AT LEAST TWO CONTAGIOUS CANCER CELL LINEAGES CAN TRANSMIT DISSEMINATED NEOPLASIA IN COCKLES Cerastoderma edule  

Antonio Villalba*, Michael J. Metzger, David Iglesias, María J. Carballal, Stephen P. Goff
 
Centro de Investigacións Mariñas
Consellería do Medio Rural e do Mar da Xunta de Galicia
Aptdo. 13, 36620 Vilanova de Arousa, Spain
antonio.villalba.garcia@xunta.es

Disseminated neoplasia (DN) has been reported in cockles from Ireland, France and Spain. This disease is hyperendemic in some cockle beds of Galicia (NW Spain), causing high mortality episodes. Transmission of DN from diseased to healthy cockles was demonstrated. Both disease dynamics and its transmissibility suggested infectious etiology; association of DN with reverse transcriptase activity led to the hypothesis of the involvement of a retrovirus. However, recent findings of the retrotransposon Steamer as responsible of reverse transcriptase activity in leukemia-affected soft-shell clams Mya arenaria as well as the evidence that leukemia spreads as clonal contagious cancer cells through M. arenaria populations raised the need of revisiting cockle DN etiology. Thus, cockles were collected from a DN-affected bed in Ría de Arousa (Galicia) DN and diagnosed for DN by examining stained hemolymph "cytospins" with light microscopy. Samples of hemolymph (as source of neoplastic cells) and mantle pieces (as source of normal cells) were taken for molecular analysis from 6 healthy cockles, 3 cockles with light DN, 3 cockles with moderate DN and 3 cockles with heavy DN. Additionally, standard histological sections from each cockle were analyzed. Molecular analysis of hemolymph and mantle cells from each cockle involved searching for Steamer-like sequences by PCR with degenerate RT-IN primers, and genotyping of 9 nuclear microsatellite loci and of SNPs of the mitochondrial cytochrome oxidase I gene. Results showed the presence of a Steamer-like retrotransposon not highly amplified in neoplastic cells. Analysis of microsatellite and SNP markers showed that genotypes of hemolymph and mantle cells were matched in healthy cockles as well as in cockles with light DN, whereas genotypes of hemolymph cells were different from those of mantle cells in cockles with moderate and heavy DN. As in M. arenaria, the fact that the neoplastic cells have genotype different from that of normal cells in the same individual indicates that cockles become diseased by acquisition of foreign neoplastic cells. A neighbor joining tree based on the 9 microsatellite loci showed that the haemolymph cell genotypes of cockles with moderate and heavy DN were distributed in two independent clusters, which suggests two independent neoplastic cell lineages, L1 and L2. Consistently, histological examination allowed distinguishing two morphological types of disseminated neoplasia: cockles showing tighter neoplastic cell arrangement, with rounded nucleus, corresponded to lineage L1 and cockles showing looser neoplastic cells, with pleomorphic nucleus (rounded, elongated, kidney shaped) with patent nucleolus, corresponded to lineage L2. In contrast to the case of M. arenaria, in which leukemia from East Canada to New York has likely a single clonal origin, our results indicate that DN is transmitted in cockle populations by at least two contagious cancer cell lineages.