Ellen K. Bayne

Impaired neuropathic pain responses in mice lacking the chemokine receptor CCR2

Mice lacking the chemokine receptor chemotactic cytokine receptor 2 (CCR2) have a marked attenuation of monocyte recruitment in response to various inflammatory stimuli and a reduction of inflammatory lesions in models of demyelinating disease. In the present study, we compared nociceptive responses in inflammatory and neuropathic models of pain in CCR2 knockout and wild-type mice. In acute pain tests, responses were equivalent in CCR2 knockout and wild-type mice. In models of inflammatory pain, CCR2 knockout mice showed a 70% reduction in phase 2 of the intraplantar formalin-evoked pain response but only a modest (20–30%) and nonsignificant reduction of mechanical allodynia after intraplantar Freunds adjuvant (CFA). In a model of neuropathic pain, the development of mechanical allodynia was totally abrogated in CCR2 knockout mice. CFA administration induced marked up-regulation of CCR2 mRNA in the skin and a moderate increase in the sciatic nerve and dorsal root ganglia (DRG). In response to nerve ligation, persistent and marked up-regulation of CCR2 mRNA was evident in the nerve and DRG. Disruption of Schwann cells in response to nerve lesion resulted in infiltration of CCR2-positive monocytes/macrophages not only to the neuroma but also to the DRG. Chronic pain also resulted in the appearance of activated CCR2-positive microglia in the spinal cord. Collectively, these data suggest that the recruitment and activation of macrophages and microglia peripherally and in neural tissue may contribute to both inflammatory and neuropathic pain states. Accordingly, blockade of the CCR2 receptor may provide a novel therapeutic modality for the treatment of chronic pain.

Susceptibility of stromelysin 1-deficient mice to collagen-induced arthritis and cartilage destruction

OBJECTIVE It has long been proposed that stromelysin is one of the major degradative matrix metalloproteinases responsible for the loss of cartilage in rheumatoid arthritis (RA) and osteoarthritis (OA). This hypothesis was tested by examining the arthritic paws of stromelysin 1 (SLN1)-deficient mice for loss of cartilage and for generation of neoepitopes that would be indicative of aggrecan cleavage. METHODS The SLN1 gene was inactivated in murine embryonic stem cells, and knockout mice deficient in SLN1 activity were bred onto the B10.RIII background. The incidence and severity of collagen-induced arthritis (CIA) were compared in wild-type and knockout mice. Paws from mice with CIA were examined for loss of cartilage and for proteoglycan staining, as well as for the generation of the neoepitope FVDIPEN341. RESULTS SLN1-deficient mice developed CIA, as did the wild-type N2 mice. Histologic analyses demonstrated no significant differences among the B10.RIII, wild-type, and knockout mice in loss of articular cartilage and proteoglycan staining. No decrease in the FVDIPEN341 epitope was observed in the SLN1-deficient mice. CONCLUSION Disruption of the SLN1 gene neither prevents nor reduces the cartilage destruction associated with CIA. Moreover, SLN1 depletion does not prevent the cleavage of the aggrecan Asn341-Phe342 bond.

VDIPEN, a metalloproteinase-generated neoepitope, is induced and immunolocalized in articular cartilage during inflammatory arthritis.

The destruction of articular cartilage in immune inflammatory arthritic disease involves the proteolytic degradation of its extracellular matrix. The role of activated matrix metalloproteinases (MMPs) in the chondrodestructive process was studied by identifying a selective cleavage product of aggrecan in murine arthritis models initiated by immunization with either type II collagen or proteoglycan. We conducted semiquantitative immunocytochemical studies of VDIPEN341 using a monospecific polyclonal antibody requiring the free COOH group of the COOH-terminal Asn for epitope detection. This antibody recognizes the aggrecan G1 domain fragment generated by MMP [i.e., stromelysin (SLN) or gelatinase A] cleavage of aggrecan between Asn341-Phe342 but does not recognize intact aggrecan. VDIPEN was undetectable in normal mouse cartilage but was observed in the articular cartilage (AC) of mice with collagen-induced arthritis 10 d after immunization, without histological damage and clinical symptoms. This aggrecan neoepitope was colocalized with high levels of glycosaminoglycans (GAGs) in pericellular matrices of AC chondrocytes but was not seen at the articular surface at this early time. Digestion of normal (VDIPEN negative) mouse paw cryosections with SLN also produced heavy pericellular VDIPEN labeling. Computer-based image analysis showed that the amount of VDIPEN expression increased dramatically by 20 d (70% of the SLN maximum) and was correlated with GAG depletion. Both infiltration of inflammatory cells into the synovial cavity and early AC erosion were also very prominent at this time. Analysis of adjacent sections showed that both induction of VDIPEN and GAG depletion were strikingly codistributed within sites of articular cartilage damage. Similar results occurred in proteoglycan-induced arthritis, a more progressive and chronic model of inflammatory arthritis. These studies demonstrate for the first time the MMP-dependent catabolism of aggrecan at sites of chondrodestruction during inflammatory arthritis.

Kinetics of aggrecanase- and metalloproteinase-induced neoepitopes in various stages of cartilage destruction in murine arthritis.

OBJECTIVE Two major cleavage sites, one mediated by metalloproteinases (MMPs) and the other by an as-yet unidentified enzyme termed aggrecanase, have been observed in aggrecan. To learn more about the relative contribution of these enzymes during cartilage degradation, this study assessed the occurrence of both specific neoepitopes in cartilage during murine arthritis and examined the correlation between neoepitope formation and different aspects of cartilage damage. METHODS Reversible cartilage damage was induced in mice in the zymosan-induced arthritis (ZIA) model, partly irreversible cartilage damage in the antigen-induced arthritis (AIA) model, and irreversible, destructive cartilage damage in the collagen-induced arthritis (CIA) model. Immunolocalization techniques were used to detect the specific C-terminal neoepitopes VDIPEN (MMPS) and NITEGE (aggrecanase). RESULTS In normal cartilage from young adult mice, no VDIPEN epitopes were detected, but a limited amount of NITEGE epitopes were already present. During the early phase of proteoglycan (PG) depletion, NITEGE expression was raised substantially in all arthritis models. VDIPEN epitopes were not detected in this early phase of cartilage destruction. When PG depletion progressed toward advanced cartilage damage, VDIPEN epitopes were induced. During ZIA, minimal induction of VDIPEN was observed, whereas in AIA, strong, but partly reversible, VDIPEN staining was evident, and in CIA, an extensive presence and persistence of the MMP-induced neoepitope was seen. When VDIPEN epitopes were intensely present, NITEGE epitopes were greatly reduced at that site in the cartilage. CONCLUSION Presence of VDIPEN epitopes in cartilage correlated with severe cartilage damage, but these epitopes were not detected during early PG degradation. This suggests a limited role for VDIPEN-inducing MMPs in early PG degradation during murine arthritis. In contrast, aggrecanase epitopes were induced before the appearance of VDIPEN epitopes, but they disappeared with progression of cartilage damage.

Cleavage of aggrecan at the Asn341–Phe342 site coincides with the initiation of collagen damage in murine antigen‐induced arthritis: A pivotal role for stromelysin 1 in matrix metalloproteinase activity

OBJECTIVE The destruction of articular cartilage during arthritis is due to proteolytic cleavage of the extracellular matrix components. This study investigates the kinetic involvement of metalloproteinases (MMPs) in the degradation of the 2 major cartilage components, aggrecan and type II collagen, during murine antigen-induced arthritis (AIA). In addition, the role of stromelysin 1 (SLN-1) induction of MMP-induced neoepitopes was studied. METHODS VDIPEN neoepitopes in aggrecan and collagenase-induced COL2-3/4C neoepitopes in type II collagen were identified by immunolocalization. Stromelysin 1-deficient knockout (SLN1-KO) mice were used to study SLN-1 involvement. RESULTS In AIA, the VDIPEN epitopes in aggrecan appeared after initial proteoglycan (PG) depletion. The collagenase-induced type II collagen neoepitopes colocalized with VDIPEN epitopes. Remarkably, cartilage from arthritic SLN1-KO mice showed neither the induction of VDIPEN nor collagen cleavage-site neoepitopes during AIA, suggesting that stromelysin is a pivotal mediator in this process. PG depletion, as measured by the loss of Safranin O staining, was similar in SLN1-KO mice and wild-type strains. Furthermore, in vitro induction of VDIPEN epitopes in aggrecan and COL2-3/4C epitopes in type II collagen, on exposure of cartilage to interleukin-1, could not be accomplished in SLN1-KO mice, whereas intense staining was achieved for both epitopes in cartilage of wild-type strains. CONCLUSION This study emphasizes that SLN-1 is essential in the induction of MMP-specific aggrecan and collagen cleavage sites during AIA. It suggests that SLN-1 is not a dominant enzyme in PG breakdown, but that it activates procollagenases and is crucial in the initiation of collagen damage.

Myogenesis and histogenesis of skeletal muscle on flexible membranes in vitro.

SummaryPrimary muscle cell cultures consisting of single myocytes and fibroblasts are grown on flexible, optically clear biomembranes. Muscle cell growth, fusion and terminal differentiation are normal. A most effective membrane for these cultures is commercially available Saran Wrap. Muscle cultures on Saran will, once differentiated, contract vigorously and will deform the Saran which is pinned to a Sylgard base. At first, the muscle forms a two-dimensional network which ultimately detaches from the Saran membrane allowing an undergrowth of fibroblasts so that these connective tissue cells completely surround groups of muscle fibers. A three-dimensional network is thus formed, held in place through durable adhesions to stainless steel pins. This three-dimensional, highly contractile network is seen to consist of all three connective tissue compartments seenin vivo, the endomysium, perimysium and epimysium. Finally, this muscle shows advanced levels of maturation in that neonatal and adult isoforms of myosin heavy chain are detected together with high levels of myosin fast light chain 3.

Interleukin-1 receptor antagonist prevents expression of the metalloproteinase-generated neoepitope VDIPEN in antigen-induced arthritis

OBJECTIVE To investigate the relationship between occurrence of the matrix metalloproteinase-generated neoepitope VDIPEN and proteoglycan (PG) loss in arthritis, and to examine the role of interleukin-1 (IL-1) in VDIPEN expression. METHODS VDIPEN expression was investigated in murine antigen-induced arthritis by immunolocalization studies on joint sections. The involvement of IL-1 in VDIPEN expression was studied by blocking of IL-1 using IL-1 receptor antagonist (IL-1Ra). RESULTS Profound PG loss was evident early in arthritis, without significant VDIPEN expression. Full expression of the neoepitope appeared after a few days, when PG depletion was severe, and disappeared at late stages when cartilage showed recovery from PG depletion. At sites where chondrocyte death occurred and cartilage did not recover from the initial cartilage depletion, VDIPEN expression remained present. Prophylactic IL-1Ra treatment of arthritic mice resulted in almost complete prevention of VDIPEN expression. However, IL-1Ra had only a minor effect on PG depletion, emphasizing that there is no correlation between VDIPEN and early PG depletion. CONCLUSION This study indicates that IL-1 is involved in VDIPEN expression. Although VDIPEN-inducing metalloproteinases do not seem to be involved in early PG depletion during antigen-induced arthritis, metalloproteinase neoepitopes are present when PG depletion is severe.

Specific bioactivities of monocyte-derived interleukin 1 alpha and interleukin 1 beta are similar to each other on cultured murine thymocytes and on cultured human connective tissue cells.

In this report we compare the bioactivities of pure, human monocyte-derived interleukin 1 (IL-1) alpha and beta in the standard murine thymocyte proliferation assay, a human dermal fibroblast proliferation assay, and in an assay measuring stimulation of prostaglandin E2 (PGE2) release from human rheumatoid synoviocytes. In each case the different species of IL-1 produced saturable stimulation and gave similar dose response curves. Half-maximal stimulation was observed at average IL-1 concentrations of 29 pM in the thymocyte assay, 2 pM in the dermal fibroblast proliferation assay, and 5 pM in the synovial cell assay. Our results show that native, monocyte-derived IL-1 alpha and IL-1 beta are both potent stimulators of connective tissue cells and that the specific bioactivities of these molecules are similar to each other in tests on human connective tissue cells, as well as on murine lymphoid cells.

Identification of a high-affinity receptor for interleukin 1 alpha and interleukin 1 beta on cultured human rheumatoid synovial cells.

In this report the binding of recombinant human interleukins 1 alpha and 1 beta (rIL-1 alpha and rIL-1 beta) to primary cultures of human rheumatoid synovial cells is measured and compared to the concentrations of these mediators required for stimulation of PGE2 production by these same cells. The average concentration of IL-1 alpha required for half-maximal stimulation of PGE2 was 4.6 +/- 1.5 pM (+/- SEM) (n = 6), whereas for IL-1 beta half-maximal stimulation was observed at a concentration of 1.3 +/- 0.24 pM (n = 6). Both direct and competitive binding experiments were performed. In direct binding experiments, IL-1 alpha bound with a Kd of 66 pM (n = 1), while IL-1 beta bound with a Kd of 4 pM (n = 2). In competitive binding experiments, IL-1 alpha inhibited binding of 125I-IL-1 alpha with a Ki of 33-36 pM (n = 2) and binding of 125I-IL-1 beta with a Ki of 51-63 pM (n = 2). IL-1 beta inhibited binding of 125I-IL-1 alpha with a Ki of 2-3 pM (n = 2) and binding of 125I-IL-1 beta with a Ki of 7 pM (n = 2). The binding data were best fit by a model specifying a single class of receptors with homogeneous affinity for either IL-1 alpha or IL-1 beta and with an abundance of 3,000-14,000 sites per cell. Autoradiography showed that the vast majority of the synoviocytes within the cultures possessed IL-1 receptors. Comparison of biological response curves with the binding curves indicates that the observed receptors exhibit sufficiently high affinity to mediate the response of human synoviocytes to low picomolar concentrations of IL-1 alpha and IL-1 beta.

Aggrecan degradation in osteoarthritis and rheumatoid arthritis

Aggrecan turnover is critically important to maintain extracellular matrix homeostasis in articular cartilage. Cartilage aggrecan metabolism has been studied in chondrocyte cell cultures, cartilage explant cultures, and in whole animal models. In many of these studies, matrix metalloproteinases (MMPs) are proposed to degrade cartilage aggrecan. MMP expression appears elevated in joint tissues from patients with osteoarthritis (OA) and rheumatoid arthritis (RA) (Dean et al. 1989, Wolfe et al. 1993) and inhibitors of both MMPs and thiol proteinases have been shown to block aggrecan (Buttle et al. 1992) and type II collagen (Mort et al. 1993) degradation in cartilage explant cultures. Using antibodies and cDNA probes, elevations in expression and concentration of many of these enzymes in animal models and in OA and RA have been described. Human cartilage aggrecan has now been cloned and sequenced (Doege et al. 1991). This information, in combination with the ability to sequence aggrecan and aggrecan fragment...