Effects of movement and weightbearing on the glycosaminoglycan content of sheep articular cartilage

Effects of movement and weightbearing on the glycosaminoglycan content of sheep articular cartilage

OR IG I N A l ARTie LE Effects of movement and weightbearing on the glycosaminoglycancontent of sheep articular cartilage K Houlbrooke KVause J ml,a...

950KB Sizes 0 Downloads 1 Views


Effects of movement and weightbearing on the glycosaminoglycancontent of sheep articular cartilage

K Houlbrooke KVause J ml,att.h~~~me..'Uj~~

Immobilised non-weightbearing joints show significant loss of proteoglycan. ,It is unclear whether this is due toa lack of compression, a lack of movement, or both. Evidence suggests thatmovementalone isinsufficientto maintain proteoglycan levels. To investigate this problem, 20 sheep were divided into four groups and the right forelegs subjected to(1) normal weightbearingandmovement, (2) movement withoutweightbearing, (3) weightbearingwithout movementand(4) non-weightbearing without movement. After one month, full thickness core samples of cartHage taken from the radiocarpal joints of both forelegs were analysed for glycosaminoglycan (GAG )content.lmmobilised non..weightbearing joints showed asignificant loss of GAG compared with 'the other groups. More importantly, movement alone without weightbearingwas sufficient tomaintai n GAG content, as wasweightbearing without movement. [Houlbrooke K, Vause K,Merrilees MJ: Effects of movement and weightbearing on theglycosaminoglycan content of sheep articular cartilage. Australian Journal of Physiotherapy 36: 88-91, 1990]

Key Words: Cartilage. articular; Glycosaminoglycans; Immob~lisation;




K Houlbrooke, NZRP, ADP(MT), DipMT, is an occupational physiotherapist for Air New Zealand, Auckland, New Zealand. KVause, NZRP,ADP(MT},DipMT, is aphysiotherapistin private practice in MountMaunganui, New Zealand. MJ Merrilees, BSc(Hons), PhD, is Associate Professorinthe Departmentof Anatomy, School of Medicine, University of Auckland, Private Bag, Auckland, New Zea land.(correspondence)

hen a normally weightbearing joint is subjected to a period of immobilisation and nonweightbearing, there is a significant loss of matrix proteoglycans of the articular cartilage (Caterson and Lowther 1978, Palmoski et al 1979 and 1980, McDonough 1981, Lowther 1985, Kiviranta 1987). Since the ability of articular cartilage to withstand compressional forces.is dependent on an adequate level of matrix proteoglycans, maintenance of proteoglycans is crucial for proper function. Decreased levels may predispose the cartilage to damage if load is suddenly applied. This is, of course, of concern to physiotherapists who frequently have to evaluate and treat patients who present with a limited range of joint movement following a period of immobilisation and non-weightbearing. What is not clear, however, is whether the loss ofproteoglycans, or their constituent glycosaminoglycans (GAG), is due to non-weightbearing, or immobilisation, or both. Caterson and Lowther (1978) demonstrated, by the immobilisation ofthe sheep foreleg in a bent leg plaster for one month, that the combination ofimmobilisation and non..;weightbearing resulted ina 20 per cent decrease in the GAG content of the articular cartilage and a 47 per cent decrease in synthesis. In a similar study, Palmoski et al (1979) immobilised the right hind leg in four dogs and showed a 20 per cent decrease in GAG content and a 41 per cent decrease in synthesis over a six

week period. In a further study, Palmoski etal (1980) amputated the lower right hind foot in four dogs, allowing them to move the ankle joint normally duringambulation but not to bear weight. These joints showed a similar decreasein GAG and the authors concluded that the maintenance of the articular surface· was primarily due to compression rather than joint movement. Amputation, however, is a traumatic event and it is debatable whether the loss of GAG under these circumstances can be extrapolated to patients who maybe capable of exercising joints of whole limbs in the absence of weightbearing. To further investigate whether movement plays a role in maintaining GAG content in the absence of normal weightbearing,we have examined the GAG content of sheep radiocarpal joints subjected to four variations of movement and weightbearing.

Method Twenty female two.-;year-old sheep of the same breed and approximately the same weight (mean = 781bs, range = 121bs) were randomly assigned into four· groups: Group 1. Fully weighthearing .and free to move normally


Group 2. Non-weightbearing but with free joint movement (SWING). The right foreleg was supported by a rubber sling so as to allow


free movement ofthe radiocarpal joint (Figure I). Group 3. Fully weightbearing without joint movement (STRAIGHT). The right foreleg was immobilised in a weightbearing straightleg plaster (Figure 1). Group 4.Non...weightbearing without joint movement (BENT). The right foreleg wasimmobilised ina bentleg,non-weightbearing plaster cast (Figure I). Groups 3 and 4 were plastered with one layer each ofSotban, Gypsona and Cristona (Smith and Nephew Pharmaceuticals). No signs of deterioration of these materials were observed over the course of the study period. All sheep were grazed together and inspected daily. After one month, all sheep were reweighed and slaughtered under normal abattoir conditions. The front left and right joints were removed intact and frozen at -20°C. The left joint was used both as a group control and as an internal control for each sheep. The joints were thawed and opened under sterile conditions. Two £ul1thiclmesscore samples (approximately 5mm diameter) of cartilage were removed from the same locations of each joint {Figure 2),pooled, homogenised, and dried at 100°C overnight. The average dry weight of the combined samples from each sheep was 3.8 mg. The glycosaminoglycancontent was determined electrophoretically. The dried samples were rehydrated in 6 ml of trisbuffer for six hours before adding one ml of protease (Sigma Type VI) in buffer at a concentration of 15 mg/m1.Samples were digested for 16 hours at 50°C. Four ml of 28 per cent NaCI and O.18ml of3N acetic acid were then added and the samples heated to 100 c for five minutes. After rapid cooling in an ice bath, samples were centrifuged in polypropylene tubes at 30,OOOG for 20 minutes at 4°C and the supernatant


o RI


Figure,"1. 'Swing, straight, ,and bent leg conditions. Theswlnglegwasheldln po'sition by a rubber sling.

Fi,gore.2. ,·Qiagratnrnatic,viewof,an opened ,.Ild' flexedl~ft radiQcarpa,1 joint showing the twosamptingsites, one on the middle plateau of the

radius (upper) andtheotheroD the intermediate, carpal (lower).

transferred to glass tubes containing 25m! of ethanol and sodium acetate (0.8 gil). After 72 hours at 4°C, the samples were centrifuged at 2200G for 20 minutes at 4°C, the supernatant was discarded, the precipitate was taken up in I-2ml of distilled water and transferred to small test tubes for drying at 80...90°C. The samples were then taken up in 60,ul of distilled water and 1 ,ul aliquotselectrophoresed on cellulose acetate membranes using a 'Beckman

Microzone system, a 0.2 M zinc sulphate buffer and a front current of 1 pAlcm for 75 minutes. Membranes were stained for 8 min in 1 percent Alcian blue made up in a 1: 1 ethanol/sodium acetate (4.1 gil) solution (pH 5.8) and excess stain was removed by rinsing in an aqueous solution of 5 per cent acetic acid and 10 per cent ethanol. Membranes were dried between two glass plates. GAG bands corresponding to hyaluronic acid (HA) and chondroitin sulphates 4 and 6 (CS) were identified using co-electrophoretic standards and enzymatic digestion with Streptomyces hyaluronidase andchondroitinase ABG.Althoughchondroitin sulphates 4 and 6 could be clearly distinguished as standards, separation was less distinct in the samples and for the purposes of quantification these two GAG were combined. Since the amount of HA was less than 5 per cent ofthe total GAG, it was not analysed separately. Likewise, keratin sulphate (KS)was not analysed separately. K.S co... electrophoreses withCSand no attempt was made to distinguish between these two GAG. Thus, results are presented as total GAG. The dried strips were scanned, without clearing, on a Beckman Microzone densitometer (R.;.110) with a 600 J1m interference



grazing, the supported leg was observed on occasions to brush the ground. The leg, however, was not used for weightbearing and when ambulating, the distal part of the leg was observed to swing freely. Within all groups there was considerable variability between sheep in the amount of articular GAG and this variability and the relatively small sample size limited the power of the statistical tests.

from Page 89

filter. Calibration curves were constructed from standard GAG preparations (O.2mg/ml) and the amount of GAG expressed as J.lg. GAG per mg of dry weight.

Stati~tical methods

Comparisons between ~e groups were made using an analysis of covariance with the left radiocarpal joint as the covariate. Planned comparisons between individual pairs of groups were then made using .the Bonferroni procedure, a conservative test which requires a p value of less than 0.02 for significance. For comparisons between groups, where the change in GAG content was expressed as a change relative to the left leg of the same sheep, differences were analysed by a one-w3;Y analysis of variance.

Results -Throughout the study period, the sheep remained healthy and all groups gained weight. There were, however, small but significant ~fferences between the groups (Table 1). The control group gained significantly (p< 0.05) more weight than the swing arid bent leg groups. The straight leg group, which sHowed a weight gain similar to the

control group, also gained significantly more weight than the bent leg group with a trend (p< 0.1) towards more weight compared with the swing leg group. Sheep were observed daily for any unexpected disturbances to their behaviour. The straight leg group was the least affected and their gait was not significantly altered by the plaster cast. A mean weight gain similar to the control group is consistent with this observation. The bent leg group was more restricted in movement and the sheep ambulated on three limbs. The bent leg did not touch the ground during walking or during grazing. The swing leg group behaved similarly but when

Analysis of covariance, however, was the most appropriate with the left radiocarpal joint as the covariate. The change in GAG content in.the experimental legs was generally not large (Figure 3) but still significant on the analysis of covariance group effect (Table 2). There was a significant difference between the left and right joints in the 19 sheep (one experimental joint was lost due to autolysis as a result of overthawing). The principal finding was that the GAG content of the bent leg group was significantly reduced compared with the swing and straight leg groups. Although the GAG content of the bent leg group was not significantly different from the control group, it was nevertheless reduced and approached significance (p< 0.09). Because of the intra-group variability, comparisons were also made on the percentage change in GAG content within individual sheep using the left joint as a control for the right joint (Figure 4).



Table 1

Weight gain (lbst of sheep over the one month experimental period








1.JO } -




2.86 } -









The principal finding here was that the experimental leg of the bent leg group showed a large (40.5 per cent) and significant(p< 0.02) loss of GAG compared with their own internal controls. It should be noted that the difference between the right and left joints in this group was not due to an increase in the GAG content of the left joint. The GAG content was not significantly different from the left joints of the other groups (Figure 3). The percentage changes in GAG content between the right and left joints of individual sheep in the control, straight and swing leg groups were much smaller than for the bent leg group and were notsignificantly different from each other or from zero.

Discussion The loss of articular cartilage GAG in the·immobilisedand nonweightbearing joints (bent leg) was not unexpected and consistent with the findings of others who have investigated the effects of immobilisation andnon-weightbearing (references cited earlier). Furthermore, the amount ofGAG lost was similar to that observed by others. Of more importance, however, was the finding that maintenance of GAG in articular cartilage occurred in both the weightbearing without movement group (straight leg) and the movement without weightbearing group (swing leg), indicating thatweightbearing alone, or movement alone, is sufficient to maintain GAG content. The finding that movement alone








Table 2 Analysis of cClvariance

Dept:ndent variable = right leg Covariate = left leg Number ofsheep 19 Grorip:eff~ct / p <: /0.046 Comparison hetween groups

Control vs Bent Straight vs Bent

Swing vs Bent

p < 0.09 p < 0,02

p < 0.02

couldmaintain·GAG content is in contrast to the findings byPalmoski et al (1980), who found that the GAG content decreased in canine joints which were free to move but were not weightbearing. They concluded that movement alone was not sufficient to prevent a IQss' of GAG. Their experimental design, however, involved amputation and was different from that used in this study, although both designs may have had their deficiencies. Whereas the trauma of amputation and the altered weight may have affected the leg in their study, we cannot rule out the possibility that, in our study, the joints of the swing group were subjected to small amounts of load associated with muscle tone. Nevertheless, we believe that the experimental design used in this study was such that the results can be extrapolated to patients where non... weightbearing but moveable joints

could be considered to approximate ·conditions in the swing leg group. Furthermore, our results are consistent with those of Salter (1980), who found continuous passive motion to be ofbenefit. Thus, the practice in physiotherapy of exercising joints, even in the absence of weightbearing, with the intention of maintaining articular cartilage matrix, would seem to he worthwhile and important. The results also support the practice of maintaining the integrity of articular cartilage by encouraging regular supported weightbearing in immobilised joints.

Acknowledgements This study was supported by the New Zealand Arthritis Foundation. We also wish to acknowledge the support of ·Smith and Nephew Pharmaceuticals who donated the plastering materials and we are most grateful to Mr Doug Houlbrookewho tended the sheep during the experimental period, and to: MrMaynard Williams for statistical advice.

References Caterson B, LowtherDA (1978): Changes in the metabolism of the proteoglycans from sheep articular cartilage in response to mechanical stress. Biochimica et Biophysica Acta 540: 412422. Kiviranta I,Jurvelin], Tammi M, Saamanen A-M, Helminen HJ, (1987): Weight-bearing controls glycosaminoglycan concentration and articular cartilage thickness in the knee joints of young beagle dogs: Arthritis and Rheumatism 30: 801-809. Lowther DA (1980): The effects of compression and tension on the behaviour of connective tissues. In IdczakRM::AspectsofManipulative Therapy. Carlton: LincolnInstitute ofHealth Sciences: pp 15-21. McDonoughAL(1981): Effects ofimmobilization and exercise on articularcartilage: Areview of literature. Journal of Orthopaedic and Sports Physical Therapy 3: 2~5. PalmoskiM, Perricone E, Brandt KD (1979): Development and.reversal of a proteoglycan aggregation defect in normal canine knee cartilage after immobilisation. Arthritis and Rheumatism 22: 508-517. Palmoski MJ, ColyerRA,Brandt KD.(1980):]oint motion in the absence ofnormal loading does not maintain normal articular cartilage. Arthritis and Rhcu11llltism 23:325-334. ' Salter RB, Simmonds DF, Malcolm BW etal (1980): Thehiologicaleffectof continuous passive motion healing. on full thieknesss defeetsinarticular cartilage: An experimental investigation in the rabbit. Journal ofBoneand Joint Surgery 62A: 1232-51.