A novel highly acidic polysaccharide with inhibitory activity on calcification from the calcified scale “coccolith” of a coccolithophorid alga, Pleurochrysis haptonemofera

A novel highly acidic polysaccharide with inhibitory activity on calcification from the calcified scale “coccolith” of a coccolithophorid alga, Pleurochrysis haptonemofera

Biochemical and Biophysical Research Communications 357 (2007) 1172–1176 www.elsevier.com/locate/ybbrc A novel highly acidic polysaccharide with inhi...

367KB Sizes 0 Downloads 4 Views

Biochemical and Biophysical Research Communications 357 (2007) 1172–1176 www.elsevier.com/locate/ybbrc

A novel highly acidic polysaccharide with inhibitory activity on calcification from the calcified scale ‘‘coccolith’’ of a coccolithophorid alga, Pleurochrysis haptonemofera Noriaki Ozaki

a,b

, Shohei Sakuda b, Hiromichi Nagasawa

b,*

a b

Department of Functional Bioscience, Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Bunkyo, Tokyo 113-8657, Japan Received 10 April 2007 Available online 19 April 2007

Abstract Coccolith, a calcified scale with species-specific fine structure produced by marine unicellular coccolithophorid algae, consists of calcium carbonate (CaCO3) crystals and a small amount of organic matrices. A novel polysaccharide named coccolith matrix acidic polysaccharide (CMAP) was isolated from the coccolith of a coccolithophorid alga, Pleurochrysis haptonemofera. The structure of CMAP was determined by chemical analysis and NMR spectroscopy including COSY, TOCSY, HMQC, and HMBC to be a polysaccharide composed of the following unit: fi4) L-iduronic acid (a1fi2) meso-tartaric acid (3fi1) glyoxylic acid (1fi. It has four carboxyl groups per a disaccharide unit as observed in another polysaccharide PS-2 characterized previously in Pleurochrysis carterae. CMAP showed a strong inhibitory activity on CaCO3 precipitation. These results suggest that CMAP serves as a regulator in the calcification of the coccolith.  2007 Elsevier Inc. All rights reserved. Keywords: Acidic polysaccharide; Biomineralization; Calcification; Coccolith; Coccolithophorid alga; Pleurochrysis haptonemofera

Biomineralization is a highly complex mineralizing process well controlled by organisms. The mineral thus formed, termed biomineral, which are observed in many organisms, help maintain body structure, perceive gravity and balance, defend against enemies, and store inorganic minerals [1]. They generally contain a small amount of organic matrices, which are thought to provide a polymeric framework and to play important roles in crystal nucleation and regulation of its growth [2,3]. The resulting crystals have a specific morphology and are oriented in specific Abbreviations: CMAP, coccolith matrix acidic polysaccharide; COSY, correlation spectroscopy; HMBC, heteronuclear multiple bond correlation; HMQC, heteronuclear multiple quantum coherence; HPLC, high performance liquid chromatography; NMR, nuclear magnetic resonance; PAGE, polyacrylamide gel electrophoresis; TOCSY, total correlated spectroscopy. * Corresponding author. Fax: +81 3 5841 8022. E-mail address: [email protected] (H. Nagasawa). 0006-291X/$ - see front matter  2007 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2007.04.078

direction, which are quite different from those formed without participation of living organisms, and therefore organic matrices appear to have essential roles in the regulation of morphology and direction of crystal growth of biominerals [4]. The marine unicellular coccolithophorid algae produce elaborate calcified scales called coccolith, which consist of fine pieces of CaCO3 (calcite) crystals and are known as one of the representative biominerals. The characteristic morphologies of coccolith are genus-specific and therefore have been used as a good tool to identify the species. Coccolith is formed intracellularly in a coccolith forming vacuole derived from the Golgi apparatus under a strict biological control, secreted from the cell by exocytosis and finally arranged on the cell surface [5,6]. As in the case of other biominerals, organic matrices in coccoliths are thought to be associated with coccolith biomineralization. To date, many investigators found acidic polysaccharides from coccoliths of some species of

N. Ozaki et al. / Biochemical and Biophysical Research Communications 357 (2007) 1172–1176

coccolithophorid algae [7–10]. Water-soluble acidic polysaccharides are considered to play important roles in crystal nucleation and regulation of crystal growth in coccolith formation, since they have an inhibitory activity on CaCO3 formation and an ability to bind calcium ions. Marsh et al. isolated three acidic polysaccharides, PS-1, -2, and -3, from the coccolith of Pleurochrysis carterae and determined the structure of the most abundant polysaccharide designated PS-2 [8]. PS-2 has a unique structure with a disaccharide unit, consisting of D-glucuronic acid and its oxidative degradation product at the C2–C3 bond (meso-tartaric acid and glyoxylic acid). Immunohistochemical study suggested that PS-2 was localized in the Golgi-derived particles with 20 nm in diameter and in the crystal coat of mature coccolith [11]. However, it is still unclear how polysaccharides are used to control crystal growth of coccolith. Recently, a novel acidic polysaccharide designated CMAP (coccolith matrix acidic polysaccharide) was found in P. carterae LU strain reported preliminarily [10]. CMAP displayed inhibitory activity on CaCO3 precipitation like many other organic matrices isolated from biominerals, but PS-2 did not. Structural determination of CMAP has not yet been accomplished, because the final preparation was contaminated with a small amount of PS-2 [10]. So, we searched for strains producing CMAP but not PS-2 among Pleurochrysis genus. Finally, we found that Pleurochrysis haptonemofera contains CMAP only [12]. In this study, we determined the chemical structure of CMAP and assessed its inhibitory activity on CaCO3 crystal growth. Materials and methods Reagents. D-Glucuronic acid, and D-, L- and meso-tartaric acid were purchased from Sigma–Aldrich (Mo, USA). L-Iduronic acid was prepared by acid hydrolysis of chondroitin sulfate B purchased from Seikagaku Kogyo (Tokyo, Japan). Cultivation of coccolithophorid alga. Pleurochrysis haptonemofera was kindly provided by Dr. Fujiwara. Cells were grown in transparent, 5-l flasks each containing 4-l of a seawater-based medium described previously [13] at 19 C with continuous aeration under an 18 h light–6 h dark cycle for 8 days. The ambient intensity of white light was about 100 lE/ m2 s. Extraction and purification of CMAP from coccoliths. Coccoliths were isolated from cells of P. haptonemofera from 12-l culture essentially according to the method reported previously [8]. The coccoliths were decalcified with 0.5 M EDTA solution (pH 8.0), and the resulting EDTAsoluble fraction was dialyzed against distilled water with molecular-porous membrane tubing (MWCO 12–14000, Spectrum, CA, USA) at 4 C for overnight. The dialysate was concentrated with an Amicon YM-10 membrane and the resulting solution was applied to a TSK-gel DEAE5PW HPLC column (7.5 · 75 mm, Tosoh, Tokyo, Japan) equilibrated with 50 mM Tris–HCl (pH 8.0). CMAP was eluted with a gradient of 0– 0.5 M NaCl in the same buffer in 30 min at a flow rate of 1 ml/min. The elution was monitored by measuring the absorbance at 230 nm. Acid hydrolysis of CMAP and purification of the hydrolysate. Hydrolysis of the purified CMAP was performed with 2 M trifluoroacetic acid at 121 C for 1 h. The hydrolysate was separated by anion-exchange HPLC as described above. Thin layer chromatography. Thin layer chromatography on a silica gel plate (Wako, Tokyo, Japan) was performed with solvent systems of 6:1:3

1173

and 2:1:2 acetonitrile–water–acetic acid for hexuronic acid and tartaric acid, respectively. Measurement of NMR spectra. Purified CMAP was deuteriumexchanged by freeze-drying three times and then dissolved in D2O to a final concentration of 10 mg/ml. 1H and 13C NMR spectra of CMAP were measured at 500 and 125 MHz, respectively, in D2O on a JEOL a-500 spectrometer at 27 C. Signals at dH 2.225 and dC 31.07 for acetone were used as external standards. The 1H–1H and 1H–13C connectivities were established by two-dimensional NMR (COSY, HMQC, HMBC, and TOCSY). Assay for inhibitory activity on CaCO3 precipitation. Inhibitory activity of CMAP on CaCO3 precipitation was assessed as described previously [10] with slight modification for small-scale operation. In brief, a solution consisting of 1 ml of 20 mM NaHCO3 (pH 8.7), and 0.1 ml of H2O (control) or a test sample solution (CMAP; at final concentrations of 0.2, 0.5, and 1.0 lg/ml) was prepared. To this solution, 1 ml of 20 mM CaCl2 was added and the formation of CaCO3 precipitate was followed by the decrease of pH value. Changes in the pH value of the solution were measured every 1 min for 20 min.

Results and discussion Although acidic polysaccharides are considered to play important roles in coccolith formation, the precise role is still unclear. To clarify the relationship between chemical structure and function in crystal growth of CaCO3, an acidic polysaccharide named CMAP was purified from the coccoliths of P. carterae LU strain by assaying inhibitory activity on calcification [10]. However, structural determination of CMAP from the P. carterae LU strain was not successful because of the contamination of another polysaccharide PS-2. It was difficult to separate CMAP from PS-2 completely, due to their similarities of elution pattern on anion-exchange HPLC and in mobility on SDS–PAGE. So, we searched for a strain which produces only CMAP but not PS-2. Eventually, we found that P. haptonemofera produces CMAP only [12]. More recently, Hirokawa et al. reported that P. haptonemofera contains three acidic polysaccharides [14]. The most abundant polysaccharide among them, named Ph-PS-2, might correspond to CMAP. However, chemical structure of this compound has not yet been determined. Therefore, we tried to determine the chemical structure of CMAP and to assess its inhibitory activity on CaCO3 crystal growth. Although CMAP is a high molecular weight compound with molecular weights of 4–100 kDa as estimated by gelfiltration, only eight proton signals were observed in the region from 3.5 to 5.0 ppm in the 1H nuclear magnetic resonance (NMR) spectrum of CMAP (data not shown, but the signals in this region shown in Fig. 1), suggesting that CMAP is a polymer consisting of carbohydrate only. Comparison of the 1H NMR spectrum of CMAP with that of PS-2, most of the signals derived from the uronic acid moiety in CMAP were different from that (glucuronic acid) of PS-2 [8]. The chemical shifts of the signals suggested that the uronic acid in CMAP was iduronic acid. In order to confirm this, CMAP and chondroitin sulfate B, which contains iduronic acid were separately hydrolyzed and the hydrolysates were subjected to silica gel thin layer chromatography (TLC). The result indicated that CMAP has a

1174

N. Ozaki et al. / Biochemical and Biophysical Research Communications 357 (2007) 1172–1176

Fig. 1. Two-dimensional 1H–1H COSY spectrum of CMAP. The cross peaks between HU1 and HU2, HU2 and HU3, HU3 and HU4, HU4 and HU5 and HT2 and HT3 are observed.

compound with the same Rf value (0.40) as that of iduronic acid derived from chondroitin sulfate B. In addition, 1H NMR spectrum of the uronic acid fraction from CMAP was almost identical with that of a mixture of authentic L-iduronic acid and L-idurono-3,6-lactone from the acid hydrolysate of chondroitin sulfate B (data not shown). In the acid hydrolysate of CMAP, meso-tartaric acid was also identified by TLC. The 13C NMR spectrum of CMAP showed twelve major carbon signals (data not shown), of which two signals were observed in the anomeric region (around 100 ppm) (Fig. 2 and Fig. 3A), suggesting that CMAP has a repeated sequence of a disaccharide unit. In the region between 175 and 178 ppm (Fig. 3B), four signals were observed, indicating that CMAP might be a polymer of di-hexuronic acid in which one hexuronic acid residue was oxidatively

Fig. 2. HMQC spectrum of CMAP.

Fig. 3. HMBC spectra of CMAP (A,B). (C) Probable repeating structure of CMAP: fi4) L-iduronic acid (a1fi2) meso-tartaric acid (3fi1) glyoxylic acid (1fi.

cleaved at C2–C3 as observed in PS-2. In PS-2 obtained from P. carterae CCMP 645, one residue of a unit was cleaved to yield a dicarboxylic acid (meso-tataric acid) and a monocarboxylic acid (glyoxylic acid) [8]. Thus, the four carboxyl groups may be attributed to iduronic acid, tartaric acid and glyoxylic acid. The connectivity among the components of CMAP was established by the analysis of 1H–1H COSY spectrum (Fig. 1). The anomeric proton signal at 4.95 ppm, for iduronic acid H-1 (HU1), gave a cross-peak to H-2 of iduronic acid (HU2) at 3.63 ppm. Cross-peaks linkage beginning with the anomeric signal identified all the proton signals of iduronic acid (from HU1 to HU5). The TOCSY spectrum supported the same connectivity of the protons in iduronic acid (data not shown). The remaining cross-peaks at 4.22

N. Ozaki et al. / Biochemical and Biophysical Research Communications 357 (2007) 1172–1176

and 4.34 ppm were assigned to the protons, HT2 and HT3, attached to CT2 and CT3 carbons of tartaric acid, respectively. Since the signal at 4.77 ppm (HG1) is the only isolated proton signal, HG1 is thought to be assigned to the acetal proton of glyoxylic acid as shown in PS-2 [8]. HMQC was used to assign the 13C signals. The HMQC spectrum of CMAP (Fig. 2) showed well-resolved crosspeaks for all the eight 13C–1H one-bond couplings, while the HMBC spectrum shows two or three-bond couplings, including signals of carboxyl groups (Fig. 3A and B). The intra- and inter-unit linkages were determined by the glycosyl-linkage of the disaccharide unit in CMAP. The proton (HT2) attached to C2 carbon of tartaric acid has a long-range coupling to C1 of iduronic acid (CU1) (Fig. 3A). These results indicating that CMAP has a linkage of iduronic acid (1fi2) meso-tartaric acid. The acetal carbon of glyoxylic acid (CG1) has a long-range coupling to both HT3 of tartaric acid and HU4 of iduronic acid, suggesting a linkage of meso-tartaric acid (3fi1) glyoxylic acid (1fi4) iduronic acid. Based on these results, the linkage of CMAP was determined as fi4) iduronic acid (1fi2) mesotartaric acid (3fi1) glyoxylic acid (1fi. PS-2 comprises D-glucuronic acid and uronic acid-derivatives, meso-tartaric acid and glyoxylic acid [8]. Considering that several Pleurochrysis strains contain both CMAP and PS-2 [12], CMAP might be biosynthesized via PS-2. Glucuronyl C5-epimerase was found to catalyze the conversion of D-glucuronic acid to L-iduronic acid in heparan sulfate biosynthesis [15]. In case of heparin biosynthesis, the inversion at C-5 of D-glucuronic acid leading to L-iduronic acid occurred after the polymer is synthesized [16]. In addition, D-iduronic acid has never been found in nature. Therefore, iduronic acid in CMAP might have L configuration. One-bond 13C–1H spin-coupling constants (1JCH) have been applied to determine the configuration of anomeric carbon; 1JCH  170 Hz for a configuration and 1JCH  160 Hz for b configuration [17]. In this study, 1JCH coupling constant between H-1 and C-1 of iduronic acid was 173.9 Hz, suggesting that iduronic acid in CMAP has an a configuration. The 3JH–H values, three-bond 1H–1H spin-coupling constants, for natural and synthetic iduronic acid-containing oligosaccharide in D2O, were listed in the literature [18]. In this study, the small H1–H2 coupling constant (3JH1–H2 < 4 Hz) of iduronic acid indicate 1C4 conformations (low-energy conformers) contribute significantly to the equilibrium. The remaining vicinal coupling constants of iduronic acid in CMAP, 3JH2–H3 (<4 Hz), 3JH3–H4 (3.7 Hz), 3JH4–H5 (2.2 Hz), indicate intermediate between 1 C4 and 2S0 conformations. Since the contribution of the 2 S0 form (>50%) is reflected mainly by large 3JH2–H3 values (>6.4 Hz) in the literature [18], this value for the iduronic acid residue in CMAP suggests that the 1C4 conformation is superior to the 2S0 conformation, but the 1C4 chair form is slightly distorted as proposed previously [19]. In addition, the weak H2–H5 NOE connectivity in the NOESY spectrum of CMAP also supports the distortion of the

1175

six-membered ring (results not shown). All these data conclude that the chemical structure and possible conformation of CMAP are estimated as shown in Fig. 3C. To investigate the effect of CMAP on CaCO3 crystallization from its supersaturated solution, the pH drop assay was performed by the method of Wheeler et al. [20] with slight modification [10]. CMAP inhibited the CaCO3 precipitation in a dose-dependent manner and completely inhibited it at a concentration of 1 lg/ml (Fig. 4). Since the average molecular weight of CMAP was estimated to be about 4.8 · 104, maximum inhibition was attained at 2.1 · 10 8 M. This activity is close to, or a little stronger than GAMP (gastrolith matrix protein) in the gastrolith of crayfish [21] and C-type lectin in the barnacle [22]. CMAP is also about 10-fold more active than CAP-1 (calcification-associated peptide) isolated from the crayfish calcified exoskeleton [23] and a recombinant molecule of Nacrein (a matrix protein of pearl oyster shell) [24] with the same assay system. Surprisingly, PS-2 did not show inhibitory activity even at the concentration of 100 lg/ml [10], although PS-2 is structurally very similar to CMAP, suggesting that iduronic acid of CMAP and/or its conformational contribution to the whole molecule are important for displaying high inhibitory activity. Similar phenomenon was reported in heparin in which iduronic acid is a more potent cation chelator than glucuronic acid [25]. Furthermore, iduronic acid-rich dermatan sulfate and iduronic acid-rich heparan sulfate displayed stronger adsorption to hydroxyapatite crystals than glucuronic acid-rich chondroitin 4-sulphate [26]. These findings suggest that the strength of inhibitory activity is not simply related to the number of carboxyl groups in acidic polysaccharide, but to the conformation of the whole molecule containing iduronic acid. At present, the exact reason for the difference in inhibitory activity could not be explained clearly, but it is

Fig. 4. Inhibitory activity of CMAP on CaCO3 precipitation. Changes in the pH of the assayed solution in 20 min are shown. Solid triangles, 1 lg/ ml CMAP; solid circles, 0.5 lg/ml CMAP; solid squares, 0.2 lg/ml CMAP; open squares, control (distilled water). Results are expressed as means ± SEM (n = 3).

1176

N. Ozaki et al. / Biochemical and Biophysical Research Communications 357 (2007) 1172–1176

important to solve this problem for understanding the function of CMAP and PS-2. Although all calcified strains of coccolithophorid thus far examined had acidic polysaccharides [12], they were not found in non-calcified strain of P. haptonemofera and Emiliania huxleyi (data not shown). Marsh and Dickinson [27] reported that the PS-2 had a significant role in coccolith formation because mutants of P. carterae CCMP 645 which did not produce PS-2, were poorly mineralized, indicating that highly acidic polysaccharides such as CMAP and PS-2 are essential for coccolith formation. These highly acidic polysaccharides have never been found from other organisms. The precursor of both polymers might be a polyuronide built up of the 1fi4 linked D-glucuronic acid or the alternating residues of D-glucuronic acid and D-mannuronic acid. An enzymatic oxidative cleavage at the C2–C3 bond in every second uronic acid residue may probably result in producing CMAP and PS-2, having four carboxyl groups per a disaccharide unit. This oxidation enzyme might be a key for the coccolith formation in Pleurochrysis genus.

[9]

[10]

[11]

[12]

[13]

[14]

[15]

Acknowledgments We are grateful to Dr. Shoko Fujiwara, Tokyo University of Pharmacy and Life Science, and Dr. Kazuo Furihata, The University of Tokyo, for providing us with the strain of P. haptonemofera and valuable suggestions on NMR analyses, respectively. This work was supported by Grants-in-Aid for Scientific Research (Nos. 12NP0201 and 17GS0311) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan and the Japan Society for the Promotion of Science (JSPS), respectively. The first author (N.O.) was supported by a Research Fellowship of JSPS for Young Scientists. References [1] A.P. Wheeler, C.S. Sikes, Matrix-crystal interactions in CaCO3 biomineralization, in: S. Mann, J. Webb, R.J.P. Williams (Eds.), Biomineralization—Chemical and Biochemical Perspectives, VCH, Weinheim, 1989, pp. 95–131. [2] H.A. Lowenstam, S. Weiner, On Biomineralization, Oxford University Press, New York, 1989. [3] M.A. Borowitzka, Carbonate calcification in algae-initiation and control, in: S. Mann, J. Webb, R.J.P. Williams (Eds.), Biomineralization—Chemical and Biochemical Perspectives, VCH, Weinheim, 1989, pp. 63–94. [4] J. Aizenberg, M. Iran, S. Weiner, L. Addadi, Intracrystalline macromolecules are involved in the morphologenesis of calcitic sponge spicules, Connect. Tissue Res. 34 (1996) 255–261. [5] K.M. Wilbur, N. Watabe, Experimental studies on calcification in mollusks and the alga Coccolithus huxleyi, Ann. N. Y. Acad. Sci. 109 (1963) 82–112. [6] D.E. Outka, D.C. Williams, Sequential coccolith morphologies in Hymenomonas carterae, J. Protozool. 18 (1971) 285–297. [7] E.W. De Jong, L. Bosch, P. Westbroek, Isolation and characterization of Ca2+-binding polysaccharide associated with coccoliths of Emiliania huxleyi (Lohmann) Kampter, Eur. J. Biochem. 70 (1976) 611–621. [8] M.E. Marsh, D.K. Chang, G.C. King, Isolation and characterization of a novel polysaccharides containing tartrate and glyoxylate residues

[16]

[17] [18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

from the mineralized scales of a unicellular coccolithophorid alga Pleurochrysis carterae, J. Biol. Chem. 267 (1992) 20507–20512. M. Okazaki, T. Sato, N. Muto, N. Wada, T. Umegaki, Calcified scale (coccolith) of Pleurochrysis carterae (Haptophyta): structure, crystallography, and acid polysaccharides, J. Mar. Biotechnol. 6 (1998) 16–22. N. Ozaki, S. Sakuda, H. Nagasawa, Isolation and some characterization of an acidic polysaccharide with anti-calcification activity from coccoliths of a marine alga, Pleurochrysis carterae, Biosci. Biotechnol. Biochem. 65 (2001) 2330–2333. M.E. Marsh, Polyanion-mediated mineralization-assembly and reorganization of acidic polysaccharide in the Gogi system of a coccolithophorid alga during mineral deposition, Protoplasma 177 (1994) 108–122. N. Ozaki, M. Okazaki, T. Kogure, S. Sakuda, H. Nagasawa, Structural and functional diversity of acidic polysaccharides from various species of coccolithphorid algae, Thalassas 20 (2004) 59–68. R.W. Eppley, R.W. Holmes, J.D.H. Stickland, Sinking rate of marine phytoplankton measured with a fluorometer, J. Exp. Mar. Biol. Ecol. 1 (1967) 191–208. Y. Hirokawa, S. Fujiwara, M. Tsuzuki, Three types of acidic polysaccharides assocated coccolith of Pleurochrysis haptonemofera: comparison of the biochemical characterics with those of P. carterae and analysis using fluorescein-isothiocyante-labeled lectins, Mar. Biotechnol. 7 (2005) 636–644. A. Hagner-Mcwhirter, J.P. Li, S. Oscarson, U. Lindahl, Irreversible glucuronyl C5-epimerization in the biosynthesis of heparan sulfate, J. Biol. Chem. 279 (2004) 14631–14638. M. Hook, U. Lindahl, G. Backstrom, A. Malmstrom, L-A. Fransson, Biosynthesis of Heparin III-Formation of iduronic acid residues, J. Biol. Chem. 249 (1974) 3908–3915. K. Bock, C. Pederson, A study of 13CH coupling constants in hexopyranoses, J. Chem. Soc. Perkin Trans. II (1974) 293–297. D.R. Ferro, A. Provasoli, M. Ragazzi, B. Casu, G. Torri, V. Bossenec, B. Perly, P. Sinay, M. Petitou, J. Choay, Conformer populations of L-iduronic acid residues in glycosaminoglycan sequences, Carbohydr. Res. 195 (1990) 157–167. A.S. Perlin, B. Casu, J. Tse, J.R. Sanderson, Methyl a- and b-Didopyranosiduronic acids synthesis and conformational analysis, Carbohydr. Res. 21 (1972) 123–132. A.P. Wheeler, J.W. George, C.A. Evans, Control of calcium carbonate nucleation and crystal growth by soluble matrix of oyster shell, Science 212 (1981) 1397–1398. N. Tsutsui, K. Ishii, Y. Takagi, T. Watanabe, H. Nagasawa, Cloning and expression of a cDNA encoding an insoluble matrix protein in the gastroliths of crayfish, Procambarus clarkii, Zool. Sci.16 (1999)619–628. H. Kamiya, M. Jimbo, H. Yako, K. Muramoto, O. Nakamura, R. Kado, T. Watanabe, Participation of the C-type hemolymph lectin in mineralization of the acorn barnacle Megabalanus rosa, Mar. Biol. 140 (2002) 1235–1240. H. Inoue, N. Ozaki, H. Nagasawa, Purification and structural determination of a phosphorylated peptide with anti-calcification and chitin-binding activities in the exoskeleton of the crayfish, Procambarus clarkii,, Biosci. Biotechnol. Biochem. 65 (2001) 1840–1848. H. Miyamoto, F. Miyoshi, J. Kohno, The carbonic anhydrase domain protein Nacrein is expressed in the epithelial cells of the mantle and acts as a negative regulator in calcification in the mollusk Pinctada fucata, Zool. Sci. 22 (2005) 311–315. D.M. Whitfied, J. Choay, B. Sarkar, Heavy metal binding to heparin disaccharides I. Iduronic acid is the main binding site, Biopolymers 32 (1992) 585–596. G. Embery, S. Rees, K. Rose, R. Waddington, P. Shellis, Calciumand hydroxyapatite-binding properties of glucuronic acid-rich and iduronic acid-rich glycosaminoglycans and proteoglycans, Eur. J. Oral Sci. 106 (1998) 267–273. M.E. Marsh, D.P. Dickinson, Polyanion-mediated mineralization mineralization in coccolithophore ( Pleurochrysis carterae ) variants which do not express PS2, the most abundant and acidic mineralassociated polyanion in wild-type cells, Protoplasma 199 (1997) 9–17.