- Email: [email protected]
Biochemical characterization of matrix vesicles from bone and cartilage
Metab.
Bone Dis. & Rel. Res. 5, 93-99
0221-8747183 $3.00+ .OO Copyright 0 1984 Pergamon Press Ltd.
(1983)
Printed in the USA. All rights reserved.
Biochemical Characterization of Matrix Vesicles from Bone and Cartilage A. MUHLRAD’,
A. SETTON’,
J. SELA2, I. BAB2, and D. DEUTSCH’
‘Department of Oral Biology, and ‘D/v/sion of Oral Pathology, The Hebrew Umvemty-Hadassah
Faculty of Dental Med/cine, Jerusalem, Israel.
Address for correspondence and reprints:: Prof. A. Muhlrad, Department of Oral Biology, Hebrew University-Hadassah Medicine, P0.B. li72, Jerusalem 91010, Israel.
Abstract
Faculty of Dental
1976; Muhlrad et al., 1978; Ali, 1977). These organelles are believed to play a major role in the primary mineralization of these tissues. Most of the original studies were ultrastructural and qualitative in nature. To obtain quantitative information, chemical data were also required. This information was, however, difficult to obtain because of inadequate methods for isolation and purification of matrix vesicles. In recent years such methods have been developed (Ali et al., 1970; Majeska and Wuthier, 1975; Fortunaetal., 1978; Wuthier et al., 1978; Kahn et al., 1978; Sela et al., 1978; Deutsch et al., 1981, a,b), and the biochemical results obtained from purified matrix vesicles in combination with histologic findings have provided a more detailed picture of the role of matrix vesicles in primary mineralization. These studies revealed that alkaline phosphatase activity was 15-20 times higher in matrix vesicles than in cells (Ali et al., 1970; Majeska and Wuthier, 1975; Deutsch et al., 1981, a, b). Since this phenomenon is unique to mineralizing tissue, alkaline phosphatase has been used as a marker enzyme for matrix vesicles. In addition to alkaline phosphatase other phosphatases have been found to be associated with matrix vesicles. These include inorganic pyrophosphatase and ATPase. These enzymes are believed to hydrolize inorganic pyrophosphate and ATP, which would otherwise act to inhibit calcification (Fleisch and Bisaz, 1962; Alcock, 1972; Betts et al., 1975). Acid phosphatase has also been implicated as a matrix vesicle-associated enzyme. There is some controversy, however, regarding this association (Ali et al., 1970; Thyberg and Friberg, 1972). Most of these histologic and biochemical studies were performed on cartilage (Anderson, 1967, 1969; Bonucci, 1967, 1970; Vaananen and Korhonen, 1979; Majeska and Wuthier, 1975; Fortuna et al., 1978). In recent years similar studies have also been carried out on bone (Muhlrad et al., 1978; Sela et al., 1978; Bab et al., 1979; Deutsch et al., 1981a). These results indicate that the basic process of mineralization via matrix vesicles is similar in both tissues, but some differences may exist between them. However, a thorough comparison of matrix vesicles from various mineralizing tissues has not been carried out.
Extracellular matrix vesicles from bone and epiphyseal cartilage of femur and tibia of rats were isolated by collagenase digestion (crude vesicles) and further purified by sucrose gradient centrifugation. Fractions containing cells and membranes were also isolated from the two tissues. The alkaline and acid phosphatase and ATPase activities, as well as protein content of all fractions including crude and purified matrix vesicles, were assayed. The crude vesicles from both tissues demonstrated a high alkaline phosphatase specific activity (5-20 times higher than in the cell fraction). The total enzyme activities and protein content were significantly higher in all fractions from cartilage than those from bone. A major peak of alkaline phosphatase activity and protein content was obtained following the sucrose gradient centrifugation. The position of this peak was similar for both tissues. The specific activity of alkaline phosphatase of purified matrix vesicles was significantly higher in bone than in cartilage. The phosphatase activities from cartilage and bone showed a similar pH dependence and a similar response to metal ions. Of the metal ions tested (Na+, Mg*+, Zn*+, and Ca*+) only Zn*+ (at 5 mM concentration) inhibited significantly the alkaline phosphatase activity of purified matrix vesicles. The electrophoretic profile of purified matrix vesicles showed eight major protein bands common for both tissues. In addition, cartilage vesicles appeared to possess two peptides not found in bone.
Introduction The presence of extracellular matrix vesicles in cartilage in areas undergoing mineralization was reported by Anderson (1967) and Bonucci (1967). Subsequently matrix vesicles have been identified in bone (Bernard and Pease, 1969; Schenk et al., 1970), dentin (Bernard, 1972; Eisenman and Glick, 1972; Siska and Provenza, 1972), and a number of other tissues undergoing pathologic mineralization (Kim, 93
94
A. Muhlrad et al.: Biochemical characterization
The aim of this study was to isolate, characterize, and compare matrix vesicles from cartilage and bone with regard to the enzymatic activity (in particular, alkaline phosphatase), electrophoretic mobility of the vesicular proteins, and density profile on a sucrose gradient, using methods recently developed (Deutsch et al., 1981 a).
Materials and Methods Preparation of matrix vesicles Vesicles were prepared from bone and cartilage obtained from rats weightng 150-160 g of the Hebrew Unrversity (Sabra) strain. Tubular bones, femur, and tibia were used. The cartilage tissue was dissected from eplphyseal growth plates of the same bones. The procedures used for isolation and purification of matnx vesicles from bone and cartilage were described previously (Deutsch et al.; 1981a). Briefly, bone and cartilage were cleaned from soft tissue and blood, cut into small pieces, and digested with crude collagenase. The digest was decanted and subjected to differentral centrifugation. Three fractions-cells, membranes, and crude matrix vesicleswere obtained. For further purification the crude matrix vesicles were subjected to a noncontinuous sucrose-density gradient centrifugatron and fractionated. Each fraction was analyzed for various enzyme activities and protein content.
Polyacrylamide
gel electrophoresis
Both crude and purified matrix vesicle fractions were examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) as follows. After sucrose density gradient centrifugation, the matrrx vesrcle fractions containing the peak alkaline phosphatase activity were pooled, diluted five times with Gey’s solution, and centrifuged at 170,000 g for 60 min. The pellets were then resuspended In 100 ~1of sample buffer containing 322 mM Trrs-HCI pH 6.8, 32% glycerol, 7% SDS, and 3.2% /3-mercapto-ethanol and incubated at 100°C. The sample was then placed on a 7-l 8% polyacrylamrdegradient SDS slab gel according to the method of Long et al. (1977).
Enzymatic assays. Alkaline phosphatase
activity was determined In 1 ml of 100 mMglycine-NaOH pH 10.5, 1 mM MgCI,, and 0.1 mM ZnCI, containing 6 mM Na-p-nitrophenyl phosphate, as described by Sela et al. (1978). Acidphosphafase activity was measured in samples Incubated in 1 ml of 10 mM sodium acetate pH 5.0, containing 18 mM Nap-nitrophenyl phosphate for 30 min at 37*C. The reaction was stopped by the addition of 2.0 ml 0.5N NaOH and read immediately at 405 nm. Mgz+-ATfase activity was determined in the presence of 1.6 mM ATP in 1 ml of 2 mM MgCI, 10 mM imidazole pH 7.0, and Incubated at 37QCfor30 min. P,liberatedduring the reaction wasmeasured by the method of Fiske and Subbarow (1.925). Enzyme activity was expressed as either nmol P, liberated per mg protein per min (specific activrty) or nmol P, per g tissue (total activity) in all cases. /on studies. The effect of several catrons on alkaline phosphatase actlvrty was studied in crude and purified matrix vesicle Preparatlons from bone and cartilage. Enzyme activrty was assayed In a 100 mM glycine-NaOH buffer pH 10.5, with the addition of various concentrations of the following Ions: O.O5M-2M Na+ (as NaCI), 0.05 mM-10 mM Mg2+ d(as MgCI,), 0.01 mM-5 mM Zn*+ (as zincacetate), 0.05 mM-2 mM Ca” (as CaCI,).
ph’profile. The pH dependence of phosphatase activity was measured between pH 3.5 and 11 .O at 37OC according to the procedure used for assaying acid phosphatase activity. The following buffers were used at 100 mM concentrations, with their specific ranges: sodium acetate acetic acid pH 3.5-5.5, maleate-NaOH pH 5.5-6.5, imidazole-HCI pH 6.5-7.5, Tris-HCI pH 7.5-8.5, glycine-NaOH pH 8.5-l 1 .O. The pH of the buffers was also measured at 37’C.
Protein determination. ford (1976).
This was performed
according
to Brad-
of matrix vesicles
Results Crude matrix vesicles from bone and cartilage A comparison of matrix vesicles from bone and cartilage, obtained after collagenase digestion, is shown in Tables I and II. Significantly more protein per g tissue was found in the vesicles, membranes, and cell fractions obtained from cartilage. In addition the total activity per g tissue of enzymes assayed (alkaline phosphatase, acid phosphatase, and ATPase) was significantly higher in cartilage fractions than in bone. This difference between the two tissues is particularly apparent in the cell fraction, where the levels of acid and alkaline phosphatase total activity are 12 times higher in cartilage than in bone. The specific activity of alkaline phosphatase in vesicle fractions obtained from bone was higher than that in cartilage, while in the cell fraction the specific activity of this enzyme was lower (both differences were significant). The distribution of the total enzyme activity and protein content among the various fractions obtained from bone and cartilage is presented in Table I. The distribution of protein content, acid phosphatase, and ATPase activity is almost identical for bone and cartilage. However, differences were observed in the distribution of alkaline phosphatase. In bone, 55% of the alkaline phosphatase activity was found in the matrix vesicle fraction, whereas in cartilage the vesicle fraction contained only 35% of the total alkaline phosphatase activity. Purified matrix vesicles from bone and cartilage. The distribution of matrix vesicles from both tissues following sucrose density gradient centrifugation is presented in Figure 1. The purification factor, which is based on the increase in the specific activity of alkaline phosphatase before and after sucrose gradient centrifugation (Table Ill), is similar for both tissues. The protein content and alkaline phosphatase activity from the preparations of both tissues resolved into a major peak between 20 and 35% sucrose layers. There seemed to be a slight shift in the alkaline phosphatase activity of cartilage toward the heavier density. Some alkaline phosphatase activity and protein were also found in the heavy density fractions. The purification achieved during the step of sucrose gradient centrifugation is similar for both tissues. The specific activity of alkaline phosphatase of purified matrix vesicles obtained from bone is significantly higher than that of cartilage vesicles. Effects of ions and pH on enzymatic activity of purified matrix vesicles In the range of ion concentration employed in thisexperiment, no significant effects were obtained for Nat, Mg2+, and Ca’+. However, Zn2+ in a high concentration inhibited alkaline phosphatase activity (Fig. 2). A 70% inhibition of this enzyme was observed at 5 mMZn2+ concentration. No differences were found in the effect of these ions on the behavior of alkaline phosphatase activity from both tissues. The pH dependence of the phosphatase activity from purified matrix vesicles from bone and cartilage was measured between pH 3.5 and 11.0 and is presented in Figure 3. The optimum phosphatase activity in the acid region was found to be at pH 4.0 for both bone and cartilage. Matrix vesicles from both tissues showed similar curves in this region, and both preparations demonstrated an acid phosphatase specific activity of less than 1% of the highest alkaline phosphatase specific activity measured at pH_ 11 .O. However, the maximal acid phosphatase activity, relative to alkaline phosphatase in
A. Muhlrad et al.: Biochemical characterization
95
of matrix vesicles
Table I. Comparison of enzyme activity and protein content of matrix vesicles from tubular bone and cartilage. Protein (mglg tissue)
Alkaline phosphatase (act./g tissue) (spec. act.)
Acid phosphatase (spec. act.) (act/g tissue)
ATPase (spec. act.) (act./g tissue)
Tubular bone Crude matrix
1021 + 63
0.06 & ,007
Vesicles
(11%
f
1 O/o)
Cells
0.33 (63%
f f
.03 7%)
z? 3%) 218 f 25 (12% f 1%)
Membranes
0.13 f (25% f
.02 5%)
628 (34%
f 44 + 2%)
.36 (14% 1.55 (58% .74 (28%
.06 2%) ,l 4%) .09 4%)
3986 (35% 2861 (25% 4542 (40%
I! + -t + + *
14 f 2
17766 + 1988
(55%
720 5126
+ 156 + 1890
(8% 116 (70% 35 (21 s/o
f + f + f
1 c/o) 28 17%) 6 4%)
59 (3% 1312 (74% 390 (22%
f If: f + f +
11 1o/o) 131 7%) 48 3%)
201&52 335 f 52 278 + 34
72 f 22 (25% 122 (42% 93 (32%
f f & f f
8%) 19 7%) 16 6%)
275 (25% 487 (44% 341 (31%
+ c f f f -e
113 10%) 139 13%) 92 8%)
1136+419 369 f 94 676&92
Carttlage Crude matrix Vesicles Cells Membranes
& + f + f f
589 5%) 552 5%) 479 40/o)
11765 f
1497
1820 f 319 6293 f 774
165? 14 856284 543 + 73
736*419 297 f 77 423 f 60
“Values in parenthesis represent the distnbutron of the total acttvtty among the varrous fractions in percent Results are expressed as mean ? standard error.
cartilage vesicles, was 25% higher than in their bone counterpart. In the alkaline region the optimum pH was not determined. However, the curves for both tissues were still rising at pH 11 .O. Protein profile of purified matrix vesicles The electrophoretogram of purified and crude matrix vesicles from bone and cartilage is presented in Figure 4. Major peptide bands of the following respective approximate molecular weights were observed for both long bone and cartilage purified and crude fractions: 150,000, 81,000, 71,000, 43,000, 38,000, 34,000, 32,000, and 22,000. In addition, matrix vesicles from cartilage appeared to possess two peptides not found in bone. These were observed at molecular weights of 58,000 and 42,000.
Discussion The results of this study revealed that the amount of protein and total enzyme activities in the matrix vesicle preparations obtained from cartilage were much higher than in bone. This implies that cartilage contains more vesicles per g tissue than bone. This assumption is supported by recent electron microscopic results that show that more matrix vesicles are present in cartilage than in bone per g tissue. This difference can be explained by the different patterns of mineralization of the two tissues. Generally, in bone the major part of the tissue is already mineralized and thus does not contain intact matrix Table
vesicles. The result is that only a relatively small proportion of bone is undergoing active mineralization involving matrix vesicles. In cartilage, on the other hand, as mineralization occurs, the tissue is immediately resorbed and replaced by bone, the net result being that a major part of cartilage is found in the active mineralizing state involving matrix vesicles, This difference in mineralization pattern could provide an explanation to the larger amounts of cellular protein per weight tissue found in cartilage relative to bone. The distribution pattern of matrix vesicles, following purification by sucrose density gradient centrifugation, was almost identical for the two mineralizing tissues. Based on alkaline phosphatase activity, the matrix vesicle fraction of both tissues resolved Into a major peak. The position of this major peak of alkaline phosphatase activity in both tissues seems to correspond to similar peaks obtained by Kahn et al. (1978) for fracture callus cartilage of rabbit, Wuthier et al. (1978) for chicken epiphyseal cartilage, and Deutsch et al., (1981a) for rat alveolar bone. Alkaline phosphatase activity and protein were also found in the heavier density sucrose layers. Kahn et al. (1978) Wuthler et al. (1978) Deutsch et al. (1981a) and Hirschman et al. (1982) all reported the existence of a minor matrix vestcle peak in the heavier sucrose density layers. The occurrence of matrix vesicles in both light and heavy density layers might reflect matrix vesicles at different stages of mineralization. The effect of pH and various metal cations on phosphatase activity from purified matrix vesicles of both tissues was nearly identical. A number of findings in the present study were
II. Stattstical comparison of enzyme activity and protein content of fractions obtained from tubular bone and eptphyseal cartilage. Crude matrtx vesicles
Cells
Membranes p < 0.01
Alkaline phosphatase
Total activtty Specific activity
p < 0.01
p < 0.01
p < 0.05
p < 0.01
NS
Acid phosphatase
Total activity Specific actwity
p < 0.02 p < 0.05
p < 0.01 p < 0.01
p < 0.01 p < 0.05
ATPase
Total actrvity Specific activity
p < 0.05 NS
p < 0.05 NS
p < 0.05 NS
Total
p < 0.01
p < 0.01
p < 0.01
Protein
Values obtained In the same expenments NS, nonstgnifrcant.
for bone and cartilage fractrons were paired and analyzed by Student’s paired t-test. Number of experiments, 5.
96
A. Muhlrad et al.: Biochemical characterization of matrix vesicles
=-,Lr,.-
I
mm
.
-----F_---__O__O-
so
. UY
Fig. 2. Effects of various ions on alkaline phosphatase
Fig. 1. Elution profiles of alkaline phosphatase (top) and protein (bottom) for matrix vesicles from cartilage *--+ and bone ??-e following sucrose density gradient centrifugation. Top. 100% activity = 162 and 671 nmol P/g tissue for bone and cartilage, respectively. Bottom. 100% protein = 0.008 and 0.041 mg protein/g tissue for bone and cartilage, respectively. found to differ, however, from those reported in the literature. Fortuna et al. (1980) found that above pH 10.6, alkaline phosphatase isolated from matrix vesicles of fetal, bovine epiphyseal cartilage was rapidly and irreversibly denatured, while it was our experience that the response curves for crude vesicles were still rising in the alkaline region at pH 11 .O. Gottlieb and Sussman (1968) showed that alkaline phosphatase purified from human placenta was fully active even at pH 11.5. Cyboron and Wuthier (1981), working with alkaline phosphatase purified from matrix vesicles, found the pH optimum at 10.5 and did not report any denaturation at alkaline pH up to pH 11 .O. The reason for the slight discrepancy in optimum pH between our findings and those of Cyboron and Wuthier (1981) is probably due to the different methods used for preparation of matrix vesicles and not to differences in
activity of purified matrix vesicles from bone and cartilage. Top to bottom: Na+, Mg*+, Zn*+, Ca*+. 100% activity = Na+, 33192 and 24668, Mg*+, 31460 and 23422, Zn’: 28828 and 19734, Ca*+, 29044 and 12277 nmol P,/mg/min in bone and cartilage, respectively.
the temperature of the pH optimum measurements, as the enzymatic assays were carried out at 37% by both groups. Of the metal ions studied (Na+, Zn2+, Ca2+) only Zn2+ was found to have any significant effect on alkaline phosphatase activity in either tissue. The alkaline phosphatase activity in matrix vesicles of both cartilage and bone decreased by 70% in the presence of 5 mM Zn*+, No inhibition, however, was measured up to 1 mMZn *+. In this respect we are in disagreement with Fortuna et al. (1980) and Cyboron and Wuthier (1981) who reported Zn2+ inhibition of matrix vesicle alkaline phosphatase at much lower Zn*+ concentration (about 0.02 mM inhibition constant). It should be noted, however, that Cyboron et al. (1982) in their more recent work assaying alkaline phosphatase at pH 7.5, did not find Zn*+inhibition at up to 1 mM concentration of this cation. SDS-polyacrylamide gel electrophoretograms showed eight major protein bands common for purified matrix vesicles of both cartilage and bone. Two additional minor bands, however, were found in the profile of purified matrix
Table III. Purification of matrix vesicles from tubular bone and cartilage. Alkaline phosphatase specific activity
Tubular bone Epiphyseal cartilage Results are expressed as mean f experiments, 5.
Crude matrix vesicles
Peak of the purified fractions
Purification factor
18734 + 2075 12550 + 1637
32456 f 7622 20521?2845
1.7 1.6
standard error and are based on only those crude fraction preparations where purification was later performed. Number of
A. Muhlrad et al.: Biochemical characterization
i2
’ :
of matrix vesicles
\ \ 5
Fig. 3. pH profile of phosphatase specific activity of purified matrix vesicles from bone and cartilage. 100% activity = 17815 and 14627 nmol P;/mg/min for bone and cartilage, respectively.
vesicles from cartilage that were not observed in bone. This may indicate slight differences in the protein composition of matrix vesicles obtained from the two tissues. Stein et al. (1981) also studied protein profiles of matrix vesicles from rat
97
and fetal calf growth plates by polyacrylamide gel electrophoresis and reported among others several bands similar to the major bands found in this study. These included peptides of approximately 33,000-34,000 molecular weight, which was also noted by Wuthier et al. (1978) and may correspond to glyceraldehyde3-phosphatase dehydrogenase (G3-PD) of erythrocyte membranes. The 43,000 molecular weight peptide, which was observed both in this study and in that of Stein et al. (1981) was shown by Muhlrad et al. (1978, 1982) to correspond to actin. The proteases, which are present in the crude collagenase used in this study for preparation of matrix vesicles, might partially digest the vesicular proteins and affect their electrophoretic pattern, as it was noted by Wuthier et al. (1978). However, since the membrane of the matrix vesicles is tightly closed, only those proteins that are located on the external surface could be affected (Stein et al., 1981). Therefore, it is assumed that the SDS-polyacrylamide gel electrophoretogram gives a true picture about the size and distribution of proteins in matrix vesicles. Electron microscopic studies indicated that the general ultrastructure of these organelles in both tissues is similar. In both cartilage and bone the matrix vesicles appear as rounded organelles enveloped by a trilaminar membrane, often containing electron-dense material (Deutsch et al., 1981 b). The comparison of matrix vesicles obtained from bone and cartilage suggests that the same basic mechanisms are involved in the primary mineralization of these two tissues. It also shows that the matrix vesicles, the main promoters of this process, are strikingly similar.
Fig. 4. SDS-polyacrylamide gel electrophoretogram of crude and purified matrix vesicles from bone and cartilage. A, Crude and, 6, purified matrix vesicles from bone, C, Crude and, D, purified matrix vesiclesfrom cartilage. Molecular weight markers are E, actin, F, subfragment-l of myosin, G, tryptic digest of myosin subfragment-1, H, phosphorylase, I, P-galactosidase. J, soybean trypsin inhibitor, and K, myosin.
98
References
A. Muhlrad et al.: Biochemical characterization of matrix vesicles Fortuna R., Anderson H.C., Carty R.P and Sadjera SW.: Enzymatic characterization of the matrix vesicle alkaline phosphatase isolated from bovine fetal
Alcock N.M.: Calcrfrcation of cartrlage. C/in. Orthop. Rel Res. 86:287311.1972. Ali SY: Matrix vesicles and apatite nodules rn arthritic cartilage. In: Perspec-
epiphyseal cartilage. Calcif. Tissue Int. 30:217-225, 1980. Gottlieb A.J. and Sussman H.H Human placental alkaline phosphatase:
twes in Inflammation. D.A. Willoughby, J.P Giroud and G.P Velo, eds. MTP Press Ltd.. Lancaster, UK 1977, pp. 21 I-233. Ali SY, Sajdera S.W. and Anderson H.C.: Isolation and characterization of calcifying matrix vesicles from epiphyseal cartilage. Proc. Nat/. Acad. Sci. (USA) 67 1513-l 520, 1970. Anderson H.C Electron microscopic studies of induced cartilage development and calcification. J. Cell Biol. 3581-92, 1967. Anderson H.C : Vesrcles associated with calcrfication in the matrix of epiphyseal cartilage. J. Cell Bfol. 4159-72, 1969. Bab IL, Muhlrad A. and Sela J.: Ultrastructural and biochemrcal properties of extracellular matrix vesicles in normal alveolar bone of rats. Cell Tissue Res. 202.1-7. 1979. Bernard GW.: Ultrastructural observations of initial calcification in dentin and enamel. J. Ultrasfruct. Res. 41 :I -17, 1972. Bernard GW and Pease D.C.: An electron microscopic study of Initial rntramembraneous osteogenesis. Am. J. Anat. 125:271-290. 1969. Betts F, Blumenthal N.C., Posner A.S.. Becker G.L and Lehninger A.L.: Atomrc structure of rntracellular amorphous calcium phosphate depostis. Proc. Nat/. Acad. Sci.(USA) 72:2088-2090,1975. Bonucci E Fine structure of early cartilage calcrfication. J. Ultrastruct. Res. 20:33-50. 1967. Bonucci E.: Fine structure and histochemistry of calcifying globules in epiphyseal cartilage. 2. Zelforsch Mikrosk.Anat. 103:192-217.1970. Bradford N.M.: A raprd and sensitive method for the quantitation of mrcrogram quantities of protein utilizing the principle of protern-dye binding. Anal. Biochem. 72:248-254, 1976. Cyboron GW and Wuthier R.E.: Punfication and initial characterization of intrinsic membrane-bound alkaline phosphatase from chicken epiphyseal cartilage. J. B/o/ Chem. 256:7262-7268. 1981. Cyboron GW., Vejins M.S. and Wuthier R.E.: Activity of epiphyseal cartilage membrane alkaline phosphatase and the effects of its inhibitors at physiologrcal pH. J. Biol Chem. 257:4141-4146,1982. Deutsch D., Bab ItI Muhlrad A. and Sela J.: Purification and further characterization of isolated matrix vesicles from rat alveolar bone. Metab. Bone 0s. Rel. Res. 3:209-214.1981a. Deutsch D., Sela J., Bab I., Setton A.J. and Muhlrad A.: A comparative biochemical study on extracellular matrix vesicles isolated from bone and cartrlage. In’ Matrix Vesicles: Proceedings from the ThirdlnternationalConference on Matrix Vesicles. A. Ascenzi, E Bonucci and 8. de Bernard, eds. Wrchtig Editore. srl, Milano, 1981 b, pp. 53-58. Ersenman D.R. and Glick PL.: Ultrastructure of inttial crystal formation In dentin. J. Ultrastruct. Res. 41:18-28,1972. Frske C.H. and Subbarow Y. The calorimetric determination of phosphorous. J. 61ol. Chem. 66:375-400,1925. Fleisch H. and BisazS.: Mechanism of calcification: Inhibitory role of pyrophosphate Nature (London) 195:911, 1962. Fortuna R., Anderson H C , Carty R P and Sajdera SW: The purification and molecular characterization of alkaline phosphatase from chondrocytes and matrrx vesrcles of bovrne fetal epiphyseal cartilage. Metab. Bone Dis. Rel. Res. 1:161-168, 1978.
171, 1968. Hrrschman A.. Deutsch D., Bab I., Hrrschman M., Muhlrad A. and Sela J.: Assocratron of neutral peptidase actrvrtres with alkaline phosphatase actlvrty in matrrx vesrcles from bovrne fetal mandibles. In: Current Advances in Skeletogenesis, Silberman, M and Slavkin, H.C., Excerpta Medica, Amsterdam, pp 375-381, 1982 Kahn S.E., Jafri A.M , Lewis N.J. and Arsenis C.: Purification of alkaline phosphatase from extracellular vesicles of fracture callus cartilage. Calcif. Tissue Res. 25,85-92, 1978 Kim K.M.: Calcification of matnx vesicles in human aortic valve and aortic media. Fed. Proc. 36:156-162, 1976. Long L., Fabran F. Mason D.T and Wikman-Coffelt J.: A new cardiac myosin characterized from the canine atria. flioch!m. Biophys. Res. Commun. 76:626-635, 1977 Majeska R.T and Wuthier R.E: Studies on matrix vesicles isolated from chick epiphyseal cartilage: Assocration of pyrophosphatase and ATPase activities with alkaline phosphatase. Biochim. &ophys. Acta 391:51-60, 1975. Muhlrad A., Bab I.A., Deutsch D. and Sela J.: The occurrence of actinlike protein in extracellular matrix vesicles. Cabf. Tissueht. 34:376-381, 1982. Muhlrad A., Stein H., Bab I. and Sela J.: Fine structure and enzymes of matrix vesicles rn osteosarcoma. Possible occurrence of contractile proteins. Metab. Bone D/s. Rel. Res. 1:227-233.1978. Schenk R.K., Miller J., Zinkernagel R. and Willenegger H.: Ultrastructure of normal and abnormal bone repair. Calof. Tissue Res. 4 (SuppI.): 10-I 11, 1970. SelaJ., Bab I. and Muhlrad A: Ultrastructural and biochemical characterization of extracellular matrix vesicles in healing alveolar bone sockets: Preliminary Indications for the presence of contractile proteins. Metab. Bone Dis. Rel. Res. 1:185-191, 1978. Siska R.F and Provenza DV. Initial dentin formation in human deciduous teeth. An electron microscope study. Calcif. Tissue 9:1-16, 1972. Stein R.M., Hsu HT. and Anderson H.C.: Protein profiles of isolated fetal calf and rachitic rat matrix vesicles by polyacrylamide gel electrophoresis In, Matrix Vesicles: Proceedings from the Third International Conference on Matrix Vesicles. A. Ascenzi, E. Bonucci, and B. de Bernard, eds. Wichtig Editore. srl, Milano, 1981, pp. 117-l 22. Thyberg J. and Friberg U.: Electron microscopic enzyme histochemical studies on the cellular genesis of matrix vesicles in the epiphyseal plate. J. Ultrastruct. Res. 41:43-59, 1972. Vaananen H.K. and Korhonen L.K.: Matrix vesicles in chicken epiphyseal cartilage. Separation from lysosomes and the distribution of inorganic pyrophosphatase activity Calcif. Tissueht. 26:65-72, 1979. Wuthier R.E., Lrnder R.E., Warner G.P. GoreS.T and Borg TK.: Non-enzymatrc isolation of matrix vesicles: Characterization and initial studies on %a and a2P-orthophosphate metabolism. Mefab. Bone Dis Rel. Res. 1:125-136,
Molecular werght and subunit structure
&o&/m
Biophys. Acta 160,170-
1978.
December 13, 1982 Revised: April 8, 1983 Accepted: April 25, 1983 Received:
A. Muhlrad et al.: Biochemical characterization of matrix vesicles
99
Des vesicules de la matrice extracellulaire ont BtC isolees par digestion collagenasique a partir d’os et de cartilage epiphysaire femoral et tibia1 preteves chez des rats. Ces v6stcules, recueillies P I’etat brut, ont et6 ensuite puriflees par centrlfugation en gradlent sucrose. Des fractions contenant ks cellules et las membranes ont et6 egalament isolees de ces deux tissus. Cactivlte des phosphataaes alcaline et acide I’activiti, AtPasique, ainsi que le contenu pmteique de toutes les fractions, incluant celles contenant tes vesicules matrtcttlles P I’itat brut et puriflees, ont CtC evaluees. L’activite speciftque de la phosphatase alcaline a et6 trouvee elevee dans les vesicules matrictelles non purifieestellesqu’ellesont Cteextraites desdeux Tissus (5 a 20 fois plus que dans la fraction cellulaire). Les activites enzymatiques totales et le contenu proteique etaient significativement plus elevees dans toutes les fractions issues du cartilage par rapport B celles issues de 1’0s. Un fort accroissement d’activite de la phosphatase alcaline et du contenu proteique a etCobservCaprCslacentrifugationengradientsucrose. Lasituationdespicsetaitidentiquepourlesdeux tissus. L’activftespeciflquedelaphosphatase alcaline des vesicules matricialles puriftees etait significativement plus elevee dans 1’0s que dans le cartilage. Les activites phosphatasiques du cartilage et de 1’0s ont montre une dependance identique a I’egard du pH et une reponse identique aux ions metalliques. Pam-ti les ions metalliques testes (Na, Mg, Zn et Ca), seul le zinc, a une concentration de 5mM, a inhibe signittcativement I’activite phosphatasique alcaline des vesicules matricielles puriftees. Le pmfil electmphoretique des vesicules matricielles puriflees a revel6 8 bandes communes a 1’0s et au cartilage. En outre, les vesicules provenant du cartilage se sont avere dotees de deux peptides non presents dans 1’0s.
Bone Dis. & Rel. Res. 5, 93-99
0221-8747183 $3.00+ .OO Copyright 0 1984 Pergamon Press Ltd.
(1983)
Printed in the USA. All rights reserved.
Biochemical Characterization of Matrix Vesicles from Bone and Cartilage A. MUHLRAD’,
A. SETTON’,
J. SELA2, I. BAB2, and D. DEUTSCH’
‘Department of Oral Biology, and ‘D/v/sion of Oral Pathology, The Hebrew Umvemty-Hadassah
Faculty of Dental Med/cine, Jerusalem, Israel.
Address for correspondence and reprints:: Prof. A. Muhlrad, Department of Oral Biology, Hebrew University-Hadassah Medicine, P0.B. li72, Jerusalem 91010, Israel.
Abstract
Faculty of Dental
1976; Muhlrad et al., 1978; Ali, 1977). These organelles are believed to play a major role in the primary mineralization of these tissues. Most of the original studies were ultrastructural and qualitative in nature. To obtain quantitative information, chemical data were also required. This information was, however, difficult to obtain because of inadequate methods for isolation and purification of matrix vesicles. In recent years such methods have been developed (Ali et al., 1970; Majeska and Wuthier, 1975; Fortunaetal., 1978; Wuthier et al., 1978; Kahn et al., 1978; Sela et al., 1978; Deutsch et al., 1981, a,b), and the biochemical results obtained from purified matrix vesicles in combination with histologic findings have provided a more detailed picture of the role of matrix vesicles in primary mineralization. These studies revealed that alkaline phosphatase activity was 15-20 times higher in matrix vesicles than in cells (Ali et al., 1970; Majeska and Wuthier, 1975; Deutsch et al., 1981, a, b). Since this phenomenon is unique to mineralizing tissue, alkaline phosphatase has been used as a marker enzyme for matrix vesicles. In addition to alkaline phosphatase other phosphatases have been found to be associated with matrix vesicles. These include inorganic pyrophosphatase and ATPase. These enzymes are believed to hydrolize inorganic pyrophosphate and ATP, which would otherwise act to inhibit calcification (Fleisch and Bisaz, 1962; Alcock, 1972; Betts et al., 1975). Acid phosphatase has also been implicated as a matrix vesicle-associated enzyme. There is some controversy, however, regarding this association (Ali et al., 1970; Thyberg and Friberg, 1972). Most of these histologic and biochemical studies were performed on cartilage (Anderson, 1967, 1969; Bonucci, 1967, 1970; Vaananen and Korhonen, 1979; Majeska and Wuthier, 1975; Fortuna et al., 1978). In recent years similar studies have also been carried out on bone (Muhlrad et al., 1978; Sela et al., 1978; Bab et al., 1979; Deutsch et al., 1981a). These results indicate that the basic process of mineralization via matrix vesicles is similar in both tissues, but some differences may exist between them. However, a thorough comparison of matrix vesicles from various mineralizing tissues has not been carried out.
Extracellular matrix vesicles from bone and epiphyseal cartilage of femur and tibia of rats were isolated by collagenase digestion (crude vesicles) and further purified by sucrose gradient centrifugation. Fractions containing cells and membranes were also isolated from the two tissues. The alkaline and acid phosphatase and ATPase activities, as well as protein content of all fractions including crude and purified matrix vesicles, were assayed. The crude vesicles from both tissues demonstrated a high alkaline phosphatase specific activity (5-20 times higher than in the cell fraction). The total enzyme activities and protein content were significantly higher in all fractions from cartilage than those from bone. A major peak of alkaline phosphatase activity and protein content was obtained following the sucrose gradient centrifugation. The position of this peak was similar for both tissues. The specific activity of alkaline phosphatase of purified matrix vesicles was significantly higher in bone than in cartilage. The phosphatase activities from cartilage and bone showed a similar pH dependence and a similar response to metal ions. Of the metal ions tested (Na+, Mg*+, Zn*+, and Ca*+) only Zn*+ (at 5 mM concentration) inhibited significantly the alkaline phosphatase activity of purified matrix vesicles. The electrophoretic profile of purified matrix vesicles showed eight major protein bands common for both tissues. In addition, cartilage vesicles appeared to possess two peptides not found in bone.
Introduction The presence of extracellular matrix vesicles in cartilage in areas undergoing mineralization was reported by Anderson (1967) and Bonucci (1967). Subsequently matrix vesicles have been identified in bone (Bernard and Pease, 1969; Schenk et al., 1970), dentin (Bernard, 1972; Eisenman and Glick, 1972; Siska and Provenza, 1972), and a number of other tissues undergoing pathologic mineralization (Kim, 93
94
A. Muhlrad et al.: Biochemical characterization
The aim of this study was to isolate, characterize, and compare matrix vesicles from cartilage and bone with regard to the enzymatic activity (in particular, alkaline phosphatase), electrophoretic mobility of the vesicular proteins, and density profile on a sucrose gradient, using methods recently developed (Deutsch et al., 1981 a).
Materials and Methods Preparation of matrix vesicles Vesicles were prepared from bone and cartilage obtained from rats weightng 150-160 g of the Hebrew Unrversity (Sabra) strain. Tubular bones, femur, and tibia were used. The cartilage tissue was dissected from eplphyseal growth plates of the same bones. The procedures used for isolation and purification of matnx vesicles from bone and cartilage were described previously (Deutsch et al.; 1981a). Briefly, bone and cartilage were cleaned from soft tissue and blood, cut into small pieces, and digested with crude collagenase. The digest was decanted and subjected to differentral centrifugation. Three fractions-cells, membranes, and crude matrix vesicleswere obtained. For further purification the crude matrix vesicles were subjected to a noncontinuous sucrose-density gradient centrifugatron and fractionated. Each fraction was analyzed for various enzyme activities and protein content.
Polyacrylamide
gel electrophoresis
Both crude and purified matrix vesicle fractions were examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) as follows. After sucrose density gradient centrifugation, the matrrx vesrcle fractions containing the peak alkaline phosphatase activity were pooled, diluted five times with Gey’s solution, and centrifuged at 170,000 g for 60 min. The pellets were then resuspended In 100 ~1of sample buffer containing 322 mM Trrs-HCI pH 6.8, 32% glycerol, 7% SDS, and 3.2% /3-mercapto-ethanol and incubated at 100°C. The sample was then placed on a 7-l 8% polyacrylamrdegradient SDS slab gel according to the method of Long et al. (1977).
Enzymatic assays. Alkaline phosphatase
activity was determined In 1 ml of 100 mMglycine-NaOH pH 10.5, 1 mM MgCI,, and 0.1 mM ZnCI, containing 6 mM Na-p-nitrophenyl phosphate, as described by Sela et al. (1978). Acidphosphafase activity was measured in samples Incubated in 1 ml of 10 mM sodium acetate pH 5.0, containing 18 mM Nap-nitrophenyl phosphate for 30 min at 37*C. The reaction was stopped by the addition of 2.0 ml 0.5N NaOH and read immediately at 405 nm. Mgz+-ATfase activity was determined in the presence of 1.6 mM ATP in 1 ml of 2 mM MgCI, 10 mM imidazole pH 7.0, and Incubated at 37QCfor30 min. P,liberatedduring the reaction wasmeasured by the method of Fiske and Subbarow (1.925). Enzyme activity was expressed as either nmol P, liberated per mg protein per min (specific activrty) or nmol P, per g tissue (total activity) in all cases. /on studies. The effect of several catrons on alkaline phosphatase actlvrty was studied in crude and purified matrix vesicle Preparatlons from bone and cartilage. Enzyme activrty was assayed In a 100 mM glycine-NaOH buffer pH 10.5, with the addition of various concentrations of the following Ions: O.O5M-2M Na+ (as NaCI), 0.05 mM-10 mM Mg2+ d(as MgCI,), 0.01 mM-5 mM Zn*+ (as zincacetate), 0.05 mM-2 mM Ca” (as CaCI,).
ph’profile. The pH dependence of phosphatase activity was measured between pH 3.5 and 11 .O at 37OC according to the procedure used for assaying acid phosphatase activity. The following buffers were used at 100 mM concentrations, with their specific ranges: sodium acetate acetic acid pH 3.5-5.5, maleate-NaOH pH 5.5-6.5, imidazole-HCI pH 6.5-7.5, Tris-HCI pH 7.5-8.5, glycine-NaOH pH 8.5-l 1 .O. The pH of the buffers was also measured at 37’C.
Protein determination. ford (1976).
This was performed
according
to Brad-
of matrix vesicles
Results Crude matrix vesicles from bone and cartilage A comparison of matrix vesicles from bone and cartilage, obtained after collagenase digestion, is shown in Tables I and II. Significantly more protein per g tissue was found in the vesicles, membranes, and cell fractions obtained from cartilage. In addition the total activity per g tissue of enzymes assayed (alkaline phosphatase, acid phosphatase, and ATPase) was significantly higher in cartilage fractions than in bone. This difference between the two tissues is particularly apparent in the cell fraction, where the levels of acid and alkaline phosphatase total activity are 12 times higher in cartilage than in bone. The specific activity of alkaline phosphatase in vesicle fractions obtained from bone was higher than that in cartilage, while in the cell fraction the specific activity of this enzyme was lower (both differences were significant). The distribution of the total enzyme activity and protein content among the various fractions obtained from bone and cartilage is presented in Table I. The distribution of protein content, acid phosphatase, and ATPase activity is almost identical for bone and cartilage. However, differences were observed in the distribution of alkaline phosphatase. In bone, 55% of the alkaline phosphatase activity was found in the matrix vesicle fraction, whereas in cartilage the vesicle fraction contained only 35% of the total alkaline phosphatase activity. Purified matrix vesicles from bone and cartilage. The distribution of matrix vesicles from both tissues following sucrose density gradient centrifugation is presented in Figure 1. The purification factor, which is based on the increase in the specific activity of alkaline phosphatase before and after sucrose gradient centrifugation (Table Ill), is similar for both tissues. The protein content and alkaline phosphatase activity from the preparations of both tissues resolved into a major peak between 20 and 35% sucrose layers. There seemed to be a slight shift in the alkaline phosphatase activity of cartilage toward the heavier density. Some alkaline phosphatase activity and protein were also found in the heavy density fractions. The purification achieved during the step of sucrose gradient centrifugation is similar for both tissues. The specific activity of alkaline phosphatase of purified matrix vesicles obtained from bone is significantly higher than that of cartilage vesicles. Effects of ions and pH on enzymatic activity of purified matrix vesicles In the range of ion concentration employed in thisexperiment, no significant effects were obtained for Nat, Mg2+, and Ca’+. However, Zn2+ in a high concentration inhibited alkaline phosphatase activity (Fig. 2). A 70% inhibition of this enzyme was observed at 5 mMZn2+ concentration. No differences were found in the effect of these ions on the behavior of alkaline phosphatase activity from both tissues. The pH dependence of the phosphatase activity from purified matrix vesicles from bone and cartilage was measured between pH 3.5 and 11.0 and is presented in Figure 3. The optimum phosphatase activity in the acid region was found to be at pH 4.0 for both bone and cartilage. Matrix vesicles from both tissues showed similar curves in this region, and both preparations demonstrated an acid phosphatase specific activity of less than 1% of the highest alkaline phosphatase specific activity measured at pH_ 11 .O. However, the maximal acid phosphatase activity, relative to alkaline phosphatase in
A. Muhlrad et al.: Biochemical characterization
95
of matrix vesicles
Table I. Comparison of enzyme activity and protein content of matrix vesicles from tubular bone and cartilage. Protein (mglg tissue)
Alkaline phosphatase (act./g tissue) (spec. act.)
Acid phosphatase (spec. act.) (act/g tissue)
ATPase (spec. act.) (act./g tissue)
Tubular bone Crude matrix
1021 + 63
0.06 & ,007
Vesicles
(11%
f
1 O/o)
Cells
0.33 (63%
f f
.03 7%)
z? 3%) 218 f 25 (12% f 1%)
Membranes
0.13 f (25% f
.02 5%)
628 (34%
f 44 + 2%)
.36 (14% 1.55 (58% .74 (28%
.06 2%) ,l 4%) .09 4%)
3986 (35% 2861 (25% 4542 (40%
I! + -t + + *
14 f 2
17766 + 1988
(55%
720 5126
+ 156 + 1890
(8% 116 (70% 35 (21 s/o
f + f + f
1 c/o) 28 17%) 6 4%)
59 (3% 1312 (74% 390 (22%
f If: f + f +
11 1o/o) 131 7%) 48 3%)
201&52 335 f 52 278 + 34
72 f 22 (25% 122 (42% 93 (32%
f f & f f
8%) 19 7%) 16 6%)
275 (25% 487 (44% 341 (31%
+ c f f f -e
113 10%) 139 13%) 92 8%)
1136+419 369 f 94 676&92
Carttlage Crude matrix Vesicles Cells Membranes
& + f + f f
589 5%) 552 5%) 479 40/o)
11765 f
1497
1820 f 319 6293 f 774
165? 14 856284 543 + 73
736*419 297 f 77 423 f 60
“Values in parenthesis represent the distnbutron of the total acttvtty among the varrous fractions in percent Results are expressed as mean ? standard error.
cartilage vesicles, was 25% higher than in their bone counterpart. In the alkaline region the optimum pH was not determined. However, the curves for both tissues were still rising at pH 11 .O. Protein profile of purified matrix vesicles The electrophoretogram of purified and crude matrix vesicles from bone and cartilage is presented in Figure 4. Major peptide bands of the following respective approximate molecular weights were observed for both long bone and cartilage purified and crude fractions: 150,000, 81,000, 71,000, 43,000, 38,000, 34,000, 32,000, and 22,000. In addition, matrix vesicles from cartilage appeared to possess two peptides not found in bone. These were observed at molecular weights of 58,000 and 42,000.
Discussion The results of this study revealed that the amount of protein and total enzyme activities in the matrix vesicle preparations obtained from cartilage were much higher than in bone. This implies that cartilage contains more vesicles per g tissue than bone. This assumption is supported by recent electron microscopic results that show that more matrix vesicles are present in cartilage than in bone per g tissue. This difference can be explained by the different patterns of mineralization of the two tissues. Generally, in bone the major part of the tissue is already mineralized and thus does not contain intact matrix Table
vesicles. The result is that only a relatively small proportion of bone is undergoing active mineralization involving matrix vesicles. In cartilage, on the other hand, as mineralization occurs, the tissue is immediately resorbed and replaced by bone, the net result being that a major part of cartilage is found in the active mineralizing state involving matrix vesicles, This difference in mineralization pattern could provide an explanation to the larger amounts of cellular protein per weight tissue found in cartilage relative to bone. The distribution pattern of matrix vesicles, following purification by sucrose density gradient centrifugation, was almost identical for the two mineralizing tissues. Based on alkaline phosphatase activity, the matrix vesicle fraction of both tissues resolved Into a major peak. The position of this major peak of alkaline phosphatase activity in both tissues seems to correspond to similar peaks obtained by Kahn et al. (1978) for fracture callus cartilage of rabbit, Wuthier et al. (1978) for chicken epiphyseal cartilage, and Deutsch et al., (1981a) for rat alveolar bone. Alkaline phosphatase activity and protein were also found in the heavier density sucrose layers. Kahn et al. (1978) Wuthler et al. (1978) Deutsch et al. (1981a) and Hirschman et al. (1982) all reported the existence of a minor matrix vestcle peak in the heavier sucrose density layers. The occurrence of matrix vesicles in both light and heavy density layers might reflect matrix vesicles at different stages of mineralization. The effect of pH and various metal cations on phosphatase activity from purified matrix vesicles of both tissues was nearly identical. A number of findings in the present study were
II. Stattstical comparison of enzyme activity and protein content of fractions obtained from tubular bone and eptphyseal cartilage. Crude matrtx vesicles
Cells
Membranes p < 0.01
Alkaline phosphatase
Total activtty Specific activity
p < 0.01
p < 0.01
p < 0.05
p < 0.01
NS
Acid phosphatase
Total activity Specific actwity
p < 0.02 p < 0.05
p < 0.01 p < 0.01
p < 0.01 p < 0.05
ATPase
Total actrvity Specific activity
p < 0.05 NS
p < 0.05 NS
p < 0.05 NS
Total
p < 0.01
p < 0.01
p < 0.01
Protein
Values obtained In the same expenments NS, nonstgnifrcant.
for bone and cartilage fractrons were paired and analyzed by Student’s paired t-test. Number of experiments, 5.
96
A. Muhlrad et al.: Biochemical characterization of matrix vesicles
=-,Lr,.-
I
mm
.
-----F_---__O__O-
so
. UY
Fig. 2. Effects of various ions on alkaline phosphatase
Fig. 1. Elution profiles of alkaline phosphatase (top) and protein (bottom) for matrix vesicles from cartilage *--+ and bone ??-e following sucrose density gradient centrifugation. Top. 100% activity = 162 and 671 nmol P/g tissue for bone and cartilage, respectively. Bottom. 100% protein = 0.008 and 0.041 mg protein/g tissue for bone and cartilage, respectively. found to differ, however, from those reported in the literature. Fortuna et al. (1980) found that above pH 10.6, alkaline phosphatase isolated from matrix vesicles of fetal, bovine epiphyseal cartilage was rapidly and irreversibly denatured, while it was our experience that the response curves for crude vesicles were still rising in the alkaline region at pH 11 .O. Gottlieb and Sussman (1968) showed that alkaline phosphatase purified from human placenta was fully active even at pH 11.5. Cyboron and Wuthier (1981), working with alkaline phosphatase purified from matrix vesicles, found the pH optimum at 10.5 and did not report any denaturation at alkaline pH up to pH 11 .O. The reason for the slight discrepancy in optimum pH between our findings and those of Cyboron and Wuthier (1981) is probably due to the different methods used for preparation of matrix vesicles and not to differences in
activity of purified matrix vesicles from bone and cartilage. Top to bottom: Na+, Mg*+, Zn*+, Ca*+. 100% activity = Na+, 33192 and 24668, Mg*+, 31460 and 23422, Zn’: 28828 and 19734, Ca*+, 29044 and 12277 nmol P,/mg/min in bone and cartilage, respectively.
the temperature of the pH optimum measurements, as the enzymatic assays were carried out at 37% by both groups. Of the metal ions studied (Na+, Zn2+, Ca2+) only Zn2+ was found to have any significant effect on alkaline phosphatase activity in either tissue. The alkaline phosphatase activity in matrix vesicles of both cartilage and bone decreased by 70% in the presence of 5 mM Zn*+, No inhibition, however, was measured up to 1 mMZn *+. In this respect we are in disagreement with Fortuna et al. (1980) and Cyboron and Wuthier (1981) who reported Zn2+ inhibition of matrix vesicle alkaline phosphatase at much lower Zn*+ concentration (about 0.02 mM inhibition constant). It should be noted, however, that Cyboron et al. (1982) in their more recent work assaying alkaline phosphatase at pH 7.5, did not find Zn*+inhibition at up to 1 mM concentration of this cation. SDS-polyacrylamide gel electrophoretograms showed eight major protein bands common for purified matrix vesicles of both cartilage and bone. Two additional minor bands, however, were found in the profile of purified matrix
Table III. Purification of matrix vesicles from tubular bone and cartilage. Alkaline phosphatase specific activity
Tubular bone Epiphyseal cartilage Results are expressed as mean f experiments, 5.
Crude matrix vesicles
Peak of the purified fractions
Purification factor
18734 + 2075 12550 + 1637
32456 f 7622 20521?2845
1.7 1.6
standard error and are based on only those crude fraction preparations where purification was later performed. Number of
A. Muhlrad et al.: Biochemical characterization
i2
’ :
of matrix vesicles
\ \ 5
Fig. 3. pH profile of phosphatase specific activity of purified matrix vesicles from bone and cartilage. 100% activity = 17815 and 14627 nmol P;/mg/min for bone and cartilage, respectively.
vesicles from cartilage that were not observed in bone. This may indicate slight differences in the protein composition of matrix vesicles obtained from the two tissues. Stein et al. (1981) also studied protein profiles of matrix vesicles from rat
97
and fetal calf growth plates by polyacrylamide gel electrophoresis and reported among others several bands similar to the major bands found in this study. These included peptides of approximately 33,000-34,000 molecular weight, which was also noted by Wuthier et al. (1978) and may correspond to glyceraldehyde3-phosphatase dehydrogenase (G3-PD) of erythrocyte membranes. The 43,000 molecular weight peptide, which was observed both in this study and in that of Stein et al. (1981) was shown by Muhlrad et al. (1978, 1982) to correspond to actin. The proteases, which are present in the crude collagenase used in this study for preparation of matrix vesicles, might partially digest the vesicular proteins and affect their electrophoretic pattern, as it was noted by Wuthier et al. (1978). However, since the membrane of the matrix vesicles is tightly closed, only those proteins that are located on the external surface could be affected (Stein et al., 1981). Therefore, it is assumed that the SDS-polyacrylamide gel electrophoretogram gives a true picture about the size and distribution of proteins in matrix vesicles. Electron microscopic studies indicated that the general ultrastructure of these organelles in both tissues is similar. In both cartilage and bone the matrix vesicles appear as rounded organelles enveloped by a trilaminar membrane, often containing electron-dense material (Deutsch et al., 1981 b). The comparison of matrix vesicles obtained from bone and cartilage suggests that the same basic mechanisms are involved in the primary mineralization of these two tissues. It also shows that the matrix vesicles, the main promoters of this process, are strikingly similar.
Fig. 4. SDS-polyacrylamide gel electrophoretogram of crude and purified matrix vesicles from bone and cartilage. A, Crude and, 6, purified matrix vesicles from bone, C, Crude and, D, purified matrix vesiclesfrom cartilage. Molecular weight markers are E, actin, F, subfragment-l of myosin, G, tryptic digest of myosin subfragment-1, H, phosphorylase, I, P-galactosidase. J, soybean trypsin inhibitor, and K, myosin.
98
References
A. Muhlrad et al.: Biochemical characterization of matrix vesicles Fortuna R., Anderson H.C., Carty R.P and Sadjera SW.: Enzymatic characterization of the matrix vesicle alkaline phosphatase isolated from bovine fetal
Alcock N.M.: Calcrfrcation of cartrlage. C/in. Orthop. Rel Res. 86:287311.1972. Ali SY: Matrix vesicles and apatite nodules rn arthritic cartilage. In: Perspec-
epiphyseal cartilage. Calcif. Tissue Int. 30:217-225, 1980. Gottlieb A.J. and Sussman H.H Human placental alkaline phosphatase:
twes in Inflammation. D.A. Willoughby, J.P Giroud and G.P Velo, eds. MTP Press Ltd.. Lancaster, UK 1977, pp. 21 I-233. Ali SY, Sajdera S.W. and Anderson H.C.: Isolation and characterization of calcifying matrix vesicles from epiphyseal cartilage. Proc. Nat/. Acad. Sci. (USA) 67 1513-l 520, 1970. Anderson H.C Electron microscopic studies of induced cartilage development and calcification. J. Cell Biol. 3581-92, 1967. Anderson H.C : Vesrcles associated with calcrfication in the matrix of epiphyseal cartilage. J. Cell Bfol. 4159-72, 1969. Bab IL, Muhlrad A. and Sela J.: Ultrastructural and biochemrcal properties of extracellular matrix vesicles in normal alveolar bone of rats. Cell Tissue Res. 202.1-7. 1979. Bernard GW.: Ultrastructural observations of initial calcification in dentin and enamel. J. Ultrasfruct. Res. 41 :I -17, 1972. Bernard GW and Pease D.C.: An electron microscopic study of Initial rntramembraneous osteogenesis. Am. J. Anat. 125:271-290. 1969. Betts F, Blumenthal N.C., Posner A.S.. Becker G.L and Lehninger A.L.: Atomrc structure of rntracellular amorphous calcium phosphate depostis. Proc. Nat/. Acad. Sci.(USA) 72:2088-2090,1975. Bonucci E Fine structure of early cartilage calcrfication. J. Ultrastruct. Res. 20:33-50. 1967. Bonucci E.: Fine structure and histochemistry of calcifying globules in epiphyseal cartilage. 2. Zelforsch Mikrosk.Anat. 103:192-217.1970. Bradford N.M.: A raprd and sensitive method for the quantitation of mrcrogram quantities of protein utilizing the principle of protern-dye binding. Anal. Biochem. 72:248-254, 1976. Cyboron GW and Wuthier R.E.: Punfication and initial characterization of intrinsic membrane-bound alkaline phosphatase from chicken epiphyseal cartilage. J. B/o/ Chem. 256:7262-7268. 1981. Cyboron GW., Vejins M.S. and Wuthier R.E.: Activity of epiphyseal cartilage membrane alkaline phosphatase and the effects of its inhibitors at physiologrcal pH. J. Biol Chem. 257:4141-4146,1982. Deutsch D., Bab ItI Muhlrad A. and Sela J.: Purification and further characterization of isolated matrix vesicles from rat alveolar bone. Metab. Bone 0s. Rel. Res. 3:209-214.1981a. Deutsch D., Sela J., Bab I., Setton A.J. and Muhlrad A.: A comparative biochemical study on extracellular matrix vesicles isolated from bone and cartrlage. In’ Matrix Vesicles: Proceedings from the ThirdlnternationalConference on Matrix Vesicles. A. Ascenzi, E Bonucci and 8. de Bernard, eds. Wrchtig Editore. srl, Milano, 1981 b, pp. 53-58. Ersenman D.R. and Glick PL.: Ultrastructure of inttial crystal formation In dentin. J. Ultrastruct. Res. 41:18-28,1972. Frske C.H. and Subbarow Y. The calorimetric determination of phosphorous. J. 61ol. Chem. 66:375-400,1925. Fleisch H. and BisazS.: Mechanism of calcification: Inhibitory role of pyrophosphate Nature (London) 195:911, 1962. Fortuna R., Anderson H C , Carty R P and Sajdera SW: The purification and molecular characterization of alkaline phosphatase from chondrocytes and matrrx vesrcles of bovrne fetal epiphyseal cartilage. Metab. Bone Dis. Rel. Res. 1:161-168, 1978.
171, 1968. Hrrschman A.. Deutsch D., Bab I., Hrrschman M., Muhlrad A. and Sela J.: Assocratron of neutral peptidase actrvrtres with alkaline phosphatase actlvrty in matrrx vesrcles from bovrne fetal mandibles. In: Current Advances in Skeletogenesis, Silberman, M and Slavkin, H.C., Excerpta Medica, Amsterdam, pp 375-381, 1982 Kahn S.E., Jafri A.M , Lewis N.J. and Arsenis C.: Purification of alkaline phosphatase from extracellular vesicles of fracture callus cartilage. Calcif. Tissue Res. 25,85-92, 1978 Kim K.M.: Calcification of matnx vesicles in human aortic valve and aortic media. Fed. Proc. 36:156-162, 1976. Long L., Fabran F. Mason D.T and Wikman-Coffelt J.: A new cardiac myosin characterized from the canine atria. flioch!m. Biophys. Res. Commun. 76:626-635, 1977 Majeska R.T and Wuthier R.E: Studies on matrix vesicles isolated from chick epiphyseal cartilage: Assocration of pyrophosphatase and ATPase activities with alkaline phosphatase. Biochim. &ophys. Acta 391:51-60, 1975. Muhlrad A., Bab I.A., Deutsch D. and Sela J.: The occurrence of actinlike protein in extracellular matrix vesicles. Cabf. Tissueht. 34:376-381, 1982. Muhlrad A., Stein H., Bab I. and Sela J.: Fine structure and enzymes of matrix vesicles rn osteosarcoma. Possible occurrence of contractile proteins. Metab. Bone D/s. Rel. Res. 1:227-233.1978. Schenk R.K., Miller J., Zinkernagel R. and Willenegger H.: Ultrastructure of normal and abnormal bone repair. Calof. Tissue Res. 4 (SuppI.): 10-I 11, 1970. SelaJ., Bab I. and Muhlrad A: Ultrastructural and biochemical characterization of extracellular matrix vesicles in healing alveolar bone sockets: Preliminary Indications for the presence of contractile proteins. Metab. Bone Dis. Rel. Res. 1:185-191, 1978. Siska R.F and Provenza DV. Initial dentin formation in human deciduous teeth. An electron microscope study. Calcif. Tissue 9:1-16, 1972. Stein R.M., Hsu HT. and Anderson H.C.: Protein profiles of isolated fetal calf and rachitic rat matrix vesicles by polyacrylamide gel electrophoresis In, Matrix Vesicles: Proceedings from the Third International Conference on Matrix Vesicles. A. Ascenzi, E. Bonucci, and B. de Bernard, eds. Wichtig Editore. srl, Milano, 1981, pp. 117-l 22. Thyberg J. and Friberg U.: Electron microscopic enzyme histochemical studies on the cellular genesis of matrix vesicles in the epiphyseal plate. J. Ultrastruct. Res. 41:43-59, 1972. Vaananen H.K. and Korhonen L.K.: Matrix vesicles in chicken epiphyseal cartilage. Separation from lysosomes and the distribution of inorganic pyrophosphatase activity Calcif. Tissueht. 26:65-72, 1979. Wuthier R.E., Lrnder R.E., Warner G.P. GoreS.T and Borg TK.: Non-enzymatrc isolation of matrix vesicles: Characterization and initial studies on %a and a2P-orthophosphate metabolism. Mefab. Bone Dis Rel. Res. 1:125-136,
Molecular werght and subunit structure
&o&/m
Biophys. Acta 160,170-
1978.
December 13, 1982 Revised: April 8, 1983 Accepted: April 25, 1983 Received:
A. Muhlrad et al.: Biochemical characterization of matrix vesicles
99
Des vesicules de la matrice extracellulaire ont BtC isolees par digestion collagenasique a partir d’os et de cartilage epiphysaire femoral et tibia1 preteves chez des rats. Ces v6stcules, recueillies P I’etat brut, ont et6 ensuite puriflees par centrlfugation en gradlent sucrose. Des fractions contenant ks cellules et las membranes ont et6 egalament isolees de ces deux tissus. Cactivlte des phosphataaes alcaline et acide I’activiti, AtPasique, ainsi que le contenu pmteique de toutes les fractions, incluant celles contenant tes vesicules matrtcttlles P I’itat brut et puriflees, ont CtC evaluees. L’activite speciftque de la phosphatase alcaline a et6 trouvee elevee dans les vesicules matrictelles non purifieestellesqu’ellesont Cteextraites desdeux Tissus (5 a 20 fois plus que dans la fraction cellulaire). Les activites enzymatiques totales et le contenu proteique etaient significativement plus elevees dans toutes les fractions issues du cartilage par rapport B celles issues de 1’0s. Un fort accroissement d’activite de la phosphatase alcaline et du contenu proteique a etCobservCaprCslacentrifugationengradientsucrose. Lasituationdespicsetaitidentiquepourlesdeux tissus. L’activftespeciflquedelaphosphatase alcaline des vesicules matricialles puriftees etait significativement plus elevee dans 1’0s que dans le cartilage. Les activites phosphatasiques du cartilage et de 1’0s ont montre une dependance identique a I’egard du pH et une reponse identique aux ions metalliques. Pam-ti les ions metalliques testes (Na, Mg, Zn et Ca), seul le zinc, a une concentration de 5mM, a inhibe signittcativement I’activite phosphatasique alcaline des vesicules matricielles puriftees. Le pmfil electmphoretique des vesicules matricielles puriflees a revel6 8 bandes communes a 1’0s et au cartilage. En outre, les vesicules provenant du cartilage se sont avere dotees de deux peptides non presents dans 1’0s.