Search

We report here the finding that normal, young cartilages, in distinction from all other tissues examined, have unusually high levels of n-9 eicosatrienoic (20:3 cis-delta 5,8,11) acid and low levels of n-6 polyunsaturated fatty acids (n-6 PUFA). This pattern is identical to that found in tissues of animals subjected to prolonged depletion of nutritionally essential n-6 polyunsaturated fatty acids (EFA). This apparent deficiency is consistently observed in cartilage of all species so far studied (young chicken, fetal calf, newborn pig, rabbit, and human), even though levels of n-6 PUFA in blood and all other tissues is normal. The n-9 20:3 acid is particularly abundant in phosphatidylethanolamine, phosphatidylinositol, and the free fatty acid fractions from the young cartilage. Several factors appear to contribute to the reduction in n-6 PUFA and the appearance of high levels of the n-9 20:3 acid in cartilage: 1) limited access to nutritional sources of EFA due to the impermeability and avascularity of cartilage, 2) rapid metabolism of n-6 PUFA to prostanoids by chondrocytes, and 3) a unique fatty acid metabolism by cartilage. Evidence is presented that each of these factors contributes. Previously, EFA deficiency has been shown to greatly suppress the inflammatory response of leukocytes and rejection of tissues transplanted into allogeneic recipients. Because eicosanoids, which are derived from EFA, have been implicated in the inflammatory responses associated with arthritic disease, reduction of n-6 PUFA and accumulation of the n-9 20:3 acid in cartilage may be important for maintaining normal cartilage structure.

Matrix vesicles (MV) are microstructures localized to the extracellular matrix of developing hard tissues that induce mineral formation. MV proteins are not well characterized, and little is known of how they interact with the surrounding matrix. However, recent electron microscopic studies indicate that MV interact with matrix proteins in growth plate cartilage. In the studies now reported, procedures developed for dissecting various components from isolated MV led to the discovery that two major vesicle proteins (38 and 46 kDa) are readily released from MV by low ionic strength solutions. These low ionic strength-soluble proteins (LISSP) were shown to be major fragments of the link protein (LP) and hyaluronic acid-binding region (HABR) of matrix proteoglycans: they react immunologically with highly specific monoclonal antibodies to LP and HABR, and the NH2-terminal sequence of the 38-kDa LISSP is essentially identical to residues 40-78 of chicken cartilage LP and that the 46-kDa LISSP represents HABR. Release of both LISSP is enhanced by hyaluronidase treatment, indicating anchorage by a hyaluronate-mediated mechanism. Both LP and HABR are firmly attached to MV in either isotonic or hypertonic solutions. In contrast, our other studies show that dissociation of type II collagen from MV occurs only with hypertonic salts which do not release the LISSP. Thus, strong interactions occur under physiological conditions between MV and both the proteoglycans and collagens, but these take place by different mechanisms.

Recent evidence indicates that matrix vesicles (MV) interact with cartilage-specific collagens and other matrix proteins. Both type II and X collagens bind to and cosediment with MV. Our companion study shows that MV also are tightly coupled to proteoglycan link proteins (LP) and hyaluronic acid-binding region (HABR) in cartilage matrix. Here we sought to identify proteins responsible for the nexus between MV and matrix collagens using affinity chromatography with types I, II, and X collagen-Sepharose columns. Elution with NaCl step-gradients in the presence of nonionic detergent was used to assess the affinity between the MV proteins and the covalently attached collagens. Several MV proteins were found to bind to native type I, II, and X collagens but none bound to denatured type I collagen. Alkaline phosphatase, proteoglycan LP and HABR, and the 33- and 67-kDa annexins, bound with varying affinities to the native type I, II and X columns. In particular, LP and HABR, the 67-kDa annexin, and alkaline phosphatase bound with high affinity to the cartilage-specific collagens, although LP, HABR, and a 37-kDa protein also bound less tightly to native type I collagen. Thus, several MV proteins bind specifically to native type II and X collagens and should promote interaction between MV and the extracellular matrix. Such interactions may be important in MV formation, or in MV-mediated mineralization.

Matrix vesicles (MV) initiate de novo mineralization in a variety of vertebrate-calcifying tissues. In recent studies, a quantitatively major group of MV proteins, the acidic phospholipid-dependent Ca2(+)-binding proteins (APD-CaBP) were found to be immunologically related to the annexin family of proteins that possess phospholipase A2 inhibitory activity. This finding helped explain the enrichment of phosphatidylserine as well as the presence of large amounts of complexed Ca2+ noted previously in these structures. To characterize further these annexin-like proteins, preparations of both collagenase-released MV and MV-enriched microsomes were subjected to a differential fractionation process that led to the isolation and purification to homogeneity of two of the MV APD-CaBP, a 33-kDa protein and a 36-kDa calpactin II-like protein. Polyclonal antibodies raised to each pure protein were found not to cross-react with the other, thus indicating two distinctive proteins. Measurement of the phosphatidylserine-dependent Ca2(+)-binding properties of the proteins revealed apparent Kd values of 2.5 x 10(-7) and 5.0 x 10(-7) M for the 36- and 33-kDa proteins, respectively. Such high affinities indicate that both proteins would be normally bound to the membrane of MV. Immunological studies revealed the presence of both APD-CaBP in cultured growth plate chondrocytes but not in vesicles released into the culture medium. The finding of the 33-kDa but not the 36-kDa protein in vesicles released from the calcifying matrix of the chondrocyte cultures by collagenase digestion may indicate a role for this protein in MV mineralization.

The capacity of matrix vesicles (MV) to induce mineralization under various electrolyte conditions has not been explored. Accordingly, we examined the ability of isolated MV to induce calcification using synthetic lymphs with ranges of Ca/Pi ratio, Ca2+ x Pi ion product, and pH relevant to both normal and pathological conditions. At a fixed ion product of 2.84 mM2, 45Ca2+ uptake was supported at all Ca/Pi ratios tested, with ratios of 1.3-1.4 being optimal. Rapid ion uptake became saturated at levels greater than 2.7 mM2 when studied at a fixed Ca/Pi = 1.3, indicating a rate-limiting membrane ion porter. However, treatment of MV with non-ionic detergent did not destroy their ability to induce mineralization. At constant Ca/Pi of 1.3 and Ca2+ x Pi of 2.63 mM2, maximal uptake rates occurred at pH 7.6-7.8 over a pH range of 7.0-8.0, with significant uptake being supported only over the narrow range of pH 7.4-7.8. Studies showing that the effects of pH on amorphous calcium phosphate (ACP)-mediated calcification were very similar to those of MV, indicate that a stabilized form of internal ACP may induce crystalline mineral formation during MV-mediated calcification.

The conversion of amorphous calcium phosphate (ACP) to crystalline hydroxyapatite (HA) was studied in vitro in the presence or absence of phosphatidyl serine (PS) and other phospholipids. ACP transformation and HA crystal growth were monitored by electron microscopy, selected-area electron diffraction and X-ray diffraction, and by measuring supernatant calcium and phosphate. PS, and other acidic phospholipids, had a significant stablizing effect on ACP. With performed ACP at ratios of only 1 lipid molecule per 30-50 Ca atoms, PS markedly delayed HA crystal formation. When PS was present during ACP precipitation, inhibition of conversion to HA was less pronounced, but crystal habit and aggregation were greatly altered resulting in stacks of thin, membrane-like sheets approximately 38-42 A thick. PS appeared to be most effective in blocking ACP to HA conversion when oriented primarily on the surface; it most affected subsequent crystal formation when distributed throughout the amorphous precursor. Phospholipids possessing anionic, and hence Ca-binding properties, were effective in stabilizing ACP; neutral zwitterion lipids, which have amphipathic properties but do not bind Ca, were not. In view of the presence of anionic lipids in matrix vesicles and their association with early mineral deposits, the current findings add further evidence that lipids may play a role in the control of normal mineralization in vivo.

The morphological features of avian epiphyseal cartilage have been investigated by freeze-fracture techniques. Progressive changes occurred in both the cells and the matrix during differentiation. Chondrocytes changed in shape from small flattened cells with few, short cellular processes, to enlarged ovoid cells with numerous long processes often associated with extracellular vesicles. In the matrix these vesicles appeared first in the cellular lacunae, then in the extralacunar matrix, becoming larger and more numerous. Large membrane-associated particles (MAPS) were seen on the p faces of the plasmalemma. These became progressively concentrated on and around the cellular processes, with few large MAPS being seen on the e face. Similar distribution of MAPS was seen in the matrix vesicles. Domains of hydrated proteoglycan aggregates were manifest as regular fracture patterns in the extralacunar matrix of the upper regions of the plate. Collagen fibrils progressively increased in size and state of aggregation, often being associated with matrix vesicles and in the end, with long plate-like mineral crystals. These findings, while in basic agreement with patterns observed with TEM, reveal important new features concerning cellular and matrix structure during cartilage differentiation.

Although alkaline phosphatase has been long associated with the mineralization process, its exact function remains to be elucidated. To clarify its possible role in matrix vesicle-mediated mineralization, we tested the effect of vanadate, a phosphate analogue and powerful competitive inhibitor of alkaline phosphatase activity, on calcium and phosphate uptakes by a matrix vesicle-enriched microsomal fraction. Vanadate was also tested in a hydroxyapatite-seeded ion uptake system to determine possible direct effects on mineral formation. The effect of vanadate on vesicle mineral ion uptake was complex; low dosages of vanadate (2-20 microM) were stimulatory to Ca2+ uptake, but were inhibitory to Pi. Higher dosages (greater than 67 microM) were inhibitory to both ions. The effect of vanadate on ion uptake was strongly influenced by the stage of vesicle loading; major effects were seen during the lag and early uptake phases, and minimal effects were seen in the terminal stages. Concentrations of vanadate highly inhibitory to vesicle ion uptake had minimal effects on ion accretion by a hydroxyapatite-seeded system. Inhibition of alkaline phosphatase activity by vanadate broadly paralleled inhibition of Pi and Ca2+ uptake; however, at low vanadate concentrations, inhibition of Pi uptake closely paralleled that of alkaline phosphatase. The data indicate that vanadate binds with high affinity to Pi-loading sites, blocking initial Pi uptake. Complexation between vanadate and Ca2+ may be responsible for the stimulation of Ca2+ uptake at early stages of vesicle ion loading with low levels of vanadate by enhancing binding of Ca2+ to the vesicles. It may also account for the selective inhibition of Ca2+ uptake during the rapid stage of vesicle ion loading with high levels of vanadate by reducing Ca2+ ion activity. The close parallelism between inhibition of early Pi uptake and of alkaline phosphatase activity supports the concept that alkaline phosphatase is involved in Pi transport during the early stages of matrix vesicle ion loading. However, the fact that only about half of the Pi uptake was affected by vanadate, despite the progressive inhibition of alkaline phosphatase activity, indicates that alkaline phosphatase is not solely responsible for Pi uptake by the matrix vesicle-enriched fraction.

The purpose of this study was to determine whether the accumulation of substantial levels of fluoride in the skeleton during bone formation would significantly reduce, or prevent, the occurrence of osteoporosis experimentally induced at a later time with a low mineral-containing diet. Male weanling rats were pretreated for 15 weeks with either 0 or 50 ppm F in their drinking water, followed by treatment for four to twelve weeks with diets either optimal or deficient in both Ca and P. Intake of the Ca and P deficient diet resulted in a progressive loss of bone mass (per unit volume of femur), with little effect on the remaining bone. High skeletal F offered no protective effect. Ca and P depletion caused a significant decrease in bone breaking strength but produced only slight reduction in percentage ash in bone. Fluoride offered little protection against either. Neither skeletal F nor dietary Ca and P levels influenced the concentration of Ca, P or Ca/P ratios in bone ash. Skeletal Mg levels were slightly lower in the Ca and P depleted rats; however, F treatment essentially eliminated this. Ca and P depletion resulted in higher F levels in both incisors and bone. We conclude that osteoporosis was produced by feeding the Ca and P deficient diet. High skeletal F did not prevent nor reduce the severity of this dietary-induced osteoporosis.

The mechanism of matrix vesicle (MV) formation by growth plate chondrocytes in primary cell culture was assessed both by using drugs which interfere with assembly or disassembly of microfilaments and microtubules, as well as by comparison of the composition of chondrocyte microvilli with MV. Cytochalasin D, which is known to inhibit assembly of actin microfilaments, was found to stimulate the release of alkaline phosphatase-rich MV. This stimulatory effect was confirmed by studies with [3H]palmitate- and 32P-prelabeled cells which showed that cytochalasin D enhanced the release of labeled MV. In contrast, phalloidin, which blocks disassembly of microfilaments, suppressed release of cellular alkaline phosphatase into MV. The phospholipid composition of vesicles released by cells treated with cytochalasin D and phalloidin was virtually identical with that of the controls. In contrast, colchicine, which interferes with the assembly of microtubules, was found to cause fragmentation of the cells, producing large vesicles significantly different in lipid composition from MV. Microscopic studies revealed that cytochalasin D caused marked rounding and retraction of the cells, with evidence of actin withdrawal from the cell periphery. This led to cell surface blebbing and formation of small zeiotic bodies at the tips of cell processes. In contrast, phalloidin enhanced and stabilized the actin network within the cells. Chemical analysis of microvilli prepared from isolated chondrocytes revealed high levels of alkaline phosphatase and a phospholipid composition almost identical to MV. Electrophoretic profiles of microvillar proteins were again like that of MV, except for the presence of high levels of actin. This cytoskeletal protein was nondetectable in MV. Taken together with the effects of the drugs, the data indicate that cell surface microvilli are the precursors of MV and that retraction of the supporting microfilament network is essential for the release of these structures.

The interaction between Ca2+ and phosphatidylserine was studied by 13C and 31P NMR spectroscopy, by IR analysis, by binding constant measurements, and through use of space-filling molecular models. NMR measurements of various salt forms of the lipid were made in two types of organic solvents that allowed sufficient averaging of chemical shift anisotropy and dipolar couplings to yield high resolution spectra. 13C resonances of the polar head-group carbons were broadened relative to those of the acyl chains. This was especially true in samples prepared at neutral pH where ionic interactions appeared to restrict molecular motion. In CDCl3 the marked line broadening of the resonances of the polar head-group atoms in the Ca2+ form indicated the formation of large, slow tumbling micelles. In the amphipathic solvent the large reduction in line broadening indicated the presence of freely tumbling Ca-(phosphatidylserine)2 dimeric complexes. The 2:1 binding stoichiometry and the low chemical activity of the Ca-phosphatidylserine complex support this view. Analysis of the chemical shifts of the various lipid atoms under the differing ionic environments indicates that Ca2+ enhanced the deprotonation of both the carboxyl and amino groups and stabilized the entire polar head group against the effects of changing pH. The marked upfield shift of the 31P phosphate resonance in the Ca2+ form and its insensitivity to changing pH indicate strong coordination binding. IR data indicate direct involvement of the carboxyl group in Ca2+ binding, as evidenced by the appearance of a C=O stretching mode. Binding studies indicated that the phosphate group was the primary binding force but that the carboxyl group also contributes positively. The amino group appears to exert a repulsive effect, which is supported by the chemical shift data which indicate that Ca2+ enhances the deprotonation of the amino group. Molecular models indicate direct involvement of the carboxyl and phosphate oxygens and that the amino group must be deprotonated to participate.

Alkaline phosphatase has been implicated in matrix vesicle-mediated calcification. We sought to clarify its role in this process by studying the effect of L-tetramisole, a strong uncompetitive inhibitor of this enzyme, and D-tetramisole, its inactive isomer, on the uptake of 32Pi and 45Ca by matrix vesicle-enriched microsomes obtained from chicken epiphyseal cartilage. Uptake of both 32Pi and 45Ca was inhibited in a dosage-dependent manner by the L-isomer; however, 32Pi uptake was selectively inhibited at low drug concentrations and at early time periods. With increasing incubation time, inhibition of uptake of both ions was lost. Apatite crystal growth was only weakly inhibited by both the D- and L-isomers of tetramisole. The kinetics of ion uptake by the vesicles was complex. 45Ca/32Pi uptake ratios, taken with the selective inhibiton of 32Pi uptake by L-tetramisole, indicated that early phases of vesicle ion uptake might be governed by alkaline phosphatase, a known Pi-binding protein. Later stages were clearly controlled by apatite formation. Although inhibition of vesicle uptake of 32Pi and 45Ca by graded levels of L-tetramisole paralleled inhibition of alkaline phosphatase activity, surprisingly, D-tetramisole was almost as inhibitory of ion uptake as the L-isomer. This finding, coupled with the fact that alkaline phosphatase substrates were not required for vesicle ion uptake, indicates that alkaline phosphatase hydrolase activity is not directly involved in vesicle-mediated calcification under these experimental conditions.

Microsomal fractions of cells isolated from chick epiphyseal cartilage catalyzed the synthesis of prostaglandins from radiolabeled delta8,11,14-eicosatrienoic and from arachidonic acids. In addition, the microsomal supernatants contained both 15-hydroxyprostaglandin dehydrogenase and prostaglandin 15-keto delta13,14-reductase activities. Two major classes of prostaglandins (E and F) were synthesized; however, a major product which chromatographically behaves as PGA was also isolated. Synthetase activities were analyzed for pH optima and response to known stimulators and inhibitors of prostaglandin systhesis. The different activators had varying stimulatory effects on prostaglandin synthesis; the anti-inflammatory drugs were all strongly inhibitory. Synthetase activity in the growth plate was highest in the zone of hypertrophy, declining substantially in the more heavily calcified regions. Degradative enzyme activities were highest in the zone of maturation and significantly lower in the adjacent hypertrophic zone. The net effect of these opposing activities would be to elevate prostaglandin levels at the zone of hypertrophy, a finding which suggests that prostaglandins may play a role in the modulation of epiphyseal cartilage metabolism.

Alkaline phosphatase has been purified from microsomes of chicken epiphyseal cartilage by first selectively extracting certain adventitious proteins with 0.25 M trichloroacetate. The membrane-bound enzyme was then solubilized by 1% cholate in buffered 33% saturated ammonium sulfate and purified by column chromatography on Bio-Gel A-5m, extraction with 1-butanol, and ion exchange chromatography on DEAE-Bio-Gel A. The purified alkaline phosphatase from the cartilage membrane had a subunit molecular weight of 53,000 and a holoenzyme weight of 207,000-220,000, indicating a tetramer. The pH optima for p-nitrophenylphosphate, ATP, and pyrophosphate hydrolysis were 10.3, 9.0, and 8.5, respectively. Values of Vmax (in micromoles/min/mg) were 220, 3.1, and 0.8, respectively. Substrate inhibition was pronounced at values of pH below 8.5. Inhibition of p-nitrophenylphosphate hydrolysis at pH 10.3 showed that phosphate and arsenate were competitive inhibitors (KI = 1.88 and 0.15 mM, respectively) and levamisole was an uncompetitive inhibitor (KI = 0.32 mM), while L-phenylalanine and ZnCl2 were mixed inhibitors (KI = 15.8 and 0.02 mM, respectively). Inhibition by preincubation in 1 mM EDTA was reversible by readdition of 0.25 mM MgCl2 nd 20 microM ZnCl2. The data indicate that this membrane-bound alkaline phosphatase from chicken epiphyseal cartilage is a Zn2+ and possibly Mg2+-containing enzyme. While the subunit molecular weight and kinetic properties of the enzyme are quite typical of vertebrate alkaline phosphatases, the tightness of binding to the membrane lipids, the extreme sensitivity to substrate inhibition, and the tetrameric conformation of the holoenzyme are unusual.

Inorganic pyrophosphate (PPi) and two diphosphonates, ethane-1-hydroxy-1, 1-diphosphonate (EHDP) and dichloromethylene diphosphonate (Cl2MDP), were found to inhibit in vitro mineralization induced by matrix vesicle-enriched fractions from chicken epiphyseal cartilage. Inhibitor concentrations from 0.20 to 20 microM caused a dosage-dependent decrease in 45Ca and 32Pi uptake by the vesicle fraction. These inhibitors were also tested in a hydroxyapatite (HA)-seeded system to help distinguish between effects on the mineral vs nonmineral portions of the vesicle fraction. The order of inhibition of the HA-seeded system was EHDP greater than PPi greater than Cl2MDP, except for inhibitor concentrations of 0.20 microM where EHDP was the least inhibitory. This variation may be due to differences in the binding of the inhibitors to HA crystals. In general, inhibition of HA mineralization was greatest during later time periods, whereas vesicle ion uptake was affected more during early stages of incubation. The differential effects of the three inhibitors were most obvious at the 2.0 microM concentration. With PPi substantial inhibition of HA-seeded mineralization was observed even in late stages of the study; in contrast, with time the vesicle fraction overcame this inhibition. This suggests that alkaline phosphatase, an enzyme notably enriched in matrix vesicles, catalyzed the hydrolysis of PPi, reducing its concentration to a level where mineralization could proceed. Our findings show that matrix vesicle-induced mineralization differs significantly from apatite-induced mineralization. The data support the concept that vesicle alkaline phosphatase acts, at least in part, to remove physiological crystal growth inhibitors.

Studies were made on the disposition of mineral ions in matrix vesicles (MV) and their relationship to alkaline phosphatase by treatment of MV-enriched microsomes (MVEM) with graded levels of Ca2+-chelating agents to complex accessible ions, fractionation of MVEM on hypertonic sucrose gradients at two different pH values (7.5 and 8.0) to evaluate for the presence of calcium phosphate mineral, and passage of MVEM through cation-exchange columns to determine the accessibility of the Ca2+. The effect of removal of Ca2+ and Pi on subsequent ability of MVEM to induce mineral formation from synthetic cartilage lymph was also determined. Passage through cation-exchange columns revealed that MV Ca2+ was not freely exchangeable, but coeluted in the void volume with alkaline phosphatase. However, upon incubation in synthetic cartilage lymph, progressively more Ca2+ was retained by the column. These findings indicate that, initially, the majority of Ca2+ in MVEM is internal and not readily exchangeable, but as Ca2+ accumulates, progressively more becomes external. The mineral in MV is labile and readily susceptible to loss; treatment with graded levels of EGTA removed major portions of the original Ca2+ and Pi. 45Ca uptake by these mineral-depleted MV was markedly reduced, even in the presence of alkaline phosphatase substrates. Sucrose gradient fractionation of MVEM caused extensive loss of Pi, but not Ca2+, from the low-density alkaline phosphatase-rich fractions. This reveals that Ca2+ and Pi are not initially coupled together: Pi is largely soluble, whereas Ca2+ must be tightly bound. In the high-density vesicles, large amounts of both Ca2+ and Pi are present. The slightly enhanced recovery at higher pH suggests the presence of a solid mineral phase. During mineralization by MV, Ca2+ became externalized, and concomitantly alkaline phosphatase activity declined. This suggests that a direct association exists between the enzyme and the developing mineral.

The in vivo metabolism of 32P orthophosphate into phospholipids of chondroxyte, matrix vesicle, and membrane fractions of chicken epiphyseal cartilage has been studied. Incorporation of radioactive phosphate into the total phospholipid fraction of matrix vesicles was rapid, the labeling of phosphatidylcholine and lysophosphatidylserine being even more rapid in matrix vesicles than in chondrocytes. These findings indicate that matrix vesicles are formed by a rapid, metabolically active process, and are not remnants of dead cells, as had previously been postulated by some workers. The rate of incorporation of 32P orthophosphate into phosphatidylserine and sphingomyelin of matrix vesicles was significantly slower than that of phosphatidylcholine and certain other vesicle phospholipids. These findings are paradoxical because, compared with chondrocytes, matrix vesicles were enriched in phosphatidylserine and sphingomyelin and depleted in phosphatidylcholine. These results indicate that in vesicle formation the rates of degradation of the various phospholipid classes must be markedly different: phosphatidylcholine must be degraded much more rapidly than either phosphatidylserine or sphingomyelin. Support for this comes from previous data which revealed that substantial phospholipase activity is present in epiphyseal cartilage, especially in the zone of hypertrophy where matrix vesicle formation appears to be particularly active.

Matrix vesicle-enriched fractions were isolated from different zones of epiphyseal cartilage by nonenzymatic methods involving tissue homogenization, differential centrifugation, and isosmotic Percoll gradient fractionation. Uptakes of both 32Pi and 45Ca were studied concomitantly over periods from 20 min to 24 h. Percoll density gradients separated epiphyseal microsomes into two alkaline phosphatase-rich fractions: a low-density noncalcifiable fraction (P-I), and a higher-density fraction (P-II) which readily mineralized. The P-II fraction was found only in calcifying regions of the growth plate. Based on chemical and physical properties and enzyme activities, both fractions were similar except that P-II contained significantly higher levels of mineral ions than did P-I, and had lower levels of alkaline phosphatase. The mineral appeared to be primarily in a noncrystalline form. Metabolism of 32Pi and 45Ca by P-II followed a complex kinetic pattern in which accumulation of large amounts of both ions was preceded by an initial limited burst of uptake and a lag-phase of variable duration. During mineral ion loading, the density of the P-II fraction progressively increased as evidenced by co-migration of 45Ca, 32Pi, and alkaline phosphatase to increasingly higher densities. During the period of early mineral deposition (1-5 h), Ca/P uptake ratios were very low (1.0-1.2) and X-ray diffraction patterns showed a predominantly amorphous pattern. This suggests that the mineral accumulated in matrix vesicles is initially some form of noncrystalline calcium monohydrogenphosphate. L-tetramisole, a potent inhibitor of alkaline phosphatase, inhibited accumulation of both 45Ca and 32Pi in the absence of organic P substrates, 32Pi being preferentially inhibited over 45Ca. This finding, coupled with recent studies on the behavior of alkaline phosphatase at physiological pH, suggests that the protein is not acting as a phosphohydrolase, but rather as a Pi-binding or transport agent in vesicle-mediated calcification.

Isolation of two membrane-bound alkaline phosphatase (AP) species from avian growth plate cartilage matrix vesicle (MV) fractions is described. AP was first released from the membranes by phosphatidylinositol-specific phospholipase C (PIase C), followed by chromatography on DEAE-Bio-Gel A and Reactive-Red agarose. Two AP species having apparent Mr of 81.5 and 77 kDa by SDS-PAGE were purified in high yield and specific activity by this simple method. Treatment with neuraminidase to remove sialic acid residues reduced their size slightly, but did not diminish the difference in Mr between the two species. Digestion with N-glycanase, however, decreased both AP species to a common size of 59 kDa. This reveals that both enzymes are highly glycosylated and suggests that the two forms may result from differences in degree of glycation. The amino acid compositions of the two avian enzyme forms are very similar, but are markedly enriched in serine, glycine and glutamate when compared to those reported for mammalian liver-kidney-bone AP. Possible differences in amino acid sequence between the two avian forms have not been excluded. The cross-reactivity of polyclonal antibodies to these enzymes with bovine kidney, but not intestinal AP, indicate that the avian cartilage APs are of the liver-kidney-bone isozyme type.

Activity of the bone/liver/kidney isozyme of alkaline phosphatase (AP) is known to be critical for mineralization in developing bone, although its role is unclear. The work now reported explores changes in the activity of this Zn2+-containing enzyme that occur during Ca2+ accumulation by matrix vesicles (MV). A marked loss (up to 65-70%) in AP activity was found to accompany Ca2+ accumulation by MV. These two events were highly correlated, both temporally and quantitatively. Investigation into possible causes revealed that the decline in AP activity during Ca2+ uptake was not due to action of proteases but rather resulted from interaction with the developing mineral phase, loss of metal ions (Zn2+ and Mg2+) from the active site of the enzyme, and concomitant irreversible denaturation of the enzyme. Protease inhibitors did not protect AP from loss of activity during mineralization; in contrast, protease treatments, which progressively destroyed the ability of MV to accumulate Ca2+ actually reduced loss of AP activity. These findings clearly demonstrate that AP is present at the site of MV mineralization and that its catalytic activity is profoundly reduced by the mineralization process.

Epiphyseal cartilage was fractionated into subcellular components by nonenzymatic methods, and analyzed for activity of marker enzymes, for phospholipids, and for calcium and inorganic phosphate. Alkaline phosphatase, a marker enzyme for matrix vesicles and plasma membranes, was concentrated in the 100 000 X g (microsomal) pellet and, upon subsequent fractionation, in the low-density fractions from the sucrose gradient. Mitochondrial and endoplasmic reticular enzymes were localized primarily in the 20 000 X g pellet, lysosomal enzymes predominantly in the supernate from the microsomal pellet. Two phospholipids characteristic of matrix vesicles, sphingomyelin and phosphatidylserine, were enriched in the low-density sucrose fractions; however, unlike matrix vesicles, there was no depletion in phosphatidylcholine or increase in lysophospolipids. Ca and inorganic P were concentrated in the higher-density fractions, the amounts in the lower-density fractions being somewhat lower than those seen in matrix vesicles. The alkaline phosphatase-rich, low-density fractions were thus not identical to matrix vesicles isolated by collagenase digestion, but rather appear to be composed primarily of plasma membranes. Enzyme profiles indicate they were relatively free of mitochondrial, endoplasmic reticular and lysosomal contaminants. the data further indicate that significant modification of the phospholipid, electrolyte, and possibly enzyme content of chondrocyte plasma membranes, must occur during blebbing and matrix vesicle formation.

Utilization and production of amino acids by primary cultures of chicken growth plate epiphyseal chondrocytes grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum were investigated in both short-term (6-72 h) and long-term (3-24 day) cultures. Comparative studies were made on levels of free amino acids in chicken blood plasma and serum, and in extracellular fluids from different regions of growth plate cartilage and from two types of muscle. Chondrocytes rapidly consumed glutamine from the medium, and to lesser extents, various other amino acids. In contrast, free ammonia, alanine, glycine, glutamate, proline, and aspartate were released into the medium. The utilization of certain amino acids changed, depending on the stage of culture. Initially glutamate was released into the medium but after confluency was consumed. Conversely, histidine, lysine, and phenylalanine were initially utilized but later were released into the medium. Levels of total free amino acids in extracellular fluids of cartilage and muscle were higher than those in plasma and serum, while in cartilage the levels increased progressively from the resting to the hypertrophic zones. In these sequential regions certain amino acids increased proportionally, whereas others decreased. These interrelationships generally correlated closely with metabolism of amino acids by the cultured chondrocytes. They indicate that significant differences in amino acid metabolism exist between tissue areas and are reflected in the extracellular fluid composition. Accordingly, adjustment of specific amino acids may optimize culture conditions, enabling more normal phenotypic expression in vitro.

Rhodamine 123, a laser dye, has been demonstrated to inhibit import of the precursor to pyridine dinucleotide transhydrogenase into mitochondria in rat liver cells. When rat hepatocytes were labeled with 35[S] methionine in the presence of 0.4 mM rhodamine 123, the precursor to transhydrogenase was found to have a half-life in the cytoplasm of 15 minutes as opposed to a half-life of 1-2 minutes when cells were radiolabeled in the absence of the dye. To clarify the mechanism of import inhibition, studies were initiated to assess the effect of rhodamine 123 on mitochondrial respiration. Upon addition of the dye to a mitochondrial suspension, respiration was initially enhanced, then inhibited. The inability of FCCP, a classical uncoupler, to enhance respiration during the inhibitory phase suggests that rhodamine 123 is primarily inhibiting respiration through the electron transport system rather than through the ATPase. These results suggest that rhodamine 123 may inhibit import of the transhydrogenase precursor into mitochondria by disrupting components in the mitochondrial membrane necessary for efficient import.

Release of mitochondrial calcium has been shown to occur concomitant with mineral ion loading of matrix vesicles at the onset of mineralization in epiphyseal growth plate cartilage. Matrix vesicles contain amorphous calcium phosphate (ACP), a mineral form that usually results from rapid precipitation at high initial levels of Ca2+ and/or inorganic P (Pi). Since the cytosol of growth plate chondrocytes has been found to contain high levels of Pi, rapid release of mitochondrial Ca2+ into the cytosol may cause local precipitation of calcium phosphate and thus be coupled with matrix vesicle formation. Studies were carried out to determine the kinetics and nature of mineral formation that occur when small amounts of Ca2+ are added under various conditions to a Pi buffer composed of electrolytes matched in concentrations and pH to that of the cytosol of epiphyseal chondrocytes. Depending on the manner in which Ca2+ was added, ACP, dicalcium phosphate dihydrate (DCPD), or apatite (HA) first formed. In the presence of ATP, ACP was the only solid phase detected, being stable for at least 24 h. However, in its absence, ACP rapidly transformed into DCPD. Increasing the pH of the reaction buffer from 6.9 to 7.5 increased the amount of ACP initially formed, but DCPD was consistently found upon ACP transformation. Yet at pH 8.0, ACP persisted for at least 24 h. The amount of precipitate formed was proportional to the level of added Ca2+; precipitates formed when as little as 1.0 mmole was added per liter of buffer. Our findings support the possibility that rapid release of mitochondrial Ca2+ may cause localized intracellular precipitation of ACP. Since nascent ACP is known to stimulate membrane fusion and blebbing of vesicles, these findings may explain the presence of ACP in matrix vesicles. The rapid conversion of ACP to DCPD in the absence of ATP under these conditions may also explain the reported occurrence of DCPD in samples of early mineralizing tissue.

A serum-free primary culture system for chicken growth plate chondrocytes has been developed which consistently undergoes mineral deposition. Upon attainment of confluency, the chondrocytes develop locally into multilayer cellular nodules leading to matrix calcification. Mineralization first occurs in matrix vesicles (MV) that are abundant in the extraterritorial matrix between the hypertrophic cells. Studies with 45Ca reveal that significant accumulation of Ca2+ occurs as early as day 12, continuing progressively throughout the culture period. By day 24, the nodules become densely calcified. Fourier transform infrared spectroscopy reveals the mineral to be similar to apatite, with features essentially identical to those of mineral formed by MV in vitro. The presence of ascorbate is critical to the culture system; in its absence, calcification is rarely observed. Ascorbate stimulates MV formation and synthesis of cellular protein, alkaline phosphatase, and especially types II and X collagens. In addition, there is strong evidence that the types II and X collagens are associated with MV. 1) Electron microscopy reveals MV embedded in a type II collagenous network; 2) Western blots of sodium dodecyl sulfate-polyacrylamide gel electrophoresis of MV using monospecific antibodies to types X and II collagen indicate that both collagens are present in specific MV fractions; 3) sucrose gradient purification of MV does not remove associated collagens; 4) graded salt extraction selectively releases type II collagen from MV; and 5) incubation of radiolabeled types II and X collagens with MV leads to their cosedimentation upon subsequent centrifugation. Taken together, the data suggest that coordinated synthesis of the collagens, alkaline phosphatase, MV formation, and Ca2+ accumulation by the cultures combine to induce mineral deposition in the multilayer nodules.

Matrix vesicles (MV) isolated from chicken growth plate by collagenase digestion and incubated in 45Ca-labelled synthetic cartilage lymph (SCL) rapidly induce mineral formation. 45Ca uptake occurs in three distinct stages: (1) an initial lag period of limited accumulation, (2) a period of rapid ion uptake and (3) an extended period of slower uptake. Treatment of MV with buffered aqueous 1,10-phenanthroline (OP), a metal ion chelator, eliminated the lag period, promoting immediate, enhanced Ca2+ uptake. Analysis of MV for trace metals showed them to contain relatively high concentrations of Zn (1.58 mumol/g MV) and lesser amounts of Cu (0.07 mumol/g MV). At least 30-40% of the Zn was readily extractable in isosmotic buffers. Addition of Zn to SCL at levels as low as 5 microM completely inhibited MV mineralization; addition of OP to Zn-inhibited MV restored their ability to mineralize. The findings suggest that Zn2+ ions act as an endogenous regulator of MV Ca2+ uptake and that the normal lag period results from a competition between Zn2+ and Ca2+ for high affinity Ca2+ binding sites in the MV membrane or within the MV lumen. Other metals tested included Cu2+, Pb2+ and Cd2+ which had little or no effect on MV mineralization, Mn2+, which had an intermediate effect, and Al3+, which was found to be almost as inhibitory as Zn2+. This finding may have implications for aluminum-associated osteomalacia.

Alkaline phosphatase (AP) activity in epiphyseal growth plate cartilage increases markedly during differentiation of the chondrocytes, and reaches high levels in the zone of hypertrophy where vascular penetration and provisional mineralization begin. A proteinaceous factor has been discovered in serum that stimulates the expression of AP in chicken growth plate chondrocytes when these cells are grown in serum-free media. Sera from a variety of vertebrate species (goat, fetal bovine, horse, human, and chicken) all contained detectable levels of the inducing activity. The chondrocyte AP-induction factor (CAP-IF) from fetal bovine serum was precipitated with ammonium sulfate between 33% and 50% saturation, and purified by dye-ligand affinity chromatography. The active fraction, which eluted from an Affi-Gel Blue column between 0.10 and 0.15 M NaCl, was further resolved on a QMA anion exchange column. The most active and almost homogeneous fraction contained primarily a 64.5 kDa protein; about 3 micrograms/ml medium induced 50% of the maximal level of AP induction. CAP-IF is stable to heat (100 degrees C for 3 min) and dithiothreitol (50 mM) treatment, and is only mildly inactivated by 2 h treatment with trypsin. CAP-IF caused no significant effect on cell division as measured by 3H-thymidine uptake. Time-course studies revealed that at least 18-24 h exposure of the chondrocytes to CAP-IF is required to produce major increases in AP activity. Longer exposure time generally further increases the response. Cycloheximide almost completely blocked the increase in AP activity, indicating that de novo protein synthesis is required for induction.

Matrix vesicles (MV) induce mineralization by compartmentalization of ion accumulation and crystal nucleation within membrane-enclosed extracellular microstructures. MV derive from cell surface microvilli by processes that cause selective enrichment of specific proteins, enzymes, lipids, and electrolytes. Incubated in synthetic cartilage lymph (SCL), MV accumulate Ca2+ and Pi, inducing mineral formation in a sequence of stages that can be altered by specific affectors. Rapid uptake of mineral ions by MV precedes formation of the first crystalline phase, octacalcium phosphate (OCP), which later converts to apatite (HAP). Early uptake of Ca2+ and Pi by MV is pH and protease sensitive, and is stimulated by o-phenanthroline (OP), a Zn2+ chelator. Recent studies reveal that a quantitatively major group of MV proteins bind to Ca2+ with high affinity in a lipid-dependent manner. These MV proteins appear to be involved in transport and accumulation of Ca2+ and Pi by MV, and may catalyze nucleation of the first mineral phase.

A serum-free medium for postembryonic chicken epiphyseal growth plate chondrocytes has been developed from 104 MCDB medium. To enable these fastidious cells to survive, grow, and express normal phenotype, a substantial increase over MCDB 104 in the level of many of the amino acids was required, as well as a change in the buffer system and the addition of SerXtend, a defined, serum-free product containing various growth factors, including fibroblast growth factor. Also required was the provision of cell attachment factors, either by coating culture surfaces with type II collagen, or better, by allowing the freshly released cells to recover for several hours in a medium supplemented with 10% fetal bovine serum before plating. This new serum-free medium, which we call USC-HC1, supports growth and replication, the retention of normal polygonal morphology, the expression of significant levels of cellular alkaline phosphatase activity, the production of sulfated proteoglycans, type II collagen, and the formation of alkaline phosphatase-rich matrix vesicles by the chondrocytes. The major advantage of USC-HC1, however, is that it will provide for the first time an opportunity to examine the effects of various defined growth and hormonal factors on the phenotypic expression and differentiation of growth plate chondrocytes, in the absence of the variable (stimulatory and inhibitory) factors present in fetal bovine serum.

Matrix vesicles (MV) can be readily isolated from culture media of chicken growth plate hypertrophic chondrocytes grown in primary culture. The chondrocytes maintain normal morphology and synthesize type II collagen throughout the culture period. The culture-derived MV are morphologically indistinguishable from MV seen in situ and are rich in alkaline phosphatase. Formation of alkaline phosphatase-rich MV is strongly influenced by the stage of culture: large numbers are released shortly after cell seeding; marked decline is seen during cell spreading and rapid cell division; notable resurgence in alkaline phosphatase-rich MV production occurs as the cells attain confluency. Increasing the initial chondrocyte seeding density proportionately increases MV production. Cells derived from the hypertrophic region are much more capable of forming alkaline phosphatase-rich MV than those from the proliferating zone, indicating that MV formation is dependent on cellular differentiation. MV released by the cultured chondrocytes were compared in protein and phospholipid composition and in their ability to accumulate mineral ions, with plasma membrane fractions and collagenase-released MV obtained from the same tissue. Electrophoretic patterns of proteins, and the phospholipid profiles, suggest that significant modification of the plasma membrane occurs during MV formation. The vesicles are capable of accumulating large amounts of mineral ions from a metastable synthetic cartilage lymph when supplied with alkaline phosphatase substrates. This culture system thus appears to be a useful model for isolating native MV and characterizing factors required for vesicle formation and mineralization.

Prostaglandin synthetase activity in high-speed particulate fractions of chick epiphyseal cartilage has been characterized with respect to cofactor requirements, pH optimum, buffer-ion effects, types of prostaglandins formed, and the distribution of prostaglandin synthetase activity in zones of the epiphyseal plate. Direct homogenization of cartilage was found to be more efficacious than releasing chondrocytes by enzymatic digestion for preparation of prostaglandin synthetase, a homogenization time of 4 min yielding maximal activity. The optimal incubation medium contained 50 mM Tris buffer (pH 7.5), 2.5 mM epinephrine, 1 micronM hemoglobin, 3.25 mM glutathione, 200 microgram/ml enzyme protein, and 5 micronM substrate. Glutathione was effective only if present during homogenization. Rates of PGE2 biosynthesis were linear up to 15 min and then rapidly declined, indicative of self-deactivation. The low levels of PGF2alpha formed, and their decrease after 20 min incubation, suggests the possible presence of degradative enzymes. Prostaglandin synthetase was inhibited by aspirin, indomethacin, and vitamin E, but not vitamin K1. Cation concentrations in the physiological range had only modest effects on prostaglandin biosynthesis, and then only if present during tissue homogenization. In the presence of phosphate buffer, Ca2+ was somewhat inhibitory. Since in the absence of phosphate Ca2+ had no deleterious effect, it is probably that the inhibitory effect was caused by precipitation of calcium phosphate. Hypertrophic and calcified cartilage exhibited significantly higher prostaglandin synthetase activity than the proliferating and maturing zones. The increased synthesis of prostaglandins in the low layers of the growth plate may indicate a role of these factors in chondrocyte differentiation and/or calcification.

Growth plate chondrocytes in primary culture release alkaline phosphatase (AP)-rich matrix vesicles (MV) into the culture medium. While these are thought to derive from the plasma membrane by a membrane fusion-dependent process, the mechanism is not fully understood. Recently, a cytosolic protein, synexin, has been shown to promote membrane fusion in a number of systems, and thus may be involved in MV formation. Since the action of synexin is selectively inhibited by trifluoperazine (TFP) and other phenothiazines, we examined the effects of these drugs, and imipramine, on cellular AP production and formation of AP-containing MV by cultured chondrocytes. In addition, we studied the effect on cell division, protein biosynthesis, and incorporation of palmitate into cellular and MV lipids. All of the drugs reduced cellular AP in a concentration-dependent manner; however, at the higher levels, chlorpromazine (CPZ) and TFP caused a transient sharp rise in MV AP activity. At IC50 levels, the drugs were more inhibitory to cellular AP than to MV AP, appearing to enhance significantly the transfer of available cellular AP into MV. In contrast, the drugs stimulated incorporation of [3H]palmitate into cellular lipids, but either had no effect on, or actually inhibited, incorporation of the fatty acid into MV. At these levels, the drugs had little effect on cell division and protein biosynthesis. The inhibitory effect of IC50 levels of CPZ on palmitate incorporation into MV appears to have resulted from impairment of vesicle formation per se, since at these levels the drug stimulated incorporation of the fatty acid into the cells. The transient stimulatory effect of higher levels of CPZ on MV AP levels, and the enhanced transfer of AP into MV by the drugs generally, may result from the effect of the drugs on membrane structure. Since TFP was not inhibitory to MV formation, it is doubtful that synexin was directly involved.

The effects of three vitamin D3 metabolites, 25-hydroxyvitamin D3 (25-(OH)D3), 1 alpha,25-dihydroxyvitamin D3 (1 alpha, 25-(OH)2D3), and 24R,25-dihydroxyvitamin D3 (24R,25-(OH)2D3) on the activity of alkaline phosphatase (AP), a key enzyme involved in biomineralization, have been studied in primary cultures of chicken epiphyseal growth plate chondrocytes. Dosages of 1 alpha, 25-(OH)2D3 (10(-12) to 10(-7) M) caused a progressive, dosage- and time-dependent decrease in cellular AP levels, IC50 occurring at approximately 10(-12) M. In contrast, 24R,25-(OH)2D3 at 10(-13) to 10(-10) M stimulated cellular AP activity, half-maximal stimulation occurring at about 10(-13) M. At higher levels (10(-10) to 10(-7) M), 24R,25-(OH)2D3 caused progressive reduction in AP activity. Maximal effects of 24R,25-(OH)2D3 were evident 48 h after administration of the metabolite. 25-(OH)D3 initially (24 h) caused a weak, dosage-dependent decrease in cellular AP activity, but after 48-72 h, low levels (10(-13) to 10(-11) M) caused a dosage-dependent increase in AP activity. Higher levels of 25-(OH)D (greater than 10(-10) M) were clearly inhibitory to AP. These findings reveal that the AP activity of growth plate chondrocytes is exquisitely sensitive to both 1 alpha,25- and 24R,25-(OH)2D3 but the response to each is in opposite directions. The paradoxical response of the cells to 25-(OH)D3 can be explained if the metabolite is slowly metabolized by a 24-hydroxylase to 24R,25-(OH)2D3 leading to stimulation of cellular AP.(ABSTRACT TRUNCATED AT 250 WORDS)

Matrix vesicles, extracellular microstructures known to be involved in endochondral calcification, are rich in alkaline phosphatase and have been shown to contain actin. The mechanism of matrix vesicle formation in chondrocytes is not well understood. Chondrocytes from the epiphyseal growth plate, when grown in primary culture, elaborate alkaline phosphatase-rich vesicles. We examined the distribution of the cytoskeletal proteins actin, myosin, tubulin, and vinculin at various time-points during culture using indirect immunofluorescent labeling. Concomitantly, the production of alkaline phosphatase-containing matrix vesicles was also followed. Cell morphology changed noticeably at two distinct stages during the 22-day culture period: Immediately after release from the growth plate the cells were rounded, but after 4 days of culture they began to spread out and acquire irregular shapes with distinct filopodia. By 13 days, as the cells attained confluency, they reacquired a rounded, polygonal appearance. At all time-points, tubulin was seen as a dense network of microtubules radiating from the perinuclear region throughout the cytoplasm toward the cell periphery. Initially actin was seen in filamentous form, but displayed a punctate distribution focused at contact points during the cell-spreading stage of culture. After confluency, actin was concentrated at cell-cell junctions. Initially, vinculin was diffusely distributed, but became focused in multiple adhesion plaques and at the termini of filopodia during the cell-spreading stage of culture. Following confluency vinculin became concentrated at cell-cell junctions. Myosin was observed at all time-points in small, intensely localized focal points in the cytoplasmic region of the cells and was consistently absent from the nuclear and peripheral regions. The amount of myosin in the cells increased steadily with time in culture. Elaboration of alkaline phosphatase-rich vesicles, which corresponded closely with the rounded morphology of early and late stages of culture, may be correlated with contact inhibition.

Fourier transform infrared spectroscopy (FTIR) was used to characterize the organic and mineral phases present during the induction of mineral formation by collagenase-released matrix vesicles (CRMV) during incubation in a synthetic cartilage lymph in vitro. CRMV mineralization, which occurs in the absence of alkaline phosphatase organic phosphate substrates, is characterized by an initial short lag period of limited Ca2+ accumulation, followed by a period of rapid Ca2+ uptake, and finally, by a plateau period during which Ca2+ accumulation continued at a slower rate. FTIR spectra taken at timed intervals during the induction of mineralization revealed the presence of absorptions characteristic of protein, phospholipid, and mineral components in the CRMV. These became progressively more intense with time. To reveal underlying changes occurring during the successive stages of Ca2+ accumulation, FTIR spectra of nascent (or demineralized) CRMV were computer-subtracted from subsequent spectra, nulling on the C-H stretch modes characteristic of the lipid acyl chains. These difference spectra showed little change during early Ca2+ loading, revealing that mineral ions initially accumulated in a form similar to that present in nascent matrix vesicles (MV). During the period of rapid Ca2+ uptake prior to appearance of crystalline mineral, difference spectra revealed subtle changes in the carbonyl and amide nitrogen stretch modes indicative of protein conformational changes. The first definable mineral phase appeared late in the rapid Ca2+ uptake period and was a distinct, crystalline octacalcium phosphate (OCP)-like phase. With time, the OCP-like precursor became more apatitic in character. There was no evidence that any amorphous calcium phosphate phase formed during the MV mineralization sequence. The mature MV mineral phase closely resembled hydroxyapatite formed via an OCP precursor and was similar to other biological apatites that show a substantial incorporation of carbonate.

We used embryonic skeletal cartilage known to have high levels of alkaline phosphatase activity to determine whether growing cartilage has phosphotyrosine phosphatase activity and phosphotyrosinyl histone phosphatase activity at physiologic pH. Embryonic chick pelvic cartilage and fetal pig scapular growth-plate cartilage were assayed using phosphotyrosine as substrate at pH 7.5 and the amount of tyrosine generated measured. Both cartilage models had Km for phosphotyrosine between 6 to 24 mus mol/L. Phosphotyrosine phosphatase activity correlated with alkaline phosphatase activity as assessed by (1) distribution of histologic staining for alkaline phosphatase within the cartilages, (2) hormonal stimulation of cartilage alkaline phosphatase activity in vitro, (3) comparison of alkaline phosphatase and phosphotyrosine phosphatase activities in the presence of known inhibitors (vanadate, levamisole, homoarginine, and zinc), and (4) assaying chick epiphyseal cartilage alkaline phosphatase purified to homogeneity for phosphotyrosine phosphatase activity. Areas of cartilage with elevated alkaline phosphatase activity also had raised phosphotyrosine phosphatase activity. Triiodothyronine, a known stimulator of cartilage alkaline phosphatase, increased chick cartilage alkaline phosphatase activity 88% and phosphotyrosine phosphatase activity 106%, and stimulated porcine growth-plate cartilage alkaline phosphatase activity 91% and phosphotyrosine phosphatase activity 145% after 3 days of in vitro incubation. Each of the inhibitors block alkaline phosphatase and phosphotyrosine phosphatase activities. The purified alkaline phosphatase had a Km for phosphotyrosine of 18 mus mol/L and Vmax of 5700 nmol tyrosine/mg protein/h, which is well over 1000-fold higher than the phosphotyrosine phosphatase activity found in the above preparations of pelvic and scapular cartilage.(ABSTRACT TRUNCATED AT 250 WORDS)

The roles of alkaline phosphatase and labile internal mineral in matrix vesicle-mediated mineralization have been studied by selectively releasing the enzyme from a wide variety of matrix vesicle preparations using treatment with a bacterial phosphatidylinositol-specific phospholipase C and by demineralization of the vesicles using isosmotic pH 6 buffer. Following depletion of 50-90% of the alkaline phosphatase activity or treatment with citrate buffer, the vesicles were tested for their ability to accumulate 45Ca2+ and 32Pi from a synthetic cartilage lymph. Removal of alkaline phosphatase by phospholipase C treatment caused two principal effects, depending on the matrix vesicle preparation. In rapidly mineralizing vesicle fractions which did not require organic phosphate esters (Po) to accumulate mineral ions, release of alkaline phosphatase had only a minor effect. In slowly mineralizing vesicles preparations or those dependent on Po substrates for mineral ion uptake, release of alkaline phosphatase caused significant loss of mineralizing activity. The activity of rapidly calcifying vesicles was shown to be dependent on the presence of labile internal mineral, as demonstrated by major loss in activity when the vesicles were decalcified by various treatments. Ion uptake by demineralized vesicles or those fractionated on sucrose step gradients required Po and was significantly decreased by alkaline phosphatase depletion. Uptake of Pi, however, was not coupled with hydrolysis of the Po substrate. These findings argue against a direct role for alkaline phosphatase as a porter in matrix vesicle Pi uptake, contrary to previous postulates. The results emphasize the importance of internal labile mineral in rapid uptake of mineral ions by matrix vesicles.

The effect of varying the amino acid concentrations of the culture medium on matrix vesicle formation was studied in primary cultures of chicken epiphyseal growth plate chondrocytes grown in Dulbecco's modified Eagle's medium (DME) supplemented with 10% fetal bovine serum (FBS). Decreasing the levels of free amino acids in the culture medium to levels of one-half, one quarter, and one eighth of the values normally present in DME caused a progressive decline in matrix vesicle (MV) formation. Increasing the level in the culture medium of those amino acids that are enriched in extracellular fluid (ECF) of growth plate cartilage significantly increased formation of matrix vesicles (MV), as assayed by the alkaline phosphatase (AP) activities present in high-speed sediments from spent culture media. However, adjusting the levels of all amino acids to match those of the ECF produced the greatest stimulation of MV formation. Of the amino acids that are notably enriched in ECF, glutamate (GLU), alanine (ALA), serine (SER), asparagine (ASN), and taurine (TAU) individually enhanced MV production, whereas proline (PRO), glycine (GLY), and aspartate (ASP) had essentially no effect. The simple combination of ECF levels of ALA and GLU resulted in a stimulation of MV formation equal to that observed when the eight aforementioned amino acids were elevated to ECF levels. Other combinations of ASP and GLY, or of TAU, SER, and ASN showed some stimulation, but at a lower level. Increasing the amino acid concentrations, alone or in combination, also increased the levels of cellular AP, and to a lesser extent cellular protein. While increases in cellular AP were generally correlated with increased formation of AP-rich MV, this was not uniformly true. These results indicate that in addition to hormones and growth factors, nutritional factors such as the levels of amino acids are also critical for normal phenotypic expression, growth, and matrix formation by epiphyseal chondrocytes.

The effect of synthetic human parathyroid hormone (hPTH) on the formation of matrix vesicles (MV), and on the rate of cell division, production of cellular alkaline phosphatase (AP) and protein by primary cultures of chicken epiphyseal growth plate hypertrophic chondrocytes was investigated. Addition to serum-containing or serum-free media of physiological levels of hPTH, in a range from 0.1 to 10 nM, caused a progressive decrease in the formation of AP-rich MV. However, studies on incorporation of [3H]choline into MV indicate that MV formation per se was not significantly decreased. hPTH was found to markedly decrease the expression of cellular AP, accompanied by an increase in cell division [( 3H]thymidine incorporation) and protein synthesis. Since these effects of hPTH were augmented by 3-isobutyl-1-methylxanthine (IBMX), a phosphodiesterase inhibitor, and mimicked by the cAMP analogue N6,O2'-dibutyryl-adenosine 3',5'-cyclic-monophosphate (DBcAMP), the findings clearly indicate that hPTH was acting through the classic cAMP-mediated mechanism. Inasmuch as elevation of AP in growth plate chondrocytes coincides with MV formation, maturation and hypertrophy of the cells, and induction of mineralization, the stimulation of cell division and suppression of cellular AP indicates that hPTH would cause the cells to revert to a less differentiated state. Thus, elevation in PTH, which results from lowered circulating levels of Ca2+, should inhibit mineral deposition in the growth plate. This may be a physiological protective mechanism to prevent a further drain on serum Ca2+.

Uptake of mineral ions by isolated matrix vesicles (MV) incubated in synthetic cartilage lymph follows a consistent pattern. After an initial lag period, MV rapidly accumulate large amounts of Ca2+ and Pi before the appearance of crystalline mineral. The ability of MV to accumulate Ca2+ is readily destroyed by proteases, indicating that proteins are important in Ca2+ accumulation. Since MV contain significant amounts of phosphatidylserine (PS), an acidic phospholipid with affinity for Ca2+, it seemed probable that this lipid might also contribute to Ca2+ binding. The development of methods for reproducible isolation of pure active MV enabled us to search for factors responsible for the rapid accumulation of Ca2+. Reported here are studies which reveal that a set of intensely staining MV proteins, extractable with EGTA, selectively bind to Ca2+, but only in the presence of acidic phospholipids. These 30-36-kDa proteins form readily sedimentable insoluble ternary complexes of protein, Ca2+, and lipid in the presence of low levels of Ca2+. With liposomes composed of PS, alone or in combination with phosphatidylethanolamine, submicromolar levels of Ca2+ or certain other divalent cations, but not Mg2+, are sufficient to form the complexes. The physical and chemical properties of these MV proteins appear to be like those of the calpactin family of membrane-associated proteins. In fact, these MV proteins were found to cross-react with antibodies to calpactin II. Thus, calpactins appear to be important protein constituents of avian growth plate MV. This finding helps explain the enrichment in PS previously noted in MV and may also point to the mechanism by which MV rapidly accumulate Ca2+.

The association of lipids with biological mineral formation was first indicated by histochemical observations. This was later confirmed by biochemical studies which revealed that calcium was bound to acidic lipids at sites of early mineral deposition. The morphological locus of initial mineral formation has now been shown to be membrane-enclosed vesicles in the extracellular matrix. Recent studies on isolated vesicles have documented that they are rich in lipids very similar to those previously found to be associated with newly forming mineral. The vesicle lipids are enriched in cholesterol, free fatty acids, sphingomyelin, glycolipids, lysophospholipids, and phosphatidylserine, and depleted in phosphatidylcholine and -ethanolamine. This composition is like that of the plasma membrane of cells and supports electron microscopic evidence that the vesicles arise by budding from the chondrocyte surface membrane. Metabolic studies indicate that formation of the vesicles occurs relatively rapidly (i.e., in less than 6 hours) by biochemical processes involving both lipid synthesis and degradation. In vitro studies suggest that the vesicle lipids may be involved in the initial binding and phase separation of calcium phosphate, as well as its conversion to crystalline apatite; however, final elucidation of the mechanism of action remains to be attained.

The kinetics of epiphyseal cartilage membrane-bound alkaline phosphatase activity was studied at physiological pH using p-nitrophenylphosphate and pyrophosphate (PPi) as substrates. The effect of three general types of alkaline phosphatase inhibitors was studied on both purified and membrane-bound forms of the enzyme: 1) uncompetitive inhibitors (L-tetramisole and theophylline), 2) competitive inhibitors (phosphate, arsenate, and vanadate), and 3) metal ions (Mg2+, Ca2+, and Zn2+). These studies were aimed at elucidating the physiological role of epiphyseal cartilage membrane alkaline phosphatase. Hydrolytic activity of enzyme at pH 7.5 toward both PPi and p-nitrophenylphosphate was only one-tenth and one-hundredth, respectively, of that observed at optimal pH. Arsenate (Ki - 1.0-2.7 microM) and vanadate (Ki = 0.7-1.3 microM) were powerful inhibitors of alkaline phosphatase; phosphate (Ki = 20-50 microM) was inhibitory at levels two orders of magnitude below the concentration in cartilage extracellular fluid. Neither Zn2+, Ca2+, or Mg2+ was inhibitory toward p-nitrophenylphosphate hydrolysis, whereas all three competitively inhibited PPi hydrolysis. The data suggest that formation of poorly hydrolyzable Ca.PPi, Mg2.PPi, and Zn2.PPi complexes was responsible. Inhibition of PPi hydrolysis by Ca2+ occurred at levels within the physiological range. The close similarity in inhibition of both the purified and membrane-bound forms of alkaline phosphatase at pH 7.5 indicates that interaction with the membrane does not significantly alter conformation at the active site. The data obtained suggest that under physiological conditions, cartilage membrane alkaline phosphatase would be essentially inactive as a phosphohydrolase due to its intrinsically weak activity at pH 7.5 and the strong inhibitory effect of physiological levels of phosphate and Ca2+.