The Association
of Lipid Abnormalities With Tissue Pathology Osteoarthritic Articular Cartilage
in Human
Louis Lippiello, Thomas Walsh, and Margery Fienhold Articular cartilage is one of very few body tissues uniquely characterized as having substantial stores of lipid deposits. Lipid droplets are naturally accumulated by chondrocytes and individual fatty acids have been shown to have protective as well as deleterious effects on cartilage degradation in animal models of degenerative joint disease. As a means to better assess the role of lipids in human joint pathology, a comparative analysis of fatty acids was undertaken in small segments of osteoarthritic articular cartilage. The data were assessed in terms of chondrocyte synthetic activity and histological determination of disease severity. The distribution profile of individual fatty acids in normal and osteoarthritic specimens remained constant, with palmitic, oleic, and linoleic acids representing 85% of the total fatty acids. In contrast, levels of total fatty acids were markedly increased in association with increasing degree of lesion severity. Compared with tissue from normal-aged joints, grade 0 to 1 mild lesions had elevated levels of total fatty acids, essential fatty acids, and chondrocyte synthetic activity of 80%, 312%. and 393%. respectively. More severe tissue involvement (grade 6 to 9). was associated with even greater increases of 440%. l,lOO%, and 1.150%. respectively. No change was noted in cholesterol content in any tissue. The accumulation of arachidonic acid was greater than the proportional increase in total fatty acid content and was primarily distributed into the neutral lipid fraction, where it constituted almost 62% of the fatty acid level in tissues of moderate lesion severity. There was an association of lipid accumulation in general and arachidonic acid in particular with histological severity. This association is suggestive of a lipid involvement in the chondrocyte’s response to development or progression of degenerative joint disease. Copyright 0 1991 by W.E. Saunders Company
T
HE ABUNDANCE of lipid droplets in chondrocytes has prompted a great deal of speculation on their significance in cartilage metabolism.‘~’ Bonner et al4 first documented the lipid profile of human articular cartilage, demonstrating that lipids, especially polyunsaturated fatty acids (PUFA), accumulate in normal tissue with aging. An accumulation of lipid deposits is also seen at higher than normal levels in cartilaginous tissue under conditions of trauma and in some forms of hereditary arthrosis5 Excessive accumulation of lipids occurs in states of lipoarthrosis following intra-articular injections of homologous fat.6.7In several animal instability models of arthrosis, sites of lipid accumulation generally precede local tissue degeneration. Such deposits are generally transient in nature and found in cells that are active and show no signs of degenerative changes.x.4 The role of these deposits is unknown, but it is apparent that articular cartilage has an avidity for lipid absorption, as well as an active lipid synthetic capacity.“’ Beneficial effects of lipids in joint physiology have also been observed. Treatment of familial hyperlipoproteinemia with lipid-lowering diets decreases the incidence of joint arthritic involvement.” In vivo studies by Silberberg et al”.” demonstrate a decrease in incidence of natural osteoarthrosis in an inbred strain of mice and rats fed supplements of essential fatty acids (EFA). EFA deficiency in Wistar rats is associated with a decrease in extracellular matrix synthesis. This decrease is abolished by corn oil supplementation.” Lipid droplets are also found in normal growing cartilage and in chondrocytes from auricular (elastic) cartilage, a tissue that rarely undergoes degeneration even in old age.’ In auricular cartilage, lipid deposits are increased in association with tissue regeneration.” While the presence of lipids in chondrocytes may signify a natural role in cell processes, their accumulation may also be causually related to development and/or progression of degenerative joint disease. With the exception of the documentation of levels of lipid classes, little is known of specific fatty acid changes in aging joints and in the Mefabolsm,
Vol 40, No 6 (June), 1991:
pp 571-576
development of osteoarthrosis. It is interesting to note that tissues from degenerating joints exhibit an acceleration in metabolism, an effect reproduced in vitro with normal chondrocytes supplemented with exogenous EFA.” The similarities of the observed metabolic consequences of the osteoarthrotic state and in vitro reproducibility of some of these changes by supplemental EFA are significant observations. Lipid deposition seen early in the disease process, before histological change, is conceived to reflect a lipid involvement in pathogenesis. Because cellular lipids also play a role in tissue repair processes and environmental perturbation is known to induce a lipid compensatory response, ” the observed lipid changes may represent a focal point in the chondrocyte’s adaptive (reactive) status. To further define the cellular “reaction” in osteoarthritis, the fatty acid profile of human articular cartilage at different stages of the disease process was documented. The data were correlated with histological observations of disease severity and cellular matrix metabolism and compared with a similar analysis of tissues from normal-aged joints. MATERIALS AND METHODS Twenty-nine l-cm’ sections were removed from 21 femoral heads obtained from patients undergoing hip arthroplasties and prosthetic replacement for degenerative joint disease. An additional five sites were obtained from three aged “normal” specimens resected from patients undergoing replacement for femoral neck
From the Orthopaedic Research Laboratories. Department of Orthopaedic Surgery, Universityof Nebraska Medical Center, Omaha. NE. Supported by a Grant from the National Institute of Arthritis and Musculoskeletal and Skin Diseases. Address reprint requests to Louis Lippiello. PhD, Department of Orthopaedic Surgery University of Nebraska Medical Center, 600 S 42nd St, Omaha. NE 68198-4505. Copyright 0 1991 by U%. Saunders Company 0026-0495191/4006-0004$03.OOiO 571
572
LIPPIELLO, WALSH, AND FIENHOLD
fractures. Full-depth specimens of articular cartilage were resected from multiple sites with care exercised to exclude osteophytic cartilage, obvious areas of fibrous infiltration, and severely degenerated tissue. Tissue sampling in our study was primarily derived from less involved non-weight-bearing areas. Segments from each site were subdivided as follows: a l-mm sample through the center of the site was processed for histology, the remaining tissue was diced, and one half frozen for lipid analysis. with the remaining tissue being used for metabolic studies.
Histologic Analysis Each sample was fixed for 24 hours in phosphate-buffered 10% formalin and 5- to g-pm paraffin-embedded sections dehydrated and stained with Safranin-0, fast green, and iron hematoxylin. The degree of lesion severity was determined by the method of Mankin et al.‘”
Lipid Analysis Total lipids were extracted by the method of Bligh and Dyer”’ with P-hydroxytoluene antioxidant added to a final concentration of 0.02%. Fatty acids were analyzed as their methyl ester derivatives following reaction with boron trifluoride in methanol (14% wt/vol). The esters were separated and quantitated on a HewlettPackard GC (Avondale. PA) equipped with a flame ionization detector and interfaced with a Hewlett-Packard 3392A integrator. A 6 ft x 0.25-in glass prepacked column with 10% SP-2330.100/120 mesh Chromosorb W (Supelaco, Bellefonte, PA) was eluted with a programmed g-minute temperature gradient of 175” to 200°C. Data were corrected for recovery based on an external standard and expressed as nmol/L fatty acid per mg wet tissue weight. Aliquots of lipid extracts from osteoarthritic samples were pooled according to the degree of lesion severity and subjected to silica gel chromatography (Sep-Pak, Waters Associates, Waltham, MA). Neutral lipids and phospholipids were selectively eluted with chloroform and methanol, respectively. Each fraction was derivitized and fatty acid composition determined by gas chromatography. The data were expressed as a percent distribution of each fatty acid. Cholesterol content was assayed with Sigma kit no. 351 (Sigma Chemical, St Louis, MO).
Metabolic Analysis Immediately upon resection, five to eight segments of cartilage from each site were incubated for 4 hours in Ham’s F-12 medium containing 10% fetal calf serum, 5 uCi/mL “SO, (New England Nuclear, Boston, MA, specific activity, 1.200 CiimmoliL). and penicillin plus streptomycin. Only incorporated radioactivity into tissue macromolecules was assayed. Incubated tissue was fixed for 24 hours in 10% formalin containing 10 mmol/L sodium sulfate, dehydrated, lyophilized, and separately weighed to the nearest microgram. Individual segments were solubilized with Soluene (New England Nuclear) and radioactivity monitored by liquid scintillation counting. The data were expressed as cpmimg dry tissue weight. Total fatty acids, counts per minute of isotope uptake, and EFA of the four groups were analyzed with one-way ANOVA. The tissues from osteoarthritic joints catagorized as 0 to 1,2 to 5, or 6 to 9 based on histologic grading of lesion severity, were contrasted with the normal group by using Scheffe’s multiple comparison procedure. These contrasts were also accertained by using the Mann-Whitney nonparametric procedure. RESULTS
Individual osteoarthrotic
segments of femoral head cartilage from joints were placed into one of three groups
based on histological evaluation of disease severity. With the exclusion of osteophytic cartilage and severely degenerating tissue, the individual specimens were grouped as grade 0 to 1 (low disease activity), 2 to 5 (mild tissue involvement), and 6 to 9 (moderate disease). The median grade and other tissue sampling characteristics are presented in Table 1. All tissues from normal joints were also evaluated histologically and each warranted the grade of 0. Calculation of lipid content based solely on fatty acid analysis was expectedly lower than published lipid values (0.09% to 0.18% v 0.32% to 3.48%). The distribution profile of fatty acids from total lipid extracts of all osteoarthritic samples was remarkably similar and when analyzed by the one-way ANOVA with Scheffe’s multiple comparison procedure, did not differ significantly from similar preparations of normal joints (Fig 1). The predominant fatty acids were palmitic (16:O). oleic (lS:l), and linoleic (l&2) acids, which collectively represented approximately 85% of the total fatty acid content. Total nmol/L of fatty acid per mg wet tissue weight increased significantly with lesion severity (Fig 2). All fatty acids increased in direct proportion to the increment of grade. However, there was a prominent elevation in 20:4 content in all osteoarthritic tissues, which was increased by 284% in grade 6 to 9 (P < .002), 156% in grade 2 to 5 (P < .009). and 150% in grade 0 to 1 (P < .Ol). Statistical significance was determined by Student’s t test. Quantitation of fatty acids based on degree of saturation (polyenoic, monoenoic, and saturated), is presented in Fig 3. There was no difference in percent distribution of fatty acids types between controls (normal-aged tissue) and tissues grouped according to lesion severity. This percent distribution of polyenoic, monoenoic, and saturated fatty acids of 20:43:36 mimicked previous reports for articular cartilage (15:46:39) and was similar to that found in human serum (25:40:35).“‘-” The ratio of saturated to unsaturated fatty acid remained constant at 0.55 in grades 0 to 1 and 2 to 5. but was increased to 0.64 in grades 6 to 9. The ratio of 18:2 to cholesterol significantly increased with increasing lesion severity. Since the cholesterol content remained constant at all levels (0.81 to 1.22 nmol/L), the altered ratio reflected an increase in 18:2 content associated with increasing lesion severity. The magnitude of increase in EFA content in osteoarthritic tissues was more than a reflection of the overall increase in total fatty acid content. EFA as a percentage of total fatty acids increased from 10% (normal) to 22% (P < .OOl) in grade 6 to 9. The estimate of the slope obtained from a regression analysis of EFA as a dependent variable and total fatty acids as an independent variable was 0.160 U, which is significantly different ( < 1.0) at P < ,001. Table 1. Tissue Sampling Characteristics Histological
Meall
No. of
Mean Grade
Grade
Age
Sites
(SD)
Normal
84
5
0
O-l
66
5
0.6 + 0.4
1
2-5
72
13
3.2 + 1.0
4
6-9
66
11
7.9 2 1.0
8
Medmn
0
LIPIDS IN OSTEOARTHRITIC
573
CARTILAGE
60
; 111 Fig 1.
Three-dimensional
resentation
of fatty acid distribu-
tion in normal-aged tilage
and
osteoarthritic according
4G
rep-
in
articular specimens
joints
to degree
carof
separated
011
2-5
6-9
of lesion seHISTOLOGIC GRADE
verity.
The R? value of .77 reflects a strong and disproportionate increase in EFA relative to total fatty acids. Hence, a selective accumulation of 20:4 occurs in association with increasing pathology. A surprising change was noted in EFA distribution. When the neutral and phospholipid fractions were analyzed separately the EFA content was markedly elevated in the neutral lipid fraction in direct proportion to lesion severity. Most other fatty acids were depressed in the neutral lipid fraction to the extent that 20:4 accounted for 62% of the total fatty acid content. A slight increase in 18:2 and larger decrease in 20:4 content was observed in the phospholipid fraction (Fig 4). A significant increase in metabolic activity was also associated with increasing lesion severity (Table 2). Incorporation of 35sulfate into matrix proteoglycans was greater in all osteoarthritic tissues compared with normal tissue, attaining a level of 1l-fold greater in tissues graded 6 to 9. Hence, chondrocyte synthetic activity, total lipid content, and EFA content were significantly increased in tissues in association with increases in lesion severity. DISCUSSION A unique system of quantitating histological parameters indicative of disease severity in degenerating articular cartilage provides the framework for defining “stages” of the disease process. The system has received substantial support in the literature, confirming that in the early to middle stages of osteoarthritis a gradation of tissue pathology can be defined. Briefly, the system is based on a numerical scale using the criteria of surface irregularities, matrix staining intensity, cell morphology, and cell density. By deliberately excluding tissues undergoing severe degen-
N&d
o-1
2-5
eration, we have attempted to minimize the effect of selective removal of superficial zone lipids associated with cartilage surface erosion. This is the first demonstration of the existence of a marked and graded increase in total fatty acids and selective accumulation of 20:4 arachidonic acid in articular cartilage from osteoarthritic joints. The results also confirm and extend previous observations of associations among chondrocyte metabolism, matrix composition, and lesion severity.‘” Our data suggest that lipid content, including EFA, is present at levels proportionate to lesion severity and is also related to cell synthetic activity. However, we cannot at this time determine whether cartilage metabolic dysregulation, as defined by Balustra et al,*” results from increases in lipid accumulation or whether lipid accumulation occurs as a result of metabolic dysregulation. A significant finding in this study was the selective accumulation of EFA in tissue from osteoarthritic joints (+l,lOO% in grade 6 to 9) at disproportionate levels to the +440% increase in total fatty acids. Our data do not support that of Adkisson et al,” who found that EFA were the predominant fatty acids in aged and diseased articular cartilage. Despite the substantial increase in EFA in specimens of moderate lesion severity, palmitic and oleic acid still constituted 66% of the total fatty acid content. This discrepancy is most likely due to tissue sampling, since we excluded those sites undergoing a severe degeneration. Studies by Kirkpatrick et al further suggest that accumulation of 20:4, at least in normal tissue, may elicit an increase in intracellular storage of lipids in general, rendering the tissue susceptible to damage.” Our results indicate that 20:4 accumulation occurs primarily in the neutral lipid fraction and, in fact, levels in the phospholipid fraction
6-9
Fig 2. specimens
HISTOLOGIC GRADE
Fatty
acid
expressed
par wet tissue weight.
content
of
as nmol/L
574
LIh’IELLO,
nmoles
FA/mg
wet
WALSH, AND FIENHOLD
Table 2. Characteristic Changes in Lipids Associated With Lesion
weight
Severity and “SO, Uptake
Grout
Total FattyAcids*
Normal
0.81 ? 0.31
335 2 150
O-1
1.46 2 0.43
1,650 2 400
P < .02 2-5 6-9
%O, cpmlmgt
P < ,014
EFA* 0.08 0.33 P < .Ol
2.30 ? 0.90
2,550 + 745
0.59
P < ,008
P < .04
P < ,001
4.33 + 2.6
4,150 * 1,400
0.96
P < ,001
P < ,006
P < ,001
NOTE. Statistical analysis based on comparison to “normal” tissue. Significance determined by Student’s t test or Mann-Whitney one-tailed 486
221
419
534
891
1010
798
2-5
o-1 Lesion
q polyenoic
I
2113
,859
6-9
*nmol/L per mg wet tissue weight. tcpm per mg dry tissue weight.
Severity fl saturated
monoenoic
Fig 3. Fatty acid content of tissues based on degree of unsaturation (polyenoic, monoenoic, and saturated).
decrease with increasing disease severity. No attempt was made in these studies to analyze the fatty acid distribution in cells versus the extracellular matrix. The values reported represent total tissue fatty acid content. It should be recognized that different biologic effects can be elicited when lipids accumulate intracellularly or extracellularly. For example, localization in the matrix may affect the chemical properties of proteoglycans and collagen. It is well NEUTRAL
;
60
nonparametric test.
known that these macromolecules in other tissues bind and indeed have an affinity for lipids.” Interestingly, the focal lesions of atherosclerosis like those in osteoarthrosis are characterized by a diminution of glycosaminoglycans, increased synthetic activity, and increased lipid material in association with increasing lesion severity.‘j Moreover, a topological relationship has been defined between lipids and polysaccharides in these lesions and suggestions made on the effect of lipids on proteoglycan aggregation and changes in surface properties of the macromolecules.2” Increased amounts of lipids have been demonstrated by
LIPID
-
e
O-l
2-s
6-9
HISTOLOGIC CR*OE
PHOSPHOLIPID
60
0 O-l
2-6 Hl6TOLOOlC
6-9 ORmE
Fig 4. Fatty acid distribution in osteoarthritic specimens after separation of lipid extracts into neutral and phospholipid fractions.
LIPIDS IN OSTEOARTHRITIC
575
CARTILAGE
histologic techniques in the extracellular matrix in a canine model of osteochondritis,z4 but similar demonstrations in human osteoarthritic samples are lacking. Our data do not allow for a determination of lipid distribution in articular cartilage and it may very well be that the increased amounts of lipids observed in osteoarthrotic tissue simply reflects the increased cellularity of the lesioned tissue. In this respect, it is controversial whether a real increase in cellularity occurs. Elevations in labeled thymidine uptake reflect increased DNA synthesis and observations of cell cloning (clusters) are readily apparent in histologic sections. However, these indices of increased cellularity must be tempered by observations of increased remodeling of the cartilage and localized areas of cell necrosis.25 Actual measurement of DNA levels in normal and osteoarthrotic specimens have indicated both increases” and no significant changes in cellular-
ity.‘h Intracellular lipid accumulation resulting from tissue exposure to a traumatic environment or artificially in an in vitro system of supplementation is a characteristic phenomenon of most cells.” Cell protection against such accumulation does not appear to exist, and while the consequences of lipid accumulation may vary dependent on the organ studied, their presence appears to be independent of the state of nutrition.‘” Altered joint mechanics could result in a compensatory cartilage response similar to that which occurs in atherosclerosis. In the latter case, an altered strain leads to lipid accumulation in endothelial and smooth muscle cells.” Several hypotheses have been suggested to explain the source of the accumulated lipids in articular cartilage as being extraneous in origin. Of these possibilities, exposure to abundant lipids in the environment is the most important.‘O Although the lipid content of synovial fluid is low, in rheumatoid and osteoarthritic joints there are elevated levels of several lipid classes.‘l Increased phospholipase activity and hydrolysis of adipose tissue (fat pad) triglycerides may also be a local source of synovial EFA’ and, in fact, pancreatic arthropy is thought to be mediated by locally derived free fatty acids.32 Lipid-laden sinus tracts that directly enter the joint space have been described,3” which, combined with the possibility of local fat necrosis, may function in elevating synovial fluid fatty acids. There is no doubt that chondrocytes have the ability to synthesize lipids”’ and that accumulation could occur due to a metabolic change within cells that facilitates intracellular lipid synthesisz8 In many cell types, the accumulated lipids appear as cytoplasmic droplets primarily composed of triglycerides and cholesterol esters. Histochemical analysis of chondrocyte lipids has similarily identified the presence of neutral triglycerides within chondrocyte droplets.’ The inability to utilize such lipids for energy purposes results in sequesteration in the cytoplasma and appearance as membrane-bound lipid droplets.“4.‘5 The selective accumulation of EFA as a component of sequestered lipids may provoke cellular responses by virtue of serving as eicosanoid precursors. EFA are conceived to function primarily as a source of eicosanoids, “local hor-
mones elaborated by cells which function to restore homeostatic conditions following cell perturbation.“3h Chondrocytes “loaded” in vitro with 20:4 demonstrate an acceleration in metabolism associated with large increases in PGE, synthesis.lh Oleic acid “loading” does not alter chondrocyte metabolism and indomethacin abolishes the 20:4 response. This strongly suggests that eicosanoid synthesis by chondrocytes is elevated due to the increased accumulation of the precursor 20:4 and assumes that 20:4, which is contained in lipid droplets, is an available substrate for phospholipase and oxygenase activity. An EFA deficiency state is associated with a depressed metabolism of proteoglycans as is in vivo substitution of n-6 fatty acids with n-3 fatty acids.” If EFA are necessary ingredients in an eicosanoid-mediated chondrocyte’s reparative response, elevated levels may enhance this effect mechanism. This scenario is supported by data suggesting that polyunsaturated fatty acids have a protective effect on development of osteoarthritis in bred mice and rats.” Alternatively, the protective effects of corticosteroids on development of cartilage lesions in the Pond-Nuki dog model of osteoarthritis38 and the proteoglycan inhibitory effect of nonsteroid anti-inflammatory agents” suggest that eicosanoid inhibition may be detrimental. This contradicts the premise that eicosanoid-induced increases in chondrocyte metabolism are beneficial in maintaining cartilage integrity. These opposing functions may be reconciled if one considers that steroidal and nonsteroidal effects operate primarily through inhibition of cartilage degradative activity, whereas EFA and eicosanoid-induced increases in metabolism operate through synthetic processes. Also, there is ample documentation that application of exogenous eicosanoids from in vitro systems exert cartilage inhibitory effects. Such activity differs from that observed when endogenous eicosanoid synthesis is stimulated suggesting different biological effects are manifested depending on the source of the eicosanoids. Whatever the mechanism, it is apparent that articular chondrocytes have available a large resource of EFA, especially 20:4. This resource is associated naturally with aging4,” and in accelerated fashion, perhaps by a different mechanism, in osteoarthritic lesions. To what extent the cells utilize EFA for metabolic modulation is unknown. The accumulated EFA and lipids could be solely a consequence of the tissue’s avidity for lipid adsorption, a genetic phenomenon similar to that proposed to exist for liver cellsJ” Further investigation is required to determine whether the selective accumulation of EFA is protective or deleterious to articular tissues in degenerative joint disease. REFERENCES
1. Montagna W: Glycogen and lipids in human cartilage, with some cytochemical observations on the cartilage of the dog, cat and rabbit. Anat Ret 103:77-92. 1949 2. Hart JAL: Age changes in articular cartilage with special reference to the lipid content. J Anat 103:221-227.1968 3. Collins DH, Ghadially FN, Meachim G: lntraarticular lipids of cartilage. Ann Rheum Dis 24:123-135. 1965
576
4. Bonner WM, Jonsson H, Malanos C, et al: Changes in the lipids of human articular cartilage with age. Arthritis Rheum 18461-473, 1975 5. Stanescu R, Stanescu V, Maroteaux P, Peyron, J et al: Constitutional articular cartilage dysplasia with accumulations of complex lipids in chondrocytes and precocious arthrosis. Arthritis Rheum 24:965-968,198l 6. Ghadially FN, Mehta PN, Kirkaldy-Willis WH: Ultrastructure of articular cartilage in experimentally produced lipoarthrosis. J Bone Joint Surg 52A:1147-1158,197O 7. Stockwell RA, Spring R: Glycosaminoglycan content and cell density of rabbit articular cartilage in experimental lipoarthrosis. J Anat 133:309-315, 1981 8. Lukoschek M, Schaffler MB, Burr DB, et al: Synovial membrane and cartilage changes in experimental osteoarthrosis. J Orthop Res 6:475-492.1988 9. Stockwell RA. Billingham MEJ, Muir H: Ultrastructural changes in articular cartilage after section of the anterior cruciate ligament of the dog knee. J Anat 136:425-439,1988 10. Nagao M, Ishii S, Murata Y, et al: Effect of extracellular fatty acids on lipid metabolism in cultured rabbit articular chondrocytes. J Orthop Res (in press) 11. Buckingham RB, Bole GG, Bassett DR: Polyarthritis associated with type IV hyperlipoproteinemia. Arch Intern Med 135:286290,1975 12. Silberberg M, Silberberg R: Effect of high fat diet on the joints of aging mice. Arch Pathol50:828-846,195O 13. Silberberg M, Silberberg R, Orcutt B: Modifying effect of linoleic acid on articular aging and osteoarthrosis in lard-fed mice. Gerontologia 11:179-187, 1965 14. Denko CW: Prostaglandins, essential fatty acids and 35sulfate incorporation into cartilage. Arthritis Rheum 14:379, 1971 (abstr) 15. Labandeira-Garcia J, Guerra-Seijas M: Intracellular lipids in rabbit ear cartilage during tissue regeneration. Acta Anat 127:249-254, 1986 16. Lippiello L, Ward MA: Lipid and cell metabolic changes associated with essential fatty acid enrichment of articular chondrocytes. Proc Sot Exp Biol Med (submitted) 17. Thompson GA Jr: Effects of environmental factors on lipid metabolism, in Thompson GA (ed): The Regulation of Membrane Lipid Metabolism. Boca Raton, FL, CRC, 1980, pp 171-196 18. Mankin HJ, Dorfman H, Lippiello L, et al: Biochemical and metabolic abnormalities in articular cartilage from osteoarthritic human hips. II. Correlation of morphology with biochemical and metabolic data. J Bone Joint Surg 53A:523-537, 1971 19. Bligh EG. Dyer WJ: A rapid method of total lipid extraction and purification. Can J Biochem Physiol37:911-917. 1959 20. Bulstra SK, Buurman WA, Walenkamp GH, et al: Metabolic characteristics of in vitro cultured human chondrocytes in relation to the histopathologic grade of osteoarthritis. Clin Orthop Rel Res 242:294-302,1989 21. Adkisson HD, Trancik TM, Zarrinkar PP, et al: Accumulation of essential fatty acids in hyaline cartilage with aging and osteoarthritis. Trans Orthop Res Sot 15:335, 1990 (abstr)
LIPPIELLO, WALSH, AND FIENHOLD
22. Kirkpatrick CJ. Mohr W, Haferkamp 0: Lipid storage in cultured articular chondrocytes due to prostanoid precursors and a prostanoid synthesis inhibitor. Cell Tissue Res 224:441-448, 1982 23. Camejo G: The interaction of lipids and lipoproteins with the intercellular matrix of arterial tissue: Its possible role in atherogenesis. Lipid Res 19:1-53.1982 24. Kincaid SA: Lipids of normal and osteochondritic cartilage of the immature canine humeral head. Am J Vet Res 46:1060-1065. 1985 25. Ali SY, Evans L: Enzymatic osteoarthritis. Fed Proc 32:1494-1498, 26. Mankin thritis. Orthop
degradation 1973
HJ: Biochemical and metabolic Clin North Am 2:19-31, 1971
of cartilage aspects
in
of osteoar-
27. Cawson RA, McCracken AW, Marcus PB: Pathologic Mechanisms and Human Disease. St Louis, MO, Mosby, 1982, pp 14-32 28. Spector AA, Mathur SN, Haduce TL, et al: Lipid nutrition and metabolism of cultured mammalian cells. Prog Lipid Res 19:155-186, 1981 29. Lupu F, Danaricu I, Simionescu N: Development of intracellular lipid deposits in the lipid-laden cells of atherosclerotic lesion. Atherosclerosis 67:127-142, 1987 30. Bole GG: Synovial patients with rheumatoid 1962
fluid lipids in normal individuals and arthritis. Arthritis Rheum 5:589-601,
31. Chung AC, Shanahan JR, Brown EM Jr: Synovial fluid lipids in rheumatoid and osteoarthritis. Arthritis Rheum 5:176-183, 1962 32. Smukler NM, Schumacher HR, Pascual E. et al: Synovial fat necrosis associated with ischemic pancreatic disease. Arthritis Rheum 23:355-360, 1979 33. Simkin PA, Brunzell JD, Wisner D, et al: Free fatty acids in the pancreatitic arthritis syndrome. Arthritis Rheum 26:127-332, 1983 34. Nahir AM: Aerobic glycolysis: A study of human cartilage. Cell Biochem Funct 5:109-112, 1987
articular
35. Dunham J, Dodds RA. Nahir AM, et al: Aerobic glycolysis of bone and cartilage: The possible involvement of fatty acid oxidation. Cell Biochem Funct 1:168-172, 1983 36. Markelonis G, Garbus J: Alterations of intracellular oxidative metabolism as stimuli evoking prostaglandin synthesis. Prostaglandins 10:1087-1106. 1975 37. Lippiello L, Fienhold M, Grandjean C: Metabolic and ultrastructural changes in articular cartilage of rats fed dietary supplements of omega-3 fatty acids. Arthritis Rheum 33:10291036,199O 38. Pelletier J-P, Martel-Pelletier J: Protective effects of corticosteroids on cartilage lesions and osteophyte formation in the Pond-Nuki dog model of osteoarthritis. Arthritis Rheum 32:181193,1989 39. Brandt KD, Palmoski M: Effects of nonsteroidal antiinflammatory drugs on proteoglycan metabolism in articular cartilage. Semin Arthritis Rheum 11:133-134. 1980 40. Mackenzie CG, Mackenzie JB. Reiss OK: Isolation and properties of a cell from liver characterized by lipid-rich particles. J Cell Biol 14:269-279,1962
of Lipid Abnormalities With Tissue Pathology Osteoarthritic Articular Cartilage
in Human
Louis Lippiello, Thomas Walsh, and Margery Fienhold Articular cartilage is one of very few body tissues uniquely characterized as having substantial stores of lipid deposits. Lipid droplets are naturally accumulated by chondrocytes and individual fatty acids have been shown to have protective as well as deleterious effects on cartilage degradation in animal models of degenerative joint disease. As a means to better assess the role of lipids in human joint pathology, a comparative analysis of fatty acids was undertaken in small segments of osteoarthritic articular cartilage. The data were assessed in terms of chondrocyte synthetic activity and histological determination of disease severity. The distribution profile of individual fatty acids in normal and osteoarthritic specimens remained constant, with palmitic, oleic, and linoleic acids representing 85% of the total fatty acids. In contrast, levels of total fatty acids were markedly increased in association with increasing degree of lesion severity. Compared with tissue from normal-aged joints, grade 0 to 1 mild lesions had elevated levels of total fatty acids, essential fatty acids, and chondrocyte synthetic activity of 80%, 312%. and 393%. respectively. More severe tissue involvement (grade 6 to 9). was associated with even greater increases of 440%. l,lOO%, and 1.150%. respectively. No change was noted in cholesterol content in any tissue. The accumulation of arachidonic acid was greater than the proportional increase in total fatty acid content and was primarily distributed into the neutral lipid fraction, where it constituted almost 62% of the fatty acid level in tissues of moderate lesion severity. There was an association of lipid accumulation in general and arachidonic acid in particular with histological severity. This association is suggestive of a lipid involvement in the chondrocyte’s response to development or progression of degenerative joint disease. Copyright 0 1991 by W.E. Saunders Company
T
HE ABUNDANCE of lipid droplets in chondrocytes has prompted a great deal of speculation on their significance in cartilage metabolism.‘~’ Bonner et al4 first documented the lipid profile of human articular cartilage, demonstrating that lipids, especially polyunsaturated fatty acids (PUFA), accumulate in normal tissue with aging. An accumulation of lipid deposits is also seen at higher than normal levels in cartilaginous tissue under conditions of trauma and in some forms of hereditary arthrosis5 Excessive accumulation of lipids occurs in states of lipoarthrosis following intra-articular injections of homologous fat.6.7In several animal instability models of arthrosis, sites of lipid accumulation generally precede local tissue degeneration. Such deposits are generally transient in nature and found in cells that are active and show no signs of degenerative changes.x.4 The role of these deposits is unknown, but it is apparent that articular cartilage has an avidity for lipid absorption, as well as an active lipid synthetic capacity.“’ Beneficial effects of lipids in joint physiology have also been observed. Treatment of familial hyperlipoproteinemia with lipid-lowering diets decreases the incidence of joint arthritic involvement.” In vivo studies by Silberberg et al”.” demonstrate a decrease in incidence of natural osteoarthrosis in an inbred strain of mice and rats fed supplements of essential fatty acids (EFA). EFA deficiency in Wistar rats is associated with a decrease in extracellular matrix synthesis. This decrease is abolished by corn oil supplementation.” Lipid droplets are also found in normal growing cartilage and in chondrocytes from auricular (elastic) cartilage, a tissue that rarely undergoes degeneration even in old age.’ In auricular cartilage, lipid deposits are increased in association with tissue regeneration.” While the presence of lipids in chondrocytes may signify a natural role in cell processes, their accumulation may also be causually related to development and/or progression of degenerative joint disease. With the exception of the documentation of levels of lipid classes, little is known of specific fatty acid changes in aging joints and in the Mefabolsm,
Vol 40, No 6 (June), 1991:
pp 571-576
development of osteoarthrosis. It is interesting to note that tissues from degenerating joints exhibit an acceleration in metabolism, an effect reproduced in vitro with normal chondrocytes supplemented with exogenous EFA.” The similarities of the observed metabolic consequences of the osteoarthrotic state and in vitro reproducibility of some of these changes by supplemental EFA are significant observations. Lipid deposition seen early in the disease process, before histological change, is conceived to reflect a lipid involvement in pathogenesis. Because cellular lipids also play a role in tissue repair processes and environmental perturbation is known to induce a lipid compensatory response, ” the observed lipid changes may represent a focal point in the chondrocyte’s adaptive (reactive) status. To further define the cellular “reaction” in osteoarthritis, the fatty acid profile of human articular cartilage at different stages of the disease process was documented. The data were correlated with histological observations of disease severity and cellular matrix metabolism and compared with a similar analysis of tissues from normal-aged joints. MATERIALS AND METHODS Twenty-nine l-cm’ sections were removed from 21 femoral heads obtained from patients undergoing hip arthroplasties and prosthetic replacement for degenerative joint disease. An additional five sites were obtained from three aged “normal” specimens resected from patients undergoing replacement for femoral neck
From the Orthopaedic Research Laboratories. Department of Orthopaedic Surgery, Universityof Nebraska Medical Center, Omaha. NE. Supported by a Grant from the National Institute of Arthritis and Musculoskeletal and Skin Diseases. Address reprint requests to Louis Lippiello. PhD, Department of Orthopaedic Surgery University of Nebraska Medical Center, 600 S 42nd St, Omaha. NE 68198-4505. Copyright 0 1991 by U%. Saunders Company 0026-0495191/4006-0004$03.OOiO 571
572
LIPPIELLO, WALSH, AND FIENHOLD
fractures. Full-depth specimens of articular cartilage were resected from multiple sites with care exercised to exclude osteophytic cartilage, obvious areas of fibrous infiltration, and severely degenerated tissue. Tissue sampling in our study was primarily derived from less involved non-weight-bearing areas. Segments from each site were subdivided as follows: a l-mm sample through the center of the site was processed for histology, the remaining tissue was diced, and one half frozen for lipid analysis. with the remaining tissue being used for metabolic studies.
Histologic Analysis Each sample was fixed for 24 hours in phosphate-buffered 10% formalin and 5- to g-pm paraffin-embedded sections dehydrated and stained with Safranin-0, fast green, and iron hematoxylin. The degree of lesion severity was determined by the method of Mankin et al.‘”
Lipid Analysis Total lipids were extracted by the method of Bligh and Dyer”’ with P-hydroxytoluene antioxidant added to a final concentration of 0.02%. Fatty acids were analyzed as their methyl ester derivatives following reaction with boron trifluoride in methanol (14% wt/vol). The esters were separated and quantitated on a HewlettPackard GC (Avondale. PA) equipped with a flame ionization detector and interfaced with a Hewlett-Packard 3392A integrator. A 6 ft x 0.25-in glass prepacked column with 10% SP-2330.100/120 mesh Chromosorb W (Supelaco, Bellefonte, PA) was eluted with a programmed g-minute temperature gradient of 175” to 200°C. Data were corrected for recovery based on an external standard and expressed as nmol/L fatty acid per mg wet tissue weight. Aliquots of lipid extracts from osteoarthritic samples were pooled according to the degree of lesion severity and subjected to silica gel chromatography (Sep-Pak, Waters Associates, Waltham, MA). Neutral lipids and phospholipids were selectively eluted with chloroform and methanol, respectively. Each fraction was derivitized and fatty acid composition determined by gas chromatography. The data were expressed as a percent distribution of each fatty acid. Cholesterol content was assayed with Sigma kit no. 351 (Sigma Chemical, St Louis, MO).
Metabolic Analysis Immediately upon resection, five to eight segments of cartilage from each site were incubated for 4 hours in Ham’s F-12 medium containing 10% fetal calf serum, 5 uCi/mL “SO, (New England Nuclear, Boston, MA, specific activity, 1.200 CiimmoliL). and penicillin plus streptomycin. Only incorporated radioactivity into tissue macromolecules was assayed. Incubated tissue was fixed for 24 hours in 10% formalin containing 10 mmol/L sodium sulfate, dehydrated, lyophilized, and separately weighed to the nearest microgram. Individual segments were solubilized with Soluene (New England Nuclear) and radioactivity monitored by liquid scintillation counting. The data were expressed as cpmimg dry tissue weight. Total fatty acids, counts per minute of isotope uptake, and EFA of the four groups were analyzed with one-way ANOVA. The tissues from osteoarthritic joints catagorized as 0 to 1,2 to 5, or 6 to 9 based on histologic grading of lesion severity, were contrasted with the normal group by using Scheffe’s multiple comparison procedure. These contrasts were also accertained by using the Mann-Whitney nonparametric procedure. RESULTS
Individual osteoarthrotic
segments of femoral head cartilage from joints were placed into one of three groups
based on histological evaluation of disease severity. With the exclusion of osteophytic cartilage and severely degenerating tissue, the individual specimens were grouped as grade 0 to 1 (low disease activity), 2 to 5 (mild tissue involvement), and 6 to 9 (moderate disease). The median grade and other tissue sampling characteristics are presented in Table 1. All tissues from normal joints were also evaluated histologically and each warranted the grade of 0. Calculation of lipid content based solely on fatty acid analysis was expectedly lower than published lipid values (0.09% to 0.18% v 0.32% to 3.48%). The distribution profile of fatty acids from total lipid extracts of all osteoarthritic samples was remarkably similar and when analyzed by the one-way ANOVA with Scheffe’s multiple comparison procedure, did not differ significantly from similar preparations of normal joints (Fig 1). The predominant fatty acids were palmitic (16:O). oleic (lS:l), and linoleic (l&2) acids, which collectively represented approximately 85% of the total fatty acid content. Total nmol/L of fatty acid per mg wet tissue weight increased significantly with lesion severity (Fig 2). All fatty acids increased in direct proportion to the increment of grade. However, there was a prominent elevation in 20:4 content in all osteoarthritic tissues, which was increased by 284% in grade 6 to 9 (P < .002), 156% in grade 2 to 5 (P < .009). and 150% in grade 0 to 1 (P < .Ol). Statistical significance was determined by Student’s t test. Quantitation of fatty acids based on degree of saturation (polyenoic, monoenoic, and saturated), is presented in Fig 3. There was no difference in percent distribution of fatty acids types between controls (normal-aged tissue) and tissues grouped according to lesion severity. This percent distribution of polyenoic, monoenoic, and saturated fatty acids of 20:43:36 mimicked previous reports for articular cartilage (15:46:39) and was similar to that found in human serum (25:40:35).“‘-” The ratio of saturated to unsaturated fatty acid remained constant at 0.55 in grades 0 to 1 and 2 to 5. but was increased to 0.64 in grades 6 to 9. The ratio of 18:2 to cholesterol significantly increased with increasing lesion severity. Since the cholesterol content remained constant at all levels (0.81 to 1.22 nmol/L), the altered ratio reflected an increase in 18:2 content associated with increasing lesion severity. The magnitude of increase in EFA content in osteoarthritic tissues was more than a reflection of the overall increase in total fatty acid content. EFA as a percentage of total fatty acids increased from 10% (normal) to 22% (P < .OOl) in grade 6 to 9. The estimate of the slope obtained from a regression analysis of EFA as a dependent variable and total fatty acids as an independent variable was 0.160 U, which is significantly different ( < 1.0) at P < ,001. Table 1. Tissue Sampling Characteristics Histological
Meall
No. of
Mean Grade
Grade
Age
Sites
(SD)
Normal
84
5
0
O-l
66
5
0.6 + 0.4
1
2-5
72
13
3.2 + 1.0
4
6-9
66
11
7.9 2 1.0
8
Medmn
0
LIPIDS IN OSTEOARTHRITIC
573
CARTILAGE
60
; 111 Fig 1.
Three-dimensional
resentation
of fatty acid distribu-
tion in normal-aged tilage
and
osteoarthritic according
4G
rep-
in
articular specimens
joints
to degree
carof
separated
011
2-5
6-9
of lesion seHISTOLOGIC GRADE
verity.
The R? value of .77 reflects a strong and disproportionate increase in EFA relative to total fatty acids. Hence, a selective accumulation of 20:4 occurs in association with increasing pathology. A surprising change was noted in EFA distribution. When the neutral and phospholipid fractions were analyzed separately the EFA content was markedly elevated in the neutral lipid fraction in direct proportion to lesion severity. Most other fatty acids were depressed in the neutral lipid fraction to the extent that 20:4 accounted for 62% of the total fatty acid content. A slight increase in 18:2 and larger decrease in 20:4 content was observed in the phospholipid fraction (Fig 4). A significant increase in metabolic activity was also associated with increasing lesion severity (Table 2). Incorporation of 35sulfate into matrix proteoglycans was greater in all osteoarthritic tissues compared with normal tissue, attaining a level of 1l-fold greater in tissues graded 6 to 9. Hence, chondrocyte synthetic activity, total lipid content, and EFA content were significantly increased in tissues in association with increases in lesion severity. DISCUSSION A unique system of quantitating histological parameters indicative of disease severity in degenerating articular cartilage provides the framework for defining “stages” of the disease process. The system has received substantial support in the literature, confirming that in the early to middle stages of osteoarthritis a gradation of tissue pathology can be defined. Briefly, the system is based on a numerical scale using the criteria of surface irregularities, matrix staining intensity, cell morphology, and cell density. By deliberately excluding tissues undergoing severe degen-
N&d
o-1
2-5
eration, we have attempted to minimize the effect of selective removal of superficial zone lipids associated with cartilage surface erosion. This is the first demonstration of the existence of a marked and graded increase in total fatty acids and selective accumulation of 20:4 arachidonic acid in articular cartilage from osteoarthritic joints. The results also confirm and extend previous observations of associations among chondrocyte metabolism, matrix composition, and lesion severity.‘” Our data suggest that lipid content, including EFA, is present at levels proportionate to lesion severity and is also related to cell synthetic activity. However, we cannot at this time determine whether cartilage metabolic dysregulation, as defined by Balustra et al,*” results from increases in lipid accumulation or whether lipid accumulation occurs as a result of metabolic dysregulation. A significant finding in this study was the selective accumulation of EFA in tissue from osteoarthritic joints (+l,lOO% in grade 6 to 9) at disproportionate levels to the +440% increase in total fatty acids. Our data do not support that of Adkisson et al,” who found that EFA were the predominant fatty acids in aged and diseased articular cartilage. Despite the substantial increase in EFA in specimens of moderate lesion severity, palmitic and oleic acid still constituted 66% of the total fatty acid content. This discrepancy is most likely due to tissue sampling, since we excluded those sites undergoing a severe degeneration. Studies by Kirkpatrick et al further suggest that accumulation of 20:4, at least in normal tissue, may elicit an increase in intracellular storage of lipids in general, rendering the tissue susceptible to damage.” Our results indicate that 20:4 accumulation occurs primarily in the neutral lipid fraction and, in fact, levels in the phospholipid fraction
6-9
Fig 2. specimens
HISTOLOGIC GRADE
Fatty
acid
expressed
par wet tissue weight.
content
of
as nmol/L
574
LIh’IELLO,
nmoles
FA/mg
wet
WALSH, AND FIENHOLD
Table 2. Characteristic Changes in Lipids Associated With Lesion
weight
Severity and “SO, Uptake
Grout
Total FattyAcids*
Normal
0.81 ? 0.31
335 2 150
O-1
1.46 2 0.43
1,650 2 400
P < .02 2-5 6-9
%O, cpmlmgt
P < ,014
EFA* 0.08 0.33 P < .Ol
2.30 ? 0.90
2,550 + 745
0.59
P < ,008
P < .04
P < ,001
4.33 + 2.6
4,150 * 1,400
0.96
P < ,001
P < ,006
P < ,001
NOTE. Statistical analysis based on comparison to “normal” tissue. Significance determined by Student’s t test or Mann-Whitney one-tailed 486
221
419
534
891
1010
798
2-5
o-1 Lesion
q polyenoic
I
2113
,859
6-9
*nmol/L per mg wet tissue weight. tcpm per mg dry tissue weight.
Severity fl saturated
monoenoic
Fig 3. Fatty acid content of tissues based on degree of unsaturation (polyenoic, monoenoic, and saturated).
decrease with increasing disease severity. No attempt was made in these studies to analyze the fatty acid distribution in cells versus the extracellular matrix. The values reported represent total tissue fatty acid content. It should be recognized that different biologic effects can be elicited when lipids accumulate intracellularly or extracellularly. For example, localization in the matrix may affect the chemical properties of proteoglycans and collagen. It is well NEUTRAL
;
60
nonparametric test.
known that these macromolecules in other tissues bind and indeed have an affinity for lipids.” Interestingly, the focal lesions of atherosclerosis like those in osteoarthrosis are characterized by a diminution of glycosaminoglycans, increased synthetic activity, and increased lipid material in association with increasing lesion severity.‘j Moreover, a topological relationship has been defined between lipids and polysaccharides in these lesions and suggestions made on the effect of lipids on proteoglycan aggregation and changes in surface properties of the macromolecules.2” Increased amounts of lipids have been demonstrated by
LIPID
-
e
O-l
2-s
6-9
HISTOLOGIC CR*OE
PHOSPHOLIPID
60
0 O-l
2-6 Hl6TOLOOlC
6-9 ORmE
Fig 4. Fatty acid distribution in osteoarthritic specimens after separation of lipid extracts into neutral and phospholipid fractions.
LIPIDS IN OSTEOARTHRITIC
575
CARTILAGE
histologic techniques in the extracellular matrix in a canine model of osteochondritis,z4 but similar demonstrations in human osteoarthritic samples are lacking. Our data do not allow for a determination of lipid distribution in articular cartilage and it may very well be that the increased amounts of lipids observed in osteoarthrotic tissue simply reflects the increased cellularity of the lesioned tissue. In this respect, it is controversial whether a real increase in cellularity occurs. Elevations in labeled thymidine uptake reflect increased DNA synthesis and observations of cell cloning (clusters) are readily apparent in histologic sections. However, these indices of increased cellularity must be tempered by observations of increased remodeling of the cartilage and localized areas of cell necrosis.25 Actual measurement of DNA levels in normal and osteoarthrotic specimens have indicated both increases” and no significant changes in cellular-
ity.‘h Intracellular lipid accumulation resulting from tissue exposure to a traumatic environment or artificially in an in vitro system of supplementation is a characteristic phenomenon of most cells.” Cell protection against such accumulation does not appear to exist, and while the consequences of lipid accumulation may vary dependent on the organ studied, their presence appears to be independent of the state of nutrition.‘” Altered joint mechanics could result in a compensatory cartilage response similar to that which occurs in atherosclerosis. In the latter case, an altered strain leads to lipid accumulation in endothelial and smooth muscle cells.” Several hypotheses have been suggested to explain the source of the accumulated lipids in articular cartilage as being extraneous in origin. Of these possibilities, exposure to abundant lipids in the environment is the most important.‘O Although the lipid content of synovial fluid is low, in rheumatoid and osteoarthritic joints there are elevated levels of several lipid classes.‘l Increased phospholipase activity and hydrolysis of adipose tissue (fat pad) triglycerides may also be a local source of synovial EFA’ and, in fact, pancreatic arthropy is thought to be mediated by locally derived free fatty acids.32 Lipid-laden sinus tracts that directly enter the joint space have been described,3” which, combined with the possibility of local fat necrosis, may function in elevating synovial fluid fatty acids. There is no doubt that chondrocytes have the ability to synthesize lipids”’ and that accumulation could occur due to a metabolic change within cells that facilitates intracellular lipid synthesisz8 In many cell types, the accumulated lipids appear as cytoplasmic droplets primarily composed of triglycerides and cholesterol esters. Histochemical analysis of chondrocyte lipids has similarily identified the presence of neutral triglycerides within chondrocyte droplets.’ The inability to utilize such lipids for energy purposes results in sequesteration in the cytoplasma and appearance as membrane-bound lipid droplets.“4.‘5 The selective accumulation of EFA as a component of sequestered lipids may provoke cellular responses by virtue of serving as eicosanoid precursors. EFA are conceived to function primarily as a source of eicosanoids, “local hor-
mones elaborated by cells which function to restore homeostatic conditions following cell perturbation.“3h Chondrocytes “loaded” in vitro with 20:4 demonstrate an acceleration in metabolism associated with large increases in PGE, synthesis.lh Oleic acid “loading” does not alter chondrocyte metabolism and indomethacin abolishes the 20:4 response. This strongly suggests that eicosanoid synthesis by chondrocytes is elevated due to the increased accumulation of the precursor 20:4 and assumes that 20:4, which is contained in lipid droplets, is an available substrate for phospholipase and oxygenase activity. An EFA deficiency state is associated with a depressed metabolism of proteoglycans as is in vivo substitution of n-6 fatty acids with n-3 fatty acids.” If EFA are necessary ingredients in an eicosanoid-mediated chondrocyte’s reparative response, elevated levels may enhance this effect mechanism. This scenario is supported by data suggesting that polyunsaturated fatty acids have a protective effect on development of osteoarthritis in bred mice and rats.” Alternatively, the protective effects of corticosteroids on development of cartilage lesions in the Pond-Nuki dog model of osteoarthritis38 and the proteoglycan inhibitory effect of nonsteroid anti-inflammatory agents” suggest that eicosanoid inhibition may be detrimental. This contradicts the premise that eicosanoid-induced increases in chondrocyte metabolism are beneficial in maintaining cartilage integrity. These opposing functions may be reconciled if one considers that steroidal and nonsteroidal effects operate primarily through inhibition of cartilage degradative activity, whereas EFA and eicosanoid-induced increases in metabolism operate through synthetic processes. Also, there is ample documentation that application of exogenous eicosanoids from in vitro systems exert cartilage inhibitory effects. Such activity differs from that observed when endogenous eicosanoid synthesis is stimulated suggesting different biological effects are manifested depending on the source of the eicosanoids. Whatever the mechanism, it is apparent that articular chondrocytes have available a large resource of EFA, especially 20:4. This resource is associated naturally with aging4,” and in accelerated fashion, perhaps by a different mechanism, in osteoarthritic lesions. To what extent the cells utilize EFA for metabolic modulation is unknown. The accumulated EFA and lipids could be solely a consequence of the tissue’s avidity for lipid adsorption, a genetic phenomenon similar to that proposed to exist for liver cellsJ” Further investigation is required to determine whether the selective accumulation of EFA is protective or deleterious to articular tissues in degenerative joint disease. REFERENCES
1. Montagna W: Glycogen and lipids in human cartilage, with some cytochemical observations on the cartilage of the dog, cat and rabbit. Anat Ret 103:77-92. 1949 2. Hart JAL: Age changes in articular cartilage with special reference to the lipid content. J Anat 103:221-227.1968 3. Collins DH, Ghadially FN, Meachim G: lntraarticular lipids of cartilage. Ann Rheum Dis 24:123-135. 1965
576
4. Bonner WM, Jonsson H, Malanos C, et al: Changes in the lipids of human articular cartilage with age. Arthritis Rheum 18461-473, 1975 5. Stanescu R, Stanescu V, Maroteaux P, Peyron, J et al: Constitutional articular cartilage dysplasia with accumulations of complex lipids in chondrocytes and precocious arthrosis. Arthritis Rheum 24:965-968,198l 6. Ghadially FN, Mehta PN, Kirkaldy-Willis WH: Ultrastructure of articular cartilage in experimentally produced lipoarthrosis. J Bone Joint Surg 52A:1147-1158,197O 7. Stockwell RA, Spring R: Glycosaminoglycan content and cell density of rabbit articular cartilage in experimental lipoarthrosis. J Anat 133:309-315, 1981 8. Lukoschek M, Schaffler MB, Burr DB, et al: Synovial membrane and cartilage changes in experimental osteoarthrosis. J Orthop Res 6:475-492.1988 9. Stockwell RA. Billingham MEJ, Muir H: Ultrastructural changes in articular cartilage after section of the anterior cruciate ligament of the dog knee. J Anat 136:425-439,1988 10. Nagao M, Ishii S, Murata Y, et al: Effect of extracellular fatty acids on lipid metabolism in cultured rabbit articular chondrocytes. J Orthop Res (in press) 11. Buckingham RB, Bole GG, Bassett DR: Polyarthritis associated with type IV hyperlipoproteinemia. Arch Intern Med 135:286290,1975 12. Silberberg M, Silberberg R: Effect of high fat diet on the joints of aging mice. Arch Pathol50:828-846,195O 13. Silberberg M, Silberberg R, Orcutt B: Modifying effect of linoleic acid on articular aging and osteoarthrosis in lard-fed mice. Gerontologia 11:179-187, 1965 14. Denko CW: Prostaglandins, essential fatty acids and 35sulfate incorporation into cartilage. Arthritis Rheum 14:379, 1971 (abstr) 15. Labandeira-Garcia J, Guerra-Seijas M: Intracellular lipids in rabbit ear cartilage during tissue regeneration. Acta Anat 127:249-254, 1986 16. Lippiello L, Ward MA: Lipid and cell metabolic changes associated with essential fatty acid enrichment of articular chondrocytes. Proc Sot Exp Biol Med (submitted) 17. Thompson GA Jr: Effects of environmental factors on lipid metabolism, in Thompson GA (ed): The Regulation of Membrane Lipid Metabolism. Boca Raton, FL, CRC, 1980, pp 171-196 18. Mankin HJ, Dorfman H, Lippiello L, et al: Biochemical and metabolic abnormalities in articular cartilage from osteoarthritic human hips. II. Correlation of morphology with biochemical and metabolic data. J Bone Joint Surg 53A:523-537, 1971 19. Bligh EG. Dyer WJ: A rapid method of total lipid extraction and purification. Can J Biochem Physiol37:911-917. 1959 20. Bulstra SK, Buurman WA, Walenkamp GH, et al: Metabolic characteristics of in vitro cultured human chondrocytes in relation to the histopathologic grade of osteoarthritis. Clin Orthop Rel Res 242:294-302,1989 21. Adkisson HD, Trancik TM, Zarrinkar PP, et al: Accumulation of essential fatty acids in hyaline cartilage with aging and osteoarthritis. Trans Orthop Res Sot 15:335, 1990 (abstr)
LIPPIELLO, WALSH, AND FIENHOLD
22. Kirkpatrick CJ. Mohr W, Haferkamp 0: Lipid storage in cultured articular chondrocytes due to prostanoid precursors and a prostanoid synthesis inhibitor. Cell Tissue Res 224:441-448, 1982 23. Camejo G: The interaction of lipids and lipoproteins with the intercellular matrix of arterial tissue: Its possible role in atherogenesis. Lipid Res 19:1-53.1982 24. Kincaid SA: Lipids of normal and osteochondritic cartilage of the immature canine humeral head. Am J Vet Res 46:1060-1065. 1985 25. Ali SY, Evans L: Enzymatic osteoarthritis. Fed Proc 32:1494-1498, 26. Mankin thritis. Orthop
degradation 1973
HJ: Biochemical and metabolic Clin North Am 2:19-31, 1971
of cartilage aspects
in
of osteoar-
27. Cawson RA, McCracken AW, Marcus PB: Pathologic Mechanisms and Human Disease. St Louis, MO, Mosby, 1982, pp 14-32 28. Spector AA, Mathur SN, Haduce TL, et al: Lipid nutrition and metabolism of cultured mammalian cells. Prog Lipid Res 19:155-186, 1981 29. Lupu F, Danaricu I, Simionescu N: Development of intracellular lipid deposits in the lipid-laden cells of atherosclerotic lesion. Atherosclerosis 67:127-142, 1987 30. Bole GG: Synovial patients with rheumatoid 1962
fluid lipids in normal individuals and arthritis. Arthritis Rheum 5:589-601,
31. Chung AC, Shanahan JR, Brown EM Jr: Synovial fluid lipids in rheumatoid and osteoarthritis. Arthritis Rheum 5:176-183, 1962 32. Smukler NM, Schumacher HR, Pascual E. et al: Synovial fat necrosis associated with ischemic pancreatic disease. Arthritis Rheum 23:355-360, 1979 33. Simkin PA, Brunzell JD, Wisner D, et al: Free fatty acids in the pancreatitic arthritis syndrome. Arthritis Rheum 26:127-332, 1983 34. Nahir AM: Aerobic glycolysis: A study of human cartilage. Cell Biochem Funct 5:109-112, 1987
articular
35. Dunham J, Dodds RA. Nahir AM, et al: Aerobic glycolysis of bone and cartilage: The possible involvement of fatty acid oxidation. Cell Biochem Funct 1:168-172, 1983 36. Markelonis G, Garbus J: Alterations of intracellular oxidative metabolism as stimuli evoking prostaglandin synthesis. Prostaglandins 10:1087-1106. 1975 37. Lippiello L, Fienhold M, Grandjean C: Metabolic and ultrastructural changes in articular cartilage of rats fed dietary supplements of omega-3 fatty acids. Arthritis Rheum 33:10291036,199O 38. Pelletier J-P, Martel-Pelletier J: Protective effects of corticosteroids on cartilage lesions and osteophyte formation in the Pond-Nuki dog model of osteoarthritis. Arthritis Rheum 32:181193,1989 39. Brandt KD, Palmoski M: Effects of nonsteroidal antiinflammatory drugs on proteoglycan metabolism in articular cartilage. Semin Arthritis Rheum 11:133-134. 1980 40. Mackenzie CG, Mackenzie JB. Reiss OK: Isolation and properties of a cell from liver characterized by lipid-rich particles. J Cell Biol 14:269-279,1962
