Alison M. Bendele

Arthritis induced by interleukin-1 is dependent on the site and frequency of intraarticular injection

Intraarticular injection of recombinant human interleukin-1 (IL-1) in rats resulted in varying degrees of inflammatory changes depending on the site and frequency of injections. (i) Much lower amounts of IL-1 were required to elicit an inflammatory response in the ankle joints (15-3000 ng) than the knee joints (90-150 micrograms). (ii) The inflammatory response was much greater if IL-1 was administered in multiple doses as compared to a single dose injection. One day after a single injection of IL-1 (90-150 micrograms), knee joints exhibited a mild increase in volume as a consequence of edema, but at the end of 1 week, no discernible change in volume was observed. However, when the same total amount of IL-1 was injected in three doses, there was a dramatic increase in joint volume at the end of 1 week that persisted for at least 3 weeks. The increase was dose dependent. (iii) The inflammatory response was dependent on the age/weight of the rats: the older the animals the greater the response. (iv) Under conditions where IL-1 induced inflammatory changes in knee joints, recombinant tumor necrosis factor failed to induce any significant response. (v) Histological examination of the knee joints revealed distinct differences in the pathological response to the two different protocols of IL-1 administration in the knee joints. The animals injected with a single dose of IL-1 showed a mild and transient inflammation that was resolved by 2 weeks postinjection, but exhibited degenerative changes associated with focal loss of chondrocytes and proteoglycan of the knee joint cartilage, which became progressively severe. The knee joints of animals given three injections of IL-1 showed evidence of marked acute synovitis, fibroplasia, loss of proteoglycan and chondrocytes, resorption of subchondral bone, and transition of hematopoeitic marrow cells into cells of mesenchymal morphology. (vi) Examination of proteoglycan synthesis by cartilage of IL-1-injected rats revealed that within 1 day after injection, a dramatic reduction in synthesis occurred which persisted for at least 2 weeks. These studies suggest that intraarticular injection of IL-1 provides a useful rodent model for the investigation of pathological changes occurring within a localized joint as a result of acute and chronic inflammatory stimuli. Relevant aspects of the pathology of joint erosion can be demonstrated depending on the frequency of IL-1 injection.

Passive Role of Articular Chondrocytes in Quinolone-Induced Arthropathy in Guinea Pigs

The role of articular chondrocytes and matrix degrading enzymes such as collagenase and neutral protease in the pathogenesis of quinolone-induced cartilage degeneration was investigated in immature guinea pigs. Articular cartilage from nalidixic acid (NA) treated guinea pigs was examined for the presence of protease activity or the ex vivo synthesis of collagenase at various times post-treatment. Histologic evaluation of knee joints confirmed the presence of degenerative changes in the matrix, but increased collagenase synthesis or protease activity were not detected. A separate group of animals was used to determine the importance of articular chondrocytes in the lesion generation. These cells were killed by intra-articular injection of the glycolysis inhibitor iodoacetic acid (IA) prior to treatment of the animals with NA. Typical “blister-like” lesions developed in cartilage devoid of viable chondrocytes at the time of exposure to NA indicating that their presence was not required for lesion development. Cartilage exposed to IA only did not exhibit “blister-like” lesions indicating that chondrocyte death and proteoglycan loss in conjunction with frictional forces associated with load-bearing were not sufficient to induce major matrical disruptive changes during the period of this study. These results indicate that articular chondrocytes are not actively involved in inducing the degenerative changes and provide no evidence for involvement of collagenase or neutral protease in the pathogenesis of the lesion.

Effect of Weight Manipulation on Bone Loss Due to Ovariectomy and the Protective Effects of Estrogen in the Rat

SummaryWhile characterizing the effects of estrogen on an ovariectomized (OVX) rat model of bone loss, we examined several weight-matching regimens e.g.,ad libitum (feed bins continually full), weight matched (rate of weight gain for OVX and Sham-OVX groups was equalized), and weight restricted (weight gain rates for all groups were equalized to that of estrogen-treated OVX rats) for possible effects. Bone loss following ovariectomy is primarily the result of an increase in bone resorption and is extremely sensitive to the effects of estrogens. Thus, in all of our analyses, treatment with 170-estradiol served as a positive control for the prevention of bone loss. Each weight-matching study had three groups: control (Sham-OVX), OVX, and OVX + 170 estradiol (0.1 mg/kg/day), and lasted for either 2, 4, or 6 weeks. Throughout the study, each Sprague Dawley rat was weighed every other day, and following sacrifice, a femur was removed for bone mineral density (BMD) analysis at the distal metaphysis by single photon absorptiometry. Following 2 weeks of dietary modifications, no significant differences were detected in BMD among thead lib or weight matched groups. However, an estradiol-preventable reduction in BMD in restricted OVX rats was detected at 2 weeks postovariectomy. Additionally, OVX rats in all three dietary regimens displayed an estrogen-preventable reduction in proximal femur BMD at 4 and 6 weeks postovariectomy. These results indicate that a 4-week rat ovariectomized model of bone loss, under conditions of ad libitum feeding, shows great potential for pharmacologic manipulation.

Interleukin 1 enhances the development of spontaneous arthritis in MRL/lpr mice.

We previously reported that treatments with human recombinant interleukin-1 beta (rIL-1 beta) in DBA/1 mice which were suboptimally immunized with native chick type II collagen (NcII) markedly accelerated the onset of collagen-induced arthritis (CIA). In the present study, we further characterized this IL-1-mediated enhancement of murine arthritis by examining the in vivo effects of rIL-1 beta in another arthritis model, namely, the spontaneous arthritis of the MRL/lpr mouse strain. The results of these studies demonstrated that IL-1 treatments also enhanced the onset and progression of the spontaneous arthritic disease in MRL/lpr mice. A substantial proportion of the IL-1-treated MRL/lpr mice that were between 3 and 3.5 months of age exhibited swelling in either the hind or front paws. Moreover, histopathologic studies demonstrated the presence of striking alterations within the various joints of these IL-1-treated MRL/lpr mice. Such abnormalities were not detected in the non-IL-1-treated, age-matched MRL/lpr mice. Therefore, as in the experimentally induced disease of CIA, IL-1 may also be capable of contributing to the pathogenesis of the spontaneous arthritis of MRL/lpr mice.

Interleukin 1 mediated acceleration of type II collagen-induced arthritis: effects of anti-inflammatory or anti-arthritic drugs.

We previously demonstrated that treatments with rIL-1B accelerated the onset and progression of CIA in mice. In the present study, it was observed that IL-1 also enhanced the development of CIA in rats. Like the mouse model, maximal incidence (80–100%) of arthritis occurred within 7 days after the first treatment with IL-1 in rats. Thus, the acceleration of CIA by IL-1 (IL-1 CIA) may be an improved model for the rapid screening of anti-inflammatory and/or anti-arthritic drugs. As a first step to determining the utility of the IL-1 CIA model as a drug screen, we examined the ability of various known anti-inflammatory and anti-arthritic drugs to modify the IL-1 mediated enhancement of CIA in both rats and mice. The results of these studies showed that when analyzed in the IL-1 CIA model, rats and mice exhibited differences in their responses to several of these drugs. For example, dexamethasone, cyclophosphamide, azathioprine, various non-steroidal anti-inflammatory drugs (NSAIDs) as well as methotrexate were found active in the IL-1 CIA of rats. By contrast, the NSAIDs were found to be less effective in suppressing the IL-1 accelerated disease in mice. In both rats and mice, cyclosporine A and several disease modifying anti-arthritic drugs failed to the prevent the development of CIA that was potentiated by IL-1. Thus, in the IL-1 CIA model NSAIDs appeared to be less active in mice than rats. In conclusion, because of the shorter latent period required for the development of arthritis in the IL-1 treated animals, the IL-1 accelerated CIA model in both mice and rats may be useful for screening anti-inflammatory or anti-arthritic compounds.

Development of a rapid screen for detecting and differentiating immunomodulatory vs. anti-inflammatory compounds in rats.

Polyarthritis can be induced in rats using a synthetic adjuvant, N,N-dioctyldecyl-N′, N-bis(2-hydroxyethyl) propanediamine (LA) suspended in oil. The disease is morphologically indistinguishable from the classic adjuvant arthritis induced by Freunds complete adjuvant (FCA). LA injection (7.5 mg/animal) consistently induced paw swelling, splenomegaly and fibrinogen level increases at certain time points. Studies evaluating various protocols and parameters determined that a 15 day assay where agents administered from days 9 through 13, would differentiate immunomodulatory and anti-inflammatory compounds. Parameters utilized were body weight, paw volumes, spleen weights, and fibrinogen levels. Immunomodulatory agents reduce paw swelling, splenomegaly and in some cases fibrinogen levels. NSAIDS reduce paw swelling,increase splenomegaly and have no effect on fibrinogen levels. These results indicate that compounds active in the traditional FCA assay can be detected and differentiated with respect to anti-inflammatory vs. immunomodulatory activity in a rapid screen.

Pharmacological and histological examinations of regional differences of guinea-pig lung: a role of pleural surface smooth muscle in lung strip contraction

1 Parenchymal lung strip preparations have been widely used as an in vitro model of peripheral airway smooth muscle. The present study examined functional responses of 4 consecutive guinea‐pig lung parenchymal strips isolated from the central region (segment 1) to the distal edge (segment 4) of the lower lung lobe. The middle two segments were designated as segments 2 and 3. 2 Lung segments 1 and 4 exhibited significantly greater contraction than the other 2 segments to KCl when responses were expressed as mg force per mg tissue weight. Contractile responses to bronchospastic agents including histamine, carbachol, endothelin‐1, leukotrienes (LT) B4 and D4, and the thromboxane A2‐mimetic U46619 demonstrated no significant difference in EC50 values among the 4 lung segments. 3 Contractile responses of segments 1 and 4 to antigen‐challenge (ovalbumin), ionophore A23187 and substance P were significantly greater than the other 2 segments with respect to either sensitivity or maximum responsiveness. 4 U46619‐induced contractions of the 4 lung segments were relaxed in similar manner by papaverine and theophylline up to 100%, salbutamol up to 80%, and sodium nitroprusside by only 20%. In contrast, sodium nitroprusside markedly reversed U46619‐induced contraction of pulmonary arterial rings and bronchial rings. 5 Histological studies identified 2–4 layers of smooth muscle cells underlying the lung pleural surface. Mast cells were prominent in this area. Moreover, morphometric studies showed that segment 4 possessed the least amount of smooth muscle structures from bronchial/bronchiolar wall and vasculatures as compared to the other 3 segments, and a significant difference in this respect was evident between segment 1 and segment 4. 6 Since lung segments 1 and 4 are covered with larger surface area of lung pleura, the present results suggest that the significantly greater intrinsic contractile responses of segments 1 and 4 are associated with the presence of increased lung pleural surface possibly together with more mast cells. Thus, a primary contribution to the net contraction of the lung parenchymal strips may be smooth muscle from the lung pleura, alveolar ducts and interstitial contractile cells rather than from bronchi/bronchioles and micro‐vasculatures.

Effects of anti-arthritic drugs on IL-1 induced inflammation in rats

Intra-articular injection of IL-1 in rats results in profound changes in the synovial joints including edema, synovitis, cartilage degeneration and fibroblast proliferation. We have tested the efficacy of several anti-arthritic drugs in this model to characterize the mechanism of IL-1 induced inflammatory lesions. The response to IL-1 was evaluated by measurement of soft-tissue swelling and by histological scoring. Dexamethasone, and several NSAIDS were effective in reducing the soft tissue welling but only some were effective in improving the histological lesions. The slow-acting antirheumaticd-penicillamine and immunosuppressive agents such as cyclosporine A were ineffective. These observations provide further evidence thatin vivo, at least some aspects of IL-1 induced changes in rat knee joints are likely to be prostaglandin mediated.

Hepatocellular proliferation in ibuprofen-treated mice

Female B6C3F1 mice were treated with ibuprofen for 2 wk or 90 days to monitor effects on hepatocellular proliferation during acute and subchronic exposure. Proliferation was assessed by bromodeoxyuridine labeling. Mice treated with 100, 200, or 400 mg/kg ibuprofen for 2 wk had significantly increased liver weights. A dose-related increase in the number of labeled hepatocytes per 1,000 hepatocytes suggested that the weight increases were in part a result of hepatocellular proliferation. Hepatocellular hypertrophy also contributed to the increased liver size as indicated by decreases in the number of hepatocytes per high power field. Ultrastructural evaluation indicated that hepatocyte peroxisome size increased significantly in treated mice. Mice treated with 100 or 200 mg/kg ibuprofen for 90 days and given bromodeoxyuridine during the last 2 wk of ibuprofen exposure had statistically significant increases in relative liver weights. However, the number of labeled hepatocytes per 1,000 hepatocytes was not increased, and there was no evidence of hepatocellular hypertrophy. Mice given 200 mg/kg ibuprofen for 90 days had significantly decreased serum triglycerides. These findings indicate that ibuprofen treatment of mice results in hepatomegaly characterized by hepatocellular hypertrophy and hyperplasia. Peroxisomal changes may be contributory to the pathogenesis of this lesion.

Effects of Various Anti-T Cell Receptor Antibodies on the Development of Type II Collagen-Induced Arthritis in Mice

Recent genetic studies of David and coworkers suggest that subsets of T cells utilizing specific V beta TcR genes may play important roles in the susceptibility to collagen-induced arthritis (CIA). Hence, in vivo depletion of such T cell subsets may significantly affect the development of CIA. To address this possibility, we first examined the effects of in vivo treatments with various monoclonal antibodies (mAbs) that are specific for particular TcR V beta families on the induction of CIA. Results presented in this study demonstrated that treatments with either anti-V beta 6, anti-V beta 8 or anti-V beta 11 did not suppress the development of arthritis in collagen-immunized mice. While combined treatments with these V beta specific mAbs which resulted in the in vivo elimination of V beta 6+, V beta 8+ and V beta 11+ T cells were not very effective in preventing the onset of CIA, the severity of the arthritic disease was somewhat reduced in animals that had received the triad of anti-V beta mAbs. By contrast, depletion of T cells expressing the alpha beta receptors by in vivo treatments with a pan anti-alpha beta mAb significantly decreased the incidence of CIA. Therefore, although an effect on the development of CIA was achieved by in vivo treatments with a mAb that detects all alpha beta + T cells, the elimination of only a few subsets of T cells which included the V beta 6+, V beta 8+, and V beta 11+ cells did not profoundly alter the incidence of CIA.(ABSTRACT TRUNCATED AT 250 WORDS)