Abstract
Intraorganism genetic stability is assumed in most organisms. However, here we show for the first time intraorganism genetic heterogeneity in natural populations of marine sponges. A total of 36 different multilocus genotypes (MLGs) were detected in 13 individuals of Scopalina lophyropoda sampled at 4 distant points within each sponge. All genotypes (showing a mosaic distribution), were transmitted to the progeny, thus contributing to the high genetic diversity and low clonality reported for this species, although its populations are small and structured and show high fission rates. There did not seem to be intraindividual genotype conflicts; on the contrary, chimeric individuals are expected to show low mortality thanks to the differential mortality of their different MLGs. This novel mechanism may also counterbalance genetic constraints in other benthic invertebrate species. The presence of sponge chimerism also suggests that results from previous population genetics studies could have been misinterpreted.
Many marine benthic invertebrates with poor dispersal and small populations live in fragmented habitats (Duran et al. 2004; Zilberberg et al. 2006; Calderón et al. 2007), which could substantially increase their extinction risk by decreasing their ability to adapt to changing conditions (Addison and Hart 2005). However, some of these small populations did not show inbreeding and low genetic diversity as was expected (McFadden 1997; Blanquer et al. 2009; Blanquer and Uriz 2010). Previous studies on the population genetics of the Mediterranean sponge Scopalina lophyropoda have reported relatively high levels of genetic diversity, outcrossing, and heterozygote excess, despite the species having philopatric larvae (Uriz et al. 1998) and small patchy populations (Blanquer et al. 2009; Blanquer and Uriz 2010). Moreover, clonality has been found to be almost nonexistent in the populations, although sponge fissions were often observed (Blanquer et al. 2008).
We studied the mating system of this species by genotyping seven microsatellite loci in a larval-free piece of three large ripe individuals of S. lophyropoda and in ten randomly collected larvae released from each ripe individual. A total of 7of the 30 larvae genotyped showed at least one locus with both alleles different to those of the parental sponges (fig. 1, supplementary table S1, Supplementary Material online).
Percentage of Scopalina lophyrpopoda larvae showing: for some loci, both alleles different to those of their parental sponge; in all loci, both alleles as their parental sponge (i.e., clones); and, in all loci, at least one allele as in their parental sponge. Note that parental sponges were only sampled at one point.
Microsatellites are the most used markers for studies on clonality and genotypic variation among individuals and populations (Duran et al. 2004; Calderón et al. 2007; Blanquer and Uriz 2010) due to their high polymorphism and presumed intraindividual stability (Estoup and Cornuet 1999). Therefore, in an attempt to understand the origin of these “nonmaternal” genotypes in sexually produced larvae (Uriz et al. 1998), we put forward the possibility of intraorganism genetic heterogeneity (IGH). We genotyped four pieces at four distant points of 12 S. lophyropoda individuals (fig. 2) before gamete production (i.e., in autumn). A total of 36 different multilocus genotypes (MLGs) were detected in the 13 individuals genotyped (supplementary table S2A, Supplementary Material online). The individuals contained from 1 to 4 different MLGs each (fig. 3). The presence of more than four genotypes in a single individual cannot be discarded because only four samples per individual were taken. Two alleles were consistently found in each sample like in the samples of approximately 200 individuals from an earlier study (Blanquer and Uriz 2010). This could be due to the low chance of sampling the intersection between two MLGs because the fragments we used for the DNA extraction were very small (i.e., 1–2 mm2 in area) and samples were systematically taken from the sponge border. We also genotyped two individuals that were in contact but clearly separated by a “nonconfluence front.” The genotypes of these two individuals in the contact zone differed in two microsatellite loci (supplementary table S2B, Supplementary Material online).
Example of location of the four sampling points (one to four) for each Scopalina lophyropoda individual.
Percentage of sampled individuals of Scopalina lophyropoda with 1,2,3 or 4 multilocus genotypes (MLGs).
This is the first time that chimeras (i.e., individuals that have two or more different populations of genetically distinct cells) are reported in sponge populations. Previous authors have stated that multigenotype individuals would be highly improbable in sponges in natural conditions because their experiments resulted in rejection or death of fused settlers in 2 months or less (Maldonado 1998; Gauthier and Degnan 2008). However, no genetic analyses were performed in those studies, and the genetic divergence of the larvae they placed in contact was not established (Gauthier and Degnan 2008).
The several distinct genotypes in natural populations of S. lophyropoda were confined to different parts of the sponge. The analyzed samples contained both somatic cells (i.e., pinacocytes, sclerocytes, spherulous cells, and collecytes) and precursors of the germ line (i.e., archeocytes and choanocytes) close together. Moreover, the different intraorganism MLGs were able to reproduce (some released larvae did not share any allele with the piece of the parent sponge genotyped for some loci), but we never found more than two alleles for each microsatellite locus. Consequently, the examined sponges were genetic mosaics, in which each genotype appeared to function independently in terms of reproduction; however, the different intraorganism genotypes behaved as a single large individual for decisive ecological issues, such as size-specific attributes that enhance fitness (Buss 1982; Foster et al. 2002).
Earlier authors (Buss 1982; Grosberg et al. 1985; Rinkevich and Weissman 1987; Solé-Cava and Thorpe 1994; Stoner and Weissman 1996) predicted constraints for the stability of IGH in marine invertebrates. In sponges, it has been hypothesized that if chimerism occurred spontaneously, it may generate intraorganism genetic conflict that would decrease individual fitness (Solé-Cava and Thorpe 1994; Stoner and Weissman 1996; Gauthier and Degnan 2008). Colonial ascidian chimerism has been studied in depth and it has been found that both death and/or resorption of one of the chimera partners and survival of the other (Stoner and Weissman 1996), and stable chimerism with reproduction of both partners can occur (Stoner et al 1999; Rinkevich 2005). Where intraindividual genetic conflict does not seem to occur, the long-term ecological and evolutionary benefits of IGH should be considered (Pineda-Krch and Lehtila 2004). IGH can be particularly beneficial for organisms that show partial mortality, such as encrusting sponges because the different parts (genotypes) of a chimera (mosaic) may exhibit different fitness levels and mortality risks, which lowers the possibilities of total individual death. In changing environments such as seasonal seas, some genotypes may grow faster and take over the space of other less adapted genotypes. This can change over the seasons and thus enhances individual survival over the year. It could explain the high but poorly understood dynamism of many encrusting Mediterranean sponges, which change their shape and size continuously and live for decades without showing any effective growth (Blanquer et al. 2008; Teixidó et al. 2009).
No external barriers (nonconfluence fronts) among genotypes were evident in our sponge mosaics (fig. 2). Yet, there would have to be a kin-recognition system (Aanen et al. 2008) in the species in order to keep the different genotypes confined to separate sponge regions. Furthermore, not all the S. lophyropoda individuals that come into contact in the field fuse, which proves that the kin-recognition mechanism is only triggered above a genetic differentiation threshold (Aanen et al. 2008). However, the two individuals separated by a nonconfluence barrier showed fewer genotypic differences than those found within MLGs of some individuals. The individuals recognized as kin must have minimum differences in the genes related to the immune-recognition system, which are not necessarily linked with microsatellite diversity.
The origin of the natural chimeras in S. lophyropoda remains unknown and merits further studies. Among the several possible mechanisms reported to originate IGH, fusion between larvae or settlers (Solé-Cava and Thorpe 1994; Sommerfeldt et al. 2003; Hughes et al. 2004) and fusion of genetically closely related adult individuals should be considered. Another possible scenario, which has not been considered before, is that nonreleased larvae develop within the maternal tissue. Individuals of S. lophyropoda were found to contain larvae that were still functional inside their tissue after the reproductive period (Maldonado and Uriz 1999). Although larval resorption is assumed, it has never been proved. However, if this occurs, all the MLGs in the same individual would share the maternal allele which is not the case in S. lophyropoda. Fusion among individuals and development of retained larvae could, however, combine and lead to a similar scenario as that found for S. lophyropoda. Finally, sponges, like plants, show indeterminate growth and asexual reproduction (Wulff 1986); thus, we cannot totally reject that the IGH found is the result of multiple somatic mutations, which often occur in plant mosaics (Santelices 2004). Moreover, these mutations have previously been proposed to explain phenotypic variation in sponges (Neigel and Schmahl 1984). In this case, a beneficial somatic mutation in any bud or part of a sponge could spread within the population by both sexual and/or asexual reproduction. Conversely, when the mutation is deleterious, the death of a part of the organism would have little effect on the survival of the whole sponge.
The mosaic chimerism found in S. lophyropoda may also occur in other sponge species. Highly dynamic encrusting species with successive growth and shrinkage and high fusion and fission rates, like S. lophyropoda (Blanquer et al. 2008), are good candidates for fixing mosaic chimerism as a heritable trait to increase ecological success. IGH would enhance the survival of many colonial sessile invertebrates with poor dispersal abilities and small populations. However, if chimeric individuals are as widespread in marine invertebrates as increasing evidence seems to indicate (Pineda-Krch and Lehtila 2004; Rinkevich 2005), the results from previous studies on the population genetics of marine invertebrates should be considered with caution because the genetic diversity, effective size, and mating systems may have been incorrectly estimated. When IGH is suspected, sampling designs should be modified accordingly to determine the real genetic heterogeneity of the populations.
Methods
Fragments from three ripe individuals of S. lophyropoda were collected by SCUBA diving at the Blanes litoral—NE Spain (lat 41°40.12′N, long 2°47.10′E) in late July, transported individually to the laboratory, where they released larvae spontaneously. Ten larvae and a larval-free fragment of the parental tissue were taken. In autumn, before gamete production, we collected four fragments (placed at four distant directions) from 13 large individuals (700–1000 cm2 of area) from the same population.
Both, adult and larval DNA was extracted with DNeasy (QIAGEN). For individual larval DNA, the protocol was modified by reducing volumes of all reagents. Eight microsatellites, seven from Blanquer et al. (2005) and a new one, Scol_f (EMBL acc num FR773981; 146 bp, amplified at 45 °C), were genotyped.
Authors acknowledge Drs T. Villanueva and M. Swan and two anonymous reviewers for valuable comments on the manuscript. This work was supported by the Spanish Government (CICYT) (BENTHOMICS CTM2010-22218) to M.J.U.
References
Author notes
Associate editor: Billie Swalla
