Clodronate is a first-generation bisphosphonate monoclonal ... ate (0â3), and degree of SM thickening (0â2). The animals ... The two-tailed Student's t-test was used to determine ... the periarteriolar lymphatic sheath (PALS) and marginal zone.
Rheumatology 1999;38:818–825
Liposomal clodronate eliminates synovial macrophages, reduces inflammation and ameliorates joint destruction in antigen-induced arthritis P. J. Richards, A. S. Williams, R. M. Goodfellow and B. D. Williams Rheumatology Research Laboratory, University of Wales College of Medicine, Heath Park, Cardiff CF4 4XN, UK Abstract Objectives. To investigate the efficacy of a single i.v. dose of clodronate encapsulated within small unilamellar vesicles in suppressing joint inflammation and the histological progression of rat antigen-induced arthritis (AIA). Methods. Rats with AIA received a single i.v. injection of 20 mg of clodronate encapsulated within small unilamellar vesicles (SUVc) or larger multilamellar vesicles (MLVc) 7 days postarthritis induction. Free clodronate or saline were used as negative controls. Results. SUVc was shown to be more effective than MLVc, sustaining a significant reduction in knee swelling for up to 7 days after the initial systemic administration. Knee swelling in free clodronate-treated animals was not significantly affected. The increased efficacy of SUVc in reducing inflammation and joint destruction was associated with a significant depletion of resident ED1+, ED2+ and ED3+ macrophages from the synovial membrane (SM ). Conclusions. SUVc is more efficient than MLVc in reducing the severity of inflammation and joint destruction in rat AIA, and is associated with the specific elimination of macrophage subpopulations from the SM. K : Clodronate, Small unilamellar vesicles, Synovial macrophages, Reticuloendothelial system.
In many chronic inflammatory diseases, continued recruitment and activation of the monocytes/macrophages can result in tissue destruction and related pathology [1]. In the inflamed synovium of patients with rheumatoid arthritis, activated macrophages are found in abundance [2] and are present in strategic sites related to the distribution of destructive pannus [3]. Their secretory products (monokines) dominate the cytokine profile of synovial tissue and fluid [4–9]. Experimental models of arthritis have shown that macrophages are present in the inflamed synovium and that different subpopulations can be identified using different monoclonal antibodies [10–12]. Newly arrived immature macrophages/monocytes (identified with the monoclonal antibody ED1) outnumber the mature resident macrophages ( ED2+). In chronically inflamed synovial tissue, ED3+ macrophages, normally only found
in close association with T cells in lymphoid tissue, are also present [12]. Synovial macrophages are in the activated state, they express Class II antigens, and secrete tissue-damaging enzymes, the pro-inflammatory cytokines tumour necrosis factor alpha ( TNF-a), interleukin (IL)-1b and IL-6, prostaglandins and several reactive oxygen species [13]. They also show enhanced phagocytic activity. Local production of TNF-a and IL-1b by macrophages in the inflamed synovium occurs in both antigen-induced arthritis (AIA) and adjuvant arthritis (AA) models [14–17]. Macrophages in synovial tissue are highly phagocytic and are capable, therefore, of ingesting liposomes which localize to the synovium. Clodronate is a first-generation bisphosphonate which, when encapsulated within liposomes, is able to enter phagocytic cells and initiate apoptosis [18–22]. Systemically administered clodronate encapsulated in large multilamellar vesicles (MLV ), sterically stabilized with polyethylene glycol MS400 stearate (PEG-S), is capable of suppressing paw inflammation in rat AA [23]. These liposomes exerted their effect via a central
Submitted 15 December 1998; revised version accepted 24 March 1999. Correspondence to: P. J. Richards.
818
© 1999 British Society for Rheumatology
Liposomal clodronate in antigen-induced arthritis
immunoregulatory mechanism rather than by the removal of macrophages from synovial tissue. It is known that liposome size influences their localization to synovial tissue [24]. Small unilamellar vesicles (SUV ) have been shown to accumulate within inflamed paws of rats with AA to a much greater extent than large MLV. Systemically administered clodronate encapsulated in SUV can produce complete resolution of the clinical indices of inflammation, a normalization of the histology and a resolution of the bone changes seen in AA-treated rats [25]. This is associated with a significant reduction in ED1+ synovial macrophages in the inflamed paws. In the present study, we compare the effects of a single i.v. injection of either unencapsulated clodronate, MLV encapsulating clodronate (MLVc) or SUV encapsulating clodronate (SUVc) on the severity of inflammation and degree of synovitis in rat AIA. We investigate the influence of SUVc treatment on macrophage subpopulations in organs of the reticuloendothelial system (RES) and also within the synovial membrane (SM ).
Materials and methods Animals Male inbred Lewis rats were obtained from Bantin and Kingman ( The Field Station, Grimston, Hull, UK ). The animals were housed in cages of five, allowed food and water ad libitum, and kept in the Biomedical Services Department for 1 week prior to their first immunization. The animals were housed in light/dark cycles of 12 h. Induction of arthritis Arthritis was induced in the right knee joint of each rat by following the method previously described [26 ]. Briefly, on two occasions 1 week apart, male Lewis rats (150 g) were injected s.c. with an emulsion of equal volumes of methylated bovine serum albumin (mBSA; 0.5 mg; Sigma Chemical Co.) and Freund’s complete adjuvant (0.25 mg heat-killed Mycobacterium tuberculosis; Sigma Chemical Co.). Fourteen days after the second immunization, arthritis was induced with 100 ml of mBSA (1 mg) injected intra-articularly into the right knee. The development of arthritis was monitored at regular intervals by measuring knee diameters (mean of three readings), with the joint flexed at an angle of 90°, using a digital micrometer. Liposome entrapment of clodronate MLVs and SUVs encapsulating clodronate were prepared as described previously [27]. MLVs were composed of egg phosphatidylcholine and cholesterol (molar ratio 2:1). SUVs were produced by probe sonication (MSC Soniprep 150, 10 mm probe) of MLVc composed of egg phosphatidylcholine, cholesterol and dipalmitoyl phosphatidic acid (molar ratio 7:7:1); and a mean size of 100 nm was achieved after one 6-mm-amplitude burst for 5 min and two 10-mm-amplitude bursts for 10 min. The concentration of encapsulated clodronate within both MLV and SUV was determined by calculating the
819
amount of tracer [99mTc]clodronate remaining using a Wallac 1261 multigamma counter (LKB). Treatment of AIA Seven days after arthritis induction, animals were divided into four matched groups (six rats per group). At this time (day 0), animals were injected i.v. with either 2 ml of sterile saline (0.9% w/v), free clodronate (20 mg), MLVc (20 mg) or SUVc (20 mg). To assess the effect of the respective treatments, joint swelling was expressed as the difference in diameters between right and left knees. Histological grading of knee joint sections Three rats from each treatment group were killed 3 days after the respective treatments (10 days after arthritis induction). The knee joints were removed, trimmed, and simultaneously fixed and decalcified in Decalcifier IB (Surgipath). Joints were embedded in paraffin wax, then sectioned in the sagittal plane at 5 mm and stained with haematoxylin and eosin (H&E ). All sections were coded prior to assessment to eliminate observer bias and subsequently scored by an independent observer. The sections were graded subjectively using three parameters: degree of cartilage destruction and bone erosions (0–3), severity of synovial infiltration and inflammatory exudate (0–3), and degree of SM thickening (0–2). Immunohistochemistry Three rats from different treatment groups were killed 3 days after the respective treatments (10 days after arthritis induction). Cryostat sections (8 mm) of liver, spleen and undecalcified knee joints mounted on Superfrost/Plus slides (Scientific Lab. Supplies Ltd, Nottingham, UK ) were used for immunohistochemical analysis. The following monoclonal antibodies (mAb) were used for the identification of macrophage subpopulations: ED1 for immature resident monocyte/macrophages; ED2 and ED3 for mature resident macrophages (all from Serotec, UK ). All mAb were diluted in 1% BSA in phosphate-buffered saline (pH 7.4) (PBS/1% BSA) where indicated. Prior to immunostaining, sections were fixed in acetone for 10 min at 4°C and allowed to air dry. Nonspecific staining was blocked by incubation with normal rabbit serum (NRS; 1:10 in PBS/1% BSA) for 30 min in a humidity chamber at room temperature. The NRS was then removed and mAb ED1 (1:200), ED2 (1:400) or ED3 (1:500) added for 1 h at room temperature. After three, 1 min washes in PBS, sections were incubated with horseradish peroxidase-conjugated ratabsorbed rabbit anti-mouse IgG (1:50) for 1 h at room temperature. After washing, peroxidase was developed in 100 ml of PBS containing 50 mg of diaminobenzidine and 40 ml of 30% H O . For controls, the same staining 2 2 procedure was performed, but the specific mAb were replaced by an isotype-matched mouse mAb at identical concentrations.
820
P. J. Richards et al.
Image analysis Macrophage subpopulations were analysed quantitatively using a modification of the method previously described [28]. Briefly, slides were imaged using a Leica DMLB light microscope (Milton Keynes, UK ) with a ×10 objective and analysed with the computer-based image analysis system Improvision Density Slicing (OpenLab, Coventry, UK ). Illumination voltage, camera set-up and calibration parameters were kept constant throughout all measurements. Minimum object boundaries were defined with 35 pixels and manual correction of selected fields was performed to achieve a complete match between the visual screen mask and the original microscope fields. In each slide, at least three consecutive representative fields were evaluated, and the mean area of brown immunoperoxidase staining determined. A total of three rats per treatment group and three sections per rat were analysed. Statistical analysis The two-tailed Student’s t-test was used to determine whether knee swelling was significantly different between saline-, free clodronate-, MLVc- or SUVc-treated rats, where a P value of
