in this individual, which was diagnosed by cytogenetics.

Clam speciesSpecimen originSpecimen codeDiagnosisFoot readsHaemolymph reads
V. verrucosaGalicia, SpainERVV17-2995N3833 M919 M
V. verrucosaGalicia, SpainERVV17-2997N1766 M598 M
V. verrucosaGalicia, SpainERVV17-3193N3739 M850 M
V. verrucosaBalearic Islands, SpainEMVV18-376N1784 M849 M
V. verrucosaBalearic Islands, SpainEMVV18-391N3617 M623 M
V. verrucosaBalearic Islands, SpainEMVV18-395N3697 M679 M
V. verrucosaBalearic Islands, SpainEMVV18-400N1782 M1133 M
V. verrucosaGalicia, SpainEVVV11-02N#743 M*–*
V. verrucosaSplit, CroatiaCSVV18-1052Healthy161 M
V. verrucosaBalearic Islands, SpainEMVV18-385Healthy143 M
V. verrucosaGranville, FranceFGVV18-183Healthy752 M
V. verrucosaCarna, IrelandIGVV19-666Healthy155 M
V. verrucosaOeiras, PortugalPLVV18-2249Healthy163 M
C. gallinaS.Benedetto, ItalyIMCG15-69Healthy147 M
C. gallinaCadiz, SpainECCG15-201Healthy752 M
C. striatulaGalicia, SpainEVCS14-09Healthy706 M

These results suggested cancer contagion between the two clam species of the family Veneridae. Hence, to decipher the origins of this clam neoplasia, we further analysed the mitochondrial DNA (mtDNA) from the two species involved and the tumours. Firstly, we performed multiplatform genome sequencing, including Illumina short reads and Oxford Nanopore long reads, on canonical individuals from the two species to obtain a preliminary assembly of the mitogenomes of V. verrucosa and C. gallina. These reconstructions resulted in 18,092- and 17,618-bp long mtDNA genomes for the warty venus and the striped venus clam, respectively (Figure 2—figure supplement 1). The comparative analysis of the nucleotide sequences from both mitogenomes confirms that, although both species are relatively close within the subfamily Venerinae (Canapa et al., 1996), they represent distinct sister species, showing a Kimura’s two-parameter nucleotide distance (K2P) equal to 21.13%. Then, we mapped the paired-end sequencing data from the six neoplastic specimens with evidence of interspecies cancer transmission onto the two reconstructed species-specific mtDNA genomes. This approach confirmed the coexistence of two different mtDNA haplotypes in the six examined neoplastic samples, matching the canonical mtDNA genomes from the two clam species. For example, in a N2-stage specimen (EMVV18-400), this analysis revealed different proportion of tumour and host mtDNA molecules in the two tissue types sequenced (Figure 2d). Here, the striped venus mtDNA results the most abundant in the haemolymph, in which tumour cells are dominant over the remaining cell types, and the lower in the matched-normal tissue (i.e. infiltrated foot), where tumour cells represent a minor fraction of the total. Similar results were obtained for the remaining five neoplastic individuals (Figure 2—figure supplement 2).

To further investigate the evolutionary origins and geographic spread of this cancer, we sequenced with Illumina paired-ends an additional set of eight healthy (i.e. non-neoplastic) clams from three different Veneridae species, including five more warty venus specimens (EMVV18-385, IGVV19-666, FGVV18-183, CSVV18-1052, and PLVV18-2249) from five different countries, two striped venus specimens (IMCG15-69 and ECCG15-201) from two countries, and one specimen (EVCS14-09) from its sibling species Chamelea striatula, a type of striped venus clam that inhabits the Atlantic Ocean from Norway to the Gulf of Cadiz in Spain. This made a total of 16 Veneridae specimens sequenced, all listed in Table 1 (see also Supplementary file 1). The complete mitochondrial genomes from all tumoural and healthy V. verrucosa specimens (13 individuals), 2 C. gallina, and 1 from its sibling species C. striatula, were individually de novo assembled from the sequencing reads. As expected, this approach reconstructed two different haplotypes in six out eight sequenced neoplastic animals, supporting the presence of mtDNA from two different species. Despite the high sequencing coverage obtained for these individuals (Table 1), we did not find foreign reads in the N1 tumours (ERVV17-2997 and EMVV18-373), most likely due to a low proportion of neoplastic cells in the haemolymph and the matched-normal tissue. Then, we performed a phylogenetic analysis based on the alignment of these mitochondrial genomes (13 coding and 2 RNA gene sequences, altogether encompassing ~14 kb). The results show that tumour and non-tumour sequences from neoplastic warty venus specimens define two well-differentiated clades, and that tumoural warty venus sequences are all identical and closer to striped venus mtDNA than to its own (warty venus) (Figure 2e). Overall, these data support the existence of a single cancer clone originated in the striped venus clam C. gallina that was transmitted to V. verrucosa.

Transmissible cancers are known to occasionally acquire mitochondria from transient hosts (Strakova et al., 2016; Strakova et al., 2020), which can lead to misinterpretation of their evolutionary history. Thus, we looked for nuclear markers to confirm the striped venus origin of this cancer lineage. We performed a preliminary draft assembly of the warty venus and the striped venus nuclear ‘reference’ genomes, using the paired-end sequencing data from two non-neoplastic animals. Then, we used bioinformatic approaches to find single copy nuclear genes that were homologous between the two species, identifying two confident candidate genomic regions (see Methods): a 2.9-kb long region from DEAH12, a gene that encodes for an ATP-dependent RNA helicase, and a 2.2-kb long fragment from the Transcription Factor II Human-like gene, TFIIH. With the idea of finding differentially fixed single-nucleotide variants (SNVs) between both species, we performed PCR amplification and capillary sequencing on a 441 bp fragment from the DEAH12, and a 559 bp fragment from TFIIH, in 2 cohorts of non-neoplastic warty venus specimens (12 for DEAH12 and 15 for TFIIH), 2 cohorts of non-neoplastic striped venus (9 for DEAH12 and 12 for TFIIH), and 1 specimen of its sister species C. striatula. This analysis provided 14 and 15 sites, respectively, for the DEAH12 and the TFIIH loci, with fixed SNVs (allele frequency >95%) that allowed to discriminate between the 3 relevant species and the tumour (Figure 3a). These variants were employed to identify the Illumina reads from each sequenced warty venus neoplastic specimens that were specific for either warty venus or striped venus, which allowed to obtain the consensus sequences that corresponded to the tumour tissue and the non-affected tissue from each neoplastic individual. At the end of this process, we performed Maximum Likelihood phylogenetic reconstructions from these individual nuclear consensus sequences. On the one hand, the phylogeny for the DEAH12 locus confirmed both the monophyly of the tumoural sequences and their closer relationship to C. gallina than to the host species (Figure 3b), which were also observed in the mtDNA analysis. However, the phylogeny derived from the TFIIH locus showed that, although the tumours remained monophyletic, they were positioned in a basal branch relative to C. gallina and V. verrucosa (Figure 3b). Hence, to resolve these differences we also obtained a multilocus species tree based on the alignment of both the mtDNA and the two nuclear genes. This new phylogeny confirmed that warty venus tumours are closer to striped venus specimens than to non-neoplastic warty venus sequences from the same diseased specimens, while the non-neoplastic sequences conformed a more distant warty venus lineage (Figure 3c).

Figure 3 with 1 supplement see all
Nuclear DNA sequencing and phylogenetic analyses confirm a single cancer lineage spreading in populations of the warty venus (V. verrucosa) that originated in the striped venus (C. gallina).

(a) Single-nucleotide variants discriminating between V. verrucosa tumours and the three canonical species (V. verrucosa, C. gallina, and C. striatula) along a 441- and a 559-bp long fragments of nuclear genes DEAH12 and TFIIH, respectively. (b) Maximum Likelihood molecular phylogenies based on the two fragments of the nuclear DNA markers DEAH12 and TFIIH. Bootstrap support values (500 replicates) from Maximum Likelihood analyses above 50 are shown on the corresponding branches. (c) Multispecies coalescent (MSC) tree of V. verrucosa, their tumours and Chamelea sp. based on the entire mitochondrial DNA (mtDNA) and the two nuclear markers, DEAH12 and TFIIH. A maximum clade credibility (MCC) tree is shown, with posterior probabilities below the branches, and 95% highest probability density (HPD) intervals of node heights as grey bars. The trees distribution shown includes 1000 trees and represents the range of alternative topologies, in which blue is the most common set of topologies, red the second most common one, and green the remaining. (d) Fluorescence in situ hybridization (FISH) to specifically detect the satellite DNA CL4 in one V. verrucosa tumour and healthy specimens from the species C. gallina and V. verrucosa shows probes accumulate in heterochromatic regions, mainly in subcentromeric and subtelomeric positions, from the chromosomes of the tumour and the healthy C. gallina tested but not in healthy V. verrucosa.

To obtain further evidence on the striped venus origin of this clam’s neoplasia, we performed a comparative screening of tandem repeats in the genomes of C. gallina and V. verrucosa using fluorescence in situ hybridization (FISH) (Figure 3d; Figure 3—figure supplement 1). We focused on two satellite DNA repeats, namely CL4 and CL17. The satellites represent repeats of 332- and 429-bp long monomers, respectively, and were identified in a preliminary bioinformatics screening of the striped venus reference genome (see methods). This FISH approach revealed that the mentioned repeats are very abundant in heterochromatic regions from the genomes of the canonical striped venus and the neoplastic warty venus specimens tested (Figure 3d; Figure 3—figure supplement 1). However, the repeats were absent in the metaphases from all the healthy warty venus individuals (Figure 3d; Figure 3—figure supplement 1). These results suggest that the relevant chromosomes with CL4 and CL17 satellites found in neoplastic warty venus specimens derive from C. gallina, supporting that a tumour originated in C. gallina was transmitted to V. verrucosa.

To find out whether this cancer is present in the clam species where it first arose, we performed a screening for its presence in natural populations of striped venus clams from the species C. gallina (n = 213) and C. striatula (n = 9) at five additional sampling points across two countries (Supplementary file 1), including Spain (n = 115) and Italy (n = 107). Histological analyses did not show any traces of HN in these specimens. The virtual absence of this tumour in natural populations of striped venus clams may suggest that today this leukaemia is being mainly, if not exclusively, transmitted between specimens of the recipient species, warty venus. However, further sampling in other regions across the striped venus area of distribution may be necessary to confirm these findings.

Overall, the results provided here reveal the existence of a transmissible leukaemia originated in a striped venus clam, most likely C. gallina, which was transmitted to a second species, the warty venus clam (V. verrucosa), and among whose specimens it currently propagates. We identified this parasitic cancer in warty venus clams from two sampling points that are more than 1000 nautical miles away in the coasts of Spain, bathed by two different seas, the Atlantic Ocean and the Mediterranean Sea. The analysis of mitochondrial and nuclear gene sequences revealed no nucleotide diversity within the seven tumours sequenced, which supports that all belong to the same neoplastic lineage that spreads between Veneridae clams in the Seas of Southern Europe. Although we ignore the age of this cancer clone, we can confirm it arose before 2011, when the neoplastic warty venus specimen EVVV11-02 was collected. The apparent lack of genetic variation between all tumours, even from distant sampling points, suggests either that this cancer is very recent, or that it may have been unintentionally scattered by the action of man, a way of transmission that has been proposed for other bivalve transmissible cancers (Yonemitsu et al., 2019).

Materials and methods

Sampling of clam specimens

Request a detailed protocol

We collected 570 clam specimens from three different species, from the following countries and locations (Supplementary file 1). V. verrucosa clams were collected in Spain (Galicia, n = 90; Balearic Islands, n = 67), France (Granville, n = 100), Croatia (Split, n = 18), Portugal (Oeiras, n = 19), and Ireland (Carna, n = 50). C. gallina clams were collected in Spain (Cadiz, n = 50; Mallorca, n = 50) and Italy (Naples, n = 50; Cattolica, n = 57). C. striatula clams were collected in Spain (Combarro, n = 9). Additionally, we recruited samples from the following specimens from private collections: one V. verrucosa clam collected in 2011 in Spain (Islas Cies), four C. gallina collected in 2015 in Italy (San Benedetto de Tronto), five C. gallina collected in 2015 in Spain (Huelva), and one C. striatula collected in 2014 in Spain (Marin).

Diagnosis of HN

Request a detailed protocol

We followed standard cytological and/or histological protocols to test and diagnose HN in the clam specimens. However, only histological examination resulted decisive for the diagnosis, particularly in early stages of the disease. Briefly, for each animal, we extracted 300–2000 ml of haemolymph from the posterior adductor muscle using a 5 ml syringe with a 23 G needle. The haemolymph (50 ml) was diluted in cold Alserver’s antiaggregant solution to a 1:4 concentration, and spotted by centrifugation (130 × g, 4°C, 7 min) onto a microscope slide using cytology funnel sample chambers to produce a cell monolayer. Haemolymph smears were fixed and stained with Hemacolor solutions from Sigma-Aldrich and subsequently examined with a light microscope for the diagnosis of HN. Tissues (visceral mass, gills, mantle, and foot) were dissected, fixed in Davidson’s solution and embedded in paraffin. Then, 5-mm thick sections from each tissue were microdissected and stained with Harris’ haematoxylin and eosin and examined using a light microscope for histopathological analysis. HN was diagnosed and classified according to three disease stages (i.e. N1, N2, or N3) as follows. N1 stage: small groups of leukaemic cells were detected only in the vessels of the gills and in the connective tissue surrounding the digestive tubules. N2 stage: leukaemic cells spread to different organs, conforming small groups in the connective tissue that surrounds the digestive gland and the gonadal follicles, branchial sinuses, and mantle. N3 stage: leukaemic cells invade the filaments, completely deforming the plica structure in the gill, invade the connective tissue surrounding the gonadal follicles and the digestive gland; in the mantle, they invade the connective tissue, but in the muscle fibres of the mantle and foot, cells appear isolated or in small groups and in lower intensity than in other tissues.

Electron microscopy analysis

Request a detailed protocol

Four V. verrucosa specimens (two non-neoplastic, ERVV17-2993 and ERVV17-2992, and two with high grade of HN, ERVV17-2995 and ERVV17-3193) were processed for transmission electron microscopy as follows: 2 mm sections of gills and digestive glands were fixed in 2.5% glutaraldehyde seawater for 2 hr at 4°C. Then, tissues were post-fixed in 1% osmium tetroxide in sodium cacodylate solution and embedded in Epon resin. Ultrathin sections were stained with uranyl acetate and lead citrate and examined in a JEM-1010 transmission electron microscope.

Cytogenetics

Request a detailed protocol

Mitotic chromosomes of a neoplastic V. verrucosa specimen (EVVV11-02) were obtained as follows. After colchicine treatment (0.005%, 10 hr), gills were dissected, treated with a hypotonic solution, and fixed with ethanol and acetic acid. Small pieces of fixed gills were disaggregated with 60% acetic acid to obtain cell suspensions that were spread onto preheated slides. Chromosome preparations were stained with DAPI (0.14 mg/ml) and PI (0.07 mg/ml) for 8 min, mounted with antifade medium, and photographed. A comparative screening of tandem repeats was performed on the genomes of C. gallina and V. verrucosa using RepeatExplorer (Novák et al., 2010) on a merged short-read dataset of both species (500,000 reads each). Short reads of healthy and neoplastic animals were mapped onto both satellite consensus sequences using BWA, filtered according to their mapping quality (q > 60 and AS >70) and their abundance assessed by means of samtools/bamtools. Satellites CL4 and CL17 were selected for FISH purposes and FISH probes were PCR amplified (CL4F: TCAGAAACCGCTATTTTTCAC, CL4R: AAATGATGCTACGAACCTCC and CL17F: ATTCCAGAAATGTACATGAACAC, CL17R: ATTTTTGCACCAGATGTTCAC, respectively) and directly labelled with digoxigenin-11-dUTP (10× DIG Labeling Mix, Roche Applied Science). FISH experiments were performed as described in reference (García-Souto et al., 2015).

De novo assembly of mitochondrial genomes and annotation

Request a detailed protocol

In total, we performed whole-genome sequencing on 23 samples from 16 clam specimens, which includes 8 neoplastic and 8 non-neoplastic animals by Illumina paired-end libraries of 350 bp insert size and reads 150 bp long. First we assembled the mitochondrial genomes of one V. verrucosa (FGVV18_193), one C. gallina (ECCG15_201), and one C. striatula (EVCS14_02) specimens with MITObim v1.9.1 (Hahn et al., 2013), using gene baits from the following Cox1 and 16S reference genes to prime the assembly of clam mitochondrial genomes: V. verrucosa (Cox1, with GenBank accession number KC429139; and 16S: C429301), C. gallina (Cox1: KY547757, 16S: KY547777), and C. striatula (Cox1: KY547747, 16S: KY547767). These draft sequences were polished twice with Pilon v1.23 (Walker et al., 2014), and conflictive repetitive fragments from the mitochondrial control region were resolved using long read sequencing with Oxford Nanopore technologies (ONT) on a set of representative samples from each species and tumours. ONT reads were assembled with Miniasm v0.3 (Li, 2016) and corrected using Racon v1.3.1 (Vaser et al., 2017). Protein-coding genes, rDNAs and tDNAs were annotated on the curated mitochondrial genomes using MITOS2 web server (Bernt et al., 2013), and manually curated to fit ORFs as predicted by ORF-FINDER (Rombel et al., 2002). Then, we employed the entire mtDNAs of V. verrucosa (FGVV18_193) and C. gallina (ECCG15_201) as ‘references’ to map reads from individuals with neoplasia, filter reads matching either mitogenome and assemble and polish their two (healthy and tumoural) mitogenomes individually as above. Further healthy individuals were later sequenced and their mitogenomes assembled, to further investigate the geographic and taxonomic spread of this neoplasia.

Analysis of Cox1 sequences

Request a detailed protocol

We retrieved a dataset of 3745 sequences comprising all the barcode-identified venerid clam Cox1 fragments available from the Barcode of Life Data System (BOLD, http://www.boldsystemns.org/). Redundancy was removed using CD-HIT (Fu et al., 2012), applying a cut-off of 0.9 sequence identity, and sequences were trimmed to cover the same region. Whole-genome sequencing data from both healthy and tumoural warty venus clams were mapped onto this dataset, containing 118 venerid species-unique sequences, using BWA-mem, filtering out reads with mapping quality below 60 (-q60), and quantifying the overall coverage for each sequence with samtools idxstats. PCR primers were designed with Primer3 v2.3.7 (Kõressaar et al., 2018) to amplify a fragment of 354 bp from the Cox1 mitochondrial gene of V. verrucosa and C. gallina (F: CCT ATA ATA ATT GGK GGA TTT GG, R: CCT ATA ATA ATT GGK GGA TTT GG). PCR products were purified with ExoSAP-IT and sequenced by Sanger sequencing.

Mitochondrial genome coverage analysis

Request a detailed protocol

We further mapped the paired-end sequencing data from healthy and neoplastic tissues from all neoplastic samples onto the ‘reference’ mitochondrial genomes of V. verrucosa and C. gallina (two of the previously assembled ones, FGVV18_193 and ECCG15_201) using BWA-mem v0.7.17-r1188 (Li and Durbin, 2009) with default parameters. Duplicate reads were marked with Picard 2.18.14 and removed from the analysis. Read coverage depth was computed with samtools v1.9 (Li et al., 2009), summarized by computing the average in windows of 100 bp size and plotted with R v3.5.3.

Draft assembly of nuclear reference genomes, identification of variable single copy orthologous nuclear loci and SNPs

Request a detailed protocol

We ran the MEGAHIT v1.1.3 assembler (Li et al., 2015) on the Illumina paired-end sequencing data to obtain partial nuclear genome assemblies of V. verrucosa (FGVV18_193), C. gallina (ECCG15_201), and C. striatula (EVCS14_02). Then, single copy genes were predicted with Busco v.3.0.2 (Seppey et al., 2019). Candidate genes were considered if they (1) were present in the genomes of the three species, and (2) showed variant allele frequencies (VAFs) at exclusively 0, 0.5, or 1.0 in all the sequenced healthy (non-neoplastic) specimens. Under this criteria, two loci were finally selected: a 3914-bp long fragment of DEAH12, a gene encoding for an ATP-dependent RNA helicase and a 2.2-kp length fragment of the Transcription Factor II Human-like gene, TFIIH. PCR primers were designed with Primer3 v2.3.7 to amplify and sequence a 441-bp region of the DEAH12 nuclear gene (DEAH12_F: AGGTATGCTGAAACAAACACTT and DEAH12_R: ACGACAAATTTGATACCTGGAAT) and a 559-bp fragment of the TFIIH gene (TFIIH_F: TGGCATCTTTGTTACGGAC and TFIIH_R: CTTGTGRTTCTGTATCTGATCAATAA) on neoplastic specimens from V. verrucosa and healthy animals from both species (DEAH12: 11 V. verrucosa and 9 C. gallina; TFIIH: 15 V. verrucosa and 12 C. gallina). We screened for differentially fixed SNVs between both species using the dapc function in the R package Exploratory Analysis of Genetic and Genomic Data adegenet (Jombart and Ahmed, 2011). These variants were later employed to filter the Illumina short reads matching either V. verrucosa or C. gallina genotypes from the neoplastic animals, and to obtain consensus sequences from tumour and healthy tissue in each sequenced specimen. Read filtering was performed with samtools fillmd, while GATK mutect2 (Benjamin et al., 2019) was used for variant calling. Only variants with VAFs close to fixation (>0.9) were considered when building the consensus sequences.

Phylogenetic analyses

Request a detailed protocol

Mitochondrial sequences for 13 coding genes and 2 rDNA genes from the 23 recovered mitogenomes (6 neoplastic, 17 from host and healthy specimens) were extracted from the paired-end sequencing data by mapping reads onto the previously reconstructed canonical mtDNAs for V. verrucosa and C. gallina (Figure 2—figure supplement 1), concatenated, and subjected to multiple alignment with MUSCLE v3.8.425 (Edgar, 2004). The best-fit model of nucleotide substitution for each individual gene was selected using JModelTest2 (Darriba et al., 2012) and a partitioned Bayesian reconstruction of the phylogeny was performed with MrBayes v3.2.6 (Ronquist et al., 2012). Two independent Metropolis-coupled Markov Chain Monte Carlo (MCMC) analyses with four chains in each were performed. Each chain was run for 10 million generations, sampling trees every 1000 generations. Convergence of runs was assessed using Tracer (Rambaut et al., 2018).

DEAH12 and TFIIH sequences were subjected to multiple alignment using MUSCLE v3.8.425. Then, a ‘species/population tree’ was inferred with the starBEAST multispecies coalescent model, as implemented in BEAST v2.6.2 (Bouckaert et al., 2019). This analysis was performed using a Yule speciation prior and strict clock, with the best-fit model of nucleotide substitution obtained with jModelTest2 on both the concatenated mitochondrial haplotypes (13 protein-coding and 2 rRNAs genes) and unphased data from DEAH12 and TFIIH nuclear fragments. The four mitochondrial groups observed on the mitogenome analysis (V. verrucosa, C. gallina, C. striatula, and Tumour) were defined as tips for the species tree. A single MCMC of 10 million iterations, with sampling every 1000 steps, was run. A burn-in of 10% was implemented to obtain ESS values above 200 with Tracer v1.7.1 and the resulting posterior distributions of trees were checked with DENSITREE v2.1 (Bouckaert, 2010). A maximum clade credibility tree was obtained with TreeAnnotator (Bouckaert et al., 2019) to summarize information on topology, with 10% burn-in and Common Ancestors for the node heights.

Data availability

Nucleotide data for the mitochondrial DNA assemblies has been uploaded to GenBank under accession numbers MW662590-MW662611. These correspond to reference healthy animals of Venus verrucosa (MW662590, MW662593, MW662607, MW662608 and MW662610), Chamelea gallina (MW662591 and MW662609), Chamelea striatula (MW662611) and to both normal (MW662592, MW662595, MW662597, MW662599, MW662601, MW662602, MW662604, MW662606) and tumoral (MW662594, MW662596, MW662598, MW662600, MW662603, MW662605) mitochondrial DNAs of neoplastic animals. Genomic data (Illumina paired-end sequencing) from the mitogenomes analysed in this manuscript are allocated in DRYAD (https://doi.org/10.5061/dryad.zcrjdfn9v).

The following data sets were generated
    1. Garcia-Souto D
    2. Diaz-Costas S
    3. Bruzos AL
    4. Rocha S
    5. Pequeno A
    6. Roman-Lewis CF
    7. Costas D
    8. Rodriguez-Castro J
    9. Villanueva A
    10. Silva L
    11. Valencia JM
    12. Annona G
    13. Tarallo A
    14. Richardo F
    15. Bratos-Cetinic A
    16. Posada D
    17. Tubio JM
    (2022) GenBank
    ID MW662594. Venus verrucosa mitochondrion, complete genome EMVV18_391_Tumour Transmissible neoplasia derived from Chamelea gallina, neoplastic hemocytes.
    1. Garcia-Souto D
    2. Diaz-Costas S
    3. Bruzos AL
    4. Rocha S
    5. Pequeno A
    6. Roman-Lewis CF
    7. Costas D
    8. Rodriguez-Castro J
    9. Villanueva A
    10. Silva L
    11. Valencia JM
    12. Annona G
    13. Tarallo A
    14. Richardo F
    15. Bratos-Cetinic A
    16. Posada D
    17. Tubio JM
    (2022) NCBI GenBank
    ID MW662596. Venus verrucosa mitochondrion, complete genome EMVV18_395_Tumour Transmissible neoplasia derived from Chamelea gallina, neoplastic hemocytes.
    1. Garcia-Souto D
    2. Diaz-Costas S
    3. Bruzos AL
    4. Rocha S
    5. Pequeno A
    6. Roman-Lewis CF
    7. Costas D
    8. Rodriguez-Castro J
    9. Villanueva A
    10. Silva L
    11. Valencia JM
    12. Annona G
    13. Tarallo A
    14. Richardo F
    15. Bratos-Cetinic A
    16. Posada D
    17. Tubio JM
    (2022) NCBI GenBank
    ID MW662598. Venus verrucosa mitochondrion, complete genome EMVV18_400_Tumour Transmissible neoplasia derived from Chamelea gallina, neoplastic hemocytes.
    1. Garcia-Souto D
    2. Diaz-Costas S
    3. Bruzos AL
    4. Rocha S
    5. Pequeno A
    6. Roman-Lewis CF
    7. Costas D
    8. Rodriguez-Castro J
    9. Villanueva A
    10. Silva L
    11. Valencia JM
    12. Annona G
    13. Tarallo A
    14. Richardo F
    15. Bratos-Cetinic A
    16. Posada D
    17. Tubio JM
    (2022) NCBI GenBank
    ID MW662600. Venus verrucosa mitochondrion, complete genome ERVV17_2995_Tumour Transmissible neoplasia derived from Chamelea gallina, neoplastic hemocytes.
    1. Garcia-Souto D
    2. Diaz-Costas S
    3. Bruzos AL
    4. Rocha S
    5. Pequeno A
    6. Roman-Lewis CF
    7. Costas D
    8. Rodrigues-Castro J
    9. Villanueva A
    10. Silva L
    11. Valencia JM
    12. Annona G
    13. Tarallo A
    14. Ricardo F
    15. Bratos-Cetinic A
    16. Posado D
    17. Tubio JM
    (2022) NCBI GenBank
    ID MW662603. Venus verrucosa mitochondrion, complete genome ERVV17_3193_Tumour Transmissible neoplasia derived from Chamelea gallina, neoplastic hemocytes.
    1. Garcia-Souto D
    2. Diaz-Costa S
    3. Bruzos AL
    4. Rocha S
    5. Pequeno A
    6. Roman-Lewis CF
    7. Costas D
    8. Rodriguez-Castro J
    9. Villanueva A
    10. Silva L
    11. Valencia JM
    12. Annona G
    13. Tarallo A
    14. Richardo F
    15. Bratos-Cetinic A
    16. Posada D
    17. Tubio JM
    (2022) NCBI GenBank
    ID MW662605. Venus verrucosa mitochondrion, complete genome EVVV11_2_Tumour Transmissible neoplasia derived from Chamelea gallina, neoplastic hemocytes.

References

Decision letter

  1. Michael J Metzger
    Reviewing Editor; Pacific Northwest Research Institute, United States
  2. Detlef Weigel
    Senior Editor; Max Planck Institute for Developmental Biology, Germany
  3. Michael J Metzger
    Reviewer; Pacific Northwest Research Institute, United States
  4. Ariberto Fassati
    Reviewer; University College London, United Kingdom
  5. Nicolas Bierne
    Reviewer; Univ Montpellier, CNRS-EPHE-IRD, France

Our editorial process produces two outputs: i) public reviews designed to be posted alongside the preprint for the benefit of readers; ii) feedback on the manuscript for the authors, including requests for revisions, shown below. We also include an acceptance summary that explains what the editors found interesting or important about the work.

Decision letter after peer review:

Thank you for submitting your article "Mitogenome sequencing of marine leukemias reveals cancer contagion between clam species in the Seas of Southern Europe" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, including Michael J Metzger as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Detlef Weigel as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Ariberto Fassati (Reviewer #2); Nicolas Bierne (Reviewer #3).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

There are a few sections where there are either unclear methods or the methods do not quite match the descriptions of the results.

1. Further explanation is required for the methods of the initial analysis of cox1 sequences.

2. There appears to be inconsistency between the descriptions of the de novo mitogenome assembly in the methods and in the results. Please clarify what methods were done and which samples were used for de novo mitogenome assemblies.

3. The statement that no nucleotide divergence was identified needs further explanation. It is unclear whether or how the authors looked for any additional somatic mutations within the cancer lineage in the DEAH12 locus, as the methods only describe analysis of SNVs that were fixed within each normal host population. The methods for mtDNA sequence extraction were also unclear and it is unclear whether the authors looked for somatic SNVs in the intergenic regions of the mtDNA in addition to the gene regions specified in the text. Since high-coverage whole genome sequence was generated, the claim can be read that no nucleotide divergence exists within the whole genomes-the statement should clarify which regions showed no evidence of nucleotide divergence and the methods should explain how the authors looked for SNVs.

4. The data for N1 animals requires further explanation. In particular, the data in the pie charts in Figure 2 do not allow the determination of whether the C. gallina cancer is completely absent or present with a low number of reads. It would help to know whether the number of C. gallina reads was greater than that found in analysis of normal samples and whether the tumor samples had more cancer-associated reads than the host samples. If the data are not consistent with a low level of C. gallina, then alternate explanations should be provided.

5. The authors present short-read sequencing at a considerable depth. Please explain further why only a single nuclear gene was chosen and what the criteria were for this selection.

6. Please include discussion of the divergence between V. verrucosa and C. gallina in the mitochondrial and nuclear regions analyzed. This would allow an appreciation of the genetic distance between the original host and the current host species. Additionally, if the high divergence prevented mapping all reads to a single mtDNA reference to analyze VAFs, please clarify this.

Reviewer #1 (Recommendations for the authors):

The manuscript is straightforward and overall the data support the conclusions. The two methods issues should be clarified and the claim about the lack of divergence between the cancer samples should be clarified as well.

Reviewer #2 (Recommendations for the authors):

The study would be strengthened by:

1. Analysing samples ERVV17-2997 and EMVV18-376 in greater depth.

2. Expanding the SNPs analysis to a few additional nuclear genes.

3. Citing correctly in the Introduction the first paper demonstrating the clonal origin of CTVT (Murgia et al. Cell 2006).

Reviewer #3 (Recommendations for the authors):

1. The mitogenome analysis would probably be more convincing by plotting variant allele fractions rather than mapping depth. Could you please explain if this was not possible, or why you prefered to present alternative mapping depth to the two reference mitogenomes? Alternatively provides VAF plots that, I think, will be more straightforward to catch for readers.

2. I really didn't understand how you came with analysing only a single nuclear gene. I initially thought at my first reading that you did genome skimming, but your sequencing coverage is very good. Out of the mountain came a mouse. You really need to explain what happened.

You said, “showed variant allele frequencies (VAF) at exclusively 0, 0.5 or 1.0 in all the sequenced healthy (non-neoplastic) specimens”. I didn’t understand what was the constrained imposed here, could you explain? You said, "A single gene was selected by these criteria?" which criteria exactly? How do you explain you selected only one gene given the data you have?

3. At the end you have two non-recombining loci, mitogenome and DEAH12. The multispecies coalescent approach was developed to analyze many independent genealogies, not just two. To my point of view this analysis bring nothing useful, and rather could be badly perceived by specialists.

4. In the abstract you said "leukemia-like transmissible cancers, called hemic neoplasia". I agree we have four decades of research on hemic/disseminated neoplasia, but only genetic analysis can prove it is a transmissible cancer (exactly what you did in this study) and it was never done before Metzger et al.'s Cell and Nature papers in 2015 and 2016. In addition, non-transmissible neoplasia have been described (in mussels: Yonetmitsu et al. 2019 ELife, Hammel et al. 2021 BioRXiv). Therefore, we don't know if studies that described hemic/disseminated neoplasia before dealt with a transmissible cancer, and we cannot equal hemic/disseminated neoplasia with transmissible cancer.

5. "beyond the death of the individual that spawned them" I would not use "spawned".

6. " hemic neoplasia has a clonal transmissible behaviour (Metzger et al., 2015), in which the haemocytes (i.e., the cells that populate the haemolymph and play a role in the immune response) are likely to be transmitted through marine water."

As said above, not all hemic neoplasia are transmissible. In addition, neoplastic cells are not necessarily haemocytes (probably not I'd say, but we don't know indeed).

7. "animals die because of the infection". Remissions have been described (see e.g. Burioli et al. JIP in mussels).

8. "leukemic haemocytes" -> leukemic cells.

9. “representing interesting models for the understanding of the genetic causes of cancer transmissibility and metastasis” Why? It adds a new twist on transmissibility but does not help to understand the first transmission. And metastasis? I don’t see why.

10. “Marine transmissible cancers likely move using ocean currents to colonize new regions”.

It's a little bit too imagery I'd say. Given how water is diffusive and how cancer cells sediment, it is much more likely that transmissible cancers spread progressively from host to nearby host, by small step.

11. "using functions in the R package" which functions? makefreq?

12. "Mitochondrial sequences for 13 coding genes and two rDNA genes from 16 specimens (six neoplastic, 10 non-neoplastic) were extracted from the paired-end sequencing data" Explanations are needed here. How did you sort the two mitogenomes in cancerous individuals? Was differential mapping sufficient? Or did you call variants? Or did you assemble the two mitogenomes with megahit?

13. It would be nice to provide a phylogentic tree with more venerid species from public databases, eg using COI-16S genes only. Could be a supplementary figure.

https://doi.org/10.7554/eLife.66946.sa1

Author response

Essential revisions:

There are a few sections where there are either unclear methods or the methods do not quite match the descriptions of the results.

1. Further explanation is required for the methods of the initial analysis of cox1 sequences.

We retrieved a dataset of 3,745 sequences comprising all the barcode-identified venerid clam Cox1 fragments available from the Barcode of Life Data System (BOLD, http://www.boldsystemns.org/). Redundancy was removed using CD-HIT (Fu, et al. 2012), applying a cut-off of 0.9 sequence identity, and sequences were trimmed to cover the same region. Whole-genome sequencing data from both healthy and tumoral warty venus clams was mapped onto this dataset, containing 118 venerid species-unique sequences, using BWA-mem, filtering out reads with mapping quality below 60 (-q60) and quantifying the overall coverage for each sequence with samtools idxstats. PCR primers were designed with Primer3 v2.3.7 (Koressaar, et al. 2018) to amplify a fragment of 354 bp from the Cox1 mitochondrial gene of V. verrucosa and C. gallina (F: CCT ATA ATA ATT GGK GGA TTT GG, R: CCT ATA ATA ATT GGK GGA TTT GG). PCR products were purified with ExoSAP-IT and sequenced by Sanger sequencing.

We have included this new information in the methods section (page 22).

2. There appears to be inconsistency between the descriptions of the de novo mitogenome assembly in the methods and in the results. Please clarify what methods were done and which samples were used for de novo mitogenome assemblies.

In total, we performed whole-genome sequencing on 23 samples from 16 clam specimens, which includes eight neoplastic and eight non-neoplastic animals by Illumina paired-end libraries of 350 bp insert size and reads 150 bp long. First we assembled the mitochondrial genomes of one V. verrucosa (FGVV18_193), one C. gallina (ECCG15_201) and one C. striatula (EVCS14_02) specimens with MITObim v1.9.1 (Hahn, et al. 2013), using gene baits from the following Cox1 and 16S reference genes to prime the assembly of clam mitochondrial genomes: V. verrucosa (Cox1, with GenBank accession number KC429139; and 16S: C429301), C. gallina (Cox1: KY547757, 16S: KY547777) and C. striatula (Cox1: KY547747, 16S: KY547767). These draft sequences were polished twice with Pilon v1.23 (Walker, et al. 2014), and conflictive repetitive fragments from the mitochondrial control region were resolved using long read sequencing with Oxford Nanopore technologies (ONT) on a set of representative samples from each species and tumours. ONT reads were assembled with Miniasm v0.3 (Li 2016) and corrected using Racon v1.3.1 (Vaser, et al. 2017). Protein-coding genes, rDNAs and tRNAs were annotated on the curated mitochondrial genomes using MITOS2 web server (Bernt, et al. 2013), and manually curated to fit ORFs as predicted by ORF-FINDER (Rombel, et al. 2002). Then, we employed the entire mitochondrial DNAs of V. verrucosa (FGVV18_193) and C. gallina (ECCG15_201) as “references” to map reads from individuals with neoplasia, filter reads matching either mitogenome and assemble and polish their two (healthy and tumoral) mitogenomes individually as above. Further healthy individuals were later sequenced and their mitogenomes assembled, to further investigate the geographic and taxonomic spread of this neoplasia.

We have included this information in the methods section (page 21-22), and in the results (pages 7 and 8). mtDNA annotations are now shown in Figure 2—figure supplement 1. Nucleotide data for the mitochondrial DNA assemblies has been uploaded to GenBank under accession numbers MW662590-MW662611 and will be released upon publication or request.

3. The statement that no nucleotide divergence was identified needs further explanation. It is unclear whether or how the authors looked for any additional somatic mutations within the cancer lineage in the DEAH12 locus, as the methods only describe analysis of SNVs that were fixed within each normal host population. The methods for mtDNA sequence extraction were also unclear and it is unclear whether the authors looked for somatic SNVs in the intergenic regions of the mtDNA in addition to the gene regions specified in the text. Since high-coverage whole genome sequence was generated, the claim can be read that no nucleotide divergence exists within the whole genomes-the statement should clarify which regions showed no evidence of nucleotide divergence and the methods should explain how the authors looked for SNVs.

We obtained a preliminary nuclear assembly using short-reads only. Obviously, the resulting assemblies are fragmented and incomplete. This has limited the identification of candidate regions shared by the three genomes (V. verrucosa and both Chamelea clams). Out of the 44 candidate nuclear fragments we tested, only two (DEAH12 and TFHII) turned out to give good PCR products, adequate for Sanger sequencing. As mentioned above, we now provide additional data on a second gene (TFIIH), identified and selected on the same basis as DEAH12. We find 14 and 15 sites, respectively, for the DEAH12 and the TFIIH loci, with fixed SNVs (allele frequency >95%) that allowed to discriminate between the three relevant species (V. verrucosa, C. gallina and C. striatula) and the tumour. These diagnostic nucleotides were then used to filter the reads from individuals with neoplasia harbouring both DNA’s. Variation within the host lineage but not within the tumour was found along the nuclear DNA fragments employen in the ML phylogenies (see Author response image 1).

Author response image 1
Molecular phylogenies based on the two selected nuclear markers.

(a) DEAH12 gene and (b) TFIIH gene, and diagnostic loci discriminating among species and tumour. Bootstrap support values (500 replicates) from ML analyses above 50 are shown above the corresponding branches. Note all diagnostic nucleotides are identical between tumours (black dots).

Regarding the mtDNA, firstly, we assembled the mitochondrial genomes of one V. verrucosa (FGVV18_193), one C. gallina (ECCG15_201) and one C. striatula (EVCS14_02) specimens with MITObim v1.9.1 (Hahn, et al. 2013). Then, we employed the entire mitochondrial DNAs from V. verrucosa (FGVV18_193) and C. gallina (ECCG15_201) as “references” to map reads from individuals with neoplasia, filter reads matching either mitogenome and assemble and polish their two (healthy and tumoral) mitogenomes individually as above. Further healthy individuals were later sequenced and their mitogenomes assembled, to further investigate the geographic and taxonomic spread of this neoplasia. Despite the usefulness of the mitochondrial control region (CR) to detect differences among lineages, we refrained from using it for two reasons. (1) The CR shows considerable variation in both length and sequence among the three species, making their alignment difficult (in fact, previous phylogenetic studies based on whole mitochondrial DNA sequences in Veneridae excluded the CR: https://doi.org/10.1111/zsc.12454), and (2) the CR contains quasi-but-not-identical tandem repeats, as a other mollusks (i.e., the Venerid Dosinia clams https://doi.org/10.1371/journal.pone.0196466 or the Littorina marine snails https://doi.org/10.1016/j.margen.2016.10.006). In our case, repeats are larger than the short-reads insert size, and even though we could infer them by means of long read sequencing, polishing the resulting consensus sequences to overcome the intrinsic error rate of those lectures would yield inconclusive results, hindering the comparison between normal and tumoral haplotypes.

We updated the methods for the mitochondrial DNA analyses (pages 21-22) and the nuclear DNA analyses (page 23). We now include new data in the results and discussion (pages 9-10).

4. The data for N1 animals requires further explanation. In particular, the data in the pie charts in Figure 2 do not allow the determination of whether the C. gallina cancer is completely absent or present with a low number of reads. It would help to know whether the number of C. gallina reads was greater than that found in analysis of normal samples and whether the tumor samples had more cancer-associated reads than the host samples. If the data are not consistent with a low level of C. gallina, then alternate explanations should be provided.

Despite being sequenced at high coverage (30 Gb), we found no C. gallina reads in the sequenced N1 animals (ERVV17-2997 and ERVV18-373), most likely due to the presence of a low proportion of neoplastic cells in the haemolymph and the matched-normal tissue relative to normal cells. This is something common in some N1 tumours from hemic neoplasia, where only a few neoplastic cells are found.

We now provide more details on page 8.

5. The authors present short-read sequencing at a considerable depth. Please explain further why only a single nuclear gene was chosen and what the criteria were for this selection.

We obtained a preliminary nuclear assembly using short-reads only. Obviously, the resulting assemblies are fragmented and incomplete. This has limited the identification of candidate regions shared by the three genomes (V. verrucosa and both Chamelea clams). Out of the 44 candidate nuclear fragments we tested, only two (DEAH12 and TFHII) turned out to give good PCR products, adequate for Sanger sequencing. As mentioned above, we now provide additional data on a second gene (TFIIH), identified and selected on the same basis as DEAH12. Individual ML phylogenies for these two fragments evidenced that tumours cluster together and separately from the host species and, in the case of DEAH12, closer to C. gallina. The MSC phylogeny was rebuilt including this new nuclear fragment.

In addition, we conducted a comparative screening of tandem repeats on the genomes of C. gallina and V. verrucosa. Two DNA satellites, namely CL4 and CL17, of, respectively, 332 and 429 bp monomer size, were very abundant in C. gallina and in the tumoral animals, but absent from all healthy V. verrucosa specimens. FISH probes designed for these satellites mapped on the heterochromatic regions, mainly in subcentromeric and subtelomeric positions, of both C. gallina and the neoplastic metaphases found in V. verrucosa, but were absent from the normal metaphases of the host species V. verrucosa. These results were consistent with the genomic abundance of these satellites in the NGS data and strongly suggest that these chromosomes derive from C. gallina.

We include the analysis of one additional nuclear locus, TFIIH (pages 9-10). We have obtained new ML and MSC phylogenies including this new locus (pages 9-10, figures 3b-c). Additional FISH approach looking for satellite DNA CL4 and CL11 was performed (page 10, figure 3d, Figure 3—figure supplement 1). The methods section has been updated accordingly (pages 20-21, 23-24).

6. Please include discussion of the divergence between V. verrucosa and C. gallina in the mitochondrial and nuclear regions analyzed. This would allow an appreciation of the genetic distance between the original host and the current host species. Additionally, if the high divergence prevented mapping all reads to a single mtDNA reference to analyze VAFs, please clarify this.

While the taxonomy and phylogeny of Venerids are under debate, C. gallina and V. verrucosa are well-differentiated species. Some authors consider that they belong to different subfamilies according to their external morphology, Chioninae and Venerinae (see World Register of Marine Species), while others consider that they pertain to Venerinae (Huber’s Compendium of Bivalves; Canappa et al. 1996; Chen et al. 2011). We have now estimated the genetic distance between these two species (23% in the assembled mtDNA fragment) and performed the following molecular phylogeny of 2045 venerids non-redundant COI gene sequences (see Author response image 2).

Author response image 2
Molecular phylogenetic analysis by Maximum Likelihood method across Venerids.

a) Maximum likelihood (ML) molecular phylogenetic tree of Veneridae based on a 323 bp alignment including 2045 non-redundant COI gene sequences recovered from Genbank/Boldsystems and all sequences derived from this work. Highlighted clades A and B according to Chen et al. 2011. Bootstrap support values (500 replicates) from ML analyses above 50 are shown above the corresponding branches. b) Zoom-in to the cluster of the venerid harbouring the neoplastic haplotype, both Chamelea species and Venus verrucosa in the previous tree (asterisk).

We have included new information on page 7.

Reviewer #2 (Recommendations for the authors):

The study would be strengthened by:

1. Analysing samples ERVV17-2997 and EMVV18-376 in greater depth.

Unfortunately, we were not able to perform further sequencing on these samples. We believe the absence of sequencing reads from the tumours is a consequence of a low proportion of neoplastic cells in the host tissues. Despite the high sequencing coverage obtained for the sequenced individuals, we did not find foreign reads in the N1 tumours (ERVV17-2997 and EMVV18-73) to mitochondrial nor nuclear (i.e., DEAH12, TFHII) level. This is most likely due to a very low proportion of neoplastic cells in their tissues.

We have added a sentence on page 8 that discuss this issue.

2. Expanding the SNPs analysis to a few additional nuclear genes.

We obtained a preliminary nuclear assembly using short-reads only. Obviously, the resulting assemblies are fragmented and incomplete. This has limited the identification of candidate regions shared by the three genomes (V. errucose and both Chamelea clams). Out of the 44 candidate nuclear fragments we tested, only two (DEAH12 and TFHII) turned out to give good PCR products, adequate for Sanger sequencing. As mentioned above, we now provide additional data on a second gene (TFIIH), identified and selected on the same basis as DEAH12. Individual ML phylogenies for these two fragments evidenced that tumours cluster together and separately from the host species and, in the case of DEAH12, closer to C. gallina. The MSC phylogeny was rebuilt including this new nuclear fragment.

In addition, we conducted a comparative screening of tandem repeats on the genomes of C. gallina and V. errucose. Two DNA satellites, namely CL4 and CL17, of, respectively, 332 and 429 bp monomer size, were very abundant in C. gallina and in the tumoral animals, but absent from all healthy V. errucose specimens. FISH probes designed for these satellites mapped on the heterochromatic regions, mainly in subcentromeric and subtelomeric positions, of both C. gallina and the neoplastic metaphases found in V. errucose, but were absent from the normal metaphases of the host species V. errucose. These results were consistent with the genomic abundance of these satellites in the NGS data and strongly suggest that these chromosomes derive from C. gallina.

We include the analysis of one additional nuclear locus, TFIIH (pages 9-10). We have obtained new ML and MSC phylogenies including this new locus (pages 9-10, figures 3b-c). Additional FISH approach looking for satellite DNA CL4 and CL11 was performed (page 10, figure 3d, Figure 3—figure supplement 1). The methods section has been updated accordingly (pages 20-21, 23-24).

3. Citing correctly in the Introduction the first paper demonstrating the clonal origin of CTVT (Murgia et al. Cell 2006).

The reference is now included (page 4).

Reviewer #3 (Recommendations for the authors):

1. The mitogenome analysis would probably be more convincing by plotting variant allele fractions rather than mapping depth. Could you please explain if this was not possible, or why you errucose to present alternative mapping depth to the two reference mitogenomes? Alternatively provides VAF plots that, I think, will be more straightforward to catch for readers.

Variant allele frequency (VAF) plots are a good approach to see the proportions of two cell types from the same species mapping onto the same genome. However, in this case, tumor and host cells do not map onto the same genome (mitochondrial genomes from these two species are far too divergent from each other, showing K2P nucleotide distance of 21.13%). In each tissue (haemolymph or foot) some read sequences are closer to the C. gallina (allegedly tumor cells) and others to V. errucose (ie. Host cells). Hence, VAF values will be mostly ~1 (haploid genome), not contributing to support or reject the hypothesis (see Author response image 3). However, mapping depth plots show how the proportion of reads of each tissue increase and decrease depending on the reference genome where the reads map (see Figure 2D).

Author response image 3
For the warty venus neoplastic specimen EMVV18-400, the figure shows the comparison of the effects of plotting the variant allele frequency (VAF).

We included a sentence about the divergence between mtDNA genomes on page 7.

2. I really didn’t understand how you came with analysing only a single nuclear gene. I initially thought at my first reading that you did genome skimming, but your sequencing coverage is very good. Out of the mountain came a mouse. You really need to explain what happened.

We obtained a preliminary nuclear assembly using short-reads only. Obviously, the resulting assemblies are fragmented and incomplete. This has limited the identification of candidate regions shared by the three genomes (V. errucose and both Chamelea clams). Out of the 44 candidate nuclear fragments we tested, only two (DEAH12 and TFHII) turned out to give good PCR products, adequate for Sanger sequencing. As mentioned above, we now provide additional data on a second gene (TFIIH), identified and selected on the same basis as DEAH12. Individual ML phylogenies for these two fragments evidenced that tumours cluster together and separately from the host species and, in the case of DEAH12, closer to C. gallina. The MSC phylogeny was rebuilt including this new nuclear fragment.

In addition, we conducted a comparative screening of tandem repeats on the genomes of C. gallina and V. errucose. Two DNA satellites, namely CL4 and CL17, of, respectively, 332 and 429 bp monomer size, were very abundant in C. gallina and in the tumoral animals, but absent from all healthy V. errucose specimens. FISH probes designed for these satellites mapped on the heterochromatic regions, mainly in subcentromeric and subtelomeric positions, of both C. gallina and the neoplastic metaphases found in V. errucose, but were absent from the normal metaphases of the host species V. errucose. These results were consistent with the genomic abundance of these satellites in the NGS data and strongly suggest that these chromosomes derive from C. gallina.

We include the analysis of one additional nuclear locus, TFIIH (pages 9-10). We have obtained new ML and MSC phylogenies including this new locus (pages 9-10, figures 3b-c). Additional FISH approach looking for satellite DNA CL4 and CL11 was performed (page 10, figure 3d, Figure 3—figure supplement 1). The methods section has been updated accordingly (pages 20-21, 23-24).

You said, “showed variant allele frequencies (VAF) at exclusively 0, 0.5 or 1.0 in all the sequenced healthy (non-neoplastic) specimens”. I didn’t understand what was the constrained imposed here, could you explain? You said, “A single gene was selected by these criteria?” which criteria exactly? How do you explain you selected only one gene given the data you have?

We now provide additional data on a second gene (TFHII), identified and selected on the same basis as DEAH12. Candidate genes were considered if they (1) were present in the genomes of the three species, and (2) showed variant allele frequencies (VAF) at exclusively 0, 0.5 or 1.0 in all the sequenced healthy (non-neoplastic) specimens. Regarding the 0, 0.5, 1.0 frequencies, we are looking for single-copy genes in the context of a diploid genome. Thus, for a given single-nucleotide variant (SNV) the allele frequency indicates the relative number of copies of one allele relative to the other. Homozygous SNVs have allele frequencies near 0 (AA) or 1 (BB). Heterozygous two-copy have allele frequencies near 0.5 (AB). Allelic imbalance results in intermediate values. For example, a SNV present in three copies will have four possible genotypes (AAA, AAB, ABB, BBB), thus giving possible allele frequencies of 0, 0.33, 0.67 or 1.

These criteria are defined in methods (page 23).

3. At the end you have two non-recombining loci, mitogenome and DEAH12. The multispecies coalescent approach was developed to analyze many independent genealogies, not just two. To my point of view this analysis bring nothing useful, and rather could be badly perceived by specialists.

As mentioned above (point 3), we now added additional data on another nuclear marker, as well as on two satellite DNAs. The multispecies coalescent approach can be applied to any number of loci, starting with one. A different thing is that the accuracy of the species tree estimate logically increases with the number of independent loci. Due to potential incomplete lineage sorting, we still prefer this kind of approach for two (now three) loci than a concatenated tree. Please, look at the single-locus analysis in Figure 3 and the preceding paragraph from the authors of BEAST at https://academic.oup.com/mbe/article/29/8/1969/1044583.

No specific action was taken for this query. However, we are happy to discuss this further if required by the reviewer.

4. In the abstract you said “leukemia-like transmissible cancers, called hemic neoplasia”. I agree we have four decades of research on hemic/disseminated neoplasia, but only genetic analysis can prove it is a transmissible cancer (exactly what you did in this study) and it was never done before Metzger et al.’s Cell and Nature papers in 2015 and 2016. In addition, non-transmissible neoplasia have been described (in mussels: Yonetmitsu et al. 2019 Elife, Hammel et al. 2021 BioRXiv). Therefore, we don’t know if studies that described hemic/disseminated neoplasia before dealt with a transmissible cancer, and we cannot equal hemic/disseminated neoplasia with transmissible cancer.

We agree with the reviewer that not all hemic neoplasia tumours are transmissible. We cannot change the abstract due to length restrictions, but we have changed the introduction to make this clear (“Some HNs have been proven to have a clonal transmissible behaviour…”).

We include a new sentence on page 4.

5. “beyond the death of the individual that spawned them” I would not use “spawned”.

Now the text reads as follows “…meaning that they can survive beyond the death of their hosts…” (page 4).

6. “ hemic neoplasia has a clonal transmissible behaviour (Metzger et al., 2015), in which the haemocytes (i.e., the cells that populate the haemolymph and play a role in the immune response) are likely to be transmitted through marine water.”

As said above, not all hemic neoplasia are transmissible. In addition, neoplastic cells are not necessarily haemocytes (probably not I’d say, but we don’t know indeed).

Now the text reads as follows “…in which neoplastic cells, most likely haemocytes…” (page 4).

7. “animals die because of the infection”. Remissions have been described (see e.g. Burioli et al. JIP in mussels).

Now the text reads as follows “…although remissions have also been described (Burioli, et al. 2019)…” (page 4).

8. “leukemic haemocytes” -> leukemic cells.

Now the text reads as follows “…Despite the observation that leukemic cells are typically transmitted…” (page 4).

9. "representing interesting models for the understanding of the genetic causes of cancer transmissibility and metastasis" Why? It adds a new twist on transmissibility but does not help to understand the first transmission. And metastasis? I don't see why.

This has been removed from pages 4-5.

10. "Marine transmissible cancers likely move using ocean currents to colonize new regions".

It's a little bit too imagery I'd say. Given how water is diffusive and how cancer cells sediment, it is much more likely that transmissible cancers spread progressively from host to nearby host, by small step.

This has been removed from pages 4-5.

11. "using functions in the R package" which functions? makefreq?

We screened for differentially fixed single-nucleotide variants between both species using the dapc function in the R package Exploratory Analysis of Genetic and Genomic Data adegenet.

A new sentence has been included on page 23.

12. "Mitochondrial sequences for 13 coding genes and two rDNA genes from 16 specimens (six neoplastic, 10 non-neoplastic) were extracted from the paired-end sequencing data" Explanations are needed here. How did you sort the two mitogenomes in cancerous individuals? Was differential mapping sufficient? Or did you call variants? Or did you assemble the two mitogenomes with megahit?

All healthy animals were assembled de novo using Mitobin and polished with Pilon. Two of them, C. gallina (ECCG15_201) and V. verrucosa (FGVV18_193), were initially obtained and used as representative sequences onto which short reads of the tumoral animals were mapped to. This differential mapping was enough to filter reads matching either mitogenome, and extract them from the original fastq files for all tumoral animals and assemble and polish neoplastic and normal mitogenomes individually for each specimen with neoplasia. Then, further healthy individuals (from V. verrucosa, C. gallina and C. striatula) were sequenced to investigate the evolutionary origins of this cancer.

All sequenced specimens/tissues are listed in Table 1. We also revised this in the results (pages 7-8) and methods (pages 21-22).

13. It would be nice to provide a phylogentic tree with more venerid species from public databases, eg using COI-16S genes only. Could be a supplementary figure.

While the taxonomy and phylogeny of Venerids are under debate, C. gallina and V. verrucosa are well-differentiated species. Some authors consider that they belong to different subfamilies according to their external morphology, Chioninae and Venerinae (see World Register of Marine Species), while others consider that they pertain to Venerinae (Huber’s Compendium of Bivalves; Canappa et al. 1996; Chen et al. 2011). We have now estimated the genetic distance among these two species (23% in the mtDNA). As for the phylogeny, we honestly believe this is a bit tricky to discuss in the main text, because it would break the flow of the story, and it would be out of the scope of this manuscript. However, we have performed the phylogeny, which is presented here for your review (see Author response image 2).

We have included new information on page 7.

https://doi.org/10.7554/eLife.66946.sa2

Article and author information

Author details

  1. Daniel Garcia-Souto

    1. Genomes and Disease, Centre for Research in Molecular Medicine and Chronic Diseases (CIMUS), Universidade de Santiago de Compostela, Santiago de Compostela, Spain
    2. Department of Zoology, Genetics and Physical Anthropology, Universidade de Santiago de Compostela, Santiago de Compostela, Spain
    3. Cancer Ageing and Somatic Mutation Programme, Wellcome Sanger Institute, Cambridge, United Kingdom
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review and editing
    Contributed equally with
    Alicia L Bruzos and Seila Diaz
    For correspondence
    daniel.garcia.souto@usc.es
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-0997-8799
  2. Alicia L Bruzos

    1. Genomes and Disease, Centre for Research in Molecular Medicine and Chronic Diseases (CIMUS), Universidade de Santiago de Compostela, Santiago de Compostela, Spain
    2. Department of Zoology, Genetics and Physical Anthropology, Universidade de Santiago de Compostela, Santiago de Compostela, Spain
    Contribution
    Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Validation, Writing – original draft
    Contributed equally with
    Daniel Garcia-Souto and Seila Diaz
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-4362-545X
  3. Seila Diaz

    Genomes and Disease, Centre for Research in Molecular Medicine and Chronic Diseases (CIMUS), Universidade de Santiago de Compostela, Santiago de Compostela, Spain
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft
    Contributed equally with
    Daniel Garcia-Souto and Alicia L Bruzos
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-6607-290X
  4. Sara Rocha

    Phylogenomics Lab, Universidade de Vigo, Vigo, Spain
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-9705-1511
  5. Ana Pequeño-Valtierra

    Genomes and Disease, Centre for Research in Molecular Medicine and Chronic Diseases (CIMUS), Universidade de Santiago de Compostela, Santiago de Compostela, Spain
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – original draft, Writing – review and editing
    Competing interests
    No competing interests declared
  6. Camila F Roman-Lewis

    CINBIO, Universidade de Vigo, Vigo, Spain
    Contribution
    Investigation, Methodology
    Competing interests
    No competing interests declared
  7. Juana Alonso

    1. CINBIO, Universidade de Vigo, Vigo, Spain
    2. Galicia Sur Health Research Institute (IIS Galicia Sur), SERGAS-UVIGO, Vigo, Spain
    Contribution
    Investigation, Methodology
    Competing interests
    No competing interests declared
  8. Rosana Rodriguez

    Centro de Investigación Mariña, Universidade de Vigo, ECIMAT, Vigo, Spain
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – original draft, Writing – review and editing
    Competing interests
    No competing interests declared
  9. Damian Costas

    Centro de Investigación Mariña, Universidade de Vigo, ECIMAT, Vigo, Spain
    Contribution
    Methodology, Project administration, Resources
    Competing interests
    No competing interests declared
  10. Jorge Rodriguez-Castro

    Genomes and Disease, Centre for Research in Molecular Medicine and Chronic Diseases (CIMUS), Universidade de Santiago de Compostela, Santiago de Compostela, Spain
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – original draft, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-2912-9601
  11. Antonio Villanueva

    Centro de Investigación Mariña, Universidade de Vigo, ECIMAT, Vigo, Spain
    Contribution
    Methodology, Project administration, Resources
    Competing interests
    No competing interests declared
  12. Luis Silva

    Instituto Español de Oceanografía (IEO), Centro Oceanográfico de Cádiz, Cádiz, Spain
    Contribution
    Methodology, Project administration, Resources, Writing – review and editing
    Competing interests
    No competing interests declared
  13. Jose Maria Valencia

    1. Laboratori d’Investigacions Marines i Aqüicultura, (LIMIA) - Govern de les Illes Balears, Port d'Andratx, Balearic Islands, Spain
    2. Instituto de Investigaciones Agroambientales y de Economía del Agua (INAGEA) (INIA-CAIB-UIB), Palma de Mallorca, Balearic Islands, Spain
    Contribution
    Resources, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-1912-1975
  14. Giovanni Annona

    Stazione Zoologica Anton Dohrn, Napoli, Italy
    Contribution
    Resources, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-7806-6761
  15. Andrea Tarallo

    Stazione Zoologica Anton Dohrn, Napoli, Italy
    Contribution
    Resources, Writing – review and editing
    Competing interests
    No competing interests declared
  16. Fernando Ricardo

    ECOMARE, Centre for Environmental and Marine Studies (CESAM), Department of Biology, University of Aveiro, Santiago University Campus, Aveiro, Portugal
    Contribution
    Resources, Writing – review and editing
    Competing interests
    No competing interests declared
  17. Ana Bratoš Cetinić

    Department of Aquaculture, University of Dubrovnik, Dubrovnik, Croatia
    Contribution
    Resources, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-2928-4858
  18. David Posada

    1. CINBIO, Universidade de Vigo, Vigo, Spain
    2. Galicia Sur Health Research Institute (IIS Galicia Sur), SERGAS-UVIGO, Vigo, Spain
    3. Department of Biochemistry, Genetics and Immunology, Universidade de Vigo, Vigo, Spain
    Contribution
    Investigation, Methodology, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-1407-3406
  19. Juan Jose Pasantes

    1. Department of Biochemistry, Genetics and Immunology, Universidade de Vigo, Vigo, Spain
    2. Centro de Investigación Mariña, Universidade de Vigo, Vigo, Spain
    Contribution
    Investigation, Methodology, Resources, Supervision, Writing – review and editing
    Competing interests
    No competing interests declared
  20. Jose MC Tubio

    1. Genomes and Disease, Centre for Research in Molecular Medicine and Chronic Diseases (CIMUS), Universidade de Santiago de Compostela, Santiago de Compostela, Spain
    2. Department of Zoology, Genetics and Physical Anthropology, Universidade de Santiago de Compostela, Santiago de Compostela, Spain
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – original draft, Writing – review and editing
    For correspondence
    jose.mc.tubio@usc.es
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-3540-2459

Funding

H2020 European Research Council (716290)

  • Jose MC Tubio

MINECO PhD fellowship (BES-2016–078166)

  • Alicia L Bruzos

Xunta de Galicia (ED481B/2018/091)

  • Daniel Garcia-Souto

European Research Council (617457)

  • David Posada

Spanish Ministry of Economy and Competitiveness (MINECO) (PID2019-106247GB-I00)

  • David Posada

European Union’s Horizon 2020 research and innovation programme (730984)

  • Seila Diaz

CESAM (UIDP/50017/2020 + UIDB/50017/2020)

  • Fernando Ricardo

The funders had no role in study design, data collection, and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank the Galicia Supercomputing Centre (CESGA) for the availability of informatic resources. JMCT, SR, SD, and JT are supported by European Research Council (ERC) Starting Grant 716,290 SCUBA CANCERS. ALB is supported by MINECO PhD fellowship BES-2016-078166. DG-S is supported by postdoctoral contract ED481B/2018/091 from Xunta de Galicia. DP is supported by ERC grant ERC-617457-PHYLOCANCER and by Spanish Ministry of Economy and Competitiveness (MINECO) grant PID2019-106247GB-I00. This research was partially funded by the European Union’s Horizon 2020 research and innovation programme under grant agreement 730984, ASSEMBLE Plus project. CESAM got financial support from FCT/MEC (UIDP/50017/2020, UIDB/50017/2020).

Ethics

No particular ethical aspects are necessary for bivalves research in the laboratory.

Senior Editor

  1. Detlef Weigel, Max Planck Institute for Developmental Biology, Germany

Reviewing Editor

  1. Michael J Metzger, Pacific Northwest Research Institute, United States

Reviewers

  1. Michael J Metzger, Pacific Northwest Research Institute, United States
  2. Ariberto Fassati, University College London, United Kingdom
  3. Nicolas Bierne, Univ Montpellier, CNRS-EPHE-IRD, France

Publication history

  1. Received: January 27, 2021
  2. Preprint posted: March 10, 2021 (view preprint)
  3. Accepted: December 4, 2021
  4. Version of Record published: January 18, 2022 (version 1)
  5. Version of Record updated: January 27, 2022 (version 2)

Copyright

© 2022, Garcia-Souto et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,563
    Page views
  • 150
    Downloads
  • 7
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Daniel Garcia-Souto
  2. Alicia L Bruzos
  3. Seila Diaz
  4. Sara Rocha
  5. Ana Pequeño-Valtierra
  6. Camila F Roman-Lewis
  7. Juana Alonso
  8. Rosana Rodriguez
  9. Damian Costas
  10. Jorge Rodriguez-Castro
  11. Antonio Villanueva
  12. Luis Silva
  13. Jose Maria Valencia
  14. Giovanni Annona
  15. Andrea Tarallo
  16. Fernando Ricardo
  17. Ana Bratoš Cetinić
  18. David Posada
  19. Juan Jose Pasantes
  20. Jose MC Tubio
(2022)
Mitochondrial genome sequencing of marine leukaemias reveals cancer contagion between clam species in the Seas of Southern Europe
eLife 11:e66946.
https://doi.org/10.7554/eLife.66946

Further reading

    1. Cancer Biology
    2. Computational and Systems Biology
    Pan Cheng, Xin Zhao ... Teresa Davoli
    Research Article

    How cells control gene expression is a fundamental question. The relative contribution of protein-level and RNA-level regulation to this process remains unclear. Here, we perform a proteogenomic analysis of tumors and untransformed cells containing somatic copy number alterations (SCNAs). By revealing how cells regulate RNA and protein abundances of genes with SCNAs, we provide insights into the rules of gene regulation. Protein complex genes have a strong protein-level regulation while non-complex genes have a strong RNA-level regulation. Notable exceptions are plasma membrane protein complex genes, which show a weak protein-level regulation and a stronger RNA-level regulation. Strikingly, we find a strong negative association between the degree of RNA-level and protein-level regulation across genes and cellular pathways. Moreover, genes participating in the same pathway show a similar degree of RNA- and protein-level regulation. Pathways including translation, splicing, RNA processing, and mitochondrial function show a stronger protein-level regulation while cell adhesion and migration pathways show a stronger RNA-level regulation. These results suggest that the evolution of gene regulation is shaped by functional constraints and that many cellular pathways tend to evolve one predominant mechanism of gene regulation at the protein level or at the RNA level.

    1. Cancer Biology
    2. Chromosomes and Gene Expression
    Arnaud Carrier, Cécile Desjobert ... Paola B Arimondo
    Research Article

    Aberrant DNA methylation is a well‑known feature of tumours and has been associated with metastatic melanoma. However, since melanoma cells are highly heterogeneous, it has been challenging to use affected genes to predict tumour aggressiveness, metastatic evolution, and patients' outcomes. We hypothesized that common aggressive hypermethylation signatures should emerge early in tumorigenesis and should be shared in aggressive cells, independent of the physiological context under which this trait arises. We compared paired melanoma cell lines with the following properties: (i) each pair comprises one aggressive counterpart and its parental cell line, and (ii) the aggressive cell lines were each obtained from different host and their environment (human, rat, and mouse), though starting from the same parent cell line. Next, we developed a multi-step genomic pipeline that combines the DNA methylome profile with a chromosome cluster-oriented analysis. A total of 229 differentially hypermethylated genes were commonly found in the aggressive cell lines. Genome localization analysis revealed hypermethylation peaks and clusters, identifying eight hypermethylated gene promoters for validation in tissues from melanoma patients. Five CpG identified in primary melanoma tissues were transformed into a DNA methylation score that can predict survival (Log-rank test, p=0.0008). This strategy is potentially universally applicable to other diseases involving DNA methylation alterations.