Matrix VoL 11/1991, pp. 276-281 © 1991 by Gustav Fischer Verlag, Stuttgart
Stimulation of Cartilage Inorganic Pyrophosphate Elaboration by Ascorbate LAWRENCE M. RYAN, INDIRA KURUP and HERMAN S. CHEUNG Division of Rheumatology, Department of Medicine, Medical College of Wisconsin, Milwaukee, WI, 53226, USA.
Abstract
Ascorbate (0- 500 !tM) stimulates synthesis and secretion of collagen by cartilage explants in a dose-dependent fashion as estimated by eH] proline incorporation. Concurrent elaboration of inorganic pyrophosphate (PPi) parallels [3H] proline incorporation « 0.001, Wilcoxon rank sum). The effect on proline and PPi is abolished by ascorbate oxidase. Another reducing agent, Vitamin E, did not promote PPi accumulation about cartilage. Inhibitors of collagen synthesis, including monensin, tumor necrosis factor a, and 2' ,2' dipyridyl also inhibited PPi elaboration. Cycloheximide (1 !tg/ml) inhibited the ascorbate stimulated PPi elaboration 54% but did not attenuate the secretion of PPi by unstimulated cartilage. Cosecretion of collagen and PPi by chondrocytes may explain these results. Moreover, at least two pathways exist for PPi elaboration, a cycloheximide sensitive path and another independent of new protein synthesis. Key words: ascorbate, cartilage, inorganic pyrophosphate.
Introduction
Cartilages of patients with articular chondrocalcinosis contain calcium pyrophosphate dihydrate (CPPD) crystals. CPPD crystallogenesis, an extracellular event, requires extracellular inorganic pyrophosphate (PPi). Such PPi might form extracellularly. For example, the chondrocyte ectoenzyme, nucleoside triphosphate pyrophosphohydrolase (NTPPPH), generates PPi and nucleoside monophosphate from nucleoside triphosphates (NTP). (Ryan et a1., 1984; Howell et a1., 1984) Potential sources of substrate for this ectoenzyme include synovial fluid NTP or NTP released from stimulated, damaged, or traumatized chondrocytes. Alternatively, PPi made intracellularly may traverse the cell membrane to reach the site of crystal formation. Intracellular PPi is a byproduct and regulator of multiple biosynthetic reactions, a concept introduced by
Kornberg (1962). We have postulated cosecretion of matrix-destined molecules with their synthetic coproduct, PPi (Ryan et a1., 1981; 1985). In cartilage collagen and glycosaminoglycan are major extracellular matrix constituents. The coexport thesis has been disproved for sulfated glycosaminoglycan, secretion of which is neither sufficient nor necessary for the release of PPi by cartilage organ cultures (Prins et a1., 1986; Ryan et a1., 1990). In order to further test the cosecretion hypothesis, we have studied the accumulation of extracellular PPi about cartilage explants as a function of secretion of the other major matrix constituent, collagen. We report that ascorbate (asc) stimulation of collagen production and secretion by cartilage cells is accompanied by increased PPi elaboration. When collagen synthesis and secretion is inhibited PPi accumulation around cartilage explants decreases. Materials and Methods
Presented in part at the 53rd Annual Scientific Meeting, American College of Rheuma tology, Cincinnati, 0 H, June, 1989.
Dulbecco's modified Eagle's medium (DMEM), fetal calf serum (FCS), Pen-Strep-Fungizone (PSF), and N-(2-
Stimulation of Cartilage Inorganic Pyrophosphate hydroxyethyl) piperazine' -(2-ethanesulfonic acid) (HEPES) buffer were obtained from Grand Island Biological. eH] proline was from Amersham. [14C] adenine, [32p] PPi, and 35 [ S] S04 were from New England Nuclear. L- and D-asc were from Kodak Laboratories. Asc oxidase, vitamin E (Da-tocopherol acetate), cycloheximide, 2,2-'dipyridyl, and monensin were from Sigma. Human recombinant gamma interferon was supplied by Collaborative Research. Human recombinant tumor necrosis factor a (TNFa) was from R and D Systems. Chromatographically purified Clostridial collagenase was from Worthington. Minced cartilage from freshly slaughtered pig distal femur was cultured in DMEM supplemented with 10% FCS and 1 % PSF for 24 h. Then cultures were incubated for 24 h. in HEPES-buffered DMEM without FCS before replacing with experimental media. Medium was then changed to DMEM with 10% FCS heat-inactivated (HI) (at 56°C for 1 h to remove pyrophosphatase activity) containing 1 % PSF, and 50 mM HEPES-buffer, pH 7.4 and additives being tested. All cultures were performed at 37°C in 10% CO 2 atmosphere. [3H] proline incorporation into collagen was determined after incubation of cartilage with 20 f!Ci/ml in a 20-fold (vol!wet weight) volume of 50 mM HEPES-buffered DMEM containing HI FCS and PSF by the a modified method of Benya and Nimni (1979). Briefly, after labeling, media and washes were adjusted to 30% saturation with ammonium sulfate, incubated at 4 °C for 18 h, and precipitate collected. Precipitate was dissolved in 0.5 M acetic acid, adjusted to 100 f!glml pepsin, and incubated for 18 h at 4 dc. The pepsin digest was dialyzed against 2 M NaCI at 4°C to reprecipitate collagen. The precipitated collagen was collected by centrifugation, dissolved in 0.5 M acetic acid and counted. The collagen nature of this preparation was demonstrated by susceptibility to digestion with protease free specific collagenase. [3H] proline incorporated into explant collagen was partly pepsin extractable and partly resistant to pepsin extraction. Pepsin extractable [3H] proline-labelled collagen was quantified by incubating cartilage with 1 mglml pepsin in 0.5 M acetic acid for 48 h. Pepsin extract was removed and cartilage further extracted for 24 h with 50 mM Tris-CI in 1 M NaCI at pH 7.5. The two extracts were combined, neutralized, and collagen precipitated by dialysis against 4.4 M NaCl at 4°C overnight. Precipitate was dissolved in 0.5 M acetic acid, clarified by centrifugation, and counted. These pepsin-extracted counts were collagenase sensitive. Residual eH] proline-counts in the extracted cartilage were also 96-98% collagenase sensitive. Therefore, the pepsin resistant and collagenase sensitive counts were also measured after hydrolysis by incubation with 6N HCI at 108°C for 1 h. Reported values for [3H] proline incorporation into medium and tissue collagen are the sum of all three compartments, namely, media, cartilage extract, and cartilage
277
hydrolysate. In short term organ culture most collagen synthesized by hyaline articular cartilage is type II (Benya and Nimni, 1979). In 3 organ cultures we determined that the increase in [3H] proline incorporation in the presence of asc was mostly (73, 80 and 87%) into collagenase sensitive protein. 35 [ S] S04 incorporation as a reflection of sulfated glycosaminoglycan synthesis and secretion was measured as described in detail elsewhere (Prins et aI., 1986). PPi was measured by the radiometric UDPG pyrophosphorylase method as previously described (Cheung and Suhadolnik, 1977). All standard curves were prepared in the medium being tested, since various additives altered the slope of the standard curve. Determination of hydrolysis of trace [32 p] PPi in medium was by an adaptation of the method of Sugini and Miyosho (Sugino and Miyoshi, 1964). Ecto-NTPPPH was assayed by incubation of cartilage with the artificial substrate p-nitrophenyl thymidine monophosphate (PNTMP) (Rachow and Ryan, 1985). One mM PNTMP in Hank's balanced salt solution buffered with 25 mM HEPES to pH 7.4 was prepared and a 20-fold (vol! wt) volume placed over cartilage. The reaction was stopped after 1 h by adding a 4-fold volume of 0.1 N NaOH and concentration of p-nitrophenyl generated read at OD 41O . Results Asc increases PPi elaboration by cartilage. When 50 f!M L-asc was included in medium, accumulation of PPi increased 97% (medium PPi concentration 5.8 ± 0.5 SEM f!M without asc, 11.4 ± 1.1 f!M with asc, n = 8) (p = 0.01, Wilcoxon signed rank test) after 48 h (TableI). Cartilage from each of the animals tested elaborated more PPi with asc than without. Consistently elevated [PPi] was evident at 24 h and persisted through at least 96 h. Excess PPi accumulation was not caused by decreased hydrolysis of extracellular PPi, since trace [32 p] PPi added to the medium was hydrolysed similarly in the asc-containing and -lacking cultures (data not shown). D- and L-asc were similarly effective PPI stimulants (Fig. 1). The effect of asc was dose dependent up to 500 f!M (Fig. 1). Asc oxidase 0.5 U/ml prevented the enhanced PPi accumulation by 100 f!M D- or Lasc (medium [PPi] at 48 h being 8.9 f!M for control with oxidase, 11.5 for D-asc, 8.6 D-asc plus oxidase, 14.6 for Lasc, and 9.9 for L-asc plus oxidase, n = 3 for each group). Since vitamin C is a reducing agent, altered PPi metabolism might be nonspecific, related to redox characteristics. The effect of another reductant, vitamin E was studied. At concentrations of 200 and 500 f!glml tocopherol acetate did not promote medium PPi accumulation compared to control cultures (data not shown) containing identical concentrations of acetate and alcohol (tocopherol solvent). To determine whether matrix molecule synthesis was
278
L. M. Ryan et al.
Table I. Effect of I.-Ascorbate Treatment on Cartilage Medium Inorganic Pyrophosphate Concentration at 48 Hours.
30,---------------------------~
Medium [Pyrophosphate) ([!M) - ascorbate + ascorbate
6.1a 5.1
10.4 12.6 13.8 16.6 11.9 9.6 8.0 8.0
8.9
6.4 5.5 4.5 4.7 5.0 a
a
20
iii a
10 !p
20~--------------------------'
C!I D-ASC
25uM
SOuM
Fig. 1. Medium inorganic pyrophosphate concentration after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) wolume of Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal calf serum and varying concentrations of ascorbate. PPi = inorganic pyrophosphate L-ASC = L-ascorbate D-ASC = D-ascorbate Error bars represent standard error of the mean, n = 3.
increased by asc, glycosaminoglycan and collagen metabolism were studied. Asc did not significantly increase 35 [ S]S04 incorporation into cartilage matrix or medium at 24 (n = 6) or 48 (n = 4) hours (data not shown). As expected L-asc increased [3H] proline incorporation by 169 ± 9% (± SEM, n = 11). D-asc had comparable effect. And ascorbate oxidase blocked the increases in proline incorporation and PPi elaboration. In 11-matched specimens PPi elaboration was compared to proline incorporation. PPi production increased 91 ± 10% (± SEM) while proline incorporation was enhanced 169 ± 9%. As for PPi accumulation, so also proline incorporation was dependent on L-asc concentration. Proline incorporation and PPi elaboration in 25 cartilages (11 untreated, 11 L-asc treated, and 3 D-asc treated) were significantly correlated (Fig. 2) (p < 0.001, Spearman's rank correlation).
aa a
a a
a
I
20
10
Each value represents the mean of triplicate determinations on cartilages from different animals.
a
a cP
rf
a aa aa
a
40
30
50
3H-Proline, cpm/ug Protein
Fig. 2. Medium inorganic pyrophosphate concentration versus [3Hl proline incorporation after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) volume of Dulbecco's modified Eagle's medium supplemented with 10'10 heat-inactivated fetal calf serum, with or without ascorbate. rs = 0.8, p < 0.001 (Spearman's rank correlation).
To further study the relationship between collagen production and PPi elaboration we determined the effect on medium [PPi] of inhibiting collagen synthesis and secretion using cycloheximide, gamma interferon, monensin, TNFa, and 2,2' dipyridyl. A nonselective protein synthesis inhibitor, cycloheximide (1 [lg/ml), inhibited proline incorporation by asc-treated cartilages by 88%, glycine incorporation by 97% and sulfation by 99%. Simultaneous medium PPi concentration after 48 hours was 54% less than cultures without cycloheximide. The cartilage
20~-------------------------------,
seRUM
ASC
CHX
t
t
+
t
+
t
5
50
t
+
t t 5
t t 50
Fig. 3. Medium inorganic pyrophosphate concentration after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) volume of Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal calf serum and ascorbate (50 [!glml) or cycloheximide (0- 50 [!glml). PPi = inorganic pyrophosphate ASC = L-ascorbate CHX = cycloheximide concentration Error bars represent standard error of the mean, n = 3.
Stimulation of Cartilage Inorganic Pyrophosphate remained viable as demonstrated by the ability of explants to recover proline incorporation when removed from the cycloheximide. Interestingly, cycloheximide inhibited only the increment in PPi elaboration induced by asc, but did not inhibit the basal secretion of PPi by unstimulated cartilages (Fig. 3). Interferon-gamma reportedly diminishes collagen production by cultured fibroblasts and chondrocytes (Goldring et aI., 1986; Granstein et aI., 1987). But in porcine cartilage human recombinant interferon-gamma failed to suppress either PPi production or proline incorporation (data not shown). Moreover, human gamma interferon did not reduce collagen secretion by porcine fibroblasts, suggesting that species differences account for the inefficacy. TNFa suppresses collagen release by fibroblasts (Mauviel et aI., 1988). Human recombinant TNFa decreased media PPi concentrations and proline incorporation in asc-stimulated cartilage organ cultures (Table II). Since TNFa also
279
stimulates production of prostaglandin E2 , we determined whether indomethacin (10 ILM) would abrogate the TNFa inhibition. Indomethacin did not reverse the inhibition of PPi elaboration or eH) proline incorporation (data not shown). The monovalent cation ionophore, monensin, in a dose dependent fashion decreased medium [PPi] and proline incorporation (Fig. 4). Since monensin may interfere with cycling of ectoenzymes, (Basu et aI., 1981) including NTPPPH which generates extracellular PPi, we determined the ecto-NTPPPH activities of cartilages after a 48-hour monensin incubation. Ecto-NTPPPH was unaltered by monensin (data not shown). Finally, the relatively specific inhibitor of collagen production, 2,2' dipyridyl (Bhatnagar and Prockop, 1966), decreased PPi accumulation and proline incorporation by asc-stimulated cultures (Fig. 5). 2o.-------------------------~20
DpPI
o
3H Proline
ASC
SuM
Table II. Inorganic Pyrophosphate in Cartilage Medium and eH)Proline Incorporation by Cartilage Treated with Tumor Necrosis Factor a. TNFa (ng/m!)
PPi ([tM) 48h
0 1 10 100
12.1 10.5 8.6 10.3
± ± ± ±
2.0 a 2.2 1.6 2.4
PPi ([tM) 72h
[3H)-proline (cpm/mg· 103 ) 48 h
16.1±2.1 11.9 ± 2.2 7.4 ± 1.2 4.8 ± 1.2
19.0 ± 14.1 ± 9.0 ± 7.4 ±
0.3 0.2 0.4 0.3
Control
All values are x ± SEM, n = 3. Porcine articular cartilage treated with recombinant human tumor necrosis factor was incubated in asc-containing (50 [tg/ml) medium and ambient [PPi) measured at 48 and 72 hours. Collagen production was estimated by incorporation of [-'Hl-proline.
a
~------------------------~~ 3H Proline
20
l
i' n
SOuM
SOOuM
Fig. 5. Medium inorganic pyrophosphate concentration and eHl proline incorporation after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) volume of Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal calf serum. All cultures except control contain 50 [tg/ml ascorbate and 0-500 [tM 2,2'dipyridyl. PPi = inorganic pyrophosphate Error bars represent standard error of the mean, n = 3.
Discussion ~
r
.....
1. . . . s
.8 §.
10 ~ ......
f o Fig. 4. Medium inorganic pyrophosphate concentration and eH) proline incorporation after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) volume of Dulbecco's modified Eagle's medium supplemented with 10% heat-inaccivated fetal calf serum, ascorbate (50 [tg/ml), and monensin (0-10 [tM). PPi = inorganic pyrophosphate Error bars represent standard error of the mean, n = 3.
Since CPPD crystals form interstitially, extracellular PPi is essential. An extra-articular source of cartilage PPi seems unlikely in view of the consistent downgradient in PPi concentration between synovial fluid and plasma (Silcox and McCarty, 1974). Among potential intra-articular sources, the chondrocyte alone elaborates extracellular PPi (Ryan et aI., 1981; Howell et aI., 1975). But PPi synthesized within the cell diffuses across biomembranes poorly; thus, simple diffusion could not explain the appearance of intracellularly generated PPi in the cartilage matrix. We have postulated previously that matrix molecules and PPi formed as their synthetic coproduct might be coexported. This hypothesis has been disproved for one prominent matrix component, glycosaminoglycan (Ryan et aI., 1990). Our present findings suggest a link between PPi elaboration and
280
L. M. Ryan et al.
production of the other major cartilage matrix molecule, collagen. When collagen secretion was stimulated by asc, PPi elaboration increased. Both responses were dose-dependent and significantly correlated. The effects of asc were not likely due to any contaminant in the vitamin preparation, since collagen and PPi accrual were both blocked by treatment with asc oxidase which generates biologically active but unstable asc derivatives. Extracellular PPi may accrue due to other asc influences. Asc is a nonspecific reducing agent. Thus it was important to demonstrate that another reductant, a-tocopherol, had no similar impact on medium PPi concentration or proline incorporation. If collagen production and PPi liberation are linked, then inhibiting collagen synthesis should decrease extracellular PPi elaboration. Unfortunately all inhibitors have some degree of nonspecificity and results must be interpreted with that in mind. Nonetheless, treating asc-stimulated cartilage with collagen synthesis or secretion inhibitors decreased proline incorporation and PPi elaboration in parallel. Additionally, these inhibitors act at different levels of collagen processing. For example cycloheximide is a classic translation inhibitor, while 2,2-dipyridyl probably acts by inhibiting iron dependent posttranslational modification of pro collagen (Bhatnagar et al., 1966). Each decreases collagen secretion and both diminish PPi elaboration. Most interesting was the observation that basal production of PPi by unstimulated cells is not diminished by cycloheximide, while the increment in PPi elaboration following asc stimulation is abolished by cycloheximide. Thus, at least two pathways for PPi extrusion, one protein synthesisdependent and one independent of new protein synthesis, coexist in cartilage. The latter basal production is reportedly abolished by treating chondrocytes with sufficient trypsin to remove ecto-NTPPPH activity (Prins et al., 1986). This component may be generated by ectoenzymes which salvage nucleoside triphosphates that have leaked from cells. The cycloheximide-sensitive pathway is dependent on new protein synthesis and may represent cosecretion of PPi with collagen. Although asc stimulates collagen production by cartilage, it may have other unrelated and unexamined effects which are responsible for PPi secretion. For instance, we have reported that growth factors stimulate PPi elaboration by cartilage (Rosenthal et al., 1989; 1990), although the intracellular signalling pathway is uncertain. Asc may stimulate PPi secretion through these pathways. Clinical association of osteoarthritis and CPPD deposition has been long recognized. Recent reports associate premature and severe osteoarthritis and/or dysplasia transmitted in an autosomal dominant pattern with a specific allele in restriction fragment length polymorphism for the type II procollagen gene (Knowlton et al., 1990). Analysis for similar genomic type II collagen abnormalities in CPPD deposition associated with degenerative disease or in those
kindreds with autosomal dominant transmission of CPPD deposition (Ryan and McCarty, 1988) may shed light on the pathogenesis of this disease. Acknowledgements This work is supported in part by grant from the Wisconsin Chapter of the Arthritis Foundation, by a Biomedical Research Center Grant from the Arthritis Foundation, and by Public Health Service Grants AR 38656 and AR 38421 from NIH. Dr. Ryan is recipient of Research Career Development Award from the National Institute of Health. References Ryan, L.M., Wortmann, R.L., Karas, B. and McCarty, D.J.: Cartilage nucleoside triphosphate (NTP) pyrophosphohydrolase I. Identification as an ecto-enzyme. Arthr. Rheum. 27: 404-409, 1984. Howell, D. S., Pelletier, J. P., Martel-Pelletier, J., Morales, S. and Muniz, 0.: Nucleoside triphosphate (NTP) pyrophosphohydrolase in chondrocalcinotic and osteoarthritic cartilages. II. Further studies on histologic and subcellular distribution. Arthr. Rheum. 27: 193-199, 1984. Kornberg, A.: On the metabolic significance of phosphorolytic and pyrophosphorolytic reactions. In: Horizons in Biochemistry, ed. by Kasha, M. and Pullman, D., Academic Press, New York, 1962, pp. 251-264. Ryan, L.M., Cheung, H.S. and McCarty, D.J.: Release of pyrophosphate by normal mammalian articular hyaline and fibrocartilage in organ cultures. Arthr. Rheum. 24: 1522-1527,1981. Ryan, L.M., Wortmann, R.L., Karas, B. and McCarty, D.J.: Cartilage nucleoside triphosphate pyrophosphohydrolase. II. Role in extracellular pyrophosphate generation and nucleotide metabolism. Arthr. Rheum. 28: 413-418,1985. Prins, A., Kiljan, E., van de Stadt, R. and van der Korst, J.: Inorganic pyrophosphate release by rabbit articular chondrocytes in vitro. Arthr. Rheum. 29: 1485 -1492, 1986. Ryan, L.M., Kurup, I., McCarty, D.J. and Cheung, H.S.: Cartilage inorganic pyrophosphate elaboration is independent of glycosaminoglycan synthesis. Arthr. Rheum. 33: 235 - 240, 1990. Benya, P.D. and Nimni, M.E.: The stability of the collagen phenotype during stimulated collagen, glycosaminoglycan, and DNA synthesis by articular cartilage organ culture. Arch. Biochem. Biophys. 192: 327-335,1979. Prins, A.P.A., Kiljan, E., van de Stadt, R.J. and van der Korst, J. K.: Inorganic pyrophosphate release by rabbit articular chondrocytes in vitro. Arthr. Rheum. 29: 1485 -1492,1986. Cheung, C. P. and Suhadolnik, R.J.: Analysis of inorganic pyrophosphate at the picomole level. Anal. Biochem. 83: 61-63, 1977. Sugino, Y. and Miyoshi, Y.: The specific precipitation of orthophosphate and some biochemical applications. J. Bioi. Chem. 239: 2360-2364,1964. Rachow, J. W. and Ryan, L. M.: Partial characterization of synovial fluid nucleotide pyrophosphohydrolase. Arthr. Rheum. 28: 1377-1383,1985. Goldring, M. B., Sandell, L. J., Stephenson, M. L. and Krane, S. M.: Immune interferon suppresses levels of procollagen mRNA and type II collagen synthesis in cultured human articular and costal chondrocytes. J. Bioi. Chem. 261: 9049-9056, 1986.
Stimulation of Cartilage Inorganic Pyrophosphate Granstein, R.D., Murphy, G.F., Margolis, R.]., Byrne, M.H. and Amento, E. P.: Gamma-interferon inhibits collagen synthesis in vivo in the mouse. J. Clin. Invest. 79: 1254-1258, 1987. Mauviel, A., Daireaux, M., Redini, F., Galera, P., Loyau, G. and Pujol, ].-P.: Tumor necrosis factor inhibits collagen and fibronectin synthesis in human dermal fibroblasts. FEBS Lett. 236: 47-52,1988. Basu, S. K., Goldstein,]. L., Anderson, R. G. W. and Brown, M. S.: Monensin interrupts the recycling of low density lipoprotein receptors in human fibroblasts. Cell 24: 493-501,1981. Bhatnagar, R.S. and Prockop, D.].: Dissociation of the synthesis of sulphated mucopolysacchrides and the synthesis of collagen in embryonic cartilage. Biochem. Biophys. Acta 130: 383 - 392, 1966. Silcox, D. C. and McCarty, D.].: Elevated inorganic pyrophosphate levels in synovial fluid in osteoarthritis and pseudogout. J. Lab. Clin. Med. 83: 518-531,1974. Howell, D. S., Muniz, 0., Pita,]. C. and Enis, ]. E.: Extrusion of pyrophosphate into extracellular media by osteoarthritic cartilage incubates. J. Ciin. Invest. 56: 1473-1480, 1975.
281
Rosenthal, A. K., Cheung, H. S. and Ryan, L. M.: Augmentation of inorganic pyrophosphate elaboration in cartilage by serum factors. Arch. Biochem. Biophys. 272: 386-392,1989. Rosenthal, A.K., Cheung, H.S. and Ryan, L.M.: TGF beta and EGF increase pyrophosphate production by articular cartilage. Clin. Res. 38: 237A, 1990. Knowlton, R. G., Katzenstein, P. L., Moskowitz, R. W., Weaver, E.J., Malemud, c.]., Pathria, M.N., Jimenez, S.A. and Prockop, D.].: Genetic linkage of a polymorphism in the type II procollagen gene (COL2A1) to primary osteoarthritis associated with mild chondrodysplasia. New Eng. J. Med. 322: 526- 530, 1990. Ryan, L. M. and McCarty, D.].: Calcium pyrophosphate dihydrate crystal deposition disease; pseudogollt; articular chondrocalcinosis, In: Arthritis and Allied Conditions, ed. by McCarty, D.]., Lea and Febiger, Philadelphia, 1988. L. Ryan, M. D., Professor and Chief of Rheumatology, Medical College of Wisconsin, 8100 W. Wisconsin Avenue, Milwaukee, WI 53226, USA.
Stimulation of Cartilage Inorganic Pyrophosphate Elaboration by Ascorbate LAWRENCE M. RYAN, INDIRA KURUP and HERMAN S. CHEUNG Division of Rheumatology, Department of Medicine, Medical College of Wisconsin, Milwaukee, WI, 53226, USA.
Abstract
Ascorbate (0- 500 !tM) stimulates synthesis and secretion of collagen by cartilage explants in a dose-dependent fashion as estimated by eH] proline incorporation. Concurrent elaboration of inorganic pyrophosphate (PPi) parallels [3H] proline incorporation « 0.001, Wilcoxon rank sum). The effect on proline and PPi is abolished by ascorbate oxidase. Another reducing agent, Vitamin E, did not promote PPi accumulation about cartilage. Inhibitors of collagen synthesis, including monensin, tumor necrosis factor a, and 2' ,2' dipyridyl also inhibited PPi elaboration. Cycloheximide (1 !tg/ml) inhibited the ascorbate stimulated PPi elaboration 54% but did not attenuate the secretion of PPi by unstimulated cartilage. Cosecretion of collagen and PPi by chondrocytes may explain these results. Moreover, at least two pathways exist for PPi elaboration, a cycloheximide sensitive path and another independent of new protein synthesis. Key words: ascorbate, cartilage, inorganic pyrophosphate.
Introduction
Cartilages of patients with articular chondrocalcinosis contain calcium pyrophosphate dihydrate (CPPD) crystals. CPPD crystallogenesis, an extracellular event, requires extracellular inorganic pyrophosphate (PPi). Such PPi might form extracellularly. For example, the chondrocyte ectoenzyme, nucleoside triphosphate pyrophosphohydrolase (NTPPPH), generates PPi and nucleoside monophosphate from nucleoside triphosphates (NTP). (Ryan et a1., 1984; Howell et a1., 1984) Potential sources of substrate for this ectoenzyme include synovial fluid NTP or NTP released from stimulated, damaged, or traumatized chondrocytes. Alternatively, PPi made intracellularly may traverse the cell membrane to reach the site of crystal formation. Intracellular PPi is a byproduct and regulator of multiple biosynthetic reactions, a concept introduced by
Kornberg (1962). We have postulated cosecretion of matrix-destined molecules with their synthetic coproduct, PPi (Ryan et a1., 1981; 1985). In cartilage collagen and glycosaminoglycan are major extracellular matrix constituents. The coexport thesis has been disproved for sulfated glycosaminoglycan, secretion of which is neither sufficient nor necessary for the release of PPi by cartilage organ cultures (Prins et a1., 1986; Ryan et a1., 1990). In order to further test the cosecretion hypothesis, we have studied the accumulation of extracellular PPi about cartilage explants as a function of secretion of the other major matrix constituent, collagen. We report that ascorbate (asc) stimulation of collagen production and secretion by cartilage cells is accompanied by increased PPi elaboration. When collagen synthesis and secretion is inhibited PPi accumulation around cartilage explants decreases. Materials and Methods
Presented in part at the 53rd Annual Scientific Meeting, American College of Rheuma tology, Cincinnati, 0 H, June, 1989.
Dulbecco's modified Eagle's medium (DMEM), fetal calf serum (FCS), Pen-Strep-Fungizone (PSF), and N-(2-
Stimulation of Cartilage Inorganic Pyrophosphate hydroxyethyl) piperazine' -(2-ethanesulfonic acid) (HEPES) buffer were obtained from Grand Island Biological. eH] proline was from Amersham. [14C] adenine, [32p] PPi, and 35 [ S] S04 were from New England Nuclear. L- and D-asc were from Kodak Laboratories. Asc oxidase, vitamin E (Da-tocopherol acetate), cycloheximide, 2,2-'dipyridyl, and monensin were from Sigma. Human recombinant gamma interferon was supplied by Collaborative Research. Human recombinant tumor necrosis factor a (TNFa) was from R and D Systems. Chromatographically purified Clostridial collagenase was from Worthington. Minced cartilage from freshly slaughtered pig distal femur was cultured in DMEM supplemented with 10% FCS and 1 % PSF for 24 h. Then cultures were incubated for 24 h. in HEPES-buffered DMEM without FCS before replacing with experimental media. Medium was then changed to DMEM with 10% FCS heat-inactivated (HI) (at 56°C for 1 h to remove pyrophosphatase activity) containing 1 % PSF, and 50 mM HEPES-buffer, pH 7.4 and additives being tested. All cultures were performed at 37°C in 10% CO 2 atmosphere. [3H] proline incorporation into collagen was determined after incubation of cartilage with 20 f!Ci/ml in a 20-fold (vol!wet weight) volume of 50 mM HEPES-buffered DMEM containing HI FCS and PSF by the a modified method of Benya and Nimni (1979). Briefly, after labeling, media and washes were adjusted to 30% saturation with ammonium sulfate, incubated at 4 °C for 18 h, and precipitate collected. Precipitate was dissolved in 0.5 M acetic acid, adjusted to 100 f!glml pepsin, and incubated for 18 h at 4 dc. The pepsin digest was dialyzed against 2 M NaCI at 4°C to reprecipitate collagen. The precipitated collagen was collected by centrifugation, dissolved in 0.5 M acetic acid and counted. The collagen nature of this preparation was demonstrated by susceptibility to digestion with protease free specific collagenase. [3H] proline incorporated into explant collagen was partly pepsin extractable and partly resistant to pepsin extraction. Pepsin extractable [3H] proline-labelled collagen was quantified by incubating cartilage with 1 mglml pepsin in 0.5 M acetic acid for 48 h. Pepsin extract was removed and cartilage further extracted for 24 h with 50 mM Tris-CI in 1 M NaCI at pH 7.5. The two extracts were combined, neutralized, and collagen precipitated by dialysis against 4.4 M NaCl at 4°C overnight. Precipitate was dissolved in 0.5 M acetic acid, clarified by centrifugation, and counted. These pepsin-extracted counts were collagenase sensitive. Residual eH] proline-counts in the extracted cartilage were also 96-98% collagenase sensitive. Therefore, the pepsin resistant and collagenase sensitive counts were also measured after hydrolysis by incubation with 6N HCI at 108°C for 1 h. Reported values for [3H] proline incorporation into medium and tissue collagen are the sum of all three compartments, namely, media, cartilage extract, and cartilage
277
hydrolysate. In short term organ culture most collagen synthesized by hyaline articular cartilage is type II (Benya and Nimni, 1979). In 3 organ cultures we determined that the increase in [3H] proline incorporation in the presence of asc was mostly (73, 80 and 87%) into collagenase sensitive protein. 35 [ S] S04 incorporation as a reflection of sulfated glycosaminoglycan synthesis and secretion was measured as described in detail elsewhere (Prins et aI., 1986). PPi was measured by the radiometric UDPG pyrophosphorylase method as previously described (Cheung and Suhadolnik, 1977). All standard curves were prepared in the medium being tested, since various additives altered the slope of the standard curve. Determination of hydrolysis of trace [32 p] PPi in medium was by an adaptation of the method of Sugini and Miyosho (Sugino and Miyoshi, 1964). Ecto-NTPPPH was assayed by incubation of cartilage with the artificial substrate p-nitrophenyl thymidine monophosphate (PNTMP) (Rachow and Ryan, 1985). One mM PNTMP in Hank's balanced salt solution buffered with 25 mM HEPES to pH 7.4 was prepared and a 20-fold (vol! wt) volume placed over cartilage. The reaction was stopped after 1 h by adding a 4-fold volume of 0.1 N NaOH and concentration of p-nitrophenyl generated read at OD 41O . Results Asc increases PPi elaboration by cartilage. When 50 f!M L-asc was included in medium, accumulation of PPi increased 97% (medium PPi concentration 5.8 ± 0.5 SEM f!M without asc, 11.4 ± 1.1 f!M with asc, n = 8) (p = 0.01, Wilcoxon signed rank test) after 48 h (TableI). Cartilage from each of the animals tested elaborated more PPi with asc than without. Consistently elevated [PPi] was evident at 24 h and persisted through at least 96 h. Excess PPi accumulation was not caused by decreased hydrolysis of extracellular PPi, since trace [32 p] PPi added to the medium was hydrolysed similarly in the asc-containing and -lacking cultures (data not shown). D- and L-asc were similarly effective PPI stimulants (Fig. 1). The effect of asc was dose dependent up to 500 f!M (Fig. 1). Asc oxidase 0.5 U/ml prevented the enhanced PPi accumulation by 100 f!M D- or Lasc (medium [PPi] at 48 h being 8.9 f!M for control with oxidase, 11.5 for D-asc, 8.6 D-asc plus oxidase, 14.6 for Lasc, and 9.9 for L-asc plus oxidase, n = 3 for each group). Since vitamin C is a reducing agent, altered PPi metabolism might be nonspecific, related to redox characteristics. The effect of another reductant, vitamin E was studied. At concentrations of 200 and 500 f!glml tocopherol acetate did not promote medium PPi accumulation compared to control cultures (data not shown) containing identical concentrations of acetate and alcohol (tocopherol solvent). To determine whether matrix molecule synthesis was
278
L. M. Ryan et al.
Table I. Effect of I.-Ascorbate Treatment on Cartilage Medium Inorganic Pyrophosphate Concentration at 48 Hours.
30,---------------------------~
Medium [Pyrophosphate) ([!M) - ascorbate + ascorbate
6.1a 5.1
10.4 12.6 13.8 16.6 11.9 9.6 8.0 8.0
8.9
6.4 5.5 4.5 4.7 5.0 a
a
20
iii a
10 !p
20~--------------------------'
C!I D-ASC
25uM
SOuM
Fig. 1. Medium inorganic pyrophosphate concentration after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) wolume of Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal calf serum and varying concentrations of ascorbate. PPi = inorganic pyrophosphate L-ASC = L-ascorbate D-ASC = D-ascorbate Error bars represent standard error of the mean, n = 3.
increased by asc, glycosaminoglycan and collagen metabolism were studied. Asc did not significantly increase 35 [ S]S04 incorporation into cartilage matrix or medium at 24 (n = 6) or 48 (n = 4) hours (data not shown). As expected L-asc increased [3H] proline incorporation by 169 ± 9% (± SEM, n = 11). D-asc had comparable effect. And ascorbate oxidase blocked the increases in proline incorporation and PPi elaboration. In 11-matched specimens PPi elaboration was compared to proline incorporation. PPi production increased 91 ± 10% (± SEM) while proline incorporation was enhanced 169 ± 9%. As for PPi accumulation, so also proline incorporation was dependent on L-asc concentration. Proline incorporation and PPi elaboration in 25 cartilages (11 untreated, 11 L-asc treated, and 3 D-asc treated) were significantly correlated (Fig. 2) (p < 0.001, Spearman's rank correlation).
aa a
a a
a
I
20
10
Each value represents the mean of triplicate determinations on cartilages from different animals.
a
a cP
rf
a aa aa
a
40
30
50
3H-Proline, cpm/ug Protein
Fig. 2. Medium inorganic pyrophosphate concentration versus [3Hl proline incorporation after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) volume of Dulbecco's modified Eagle's medium supplemented with 10'10 heat-inactivated fetal calf serum, with or without ascorbate. rs = 0.8, p < 0.001 (Spearman's rank correlation).
To further study the relationship between collagen production and PPi elaboration we determined the effect on medium [PPi] of inhibiting collagen synthesis and secretion using cycloheximide, gamma interferon, monensin, TNFa, and 2,2' dipyridyl. A nonselective protein synthesis inhibitor, cycloheximide (1 [lg/ml), inhibited proline incorporation by asc-treated cartilages by 88%, glycine incorporation by 97% and sulfation by 99%. Simultaneous medium PPi concentration after 48 hours was 54% less than cultures without cycloheximide. The cartilage
20~-------------------------------,
seRUM
ASC
CHX
t
t
+
t
+
t
5
50
t
+
t t 5
t t 50
Fig. 3. Medium inorganic pyrophosphate concentration after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) volume of Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal calf serum and ascorbate (50 [!glml) or cycloheximide (0- 50 [!glml). PPi = inorganic pyrophosphate ASC = L-ascorbate CHX = cycloheximide concentration Error bars represent standard error of the mean, n = 3.
Stimulation of Cartilage Inorganic Pyrophosphate remained viable as demonstrated by the ability of explants to recover proline incorporation when removed from the cycloheximide. Interestingly, cycloheximide inhibited only the increment in PPi elaboration induced by asc, but did not inhibit the basal secretion of PPi by unstimulated cartilages (Fig. 3). Interferon-gamma reportedly diminishes collagen production by cultured fibroblasts and chondrocytes (Goldring et aI., 1986; Granstein et aI., 1987). But in porcine cartilage human recombinant interferon-gamma failed to suppress either PPi production or proline incorporation (data not shown). Moreover, human gamma interferon did not reduce collagen secretion by porcine fibroblasts, suggesting that species differences account for the inefficacy. TNFa suppresses collagen release by fibroblasts (Mauviel et aI., 1988). Human recombinant TNFa decreased media PPi concentrations and proline incorporation in asc-stimulated cartilage organ cultures (Table II). Since TNFa also
279
stimulates production of prostaglandin E2 , we determined whether indomethacin (10 ILM) would abrogate the TNFa inhibition. Indomethacin did not reverse the inhibition of PPi elaboration or eH) proline incorporation (data not shown). The monovalent cation ionophore, monensin, in a dose dependent fashion decreased medium [PPi] and proline incorporation (Fig. 4). Since monensin may interfere with cycling of ectoenzymes, (Basu et aI., 1981) including NTPPPH which generates extracellular PPi, we determined the ecto-NTPPPH activities of cartilages after a 48-hour monensin incubation. Ecto-NTPPPH was unaltered by monensin (data not shown). Finally, the relatively specific inhibitor of collagen production, 2,2' dipyridyl (Bhatnagar and Prockop, 1966), decreased PPi accumulation and proline incorporation by asc-stimulated cultures (Fig. 5). 2o.-------------------------~20
DpPI
o
3H Proline
ASC
SuM
Table II. Inorganic Pyrophosphate in Cartilage Medium and eH)Proline Incorporation by Cartilage Treated with Tumor Necrosis Factor a. TNFa (ng/m!)
PPi ([tM) 48h
0 1 10 100
12.1 10.5 8.6 10.3
± ± ± ±
2.0 a 2.2 1.6 2.4
PPi ([tM) 72h
[3H)-proline (cpm/mg· 103 ) 48 h
16.1±2.1 11.9 ± 2.2 7.4 ± 1.2 4.8 ± 1.2
19.0 ± 14.1 ± 9.0 ± 7.4 ±
0.3 0.2 0.4 0.3
Control
All values are x ± SEM, n = 3. Porcine articular cartilage treated with recombinant human tumor necrosis factor was incubated in asc-containing (50 [tg/ml) medium and ambient [PPi) measured at 48 and 72 hours. Collagen production was estimated by incorporation of [-'Hl-proline.
a
~------------------------~~ 3H Proline
20
l
i' n
SOuM
SOOuM
Fig. 5. Medium inorganic pyrophosphate concentration and eHl proline incorporation after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) volume of Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal calf serum. All cultures except control contain 50 [tg/ml ascorbate and 0-500 [tM 2,2'dipyridyl. PPi = inorganic pyrophosphate Error bars represent standard error of the mean, n = 3.
Discussion ~
r
.....
1. . . . s
.8 §.
10 ~ ......
f o Fig. 4. Medium inorganic pyrophosphate concentration and eH) proline incorporation after 48 hours of porcine hyaline articular cartilage culture in a 20-fold (vol/weight) volume of Dulbecco's modified Eagle's medium supplemented with 10% heat-inaccivated fetal calf serum, ascorbate (50 [tg/ml), and monensin (0-10 [tM). PPi = inorganic pyrophosphate Error bars represent standard error of the mean, n = 3.
Since CPPD crystals form interstitially, extracellular PPi is essential. An extra-articular source of cartilage PPi seems unlikely in view of the consistent downgradient in PPi concentration between synovial fluid and plasma (Silcox and McCarty, 1974). Among potential intra-articular sources, the chondrocyte alone elaborates extracellular PPi (Ryan et aI., 1981; Howell et aI., 1975). But PPi synthesized within the cell diffuses across biomembranes poorly; thus, simple diffusion could not explain the appearance of intracellularly generated PPi in the cartilage matrix. We have postulated previously that matrix molecules and PPi formed as their synthetic coproduct might be coexported. This hypothesis has been disproved for one prominent matrix component, glycosaminoglycan (Ryan et aI., 1990). Our present findings suggest a link between PPi elaboration and
280
L. M. Ryan et al.
production of the other major cartilage matrix molecule, collagen. When collagen secretion was stimulated by asc, PPi elaboration increased. Both responses were dose-dependent and significantly correlated. The effects of asc were not likely due to any contaminant in the vitamin preparation, since collagen and PPi accrual were both blocked by treatment with asc oxidase which generates biologically active but unstable asc derivatives. Extracellular PPi may accrue due to other asc influences. Asc is a nonspecific reducing agent. Thus it was important to demonstrate that another reductant, a-tocopherol, had no similar impact on medium PPi concentration or proline incorporation. If collagen production and PPi liberation are linked, then inhibiting collagen synthesis should decrease extracellular PPi elaboration. Unfortunately all inhibitors have some degree of nonspecificity and results must be interpreted with that in mind. Nonetheless, treating asc-stimulated cartilage with collagen synthesis or secretion inhibitors decreased proline incorporation and PPi elaboration in parallel. Additionally, these inhibitors act at different levels of collagen processing. For example cycloheximide is a classic translation inhibitor, while 2,2-dipyridyl probably acts by inhibiting iron dependent posttranslational modification of pro collagen (Bhatnagar et al., 1966). Each decreases collagen secretion and both diminish PPi elaboration. Most interesting was the observation that basal production of PPi by unstimulated cells is not diminished by cycloheximide, while the increment in PPi elaboration following asc stimulation is abolished by cycloheximide. Thus, at least two pathways for PPi extrusion, one protein synthesisdependent and one independent of new protein synthesis, coexist in cartilage. The latter basal production is reportedly abolished by treating chondrocytes with sufficient trypsin to remove ecto-NTPPPH activity (Prins et al., 1986). This component may be generated by ectoenzymes which salvage nucleoside triphosphates that have leaked from cells. The cycloheximide-sensitive pathway is dependent on new protein synthesis and may represent cosecretion of PPi with collagen. Although asc stimulates collagen production by cartilage, it may have other unrelated and unexamined effects which are responsible for PPi secretion. For instance, we have reported that growth factors stimulate PPi elaboration by cartilage (Rosenthal et al., 1989; 1990), although the intracellular signalling pathway is uncertain. Asc may stimulate PPi secretion through these pathways. Clinical association of osteoarthritis and CPPD deposition has been long recognized. Recent reports associate premature and severe osteoarthritis and/or dysplasia transmitted in an autosomal dominant pattern with a specific allele in restriction fragment length polymorphism for the type II procollagen gene (Knowlton et al., 1990). Analysis for similar genomic type II collagen abnormalities in CPPD deposition associated with degenerative disease or in those
kindreds with autosomal dominant transmission of CPPD deposition (Ryan and McCarty, 1988) may shed light on the pathogenesis of this disease. Acknowledgements This work is supported in part by grant from the Wisconsin Chapter of the Arthritis Foundation, by a Biomedical Research Center Grant from the Arthritis Foundation, and by Public Health Service Grants AR 38656 and AR 38421 from NIH. Dr. Ryan is recipient of Research Career Development Award from the National Institute of Health. References Ryan, L.M., Wortmann, R.L., Karas, B. and McCarty, D.J.: Cartilage nucleoside triphosphate (NTP) pyrophosphohydrolase I. Identification as an ecto-enzyme. Arthr. Rheum. 27: 404-409, 1984. Howell, D. S., Pelletier, J. P., Martel-Pelletier, J., Morales, S. and Muniz, 0.: Nucleoside triphosphate (NTP) pyrophosphohydrolase in chondrocalcinotic and osteoarthritic cartilages. II. Further studies on histologic and subcellular distribution. Arthr. Rheum. 27: 193-199, 1984. Kornberg, A.: On the metabolic significance of phosphorolytic and pyrophosphorolytic reactions. In: Horizons in Biochemistry, ed. by Kasha, M. and Pullman, D., Academic Press, New York, 1962, pp. 251-264. Ryan, L.M., Cheung, H.S. and McCarty, D.J.: Release of pyrophosphate by normal mammalian articular hyaline and fibrocartilage in organ cultures. Arthr. Rheum. 24: 1522-1527,1981. Ryan, L.M., Wortmann, R.L., Karas, B. and McCarty, D.J.: Cartilage nucleoside triphosphate pyrophosphohydrolase. II. Role in extracellular pyrophosphate generation and nucleotide metabolism. Arthr. Rheum. 28: 413-418,1985. Prins, A., Kiljan, E., van de Stadt, R. and van der Korst, J.: Inorganic pyrophosphate release by rabbit articular chondrocytes in vitro. Arthr. Rheum. 29: 1485 -1492, 1986. Ryan, L.M., Kurup, I., McCarty, D.J. and Cheung, H.S.: Cartilage inorganic pyrophosphate elaboration is independent of glycosaminoglycan synthesis. Arthr. Rheum. 33: 235 - 240, 1990. Benya, P.D. and Nimni, M.E.: The stability of the collagen phenotype during stimulated collagen, glycosaminoglycan, and DNA synthesis by articular cartilage organ culture. Arch. Biochem. Biophys. 192: 327-335,1979. Prins, A.P.A., Kiljan, E., van de Stadt, R.J. and van der Korst, J. K.: Inorganic pyrophosphate release by rabbit articular chondrocytes in vitro. Arthr. Rheum. 29: 1485 -1492,1986. Cheung, C. P. and Suhadolnik, R.J.: Analysis of inorganic pyrophosphate at the picomole level. Anal. Biochem. 83: 61-63, 1977. Sugino, Y. and Miyoshi, Y.: The specific precipitation of orthophosphate and some biochemical applications. J. Bioi. Chem. 239: 2360-2364,1964. Rachow, J. W. and Ryan, L. M.: Partial characterization of synovial fluid nucleotide pyrophosphohydrolase. Arthr. Rheum. 28: 1377-1383,1985. Goldring, M. B., Sandell, L. J., Stephenson, M. L. and Krane, S. M.: Immune interferon suppresses levels of procollagen mRNA and type II collagen synthesis in cultured human articular and costal chondrocytes. J. Bioi. Chem. 261: 9049-9056, 1986.
Stimulation of Cartilage Inorganic Pyrophosphate Granstein, R.D., Murphy, G.F., Margolis, R.]., Byrne, M.H. and Amento, E. P.: Gamma-interferon inhibits collagen synthesis in vivo in the mouse. J. Clin. Invest. 79: 1254-1258, 1987. Mauviel, A., Daireaux, M., Redini, F., Galera, P., Loyau, G. and Pujol, ].-P.: Tumor necrosis factor inhibits collagen and fibronectin synthesis in human dermal fibroblasts. FEBS Lett. 236: 47-52,1988. Basu, S. K., Goldstein,]. L., Anderson, R. G. W. and Brown, M. S.: Monensin interrupts the recycling of low density lipoprotein receptors in human fibroblasts. Cell 24: 493-501,1981. Bhatnagar, R.S. and Prockop, D.].: Dissociation of the synthesis of sulphated mucopolysacchrides and the synthesis of collagen in embryonic cartilage. Biochem. Biophys. Acta 130: 383 - 392, 1966. Silcox, D. C. and McCarty, D.].: Elevated inorganic pyrophosphate levels in synovial fluid in osteoarthritis and pseudogout. J. Lab. Clin. Med. 83: 518-531,1974. Howell, D. S., Muniz, 0., Pita,]. C. and Enis, ]. E.: Extrusion of pyrophosphate into extracellular media by osteoarthritic cartilage incubates. J. Ciin. Invest. 56: 1473-1480, 1975.
281
Rosenthal, A. K., Cheung, H. S. and Ryan, L. M.: Augmentation of inorganic pyrophosphate elaboration in cartilage by serum factors. Arch. Biochem. Biophys. 272: 386-392,1989. Rosenthal, A.K., Cheung, H.S. and Ryan, L.M.: TGF beta and EGF increase pyrophosphate production by articular cartilage. Clin. Res. 38: 237A, 1990. Knowlton, R. G., Katzenstein, P. L., Moskowitz, R. W., Weaver, E.J., Malemud, c.]., Pathria, M.N., Jimenez, S.A. and Prockop, D.].: Genetic linkage of a polymorphism in the type II procollagen gene (COL2A1) to primary osteoarthritis associated with mild chondrodysplasia. New Eng. J. Med. 322: 526- 530, 1990. Ryan, L. M. and McCarty, D.].: Calcium pyrophosphate dihydrate crystal deposition disease; pseudogollt; articular chondrocalcinosis, In: Arthritis and Allied Conditions, ed. by McCarty, D.]., Lea and Febiger, Philadelphia, 1988. L. Ryan, M. D., Professor and Chief of Rheumatology, Medical College of Wisconsin, 8100 W. Wisconsin Avenue, Milwaukee, WI 53226, USA.
