Induction of osteoarthrosis in the guinea pig knee by papain

Induction of osteoarthrosis in the guinea pig knee by papain Sigvard Kopp, Christina Mejersji, and Elisabet Clemensson, Giiteborg. Sweden DEPARTMENT G6TElBORG

OF STOMATOGNATHIC

PHYSIOLOGY.

FACULTY

OF ODONTOLOGY,

UNIVERSITY

OF

The main purpose of this investigation was to develop a model for experimental induction of osteoarthrosis. A previously described method using papain was tried on guinea pig knee joints. Eighteen adult guinea pigs were given papain intra-articularly in the right joints; the left joints were used as controls. The animals were killed after 6 hours, 1 week, 2 weeks, 2 months, 4 months, 6 months, 6 months, and 10 months. Specirmens of the articular cartilage were removed for histologic and histochemical investigation. Microscopic surface irregularities could be observed in the animals after 6 hours, 1 week, and 2 weeks, and again after 8 and 10 months. Histochemical examination of the sections from the experimental joints indicated a loss and degradation of the sulfated glycosaminoglycans. This loss was evident after 6 hours, 1 week, 2 weeks, and 6 months. The first osteoarthrotic changes were observed macroscopically after 8 monthls. Radiographic changes in the experimental joints could be observed in all animals killed after 10 monthls. It was concluded that osteoarthrosis similar to that occurring in humans can be induced by this method.

H

uman temporomandibular joint (TMJ) osteoarthrosis has been extensively studied in clinical, radiographic, and pathologic aspects,but there is a lack of knowledge as to how the articular tissues react to the treatments used in the clinic. An experimental animal model is needed to investigate this issue further. Many methods of inducing osteoarthrosis in animal joints have been described in the literature, and they can broadly be divided into metabolic, endocrine, chemical, biomechanical, and immunologic methods. For additional information, the reader is referred to a review by Moskowitz.’ The biomechanical methods are probably most realistic but require relatively complicated surgical manipulations.2*3 The chemical methods generally entail intra-articular injection of various substances and are relatively easy to perform. The changes resulting from injection of papain into rabbit knee joints have been similar to those seen in human osteoarthrosis’6 and have been interpreted as the effect of normal joint loading on cartilage damaged by papain. Osteoarthrosis has also consistently been produced by papain in the TMJ of adult rabbits.’ This method .usingintra-articular injections of papain was chosen to induce osteoarthrosis in the guinea pig knee. A pilot study had shown that it is difficult to inject accurately into the TMJ of the guinea pig

becauseof its anatomy and small size. This joint was therefore not suitable for this study. The primary aim of this study was to develop an experimental model for induction of osteoarthrosis, with special emphasis on the early his&chemical changesin the joint cartilage. The model will later be used to study the effect of various treatments on these changes. MATERIAL AND METHODS Animals

Eighteen adult (approximately 1 year old) male guinea pigs weighing between 820 and 1195 grams (mean 1,006 grams) were used. The intra-articular injections were performed with the animals under general anesthesia attained by means of pentobarbital (Mebumal Vet, ACO, 60 mg./ml.) administered intraperitoneally (25 to 30 mg./kg.). Intra-articular

injections

of papain

Papain (crystallized twice from Papaya Latex; Sigma P 3 125) was dissolved in acetate buffer (pH 5.6) with addition of thymol(O.01 percent) to inhibit bacterial growth. A 2 percent solution was made by dilution in saline solution. m-cysteine hydrochloride hydrate was used as activator in a concentration of 0.05 mol./L. (Sigma C 8256) and was added to the enzyme solution immediately before injection. The 259

260

Kopp, Mejersjii,

and Clemensson

Oral Surg. March, I983

Fig. 1. A, Slight changesof the joint surfaces after 10 months and two papain injections. The joint surface of the medial condyle of the femur is thickened and fibrillated (arrow). B, The control joint of the same animal showing a normal appearance.

enzyme preparation plus activator was injected into

the right knee joint (0.10 + 0.05 ml.) with aseptic technique, whereas the left joint was not subject to any treatment but was used as a control joint. The injection was made from the lateral side behind the superior part of the patella and musculus quadricipitis into the suprapatellar bursa. The outer diameter of the cannula was 0.4 mm. The animals were given either one or two injections and were killed 6 hours, 1 week, 2 weeks, 2 months, 4 months, 6 months, 8 months, and 10 months after the first injection. The second injection was given 1 month after the first (Table I). One of the animals died accidentally 1 hour after the first injection. Macroscopic

evaluation

of articular

surfaces

The knee joints were dissected immediately after death, and the articular surfaces were examined for macroscopic lesions with the naked eye. The surface changeswere described as increased thickness of the articular cartilage, superficial fraying, fissuring and flaking, erosion (loss) of cartilaginous tissue, and bone exposure. Artiiular

surface specimens

Immediately after the macroscopic evaluation, full-thickness soft tissue specimens, approximately 2 x 2 mm., were cut out of the central part of the summit of the medial femoral condyle, which is supposed to be subject to the most weight bearing. Specimensof adjacent synovial membrane were also removed. The specimens were put in Histocon (a transport medium manufactured by Histo-Lab, Bethlehem Trading, Ltd., Giiteborg, Sweden) before being embedded in OCT compound (same manufacturer) and frozen in isopentane and liquid nitrogen.

They were then kept frozen at a temperature of -70” C. until they were sectioned in a cryostat set on 7 pm. Microscopic

evaluation

of articular

surface

Sections of the specimens from the femoral condyle were stained in Mayer’s hemalum-eosin solution* and examined under the light microscope for

surface irregularities and cell content. Sections from the experimental and the control knees were stained in a parallel manner in the same solution. Toluidine

blue staining

Cryostat sections were immersed in a solution of 0.1% toluidine blue 0 (No. 1273, E. Merck AG, Darmstadt, West Germany) in OSN HCl to stain the sulfated glucosaminoglycans (GAG) metachromatitally. The staining procedure used has been described by Chayen and co-workers.9 The sections were examined for metachromasia intercellularly and pericellularly. Alcian blue staining

Sections were also stained with alcian blue 8 GS (Chroma Gesellschaft, Stuttgart-Untertiirkheim, West Germany). Increasing amounts of magnesium chloride were added to the dye solution (0.05% alcian blue in 0.025M acetate buffer, pH 5.6) to differentiate the GAG according to the critical electrolyte concentration (CEC) method.“‘, ” The following molar concentrations of MgClz were used: 0.00, 0.10, 0.15, 0.25, 0.35, 0.45, 0.55, 0.65, 0.75, 0.85, and 1.00. The value of the CEC is given as the mean of the two concentrations of MgCl, between which the alcian blue staining disappears. The CEC determinations were made intercellularly and peri-

Volume

55

Number

3

Induction of osteoarthrosis 261

Table 1.Changes induced in articular cartilage of guinea pig knee after various time intervals and numbers of papain injections

I

Deviation in shape/macroscopic lesions in the articular surface

Microscopic surface lesions (irregularities)

No.

Interval

Number of injections

1 2 3

1 hr. 6 hr.

1 I

6 hr.

1

4

I wk. 1 wk. 2 wk. 2 wk. 2 mcl.

I I 1 1 1

2 mo. 4 mci.

2

+

I

+

4 mci. 6 mo. 6 mo.

2

5 6 I 8 9 IO II

12 13 14 15 16 17

IO mo. 10 mo.

18

IO mo.

+ + + + + +

M

I-

I 1

I

I-

1

1

IIIII-

P+ P+ P+ P+

II-

P+ P+

I

II-

P+ P+

1

I-

2 2

II-

2 1

2

2 3 3 2

I

1

1

1 1

2

2 DS, F, I

+ +

DS, BE, 1 DS, E, I

+ +

DS, E, I

+

P+

Radiographic changes

1

2

1

2 2 2 2

Metachromasia

P

1

I

8 mo. 8 mo.

_

Inflammatory cells in synovial membrane*

I

E, C S DS, S DS, C, S

P = Polymorphonttclear cells; M = mononuclear cells; I- = metachromatic reactions absent intercellularly; P+ = metachromatic reactions present pericellularly; DS = deviation in shape; F = fibrillation of cartilage; I = increased thickness of articnlar cartilage; BE = cartilage erosion down to bone exposure; E = erosion of the cortical outlining; C = cystic destruction of bone; S = bone sclerosis. *The scale used for estimating the number of inflammatory cells was as follows: I = solitary cells; 2 = small number of cells; 3 = moderate number of cells; 4 = large number of cells.

cellularly in two different zones of the cartilage: (1) the cartilage zone, which comprises the outer two thirds of the articular surface, and (2) the mineralized cartilage zone, which comprises the deeper third. The reproducibility of the CEC determinations was investigated by duplicate readings of the sections with a l-month interval. The last reading was used in the analyses. Microscopic: evaluation

of synovial membrane

Sections stained with hematoxylin-eosin mixture were examined for inflammatory cells. The type of cells were identified (polymorphonuclear or mononuclear) and the amount of cells were estimated throughout the whole section according to the following scale: 1. Solitary cells 2. Small number of cells 3. Moderate number of cells 4. Large number of cells Radiographic

examination

The knee joints were first exposed en bloc on ordinary dental film (Kodak periapical ultraspeed

film, DF 57, CAT 165 8210). The femoral component was then cut sagittally in the midline into medial and lateral parts that were exposedseparately. The radiographs were examined for deviation in shape (signs of remodeling) and structural changes (erosions, sclerosis, cysts). All assessments were made by two experienced observers in cooperation and performed as comparisons between the experimental and control joints. RESULTS Macroscopic

changes

The first obvious osteoarthrotic changes were observed after 8 months and two injections as extensive fibrillation of the articular surfaces, which were also thickened in some areas. After 10 months and two injections, extensive erosion of cartilage with bone exposure was observed in one of the animals. The other two showed a marked thickening of the articular surface in someareas and erosions in others (Fig. 1). The articular surface area was also extended (Table I). Two of the control joints showed slight fibrillation after 10 months, and the remaining 16 control joints were normal.

262

Kopp. Mejersjii, and Ciemensson

Oral Slug. March,

1983

Fig. 2. Photomicrographs showing specimensstained with toluidine blue 0 in OSN HCI. A, Experimental joint 1 week after papain injection, with superficial fraying of the joint surface and loss of metachromasia (arrows] and cell necrosis. B, The joint surface specimen from the control joint of the same animal. (Magnification, x230.)

Fig. 3. The reproducibility of the determination of the CEC of alcian blue staining between two readings with a l-month interval. A, Intercellular staining. B, Pericellular staining. Microscopic

changes

Surface irregularities could be observed in all animals after 6 hours, 1 week, and 2 weeks. This finding was also made in the 2-month animal receiving two injections but not after 4 and 6 months. The surface irregularities then reappeared in the animals 8 and 10 months after induction. A marked thickening of the articular surfaces could be observed in the &month animal receiving two injections and in all the animals after 10 months (Table I). In two of the animals going 10 months, chondrocyte necrosis could

be observed in the surface zone of the cartilage. All specimens from the control joints were normal. Metachromasia

All sections from the control joints showed an even intercellular metachromasia except for the outermost surface lining. Sections from the right experimental knee, on the other hand, showed markedly less intercellular metachromasia in the animals killed after 6 hours, 1 week, and 2 weeks, reqectively, as well as in the animal receiving two injections

Znduction of osteoarthrosis 263

Volume 55 Number 3

and hilled after 2 months (Fig. 2). A difference in intercellular metachromasia was also evident in the animals after 8 months. Pericellular metachromasia was found to be increased in the right knee in the animals hilled after 1 and 2 weeks. The same finding was made in the 2-month animal receiving two injections. It then reappeared in the animals killed after 8 months. Pericellular metachromasia was absent in the animals after 10 months. Alcian blue staining

In the 36 specimens examined twice for CEC of alcian blue staining, a total of 288 readings were made. Total agreement was found for 49 percent (35/72) of the duplicate readings of intercellular staining (Fig. 3, A) and for 60 percent (43/72) of the duplicate rea.dings of pericellular staining (Fig. 3, B). The standard deviation of a single determination of the CEC was 0.7 degree on the 1l-point scale for both intercellular and pericellular staining. There was no significant difference between the CEC values of the experimental and control joints as a whole, but there were some notable differences in animals after certain periods. Becauseof the error in determining the CEC value, only differences of two or more points on the scale were considered. A lower value of the intercellular CEC in the cartilage zone was observedin the experimental joint of both animals after 6 hours. A lower value of the pericellular CEC in the experimental joint was observed in 1.~0 animals of the lo-month group. There was no consistent difference in CEC of intercellular alcian blue staining between experimental and control joints in the mineralized cartilage zone, but the pericellular staining had a higher value in the experimental joints after 6 hours and a lower value in two of the experimental joints after 10 months. Synovial membrane

No animal showed an intense inflammation ‘of the synovial membrane. A gradual shift from polymorphonuclear cells to mononuclear cells occurred after the first 2 weeks. The samples from the synovial membrane 1 and 2 weeks after induction showed the strongest inflammatory reaction. The mononuclear reaction was dominated by plasma cells (Table I). Three control knees also showed a slight inflammatory reaction. Radiographic

changes

Structural changes in the mineralized tissues of the experimental kneejoints could be observedradio-

Fig. 4. A, Radiographicchangesafter 10 months and two injections.The cortical outline is diffuse,especiallyin the femoralcomponent,and there is subchondralsclerosis in the tibia1component.Cystic destructionscan he seenin the central and lateral parts of the femur {urrows~.B, Radiographicappearanceof the control joint of the same animal (frontal plane).

graphically in the animals after 10 months and two injections (Fig. 4; Table I). The changeswere erosive in character in one animal; i.e., there was a demineralization of the cortical outline of the surface. Both the femoral and tibia1 surfaces were involved, especially the medial femoral condyle, which also demonstrated cystic destructions surrounded by sclerotic parts in the interior of the bone. The other two animals demonstrated increase in size and sclerosis of the medial femoral condyle and the corresponding part of the tibia. In one of these two experimental joints the lateral femoral condyle was also sclerotic, whereas the other demonstrated cystic destructions. The control joints showed a normal radiographic appearance. DISCUSSION

According to current thinking, the basic macroscopic’2and microscopicI3714 features of osteoarthrosis in the TMJ are similar to those in other synovial joints. The aim of the investigation was to develop a model of osteoarthrosis that would reproduce especially the early histochemical changes in the soft tissue joint surface. As a means of inducing osteoarthrosis, papain was

264

Kopp, Mejersjii,

and Clemensson

Oral Surg. March. 1983

Fig. 5. Photomicrograph showing alcian blue staining of a control joint specimen, with A, 0.45M; B, OSSM; C, 0.65M; and D, 0.85M MgCI,. Note the strong pericellular staining. (Magnification, X230.)

chosen becauseof the catabolic effect on the proteoglycan molecule and its consistency and simple management. The amount of enzyme used in this study was calculated according to the size of the joint cavity of the guinea pig knee compared to that of the animals used in previous studies. Possible disadvantages of the method are that the diseaseis induced in an unrealistic way by the introduction of foreign proteins into the joint cavity and that it elicits an inflammatory response. The acute phase of the cellular inflammatory responsein the synovial membrane lasts for at least 2 weeks,and the joint swelling subsides after a week. The introduction of papain

and cysteine into the tissues gives rise to a heavy immune response, especially after two injections, which ultimately leads to their removal. The initial damage to the surface lining and the ground substance of the cartilage is probably produced by the enzymatic action of papain, perhaps aggravated by the subsequent inflammation. Microscopic surface lesions develop rapidly and consistently in association with loss of intercellular sulfated GAG. The initial development here is in accordance with the hypothesis of pathogenesis presented by Freeman.‘j These osteoarthrotic changes can be demonstrated for up to 2 weeks. The absence

Volume Number

Induction of osteoarthrosis 265

55 3

of changes after 4 to 6 months cannot be explained with certainty, but it might be speculated that a temporary healing of the initial damage occurs. This healing might be a result of the increased synthesis of sulfated GAJG, as demonstrated by the marked increase in pericellular metachromasia. In addition, the inflammatory reaction of the synovial membrane is low during this period. Macroscopic and radiographic signs of osteoarthrosis develop relatively late, after 8 and 10 months, respectively. This later development is probably due to the mechanical insult of normal loading on the already damaged cartilage. It seems that the initial damage to the cartilage produced by the enzymatic degradation leads to destruction of the cartilage after a period of function, in spite of attempts to repair. This conclusion is in agreement with previous results obtained with this method.4The incipient osteoarthrotic changes in the two control joints after 10 months are probably due to increased functional demand as a consequenceof the gross damage to the experimental joints. The increased synthesis of sulfated GAG, as indicated by increased pericellular metachromasia, is probably responsible for the thickening of cartilage and the exten.sionof the cartilage surface area found after 8 to 10 months. This reaction is probably an attempt to replace lost tissue and to restore the functional capacity of the cartilage. This attempt, however, has terminated and failed after 10 months, since chondrocyte necrosis is present in the surface layer of the cartilage and the synthesis of sulfated GAG has ceased. The loss of sulfated GAG from the cartilage zone of the joint surface was obvious, as demonstrated by loss of metachromasia. This loss is consistent with the early phase of the natural disease process’3*‘4*16 and could already be demonstrated after 6 hours. The presenceof normal metachromasia in the specimens from animals killed 4 to 6 months after the induction may again be explained as a temporary substitution. The values of the CEC of the intercellular and pericellular alcian blue staining in both the cartilage zone and mineralized cartilage zone indicated the presenceof c.hondroitin sulfate (Fig. 5). The slightly higher values in the pericellular region are probably due to a higher charge density and/or a higher molecular weight. The presence of keratan sulfate was indicatecl in the mineralized cartilage zone. The CEC values of intercellular and pericellular staining in the cartilage zone seemedto be reduced in the experimental joints, especially in the animals killed after 6 hours. This finding is probably

explained by the enzymatic degradation of the proteoglycans, leading to less negative charge density and a lower molecular weight of the macromolecules. The radiographic signs of osteoarthrosis appeared almost at the same time as the macroscopic changes (10 months). The subchondral bone sclerosis probably developed as a response to increased loading of the bone caused by the cartilage destruction. Subchondral bone sclerosis is regarded as a characteristic sign of osteoarthrosis.” It is impossible from this study to evaluate the difference in effect between one and two injections of papain. It may be concluded from this study that osteoarthrosis of a character similar to that occurring in humans can be induced consistently in the guinea pig knee by intra-articular injections of papain. Early histologic and histochemical changesin the cartilage, as well as a synovitis, develop in 6 hours, whereas macroscopic surface changes and radiographic changes develop in 10 months.

REFERENCES 1. Moskowitz, R. W.: Experimental Models of Degenerative Joint Disease, Semin. Arthritis Rheum. 2195 I 16, 1972. 2. Telhag, H., and Lindberg, L.: A Method for Inducing Osteoarthritic Changes in Rabbit’s Knees, Clin. Orthop. 86~214-223, 1972. 3. Reimann, I.: Experimental Osteoarthritis of the Knee in Rabbits Induced by Alteration of the Load-bearing, Acta Orthop. &and. 44496-504, 1973. 4. Murray, D. G.: Experimentally Induced Arthritis Using Intra-articular Papain, Arthritis Rheum. 121 I-219, 1964. Degenerative Arthritis of the 5. Bentley, G.: Papain-induced Hip in Rabbits, J. Bone Joint Surg. 53(B):324-337, 1971. L.: Degenerative Joint Disease of 6. Scheck, M., and Sakovich, the Canine Hip, Clin. Orthop. 86:115-120, 1972. V., Alhopuro, S., and Ranta, R.: Papain-induced I. Ritsill, Arthritis of the Temporomandibular Joint and Its Effect on the Occlusion in Growing and Adult Rabbits: A Preliminary Report, Proc. Finn. Dent. Sot. 69:116-l 19, 1973. 8. Romeis, B.: Mikroskopische Technik. 16. Auflage, Munich, 1968, R. Oldenbourg Verlag, p. 164. 9. Chayen, J., Bitensky, L., and Butcher, R.: Practical Histochemistry, London, 1973, John Wiley & Sons, Inc. IO. Scott, J. E., and Darling, J.: Differential Staining of Acid Glucosaminoglycans (Mucopolysaccharides) by Alcian Blue in Salt Solutions, Histochemie 5~221-233, 1965. II. Scott, J. F.: Affinity, Competition and Specific Interactions in the Biochemistry and Histochemistry of Polyelectrolytes, Biochem. Sot. Trans. 1:787-806, 1973. 12. (iberg, T., Carlsson, G. E., and Fajers, C. M.: The Temporomandibular Joint: A Morphologic Study on a Human Autopsy Material, Acta Odontol. Stand. 29349-384, 1971. 13. Kopp, S.: Topographical Distribution of Sulphated Glucosaminoglycans in Human Temporomandibular Joint Disks: A Histochemical Study of Autopsy Material, J. Oral Pathol. 5~265-216, 1916. 14. Kopp, S.: Topographical Distribution of Sulphated Glucosaminoglycans in the Surface Layers of the Human Tempo-

266

Kopp, Mejersjii,

and Clemensson

romandibular Joint: A Histochemical Study of an Autopsy Material, J. Oral Pathol. 7~283-294, 1978. 15. Freeman, M. A. R.: The Pathogenesis of Osteoarthrosis: A Hypothesis. In Appley, A. G., editor: Modern Trends in Orthopaedics, vol. 6, London, 1972, Butterworth & Company, pp. 40-94. 16. Freeman, M. A. R., and Meachim, G.: Ageing, Degeneration and Remodelling of Articular Cartilage. In Freeman, M. A. R., editor: Adult Articular Cartilage, Oxford, 1973, Alden Press, pp. 287-329.

Oral Surg. March. I983 17. Martel, W.: Radiology of the Rheumatic Diseases, J. A. M A. 224:791-798, 1973. Reprint requests to. Professor Sigvard Kopp Department of Stomatognathic Faculty of Odontology University of Lund Carl Gustafs v&g 34 214 21 Malmii, Sweden

Physiology