Lubrication of animal joints. iii. the effect of certain chemical alterations of the cartilage and lubricantкод для вставкиСкачать
Lubrication of Animal Joints. 111. The Effect of Certain Chemical Alterations of the Cartilage and Lubricant By FRANK C. LI" The coefficient of friction of dog ankle joints measured in oitro was found to be greatly affected by certain enzymatic treatments of bovine synovial mucin. Testicular hyaluronidase had no effect on its lubricating action even though it abolished the viscosity. By contrast, tryptic digestion destroyed the ability of the mucin to lubricate without reducing the viscosity. These observations indicate that a protein moiety is an intrinsic component of the mucin molecule; and that, together with an only partly polymerized hyaluronate, it is responsible for the mucin's function as a lubricant. The protein may serve as a prosthetic group for adsorption of the mucin onto the surface of the cartilage. Sodium heparin inhib- U AND ERICL. RADIN ited lubrication by the m u c h In mucinfree buffers, the pH and molarity of the lubricant bath altered both the friction and deformability of the cartilage. The change in friction was not consistently related to the change in deformability. The amphoteric effect of pH on the friction raises the possibility that electrostatic surface effects contribute to adhesive friction. Formalin made the cartilage rigid and increased the coefficient of friction. It did not, however, have much effect on the lubricating action of the mucin. A small number of pathological human synovial fluids were studied and found to lubricate as well or better than healthy bovine material. the lubrication of animal joints was believed to be hydrodynamic and dependent on the viscosity of synovial fluid. It has now been established that a combination of hydrostatic and boundary mechanisms is mainly responsible for the low friction.1-5Joints are slippery even when lubricated with saline solutions, but the addition of synovial much lowers the coefficient of friction by about 45 per cent.5 McCutchen4 and L i n d have shown that when the viscosity of synovial fluid is reduced to that of water by digestion with hyaluronidase, it loses little or none of this lubricating advantage. This paper deals with certain chemical From the Section on Rheumatic Diseases, Laboratory of Experimental Pathology, National Institute of Arthritis and Metabolic Diseases. Requests for reprints should be addressed to Section on Rheumatic Diseases, 10:3N-114, N.I.A.M.D., Bethesda, Md.20014. FRANK C. LINN,B.A. IN M.E.: Research Engineer (retired}, Section on Rheumatic Diseases, Laboratory of Experimental Pathology, National I n stitute of Arthritis and Metabolic Diseases, National Znstitutes of Health, Bethesda, Maryland 20014. ERIC L. RADIN, M.D.: Formerly Captain 674 NTIL RECENTLY U.S.A.F., M.C., Malcolm Grow Clinical Center, Andrew Air Force Base, Maryland; Guest Inoestigator, Section on Rheumatic Diseases, Laboratory of Experimental Pathology, National Institute of Arthritis and Metabolic Diseases, National Znstitutes of Health, Bethesda, Maryland. Presently Instructor in Orthopedic Surgery, Haruard Medical School. Address: Massachusetts General Ilo.cpita1, Boston, Mass. 02114. ARTHRITISAND RHEUMATISM, VOL.11, No. 5 (OCTOBER 1968) 675 LUBRICATION OF ANIMAL JOINTS. 111 Table 1.-E#eet Lxge8tion: Ezperiment no.: Lubricant Buffer Control mucin Buffer Digested mucin Buffer of Terticular Hyaluronidaae Digation on Lubrication by Synovial M u c h Mild (3 hours, 23"C, p H 7.2) Exhaustive (24 hours, 37"C, p H 5.5) 1 2 3 4 Coef. of Relative friction viscosity Coef. of Relative friction viscosity Coef. of Relative friction viscosity Coef. of Relative friction viscosity .0077* .0052 .0076 .0056 .0076 .0083 2.38 .o040 1.00 .0082 ,0041 .0086 Uronic acid (micmgrams/ml. ) Acetylglucostlmine (micrograms/ml.) .0061 3.27 .oO40 1.15 .0062 .0057 ,0059 .0063 3.50 .0035 3.31 1.05 .0061 .0043 .0065 1.07 240 245 240 265 *A different ankle was used in each experiment. alterations of articular cartilage and mucin which further elucidate their particular contributions to joint lubrication. MATERIALS AND METHODS Lubrication was tested by measuring in uitro the coefficient of friction of dog ankles during continuous oscillation under constant load. The apparatus, techniques for conducting the tests and analyzing the results have been established and a high degree of reproducibility has been obtained.0 Thc friction tests were conducted at 23 C with a load of 40 pounds at 40 cycles per minute. Deformation of the joint cartilage also was measured by previously described techniques.0 Bovine synovial fluid was obtained from the hocks of freshly slaughtered calves. Cells were removed by centrifugation in the cold at 6000 rpm for 20 minutes. The fluid was then stored at 4 C until just before testing. A mucin fraction was prepared by filtering the centrifuged fluid through a 220 mp filter (Millipore Filter Corp., Bedford, Mass.) and redissolving it in cold buffer. Because phosphate buffer is known to cause slow degradation of synovial mucin,' 0.155 M veronate buffer (.088M NaCl, .00408M Na 5-5 diethylbarbiturate, .0625M HCI, pH 7.2) was employed. Friction measurements were carried out with veronate buffer before and after each test to check the reliability of the cartilage and measuring devices. After each test, the surfaces were washed with buffer and wiped clean with tissue paper. The viscosity of the lubricants relative to water was measured by a capillary flow method ancl pFI with a glass electrode. The effects of p H and ionic strength were determined by using a series of buffers as lubricants without synovial mucin. The tissues were equilibrated with each buffer by oscillating them under load for 7 minutes and then replacing the buffer with a second aliquot of the same fluid. Synovial fluid was subjected to mild and to exhaustive digestion with testicular hyaluronidase (Worthington Biochemical Corp., Freehold, N.J., HSE 7BA) in concentrations of 75 USP units per ml. The mild conditions were a digestion time of 3 hours as 23 C, p H 7.2; the exhaustive conditions were 24 hours at 37 C, pH 5.5. Completeness of the digestion was determined by measuring the uronic acids and acetylglucosamineg before and after treatment with the enzyme. At the suggestion of Dr. David Hamerman, the mucin was prevented from clotting at pH 5.5 by first dialyzing it against 0.5 M NaCl solution. Before testing, the pH was restored to 7.2 and the molarity to .155 by dialyzing against veronate buffer. Streptococcal hyaluronidase (Wyeth Corp., Philadelphia, Pa.) was found unsuitable for these experiments because it acted on the cartilage and itself alterecl the friction characteristics. Tryptic digestion of the synovial mucin was carried out with bovine pancreatic trypsin (Worthington Biochemical Corp., Freehold, N.J., TRL 7AA) in a concentration of 60 turbidity reducing units per ml. for 72 hours at 4 C, pH 7.2. The cartilage was protected from tryptic digestion by addition of an excess of soybean -sin inhibitor (Worthington Biochemical Corp., SI 61B) to the synovial digestate. Sodium heparin, without added preservatives, was obtained from two independent sources (Up- 676 LI" AND W I N Table t.-Efecr of Tryptic Digestion on Lubrication b y Synovial Mucin Experiment m.: Lubricant 2 1 Coef. of friction Relative viscosity Coef. of friction 3 Relative viscosity Coef. of friction Relative viscosity 2.37 .0056 .0031 .0052 2.49 ~~ Buffer Mucin and inhibitor Buffer Digested much and inhibitor Buffer .0103* .0059 .0098 .0089 .0100 3.18 3.19 .0053 .0028 .0055 .0054 .0053 2.38 .0048 .0054 2.49 *A different ankle w&s used in each experiment. range of 3 to 9.5 in .155 M acetate and veronate buffers. When acetate solutions were employed, the value was quite constant between pH 6.2 and 8 (Fig. 1).Thc RESULTS coefficient of friction was highest at about Enzymatic digestion of mucin. Partial di- pH 5 and dropped perceptibly above and gestion of whole synovial fluid or its mucin below this hydrogen ion level. There was fraction by testicular hyaluronidase, suffi- a slight but progressive decrease in the cient to reduce the viscosity to that of the friction obtained with veronate buffers bebuffer sohtion, had no effect on their ability tween pH 6.5 and 9.5. Differing dissociation to lubricate the cartilage (Table 1). Com- constants of the two buffers may account plete digestion of synovial mucin, though for slight variation in the friction values not of whole synovial fluid, increased the of comparable pH. Deformation of the cartilage, as it was coefficient of friction appreciably but it still compressed, increased when pH fell below was considerably below values obtained 5.5. Above this value, it was constant (Fig. with the control buffer solution. Protein was 1). This relationship held both in tests constill precipitable with 10 per cent trichlorducted after the joint was loaded with 40 acetic acid in this exhaustively digested pounds for 7 minutes or with only a tare mucin. load. Tryptic digestion of synovial mucin did When the pH was kept constant at 7.2 not alter its viscosity but consistently caused varying the molarity of acetate buffer over the coefficient of friction to rise to values a range of .002 to 1.0 also affected the close to those obtained with the buffer solutions (Table 2). This effect was not ob- coefficient of friction. It was lowest at conserved when whole synovial fluid rather centrations of .4 to .6 moles per liter. Dethan mucin was subjected to digestion. formation of the cartilage increased with Soybean inhibitor, by itself or added to the molarity (Fig. 2). The changes in friction and deformability synovial fluid or mucin, had no effect on of the cartilage produced by varying the the friction. pH and molarity were reversible. Fresh frozen citrated plasma and fresh Znhibitory efect of heparin. Heparin, in serum had no lubricating advantage over concentration of at least 18 USP units per buffer solution. Effect of p H and molarity. The coeffi- ml., decreased the lubricating effectiveness cient of friction varied with pH over n of whole synovial fluid or its mucin. This john Co., Kalamazoo, Mich., and Mann Research Laboratories, New York, N.Y.). Both preparations behaved identically and no distinction is made between them in the results. 677 LUBRICATION OF ANIMAL JOINTS. 111 0 HCf - Veronof No Buffer Aceioie Buffer E .OlO 0 rn n .008 2 ,006 on 0 rn z -I 1 ; ob: b b i .05 I I : 1, f d ,bl 0 VH Fig. 1.-Effect of pH on coefficient of frictionand deformation of cartilage. inhibitory effect was incomplete but related novial fluid as lubricants (Fig. 4). Synovial to the amount of heparin (Fig. 3). Even at fluid still retained, however, considerable a level of 290 units per ml., the average lubricating advantage under these condimaximal reduction was never greater than tions. Lubricating ability of human synovial 20 per cent of the lubricating advantage. Effect of formalin fixation of cartilage. fluids. Synovial fluid from the knees of 2 Fixation of the cartilage in 10 per cent patients with rheumatoid arthritis and 2 neutral formalin for 7 days greatly reduced with traumatic effusions secondary to a torn the deformation of the cartilage. It also re- medical meniscus lubricated with as low or sulted in a several-fold increase in the co- lower coe5cients of friction than did the efficient of friction, both with buffer or sy- control bovine synovial fluid (Table 3). -.020 .020 - . o.ole le ; n 2 g z -I - ,016 ,016 2 a n LL 0 I) 2 .05 -.014 ,014 zi 4 -I I I a I I I 0.3 0.4 0.5 a 8 4 015 0.6 d6 0.7 017 o(8 CONCENTRATION, Moles per liter ACETATE BUFFER 8 - .01 .012 2 I 1 0.9 1.0 It0 Fig. 2.-Effect of molarity on coefficient of friction and deformation of cartilage. LI" AND W I N .010 Saline t Heparin Synovia t Heparin w 0 8 'i,,, 0 100 50 150 250 200 300 HEPARIN, U. S. I? UNITS / M L Fig. 3.-Inhibition of synovial fluid lubrication by heparin. The values obtained for undiluted synovia remained constant throughout the test. The viscosity of the human fluids was higher than that of the bovine. Efect of diluting synovial fluid. Bovine synovial fluid retained its lubricating effect until it was diluted with more than 3 parts of veronate buffer despite the fact that the relative viscosity decreased ( Fig. 5 ) . DISCUSSION The present experiments show that the friction of joints is greatly affected by certain enzymatic treatments or addition of heparin to synovial mucin; by the pH and molarity of the lubricating bath; and by formalin fixation of the articular cartilage. The lubricating advantage of synovial Table 3.-Lubricating Patient: Diagnosis: Lubricant Buffer Patient's synovial fluid Buffer Bovine synovial fluid Buffer 32 yr. male Torn medial meniscus mucin is not related to its viscosity. Digestion of the mucin by testicular hyaluronidase abolished the high viscosity but did not reduce its lubricating ability. By contrast, tryptic digestion, although not affecting the viscosity, completely destroyed the lubricant action. This effect of trypsin was observed on isolated synovial mucin although not on whole synovial fluid. The explanation undoubtedly has to do with the anti-tryptic activity of the synovial fluid.1° Opinion is divided as to whether protein is an integral part of the mucin molecule or a contaminant.ll Trypsin together with chymotrypsin removes 65 per cent of pro- Action of Human Synodal Fluid8 Coef. of Relative Coef. of Relative friction viscosity friction viscosity ,0077* ,0026 .0077 ,0043 ,0078 61.40 5.11 46 yr. female Rheumatoid arthritis 51 yr. male Rheumatoid arthritis Coef. of Relative friction viscosity Coef. of Relative friction viscosity 21 yr. male Torn medial meniscus ,0073 .0035 ,0077 ,0047 ,0075 40.05 5.20 * A different dog ankle was used to test the fluid of each patient. ,0098 ,0049 .0096 ,0062 ,0095 27.55 5.17 .0083 ,0043 .0078 ,0049 ,0083 20.30 6.30 679 LUBRICATION OF ANIMAL JOMTS. I11 .OL SALl NE 0: P ' I- SY NOW A ua LL LL 0 k - .02 uLL LL W 0 u .01 C Unfixed 1 Unfixed Fig. 4.-Effect of formalin fixation on Lubrication. tein from hyaluronate-protein complex of synovial fluid.12 The fact that trypsin abolishes the lubricating action of the mucin siiggests that a protein moiety is a necessary component of the mucin. One possibility is that protein may serve to attach the hyaluronate to the cartilage. Adsorption of the mucin is an essential prerequisite for boundary lubrication. In experiments concurrent with these, Wilkins has observed similar effects of enzymatic digestion on the ability of synovial mucin to lubricate a rubber-glass friction system.13 Since viscosity is related to the degree of polymerization of the hyalur~nate,'~ other things being equal, the data obtained with hyaluronidase suggest that highly polymerized hyaluronate is not necessary for synovial mucin to lubricate. Exhaustive digestion with testicular hyaluronidase did cause a rise in the coefficient of friction. The completeness of the digestion by hyaluronidase was attested to by the acetylglucosamine determination (Table 1).It is likely that some partly polymerized hyaluronate, in conjunction with protein, is responsible for the lubricating action of synovial mucin. Our tests show that blood proteins are not lubricants. That ovalbumin or purified hyaluronate solutions alone have no lubricating ability has been demonstrated previously.5 Heparin is a sulfated mucopolysaccharide. It may conceivably act as an inhibitor in this situation by blocking the attachment sites on the cartilage for the boundary lubricant, or it may in some manner react with the lubricant. It has been reported that heparin competes with hyaluronate of streptococcal origin in certain other systems.];' 25 50 75 PERCENT SYNOVIA IN THE SOLUTION 100 Fig. 5.-Lubricating ability and viscosity of synovial fluid at various dilutions. 680 The variation in the friction with pH and molarity indicates that these factors act on the articular cartilage itself. The elastic properties of this tissue and its lubrication depend largely on the fixed anionic charges of its ground substance."' McCutchen has proposed that cartilage acts as a self-pressurized sponge. It has been established that fluid flows from cartilage under pressure; that this flow can be enhanced by increasing the molarity of the interstitial fluidlT; that polyvalent cations increase deformability of the cartilage and trivalent ones depress elastic recovery.18 The deformability of articular cartilage under these conditions was studied because it is relevant to several known aspects of the lubrication of joints. Interstitial fluid is held osmotically and its displacement during loading is necessary for the boundary mechanism, inasmuch as synovial fluid lubricates deformable but not rigid beari n g ~ . ~"Plowing" -~ friction" is associated with compressive deformation of the cartilage and there is a small shear component.G The slow rise in deformation of the cartilage as the ionic strength of the immersion bath increased is in agreement with previous report^.^^'^^^^ Although swelling of articular cartilage in the presence of acid was not as pronounced as in other collagenous tissues, a distinct softening, i.e., an increase in deformation per unit load, was observed below pH 5. Under most conditions, softening was associated with a reduction in friction. Nevertheless the relationship between deformability and friction was not consistent: softening induced by molar concentrations greater than 0.5 were associated with some increase of friction. This suggests that 'Plowing involves the expenditure of energy to deform the cartilage compressively as the joint oscillates, i s . , the force necessary to move the depressions which the depressible bearing siirfnces make in each other. LI" AND W I N factors other than deformability of the cartilage may also be involved. One possiblity is that there is a surface phenomenon related to the electrostatic charges on the cartilages. This is suggested by the amphoteric character of the relationship between the coefficient of friction and pH. Maximum friction was observed at pH 5; this decreased at higher and lower hydrogen ion concentrations. It may be that the surface was isoelectric at pH 5 and positively or negatively charged at other points on the curve. The rubbing surfaces might now be separated by their mutual electrostatic repulsion. Since these experiments were carried out in the absence of mucin, no conclusions can be drawn about this phenomenon under conditions of sy novial fluid lubrication. The nature of the sites on the cartilage surface for attaching synovial much is not known. Although formalin greatly stiffened cartilage and increased the coefficient of friction by a factor of at least three, it did little to impair boundary lubrication by the mucin. One may conclude then that the attachment sites are not susceptible to blocking by formaldehyde. The human synovial fluids lubricated better than the bovine. Even though they were pathological, their viscosity was higher than that of the calf material-a well recognized species difference. Considerable dilution of the bovine fluid could be made before it failed to lubricate. Although viscosity itself is not a determinant of lubrication by synovial fluid, these observations raise the possibility that the amount of synovial mucin in joints may greatly exceed the requirements for reducing friction ,and thus be a safety factor. It is conceivablc that long-term clinical effects of small changes in lubrication are not adequately represented by these acute experiments, but we have found bovine synovial fluid 681 LUBRICATION OF ANIMAL JOINTS. 111 worked well in this test system over a period of 36 hours of continuousoperation, ACKNOWLEDGMENTS We thank Drs. Leon Sokoloff and David Hamerman for valuable contributions to this paper. Dr. Charles W. McCutchen gave helpful criticisms. Kenneth T. Cullen provided capable technical assistance. Dr. Roger Winand made the chemical analyses. Streptococcal hyaluronidase was contributed by Dr. George H. Warren, Wyeth Laboratories, and heparin by the Upjohn Company. SUMMARIOIN INTERLINGUA Esseva trovate que le coefficiente de friction in le articulationes talocrural del can mesurate in vitro es afficite marcatemente per certe tractamentos enzymatic a base de bovin mucina synovial. Hyaluronidase testicular habeba nulle effecto super le efficacitate del agente lubricatori ben q u e ill0 aboleva su viscositate. Per contrasto con isto, digestion tryptic destrueva le capacitate lubricatori del mucina sin reducer su viscositate. Iste observationes indica que un ingrediente proteinic es un componente intrinsec del molecula de mucina e que, insimul con un solo partialmente polymerisate hyaluronato, ille ingrediente es responsabile pro le function lubricatori del mucina. I1 es possibile q u e le proteina servi como gruppo prosthetic pro le adsorption del mucina a1 superficie del cartilagine. Heparina a natrium inhibiva le effecto lubricatori del mucina. In tampones libere de mucina, tanto le pH como etiam le molaritate del banio lubricatori alterava non solmente le friction sed etiam le deformabilitate del cartilagine. Le alteration in friction non esseva systematicamente relationate con le alteration in deformabilitate. Le effecto amphoteric del pH super le friction suggestiona le possibilitate que effectos electrostatic superficial contribue a1 friction adhesive. Formalina rendeva le cartilagine rigide e augmentava le coefficiente de friction. Tamen, ill0 non exerceva un marcate effecto super le action lubricatori del mucina. Un micre numero de specimens pathologic de human fluido synovial esseva studiate e il esseva trovate que illos lubricava tanto ben como o melio q u e normal material bovin. REFERENCES 1. Charnley, J.: Lubrication in synovial joints. 9. Ressing, J. L., Strominger, J. L. and Leloir, In: Proceedings of the Symposium on Biomechanics. London, Institute of Mechanical Engineers, 1959, pp. 12-22. 2. McCutchen, C. W.: Mechanism of animal joints. Sponge-hydrostatic and weeping bearings. Nature 184:1284, 1959. 3. McCutchen, C. W.: The frictional propertie.; of animal joints. Wear 5:1, 1963. 4. McCutchen, C. W.: Boundary lubrication by synovial fluid: demonstration and possible osmotic cxplauation. Fed. Proc. 25: 1061, 1966. 5. Linn, F. C.: Lubrication of animal joints. 11. The mechanism. J. Bioinech. (in press). 6. Linn, F. C.: Lubrication of animal joints. I. The arthrotripsometer. J. Bone Joint Surg. 49A: I,. F.: A modified colorimetric method for the es- 1079,1967. 7. Pigman, W. and Rizvi, S.: Hyaluronic acid and the ORD reaction. Biochem. Biophys. Res. Commun. 1:39, 1959. 8. Dische, Z.: A new specific color reaction of hexuronic acids. J. Biochem. (Engl.) 167:189, 1947. timation of N-acetylamino sugars. J. Biochem. (Engl.) 217:959, 1955. 10. Holmes, W. J., Jr., Keefer, C. S. and Myers, W. K.: Antitryptic activity of synovial fluid in patients with various types of arthritis. J. Clin. Invest. 14:124, 1935. 11. Hamerman, D., Rojkind, M. and Sandson, J.: Protein bound to hyaluronate: chemical and immunological studies. Fed. Proc. 25:1040, 1966. 12. Ogston, A. G . and Sherman, T. F.: Degradation of hyaluronic acid complex of synovial fluid by proteolytic enzymes and by EDTA. J. Biochem. (Engl.) 72:301, 1959. 13. Wilkins, J. F.: Destruction of synovial 11111cin boundary lubrication by proteolytic enzymes. (In preparation) 14. Ogston, A. G. and Stanier, J. C.: The dimensions of the particle of the hyaluronic acid complex in synovial fluid. Biochem. J. 49:585, 19 1 15. McClean, D.: The in uioo clecapsulation of 682 streptococci by hyaluronidase. J. Path. Bact. 54: 284,1942. 16. Schubert, M.: Structure of connective tissue. A chemical point of view. Fed. Proc. 25:1047. 1966. 17. Linn, I;. C. and Sokoloff, L.: Movement ant1 composition of interstitial flriid ol cartilage. Arthritis Rheum. 8:481, 1965. LINN AND W I N 18. Sokoloff, L.: Elasticity of articular cartilage: effect of ions and viscous solutions. Science 141: 1055, 1963. 19. Elmorc, S. M., Sokoloff, L., Norris, G., and Ca. iiieci, P.: The nature of “imperfect” elasticity of articiilar cartilage. J. Appl. Physiol. 18393, 1963.