© Birkhäuser Verlag, Basel, 2006 Inflamm. res. 55 (2006) 60–65 1023-3830/06/020060-06 DOI 10.1007/s00011-005-0014-4

Inflammation Research

Comparison of the pharmacology of hydroxamate- and carboxylate-based matrix metalloproteinase inhibitors (MMPIs) for the treatment of osteoarthritis M. J. Janusz, E. B. Hookfin, K. K. Brown, L. C. Hsieh, S. A. Heitmeyer, Y. O. Taiwo, M. G. Natchus, S. Pikul, N. G. Almstead, B. De, S. X. Peng, T. R. Baker and V. Patel Procter & Gamble Pharmaceuticals Inc., 8700 Mason-Montgomery Rd., Mailbox 1069, Mason OH 45040-9462, USA, Fax: ++513 622 3681; e-mail: [email protected] Received 12 April 2005; returned for revision 24 July 2005; accepted by J. Skotnicki 20 October 2005

Abstract. Objective and Design: Hydroxamic-and carboxylic-acid based matrix metalloproteinase inhibitors (MMPIs) were compared for their potency against various MMPs, pharmacodynamic properties and in vivo efficacy in a model of cartilage degeneration. Materials and Methods: The MMPIs were evaluated for their ability to inhibit human MMPs using the quenched fluorescence assay. The ability of the MMPIs to inhibit the degeneration of the knee joint was evaluated in rats injected intraarticularly with iodoacetate. The amount of MMPI in the plasma and cartilage was determined using liquid chromatography/mass spectrometry/mass spectrometry (LC/ MS/MS). Plasma protein binding was measured by ultrafiltration and unbound MMPI was quantitated using HPLC. Results: The hydroxamic acid based inhibitor PGE-3321996 and the carboxylic acids PGE-2909492 and PGE-6292544 were potent MMP-13 inhibitors, but only the hydroxamic acid PGE 3321996 demonstrated significant inhibition of knee degeneration in the rat iodoacetate model. Both of the carboxylic acids demonstrated superior pharmacokinetic properties and established much higher plasma concentrations than the hydroxamic acid. However, neither of the carboxylic acids was detectable in the cartilage, whereas, the hydroxamic acid was present in both the cartilage and the plasma. The carboxylic acid based MMPIs also demonstrated higher plasma protein binding (>99 %) than the hydroxamic acid (79 %). Conclusions: Carboxylic acid-based MMPIs were identified that had superior in vivo plasma exposure compared to a hydroxamic acid inhibitor but lacked in vivo efficacy in the rat iodoacetate model of cartilage degeneration. The lack of in vivo efficacy of the carboxylic acid based MMPIs were probably due to their lack of cartilage penetration which was related to their physicochemical properties.

Correspondence to: M. J. Janusz

Key words: Matrix metalloproteinase inhibitors – Hydroxamic acid – Carboxylic acid

Introduction The matrix metalloproteinases (MMPs) are a large family of zinc and calcium dependent neutral endopeptidases [1] that are involved in extracellular matrix turnover in both normal physiological processes and in pathological conditions. The MMPs are regulated at the level of synthesis, activation and by direct inhibition by endogenous inhibitors. Unregulated MMP activity has been implicated in the cartilage matrix degradation that occurs in osteoarthritis (OA) and rheumatoid arthritis (RA) [2–8]. Many types of MMP inhibitors have been synthesized as potential therapeutic agents and the types of compounds and their activities have been reviewed [9]. Synthetic inhibitors of the MMPs employ a number of chelating groups to coordinate the catalytic zinc in the active site including hydroxamates, carboxylates, thiol and phosphorous-based ligands. The hydroxamic acid functionality has traditionally been the ligand of choice for the most potent MMP inhibitors. A number of potent hydroxamic acid based MMP inhibitors have been shown to be active in animal models of arthritis. The hydroxamates GI-168, BB-1101 and BB-1433 have been shown to inhibit cartilage and bone destruction in the rat adjuvant model of arthritis [10, 11]. The hydroxamate Ro 32-3555 inhibited the degradation of cartilage in a model of monoarthritis induced by the intraarticular injection of Propionibacterium acnes [12]. Several different hydroxamic acid based inhibitors have been shown to reduce joint degeneration in models of osteoarthritis including the rat iodoacetate model [13, 14], the STR/ORT spontaneous mouse model [15], and a guinea pig meniscectomy model [16]. There is considerably less information on the effect of carboxylic acid based inhibitors in animal models of

Vol. 55, 2006

Comparison of the pharmacology of hydroxamate-and carboxylate-based matrix metalloproteinase inhibitors (MMPIs)

arthritis. BAY 12-9566 was reported in an abstract [17] to inhibit cartilage degradation in canine and guinea pig medial menisectomy models. Recently, this compound was shown to reduce proteoglycan turnover in patients that received compound for 3 weeks prior to knee surgery for OA [18]. In this study, we describe the in vitro and in vivo activity of representative MMPIs of the hydroxamic and carboxylic acid classes. The hydroxamic acid MMPI was efficacious in the rat iodoacetate model of cartilage degeneration, whereas, the carboxylic acid based MMPIs were not efficacious in this model even though they displayed superior pharmacokinetic properties compared to the hydroxamic acid MMPI. We present data on the importance of the physical properties of MMPIs for obtaining chondroprotective effects in vivo. Materials and methods Rat iodoacetate model Sprague-Dawley male rats weighing 220-230 grams (Harlan, Indianapolis, IN) were housed singly in wire cages in sanitary ventilated animal rooms with controlled temperature, humidity and regular light cycles. Rodent chow (Ralston-Purina, Richmond, IN) and water were available ad libitum. Animals were acclimated for at least one week before use. All animals were housed, fed, and handled in compliance with the standards set forth by the Animal Welfare Act as amended. Where standards are not indicated in the Animal Welfare Act the recommendations on HHS Publications (NIH) No. 85-23, “Guide for the Care and Use of Laboratory Animals” were followed. Arthritis was induced by a single intraarticular injection of iodoacetate into the knee joint of rats anesthetized using (3: 1) CO2/O2 as described previously [13] A 10 mg/ml concentration of monosodium iodoacetate (MIA) (Aldrich Chemical, Milwaukee, WI) was prepared using injectable saline as the vehicle. After appropriate anesthesia each rat was positioned on its back and the left leg was flexed 90 degrees at the knee. The patellar ligament was palpated below the patella and the injection was made into this region. Each rat received 0.025 ml intraarticular injection into the left knee using a glass gas tight syringe with a 27 gauge ½ inch needle. Care was taken not to advance the needle in too far into the cruciate ligaments. Animals were sacrificed 21 days after iodoacetate injection and the left knees of the sacrificed animals were disarticulated and the tibial plateau imaged using an Optimas image analyzer. The tibial plateau was used for image analysis because it provided a relatively flat surface compared with the femoral condyles, allowing the image analysis camera to focus on the entire cartilage surface. The severity of damage in the magnified images was assessed by three independent observers in a

A

B

61

blinded manner using a scale of (0–4) of increasing severity (0 = normal; 4 = maximum severity) as described previously [13] and illustrated in Figure 1. Data were analyzed using a nonparametric procedure (Wilcoxon rank sum). The data are expressed as the mean ± S.E.M, n = 15 and values exhibiting a statistical difference from the vehicle treated control (p < 0.05) are denoted with an asterisk.

MMP inhibition assay Previously, we have described the inhibition of MMP activity in detail [13] as well as the preparation of the human recombinant enzymes used in these studies [19]. The MMP inhibitors used in the present study were evaluated for their ability to inhibit human MMPs using the quenched fluorescence assay [20] modified to fit a 96-well format. Briefly, MMPs 1, 2, 3, 7, 8, 9 and 13 were used at final concentrations of 8 nM, 1 nM, 16 nM, 2 nM, 4 nM, 0.75 nM and 0.5 nM, respectively. The MMP assays were performed using the fluorogenic substrate Mca-Pro-Leu-Gly-LeuDpa-Ala-Arg-NH2 at a concentration of 4 µM at 25 °C. The assay buffer was 50 mM Tris, pH 7.5, 0.2 M NaCl, 5 mM CaCl2 and 0.02 % Brij-35. The increase in fluorescence due to cleavage of the substrate (Gly-Leu bond) was monitored kinetically for 30 min with a BMG Fluostar fluorescence plate reader (lex328 nm, lem 393 nm) (BMG LabTechnology GmbH, Offenburg, Germany). Each 96 well microtiter plate contained 100 µl of substrate and 50 µl of enzyme in each well. 50 µl of MMP inhibitor was added to each well (except for positive control) to give a final volume of 200 µl/well. The MMP inhibitors were tested at 8 different concentrations and an IC50 was calculated using the formula: Vi/Vo = 1/1+ [I]/IC50 where Vi is the initial velocity of substrate cleavage in the presence of inhibitor at concentration [I] and Vo is the initial velocity in the absence of inhibitor.

Quantitation of MMPIs in plasma and cartilage. Plasma and cartilage were evaluated for exposure to the MMPIs at the end of the rat iodoacetate studies. Samples (plasma & cartilage) were taken from study animals 1, 2, 4, 8 and 12 h after a last dose of MMPI. Both plasma and cartilage samples were collected from each of the three animals per time period. After analysis plasma and cartilage values were averaged to generate a composite profile. Liquid chromatography/mass spectrometry/mass spectrometry (LC/MS/MS) methods were utilized to determine MMPI levels in rat plasma and cartilage samples. The plasma samples were vortexed prior to sampling and then prepared utilizing a protein precipitation step with a five-fold excess of acetonitrile. Cartilage samples were pulverized in tissue grinders prior to a double liquid extraction with methanol. Following centrifugation, the supernatants from plasma and cartilage extracts were dried and reconstituted in diluent no stronger than the Fig. 1. Scoring of joint degeneration in the rat iodoacetate model. Cartilage damage was assessed from images captured and magnified using an image analyzer by three independent observers in a blinded manner using a scale of 0–4 of increasing severity. A) Shows an image of a normal joint with no degeneration; B) shows severe cartilage damage (score of 4) involving the majority of the tibial plateau with large erosions extending down to the subchondral bone.

62

M. J. Janusz et al.

mobile phase. The assay was performed using a gradient reversed-phase HPLC separation followed by a selective tandem mass spectrometry detection (ionspray introduction followed by selected reaction monitoring or SRM) using an API 3000(Applied Biosystems, Toronto, Canada). An SRM channel was monitored for the analyte and an internal standard (IS) was utilized.

HO N H

O OO S N

Inflamm. res.

O

O

Determination of plasma protein binding

O Plasma protein binding of MMPIs was measured by ultrafiltration where unbound drug can pass through a membrane and its concentration measured in the ultrafiltrate (21). Solutions of the MMPIs were prepared in phosphate buffered saline (nonspecific binding) or rat plasma and were incubated for 15 min at 37 °C followed by ultrafiltration through an Amicon micropartition filter (30 kDa cutoff). The ultrafiltrates were collected, centrifuged at 2000 rpm for 25 min and the amount of unbound compound was quantitated by HPLC.

PGE-3321996

OMe

Pharmacokinetic Evaluations MMPI plasma concentration-time data were analyzed by noncompartmental pharmacokinetic methods. The areas under the plasma concentration – time curve from 0–12 h (AUC) – were determined using the linear-trapezoidal method (software: AUC V5.0). tn

AUC =

n–1

∫ C dt = ∑ n

0

i=0

O O S O NH

HO

ti + 1 – ti × (Ci + Ci + 1) 2

Where Cn is the final observed concentration at time tn.

PGE-2909492

The average plasma concentration (Cav) was also calculated. Cav = AUC/r Where r is the dosing interval (12 h).

O

Concentration ng/ml

10000

O HO

1000

H N

SO2

S 100

10

PGE-6292544

1 0

2

4

6

8

10

12

Fig. 2. Structures of the hydroxamic acid MMPI PGE-3321996 and the carboxylic acid MMPI’s PGE-2909492 and PGE-6292544.

Time (Hr) Fig. 3. Pharmacokinetic profile of the MMPIs in the rat. The mean plasma concentration of PGE-3321996 (open circles), PGE-2909492 (open squares), and PGE-6292544 (closed squares) following oral administration of 25 mg/kg in the rat iodoacetate model (summarized in Table 2) are shown. The cartilage concentrations of PGE-3321996 are shown as closed circles. The cartilage concentration of PGE-2909492 and PGE 6292544 were below the level of detection. The data are the mean from 3 rats sacrificed at various times after receiving the last dose of MMPI.

Matrix metalloproteinase Inhibitors The matrix metalloproteinase inhibitors utilized in these studies were synthesized at Procter and Gamble Pharmaceuticals. The structures of these compounds are shown in Figure 2 were verified by 1H nuclear magnetic resonance, mass spectroscopy and elemental analysis.

Vol. 55, 2006

Comparison of the pharmacology of hydroxamate-and carboxylate-based matrix metalloproteinase inhibitors (MMPIs)

63

Table 1. The inhibition of MMPs by hydroxamic and carboxylic acid based MMPIs IC50 (nM) MMPI

MMP-1

MMP-2

MMP-3

MMP-7

MMP-8

MMP-9

MMP-13

PGE-3321996

946 ± 91

0.5 ± 0.1

7.2 ± 1.6

2036 ± 235

0.7 ± 0.2

0.8 ± 0.2

0.6 ± 0.1

PGE-2909492

1961 ± 425

13 ± 0.7

2283 ± 418

> 10000

9.4 ± 3.5

553 ± 123

21 ± 1.8

PGE-6292544

3508 ± 425

24 ± 1.5

1693 ± 537

> 10000

334 ± 0.1

797 ± 304

29 ± 4.7

The MMP inhibitory activity of the MMPIs was measured using the quenched fluorescence assay with Mca-Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH2 as fluorogenic substrate (20). Cleavage of the substrate by MMPs was monitored kinetically for 30 min with a fluorescence plate reader. Data are the mean ±S.D. from 3 experiments.

The data are expressed as the mean % inhibition of joint damage in the iodoacetate model ± S.E.M. The asterisk indicates statistical significance P < 0.05 using a non-parametric procedure (Wilcoxon rank sum). 2 Plasma and cartilage concentrations of MMPI are expressed as the average concentration of compound (Cav) over a 12 h dosing period as calculated using the area under the curve. 3 N.D. means MMPI was not detectable. The limit of detection for PGE6292544 was 222 pM and PGE-2909492 was 58 pM.

strated statistically significant inhibition of iodoacetate induced joint damage, whereas neither of the carboxylic acids demonstrated statistically significant efficacy in this model (Table 2). Plasma and cartilage from the rat iodoacetate studies were collected at the end of the studies to determine exposure levels to the compounds. The plasma exposure curves for these compounds are presented in Figure 2 and summarized in Table 2. The average plasma concentrations of the carboxylic acid MMPIs PGE-6292544 and PGE-2909492 over a 12 h period (Cav) were 2047 ng/ml (3937 nM) and 198 ng/ml (439 nM), respectively which was considerably higher than the plasma Cav for the hydroxamate PGE3321996 which was 25 ng/ml (61 nM) (Table 2). However, neither of the carboxylic acids was detectable in cartilage, whereas, the cartilage concentrations of the hydroxamic acid were easily detectable and were higher than the plasma concentrations (Table 2).

Results

The role of the physicochemical properties of the MMPIs in cartilage penetration

Table 2. Inhibition of joint degeneration in the rat iodoacetate model by hydroxamic but not by carboxylic acid based MMPIs. MMPI

Joint MMPI Degeneration Plasma Conc. (% Inhibition)1 (Cav) (ng/ml)2

PGE-3321996

28 ± 2*

PGE-2909492

0

PGE-6292544

12 ± 2

25

MMPI Cartilage Conc. (Cav) (nM)2 282

198

N.D.3

2047

N.D.

1

MMP inhibitory profile of selected carboxylic-acid and hydroxamic based MMPIs. The MMP inhibitory profile of the two carboxylic acid-based MMPIs PGE-6292544 and PGE-2909492 and a hydroxamic acid-based inhibitor PGE-3321996 were compared under identical conditions using the quenched fluorescence assay. The carboxylic acid-based inhibitors were potent low nanomolar inhibitors of MMPs 2, 8 and 13 whereas, the hydroxamate PGE-3321996 was a very potent low nanomolar inhibitor of MMPs 2, 3, 8, 9 and 13 (Table 1). All of the MMPIs were weak inhibitors of the shallow binding pocket MMPs 1 and 7 (Table 1). The efficacy of the MMPIs in the rat iodoacetate model of joint degeneration Intraarticular injection of sodium iodoacetate into the knee joint of rats results in rapid degeneration of the cartilage and subchondral bone [13, 22, 23]. The effect of oral administration of the MMPIs twice daily (25 mg/kg) on degeneration of the knee joint in iodoacetate injected rats was assessed. The hydroxamic acid-based MMPI PGE-3321996 demon-

The physicochemical properties of compounds can have profound effects on their ability to achieve efficacious concentrations in their target organs. Since all of the MMPIs used in this study were small molecule inhibitors of similar molecular weight (~ 400–500) the size of the compounds were not a significant factor in cartilage penetration. Several characteristics of the carboxylic acid MMPIs could be responsible for their lack of exposure in cartilage in the rat iodoacetate model. It has been shown previously in a series of hydroxamic MMPIs that increasing hydrophilicity increased the permeability coefficient in cartilage [24]. The hydroxamic acid MMPI PGE-3321996 had a calculated Log P (cLog P) of 1.7 which was significantly more hydrophilic than the carboxylic acid MMPIs PGE-2909492 and PGE6292544 which had cLog Ps of 5.95 and 6.34, respectively (Table 3). The water solubility at pH 7.0 of the hydroxamic acid MMPI (2.4 mg/ml) was greater than that of the carboxylic acids (PGE 2909492, 0.02 mg/ml and PGE-6292544, 0.17 mg/ml) which correlates with the cLog Ps of these compounds (Table 3). Carboxylic acids are known to be high protein binders [25] so the protein binding of the MMPIs was investigated to determine if the lack of cartilage penetration by the carboxylic acid MMPIs could be due to high protein binding. The

64

M. J. Janusz et al.

Table 3. Physicochemical properties of the MMPIs. 1

MMPI

cLogP

Solubility (mg/ml) at pH 7.42

%Protein Binding3

PGE-3321996

1.9

2.4

79

PGE-2909492

5.95

0.02

99.4

PGE-6292544

6.34

0.17

99.6

1

cLogP was calculated using ACD laboratory software version 8.0 (Advanced Chemistry Development , Inc. Toronto, Canada). 2 Solubility was determined in 50 mM phosphate buffer pH 7.4, ionic strength 0.15 M at room temperature by the shake flask method after 24 h of equilibration. Concentrations were assessed for supernatants of centrifuged samples by UV-Vis spectroscopy. 3 Protein binding was assessed at 1 and 10 µM of MMPI.

hydroxamic acid MMPI PGE-3321996 demonstrated significantly lower plasma protein binding (79 %) compared to the carboxylic acid MMPIs PGE-2909492 and PGE-6292544 which demonstrated high plasma protein binding of 99.4 % and 99.6 %, respectively (Table 3). Discussion A number of studies have been published demonstrating the in vivo efficacy of hydroxamic based MMPIs in different animal models of rheumatoid and osteoarthritis [10–16]. However, there is little information on the in vivo efficacy of carboxylic acid based MMPIs in animal models of arthritis with the exception of a meeting abstract reporting the efficacy of the carboxylic acid MMPI BAY 12-9566 in canine and guinea pig medial menisectomy models [17]. Carboxylate MMPIs have been reported to have efficacy in other disease models. The carboxylic acid based MMPI PD166793 was shown to attenuate left ventricular dilation and progression of left ventricular failure in a rodent heart failure model [26]. Similarly, carboxylate MMPIs have demonstrated efficacy in a melanoma model in mice [27]. In the present study and in a previous report, we have shown a number of hydroxamic acid MMPIs to attenuate cartilage degeneration in the iodoacetate model of arthritis [13]. We and others have pursued carboxylic acid based MMPIs to address some of the pharmacokinetic liabilities observed with the hydroxamic acid MMPIs. Although the carboxylates PGE-6292544 and PGE-2909492 demonstrated superior plasma exposures in rats compared to the hydroxamate PG-3321996 (Fig. 2 and Table 2), the carboxylic acids did not demonstrate efficacy in the rat iodoacetate model. These data are consistent with the lack of efficacy we have seen with all of the carboxylic acid based MMPIs (10 different compounds) that we tested in the rat iodoacetate model (data not shown). The carboxylic acid MMPIs PGE 6202544 and PGE-2909492 were less potent inhibitors of MMP 2, 3, 9 and 13 compared to the hydroxamic acid PGE 3321996. MMP-13 is thought to be a key mediator of cartilage degradation in osteoarthritis. MMP-13 has been shown to be an important MMP for the cleavage of type II collagen (28) and overexpression of MMP-13 in transgenic mice led to the de-

Inflamm. res.

velopment of osteoarthritis (29). Knockout mice have been employed in a number of different models of rheumatoid arthritis to attempt to determine the role of various MMPs in disease pathogenesis. Antigen-induced arthritis was shown to be less severe in MMP-9 knockout mice suggesting a role for MMP-9 in arthritis whereas; arthritis was more severe in MMP-2 knockout mice (30). Similarly, collagen-induced arthritis was as severe in MMP-3 knockout mice as in wild type mice suggesting that this enzyme was not required for the arthritis process (31). However, although the carboxylic acid MMPIs were less potent than the hydroxamate the most likely reason for the lack of efficacy of the carboxylates was their inability to be distributed to the cartilage (Table 2). The efficacy of a drug is related to its ability to reach its physiological target at adequate concentrations. The upregulation and activation of MMPs in the cartilage are believed to play an important role in the cartilage degradation that occurs in osteoarthritis. The injection of iodoacetate results in an increase in MMP activity in the cartilage of rats [13]. Therefore, it is important that the MMP inhibitor gain access to the cartilage matrix. Neither of the carboxylic acids was detected in the cartilage of rats after oral dosing although they obtained high concentrations in the plasma (Table 2) and had longer half lives (data not shown) than the hydroxamic acid MMPI. There are several possible explanations for the lack of cartilage penetration of the carboxylic acid MMPIs. Previously, it was demonstrated with a series of hydroxamic acid MMPIs that increasing hydrophilicity increased the permeability coefficient in cartilage [24]. The carboxylic acid MMPIs were more hydrophobic and less soluble than the hydroxamate (Table 3) which could have adversely affected the cartilage penetration of these compounds. However, at physiological pH, the carboxylic acid MMPIs would be completely ionized and therefore may have difficulty penetrating through the negatively charged glycosaminoglycan side chains of the cartilage aggrecan. In addition, the carboxylic acids demonstrated very high protein binding compared to the hydroxamate (Table 3) which could have hindered the penetration of compound into the cartilage. These data demonstrate that the physicochemical properties of a MMPI are very important in the development of compounds that are active in disease models of osteoarthritis where cartilage penetration is important for obtaining efficacy.

References [1] Clark IM, Parker AE. Metalloproteinase: their role in arthritis and potential as therapeutic targets. Expert Opin Ther Targets 2003; 7: 19–34. [2] Gravallese EM., Darling JM, Ladd AL, Katz JN, Glimcher LH. In situ hybridization studies of stromelysin and collagenase messenger RNA expression in rheumatoid synovium. Arthritis Rheum 1991; 34: 1074–84. [3] Lohmander LS, Hoerrner LA, Lark MW. Metalloproteinases, tissue inhibitor, and proteoglycan fragments in knee synovial fluid in human osteoarthritis. Arthritis Rheum 1996; 36: 181–89. [4] Walakovits LA, Moore VL, Bhardwaj N, Gallick GS, Lark MW. Detection of stromelysin and collagenase in synovial fluid from patients with rheumatoid arthritis and posttraumatic knee injury. Arthritis Rheum 1992; 35: 35–41.

Vol. 55, 2006

Comparison of the pharmacology of hydroxamate-and carboxylate-based matrix metalloproteinase inhibitors (MMPIs)

[5] Macachren SS. Expression of metalloproteinases and metalloproteinase inhibitor in human arthritic synovium. Arthritis Rheum 1991; 34: 1085–93. [6] Mitchell PG, Magna HA, Reeves LM, Lopresti-Morrow LL, Yocum SA, Rosner PJ et al. Cloning, expression, and type II collagenolytic activity of matrix metalloproteinase-13 from human osteoarthritic cartilage. J Clin Invest 1996; 97: 761–68. [7] Moldovan F, Pelletier JM, Hambor J, Cloutier JM, Pelletier JM. Collagenase-3 (matrix metalloproteinase 13) is preferentially localized in the deep later of human arthritic cartilage in situ. In vitro mimicking effect by transforming growth factor b. Arthritis Rheum 1997; 40: 1653–61. [8] Wernicke D, Seyfert C, Hinzmann B, Gromnica-Ihle E. Cloning of collagenase 3 from the synovial membrane and its expression in rheumatoid arthritis and osteoarthritis. J Rheumatol 1996; 23: 590–95. [9] Skiles JW, Gonnella NC, Jeng AY. The design, structure and therapeutic application of matrix metalloproteinase inhibitors. Curr Med Chem 2001; 8: 425–74. [10] Conway JG, Wakefield JA, Brown RH, Marron BE, Sekut L, Stimpson SA et al. Inhibition of cartilage and bone destruction in adjuvant arthritis in the rat by a matrix metalloproteinase inhibitor. J Exp Med 1995; 182: 449–57. [11] DiMartino M, Wolff C, High W, Stroup G, Hoffman S, Laydon J et al. Anti-arthritic activity of hydroxamic acid-based pseudopeptide inhibitors of matrix metalloproteinases and TNFa processing. Inflamm Res 1997; 46: 211–15. [12] Lewis EJ, Bishop J, Bottomley KMK, Bradshaw D, Brewster M, Broadhurst MJ et al. Ro 32-3555, an orally active collagenase inhibitor, prevents cartilage breakdown in vitro and in vivo. Br J Pharmacol 1997; 121: 540–46. [13] Janusz MJ, Hookfin EB, Heitmeyer SA, Woessner JF, Freemont AJ, Hoyland JA et al. Moderation of the iodoacetate-induced experimental osteoarthritis in rats by matrix metalloproteinase inhibitors. Osteoarthritis Cartilage 2001; 9: 751–60. [14] Natchus MG, Bookland RG, De B, Almstead NG, Pikul S, Janusz MJ et al. J Med Chem 2000; 43: 4948–63. [15] Brewster M, Lewis EJ, Wilson KL, Greenham AK, Bottomley KMK. Ro-32-3555, an orally active collagenase selective inhibitor, prevents structural damage in the STR/ORT mouse model of osteoarthritis. Arthritis Rheum 1998; 1639–44. [16] Sabatini M, Lesur C, Thomas M, Chomel A, Anract P, de Nanteuil G et al. Effect of inhibition of matrix metalloproteinases on cartilage loss in vitro and in a guinea pig model of osteoarthritis. Arthritis Rheum 2005; 171–80. [17] Chau T, Jolly G, Plym JM, McHugh M, Bortolon E, Wakefield J et al. Inhibition of articular cartilage degradation in dog and guinea pig models of osteoarthrtis by the stromelysin inhibitor BAY 129566 Arthritis Rheum 1998; 41(9S): S300. [18] Leff RL, Elias I, Ionescu M, Reiner A, Poole AR. Molecular changes in human osteoarthritic cartilage after 3 weeks of oral ad-

[19]

[20] [21]

[22] [23]

[24] [25] [26]

[27]

[28]

[29]

[30]

[31]

65

ministration of BAY 12-9566, a matrix metalloproteinase inhibitor. J Rheumatol 2003; 30: 544–49. Almstead NG, Bradley RS, Pikul S, De B, Natchus MG, Taiwo YO et al. Design, synthesis, and biological evaluation of potent thiazine- and thiazepine-based matrix metalloproteinase inhibitors. J Med Chem 1999; 42: 4547–62. Knight CG, Willenbrock F, Murphy G. A novel coumarin-labelled peptide for sensitive continuous assays of the matrix metalloproteinases FEBS 1992; 296: 233–36. Craig WA, Suh B. Protein binding and antimicrobial effects: Methods for the determination of protein binding. In Antibiotics in Laboratory Medicine; V Lorian Ed.; Williams and Wilkins 1991: 367–402. Kalbhen DA, Blum U. Theroretisches Koncept und experimentelle bestatigung fur ein neues arthrose-modell am versuchstier. Arzneimittelforshung Drug Res 1977; 27: 527–31. Guingamp C, Gegout-Pottie P, Philippe L, Terlain B, Netter P, Gillet P. Mono-iodoacetate-induced experimental osteoarthritis. A dose-response study of loss of mobility, morphology and biochemistry. Arthritis Rheum 1997; 40: 1670–79. Peng SX, VonBargen EC, Bornes DM, Pikul S. Permeability of articular cartilage to matrix metalloproteinase inhibitors. Pharm Res 1998; 15: 1414–18. Kragh-Hansen U, Chuang TG, Otagiri M. Practical aspects of the ligand-binding and enzymatic properties of human serum albumin. Biol Pharm Bull 2002; 25: 695–704. Peterson JT, Hallak H, Johnson L, Li H, O’Brien PM, Sliskovic DR et al. Matrix metalloproteinase inhibition attenuates left ventricular remodeling and dysfunction in a rat model of progressive heart failure. Circulation 2001; 103: 2303–09. Le Diguarher T, Chollet AM, Bertrand M, Hennig P, Raimbaud E, Sabatini M et al. Stereospecific synthesis of 5-substituted 2bisarylthiocyclopentane carboxylic acids as specific matrix metalloproteinase inhibitors. J Med Chem 2003; 46: 3840–52. Mitchell PG, Magna HA, Reeves LM,Lopresti-Morrow LL, Yocum SA, Rosner PJ et al. Cloning, expression, and type II collagenolytic activity of matrix metalloproteinase-13 from human osteoarthritic cartilage. J Clin Invest 1996; 97: 761–8. Newhold LA, Killar L, Zhao W, Sung ML, Warner L, Kulik J et al. Postnatal expression in hyaline cartilage of constitutively active human collagenase-3 (MMP-13) induces osteoarthritis in mice. J Clin Invest 2001; 107: 35–44. Itoh T, Matsuda H, Tanioka M, Kuwabara K, Itohara S, Suzuki R. The role of matrix metalloproteinase-2 and matrix metalloproteinase-9 in antibody-induced arthritis. J Immunol 2002; 169: 2643–47. Mudgett JS, Hutchinson NI, Chartrain NA, Forsyth AJ, McDonnell J, Singer II et al. Susceptibility of stromelysin-1 deficient mice to collagen-induced arthritis and cartilage destruction. Arthritis Rheum 1998; 41: 110–21.