OSTEOARTHRITIS and CARTILAGE

Transcription

1 Osteoarthritis and Cartilage (1993) 1, Osteoarthritis Research Society /93/ $08.00/0 OSTEOARTHRITIS and CARTILAGE Alterations of proteoglycan synthesis in rabbit articular cartilage induced by intra-articular injection of papain BY SATOSHI MIYAUCHI*, AKEMI MACHIDA*, JUN'ICHI ONAYA*, TAKASHI SAKAMOTO*, KIYOCHIKA TOKUYASU* AND HISASI-II IWATA~" *Tokyo Research Institute, Seikagaku Corporation, Tokyo, Japan; t Department of Orthopaedic Surgery, Nagoya University, School of Medicine, Nagoya, Japan Summary In order to investigate the biochemical alteration of proteoglycan (PG) synthesis during cartilage repair, reversible destruction was induced by injecting papain into the knee joint cavity of rabbits. The PG synthesis in the cartilage was examined using Na 2 35SO4 and high performance liquid chromatography (HPLC). PGs labeled with 35SO42- (35S-PGs) were extracted from normal and papain-treated cartilage, and the amount of synthesis, ability to aggregate with hyaluronan (HA), and the composition of glycosaminoglycan and chondroitin sulfate isomer labeled with 35SO42- (sss-gag and 35S-CS isomer) were analyzed. Synthesis of 35S-PGs, especially those that were unable to aggregate with HA (nonaggregating 35S-PGs), increased in papain-treated cartilage compared with that in normal cartilage. The acceleration and qualitative change in PG synthesis in the papain-treated cartilage are considered to be responses to the supplementation of the loss of cartilage PGs induced by papain. The compositions of sss-gag and 35S-CS isomer of the nonaggregating 35S-PGs differed from those of 35S-PGs which were able to aggregate with HA (aggregating sss-pgs) in the papain-treated cartilage as well as in the normal cartilage. However, the compositions of both nonaggregating and aggregating ~S-PGs in the papain-treated and normal cartilage were similar. These results indicate that most of the nonaggregating sss-pgs in papain-treated cartilage have properties similar to those in normal cartilage and are not simple degradation products of aggregating 35S-PGs; they also suggest that the supplementary reaction for PG content in the cartilage during its repair process is not simple acceleration in PG turn-over but the enhancement of PG synthesis accompanied by alterations in aggregating ability and the compositions of GAG and CS isomer. Key words: Cartilage, Proteoglycans, Chondroitin sulfate, Keratan sulfate. Introduction VARIOUS methods of inducing cartilage destruction are known. The method in which proteases, such as papain or chymopapain, are injected into the joint cavity is appropriate for studies using a number of animals because the operation is simple and cartilage destruction occurs in a short period of time [1-5]. The arthritis model induced by the intra-articular injection of papain has been available for 20 years and extensive histological and biochemical studies have been performed [1-5]. This model produces cartilage destruction accompanied by a drastic decrease in proteoglycan (PG) content and abnormal mitosis of the chondrocytes [1-5]. The repair process of cartilage treated with protease depends upon the dose of protease Submitted 9 February 1993; accepted 19 November Address reprint requests to: Dr S. Miyauchi, Tokyo Research Institute, Seikagaku Corporation, , Tateno, Higashi-Yamato, Tokyo 207, Japan. injected into the joint cavity [5]. Namely, when a low dose is injected, cartilage repair is completed, although a transient decrease in PG content occurs immediately after the injection. However, when a high dose of protease is injected, irreversible cartilage destruction is induced [5]. As mentioned above, cartilage is repaired when a low dose of protease is injected. However, little information is known about the properties of PGs synthesized during this process. In this study, we attempt to understand the characteristics of PGs newly synthesized during th~ repair process in cartilage. Reversible destruction of cartilage was induced by injecting papain into the joint cavity of rabbits, and alterations in the aggregating ability and glycos~in0glycan (GAG) composition of PGs were examined using Na 2 35SO4 and high performance liquid chromatography (HPLC). Papaininduced arthritis is suitable for these objectives because this arthritis model is reversible when a low dose is injected and has extensive background data, as described above. 253

2 254 Miyauchi et al.: PG synthesis treated with papain Materials and methods INJECTION OF PAPAIN Nine male Japanese White rabbits weighing kg were used. Papain was injected according to the method of Inuoe et al. [4]. Papain powder (2.8 U/mg solid, Sigma Chemical Co., St Louis, MO, U.S.A.) was dissolved in physiological saline containing 0.4 mg/ml of L-cysteine at a concentration of 5 mg/ml. The solution was filtered through a membrane with a pore size of 0.45 #m, and 0.6 ml of the filtrate was injected with a 27 gage injection needle into the joint cavity of the left knees of the nine rabbits. The administered dose of papaimwas 3 mg/joint. The right knees of the same rabbits were used as normal controls. When papain (2.8 U/mg solid, Sigma) is injected into the joint cavity with L-cysteine at a dose of 3 mg/joint, cartilage proteins are nonspecifically degraded, and a remarkable decrease in PG content is induced [1-5]. However, we found in a preliminary study that irreversible cartilage destruction is not induced by injecting papain at this dose and that cartilage repair is completed. The phenomena during the cartilage repair process, such as the proliferation of chondrocytes and increase in PG content, can be observed 7 days after the injection of papain. The PG content recovered to the pre-injection level about 4 weeks after the injection. Furthermore, we also confirmed in a preliminary study that the activity of SH-proteases in the joint fluid decreases to a normal level 24 h after injection of papain. INJECTION OF Na Seven days after the papain injection, Na 2 3~SO4 was injected into both knees of all rabbits. Na 2 35SO4 (carrier free, concentration of radioactivity over 10 mci/ml; catalog number: JAERI S-35-1, Japan Atomic Energy Research Institute, Tokyo, Japan) was diluted with 199 medium (ph 7.2, Flow Lab., Scotland, U.K.) to a radioactivity level of 0.5 mci/ml. The solution was filtered through a membrane with a pore size of 0.45 #m, and 0.6 ml of the filtrate was injected into the joint cavities of both knees of each rabbit. The amount of Na injected was 0.3 mci/joint. DETERMINATION OF 358-PGs Three rabbits each were sacrificed 6, 24 and 48 h after the Na 2 35SO4 injection, and the femur was isolated. The cartilage of the external condyle of the isolated femur was sliced with a blade, then the slices were washed with phosphate-buffered physiological saline (PBS). They were then fixed. with ethanol, and dehydrated and dried using acetone and diethylether. After the dry weights of the cartilage slices were measured, 35S-PGs were extracted twice with 1.2 ml of 50 mm sodium acetate buffer (ph 6.0) containing 4 M guanidine hydrochloride (GuHC1) and various protease inhibitors (100'ram 6-amino-n-caproic acid, 10 mm ethylenediamine-tetraacetic acid tetrasodium salt (EDTA-Na4), 10 mm N-ethylmaleimide, and 10 mm phenylmethylsulfonyl fluoride (PMSF)), for 24 h at 4 C [6]. To remove free 3~SO42- from the extract, 0.1 ml of the solution was ultrafiltered by centrifugation (2500 g, 20 min) through an MPS-1 YMT membrane (molecular weight cut-off: , Amicon Corp., Lexingtoni MA, U.S.A.), and the membrane was washed by ~epeating this procedure twice in a similar manner u~ng 0.4 ml of PBS. The washed membrane was remsyed from the holder and placed in a scintillation Vial. PBS (1 ml) was added into the vial, and the vidt was then stirred. Next, an emulsified scintillator (6 ml, ACS-II, Amersham Japan, Tokyo, Japan) was added, and the radioactivity was measured using a liquid scintillation counter (LCS-700, Aloka, Tokyo, Japan). MEASUREMENT OF THE MOLECULAR WEIGHT DI8TRIBUTION OF 35S-PGs The distribution of molecular weight was measured using a gel permeation column (TSKgel G6000 PW, Tosoh Co. Ltd, Tokyo, Japan) attached to an HPLC system (Waters M600, Japan Millipore Ltd, Tokyo, Japan). The operating range of the molecular weight with this column was from 40 x 103 to 8000 x 103 for polyethylene glycol and from x 103 to 2 x 103 for hyaluronan (HA). The sample was eluted under both dissociative and associative conditions. We used HPLC to rapidly measure the molecular weight distribution of 35S-PG. Under dissociative conditions, to protect the HPLC pump and the column, 7 M urea was used as the elution buffer instead of 4 M GuHC1. Under associative conditions, probably because the column packing has a high affinity for proteins, aggregates of 35S-PGs with HA partially disaggregated during the separation. However, this disaggregation was prevented by adding sodium hyaluronate (Na-HA) with a molecular weight of (Seikagaku Corporation, Tokyo, Japan) to the elution buffer. Na-HA with a molecular weight of is almost completely excluded by a TSKgel G6000 PW. ~

3 Osteoarthritis and Cartilage l. 1 No Under dissociative conditions, samples were eluted with 50 mm sodium acetate buffer (ph 6.0) containing protease inhibitors and 7 M urea. Under associative conditions, samples were eluted with 50 mm sodium acetate buffer (ph 6.0) containing protease inhibitors and 10 pg/ml of Na-HA with a molecular weight of 2160 x 103. The flow rate was 1.0 ml/min. Fractions of 0.5 ml were collected, and the radioactivity in each was determined using the liquid scintillation counter. A 0.5 ml GuHC1 extract derived from the cartilage slices was ultrafiltered as described above. The residual 35S-PGs on the membrane were shakeextracted with 1 ml of each elution buffer, and the extract was centrifuged at g for 15 rain. A 0.5 ml sample of the supernatant was analyzed by the HPLC. FRACTIONATION OF 35S-PGs 35S-PGs were extracted from the normal and papain-treated cartilages obtained 24 h after the injection of Na2 35SO4 and then fractionated individually by a TSKgel G6000 PW under the associative conditions described earlier. Fractions were combined and designated as the aggregating fraction with HA; fractions were also combined and designated as the non-aggregating fraction. Therefore, a total of four combined fractions were obtained. To remove HA from the four combined fractions, 100 TRU of hyaluronidase derived from Streptomyces (Seikagaku Corp.) was added to each, and the mixtures were incubated at 37 C for 2 h. After enzymatic digestion, each mixture was dialyzed against distilled water at 4 C for 48 h, and the inner solution was lyophilized. The lyophilized powder was dissolved in 1 ml of distilled water and used to study digestion with chondroitinase ABC derived from Proteus (CHase ABC, Seikagaku Corp.) and with keratanase derived from Pseudomonas (KSase, Seikagaku Corp.). MEASUREMENT OF THE MOLECULAR WEIGHT DISTRIBUTION OF 35S-PGs AFTER ENZYMATIC DIGESTION A TSKgel G5000 PWxL (Tosoh Co. Ltd) was used in gel permeation chromatography (GPC) after enzymatic digestion. The operating range of this column was from 4x 103 to for polyethylene glycol and from 20 x 103 to x 103 for HA. The elution buffer of the GPC was 100 mm sodium phosphate buffer (ph 7.2) containing protease inhibitors, and the flow rate was 1.0 ml/min. Fractions of 0.5 ml were collected, and the radioactivity in each was measured using the liquid scintillation counter. A 0.2 ml sample of each combined fraction was mixed with 100 mm sodium phosphate buffer (ph 7.2) containing protease inhibitors and 5 U/ml of CHase ABC, or the same buffer also containing 10 U/ml of KSase, and the mixtures were incubated at 37 C for 1 h. The resulting mixtures and the untreated combined fraction were injected into the HPLC. SEPARATION OF UNSATURATED DISACCHARIDES DERIVED FROM 35S-CHONDROITIN SULFATE Unsaturated disaccharides derived from chondroitin sulfate (CS) were separated using HPLC according to the method of Yoshida et al. [7]. The column used was Radialpak p-bondapak-nh 2 (Waters, Japan Millipore Ltd). The unsaturated disaccharides derived from 3~C-CS (~ss-adi-4s and 35S-ADi-6S) were eluted with a linear mm gradient of NaH2PO 4 for 60 rain. A 0.2 ml sample of 100 mm sodium phosphate buffer (ph 7.2) containing protease inhibitors and 5 U/ml of CHase ABC was added to 0.2 ml of each combined fraction, and the mixture was incubated at 37 C for 1 h. The mixture was then ultrafiltered with an MPS-1 YMT membrane, and 0.35 ml of the filtrate was injected into the HPLC. Fractions of 0.5 ml were collected, and the radioactivity in each corresponding to the eluting position of 35S-ADi-4S or 35S-ADi-6S was determined using the liquid scintillation counter. STATISTICAL ANALYSIS The mean and standard deviation was calculated for each group, and the significant difference was determined by the Student's paired t-test, since the normal and papain-treated cartilage were obtained from the right and left knees of the same rabbit, respectively.... Results THE AMOUNT OF 35S-PG EXTRACTED FROM THE NORMAL AND PAPAIN-TREATED CARTILAGE The percentage of extracted radioactivity was about 91%, and no difference was observed between the normal and papain-treated cartilage. In the normal cartilage, the amount of synthesized 35S-PG was maximal 6 h after the injection of Na 2 35SO4 and tended to decrease gradually thereafter. In the papain-treated cartilage, the amount of synthesized 35S-PG was greater at any time point than that in the normal cartilage (Fig. 1).

4 256 Miyauchi et al.: PG synthesis treated with papain Papain-treated and in chromatograms 6 and 24 h after the injection, the latter peak was a shoulder of the former. At any time point, the chromatogram profiles were clearly different from those of the normal cartilage (Fig. 2). X r THE MOLECULAR WEIGHT DISTRIBUTION UNDER ASSOCIATIVE CONDITION I I, I Time after injection of Na235SO4 (h) FIG. 1. The amount of ass-pgs extracted from the normal and papain-treated cartilages. Each point represents the mean +_ S.D. of three animals. THE MOLECULAR WEIGHT DISTRIBUTION UNDER DISSOCIATIVE CONDITION In the normal cartilage, two peaks were obtained at fractions and In the papain-treated cartilage, the separation between peaks at fractions and was incomplete, In the normal cartilage, a peak of 85S-PG aggregating with HA was observed at fractions (V0: void volume). In the papain-treated cartilage, two peaks were present in and fractions The difference in the chromatograms between the normal and papain-treated cartilage was much more obvious than that in those obtained under the dissociative condition (Fig. 3). PROPORTIONS OF THE ~GGREGATING AND NONAGGREGATING FRACTIONS In the chromatograms shovsn in Fig. 3, fractions were combined and desig~nated as the aggregating fraction with HA, and fractions as the nonaggregating fraction. The proportion of the total represented by each of these combinations 300 i i (a) 6h h V t 300 " 48 h t i 150 0! ! 10 2" 150! 2O 30 4O 50 (b) V~ i, i i,i o 300 6h h 30( 48 h 150 \ 15 15( Fraction no. Fro. 2. Elution profiles of 3~S-PGs on a TSKgel G6000 PW under dissociative conditions, for normal (a) and papain-treated (b) cartilage. Each point represents the mean value of three animals. V 0, void volume;, total volume.

5 Osteoarthritis and Cartilage l. 1 No l (a) [6 h vt i i 24 h I (b), 10 i i 10 Y O V t 100 I, ii ~ y~ 6h - 24 h. 48 h 100 I 100 l 0 10 Fraction no. 100 L Fro. 3. Elution profiles of 35S-PGs on a TSKgel G6000 PW under associative conditions, for normal (a) and papain-treated (b) cartilage. Each point represents the mean value of three animals. V0, void volume; V t, total volume. was calculated (Fig. 4). In the normal cartilage [Fig. 4(a)], 56-61% of the total was the aggregating fraction, and the percentage changed only slightly with time. In the papain-treated cartilage [Fig. 4(a)], 45-56% of the total was the aggregating fraction, and the percentage tended to increase with time. Twenty-six to 35% of the total was eluted in the nonaggregating fraction in the 70 ii,i (a) 70 (b) 60-6O o ~o ~,40 t~ t~ ~ ao Z 20 o'i II,, Time (h) 2( 0 T ;i I 'i ' FIG. 4. Proportions of the aggregating and non-aggregating fractions in the elution profiles under associative condition. (a) Aggregating fraction; (b) non-aggregating fraction. Each point represents the mean+s.d, of three animals. * P < 0.05, ***P < 0.001, significant difference between the normal (D) and papain-treated (I) cartilage.

6 258 Miyauchi et al.: PG synthesis treated with papain (a) (b) (c) JL! > "~ (d) o 900 vt (e) yt (f) V0 yt 90( ~ ~ ( t 450 0! 10 Fraction no. FIG. 5. Elution profiles of the aggregating fraction in 35S-PGs extracted from the normal cartilage and p~pain-treated cartilage on a TSKgel G5000 PWxL. (a)-(c), a~s-pgs from normal cartilage; (d)-(f), 35S-PGs from papain-treated cartilage. (a), (d), non-treated; (b), (e), after digestion with CHase ABC; (c), (f), after digestion with CHase ABC and KSase. V0, void volume;, total volume. normal cartilage and 37-46% in the papain-treated cartilage (Fig. 4). These percentages decreased with time in both the normal and papain-treated cartilage. The percentage of the aggregating fraction was larger, while that of the nonaggregating fraction was smaller in the normal, compared with the papain-treated cartilage, at any time point. In the chromatograms at 6 and 24 h after the injection of Na 2 35SOt, significant differences were observed in the percentages of both combined fractions. 35S-GAG COMPOSITION OF 35S-PGs To clarify the differences in synthesized 35S-PGs between the normal and papain-treated cartilages, the composition of the sulfated glycosaminoglycan (S-GAG) of the aggregating or nonaggregating fractions were investigated. 35S-PGs were extracted from cartilages obtained 24 h after the injection of Na2 3~SO4 and then separated by GPC under associative conditions into two combined fractions, namely, the aggregating fraction with HA and the nonaggregating fraction. These fractions were digested with CHase ABC and KSase after digestion with hyaluronidase to remove the HA. They were then separated by GPC using a TSKgel G5000 PWxa, and the 35S-GAG composition of 85S-PGs was estimated from the changes in the chromatograms. The results are shown in Figs 5 and 6. 35S-PGs in the aggregating fraction of the normal cartilage were mostly eluted in the V0 of a TSKgel G5000 PWxL [Fig. 5(a)]. After digestion with CHase ABC, about 70% of the total radioactivity was eluted in fraction 43 (V t: total volume) [Fig. 5(b)] and after further digestion with KSase, the radioactivity eluted in the accounted for not less than 80% of the total [Fig. 5(c)]. The results with the aggregating fraction of the papain-treated cartilage are shown in Figs 5(d)-(f). The changes in the chromatograms after digesting the fractions with each enzyme were very similar to those observed with the aggregating fraction of the normal cartilage. ~ss-pgs in the nonaggregating fraction of the normal cartilage were also mostly eluted in the V 0 [Fig. 6(a)]. After digestion with CHase ABC, not less than 80% of the total radioactivity was eluted in, and small peaks remained in V 0 and frac-

7 Osteoarthritis and Cartilage l. 1 No (a) V t (b) go (c) yo yt 400 4OO 400 2OO 2O 30 4O o 900 (d) Yo V t 900 (e) Yt 900 i~ go y~ 45O 45G 450 I 10 0 Fraction no. 10 Fro. 6. Elution profiles of the non-aggregating fraction in 35S-PGs extracted from normal cartilage and papain-treated cartilage on a TSKgel G5000 PWxL. (a)-(c), ass-pgs from normal cartilage; (d)-(f), 35S-PGs from papain-treated cartilage. (a), (d), non-treated; (b), (e), after digestion with CHase ABC; (c), (f), after digestion with CHase ABC and KSase. V0, void volume;, total volume. tions [Fig. 6(b)]. After digestion with CHase ABC and KSase, the peaks in the V 0 and fractions further decreased, and about 90% of the radioactivity was eluted in the V t [Fig. 6(c)]. The results with the nonaggregating fraction of the papain-treated cartilage are shown in Fig. 6(d)-(f). The changes in the chr0matograms after digesting the fractions with each enzyme were also similar to those observed with the nonaggregating fraction of the normal cartilage, except for the disappearance of the peak observed in the V 0 after CHase ABC digestion. In the chromatograms shown in Figs 5 and 6, fractions were regarded as a high molecular weight fraction, and the percentage of radioactivity relative to the total was calculated. The Table I Percentages of high-molecular weight ~ss-pgs (fractions 21-39) in the elution profiles on a TSKgel G5000 PWxL and estimation of 35S-GAG composition Percentage of high=molecular weight 35S-PGs (%) 35S.GAG composition (%) After digestion After digestion with CHase with CHase Fraction Nontreatment ABC ABC + KSase 35S-CS 35S-KS Normal cartilage Aggregating Nonaggregating Papain-treated cartilage Aggregating Nonaggregating S-GAG composition was estimated from percentage degraded by the corresponding GAGase.

8 260 Miyauchi et al.: PG synthesis treated with papain Table II Ratios of zss-adi-4s to zss-adi-6s obtained from the aggregating and nonaggregating fractions after digestion with CHase ABC Fraction ~SS-ADi-6SffSS-ADi-4S Normal cartilage Aggregating 2.15 Nonaggregating 1.45 Papain-treated cartilage Aggregating 1.91 Nonaggregating 1.38 results are shown in Table I. In the aggregating fractions, 62.6% (normal) or 58.2% (papaintreated) of 35S-GAG was digested with CHase ABC, and 11.9% (normal) or 14.8% (papain-treated) w.as digested with KSase. In the nonaggregating fractions, 80.3% (normal) or 79.6% (papain-treated) of 35S-GAG was digested with CHase ABC, and 8.8% (normal) or 8.5~/o (papain-treated) was digested with KSase. No large difference was observed between the normal and papain-treated cartilages in the percentages of radioactivity digested with CHase ABC and KSase. COMPOSITION OF 35S-CS ISOMER IN 35S-PGs Composition of the 35S-CS isomer in 35S-PGs was estimated from the amount of 35S-labeled unsaturated disaccharides (35S-ADi-6S and 35S-ADi-4S) obtained after digestion with CHase ABC. The disaccharides of 35S-CS were separated by HPLC, and the ratio of 35S-ADi-6S to 35S-ADi-4S (~S-ADi- 6SffSS-ADi-4S) was calculated (Table II). The value of 35S-ADi-6S/~SS-ADi-4S in the aggregating fraction was about 2.0, and that in the nonaggregating fraction was about 1.4 in both the normal and papain-treated cartilage cases. Discussion As shown in Fig. 1, ~ss-pg synthesis was enhanced in the cartilage 7 days after papain injection compared with normal cartilage. This enhancement has also been found in cartilage denatured by other means, and it was explained as a reaction that supplements the loss of PGs from the damaged cartilage [8]. It has also been reported that stainability with safranin O or toluidine blue decreases and that the proliferation of chondrocytes accelerates in the cartilage 7 days after the injection of papain [1-4]. Therefore, the enhancement in 35S-PG synthesis observed in this study is also seen as a reaction for supplementing the loss of cartilage PG. Chromatograms of 35S-PGs on a TSKgel G6000 PW revealed clear differences between the normal and papain-treated cartilages (Figs 2 & 3). These were more obvious under associative conditions, and the proportion of the nonaggregating fraction in the papain-treated cartilage" was higher than that in the normal cartilage (Figs 3 & 4). The results shown in Figs 1 and 3 indicate that the proportion of nonaggregating 35S-PGs increased with the acceleration in to~al 35S-PG synthesis in the papain-treated cartilage during the repair process. In the normal cartilage, 26~35 /O of the total 35S-PGs were eluted in the non~ggregating fraction (Figs 3 & 4). It has been demonstrated using various animal cartilages that the low molecular weight PGs, which are unable to ~ggregate with HA, are also present in normal cartilage, except for the degradation products of aggrecan [9-12]. The content of these PGs in cartilage, however, is much lower than that of aggrecan, and most of these do not contain KS as the GAG chain. Because KSase sensitive 35S-GAG is present in the nonaggregating fraction of normal cartilage as shown in Fig. 6, it is suggested that most of the 35S-PGs in the nonaggregating fraction are degradation products of 3~S-aggrecan or immature 35S-aggrecan. Degradation products of aggrecan which have lost the HA binding domain are naturally unable to bind HA. In addition, newly synthesized immature aggrecan in the cartilage of the rabbit knee joint is also unable to aggregate with HA [13, 14]. As shown in Fig. 4, the proportion of 35S-PGs in the nonaggregating fraction decreased with time after the Na 2 35SOt injection in the papain-treated cartilage, and that in the aggregating fraction increased. If most of the nonaggregating 35S-PGs were immature ~S-aggrecan, this phenomenon could be explained by the conversion of nonaggregating immature 35S-aggrecan to its mature type, which was able to aggregate with HA. However, under dissociative conditions, chromatograms of 35S-PGs derived from the papain-treated cartilage differed from those of normal cartilage; the amount of small 3~S-PGs eluted in fractions increased in the papain-treated cartilage

9 Osteoarthritis and Cartilage l. 1 No compared with the normal cartilage (Fig. 2). In addition, these small 35S-PGs might also be eluted in the same fractions as non-aggregating 35S-PGs under associative conditions (Fig. 3). Because the conversion of immature aggrecan to its mature type is not accompanied by an alteration in molecular size [13, 14], these results suggest that most of the nonaggregating 35S-PGs increasing in the papain-treated cartilage are not immature 35S-aggrecan but its products of degradation. The decrease with time in the nonaggregating fraction shown in Fig. 4 can also be explained by the rapid disappearance of nonaggregating ~ss-pgs, which have a higher diffusion constant, from the papaintreated cartilage. To investigate further the origin of nonaggregating 35S-PGs observed both in the normal and papain-treated Cartilage and increasing in the papain-treated cartilage, extracted 35S-PGs were divided into aggregating and nonaggregating fractions. The composition of the 35S-GAG in these fractions was estimated by changes in the gel permeation chromatograms (GPC) obtained before and after digestion with CHase ABC and KSase. Furthermore, unsaturated disaccharides of ~ss-cs (3~S-ADi-4S and 35S-ADi-6S) were also determined after digestion with CHase ABC, and their ratios (35S-ADi 6S: ~S-ADi:4S) were calculated. KSase derived from Pseudomonas, as used in this study, does not act if galactose residues of KS are sulfated [15]. Therefore, KS in cartilage containing a number of galactose sulfate residues cannot be decomposed into its disaccharides with the KSase, but if there are any enzyme-acting sites, however few there are, the GPC performed before and after enzyme digestion should be different. About 60% of the radioactivity in the aggregating fraction and about 80% of that in the nonaggregating fraction were derived from 35S-CS. Twelve to 15% of the radioactivity in the aggregating fraction and ~/o of that in the nonaggregating fraction were derived from 35S-KS, which was sensitive to KSase derived from Pseudomonas (Figs 5 & 6, Table I). When the aggregating and nonaggregating fractions were compared, there were some differences in the percentages of ~ss-cs and ~ss-ks, but little difference was found between the normal and papaintreated cartilage. Furthermore, the values of 3~S-A Di-6S/35S-ADi-4S in the nonaggregating fraction derived from both normal and papain-treated cartilage were lower than those in the aggregating fraction, but no large differences were observed between the normal and papain-treated cartilage (Table II). Namely, these results suggest that the nonaggregating 358-PGs increasing in the papaintreated cartilage are similar to those in the normal cartilage with regard to not only the elution position on a TSKgel G6000 PW but also the compositions of 35S-GAG and 35S-CS isomer. The relatively high CS and low KS levels in the nonaggregating 35S-PGs may also be explained by the cleavage of 35S-aggrecan at the core protein between the CS- and KS-rich regions, but the difference in composition of the 35S-CS isomer cannot be explained by the simple proteolytic degradation of ~SS-aggrecan. Therefore, the difference in the composition of the 3~S-CS isomer suggests that most of the nonaggregating 35S-PGs were not simple degradation products of the aggregating 85S-aggrecan. It has been reported that some newly synthesized aggrecan is degraded soon after secretion, and as a result, it loses its aggregating ability with HA [13, 16, 17]. If this aggrecan, which is degraded soon after secretion, has the GAG composition such as high CS and low KS contents and a low value of A Di-6S/A Di-4S, the results discussed here will be easily explained. Thus, with regard to the papain-treated cartilage during the repair process, it can be considered that the synthesis of 35S-PGs, which lose the aggregating ability with HA soon after secretion, is also accelerated as one of the supplementary reactions that compensates for the loss of PGs. To demonstrate the above hypothesis, a more detailed characterization of the newly synthesized nonaggregating 35S-PGs is required. Regardless, this study shows that the supplementary reaction for PG content in the cartilage during its repair process is not simple acceleration of PG biosynthesis but its enhancement accompanied by an alteration in the aggregating ability and GAG compositions. In addition, these results also indirectly suggest the importance of alterations in GAG composition in research into PG metabolism. References 1. Bentley G. Papain-induced degenerative arthritis of the hip in rabbits. J. Bone Joint Surg [Br] 1971;53: Farkas T, Bihari-Varga M, Bir5 T. Histological and thermo-analytical study on intraarticular papain induced degradation and repair of rabbit cartilage. I. Immature animals. Ann Rheum Dis 1974;33: Farkas T, Bihari-Varga M, Bir5 T. Histological and thermo-analytical study on intraarticular papain induced degradation and repair of rabbit cartilage. II. Mature animals. Ann Rheum Dis 1976;35:23-6.

10 262 Miyauchi et al.: PG synthesis treated with papain 4. Inoue S, Glimcher MJ. The reaction of cartilage and osteophyte formation after the intraarticular injection of papain. J Jpn Orthop Assoc 1982;56: Williams JM, Uebelhart D, Ongchi DR, Kuettner KE, Thonar EJ-M. Animal models of articular cartilage repair. In: Kuettner KE, Schleyerbach R, Peyron JG, Hascall VC, Eds. Articular cartilage and osteoarthritis. New York, NY: Raven Press 1992: Oike Y, Kimata K, Shinomiya T, Nakazawa K, Suzuki S. Structural analysis of chick-embryo cartilage proteoglycans by selective degradation with chondroitin lyases (chondroitinases) and endfl-d-galactosidase (keratanase). Biochem J 1980;191: Yoshida K, Miyauchi S, Kikuchi H, Tawada A, Tokuyasu K. Analysis of unsaturated disaccl~arides from glycosaminoglycuronan by highperformance liquid chromatography. Anal Biochem 1989;177: Carney SL, Billingham MEJ, Muir H, Sandy JD. Demonstration of increased proteoglycan turnover in cartilage explants from dogs with experimental osteoarthritis. J Orthop Res 1984;2: Heinegfid DH, Hascall VC. Characteristics of the non-aggregating proteoglycans isolated from bovine nasal cartilage. J Biol Chem 1979;254: Swann DA, Powell S, Sotman S. The heterogeneity of cartilage proteoglycans. Isolation of different types of proteoglycans from bovine articular cartilage. JBiol Chem 1979;245: Swann DA, Garg HG, Sotman SL, Hermann H. Isolation and partial characterization of low density proteoglycans from bovine articular cartilage. J Biol Chem 1983;258: Vilim V, Krajickova J. Proteoglycans of human articular cartilage--identification of several populations of large and small proteoglycans and of hyaluronic acid-binding proteins in successive cartilage extracts. Biochem J 1991;273: Plaas AHK, Sandy JD. The affinity of newly synthesized proteoglycan for hyaluronic acid can be enhanced by exposure to mild alkali. Biochem J 1986;234: Sandy JD, O'Neill JR, Ratslaff LL. Acquisition of hyaluronate-binding affinity: in vivo by newly synthesized cartilage proteoglycans. Biochem J 1989;258: Nakazawa K, Suzuki S. Purification of keratan sulfate-endo-galactosidase and its action of keratan sulfate of different origin. J Biol Chem 1975;250: Sandy JD, Plaas AHK. Age:related changes in the kinetics of release of proteoglycans from normal rabbits cartilage explants. J Orthop Res 1986;4: Sandy JD. Extracellular metabolism of aggrecan. In: Kuettner KE, Schleyerbach R, Peyron JG, Hascall VC, Eds. Articular cartilage and osteoarthritis. New York, NY: Raven Press 1992:~2-33.