Body polarity and mineral selectivity in the demosponge Chondrosia reniformis.
Chondrosia reniformis is a cushion-shaped, Atlanto-Mediterranean demosponge that usually lives on shallow rocky bottoms. A section through the sponge reveals two distinct regions: a cortical zone called ectosome, and an internal zone, the choanosome, which contains the choanocyte chambers. The ectosome is composed of a layer of flattened cells, exopinacocytes, that surround dense interwoven bundles of fibrils of collagen. In many circumstances the pinacocyte layer is loose, and the collagen fibrils can be in direct contact with water (Garrone et al., 1975).
C. reniformis lacks the opaline spicules that are the main constituents of the skeleton of other demosponges; rather, the collagenous ectosome is strengthened by sand grains and exogenous spicules, which are actively incorporated by the sponge (Bavestrello et al., 1995). Studies of foreign matter incorporation have been carried out on the sponge Dysidea etheria. In this species, the particles are incorporated by contraction of the dermal membrane, which probably separates or disrupts the thin exopinacocyte layer on the dermal membrane surface (Teragawa, 1986a, b). The ectosome of C. reniformis behaves similarly (Bavestrello et al, in press).
In C. reniformis, the upper ectosome is able to select the minerals that settle on the sponge: thus, siliceous material is engulfed while the calcareous fragments that are the main sediments available in the surrounding environment are eliminated (Bavestrello et al., 1996). In contrast, however, C. reniformis settles on calcareous rocks through the partial incorporation of outgrowths of this substratum by the lower ectosome. This process suggests a polarization in the sponge body, with a specificity for the incorporation of minerals that varies from the lower ectosome to the upper one.
Indeed, the polarization of the adult sponge body was already demonstrated 45 years ago with Sycon raphanus. When specimens of these sponges were bisected transversely, both halves could develop a complete new animal with the same polarity, from osculum to base, as the original (Tuzet and Paris, 1963).
The polarization of sponges relative to their position on the substratum probably arises in the larval stage. In the amphiblastula larvae of Calcarea, the flagellate cells of the anterior pole make the initial contact with the substratum (Bergquist and Green, 1977). In Demospongiae, several authors have suggested that the coeloblastula or parenchymella larvae express an existing polarity in their attachment to the substrate. But such views are highly speculative, because distinguishing between an anterior and a posterior hemisphere in these animals is difficult (see Simpson, 1984, for a review).
The aims of this work are to verify - through laboratory experiments and manning electron microscopy (SEM) - the capacity of both kinds of ectosome of Chondrosia reniformis to develop specificity towards siliceous and calcareous materials, and to demonstrate the cellular basis of that specificity.
Materials and Methods
Along the rocky cliff of the Portofino Promontory (Ligurian Sea, Italy) Chondrosia reniformis lives from the surface to the base of the cliff (about 50 m depth). The specimens used in this study were collected during January 1997, at 10 m depth, on calcareous substrates.
We performed our experiments with specimens having a surface area of 112-156 [cm.sup.2]. These specimens were reared at 15 [degrees] C in 200-1 aquaria containing natural seawater with a salinity of 37%. The medium was aerated by bubbling, and it was replaced twice a week. The collected sponges attached to the aquarium bottom in about 10 days.
The sandy materials used in testing sponge selectivity were white polycrystalline quartz with a particle size of 0.25-0.5 mm (BDH laboratory sand); red calcareous sand of the same particle size obtained from the organ-pipe coral Tubipora musica; and fragments of a coralline alga, Lithothamnium sp., 3-5 mm in size.
To test the differences in behavior between the upper and lower surfaces of the sponge ectosome, a thin layer of a mixture (1:1) of the BDH siliceous and Tubipora calcareous sands was laid down on the upper ectosome of five specimens that had attached to the bottom of an aquarium covered by the same mixture. Two experiments were carried out with these specimens. First, cores through the sponge, from upper to lower surface, were inverted and then transplanted, upside-down, inside the same specimen (five replicas). Second, 40 half-cores (20 mm in diameter) were taken from the upper and lower ectosome and reared for 3 months [ILLUSTRATION FOR FIGURE 1 OMITTED].
Observations made with SEM allowed us to distinguish differences in the organization of the two ectosomal surfaces. The samples used for these observations were collected by scuba divers and fixed in situ in 10% formalin.
When a mixture of siliceous and calcareous sand was allowed to settle on the upper ectosome of specimens that had attached to the bottom of the aquarium, only the quartz fragments (white) were incorporated. The calcareous grains (red) were never engulfed; rather they were quickly removed from the sponge surface [ILLUSTRATION FOR FIGURE 2a OMITTED]. In contrast, the lower ectosome showed a remarkable preference for incorporating the calcareous grains of the mixture lying on the bottom [ILLUSTRATION FOR FIGURE 2b OMITTED].
The upper ectosome of newly collected specimens that had not yet attached to the aquarium bottom actively incorporated both kinds of minerals [ILLUSTRATION FOR FIGURE 2c OMITTED]. Ten days later, after attachment, the upper surface of all the tested specimens began to select siliceous material exclusively.
To verify the difference in mineral selectivity between the two sides of the ectosome, cores (about 5-6 [cm.sup.2]) extending from the upper to the lower surface were cut from five large specimens and reinserted upside-down, to produce sponges with a portion of their lower surface having an upward orientation [ILLUSTRATION FOR FIGURE 1, 2e OMITTED]. A week later, when the explants were perfectly fused in their anomalous positions and the sponges were attached to the aquarium floor, 5 g of Lithothamnium calcareous sand was laid on their surface [ILLUSTRATION FOR FIGURE 2f OMITTED]. The subsequent behavior of the two types of surface was very different. The upper ectosome (brown) behaved normally, quickly removing calcareous particles [ILLUSTRATION FOR FIGURES 2g-h OMITTED]; but the inverted, originally lower surfaces (whitish) did not move the particles, and 3 months later, they were all incorporated.
The gross morphological differences between the upper [ILLUSTRATION FOR FIGURE 3a OMITTED] and lower surfaces of the sponge [ILLUSTRATION FOR FIGURE 3b OMITTED] were supported by SEM observations: the upper ectosome is entirely covered by polygonal flattened pinacocytes and perforated by the incurrent openings (ostia) [ILLUSTRATION FOR FIGURE 3c OMITTED], whereas the pinacoderm of the lower surface is covered by a collagenous sheet that lacks ostia [ILLUSTRATION FOR FIGURE 3d OMITTED].
To verify that the polarization between the upper and lower ectosome arises very early in development, experiments were performed with half-cores; these are fractions of sponge tissue that contain a portion of choanosome covered on one side by either upper or lower ectosome [ILLUSTRATION FOR FIGURE 1 OMITTED]. In the first 2 weeks of culture, the ectosome of both kinds of free half-cores actively proliferated, and the pieces assumed a spherical shape [ILLUSTRATION FOR FIGURES 3e-f OMITTED]. The half-cores covered with lower ectosome [ILLUSTRATION FOR FIGURE 3f OMITTED] settled on the smooth bottom of the aquarium in 3-4 weeks, whereas those with the upper ectosome [ILLUSTRATION FOR FIGURE 3e OMITTED] settled after about 10-12 weeks. Furthermore, 3 months after the beginning of the experiment, the half-cores with lower ectosome were covered by a collagenous, non-cellularized layer that remained white and formed no osculum; the half-cores with the upper ectosome produced a new oscule in 12-15 weeks and showed a normal pinacoderm periorated by pores. Sometimes, when half-cores deriving from the upper and lower ectosome came in contact, they fused together; but a normal sponge was never reconstituted, although the two kinds of ectosome remained distinct on the opposite sides of the sponge [ILLUSTRATION FOR FIGURE 3g OMITTED].
The structural and functional differences between the two sides of the ectosome of the sponge Chondrosia reniformis suggest a strong polarization - upper pole versus lower pole - along the axis of the sponge. The activity of the upper ectosome is likely due to the pinacocyte-mineral interaction. More problematic is the basis for the preferences of the lower ectosome for calcareous substrata. In other sponges, a nonspecific attachment to the substratum is probably due to the secretion by the basopinacocytes of a complex basal lamella (Pavans de Ceccatty, 1981) that anchors the sponge but prevents any contact between the cells and the substratum.
The ability of the upper ectosome to discriminate between silica and carbonates is present only in attached specimens and vanishes in free, nonattached ones, which incorporate both materials indiscriminately. This distinction is interpretable if we consider the asexual reproductive strategy of the species: Sponges living on overhanging ledges or on the vaults of submarine grottos give rise to long, thin pendant filaments [ILLUSTRATION FOR FIGURE 2d OMITTED]. Cell reorganization within the apical region of these filaments produces a new, functional, but suspended animal. When the filament breaks, the bud is separated from the maternal sponge (Gaino et al., 1995); it falls and must attach quickly irrespective of the side of the ectosome that comes in contact with the substratum. This behavior indicates, not only that mineral receptors are distributed evenly on the sponge surface, but also that these receptors may be activated or deactivated under particular conditions by an environmental switch. We suppose that the mineral discrimination of the upper ectosome is switched on by the adhesion of the sponge to the bottom.
Our studies indicate two modalities of mineral incorporation that are associated with the two sides of the ectosome. The upper side collects quartz and silicates, which strengthens the collagenous structure; this is a dynamic process comprising incorporation of the particles and resizing of the quartz grains, with their elimination via the aquiferous system (Bavestrello et al., 1995). The lower side specifically engulfs the calcareous substrata, thus fixing the sponge to the bottom.
There is a rich literature about the very wide potential for cytodifferentiation in sponges (see Simpson, 1984). Our data indicate that, at least in C. reniformis, the morphological differences between the upper and lower regions of the ectosome are sharp, probably arising in a very early stage of sponge ontogeny; that is, a functionally complete specimen cannot be reconstituted from a portion of only upper or lower ectosome with its adjacent choanosome. Connes (1966, 1968) has demonstrated that the ectosome and choanosome of Tethya aurantium have different potentials for reconstructing an entire sponge, but our data provide the first indication of differences in this process associated with distinct zones of the same ectosome.
In higher metazoans, the spatiotemporal development of morphological structures is regulated by homeobox genes (Lawrence, 1992). These genes have also been observed in lower metazoans such as sponges (Kruse et al., 1994; Coutinho et al., 1994; Degnan et al., 1995; Seimiya et al., 1997), but their meaning has been obscure until now. We hypothesize that the acquisition of an axial polarity in the sponge may be controlled by these genetic structures.
This research was supported by Italian MURST funds.
Bavestrello, G., A. Arillo, U. Benatti, C. Cerrano, R. Cattaneo-Vietti, L. Cortesogno, A. Gaggero, M. Giovine, M. Tonetti, and M. Sara. 1995. Quartz dissolution by the sponge Chondrosia reniformis (Porifera, Demospongiae). Nature 378: 374-376.
Bavestrello, G., C. Cerrano, R. Cattaneo-Vletti, M. Sara, F. Calabria, and L. Cortesogno. 1996. Selective incorporation of foreign material in Chondrosia reniformis (Porifera, Demospongiae). Ital. J. Zool. 63: 215-220.
Bavestrello, G., A. Arillo, B. Calcinal, C. Cerrano, R. Cattaneo-Vietti, S. Lanza, and M. Sara. Interaction between different kinds of silica and the exo-pinacocytes of the demosponge Chondrosia reniformis. Ital. J. Zool. (in press).
Bergquist, P. R., and C. R. Green. 1977. An ultrastructural study of settlement and metamorphosis in sponge larvae. Cah. Biol. Mar. 18: 289-302.
Connes, R. 1966. Aspects morphologiques de la regeneration de Tethya lyncurium Lamarck. Bull. Soc. Zool. Fr. 91: 43-53.
Connes, R. 1968. Etude histologique, cytologique et experimentale de la regeneration et de la reproduction asexuee chez Tethya lyncurim Lamarck (= Tethya aurantium Pallas) (Demosponges). Thesis, Univ. Montpellier, France. 193 pp.
Coutinho, C., S. Vissers, and G. Van de Vyver. 1994. Evidence of homeobox genes in the freshwater sponge Ephydatia fluviatilis. Pp. 385-388 in Sponges in Time and Space, R. W. M. van Soest, Th. M. G. van Kempen, and J. C. Braekman, eds. Balkema, Rotterdam.
Degnan, B. M., S. M. Degnan, A. Giusti, and D. E. Morse. 1995. A hox/hom homeobox gene in sponges. Gene 155: 175-177.
Gaino, E., R. Manconi, and R. Pronzato. 1995. Organizational plasticity as a successful conservative tactics in sponges. Anim. Biol. 4: 31-43.
Garrone, R., A. Huc, and S. Junqua. 1975. Fine structure and physicochemical studies on the collagen of the marine sponge Chondrosia reniformis Nardo. J. Ultrastruct. Res. 52: 261-275.
Lawrence, P. A. 1992. The Making of a Fly: The Genetics of Animal Design. Blackwell Scientific, Oxford.
Kruse, M., A. Mikoc:, H. Cetkovic, V. Gamulin, B. Rinkerich, I. M. Muller, and W. E. Muller. 1994. Molecular evidence for the presence of a developmental gene in the lowest animals: identification of a homeobox-like gene in the marine sponge Geodia cydonium. Mech. Ageing Dev. 77: 43-54.
Pavans de Ceccatty, M. 1981. Demonstration of actine filaments in sponge cells. Cell Biol. Int. Rep. 5: 945-952.
Seimiya, M., H. Ishiguro, K. Miura, Y. Watanabe, and Y. Kurosawa. 1997. Homeobox-containing genes in the most primitive metazoa, the sponges. Eur. J. Biochem. 221: 219-225.
Simpson, T. L. 1984. Cell Biology of Sponges. Springer, New York.
Teragawa, C.K. 1986a. Particle transport and incorporation during skeleton formation in a keratose sponge: Dysidea etheria. Biol. Bull. 170: 321-334.
Teragawa, C. K. 1986b. Sponge dermal membrane morphology: histology of cell-mediated particle transport during skeletal growth. J. Morphol. 190: 335-347.
Tuzet, O., and J. Paris. 1963. Recherches sur la regeneration de Sycon raphanus. Vie Milieu 14: 285-298.