The generation and characterization of Tg197 human TNF transgenic mice have been previously described 20. Tg197 mice generated on CBA × C57BL/6 genetic backgrounds and littermate controls were bred and maintained at the animal facilities of the Biomedical Sciences Research Center, Alexander Fleming, Athens, Greece, under specific-pathogen-free conditions. All of the Tg197 mice typically developed polyarthritis 3–4 weeks after birth, whereas nontransgenic (wild-type) mice remained normal. Mice were given conventional oral food and water ad libitum. All procedures involving animals were in compliance with the Declaration of Helsinki principles.

A total of 44 weight-matched mice (34 Tg197 and 10 nontransgenic wild-type littermates) were divided into six groups for subsequent gross observations and histopathologic analyses – untreated Tg197 group (N = 10), P-NT.II-treated Tg197 group (N = 18), scrambled-P-NT.II-treated Tg197 group (N = 6), P-NT.II-treated wild-type group (N = 4), scrambled-P-NT.II-treated wild-type group (N = 4), and Tg197 baseline group – just before the treatment at 4 weeks of age (N = 4). Nontransgenic mice were given the same dose of P-NT.II or scrambled P-NT.II, and the same regimen of treatment, as the Tg197 mice.

P-NT.II (test peptide) and the scrambled P-NT.II (negative control peptide) were synthesized using the solid-phase method with 9-fluorenylmethoxy carbonyl chemistry and were purified and validated as described elsewhere 18. They were stored lyophilized at -20°C in sealed tubes and were dissolved freshly before use in 0.1% dimethyl sulfoxide (DMSO). Each Tg197 or wild-type mouse was given intraperitoneal injections of P-NT.II or the scrambled P-NT.II (7.5 mg/kg) in 50 μl of vehicle (0.1% final DMSO concentration), three times a week for 4 weeks (i.e. from age 4–8 weeks).

This was done by gross observations based on body-weight measurements and arthritis scoring, which were done twice weekly from 4 weeks (baseline) to 8 weeks of age (end of the study), after which all the animals were killed by CO₂ inhalation. The level of severity of clinical arthritis was evaluated based on an arthritis score (AS) taken on both ankle joints. Average scores on a scale of 0–3 were used; 1 = mild arthritis (joint swelling); 2 = moderate arthritis (severe joint swelling and deformation, no grip strength); 3 = severe arthritis (ankylosis detected on flexion, and severely impaired movement) 21.

The whole ankle joints harvested from the right side of each mouse were fixed overnight in 10% formalin, decalcified in 30% citrate-buffered formic acid for 3 days at 4°C, dehydrated in a graded series of methanol and xylene, and then embedded in paraffin. Thin sections (6 μm thick) were stained with hematoxylin and eosin, and histopathologic scorings performed under the light microscope (Leitz Aristoplan) by a blinded observer. The histopathologic score (HS) was evaluated 21 using a scale of severity ranging from 1 to 4, where 1 = hyperplasia of the synovial membrane and presence of polymorphonuclear infiltrates, 2 = pannus and fibrous tissue formation and focal subchondral bone erosion, 3 = articular cartilage destruction and bone erosion, and 4 = extensive articular cartilage destruction and bone erosion.

Arbitrary scores were used to assess the extent of synovitis, cartilage destruction, and bone erosion. Semiquantitative scores from 0 to 4 were used for each histopathologic parameter 22. Synovitis: 0 = normal; 1 = mild synovial hypertrophy (<5 cell layers) with few inflammatory cells; 2 = moderate synovial hypertrophy (<20 cell layers) with accumulation of inflammatory cells into intrasynovial cysts; 3 = pannus and fibrous tissue formation; and 4 = pannus and fibrous tissue formation on both sides of the ankle joint. Cartilage damage: 0 = intact; 1 = minor (<10%); 2 = moderate (10–50%); 3 = high (50–80%); and 4 = severe (80–100%). Bone erosions: 0 = normal; 1 = mild (focal subchondral erosion); 2 = moderate (multiple subchondral erosions); 3 = high (as above + focal erosion of talus); and 4 = maximum (multiple erosions of tarsal and metatarsal bones).

Ankle joints dissected from the left hind leg of each mouse were split open longitudinally through the midline between the tibia and the talus, prefixed overnight with 2.5% glutaraldehyde in phosphate buffer, pH 7.4, and rinsed with the buffer. After they had been postfixed with 1% osmium tetroxide in phosphate buffer for 2 hours, they were dehydrated in a graded series of ethanol and embedded in epoxy resin (Araldite). Semithin sections (1.0 μm) were cut and stained with methylene blue to reveal their orientation for ultrathin sectioning and for histopathologic scoring under the light microscope. Ultrathin sections (80–90 nm) were then cut with an ultramicrotome (Ultracut E; Riechert-Jung, Leica, Vienna, Austria), mounted on copper grids, counterstained with uranyl acetate and lead citrate, and evaluated in the electron microscope (CM120 Biotwin; FEI Company, Electron Optics, Eindhoven, The Netherlands).

sPLA₂ was measured in the serum of transgenic (Tg197) mice and nontransgenic wild-type controls, using an Escherichia coli membrane assay as described previously 18. In brief, [³H]arachidonate-labeled E. coli membrane suspension (5.8 μCi/μmol, PerkinElmer Life Sciences, Inc, Boston, MA, USA) was used as substrate, and 25 mM CaCl₂-100 mM Tris/HCl (pH 7.5) as assay buffer. The reaction mixture, containing substrate (20 μl) and either purified human synovial sPLA₂ standard (1–80 ng/ml; Cayman Chemical Company, Ann Arbor, MI, USA) or serum (10 μl), in a final volume of 250 μl in assay buffer, was incubated at 37°C for 1 hour, and the reaction was terminated with 750 μl of chilled phosphate-buffered saline containing 1% bovine serum albumin. Aliquots (500 μl) of the supernatant were then taken, for measurement of the amount of [³H]arachidonate released from the E. coli membrane using liquid scintillation counting (LS 6500 Scintillation Counter; Beckman Inc., Fullerton, CA, USA). The amount of sPLA₂ present in the serum was calculated from the standard curve and is expressed as ng/ml ± SEM.

The murine macrophage cell line J774 (American Type Culture Collection, Manassas, VA, USA) was cultured at 37°C in humidified 5% CO₂/95% air in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, 2 mM glutamine, 20 mM HEPES, 100 IU/ml penicillin, and 100 μg/ml streptomycin. After growing to confluence, the cells were dislodged by scraping, plated in 12 culture wells at a density of 5 × 10⁵ cells/ml per well, and allowed to adhere for 2 hours. Thereafter, the medium was replaced with fresh medium containing lipopolysaccharide (LPS) (2 μg/ml) and one of the PLA₂ inhibitors (P-NT.II, scrambled P-NT.II, or LY315920 [Lilly Research Laboratories, Indianapolis, IN, USA], dissolved in DMSO [final concentration 0.1% v/v]). Peptides were tested at various concentrations ranging from 0.01 to 40 μM. After incubation in 5% CO₂/95% air at 37°C for 20 hours, culture medium supernatants were collected and stored frozen (-80°C) until use. In parallel experiments, cells were stimulated with mouse recombinant TNF (10 ng/ml; Sigma, St. Louis, MO, USA) for 20 hours, in the presence or absence of 10 μM P-NT.II or LY315920 dissolved in DMSO (0.1% final concentration). Culture medium supernatants were collected after centrifugation (10,000 g, 4°C, 15 min) and stored at -80°C prior to measurement of prostaglandin E₂ (PGE₂).

XTT (sodium 3'-[(phenyl amine carboxyl)-3,4-tetrazolium]-bis(4-methoxy-nitro) benzene sulfonic acid hydrate) Cell Proliferation Kit II (Roche Applied Science) was used to assess the possible cytoxic effect of the peptide P-NT.II on the mouse macrophage J774 cell line.

PGE₂ (EIA kit-monoclonal; Cayman) concentrations were measured in the cell-culture supernatants in accordance with the manufacturer's instructions.

Statistical analyses were performed using GraphPad Prism software to calculate the means and SEMs. Group means were compared by using one-way analysis of variance (ANOVA), followed by Bonferroni's multiple comparison post test to identify statistically significant differences (i.e. P < 0.05).

Figure 1 shows the histologic features of ankle joints of Tg197 mice in the untreated, P-NT.II-treated, and scrambled-P-NT.II-treated groups. Gross histologic findings of the three experimental groups are summarized in Table 1. At 8 weeks of age, ankle joints from the untreated Tg197 group were moderately (90% with HS 3) to severely (10% with HS 4) damaged, with pannus and fibrous tissue formation and focal subchondral bone erosion. Articular cartilage destruction and bone erosion were observed in 90% of those joints (Fig. 1a,1b). In contrast, all the articular cartilage surfaces and associated synovial linings of the ankle joints of the P-NT.II-treated group at 2 weeks post-treatment (i.e. age 6 weeks) were only mildly affected (HS 2), with no evidence of cartilage or bone erosion (Fig. 1c), and 25% of joints were affected moderately (HS 3) at 4 weeks post-treatment (i.e. age 8 weeks) (Fig. 1d). In contrast, 83.3% of joints of scrambled-P-NT.II-treated Tg197 mice at 8 weeks of age were moderately damaged (HS 3) (Fig. 1e,1f), with histologic features similar to those of the untreated Tg197 mice. Although the disease, as assessed by the HS, was significantly lower in the P-NT.II-treated group than in the untreated or scrambled-P-NT.II-treated groups, visual disease scores (ASs) did not correlate well with the HS. In contrast to HSs, ASs of mice treated with P-NT.II did not significantly differ from those of the untreated or scrambled-P-NT.II-treated group (Fig. 2).

To assess specific effects of the peptide P-NT.II on synovitis, cartilage destruction, and bone erosion, we conducted a semiquantitative scoring analysis for each of these pathologic parameters. P-NT.II treatment in Tg197 mice resulted in a significant reduction (P < 0.05) in all three analytical HSs as compared with those of untreated or scrambled-P-NT.II-treated Tg197 mice, which all developed synovitis with severe articular cartilage degradation and bone erosions (Fig. 3). Statistical analysis revealed a greater beneficial effect of P-NT.II on cartilage destruction and bone erosion (**P < 0.01 versus untreated or scrambled-P-NT.II-treated groups for both parameters) than on synovitis (*P < 0.05 versus untreated or scrambled-P-NT.II-treated groups).

Articular cartilage in the ankle joints of all untreated Tg197 mice was generally damaged at 8 weeks of age (Fig. 4c,4d,4e,4f) as compared with normal morphology seen in control, wild-type mice (Fig. 4a). No significant ultrastructural changes in the nucleus and plasma membrane were noted at the cellular level in the articular cartilage of untreated Tg197 mice at age 4 weeks (baseline) except for some minor changes including vacuoles, dilated cisternae, and the presence of granular materials seen inside the cytoplasm (Fig. 4b). In the 8-week-old mice, the chondrocytes on the surface of the superficial cartilage layer had become necrotic, with alterations of cartilage developed in most cases (Fig. 4c,4d,4e,4f). The cell body and nucleus of some chondrocytes became large and rounded, resulting in vacuolation, and the cytoplasm was transparent, with an accumulation of intracytoplasmic filaments (Fig. 4c). Degenerating chondrocytes with greatly vacuolated cytoplasm and pyknotic nuclei (Fig. 4d), and chondrocytes with complete loss of nuclei and disrupted rough endoplasmic reticulum (r-ER) (Fig. 4e,4f), were also observed. In contrast, the ultrastructural features of chondrocytes 1–4 weeks after P-NT.II treatment (i.e. age 5–8 weeks; Fig. 5a) did not substantially differ from those seen in the joints of normal wild-type mice (Fig. 4a). Most of them had a prominent nucleus, lined by plasma membrane with short cytoplasmic protrusions, and vacuoles, r-ER, and mitochondria in the cytoplasm. The ultrastructure of chondrocytes of the scrambled-P-NT.II-treated joints at 8 weeks of age (Fig. 5b) were more or less similar to those described for untreated Tg197 mice with degenerating features such as the greatly vacuolated cytoplasm and pyknotic nuclei (cf. Fig. 4d) or loss of nucleus, disrupted r-ER (cf. Fig. 4f), and swollen mitochondria with distorted cristae (cf. Fig. 4c).

The early response of the synovial membrane in the untreated Tg197 mice at age 4 weeks (baseline) was synovial hyperplasia, with the presence of type A and B synovial cells along with inflammatory cells such as lymphocytes, macrophages, and mast cells. Type A cells were similar to macrophage cells and were characterized by many vesicles, vacuoles, and a higher number of cell processes. Type B cells were similar to fibroblast cells and contained small vesicles and r-ER. The later response (at ≥ 5 weeks of age) included degeneration of synovial cells, with swollen mitochondria and cell fragmentations. In areas of high inflammation, the synovial tissue (mostly type A cells) had proliferated into the articular cavity (Fig. 6a). Type A and B cells in the synovium were no longer distinguishable at age 6 weeks and thereafter. The synovial membrane was lined by closely packed elongated synoviocytes which were sealed by junctional systems of the adherent type (Fig. 6b). Large amounts of fibrin deposition on the synovial surface could be seen, and the two opposing, flattened synoviocytes with fibrin between them indicated the existence of synovial adhesion (Fig. 6d). Also, degenerating synoviocytes with disintegrating nuclei and vacuolated cytoplasm were randomly seen in the synovium (Fig. 6c). Synoviocytes appeared flattened, and partially degranulated mast cells were seen under the synovium (Fig. 6e).

P-NT.II treatment tended to decrease the number of inflammatory cells, with less degeneration of synovial cells and cell fragmentation seen in the joints of the treated group (Fig. 7b). The peptide P-NT.II retained at least the basic structural organization of the synovial membrane seen in the control wild-type mice (Fig. 7a), while the synoviocytes from mice treated with scrambled P-NT.II (Fig. 7c) were structurally indistinguishable from those seen in untreated joints (cf. Fig. 6). In those joints, the synovial cells were seen lining up close together, and many cell fragments resulting from degenerating cells were present in the synovium, along with infiltrating mast cells (Fig. 7c).

In a time-course study to evaluate the specific effect of peptide in modulating the serum sPLA₂ levels in Tg197 mice (Fig. 8), P-NT.II significantly suppressed the circulating sPLA₂ in the mice at age 8 weeks (P < 0.05), by comparison with the serum levels of the untreated mice of same age. In contrast, the circulating sPLA₂ of scrambled-P-NT.II-treated and untreated Tg197 mice at age 8 weeks were not significantly different (P > 0.05), thus indicating the specific effect of the peptide P-NT.II on sPLA₂ levels.

The suppressive effect of P-NT.II and sPLA₂-selective inhibitor LY315920 (Lilly) on LPS- and TNF-stimulated PGE₂ production was examined in mouse macrophage cell cultures (Fig. 9). Production of PGE₂ in the medium increased approximately sixfold from the basal level of 55 ± 6 pg/ml to 320 ± 35 and 330 ± 11 pg/ml (mean ± SD, N = 5), after 20 hours' stimulation of cultured cells with LPS (2 μg/ml) (Fig. 9a) or TNF (10 ng/ml) (Fig. 9b), respectively. When the inhibitors were coincubated with either LPS- or TNF-stimulated macrophages in the medium, both P-NT.II and LY315920 (final concentration 10 μM) dose-dependently inhibited PGE₂ production, with estimated IC₅₀ values of 25 and 30 μM, respectively. In contrast, scrambled P-NT.II (negative control) showed no inhibitory effect on either LPS- or TNF-induced PGE₂ release in the culture medium. Neither the peptide nor LY315920 affected the cellular viability, when tested by XTT assay kit at the highest concentration (40 μM) used in culture experiments (E_{492 nm} values of 0.89 ± 0.02, 0.84 ± 0. 021, and 0.92 ± 0.019 for untreated, P-NT.II-treated, and LY315920-treated cells, respectively).