Introduction
Therapeutic strategies targeting cellular components of the inflamed synovial tissue and blockage of specific inflammatory mediators have been shown to be efficacious in ameliorating inflammation and inhibiting joint destruction in patients with rheumatoid arthritis (RA) [
1,
2]. From a pharmacodynamic perspective, the treatment efficacy mainly depends upon two factors: the specificity of the drug to its molecular target, and the local concentration of the drug where it interacts with its putative target. While most of the efforts to improve RA treatment have been focused on the discovery of agents that target specific molecules or pathways and more potent therapeutic agents, the approach to manipulate the local drug concentration in the synovium and consequently potentiate the efficacy of a particular therapy has been very limited [
3‐
5].
In the past decade, several groups have developed liposomal and protein-based formulations to facilitate the targeting of drugs to arthritic joints [
6‐
9]. Recently, our team has developed the acid-labile arthrotropic macromolecular dexamethasone prodrug based on
N-(2-hydroxypropyl)methacrylamide (HPMA) copolymer (P-Dex) [
10] and has provided preliminary evidence demonstrating its superior anti-inflammatory efficacy compared with an equivalent dose of free dexamethasone (Dex) [
11‐
13]. The present studies were undertaken to delineate the mechanisms involved in the arthrotropism and joint retention of P-Dex, and its capacity to produce sustained amelioration of inflammatory arthritis.
Materials and methods
Treatment of adjuvant-induced arthritis rats with P-Dex
Adjuvant-induced arthritis (AA) rats were induced as described previously [
11]. On day 14 post induction, the rats were randomized into five groups (eight rats/group): P-Dex half-dose (equivalent Dex dose = 5 mg/kg, single intravenous (i.v.) injection), P-Dex (equivalent Dex dose = 10 mg/kg, single i.v. injection) [
13], free Dex (total dose = 10 mg/kg, four intraperitoneal injections, days 14 to 17), saline (single i.v. injection) and HPMA polymer without Dex (PHPMA, single i.v. injection; amount of polymer used is equivalent to P-Dex).
Arthritis flare of the P-Dex group was set as the experimental endpoint, at which time the hind limbs were isolated for bone mineral density (BMD) and histology evaluations. The BMD was measured from the distal tibia to the phalanges of the paw using a pDEXA® Sabre™ X-ray bone densitometer (Norland Medical System, Inc., Fort Atkinson, WI, USA).
All animal experiments were performed according to a protocol approved by the University of Nebraska Medical Center Institutional Animal Care and Use Committee.
Clinical measurements
The articular index score was measured during the treatment as described previously [
13]. The scores were applied to each hind limb by the same observer (LDQ) from day 8 to day 34 post-arthritis induction, and the sum of the two hind limb scores for each animal were recorded. The ankle diameter (medial to lateral) was measured using a digital caliper (World Prescision Instruments, Inc., Saraspta, FL, USA).
Histological analysis
The isolated hind limbs were fixed with buffered formalin and were decalcified. Thin sections (5 μm) were cut approximately 200 μm apart and were H & E stained. The joints were histologically graded by a pathologist (SML), who was blind to the treatment groups, using a scoring system adapted from previous work [
13]. Each histopathologic feature was graded as follows: synovial cell lining hyperplasia (0 to 2); pannus formation (0 to 3); mononuclear cell infiltration (0 to 3); polymorphonuclear leukocytes infiltration in periarticular soft tissue (0 to 3); cellular infiltration and bone erosion at distal tibia (0 to 3); and cellular infiltration of cartilage (0 to 2). The score for every histopathologic feature was summed for each animal.
Quantitative analysis of joint vascular leakage using magnetic resonance imaging
High-resolution T
1-weighted magnetic resonance imaging (MRI) scans and T
1 maps were acquired before and after (every 10 minutes for 4 hours) the injection of DOTA-Gd
3+-labeled HPMA copolymer (P-DOTA-Gd
3+) [
11]. All MRI scans were performed on a Bruker Avance MRI and spectroscopy system (7T/21 cm; Bruker, Karlsruhe, Germany). T
1 maps were acquired using a Look-Locker technique [
14]. Patlak plots were used to estimate the tissue transfer constant of P-DOTA-Gd
3+ (
K
i
). T
1 and
K
i
maps were reconstructed using the neuropipes software suite (a software suite written in 'C' for the apodization, reconstruction, and curve-fitting of MRI images. Developed by Dr. James Ewing at Henry Ford Hospital, Detroit, MI, USA). Both unidirectional and bidirectional transfer models were used in the analyses. F-tests were performed between the two models, and each model was tested against no leakage on a pixel-by-pixel basis [
15].
Biodistribution of 125I -labeled P-Dex
Tyrosine amide-containing P-Dex (P-Dex-Tyr-NH
2) was iodinated using a standard chloramine T method [
16]. For the biodistribution study,
125I-labeled P-Dex-Tyr-NH
2 (mixed with P-Dex-Tyr-NH
2 without
125I, Dex equivalent dose = 5 mg/kg) was administered to AA rats and healthy rats (six rats/group) via tail vein injection. The animals were euthanized 24 hours post administration. Major organs and tissues were collected and evaluated with a γ-counter (Minaxi Auto-gamma 5000 series; Packard Instrument Company, Meriden, CT, USA).
Immunohistochemical analysis of arthritic joints
In the MRI studies, HPMA copolymer labeled with Alexa Fluor® 488 (P-Alexa) was given to AA rats simultaneously with P-DOTA-Gd3+. At 24 hours post injection, hind limbs were isolated and fixed with 0.5% paraformaldehyde in PBS and were decalcified with 10% ethylenediamine tetraacetic acid. Tissues were paraffin embedded and sections corresponding to the MRI hot-spots were selected for immunohistochemical analysis. After deparaffinization, the sections were incubated for 30 minutes with citrate buffer (10 mM, pH = 6.0) at 95°C, followed by incubation with 10% goat serum/PBS for 20 minutes at room temperature. After addition of the primary antibodies (CD68 (10 μg/ml) or prolyl 4-hydroxylase (10 μg/ml), diluted in 10% goat serum/PBS), the sections were incubated overnight at 4°C in a humidified chamber. After washing with PBS (three times), diluted phycoerythrin-labeled rabbit anti-mouse IgG secondary antibody (5 μg/ml) was added and incubated for 30 minutes in the dark at room temperature. In control experiments, primary antibodies were replaced by PBS and the samples were processed as described above. The processed tissue sections were then evaluated with confocal microscopy.
Fluorescence-activated cell sorting analysis of cells isolated from synovial tissue
At 14 days post induction, P-Alexa was given to AA rats by tail vein injection. At 24 hours post injection, ankle joint-associated soft tissues were surgically removed with a scalpel and minced aseptically. The tissues were further digested with collagenase type I (1 mg/ml; Sigma-Aldrich, St Louis, MO, USA) at 37°C for 30 minutes. After passing through a 70 μm cell strainer, a single-cell suspension (2 × 106/ml) was obtained. ACK Lysing Buffer (Quality Biological, Gaithersburg, MD, USA) was then used to remove blood cells.
For FACS evaluation, the cells were first incubated respectively with three primary antibodies - CD68 (1:100 dilution; AbD Serotec, Raleigh, NC, USA), CD11c (1:10 dilution; abcam Inc., Cambridge, MA, USA) and prolyl-4-hydroxylase (1:50 dilution; Acris Antibodies, Herford, Germany) - each for 30 minutes on ice. The cells were then incubated with phycoerythrin-labeled rabbit anti-mouse IgG secondary antibody (1:100 dilution; BD Biosciences, San Jose, CA, USA) for another 30 minutes on ice. Isotype-matched mouse IgG1 and mouse IgG2a (BD Biosciences) were used as negative controls. Following the final wash, the cells were analyzed with flow cytometry.
Macrophage and fibroblast cultures
CD14-positive monocytes were prepared from peripheral blood mononuclear cells derived from de-identified normal human donors as described previously [
17]. Cells were cultured for 24 hours at a cell density of 10
6/ml in α-MEM medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% FBS (VWR, West Chester, PA, USA) and 1% antibiotic/antimycotic (Invitrogen) in the presence of 10 ng/ml human macrophage colony-stimulating factor (Peprotech, Rocky Hill, NJ, USA) in 24-well tissue culture plates (1 ml/well). Cells were then pulsed for 4 hours with P-Dex (5 μM) or Dex (0.6 μM), washed and replenished with fresh medium. The two treatments had equivalent doses of Dex.
After 24 hours, cells were either analyzed by confocal microscopy and flow cytometry or challenged with inflammatory mediators. For confocal microscopy, the lysosomal and nuclear compartments were stained with LysoTracker Red DND-99 and Hoechst 33342 (Invitrogen), respectively. For flow cytometric analysis of apoptosis, cells were analyzed using the Vybrant Apoptosis Assay Kit #3 (Invitrogen) in accordance with the manufacturer's recommendations. For inflammatory challenge, cells were treated with 40 pg/ml lipopolysaccharide (LPS) for 6 hours, after which conditioned media were analyzed for inflammatory cytokine production (that is, TNFα and IL-6) by ELISA (BD Bioscience, San Diego, CA, USA).
Fibroblasts were isolated from periprosthetic tissues retrieved from patients undergoing total hip replacement revision surgery and from synovial membranes of RA patients. After digestion with collagenase (5 mg/ml), fibroblasts were obtained by passing through a 70 μm cell strainer. The cells were subsequently cultured at a density of 5 × 104/ml in α-MEM medium supplemented with 10% FBS and 1% antibiotic/antimycotic in 24-well tissue culture plates (1 ml/well).
Treatment with Dex or P-Dex and confocal microscopy were performed as described above for monocytes. For inflammatory challenge, fibroblasts were incubated for 24 hours with 100 ng/ml human TNFα (Peprotech), following which conditioned media were analyzed for prostaglandin E2 by ELISA (BD Bioscience) and total cellular RNA was prepared (RNeasy Mini Kit, QIAGEN, Inc., Valencia, CA, USA) and analyzed by real-time RT-PCR for matrix metalloproteinases (MMP1 and MMP3) expression.
All human cells experiments were approved by the Institutional Review Board of Hospital for Special Surgery.
Statistical methods
One-way analysis of variance was used in the data analysis, followed by a post hoc test (Student-Newman-Keuls) for multiple comparisons using Instant Biostatistics (GraphPad Software, La Jolla, CA, USA). P < 0.05 was considered statistically significant.
Discussion
In preliminary studies, we reported that a single dose of P-Dex (synthesized by either polymer analogues reaction or Dex-containing monomer copolymerization) could result in more potent resolution of joint inflammation compared with free Dex [
12,
13]. In the current investigation we have extended these observations and established the full time course of sustained anti-inflammatory efficacy for P-Dex (> 20 days), which is accompanied by protection from both bone and cartilage destruction. In addition, we demonstrated the intracellular localization of P-Dex in synovial cells after systemic delivery and established its efficacy in inhibiting proinflammatory cytokine release.
The rationale for the use of the HPMA copolymer as a drug-targeting vehicle is based on observations indicating that synovial inflammation is associated with enhanced vascular permeability to macromolecules [
18]. Our MRI data using P-DOTA-Gd
3+ quantitatively confirmed the enhanced vascular permeability of the synovial tissues to the HPMA copolymers and provides an explanation for its unique arthrotropism. Of interest, the tissues with the highest
K
i
values (hot-spots) were in proximity to regions that histologically were associated with the most severe bone and cartilage damage. This suggests that there is a relationship between the quantitative delivery of the polymer-based prodrug and the severity of synovitis, with preferential localization of the prodrug at sites of maximal inflammation. The MRI results were confirmed by biodistribution studies using
125I-labeled HPMA copolymer-dexamethasone conjugate (P-Dex-Tye-NH
2-
125I). These results provided quantitative affirmation that the distribution of P-Dex at 24 hours post administration was four or five times higher in the AA rat ankle joints compared with healthy joints.
Previous studies have not provided an explanation for the retention of macromolecules at sites of inflammation. To address this issue, we studied the cellular localization of the prodrug within inflamed joints after systemic administration of Alexa Fluor
® 488-labeled HPMA copolymer (P-Alexa). Histologic analysis of inflamed joints revealed that P-Alexa was internalized by synoviocytes. Immunohistochemical and FACS analyses identified these cells as type A synoviocytes (macrophage-like), type B synoviocytes (fibroblast-like) and dendritic cells. The internalization of different HPMA copolymers by murine macrophages and fibroblasts has been reported previously [
19,
20]. The internalization and subcellular partitioning of P-Dex, however, have not been evaluated. Our studies utilizing FITC-labeled P-Dex and immunohistochemical staining with the lysozomal marker DND-99 confirmed the co-localization of the macromolecular prodrug in a lysosomal compartment consistent with internalization via an endocytic pathway, and flow cytometric analysis showed that P-Dex uptake did not induce apoptosis. HPMA copolymers such as P-Dex are water-soluble polymers, and thus are restricted by the lysosomal membrane from escaping the endosome/lysosome compartments once internalized by the cells. Within these subcellular vesicles, P-Dex is exposed to an acidic environment (pH ~5.5); and since the Dex is conjugated to the HPMA copolymer via an acid-labile hydrazone bond, the prodrug is subject to gradual hydrolysis and subsequent release of the active drug [
12]. The presence of intracellular prodrug activation was confirmed by the capacity of the internalized P-Dex to inhibit TNFα and IL-6 release from LPS-treated macrophages and to reduce expression of MMP1 and MMP3 in RA synovial fibroblasts. Protective effects of P-Dex treatment were maintained for at least 7 days after exposure to cells
in vitro. We speculate that the sustained therapeutic effect observed in the P-Dex treatment is due to the prolonged residence of P-Dex within the synoviocytes and their gradual low pH-triggered activation within lysosomes, followed by the release of free Dex into the cytosol.
Competing interests
As a co-inventor, DW has filed a patent application related to the content of this manuscript. All remaining authors declare that they have no competing interests.
Authors' contributions
LDQ synthesized the polymer conjugates used in the study, and performed the animal treatment, immunohistochemistry and FACS experiments. He also prepared the first draft of the manuscript. PEP carried out all cell culture studies, and participated in the data interpretation and manuscript preparation. XML synthesized all the monomers and supported LDQ in the conjugate synthesis. MDB performed the MRI analysis and data interpretation. SML performed the histology evaluation. GMT designed the FACS experiments and supported LDQ in FACS data analysis. TRM participated in the data interpretation and manuscript preparation. HD designed the immunohistochemistry experiments and supported LDQ in the data analysis. SRG participated in the general design of the experiments, data interpretation and manuscript preparation. DW conceived the study, and led its design, coordination, data interpretation and manuscript preparation. All authors read and approved the final manuscript.