The effects of leucocyte elastase on the mechanical properties of adult human articular cartilage in tension

The effects of leucocyte elastase on the mechanical properties of adult human articular cartilage in tension

Biochimica et Biophysica A cta, 677 (1981) 103-108 103 Elsevier/North-HollandBiomedicalPress BBA 29711 THE EFFECTS OF LEUCOCYTE ELASTASE ON THE MECH...

432KB Sizes 0 Downloads 67 Views

Biochimica et Biophysica A cta, 677 (1981) 103-108

103

Elsevier/North-HollandBiomedicalPress BBA 29711 THE EFFECTS OF LEUCOCYTE ELASTASE ON THE MECHANICAL PROPERTIES OF ADULT HUMAN ARTICULAR CARTILAGE IN TENSION DANIEL L. BADERa,*, GEOFFREY E. KEMPSONa, ALANJ. BARRETTb and WENDYWEBBb a The Department of Medical Engineering, Level D, Centre Block, Southampton General Hospital, Southampton, S09 4XY, and b The Biochemistry Department, Strangeways Research Laboratory, Wort's Causeway, Cambridge (U.K.J

(Received February 16th, 1981)

Key worda" Elastase; Mechanical property; Articular cartilage; Tension; (Human)

The effects of leucocyte elastase on the tensile properties of adult human articular cartilage were examined in detail in 99 specimens from hip, knee and ankle joints in the age range 16-83 years. The results showed that elastase reduced the tensile stiffness of cartilage, both at low stress and at fracture. The tensile strength of car. triage was also considerably reduced by the action of elastase. Biochemical analysis of the incubation media, and the specimens, revealed that 90%, or more, of the proteoglycan was released from the cartilage, whilst the release of collagen was negli~"ole. Leucocyte elastase is known to degrade the non-helical terminal peptides of cartilage collagen molecules and thereby disrupt the main intermolecular cross-links in collage fribrils. A previous study (Kempson, G.E., Tuke, M.A., Dingle, J.T., Barrett, A.J. and Horsfield, P.H. (1976) Biochim. Biophys. Acta 428, 741-760) showed the lack of effect of proteoglycan degradation alone on the tensile strength and stiffness of cartilage. The reduction in strength and stiffness recorded in the present study can, therefore, be attributed to the action of elastase on the collagen in cartilage and it emphasises the important of covalent intermolecular cross-links to the mechanical properties of collagen fibrils.

Introduction Articular cartilage is the resilient tissue which covers the articulating surfaces of synovial joints. The main functions of cartilage are to protect the subchondral bone from mechanical damage and to provide two bearing surfaces of low friction and wear rate. Preservation of the structure and mechanical properties of normal cartilage is essential to the existence of a healthy synovial joint. The main constituents of articular cartilage are collagen fibres, proteoglycan and water. Degradation of the collagen or t h e proteoglycan is known to change the mechanical properties of cartilage and I

* Present address: The Oxford OrthopaedicEngineeringUnit, The Nuffield Orthopaedic Centre, Headington, Oxford, U.K.

render it more susceptible to mechanical damage. Any enzyme which is capable of degrading cartilage in vivo could, therefore, contribute to the tissue damage or rheamatoid arthritis and possibly osteoa~hrosis. The effects of proteoglycan degradation on the mechanical properties of cartilage have been studied [1] by use of cathepsin D. The results showed that, after 100h incubation at 37°C, 90%, or more, of the proteoglycan was released from the cartilage specimens with a negligible release of collagen. The change in tensile properties of the cartilage, which occurred as the result of this large release of proteoglycan, was limited to a reduction in the stiffness in the initial part of the stress versus strain curve. Neither the stiffness at high stresses nor the fracture stress were significantly reduced. The effects of collagen degradation on the tensile

0304-4165/81/0000-0000/$02.50 © 1981 Elsevier/North-HollandBiomedicalPress

104

properties of cartilage were also studied in the same investigation. Using purified clostridial collagenase it was shown that, after 24 h incubation at 37°C, 20%, or more, of the collagen was released from the cartriage specimens. This resulted in a considerable reduction in the tensile strength and a reduction of the stiffness at all values of stress. It was evident from this previous investigation that degradation of collagen produced an immediate deterioration in the tensile properties of articular cartilage, whereas proteoglycan degradation did not. The degradation of molecules in the cartilage collagen fibrils can occur not only by cleavage of the helical region of the molecule, such as is brought about by the specific collagenases, but also by cleavage of the non-helical terminal peptides. The latter type of degradation leaves the helical region of the molecules intact, but eliminates most of the covalent cross-links between molecules. The effect of this on the mechanical strength has not previously been determined. An enzyme which has been shown to eliminate the covalent cross-links of cartilage collagen in this way is the elastase of neutrophil leucocytes [2]. Large quantities of the enzyme are brought into the joint affected by rheumatoid arthritis [3]. Moreover, leucocytes have recently been shown in close association with eroding surfaces of cartilage [4], and it has been demonstrated that elastase can penetrate the cartilage surface [5,6]. The low grade inflammation associated with osteoarthrosis [7] may also allow neutrophil elastase to contribute to the tissue damage. It was found that, at a concentration of 1.3/xg/mg of tissue, leucocyte elastase released approx. 20% of the collagen from the articular cartilage of pigs in the age range from 6 to 12 months [2]. Similar tests on two human joints aged 55 and 67 years released 7 and 14% of the collagen, respectively. Most of the proteoglycan was also released from the cartilage by the action of elastase. It has also been shown that the creep deformation of cartilage in compression increased considerably after 12 h incubation in the presence of elastase [8]. The present investigation was performed to determine the effects of leucocyte elastase on the mechanical properties of articular cartilage in tension.

Materials and Methods

Materials Adult human articular cartilage was obtained from the femoral condyles of the knee, the femoral head of the hip and from the talus of the ankle joint. All the joints were obtained from routine post-mortem examinations, where the cause of death was unlikely to have affected the cartilage. Prior to their use, the joints were stored at -20°C, at which temperature the mechanical properties of the cartilage are not affected [9]. The India ink method [10] was used to detect early fibrillation of the articular surface of cartilage. Only areas with a normal surface were tested. Enzyme preparation Elastase was isolated from human neutrophil leucocytes as described by Saklatvala and Barrett [I 1 ]. Specimens of cartilage were exposed at 37°C to the action of leucocyte elastase at a concentration of 1.0/xg/mg of tissue in saline (0.08 M) containing 0.02 M Tris-HC1 buffer at pH 8.0 and 1,1,1 ,-trichloro2-methylpropan-2-ol (0.2%) and pentachlorophenol (1 mg/1) as preservatives. Apparatus and tension tests The apparatus and the method used for testing cartilage in tension in planes parallel to the articular surface have been described previously [12]. Four areas on the femoral condyles of the knee were selected for testing, namely the anterior and central regions of both the medial and lateral condyles. One of the two areas on each of the condyles was aligned and tested in the direction parallel to the predominant alignment of the collagen fibres in the superficial zone, whilst the other specimen was oriented perpendicularly to the collagen fibres. Four areas from the femoral head of the hip joint were selected for testing, namely the superior, inferior, anterior and posterior regions. Two regions were oriented and tested in the direction parallel to the collagen fibres, whilst the other two regions were oriented perpendicularly to the collagen fibres. Two areas from the talus of the ankle joint were tested, namely the anterior and posterior regions. One area was oriented parallel to the collagen fibres and the other perpendicular to them.

105 The cartilage was sectioned into layers, 200/.tm in thickness, from which the superficial layer and the fourth layer below the surface were selected for testing.

length). The tensile stiffness of the specimen at any particular value of stress was determined from the gradient to the stress versus strain curve at that stress. The tensile strength, or fracture stress, was the stress at which the specimen fractured.

Experimen tal procedure Before being tested mechanically, each specimen was equilibrated for at least 30 min in buffered saline. During the tension test, the specimen was immersed in Tris-HCl-buffered saline at pH 8.0 and extended at a rate of 5 mm/min up to a stress of approx. 50% of the fracture stress. This latter value was determined by fracturing adjacent control specimens. After the initial tension test, each specimen was allowed to recover for at least 30 min in the buffered saline after which time the experimental specimens were placed in 1 ml of enzyme solution and the control specimens were placed in an equal volume of buffered saline. Both the experimental and control specimens were then incubated in their respective solutions for 24 h at 37°C. After this period, they were removed from the solution, re-equilibrated for 30 rain in fresh buffered saline and re-tested in tension. After the completion of a tension test, the incubation medium was stored at -20°C and the specimen was placed in 1 ml of fresh enzyme solution or buffered saline and incubated for a further 24 h. This procedure was repeated until each specimen had been incubated for a total of 72 h. After the final test, the specimen was also stored at -20C. The results of the tension tests were presented in the form of graphs of nominal tensile stress (applied force/original cross-sectional area of gauge section) versus strain (extension of gauge length/original gauge

Chemical analysis Both the incubation media and the specimens were analysed for their proteoglycan contents (Farndale, R., Sayers, C.A. and Barrett, A.J., unpublished data). The estimation of collagen as hydroxyproline after acid hydrolysis was made by a method based on those of Burleigh et al. [13] and Tongaard [14]. The sensitive modified method allowed the quantification of as little as 1% of the collagen being solubilised. Results

Tensile strength The effect of elastase on the tensile strength of cartilage is presented in Table I. The enzyme produced a marked decrease in the tensile strength of cartilage, irrespective of the joint from which the tissue was excised, the distance of the specimen below the articular surface or the orientation of the specimen with respect to the collagen fibres in the superficial zone. After the final tension test, each specimen was treated with India ink and examined under the microscope. In no case was fibrillation of the cartilage observed.

Tensile stiffness Table II shows the results of the action of elas-

TABLE I EFFECT OF ELASTASEON THE TENSILE STRENGTHAND STIFFNESS OF CARTILAGEFROM THE HIP, KNEE AND ANKLE JOINTS Joint

Percentage reduction in tensile strength Number of specimens

Range

Percentage reduction in tensile stiffness Mean

Number of specimens

At stress of 1 MN/m2 Range

Femoral condyles of knee Femoral head of hip Talus of ankle

68 16 15

0-79 0-73 0-89

42 37 46

52 8 15

8-81 2-69 7-71

At fracture stress Range

Mean

0-80 0-49 0-88

26 22 30

Mean 40 37 23

106 TABLE II INFLUENCE OF SPECIMEN ORIENTATION WITH RESPECT TO THE COLLAGEN ALIGNMENT IN THE SUPERFICIAL ZONE AND DEPTH BELOW THE ARTICULAR SURFACE, "ON THE EFFECT OF ELASTASE ON THE TENSILE FRACTURE STRENGTH OF ARTICULAR CARTILAGE FROM THE HIP, KNEE AND ANKLE JOINTS

Orientation of specimens to surface collagen

Depth below articular surface

Parallel Parallel Perpendicular Perpendicular

Range

Number of specimens

Surface 4th layer Surface 4th layer

29 30 20 20

0-79 8-89 0-79 5-82

::

i

/

co

/

/

/"

.."

..:'•

."

(3

A ."

Fracture stress

45 39 41 43

23 23 16 13

Range

Mean

8-62 0-63 19-71 26-79

34 34 38 51

0-66 0-69 0-88 2-58

Mean

21 18 36 26

Discussion

:::

.:

Stress of 1 MN/m z

Chemical analysis The condensed results of the chemical analysis are presented in Table III. In all specimens, a large proportion of the proteoglycan was released from the matrix by the action of elastase. In contrast, the release of collagen was less than 1% in all the specimens, except those of two joints.

/



/( •

i

Number of specimens

tase on the tensile stiffness of cartilage at two stress levels namely, 1 MN/m 2 and fracture. The results showed that the tensile stiffness at both stress levels was reduced by the action of elastase, irrespective of the source of the cartilage, its location or orientation. Some typical results are presented graphically in Fig. 1.

...

15

Mean

Range

MN/m 2

to

Percentage reduction in tensile stiffness

Percentage reduction in tensile strength

z~•" •.•••

•.•••••'••¢

o g--'/'-"........ •--••••••o2f

04

or6

Tensile Strain

Fig. 1. Typical curves of tensile stress versus strain for two specimens from the superficial and deep zones, respectively. The results show the reduction in stiffness and fracture stress after 72 h incubation in elastase at 37°(2. Superficial zone: (o) before elastase treatment; (o) after 72 h in elastase. Deep zone: (A) before elastase treatment; (zx) after 72 h in elastase. Both the fracture stress and the stiffness were considerably reduced by elastase.

Collagen fibres are the main tension-resistant elements in all connective tissues• Therefore, their presence in articular cartilage suggests that tensile stresses are present despite the fact that the normal mode of loading in vivo is compressive and perpendicular to the articular surface. This deduction is supported by an analysis of the mechanics of load bearing in cartilage [15]. Any weakening of the mesh of collagen fibres in cartilage could, therefore, allow mechanical failure of the tissue to occur rapidly. A previous investigation [1] showed that degradation of the proteoglycan in cartilage reduced the tensile stiffness in the initial section of the stress/ strain curve. It did not, however, alter the tensile

107 TABLE III BIOCHEMICALRESULTSOF THE EFFECTSOF ELASTASEON HUMANARTICULARCARTILAGE Joint

Femoral condylesof knee Femoral head of hip Talus of ankle

Number of specimens

96 16 8

Percentage proteoglycanrelease in

Percentage collagenrelease in

Elastase

Elastase

Buffer

Buffer

Range

Mean

Range

Mean

Range

Mean

Range

Mean

55-99 92-99 91-99

88 97 94

13-68 6-69 7-49

37 37 27

0-11 <1 0- 7

<1 <1 <1

<1 <1 <1

<1 <1 <1

stiffness at, or near to, fracture, nor did it reduce the fracture stress. In contrast, the effects of collagen degradation on the tensile properties of cartilage were considerable, Cleavage of the helical region of the tropocollagen molecule reduced the tensile stiffness at all levels of stress and the tensile fracture stress. It was concluded that the integrity of the collagen fibres is important in maintaining the normal tensile properties of cartilage. Leucocytic elastase is known to degrade prteoglycan in cartilage [2] and a reduction in the initial tensile stiffness, which results from proteolgycan degradation, was observed in the present investigation. However, the reduction in tensile strength and stiffness at high stress only occurs when the collagen fibres are degraded. Since the main action of elastase on collagen is to disrupt the major intermolecular cross-link, it follows from the mechanical results that this cross-link is important in determining the tensile properties and the structural integrity of cartilage. The results of 'chemical analysis revealed that a negligible amount of collagen was released from cartilage by elastase despite the large changes in the mechanical properties. This result was unexpected because cleavage of the terminal peptides is thought to allow spontaneous denaturation of the tropocollagen molecule at 37°C and subsequent release from the cartilage matrix [2]. The possibility that denatured collagen might have been retained in the cartilage by secondary non-covalent cross-links was examined by treating the tissue with a 5% solution of acetic acid, 4 M guanidine hydrochloride or a 1.0% solution of sodium dodecyl sulphate. In all cases, the release of collagen was negligible. The apparent disparity in the amounts of collagen

released from human cartilage by elastase in the present study and the previous investigation may be attributed to a difference in the activity of the enzyme in the incubation media. In the present investigation, it was found that the activity of elastase in the media at the end of an incubation period was considerably lower than that at the beginning of the incubation period. Leucocyte elastase is unstable in dilute solutions at low ionic strength and it is likely that this was the cause of the loss of activity. Conclusions The present investigation has confirmed the degradative action of leucocytic elastase on the proteoglycan of adult human articular cartilage in vitro. It has also been shown that elastase considerably reduces the tensile strength and stiffness of cartilage with a negligible release of collagen. We interpret this as showing that the tensile strength of articular cartilage is very sensitive to even limited breakdown of the main covalent cross-link between collagen molecules through degradation of the terminal non.helical peptides. Barrett [3] has pointed out that very large amounts of elastase are brought into a joint affected by rheumatoid arthrisis, and that if unchecked, the enzyme would very rapidly destroy the articular cartilage. There are, however, sufficient amounts of proteinase irthibitors, notably oq-proteinase inhibitor and a2-macroglobulin, to ensure that there is seldom, ff ever, overt proteolytie activity in the synovial fluid. A more likely route for the entry of leucocyte enzymes into the cartilage matrix is directly from leucocytes adherent to the cartilage surface. Immune

108 complexes often present in cartilage in rheumatoid arthritis m a y well attract leucocytes to the tissue, and stimulate them to release their stored lysosomal enzymes there [4,6,16]. It is also interesting to note that cartilage which has become fibrillated in osteoarthrosis contains a normal a m o u n t o f collagen, but is nevertheless extremely weak in tension, like cartilage treated with elastase in our experiments. Conceivably, a proteinase with a limited action on collagen like that o f elastase may be responsible for this degradation, too.

Acknowledgement The authors acknowledge a grant from the Arthritis and Rheumatism Council in support o f this work.

References 1 Kempson, G.E., Tuke, M.A., Dingle, J.T., Barrett, A.J. and Horsfield, P.H. (1976) Bioehim. Biophys. Acta 428, 741-760 2 Starkey, P.M., Barrett, A.J. and Burleigh, M.C. (1977) Biochim. Biophys. Acta 438,386-397 3 Barrett, A.J. (1978) Argents and Action V8/1-2, 11-18

4 Molar, W. and Wessinghage, D. (1978) Zeit. Rheumatol. 37,81-86 5 Janoff, A., Feinstein, G., Malemud, C.J. and Elias, J.M. (1976) J. Clin. Invest. 57,615-624 6 Menninger, H., Putzier, R., Mohr, W., Wessinghage, D. and Tillmann, K. (1980) Zeit. Rheumatol. 38, 145-156 7 Dieppe, P.A., Huskisson, E.C. and Willoughby, D.A. (1980) in The Aetiopathogenesis of Osteoarthrosis (Nuki, G., ed.), pp. 117-123, Pitman Medical, U.K. 8 Menninger, H., Putzier, R., Mohr, W., Hering, B. and Meiran, H.D. (1979) in Biological Functions of Proteinases (Holzer, H. and Tscheache, H., eds.), pp. 196-206, Springer Verlag, F.R.G. 9 Kempson, G.E., Spivey, C.J., Swanson, S.A.V. and Freeman, M.A.R. (1971) J. Biomech. 4,597-609 10 Meachim, G. (1972) Ann. Rheum. Dis. 31,457-464 11 Saklatvala, J. and Barrett, A.J. (1980) Biochim. Biophys. Acta 615,167-177 12 Kempson, G.E., Muir, H., Pollard,C. and Tuke, M. (1973) Biochim. Biophys. Acta 297,456-472 13 Burleigh, M.C., Barrett, A.J. and Lazarus, G.S. (1974) Biochem. J. 137, 387-398 14 Tongaard, L. (1973) Scand. J. Clin. Lab. Invest. 32, 351355 15 Weightman, B. and Kempson, G.E. (1979) in Adult Articular Cartilage (Freeman, M.A.R., ed.), pp. 291333, Pitman Medical, U.K. 16 Barrett, A.J. and Saklatvala, J. (1981) in Texbook of Rheumatology (Harris, E.D., Jr., Krane, S.M., Sledge, C.B. and Ruddy, S., eds.), pp. 195-209, W.B. Saunders, Philadelphia