Growth Of Apatite Crystals Research Articles

Inducers and inhibitors of biomineralization: lessons from pathological calcification

Ectopic calcification is a common response to soft tissue injury and systemic mineral imbalance and can lead to devastating clinical consequences when present in joints, heart valves and blood vessels. We have hypothesized that mineralization of matrices in any tissue is normally controlled by a balance between procalcific and anticalcific regulatory proteins such that abnormal deposition of apatite is avoided. Alterations in this balance induced by injury, disease or genetic deficiency are postulated to induce ectopic mineral deposition. Over the past several years, we have developed in vitro and in vivo models of ectopic calcification to investigate potential inducers and inhibitors of this process. Osteopontin, a secreted phosphoprotein, has emerged as a major inhibitor of ectopic mineralization. Osteopontin is a potent inhibitor of vascular cell calcification in vitro and mice lacking osteopontin are highly susceptible to ectopic calcification. Furthermore, osteopontin treatment of biomaterials protected against ectopic mineralization. Our studies indicate that in addition to inhibiting apatite crystal initiation and growth, osteopontin stimulates resorption of ectopic calcification via peripheral macrophages and giant cells. In contrast, inorganic phosphate has emerged as a major inducer of mineralization in these systems. Elevated inorganic phosphate (Pi) was shown to induce smooth muscle cell matrix calcification with morphological properties similar to those observed in calcified human valves and atherosclerotic plaques. Furthermore, mineralization induced by inorganic phosphate was dependent on the activity of the sodium-dependent phosphate cotransporter, Pit-1. These studies implicate controlled, transcellular transport of Pi as a major requirement for matrix calcification.

Deposition of calcium phosphate coating on nanotubular anodized titanium

Nanotubular anodic oxide layer was grown on titanium surface by room temperature anodization in 0.5 M H 3PO 4 + 0.138 M fluoride solution. The inner surface of the nanotubular oxide walls contained adsorbed anions from the solution that were predominantly phosphate. Hydroxy apatite was electrodeposited on the anodized surface of Ti. Adsorbed phosphate ions facilitated nucleation of nanophase calcium phosphate material inside the nanotubular structure that resulted in vertical growth of apatite crystals. Bond strength of the calcium phosphate coating was found to be good, and cohesive mode of failure of the coating was observed after tensile testing.

Supramolecular Assembly of Amelogenin Nanospheres into Birefringent Microribbons

Although both tooth enamel and bone are composed of organized assemblies of carbonated apatite crystals, enamel is unusual in that it does not contain collagen nor does it remodel. Self-assembly of amelogenin protein into nanospheres has been recognized as a key factor in controlling the oriented and elongated growth of carbonated apatite crystals during dental enamel biomineralization. We report the in vitro formation of birefringent microribbon structures that were generated through the supramolecular assembly of amelogenin nanospheres. These microribbons have diffraction patterns that indicate a periodic structure of crystalline units along the long axis. The growth of apatite crystals orientated along the c axis and parallel to the long axes of the microribbons was observed in vitro. The linear arrays (chains) of nanospheres observed as intermediate states before the microribbon formation give an important indication as to the function of amelogenin in controlling the oriented growth of apatite crystals during enamel mineralization.

Surface chemistry of enamel apatite during maturation in relation to pH: implications for protein removal and crystal growth

Apatite crystal growth rate and morphology in mineralized tissues are considered to be controlled by surface interaction with extracellular matrix proteins. During enamel maturation where protein is finally removed from crystal surfaces to permit massive crystal growth, pH oscillates between approximately 5.8 and approximately 7.2. With this in mind, a study of enamel apatite surface chemistry in terms of local environmental pH was undertaken. Using atomic force microscopy adhesion force measurements were made between hydroxylated or carboxylated cantilever tips and maturation stage crystals between pH 2 and 10. Adhesion force increased from pH 10 to a maximum at pH 6.6 presumably due to increased hydrogen bonding due to replacement of surface cations (Na, Ca, Mg) with protons and/or protonation of phosphate per se. Below pH 6.6 adhesion force decreased and became very variable indicating that the surface had become unstable probably due to removal of fully protonated phosphate from the surface by adherence to the cantilever tip. Frictional force measurements also revealed 2-3, approximately 30 nm diameter high friction domains in bands across the crystal long axis. Their location mirrored the binding pattern of similarly sized amelogenin aggregates seen in vitro. The data suggests that specific protein binding sites may exist on crystal surfaces and may be released at lower pH by protonation which would lower cationic charge on both crystal surface and ionic charge on the protein. Instability of the crystal surface could also play a role.

Hydrolysis and cytocompatibility of zinc-containing α-tricalcium phosphate powder

Zinc is an essential trace element with stimulatory effects on bone formation in vitro and in vivo. In order to apply zinc-containing α-tricalcium phosphate (αZnTCP) to αTCP-based calcium phosphate cement, αZnTCP powders containing various amounts of zinc were prepared and the optimum zinc content was examined from the viewpoints of hydrolysis property and cytocompatibility. Single-phase αZnTCP containing zinc from 0.00 to 1.26 wt.% was obtained by heating zinc-containing β-tricalcium phosphate at 1450 °C. At a zinc content of 0.11 wt.%, there was no significant difference in the amount of calcium-deficient low-crystalline apatite formed by hydrolysis between pure αTCP and αZnTCP. The hydrolysis of αZnTCP was significantly retarded by incorporating zinc at or more than 0.26 wt.% due to the inhibitory effect of zinc on the crystal growth of apatite. The MC3T3-E1 cell densities cultured with pure αTCP and 0.11 Zn wt.% αZnTCP powders showed no significant difference in the range of powder dose from 10 to 1000 μg mL −1. The result indicated that the cytocompatibility of pure αTCP powder was comparable to that of αZnTCP at the zinc content of 0.11 wt.%. Therefore, the optimum zinc content of αZnTCP powder is less than 0.26 wt.%, and preferably at or less than 0.11 wt.% to apply to an αTCP-based calcium phosphate cement.

In Vitro and In Vivo Effects of the Overexpression of Osteopontin on Osteoblast Differentiation Using a Recombinant Adenoviral Vector

Osteopontin (OPN) is a highly acidic secreted phosphoprotein that binds to cells via an RGD (arginine-glycine-aspartic acid) cell adhesion sequence that recognizes the alphaVbeta3 integrin. OPN may regulate the formation and remodeling of bone. To elucidate the function of OPN in bone tissue, we examined the overexpression of OPN in osteoblasts in vitro and in vivo using an adenoviral vector carrying an OPN cDNA (Adv-OPN). Rat bone marrow-derived osteoblasts infected with Adv-OPN were examined by Western blotting, immunofluorescence, nodule formation measurements, assay of alkaline phosphatase (ALP) activity, and Northern blotting. The results suggested that not only osteoblast differentiation markers such as osteocalcin and ALP, but nodule formation and ALP activity are markedly enhanced by OPN overexpression in the case of viral infection. On the contrary, when Adv-OPN and uninfected osteoblasts were implanted into subcutaneous sites with a porous ceramic scaffold, the ALP activity and calcium content of the OPN-infected composite were higher than in uninfected composites, however, the differences were smaller than expected from the in vitro experiments. We speculate that the difference in the result of in vitro and in vivo experiments originates from the inhibitory effect of secreted OPN on the crystal growth of apatite in vivo, which competes with the induced activity of osteoblasts.

Control of apatite crystal growth in a fluoride containing amelogenin-rich matrix

To study how crystal growth in dental enamel is controlled by the components of the extracellular matrix, we investigated the functional roles of amelogenins and fluoride ions in apatite formation occurring through an octacalcium phosphate (OCP)-precursor pathway. Using a cation selective membrane system as a model of tooth enamel formation, we evaluated the resulting mineral habit grown in native porcine amelogenins and fluoride ions. In the absence of amelogenin and in the presence of 1 or 2 ppm F, we obtained OCP+apatite and apatite, respectively. Without amelogenins, the crystals were hexagonal prisms and cones with diameters of ∼100–200 nm. In the presence of 10% amelogenins and in the absence of fluoride, rod-like OCP with a diameter of 35 nm were obtained. Remarkably, a combination of amelogenin and fluoride created the formation of rod-like apatite crystals with dimensions similar to the former crystals. These observations indicate a cooperative role of amelogenin and fluoride in the regulation of habit, size orientation and phase of the calcium-phosphate crystals, resulting in the formation of fine rod-like apatite whose habit and orientation were similar to that of authentic tooth enamel crystals. The significant modulating effect of the amelogenin matrix combined with fluoride ions suggests the potential for this artificial system to contribute to the engineering of novel enamel-like biomaterials in vitro.

Adsorption of Amelogenin Nanopheres onto Charged Surfaces, A Model for Tooth Enamel Matrix Re-construction

AbstractAmelogenins are hydrophobic proteins that constitute more than 90% of the secretory stage enamel matrix. The assembly of amelogenin into nanospheres has been postulated to be a key factor in controlling the structural organization of the enamel extracellular matrix framework, which provides the scaffolding for the elongated and oriented growth of enamel apatite crystals. To get insight into the structure and function of amelogenin in controlling the process of crystal growth we have utilized two different approaches to investigate adsorption of amelogenin nanospheres onto charged surfaces: A) analysis of adsorption of amelogenin onto hydroxyapatite crystals by means of Langmuir Model for protein adsorption. B) analysis of amelogenin mono or multi-layer formation by sequential adsorption process onto auto-assembled polyelctrolytes films. Our data indicate that amelogenin nanospheres adsorb onto the surface of apatite crystals as binding units with defined adsorption sites. We found that amelogenin nanospheres are negatively charged and a monolayer of these nanospheres adsorbed in an irreversible way on positively ending polyelectrolyte multilayers most likely through electrostatic interactions.

Modification of calcium-phosphate coatings on titanium by recombinant amelogenin.

Amelogenin proteins, the principal components of the developing dental enamel extracellular matrix, have been postulated to facilitate the elongated and oriented growth of the carbonated apatite crystals during enamel formation. We previously reported that amelogenin caused modulation of apatite crystals nucleated on a bioactive glass (Bioglass(R)) in vitro. Here, the effects of amelogenin on the growth morphology of calcium-phosphate crystals nucleated on a titanium surface were investigated in order to gain a better understanding of the role of amelogenins during enamel biomineralization and to explore their potential application in the design and development of novel biomaterials. The dose-dependent effects of a recombinant mouse amelogenin (rM179) were found to be different from those of bovine serum albumin, which significantly inhibited apatite crystal growth and caused the octacalcium phosphate (OCP) crystals to change from a plate-like shape to a curved shape, indicating a general inhibitory effect. The effects of rM179 on the crystal growth of OCP at 12.5-100 microg/mL and of apatite at 50 microg/mL were insignificant while the apatite crystals were remarkably elongated along their c-axes upon the use of 100 microg/mL of rM179. The unique modulation of the calcium-phosphate coatings on titanium by rM179 supports the view that amelogenins have a great potential for applications designed to develop novel biomimetic materials.

Amelogenin Induces Biomimetic Mineralization at Specific pH

AbstractAmelogenin proteins are assumed to control the calcification of dental enamel with a nanoscale precision that facilitates the formation of fibrous apatite crystals organized in a remarkable microstucture. In this study, recombinant full-length human amelogenin induced protein-guided mineralization and the formation of an enamel-like composite material at specific physical-chemical conditions as observed by atomic force microscopy (AFM). Amelogenin bound specifically to fluoroapatite crystals (FAP) of a glass-ceramic substrate at Ca2+ and PO43- concentrations similar to in-vivo conditions and at pH 8. Layers up to 400 nm high, containing elongated crystals, formed on the (001)-planes of FAP within 24h in supersaturated solutions. In contrast, (hk0)-faces grew by only 10-30 nm, but showed nanospheres aligned parallel to the c-axis of FAP. At pHs different from 8, proteins bound non-specifically to substrate and layers on FAP reached only 5-15 nm thickness. Micro-Raman spectroscopy and AFM revealed the formation of a composite material that resembled a structure and composition comparable to human enamel. These observations suggest that certain conditions are required to activate amelogenin to control and promote crystal growth of apatite along the c-axis and to synthesize an enamel-like material.

Analysis of self-assembly and apatite binding properties of amelogenin proteins lacking the hydrophilic C-terminal.

Amelogenins, the major protein component of the mineralizing enamel extracellular matrix, are critical for normal enamel formation as documented in the linkage studies of a group of inherited disorders, with defective enamel formation, called Amelogenesis imperfecta. Recent cases of Amelogenesis imperfecta include mutations that resulted in truncated amelogenin protein lacking the hydrophilic C-terminal amino acids. Current advances in knowledge on amelogenin structure, nanospheres assembly and their effects on crystal growth have supported the hypothesis that amelogenin nanospheres provide the organized microstructure for the initiation and modulated growth of enamel apatite crystals. In order to evaluate the function of the conserved hydrophilic C-terminal telopeptide during enamel biomineralization, the present study was designed to analyze the self-assembly and apatite binding behavior of amelogenin proteins and their isoforms lacking the hydrophilic C-terminal. We applied dynamic light scattering to investigate the size distribution of amelogenin nanospheres formed by a series of native and recombinant proteins. In addition, the apatite binding properties of these amelogenins were examined using commercially available hydroxyapatite crystals. Amelogenins lacking the carboxy-terminal (native P161 and recombinant rM166) formed larger nanospheres than those formed by their full-length precursors: native P173 and recombinant rM179. These data suggest that after removal of the hydrophilic carboxy-terminal segment further association of the nanospheres takes place through hydrophobic interactions. The affinity of amelogenins lacking the carboxy-terminal regions to apatite crystals was significantly lower than their parent amelogenins. These structure-functional analyses suggest that the hydrophilic carboxy-terminal plays critical functional roles in mineralization of enamel and that the lack of this segment causes abnormal mineralization.

Extracellular Matrix Proteins and the Dynamics of Dentin Formation

Dentinogenesis involves controlled reactions that result in conversion of unmineralized predentin to dentin when apatite crystals are formed. This process is dynamic: Maturation events occur within predentin beginning at the proximal layer and progressing to the predentin-dentin (PD) border. One type of controlled reaction is the proteolytic processing of dentin sialophosphoprotein (DSPP) to dentin sialoprotein (DSP) and dentin phosphoprotein (DPP), by cleavage of at least three highly conserved peptide bonds. We postulate that this processing event represents an activation step, resulting in release of DPP, which is active in its effects on formation and growth of apatite crystals. Dentin matrix protein 1 (DPM1), present as a processed fragment (57-kD protein) in bone, is seen in dentin on sodium dodecyl sulfate polyacrylamide gel electrophoresis as one intact protein of 150-200 kD. Anti-57-kD antibodies elicit immunoreactivity in bone, dentin, and cellular cementum. In bone, the reactivity is associated with osteocytes and their cell processes. Similarly, dentin shows reactivity in odontoblasts, predentin, and the odontoblast processes. In summary, the processing of large sialic acid-rich proteins into smaller fragments may be an important part of the controlled conversion of predentin to dentin and osteoid to bone.

Elongated growth of octacalcium phosphate crystals in recombinant amelogenin gels under controlled ionic flow.

Amelogenin proteins constitute the primary structural entity of the extracellular protein framework of the developing enamel matrix. Recent data on the interactions of amelogenin with calcium phosphate crystals support the hypothesis that amelogenins control the oriented and elongated growth of enamel carbonate apatite crystals. To exploit further the molecular mechanisms involved in amelogenin-calcium phosphate mineral interactions, we conducted in vitro experiments to examine the effect of amelogenin on synthetic octacalcium phosphate (OCP) crystals. A 10% (wt/vol) recombinant murine amelogenin (rM179, rM166) gel was constructed with nanospheres of about 10- to 20-nm diameter, as observed by atomic force microscopy. The growth of OCP was modulated uniquely in 10% rM179 and rM166 amelogenin gels, regardless of the presence of the hydrophilic C-terminal residues. Fibrous crystals grew with large length-to-width ratio and small width-to-thickness ratio. Both rM179 and rM166 enhanced the growth of elongated OCP crystals, suggesting a relationship to the initial elongated growth of enamel crystals.

Nucleation of biomimetic Ca–P coatings on Ti6Al4V from a SBF×5 solution: influence of magnesium

Biomimetic Calcium–Phosphate (Ca–P) coatings were applied by using 5 times concentrated Simulated Body Fluid (SBF×5) using Carbon Dioxide gas. This process allows the deposition of a uniform Ca–P coating within 24 h. A previous study of our process emphasized the importance of hydrogenocarbonate ions (HCO 3 −), a crystal growth inhibitor. The aim of the present study was to investigate the role of the other crystal growth inhibitor present in SBF×5, Magnesium (Mg 2+), on the Ca–P coating formation. Several SBF×5 solutions were prepared with various Mg 2+ and HCO 3 − contents. No Ca–P deposits were detected on Ti6Al4V substrate soaked for 24 h in a Mg-free SBF×5 solution, whereas by increasing HCO 3 − content in a Mg-free SBF×5 solution, a Ca–P coating developed on Ti6Al4V substrate. Therefore, it appeared that Mg 2+ has a stronger inhibitory effect on apatite crystal growth than HCO 3 −. Nevertheless, Mg 2+ plays also another important role as suggested by depth profile X-ray Photoelectron Spectroscopy (XPS) of the Ca–P coating obtained from SBF×5 solution. Ca 2+ and Mg 2+ contents increased significantly at the titanium/coating interface. Therefore, Ca 2+ and Mg 2+ initiated Ca–P coating from SBF×5 solution. The relatively high interfacial concentration in Mg 2+ favors heterogeneous nucleation of tiny Ca–P globules onto the substrate. So physical adhesion is enhanced at the early stage of the coating formation.

Nanobacteria: controversial pathogens in nephrolithiasis and polycystic kidney disease.

Nanobacteria are unconventional agents 100-fold smaller than common bacteria that can replicate apatite-forming units. Nanobacteria are powerful mediators of biogenic apatite nucleation (crystal form of calcium phosphate) and crystal growth under conditions simulating blood and urine. Apatite is found in the central nidus of most kidney stones and in mineral plaques (Randall's plaques) in renal papilla. The direct injection of nanobacteria into rat kidneys resulted in stone formation in the nanobacteria-injected kidney during one month follow-up, but not in the control kidney injected with vehicle. After intravenous administration in rats and rabbits, nanobacteria are rapidly excreted from the blood into the urine, as a major elimination route, and damage renal collecting tubuli. Nanobacteria are cytotoxic to fibroblasts in vitro. Human kidney cyst fluids contain nanobacteria. Nanobacteria thus appear to be potential provocateurs and initiators of kidney stones, tubular damage, and kidney cyst formation. It is hypothesized that nanobacteria are the initial nidi on which kidney stone is built up, at a rate dependent on the supersaturation status of the urine. Those individuals having both nanobacteria and diminished defences against stone formation (i.e. genetic factors, diet and drinking habits) could be at high risk. Kidney cyst formation is hypothesized to involve nanobacteria-induced tubular damage and defective tissue regeneration yielding cyst formation, the extent of which is dependent on genetic vulnerability.

The ank gene story.

The underlying molecular defect resulting in the abnormal calcification observed in ank/ank mice has been identified. The responsible nonsense mutation affects the protein product of ank, resulting in diminished production of extracellular inorganic pyrophosphate, an important inhibitor of nucleation and of the growth of apatite crystals. The ank gene product is one of several cell membrane proteins, including ectonucleoside triphosphate pyrophosphohydrolase enzymes and alkaline phosphatase, that regulate extracellular inorganic pyrophosphate levels and thereby regulate mineralization.

Type I collagen influences cartilage calcification: An immunoblocking study in differentiating chick limb-bud mesenchymal cell cultures

Chick limb-bud mesenchymal cells, plated in high-density micro-mass culture, differentiate and form a matrix resembling chick epiphyseal cartilage. In the presence of 4 mM inorganic phosphate or 2.5 mM beta-glycerophosphate mineral deposits upon this matrix forming a mineralized tissue that, based on electron microscopy, x-ray diffraction and Fourier Transform Infrared microspectoscopy, is like that of chick calcified cartilage. In this culture system the initial mineral deposits are found on the periphery of the chondrocyte nodules. During differentiation of the cells in the high-density micro-mass cultures there is a switch from expression of type I collagen to type II, and then to type X collagen. However, type I collagen persists in the matrix. Because there is some debate about whether type I collagen influences cartilage calcification, an immunoblocking technique was used to determine the importance of type I collagen on the mineralization process in this system. Studies using nonspecific goat anti-chick IgG demonstrated that 1-100 ng/ml antibody added with the media after the cartilage nodules had developed (day 7) had no effect on the accumulation of mineral in the cultures. Nonspecific antibody added before day 7 blocked development of the cultures. Parallel solution based cell-free studies showed that IgG did not have a strong affinity for apatite crystals, and had no significant effect on apatite crystal growth. Type I collagen antibodies (1-200 ng/ml) added to cultures one time on day 9 (before mineralization started), or on day 11 (at the start of mineralization), slightly inhibited the accumulation of mineral. There was a statistically significant decrease in mineral accretion with 100 or 200 ng/ml collagen antibody addition continuously after these times. Fab' fragments of nonspecific and type I collagen antibodies had effects parallel to those of the intact antibodies, indicating that the decreased mineralization was not attributable to the presence of the larger, bulkier antibodies. The altered accumulation of mineral was not associated with cell death in the presence of antibody (demonstrated by fluorescent labeling of DNA) or with increased apoptosis (TUNEL-stain). In the immunoblocked cultures, EM analysis demonstrated that mineral continued to deposit on collagen fibrils, but there appeared to be fewer deposits. The data demonstrate that type I collagen is important for the mineralization of these cultures.

Inhibition of Apatite Formation by Vitronectin

This study investigated the concentration-dependent effect of vitronectin (VN), a glycoprotein of the bone matrix, on apatite formation and growth. Precipitation trials in metastable solution and in a pH-controlled solution system showed an inhibition of apatite microcrys-tal formation by VN. In the presence of biphasic calcium-phosphate ceramic, transmission electron microscopy showed a reduction of precipitated microcrystal size: precipitates were significantly smaller than in ionic simulated body fluid without proteins or in the presence of type I collagen as a negative control. Moreover, the size of the precipitated microcrystals was reduced in a dose-dependent manner. Two indirect methods showed that calcium-phosphate precipitation was inhibited by VN. It would appear that VN prevents apatite formation by inhibiting the growth of apatite crystals rather than by secondary nucleation, as in the case of osteopontin, a bone-specific protein.

Modulation of apatite crystal growth on Bioglass by recombinant amelogenin.

The effects of a recombinant mouse amelogenin (rM179) on the growth of apatite crystals nucleated on a bioactive glass (45S5 type Bioglass) surface were investigated with a view to gaining a better understanding of the role of amelogenin protein in tooth enamel formation and of its potential application in the design of novel enamel-like biomaterials. Bioglass discs were incubated in phosphate-buffered saline (PBS) to preform a calcium phosphate surface layer and subsequently immersed in blank, bovine serum albumin (BSA)- and rM179-containing supersaturated calcification solutions (SCS(B), SCS(BSA) and SCSrM179), respectively. Calcium phosphate layers formed on all the treated samples and were characterized to be apatite by X-ray diffraction and Fourier transmission infrared spectrophotometry. Under scanning electron microscopy, plate-shaped crystals (approximately 50 nm thick and 300-600 nm across) were observed on the samples after PBS incubation. The crystals grown from SCS(B) were of the typical plate shape except for an increased thickness, while needle-shaped crystals (200-300 nm long and 50-70 nm thick) were precipitated on the SCS(BSA)-immersed samples. Interestingly, it was found that the crystals deposited on the SCSrM179-immersed samples adopted an elongated, curved shape (approximately 500 nm long and approximately 120 nm thick). Further TEM observations showed that the crystals generated by the SCSrM179 immersion appeared to be composed of bundles of lengthwise crystals (15-20 nm thick) orientated parallel to one another, much alike the long and thin crystals observed in the very early stage of enamel formation. The significant modulation by the rM179 protein of apatite crystal growth is quite different from the overall inhibition observed by BSA and most likely is relevant to the specific function of the amelogenin matrix in controlling enamel crystal growth in vivo.

Effects of ionic inflow and organic matrix on crystal growth of octacalcium phosphate; relevant to tooth enamel formation

The mechanism of the lengthwise and oriented growth of enamel apatite crystals was studied on the basis of a hypothesis that (1) enamel crystals grow initially as octacalcium phosphate (OCP) as a precursor of apatite and (2) a one-directional Ca 2+ ion supply from the ameloblasts promotes the lengthwise and oriented growth of the crystals. To test this, OCP crystals were grown in a model system of enamel formation, where a cation selective membrane was used to control the Ca 2+ ion diffusion, at 37°C and at pH 6.3–7.4. The lengthwise and oriented growth of OCP was enhanced by the Ca influx from the membrane. H + ion enhanced while F −, Mg 2+ and CO 3 2− ions decreased the growth in the c-axis direction. F − induced an epitaxial overgrowth of apatite on the (1 0 0) of OCP, embedding an OCP lamella in the center of a crystal. When 5% polyacrylamide gel was attached to the membrane, OCP grew in long plate-like or ribbon-like crystals at pHs between 6.5 and 7.4. The length of these crystals was greater than that of those grown on the membrane without gel. The crystal size decreased with an increase in gel concentration from 5 to 20%. When a thin slice of tendon was used in place of the membrane, long OCP crystals grew along the collagen fibrils; these crystals were oriented with their axis of elongation perpendicular to the direction of ionic inflow. The size of crystals grown inside the collagenous matrix was much smaller than the size of crystals formed outside of the matrix and on the membrane. Thus, crystal growth was regulated by ionic diffusion and the matrix. When there were no constituents which regulated the growth direction, OCP crystals tended to grow along the direction of ionic inflow.