Collagen biosynthesis of mechanically loaded articular cartilage explants

OsteoArthritis and Cartilage (2005) 13, 906e914 ª 2005 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.joca.2005.06.001

International Cartilage Repair Society

Collagen biosynthesis of mechanically loaded articular cartilage explants1 B. Ackermann Ph.D. and J. Steinmeyer Ph.D., Professor* Orthopaedic Research Laboratories, Department of Orthopaedic Surgery, Justus-Liebig-University Giessen, Paul-Meimberg-Strasse 3, D-35385 Giessen, Germany Summary Objective: The purpose of this study was to investigate systematically the effect of load amplitudes, frequencies and load durations of intermittently applied mechanical pressure on the biosynthesis of collagen and non-collagenous proteins (NCP) as well as on the water content of cultured bovine articular cartilage explants. Methods: Cyclic compressive pressure was applied using a sinusoidal waveform of 0.5 Hz frequency with a peak stress of 0.1, 0.5 or 1.0 MPa for a period of 10 s followed by a load-free period of 10, 100 or 1000 s. These intermittent loading protocols were repeated for a total duration of 1, 3 or 6 days. During the final 18 h of experiments, the incorporation of [3H]-proline into collagen and NCP, the content of water as well as the deformation of loaded explants were determined. Results: Intermittently applied, cyclic mechanical loading of articular cartilage explants consistently reduced the relative rate of collagen synthesis compared to load-free conditions. This reduced proportion of newly synthesized collagen among newly made proteins was independent of the mechanical stimuli applied. The release of newly synthesized collagen and NCP from loaded explants into the nutrient media was unaffected by any of the loading protocols applied. In addition, quantitative data are provided showing that only high amplitudes of loads and frequencies enhanced the water content of the explants. Conclusions: Previous studies reporting that osteoarthritic cartilage in vivo can synthesize elevated amounts of collagen imply that the loading protocols chosen were inadequate for simulating in vitro osteoarthritic-like alterations of collagen synthesis. In our experiments the collagen biosynthesis of chondrocytes was only minor responsive to alterations in mechanical stimuli, applied over a wide range. Thus, our results imply that the synthesis of these structural macromolecules is under the strict control of normal chondrocytes enabling them to maintain the shape of this physical demanded tissue. ª 2005 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved. Key words: Collagen, Proteins, Cartilage, Biomechanics, Osteoarthritis.

but are likely to be related to its high hydroxylysine content which facilitates glycosylation, a property of hydrophilic tissue. Collagen types IX and XI are also cartilage-specific molecules that are found in smaller quantities of 3e10%, each depending on the cartilage source and age1,2. Collagen type XI is located largely within the fibrils, is covalently linked to collagen type II, and is believed to regulate fiber size. In contrast, collagen type IX is located on the exterior of the fibrils and functions amongst other things as a mechanism to allow collagen fibrils to interact with proteoglycan macromolecules1,2. The matrix immediately surrounding the chondrocytes forms a specialized subcompartment of articular cartilage containing small diameter collagen fibrils and the minor collagen types III, V, and VI2. Collagen type X is found in the deep, calcified zone of mature articular cartilage2. Articular cartilage destruction and the loss of its components type II collagen and proteoglycans are a characteristic feature of osteoarthritis (OA). Using human specimens obtained at autopsy and arthroplasty, it was found that in both aging and OA the first damage to collagen type II occurs in the superficial and upper middle zones and extends into the deeper zones with increasing degeneration. The initial damage is always seen around chondrocytes in the pericellular region3. OA cartilage exhibits increased levels of cleaved and denatured type II collagen4,

Introduction The uniqueness of articular cartilage lies in its remarkable elasticity, low-friction surface, and ability to withstand enormous physical forces during motion1,2. These extraordinary features are directly related to the swelling pressure resulting from the hydration of the polyanionic proteoglycan aggregates, which is in turn resisted by a highly organized network of collagen fibers. Five collagens (types II, VI, IXeXI) have been identified in articular cartilage in quantities sufficient to allow their isolation. Collagen type II is the most abundant fibrillar protein found in articular cartilage and constitutes 85e90% of the total collagen content. Collagen type II forms the backbone of the cartilage heteropolymeric fibrils1,2. The features that make collagen type II uniquely suited for cartilage are unknown 1 Source of support: This work was supported by the Federal Ministry of Education and Research (BMBF no. 0311058) and by the foundation S.E.T. *Address correspondence and reprint requests to: Professor Dr Ju¨ rgen Steinmeyer, Orthopaedic Research Laboratories, Department of Orthopaedic Surgery, Justus-Liebig-University Giessen, Paul-Meimberg-Strasse 3, D-35385 Giessen, Germany. Tel: 49-641-99-42920; Fax: 49-641-99-42939; E-mail: juergen. [email protected] Received 17 April 2004.

906

907

Osteoarthritis and Cartilage Vol. 13, No. 10 shown to result from excess matrix metalloproteinase (MMP) activity5. MMP-13 has been identified as the enzyme responsible for most of the collagen degradation5. In addition, MMP-3, -7, -9, and -14 are present in OA cartilage and can cleave in the nonhelical telopeptide of collagen type II5,6. Disruption of the collagen network allows the tissue to swell resulting in an up to 9% increased water content and in loss of proteoglycans7,8. The turnover of collagen occurs as a delicate balance between cell-mediated degradation and synthesis, both of which are probably regulated by a cascade of interdependent mechanisms. Several laboratories have reported that collagen synthesis was increased in OA cartilage9e11. Furthermore, data from animal models of OA and humans revealed that newly synthesized, radiolabeled collagen was only of type II10,12. Burton-Wurster et al.13 reported that both in vitro and in vivo degenerated cartilages of dogs with hip dysplasia deposited less newly synthesized collagen resulting in collagen depletion in the foci of degenerated cartilage. Recent studies14,15 disclosed that normal and early human OA cartilage have a low level of mRNA expression of collagen type II, thus supporting the already reported low turnover of collagen with a half-life in excess of 100 days in immature cartilage, and apparent half-lives of many years in adult tissue16,17. In contrast, late stage human OA cartilage strongly upregulated the expression of collagen types II and VI probably in an attempt to remodel and thus repair the cartilage defect16,17. Strong clinical correlations exist between abnormal patterns of mechanical loading and the later onset of cartilage degradative disease such as OA. It seems even likely that mechanical loading and joint motion play important roles in establishing the orientation and organization of the collagen fibers of the matrix18,19. For instance, after long-distance running training of dogs, the superficial collagen network of articular cartilage was either deteriorated or the fibrils were reoriented due to an altered loading pattern19. Furthermore, published data from a range of laboratories20e29 have disclosed that mechanical loading can either stimulate or inhibit the biosynthetic activity of chondrocytes. Using articular cartilage explants, these investigations have focused on water content, ion-induced swelling, and on proteoglycan as well as fibronectin metabolism, while no data have been provided about the possible impact on collagen synthesis due to load. Under in vitro conditions, cartilage plugs can be maintained in stable and controlled biochemical and physical environments allowing a precise characterization of the impact of mechanical forces on cartilage metabolism. This is the benefit gained with cartilage explant studies compared with studies carried out in vivo and, furthermore, a prerequisite for developing a mechanical induced in vitro model of OA-like cartilage which has been one of our longer-term objectives. We recently reported that intermittent loading of cartilage explants can induce an elevated synthesis and retention of fibronectin26,28, an increased synthesis and release of proteoglycans29 and even some of the same alterations in the sulfation pattern of chondroitin sulfate chains27 that have previously been shown to occur also during the development of degenerative joint diseases such as OA. Based on our previous findings, our main hypothesis was that mechanical load plays an intrinsic role in the homeostatic control of collagen synthesis that is important in normal physiology as well as in cartilage degeneration. This hypothesis was tested in the present study by systematically investigating the effect of load amplitudes, frequencies and load durations of intermittently

applied mechanical pressure on the biosynthesis of collagen and water content of cultured mature articular cartilage explants.

Materials and methods LOAD-FREE, LONG-TERM EXPLANT CULTURES

In order to determine the optimal time point to start loading the explants, the maintenance of collagen and noncollagenous protein (NCP) synthesis in bovine articular cartilage explants was investigated over a period of 14 days. Under sterile conditions, eight macroscopically healthy cartilage explants (without subchondral bone) were removed in full thickness from the weight-bearing area of the metacarpophalangeal condyles of one joint of an 18e24-month-old steer within 2 h after slaughter. Explants were washed and then cultured in 2.5 ml Ham’s F-12 nutrient media supplemented with 2.5 mM HEPES, pH 7.2, 30 mg/ml alpha-ketoglutarate, 300 mg/ml L-glutamine, 50 mg/ ml ascorbate, 20 U/ml penicillin, 10 mg/ml streptomycin, 485 mg/ml CaCl2 ! 2H2O and the serum substitute CRITSCTM Premix (Collaborative Research Inc., Bedford, USA) as previously described30. This medium was further supplemented with 1.0 mM Na2SO430e32, 2.5 mg/ml amphotericin B and 50 mg/ml gentamycin. Explants were cultured at 37(C and under 5% CO2 and 95% humidity. Cultures were subjected to medium changes every second day or on the final day of experimentation (days 1, 3, 6, 9, 14), whereby explants were incubated for 2 h in fresh media before the addition of the radioactive precursor as described below. Care was taken to keep the pre-incubation time in fresh media (2 h) and subsequent labeling period (18 h) identical in all cultures. The experiment was repeated five times using condyles from six different steers (N Z 6). PREPARATION, CULTURE AND MECHANICAL LOADING OF EXPLANTS

The mechanical loading apparatus used was designed and constructed to load living cartilage explants over extended time periods under sterile conditions, and has already been described elsewhere in detail33. Under sterile conditions, two macroscopically healthy, full-thickness cartilage explants (without subchondral bone) were removed from the weight-bearing area of one metacarpophalangeal condyle of an 18e24-month-old steer shortly after slaughter. Articular cartilage discs of 7-mm diameter were obtained using a biopsy punch. One explant was loaded and another explant from the same condyle served as a control. In order to determine the degree of compression during loading, the thickness of explants was then determined twice in the center of the cartilage using a digital caliper with a resolution of 0.01 mm and an accuracy of G0.02 mm. Cartilage explants were washed, placed into the specimen holders with the articular surface upward, and cultured in 2.5 ml of supplemented Ham’s F-12 nutrient media as described above. The loading device was then positioned into a CO2incubator for 2 h to allow equilibration with the incubator’s environment (37(C, 5% CO2, 95% humidity). Intermittently applied, uniaxial cyclic loading was introduced by using an approximately sinusoidal waveform of 0.5 Hz frequency and a peak stress of 0.1, 0.5 or 1.0 MPa for a period of 1, 3 or 6 days (Fig. 1). The cyclic loads were applied for 10 s followed by a period of unloading lasting 10, 100 or 1000 s. Explants were loaded perpendicular to their long axis in

908

B. Ackermann and J. Steinmeyer: Cartilage collagen and intermittent loading Simultaneously, 25 ml of 1% b-aminopropionitrile was added to inhibit crosslinking of the collagen and thus facilitate the extraction of newly synthesized proteins. At the end of the radiolabeling period, explants and media were harvested and stored frozen at
20(C in the presence of a 10% (v/v) protease inhibitor mixture until analysis. DETERMINATION OF WATER CONTENT

Fig. 1. Schematic illustration of the loading protocols described in Materials and methods. Intermittent cyclic mechanical loading was applied using an approximately sinusoidal waveform of 0.5 Hz frequency followed by a period of unloading. The effects of intermittent cyclic loading on the incorporation of [3H]-proline into collagen and NCP were assessed by radiolabeling the cartilage explants during the final 18 h of the experiments before tissue was harvested and processed for analyses. A: Explants were cyclically loaded for 10 s with a peak stress of 0.5 MPa followed by a period of unloading lasting 10, 100 or 1000 s. This intermittent loading was repeated X-times for the total duration of experiments lasting 3 days. B: Explants were cyclically loaded for 10 s with a peak stress of 0.1, 0.5 or 1.0 MPa followed by a period of unloading lasting 1000 s. This intermittent loading was repeated X-times for the total duration of experiments lasting 3 days. C: Explants were cyclically loaded for 10 s with a peak stress of 0.5 MPa followed by a period of unloading lasting 1000 s. This intermittent loading was repeated X-times for the total duration of experiments lasting 1, 3 or 6 days.

radially-unconfined compression. During the period of unloading, the load platen was lifted from the cartilage surface. Unloaded cartilage discs of the same condyle were cultured in identically constructed loading chambers, and served as controls. The degree of compression of cartilage explants during loading was monitored using a displacement transducer system as described elsewhere in detail33. Depending on the loading protocol, an increase in cartilage compression was recorded which during the first 8 h reached a maximum maintained for the remainder of the experiment. Media were changed on day 3 and collected media were stored frozen at
20(C in the presence of a 10% (v/v) protease inhibitor mixture containing 0.1 mM phenylmethylsulfonyl fluoride (PMSF), 200 mM ethylene diaminetetraacetic acid (EDTA), 5 mM benzamidine/HCl and 10 mM N-ethylmaleimide until analysis. Each protocol was repeated five times using condyles from six different steers (N Z 6).

RADIOLABELING OF EXPLANTS

Radiolabeling for determining the synthetic activity of the chondrocytes was performed during the final 18 h of each experiment using [3H]-proline (DuPont de Nemours GmbH, Bad Homburg, Germany) at a final concentration of 10 mCi/ml.

As evidence of collagen matrix damage, the water content of the explants was determined gravimetrically. Explants were thawed and allowed to swell to an equilibrium wet weight for 45 min at 4(C. Excess liquid was removed by blotting the specimen on filter paper, and tissue wet weights were determined. Explants were washed four times with ice-cold Hank’s balanced salt solution (HBSS) containing a protease inhibitor mixture, lyophilized overnight (model Lyovac GT2, Finn-Aqua SantasaloSohlberg GmbH, Hu¨rth, Germany), weighed again, and then stored frozen at
20(C until analysis. The difference between wet and dry weight represents the water content, and was used to calculate the percentage of water within wet weighed explants. SYNTHESIS OF COLLAGEN

Radioactive collagen (C) of cartilage explants as well as those secreted into the media were assayed as collagenase-sensitive material using a modified method according to Peterkofsky and Diegelmann34,35. Briefly, lyophilized cartilage explants were minced with a scalpel in the presence of 1.0 ml ice-cold buffer A containing 50 mM Tris/HCl (pH 7.2), 10 mM L-proline and 10% (v/v) protease inhibitor mixture. Media (2.0 ml) were first diluted with 40 ml of 2.5% bovine serum albumin and 4.0 ml ice-cold buffer B containing 75 mM Tris/HCl (pH 7.2), 15 mM L-proline and 302 mM NaCl; proteins were then precipitated by the addition of 1.2 ml of 60% (v/v) trichloroacetic acid (TCA)/ 10 mM L-proline. The mixture was kept on ice for 15 min, and proteins were pelleted by centrifugation at 23,000 ! g for 10 min at 4(C. The supernatant was discarded. Unincorporated [3H]-proline was removed completely from media pellets as well as minced cartilage explants by repeated washing and centrifugation. Explants and pellets were solubilized with 0.2 N NaOH followed by neutralization at 4(C with 0.15 N HCl and 1 M N-(2-Hydroxyethyl) piperazine-N#-(2-ethanesulfonic acid) hemisodium salt (HEPES) (pH 7.2). Then, 0.5 ml of solubilized cartilage explants or media pellets were mixed with 85 ml of a digestion solution containing 49 units of the highly purified bacterial collagenase type VII (SigmaeAldrich Chemie GmbH, Deisenhofen, Germany), 50 mM Tris/HCl (pH 7.2), 5 mM CaCl2, 30 mM benzamidine HCl, 60 mM N-ethylmaleimide, 0.6 M 6-aminohexanoic acid, and were incubated at 37(C for 90 min. The respective controls received no collagenase. Collagen digestion was terminated by the addition of 0.6 ml of 10% (v/v) TCA/0.5% (w/v) tannin and 20 ml of 2.5% (w/v) bovine serum albumin. The mixture was kept on ice for 1 h, centrifuged at 23,000 ! g for 10 min at 4(C, and the supernatant containing the digested [3H]labeled collagen was assayed for radioactivity by liquid scintillation counting. The pellet was washed again with 0.6 ml of a solution containing 5% (v/v) TCA/0.25% (w/v) tannin at 4(C for 5 min, before a centrifugation step at 23,000 ! g for 10 min at 4(C. The radioactivity within the supernatant, again containing digested [3H]-labeled collagen, was then measured. The pellets with their NCP were

Osteoarthritis and Cartilage Vol. 13, No. 10 dissolved in 1.2 ml of 0.2 N NaOH overnight at 56(C, neutralized with 50 ml of 2.4 N HCl and 0.25 ml of 1 M Tris/ HCl (pH 8.0), and the radioactivity was subsequently measured. The proportion of collagen amongst the newly synthesized proteins was calculated using the following formula that takes into account the enrichment of proline in collagen as compared to other proteins35: %CZdpm C!100=dpm CCð5:4!dpm NCPÞ

STATISTICAL ANALYSIS

Each experiment was repeated five times (N Z 6). The data obtained from loaded explants were normalized using values from unloaded controls before they were analyzed using a one-way analysis of variance (ANOVA) and posthoc testing with Scheffe´’s multiple mean comparison test to determine statistical significance. In addition, data from loaded explants were compared with those from unloaded controls using Student’s two-tailed paired t test. Differences were considered significant at P values of less than 0.05.

Results LONG-TERM CULTURE OF UNLOADED CARTILAGE EXPLANTS

The synthesis of collagen and NCP by bovine articular cartilage explants remained constant during the 14-day culture period (Fig. 2). The proportion of collagen amongst newly synthesized proteins did not change over the entire culture period (day 1: 13.0 G 3.1%; day 3: 12.8 G 2.1%; day 6: 10.7 G 3.7%; day 9: 12.1 G 2.8%; day 14: 12.3 G 2.8%). Loss of newly synthesized collagen from explants into the nutrient media was also unchanged over the entire culture period (day 1: 18.24 G 3.92; day 3: 20.7 G 6.3%; day 6: 17.4 G 5.0%; day 9: 15.4 G 4.1%; day 14: 17.9 G 3.3%).

909 rate of collagen synthesis compared to load-free conditions (Figs. 3e5). This reduced proportion of newly synthesized collagen among proteins appeared to be independent of the mechanical stimulus applied (ANOVA: P O 0.05). At the same time we determined an increased incorporation of [3H]-proline into overall proteins, whereas the incorporation of the radiolabeled precursor into collagen remained constant with a wide range of mechanical stimuli. The release of newly synthesized collagen and NCP from loaded explants into the nutrient media was unaffected by any of the loading protocols applied (data not shown). Increasing the applied magnitude of compressive load from 0.1 MPa to 1.0 MPa significantly modulated the incorporation of [3H]-proline into collagen (Fig. 3; ANOVA: P Z 0.01) and NCP (Fig. 3; ANOVA: P Z 0.038). In addition, we determined a significantly reduced relative rate of collagen synthesis by explants which was not related to the magnitude of the loads applied (Fig. 3; ANOVA: P O 0.05). Increasing the magnitude of loading did not modify the release of newly synthesized collagen from explants into nutrient media (ANOVA: P O 0.05; ratio of incubation medium collagen from loaded cultures (%)/from unloaded cultures (%): 0.1 MPa: 1.35 G 0.41; 0.5 MPa: 1.20 G 0.35; 1.0 MPa: 1.83 G 0.57). In addition, loading of explants with a maximum pressure of between 0.1 MPa and 1.0 MPa did not result in an increased water content (Table I); unloaded cartilage explants contained 71.1 G 2.2% (N Z 18) water. The duration of the load-free period significantly altered the incorporation of radiolabeled precursor into collagen (Fig. 4; ANOVA: P Z 0.012). Explants that were unloaded for 10e1000 s between each cyclic loading phase incorporated approximately half as much [3H]-proline into collagen whereas approximately 1.5e2-fold more [3H]proline was incorporated into NCP. Furthermore, a reduced proportion of newly synthesized collagen amongst all newly synthesized proteins was found in all loaded explants

COLLAGEN AND PROTEIN SYNTHESIS UNDER LOAD

Intermittently applied, cyclic mechanical loading of articular cartilage explants consistently reduced the relative

Fig. 2. Biosynthesis of collagen and NCP by bovine articular cartilage explants cultured for 14 days in serum-free and supplemented Ham’s F-12 media. Data represent the mean values G S.D. (N Z 6) of the incorporation of radioactive precursor into collagen and NCP per mg dry weight cartilage explant cultured for various times (e,e, collagen synthesis; C, synthesis of NCP).

Fig. 3. Effect of different stress levels on the [3H]-proline incorporation into collagen and NCP per mg dry weight cartilage, as well as the proportion of collagen among newly synthesized proteins (relative rate of collagen synthesis), expressed as percentage of unloaded controls. Using a 0.5 Hz frequency and a sinusoidal waveform, a maximum pressure ranging from 0.1 MPa to 1.0 MPa was cyclically applied on cartilage discs for 10 s followed by a period of unloading lasting 1000 s. Explants were intermittently loaded for 3 days. Data are mean values G S.D. (N Z 6). Statistically significant differences from unloaded control values as determined by Student’s two-tailed paired t test: *0.01 ! P % 0.05; **0.001 ! P % 0.01 (e,e, collagen synthesis; C, synthesis of NCP; –-–, relative rate of collagen synthesis).

910

B. Ackermann and J. Steinmeyer: Cartilage collagen and intermittent loading

Fig. 4. Effect of time of intermittence on the [3H]-proline incorporation into collagen and NCP per mg dry weight cartilage, as well as the proportion of collagen among newly synthesized proteins (relative rate of collagen synthesis), expressed as percentage of unloaded controls. Using a 0.5 Hz frequency and a sinusoidal waveform, a pressure of 0.5 MPa was cyclically applied on cartilage discs for 10 s followed by a period of unloading ranging from 10 s to 1000 s. Explants were intermittently loaded for 3 days. Data are mean values G S.D. (N Z 6). Statistically significant differences from unloaded control values as determined by Student’s two-tailed paired t test: *0.01 ! P % 0.05; **0.001 ! P % 0.01 (e,e, collagen synthesis; C, synthesis of NCP; –-–, relative rate of collagen synthesis).

independent of the duration of the load-free period; ANOVA revealed that increasing the duration of the load-free period between each loading cycle did not modify the release of newly synthesized collagen from explants into nutrient media compared to unloaded controls (P O 0.05; ratio of incubation medium collagen from loaded cultures (%)/from unloaded cultures (%): 10 s load-free period: 1.58 G 0.67;

100 s load-free period: 1.07 G 0.44; 1000 s load-free period: 1.20 G 0.35). Furthermore, reducing the length of the load-free period from 1000 s to 10 s resulted in a slightly but significantly increased water content compared to unloaded controls (Table I). Figure 5 demonstrates that increasing the total time of experiments from 1 to 6 days abolished the inhibitory effect of loading on the incorporation of [3H]-proline into collagen as seen on day 1 (ANOVA: P Z 0.028), and simultaneously enhanced the incorporation of the precursor into NCP (ANOVA: P Z 0.008). Again, a significantly reduced relative rate of collagen synthesis was determined independent of the duration of loading. Also, ANOVA revealed that the loss of newly synthesized collagen from explants into nutrient media remained unchanged independent of experimental duration (P O 0.05; ratio of incubation medium collagen from loaded cultures (%)/from unloaded cultures (%): 1 day: 0.78 G 0.26; 3 days: 1.20 G 0.35; 6 days: 1.28 G 0.32). Furthermore, increasing the length of loading from 1 to 6 days resulted in a slightly and significantly increased water content compared to unloaded controls (Table I). COMPRESSION OF INTERMITTENTLY LOADED CARTILAGE EXPLANTS

The degree of compression of loaded specimens, as determined after 8 h of loading, increased nonlinearly during the increase in maximum pressure from 0.1 MPa to 1.0 MPa (Table I; ANOVA: P Z 0.0001). Reducing the length of the load-free period from 1000 s to 10 s also elevated the maximum strain of cartilage explants (Table I; ANOVA: P ! 0.0001). Continuous determination of the maximum compression of explants revealed that after 8 h of loading no further increase in maximum compression occurred (Table I; ANOVA: P O 0.05). The standard deviations shown in Table I are in the range of G5%, indicating that loaded specimens, being initially 0.54 G 0.07 mm (N Z 42) thick, are compressed to a similar degree.

Discussion

Fig. 5. Effect of duration of loading on the [3H]-proline incorporation into collagen and NCP per mg dry weight cartilage, as well as the proportion of collagen among newly synthesized proteins (relative rate of collagen synthesis), expressed as percentage of unloaded controls. Using a 0.5 Hz frequency and a sinusoidal waveform, a pressure of 0.5 MPa was cyclically applied on cartilage discs for 10 s followed by a period of unloading lasting 1000 s. Explants were intermittently loaded for 1e6 days. Data are mean values G S.D. (N Z 6). Statistically significant differences from unloaded control values as determined by Student’s two-tailed paired t test: *0.01 ! P % 0.05; **0.001 ! P % 0.01 (e,e, collagen synthesis; C, synthesis of NCP; –-–, relative rate of collagen synthesis).

Even though it is well recognized that mechanical factors have a dominant impact on chondrocyte metabolic behavior both in health and diseases such as OA, this has not been addressed appropriately in many in vitro systems used for physiological, pharmacological or tissue engineering studies. In articular cartilage, several collagen types have already been identified that are synthesized and degraded in a precisely controlled manner to provide an architecture of this tissue able to withstand the enormous physical forces applied in vivo. Knowledge about the factors and mechanisms involved in controlling the turnover of collagen synthesis is a prerequisite to interpret data obtained from these in vitro studies. Following this line, the present study was the first ever study designed to investigate systematically the effects of the mechanical factors: (1) load amplitude, (2) time of intermittence and (3) load duration of mechanical applied pressure, on the biosynthesis of collagen, NCP and the water content of mature articular cartilage explants. We were able to demonstrate for the first time that the relative rate of collagen synthesis, expressed as the proportion of newly synthesized collagen among newly made proteins, is significantly reduced but not responsive to alterations of a wide range of different mechanical stimuli.

911

Osteoarthritis and Cartilage Vol. 13, No. 10 Table I Effect of intermittently applied cyclic mechanical loading on the water content and the degree of compression of cartilage explants Effect of

Loading conditions

Water content (% of control)

Compression (%)

Load amplitude (10 s loading; 1000 s unloading, duration: 3 days)

1.0 MPa 0.5 MPa 0.1 MPa

101.1 G 2.8 101.4 G 3.2 101.6 G 4.0

21.0 G 2.3** 11.4 G 3.1** 07.7 G 2.0**

Unloading period (0.5 MPa, 10 s loading; duration: 3 days)

1000 s 100 s 10 s

101.4 G 3.2 100.5 G 1.6 102.6 C 1.5*

11.4 G 3.1** 24.0 G 4.8** 23.0 G 1.7**

Duration of loading (0.5 MPa, 10 s loading, 1000 s unloading)

1 day 3 days 6 days

100.4 G 1.5 101.4 G 3.2 103.1 G 2.9*

16.1 G 3.2** 11.4 G 3.1** 18.0 G 2.6**

Cartilage explants were intermittently loaded as indicated. The water content of loaded explants is normalized by data obtained from the corresponding unloaded control explants. The degree of compression of loaded explants was determined after 8 h of loading after which the degree of maximum compression remained constant. Data are mean values G S.D. (N Z 6). Statistically significant difference from unloaded control specimens as determined by the Student’s two-tailed paired t test, *0.01 ! P % 0.05; **P % 0.001.

Loading of cartilage explants was started on the day of slaughter, since we found that the incorporation of radiolabeled precursor into collagen and NCP remained constant over a total duration of 14 days. The loads were applied intermittently because chondrocytes within articular cartilage in vivo experience complete unloading between loading cycles. Compression of cartilage in vivo occurs as the result of forces acting on this tissue; however, the pressures acting on the cartilage surface in species of widely differing size are relatively constant36. The forces themselves are generated by the weight and movement of the individual. In order to simulate the in vivo situation within a joint more closely, loads were applied intermittently using a load-controlled device33 to compress in a radiallyunconfined manner cartilage explants perpendicular to their long axis. The overall stresses acting in the bovine knee joint in vivo are unknown, but it has been calculated that the static compressive stress in the cow knee joint is approximately 0.8 MPa and in the human knee joint during walking between 0.8 MPa and 6.3 MPa36e38. The pressures used in our experiments were in the range of 0.1e1.0 MPa and include those measured in vivo; however, these strains might be quite similar or even higher than those likely to occur in vivo since we used cartilage explants without the subchondral bone and surrounding tissue. Newly synthesized collagen can differ from collagen contained within the tissue in that under a variety of in vitro circumstances such newly synthesized collagen can be significantly underhydroxylated. For instance, during the labeling period of 18 h used in our experiments, the ascorbate concentration within the media may have dropped due to its known low stability. As a result, the amount of labeled collagen synthesized cannot accurately be deduced from the amount of labeled hydroxyproline contained within a tissue. In order to circumvent this problem, we used a standard method originally developed by Peterkofsky and Diegelmann34,35 using a highly purified bacterial collagenase in order to determine the incorporation of [3H]-proline into collagen. This assay does not distinguish between labeled proline or labeled hydroxyproline, but it does separate collagen from all other known proteins because collagen is uniquely susceptible to bacterial collagenase. Differences in the rate of proline transport that alter the specific radioactivity of the intracellular pool may occur when testing the effect of loading on the incorporation of radiolabeled precursor into collagen; consequently, the

amount of radioactivity in collagen will be altered without any actual change in the rate of synthesis. Expressing collagen synthesis as a relative rate, i.e., the rate of synthesis of collagen compared to synthesis of all other cellular proteins, eliminates this problem. In our experiments, we measured for the first time the relative rate of collagen synthesis by unloaded bovine articular cartilage which during a 14-day culture period was found to lie in the range of 10.7% and 13.0%; these values are in good agreement with the relative rates of 18.8% and 14.5% found for knee articular cartilage of guinea pigs and rats, respectively39,40. Loading of explants resulted in a reduced relative rate of collagen synthesis independent of the loading protocols applied compared to controls, whereas an increased incorporation of [3H]-proline into NCP was found. In contrast, we previously reported28 that the incorporation of [3H]-phenylalanine into proteins remained constant using the same culture and loading protocols as used in this study. We, therefore, conclude that intermittent mechanical loading stimulated the proline transport into the intracellular pool. Consequently, the reduced relative rate of collagen synthesis reflects the load-induced inhibition of collagen biosynthesis as seen with all loading protocols applied. However, it is also important to determine whether a decreased collagen production resulted mainly from a decreased synthesis or an enhanced degradation. The extent of degradation of extracellular collagen was estimated from the proportion of labeled collagenase products present in the media. We determined no significant increase in the release of these fragments, which represented approximately 18% of total labeled collagen, indicating again that the reduced relative rate of collagen production was due to a load-induced inhibition of collagen synthesis. Furthermore, the constant proportion of radiolabeled collagen found within media suggests that mechanical factors applied over a wide range did not induce the expression of extracellular collagenase activity such as MMP-13 in our experiments. Rather, this quite large fraction appears to reflect a rapid intracellular degradation of a significant portion of newly synthesized procollagen prior to secretion as has been described for fibroblasts41. Furthermore, in our study b-aminopropionitrile was added during the radiolabeling period in order to prevent crosslinking of newly synthesized collagen by inhibiting the activity of lysyloxidase, thus facilitating the extraction necessary for the analysis. This additive might have altered the deposition of newly synthesized collagen into the

912

B. Ackermann and J. Steinmeyer: Cartilage collagen and intermittent loading

pericellular region42 and may also explain the relatively high portion of collagen found in media. A strict control of collagen biosynthesis with respect to type and quantity produced is required to achieve and maintain the physiological and mechanical properties of articular cartilage. Several authors9e11 reported that collagen synthesis was increased in OA cartilage. In the early phases of post-traumatic OA an increased expression of type II collagen mRNA has been found in the cartilage of canine knee joints after anterior cruciate ligament transection11. However, recently a low level of mRNA expression of collagen type II was found in normal and early human OA cartilage whereas a strongly upregulated expression of collagen types II and VI was reported only for late stages of this disease14,15. Considering this, the reduced relative rate of collagen synthesis implies that the loading protocols chosen were not sufficient to allow the simulation of in vitro OA-like alterations in collagen synthesis as seen in humans or animals. Furthermore, the method applied to measure the incorporation of radiolabeled precursor into collagen did not allow us to distinguish between different types of collagen being synthesized. An increased mRNA expression of collagen type II43,44 as well as incorporation of [3H]-proline into collagen45 were found for cultured bovine articular chondrocytes exposed to intermittently applied hydrostatic pressure43 or dynamic stretching44,45. In our study, a wide range of loading protocols were applied that resulted in intermittent compression of cartilage explants, and subsequently generation of hydrostatic pressure and even deformation of chondrocytes; however, the relative rate of collagen synthesis was reduced and not susceptible to changes in the mechanical stimuli applied. The discrepancy between our data and those obtained with cultured chondrocytes may be explained by the existence of several signal transduction pathways enabling the chondrocytes to sense each mechanical signal separately. The specific mechanisms by which chondrocytes detect mechanical signals and convert them to an intracellular biochemical response are not fully understood. However, several signal transduction pathways have already been found to be involved including integrin-mediated pathways, stretch activated or inactivated ion channels, involvement of the cytoskeleton, decreased pH, electrical streaming potentials, nitric oxide or activating second messenger systems46e49. Thus, the metabolic response of chondrocytes seems to be the result of the balance between the different signals perceived by the chondrocytes enabling them to respond appropriately depending on the mechanical demand on the tissue. Our finding that intermittent loading can selectively modulate the biosynthesis of a specific cartilage component is consistent with our earlier reported observations. We previously found that continuously or intermittently applied mechanical loading can selectively influence fibronectin synthesis without altering overall protein synthesis26,28. Furthermore, we showed that in vitro dynamic compression of cartilage tissue can significantly alter the sulfation pattern of chondroitin sulfate chain termini of proteoglycans possibly by a mechanism involving a terminal galNAc4,6S-disulfotransferase activity27. In addition, chondrocytes were stimulated to produce elongated chondroitin sulfate chains27. Our findings are in line with those of other laboratories. For instance, Kim et al.50 described that the synthesis of the link protein is affected by static loading to a much lesser degree than that of aggrecan and the small proteoglycans. Remarkably, Baskin et al.51 found that low biaxial strain of 1.8% stimulated urethral fibroblasts in vitro

to produce elevated amounts of fibronectin whereas at a high strain of 4.9% both fibronectin and collagen synthesis were significantly increased. In our experiments, an elevated water content was observed only in those explants loaded either for 6 days or with a high rate of repetitively applied loading periods. Our findings are in good agreement to our earlier reported29 ioninduced swelling of explants and could have been caused by matrix damage due to the dying superficial zone of specimens already reported earlier28. The increased water content may also be interpreted as evidence of functional matrix damage which may be due to either rupture of critical collagen fibrils, loss of minor collagens, or loss of proteoglycans from explants into nutrient media20,23,25. Early OA changes are accompanied by an increase in water content of the cartilage even though there is no significant change in proteoglycan content7. The increase in cartilage water content is not of much biochemical significance but has biomechanical importance8,24. In conclusion, the results described here show for the first time that the relative rate of collagen synthesis of articular cartilage is reduced in a manner independent of the loading configuration chosen. In addition, we provide quantitative data showing that only high amplitudes of loads and frequencies enhanced the water content of explants. In our experiments the collagen biosynthesis of chondrocytes was only minor responsive to alterations in mechanical stimuli, applied over a wide range. Thus, our results imply that the synthesis of these structural macromolecules is under the strict control of normal chondrocytes enabling them to maintain the shape of this physical demanded tissue.

Acknowledgements The authors would like to thank I. Pelzer for excellent technical assistance, Dr J.P. Keogh for assisting in the editing of the manuscript, and J. Stone for typing the manuscript. This work was supported by the Federal Ministry of Education and Research (BMBF no. 0311058) and by the foundation S.E.T.

References 1. Eyre DR, Wu JJ, Fernandes RJ, Pietka TA, Weis MA. Recent developments in cartilage research: matrix biology of the collagen II/IX/XI heterofibril network. Biochem Soc Trans 2002;30:893e9. 2. Poole AR. Cartilage in health and disease. In: Kelley WN, Harris ED, Ruddy S, Sledge CB, Eds. Textbook of Rheumatology. Philadelphia: Saunders 1997; 255e308. 3. Hollander AP, Pidoux I, Reiner A, Rorabeck C, Bourne R, Poole AR. Damage of type II collagen in aging and osteoarthritis starts at the articular surface, originates around chondrocytes, and extends into the cartilage with progressive degeneration. J Clin Invest 1995;96: 2859e69. 4. Bank RA, Krikken M, Beekman B, Stoop R, Maroudas A, Lafeber FPJG, et al. A simplified measurement of degraded collagen in tissues: application in healthy, fibrillated and osteoarthritic cartilage. Matrix Biol 1997; 16:233e43. 5. Billinghurst RC, Dahlberg L, Ionescu M, Reiner A, Bourne R, Rorabeck C, et al. Enhanced cleavage of

913

Osteoarthritis and Cartilage Vol. 13, No. 10

6.

7. 8.

9. 10.

11.

12. 13.

14.

15.

16. 17.

18.

19.

20.

21.

type II collagen by collagenases in osteoarthritic articular cartilage. J Clin Invest 1997;99:1534e45. Poole AR, Nelson F, Dahlberg L, Tchetina E, Kobayashi M, Yasuda T, et al. Proteolysis of the collagen fibril in osteoarthritis. Biochem Soc Symp 2003;70: 115e23. Mankin HJ, Thrasher AZ. Water content and binding in normal and osteoarthritic human cartilage. J Bone Joint Surg 1975;57-A:76e80. Bonassar LJ, Frank EH, Murray JC, Paguio CG, Moore VL, Lark MW, et al. Changes in cartilage composition and physical properties due to stromelysin degradation. Arthritis Rheum 1995;38:173e83. Lippiello L, Hall D, Mankin HJ. Collagen synthesis in normal and osteoarthritic human cartilage. J Clin Invest 1977;59:593e600. Eyre DR, McDevitt CA, Billingham ME, Muir H. Biosynthesis of collagen and other matrix proteins by articular cartilage in experimental osteoarthrosis. Biochem J 1980;188:823e37. Matyas JR, Adams ME, Huang D, Sandell LJ. Discoordinate gene expression of aggrecan and type II collagen in experimental osteoarthritis. Arthritis Rheum 1995;38:420e5. Hui-Chou CS, Lust G. The type of collagen made by the articular cartilage in joints of dogs with degenerative joint disease. Coll Relat Res 1982;2:245e56. Burton-Wurster N, Hui-Chou CS, Greisen HA, Lust G. Reduced deposition of collagen in the degenerated articular cartilage of dogs with degenerative joint disease. Biochim Biophys Acta 1982;718:74e84. Aigner T, Zien A, Gehrsitz A, Gebhard PM, McKenna L. Anabolic and catabolic gene expression pattern analysis in normal versus osteoarthritic cartilage using complementary DNA-array technology. Arthritis Rheum 2001;44:2777e89. Gebhard PM, Gehrsitz A, Bau B, So¨der S, Eger W, Aigner T. Quantification of expression levels of cellular differentiation markers does not support a general shift in the cellular phenotype of osteoarthritic chondrocytes. J Orthop Res 2003;21:96e101. Poole AR, Nelson F, Hollander A, Reiner A, Pidoux I, Ionescu M. Collagen II turnover in joint disease. Acta Orthop Scand 1995;266:88e91. Mouritzen U, Christgau S, Lehmann HJ, Tanko LB, Christiansen C. Cartilage turnover assessed with a newly developed assay measuring collagen type II degradation products: influence of age, sex, menopause, hormone replacement therapy, and body mass index. Ann Rheum Dis 2003;62:332e6. Kira´ly K, Hyttinen MM, Parkkinen JJ, Arokoski JA, Lapvetelainen T, To¨rro¨nen K, et al. Articular cartilage collagen birefringence is altered concurrent with changes in proteoglycan synthesis during dynamic in vitro loading. Anat Rec 1998;251:28e36. Arokoski JP, Hyttinen MM, Lapvetelainen T, Takacs P, Kosztaczky B, Modis L, et al. Decreased birefringence of the superficial zone collagen network in the canine knee (stifle) articular cartilage after long distance running training, detected by quantitative polarised light microscopy. Ann Rheum Dis 1996;55:253e64. Torzilli PA, Grigiene R, Borrelli J, Helfet DL. Effect of impact load on articular cartilage: cell metabolism and viability, and matrix water content. J Biomech Eng 1999;121:433e41. Quinn TM, Grodzinsky AJ, Hunziker EB, Sandy JD. Effects of injurious compression on matrix turnover

22.

23.

24.

25.

26. 27.

28.

29.

30.

31.

32.

33. 34.

35.

36. 37.

around individual cells in calf articular cartilage explants. J Orthop Res 1998;16:490e9. Chen CT, Burton-Wurster N, Lust G, Bank RA, TeKoppele JM. Compositional and metabolic changes in damaged cartilage are peak stress, stress-rate and loading-duration dependent. J Orthop Res 1999;17: 870e9. Chen CT, Bhargava M, Lin PM, Torzilli PA. Time, stress, and location dependent chondrocyte death and collagen damage in cyclically loaded articular cartilage. J Orthop Res 2003;21:888e98. Thibault M, Poole AR, Buschmann MD. Cyclic compression of cartilage/bone explants in vitro leads to physical weakening, mechanical breakdown of collagen and release of matrix fragments. J Orthop Res 2002;20:1265e73. Farquhar T, Xia Y, Mann K, Bertram J, Burton-Wurster N, Jelinski L, et al. Swelling and fibronectin accumulation in articular cartilage explants after cyclical impact. J Orthop Res 1996;14:417e23. Wolf A, Raiss RX, Steinmeyer J. Fibronectin metabolism of cartilage explants in response to the frequency of intermittent loading. J Orthop Res 2003;21:1081e9. Sauerland K, Plaas AH, Raiss RX, Steinmeyer J. The sulfation pattern of chondroitin sulfate from articular cartilage explants in response to mechanical loading. Biochim Biophys Acta 2003;1638:241e8. Steinmeyer J, Ackermann B, Raiss RX. Intermittent cyclic loading of cartilage explants modulates fibronectin metabolism. Osteoarthritis Cartilage 1997;5: 331e41. Steinmeyer J, Knue S, Raiss RX, Pelzer I. Effects of intermittently applied cyclic loading on proteoglycan metabolism and swelling behaviour of articular cartilage explants. Osteoarthritis Cartilage 1999;7:155e64. Burton-Wurster N, Lust G. Fibronectin and proteoglycan synthesis in long term cultures of cartilage explants in Ham’s F-12 supplemented with insulin and calcium: effects of the addition of TGF-b. Arch Biochem Biophys 1990;283:27e33. Van der Kraan PM, Vitters EL, de Vries BJ, van den Berg WB. High susceptibility of human articular cartilage glycosaminoglycan synthesis to changes in inorganic sulfate availibility. J Orthop Res 1990;8: 565e71. Verbruggen G, Malfatt A-M, Dewulf M, Broddelez C, Veys EM. Standardization of nutrient media for isolated human articular chondrocytes in gelified agarose suspension culture. Osteoarthritis Cartilage 1995;3:249e59. Steinmeyer J. A computer-controlled mechanical culture system for biological testing of articular cartilage explants. J Biomech 1997;30:841e5. Peterkofsky B, Diegelmann R. Use of a mixture of proteinase-free collagenases for the specific assay of radioactive collagen in the presence of other proteins. Biochemistry 1971;10:988e94. Peterkofsky B. Estimation of total collagen production by collagenase digestion. In: Haralson MA, Hassel JR, Eds. Extracellular Matrix, a Practical Approach. New York: Oxford University Press 1995;33e8. Simon WH. Scale effects in animal joints. I. Articular cartilage thickness and compressive stress. Arthritis Rheum 1970;3:244e56. Matthes LS, Sonstegard DA, Henke JA. Load bearing characteristics of the patello-femoral joint. Acta Orthop Scand 1977;48:511e6.

914

B. Ackermann and J. Steinmeyer: Cartilage collagen and intermittent loading

38. Hodge WA, Fijan RS, Carlson KL, Burgess RG, Harris WH, Mann RW. Contact pressures in the human hip joint measured in vivo. Proc Natl Acad Sci U S A 1986; 83:2879e83. 39. Spanheimer RG, Peterkofsky B. A specific decrease in collagen synthesis in acutely fasted, vitamin Csupplemented, guinea pigs. J Biol Chem 1985;260: 3955e62. 40. Spanheimer RG, Zlatev T, Umpierrez G, Digirolamo M. Collagen production in fasted and food-restricted rats: response to duration and severity of food deprivation. J Nutr 1991;121:518e24. 41. Bienkowski RS, Gotkin MG. Control of collagen deposition in mammalian lung. Proc Soc Exp Biol Med 1995;209:118e40. 42. Beekman B, Verzijl N, Bank RA, von der Mark K, TeKoppele JM. Synthesis of collagen by bovine chondrocytes cultured in alginate; posttranslational modifications and cellematrix interaction. Exp Cell Res 1997;237:135e41. 43. Smith RL, Rusk SF, Ellison BE, Wessells P, Tsuchiya K, Carter DR, et al. In vitro stimulation of articular chondrocyte mRNA and extracellular matrix synthesis by hydrostatic pressure. J Orthop Res 1996;14: 53e60. 44. Holmvall K, Camper L, Johansson S, Kimura JH, Lundgren-Akerlund E. Chondrocyte and chondrosarcoma cell integrins with affinity for collagen type II and their response to mechanical stress. Exp Cell Res 1995;221:496e503.

45. Toyoda T, Saito S, Inokuchi S, Yabe Y. The effects of tensile load on the metabolism of cultured chondrocytes. Clin Orthop 1999;359:221e8. 46. Das P, Schurman DJ, Smith RL. Nitric oxide and G proteins mediate the response of bovine articular chondrocytes to fluid-induced shear. J Orthop Res 1997;15:87e93. 47. Millward-Sadler SJ, Wright MO, Lee HS, Caldwell H, Nuki G, Salter DM. Altered electrophysiological responses to mechanical stimulation and abnormal signalling through a5b1 integrin in chondrocytes from osteoarthritic cartilage. Osteoarthritis Cartilage 2000; 8:272e8. 48. Wright M, Jobanputra P, Bavington C, Salter DM, Nuki G. Effects of intermittent pressure-induced strain on the electrophysiology of cultured human chondrocyte: evidence for the presence of stretch-activated membrane ion channels. Clin Sci 1996;60:61e71. 49. Boustany NN, Gray ML, Black AC, Hunziker EB. Correlation between synthetic activity and glycosaminoglycan concentration in epiphyseal cartilage raises questions about the regulatory role of interstitial pH. J Orthop Res 1995;13:733e9. 50. Kim YJ, Grodzinsky AJ, Plaas AHK. Compression of cartilage results in differential effects on biosynthetic pathways for aggrecan, link protein, and hyaluronan. Arch Biochem Biophys 1996;328:331e40. 51. Baskin L, Howard PS, Macarak E. Effect of mechanical forces on extracellular matrix synthesis by bovine urethral fibroblasts in vitro. J Urol 1993;150:637e41.