Siliceous Spicules and Skeleton Frameworks in Sponges: Origin, Diversity, Ultrastructural Patterns, and Biological Functions MARI ´ A-J. URIZ, 1 * XAVIER TURON, 2 MIKEL A. BECERRO, 1 AND GEMMA AGELL 1 1 Center for Advanced Studies (CSIC), Girona, Spain 2 Department of Animal Biology (Invertebrates), University of Barcelona, Barcelona, Spain KEY WORDS silification; spicules; ultrastructure; sponges ABSTRACT Silica deposition is a fundamental process in sponges. Most sponges in the Classes Demospongiae and Hexactinellida secrete siliceous elements, which can subsequently fuse, inter- lock with each other, or form three-dimensional structures connected by spongin. The resulting skeletal frameworks allow sponges to grow upwards and facilitate water exchange with minimal metabolic cost. Several studies on sponge skeletogenesis have been published. We are beginning to understand the mechanisms of spicule secretion and the role of spicules and skeletal frameworks in the biology, ecology, and evolution of sponges. Molecular techniques and ecological experiments have demonstrated the genetic control of the process and the contribution of environmental factors to the expression of a sponge spicule, respectively. However, other classic topics such as the role of membranes in silicon transport or whether spicules are formed in situ or secreted anywhere in the sponge mesohyl and then transported to the skeletal framework require further investigation. We review the process of silica deposition in sponges at the molecular and cellular levels, as well as the biological and ecological functions of spicules and skeletons. The genetic control of spicule shapes makes them useful in the reconstruction of sponge phylogeny, although recent experiments have demonstrated the influence of environmental factors in modulating spicule size, shape, and the pres- ence or absence of one or more spicule types. The implications of such variations in sponge taxonomy may be important. Besides supporting sponge cells, spicules can help larvae stay buoyant while in the plankton or reach the bottom at settlement, enhance reproduction success, or catch prey. Conversely, the role of spicules and skeletons in deterring predation has not been demonstrated. Knowledge of several aspects is still based on a single or a few species and extrapolations should be made only with caution. With the advent of new molecular techniques, new lines of research are presently open and active in this field. Microsc. Res. Tech. 62:279 –299, 2003. © 2003 Wiley-Liss, Inc. INTRODUCTION Sponges, regardless of the Class to which they belong (i.e., Calcarea Bowerbank, Demospongiae Sollas, or Hexactinellida Schmidt), secrete mineral or protein- aceous structures that give them a variety of three- dimensional shapes, which minimizes the metabolic cost of water exchange (Vogel, 1974; Larsen and Riis- gard, 1994; Riisgard and Larsen, 1995). Most Demospongiae and Hexactinellida produce sil- ica-made skeletons consisting of individualized ele- ments (spicules) of lengths ranging from micrometers to centimeters, which can subsequently fuse or inter- lock with each other. The two classes differ from a skeletal point of view in the number of symmetry axes of their megascleres, which are monaxons and tetrax- ons in demosponges and monaxons and triaxons in hexactinellids (Fig. 1). The high diversity of spicule shapes and sizes (Fig. 2) in both fossil and living sponges has been repeatedly reported (e.g., Hinde, 1887–1893; Hartman, 1981; Simpson, 1984) and has received particular attention in taxonomic and cladistic studies (e.g., Chombard et al., 1998; Hooper, 1990; Rosell and Uriz, 1997; Uriz and Carballo, 2001). However, the mechanisms that deter- mined such diversity remained elusive until recently. Harrison and Simpson (1976) and later authors (e.g., Garrone et al., 1981) attributed to both the protein- aceous spicule core (the axial filament) and the sur- rounding membrane (silicalemma) a role in shaping the spicules. Most of those early contributions were reviewed 20 years ago (e.g., Volcani, 1981; Simpson, 1984, 1989), and here we focus on progress since then. Recent molecular studies (Shimizu et al., 1998; Cha et al., 1999, 2000; Krasko et al., 2000) cast light on the genetic control of spicule deposition. In contrast, the role of membranes in modulating spicule ornamenta- tion (spines and swellings) or the terminal formations Contract grant sponsor: CICYT; Contract grant number: MAR98-1004-C02; Contract grant sponsor: Generalitat of Catalonia; Contract grant numbers: 1999SGR00184 and REN2001-2312-CO3/MAR and INTERREG-III-2002 (to MJU and XT). *Correspondence to: Marı ´a-J. Uriz, Center for Advanced Studies (CSIC), Acce ´s a la Cala St. Francesc, 14 17300 Blanes, Girona, Spain. E-mail: Iosune@ceab.csic.es Received 7 August 2002; accepted in revised form 20 April 2003 DOI 10.1002/jemt.10395 Published online in Wiley InterScience (www.interscience.wiley.com). MICROSCOPY RESEARCH AND TECHNIQUE 62:279 –299 (2003) © 2003 WILEY-LISS, INC.