Lawrence C. Rosenberg

Evidence for altered synthesis of type II collagen in patients with osteoarthritis.

There is evidence to suggest that the synthesis of type II collagen is increased in osteoarthritis (OA). Using an immunoassay, we show that the content of the C-propeptide of type II procollagen (CPII), released extracellularly from the newly synthesized molecule, is directly related to the synthesis of this molecule in healthy and osteoarthritic articular cartilages. In OA cartilage, CPII content is often markedly elevated (mean 7.6-fold), particularly in the mid and deep zones, reaching 29.6% of the content in newborn. Synthesis is also directly related to total collagen II content in OA, suggesting its importance in maintaining collagen content and cartilage structure. The release of CPII from cartilage is correlated directly with cartilage content. However, the increase in CPII in OA cartilage is not reflected in serum, where a significant reduction is observed. Together these studies provide evidence for alterations in procollagen II synthesis in vivo in patients with OA.

Decorin-binding adhesins from Borrelia burgdorferi

Lyme disease is a tick‐transmitted infection caused by the spirochete Borrelia burgdorferi. Ticks deposit B. burgdorferi into the dermis of the host, where they eventually become associated with collagen fibres. We demonstrated previously that B. burgdorferi is unable to bind collagen, but can bind the collagen‐associated proteoglycan decorin and expresses decorin‐binding proteins (Dbps). We have now cloned and sequenced two genes encoding the proteins, DbpA and DbpB, which have a similar structure, as revealed by circular dichroism (CD) spectroscopy of recombinant proteins. Competition experiments revealed a difference in binding specificity between DbpA and DbpB. Western blot analysis of proteinase K‐treated intact B. burgdorferi and transmission electron microscopy studies using antibodies raised against recombinant Dbps demonstrated that these proteins are surface exposed. DbpA effectively inhibits the attachment of B. burgdorferi to a decorin substrate, whereas DbpB had a marginal effect, suggesting a difference in substrate specificity between the two Dbps. Polystyrene beads coated with DbpA adhered to a decorin‐containing extracellular matrix produced by cultured skin fibroblasts, whereas beads coated with OspC did not. Taken together, these data suggest that Dbps are adhesins of the MSCRAMM (microbial surface component‐recognizing adhesive matrix molecule) family, which mediate B. burgdorferi attachment to the extracellular matrix of the host.

The inhibitory effect of cartilage proteoglycans on hydroxyapatite growth

SummaryThe calcification of connective tissues, including cartilage, is under the control of many interacting systems. Proteoglycans are thought to retard the deposition of hydroxyapatite crystals, and modification of the proteoglycans presumably facilitates mineralization in those tissues that are actively calcifying. The mechanism underlying these regulations remains speculative. This study investigates this question by comparing the inhibitory effectiveness of several macromolecules at neutral pH and approximately physiological ionic strengths. Inhibitors tested include bovine nasal proteoglycan monomer A1D1D1 and aggregate-containing A1 fractions, glycosaminoglycan chains (chondroitin 4-sulfate), and neutral dextran (as an uncharged analog). Hydroxyapatite growth was assessed either by measuring the time-dependent decreases in solution calcium and phosphate concentrations, or by determining utilization of hydroxyl ion in a pH-Stat. All species studied inhibit hydroxyapatite growth, and the extent of inhibition for each class is concentration-dependent. The proteoglycan aggregate-containing A1 fraction is more effective than the proteoglycan monomer at the same concentration, and the proteoglycan monomer is more effective than chondroitin 4-sulfate. Neutral dextran inhibits hydroxyapatite growth less effectively than proteoglycans. These results suggest that inhibition of hydroxyapatite growth by proteoglycans critically depends on both status (aggregate, monomer, etc.) and hydrodynamic size of this macromolecule, supporting the hypothesis that modification of proteoglycansin vivo functions to modulate the effectiveness of proteoglycans as a hydroxyapatite growth inhibitor.

Effects of Bone CS-Proteoglycans, DS-Decorin, and DS-Biglycan on Hydroxyapatite Formation in a Gelatin Gel

Abstract. The small leucine-rich bone proteoglycans, biglycan and decorin, can be purified by chromatography on hydroxyapatite columns, demonstrating their potential affinities for bone apatite. To determine their effects on in vitro apatite formation and growth, a mixture of the chondroitin-sulfate (CS) bone proteoglycans, or purified fractions of the dermatan sulfate (DS) containing proteoglycans, DS-decorin and DS-biglycan obtained from skin and articular cartilage, respectively, were analyzed in a gelatin gel diffusion system in which apatite formation occurs in the absence of proteins in a 3.5 day period. Low concentrations of the bone CS-proteoglycan mixture and low DS-biglycan concentrations (5–25 μg/ml) increased apatite formation relative to proteoglycan-free controls at 3.5 days. The CS-proteoglycan mixture was less effective at 50 μg/ml than at 10 μg/ml. DS-biglycan was similarly most effective at 5–25 μg/ml. At 5 days, when apatite growth and proliferation were assessed, 10 and 50 μg/ml of both CS-bone proteoglycan and DS-biglycan increased mineral yields. DS-decorin, in contrast, had no significant effect on mineral accumulation at any of these concentrations. In seeded growth experiments, 1 and 10 μg/ml CS-proteoglycan and 10 and 50 μg/ml DS-biglycan were significant effective inhibitors of mineral accretion, whereas DS-decorin showed no tendency to inhibit seeded growth. Using molar extinction coefficients to determine concentrations, the binding of DS-biglycan and DS-decorin to apatite (specific surface 54 m2/g) was determined using a Langmuir adsorption isotherm model. DS-biglycan had a greater affinity for apatite than DS-decorin (0.285 ml/μmol versus 0.0098 ml/μmol). DS-biglycan binding was more specific with fewer binding sites (3.5 μmol/m2 compared with 18.2 μmol/m2 for DS-decorin). Data suggest that of the small proteoglycans, biglycan may play a more significant role than decorin in the regulation of mineralization.

Structural Changes During Development in Bovine Fetal Epiphyseal Cartilage

Sedimentation coefficients of approximately 150 S show that proteoglycan aggregates from bovine fetal epiphyseal cartilage are exceptionally large. To determine the structural basis for the unusually large size of these proteoglycan aggregates, identify changes in proteoglycan structure with changing developmental age, and provide a basis for demonstrating the structural modifications which may occur in growth plate proteoglycan aggregates during endochondral ossification, we examined the molecular architecture and dimensions of fetal epiphyseal proteoglycans by electron microscopy. The eight bovine epiphyseal cartilages studied ranged in fetal age from 168 to 241 days. Proteoglycans were extracted in 4 M guanidinium hydrochloride containing protease inhibitors and isolated by equilibrium density gradient centrifugation under associative and dissociative conditions. Electron micrographs were made from monolayer preparations of proteoglycan-cytochrome c mixtures on nitrocellulose support films. The overall molecular architecture of the proteoglycan aggregates from fetal epiphyseal cartilages was similar to that of aggregates from other cartilages and showed a single, unbranched central hyaluronic acid filament to which many proteoglycan monomers were attached. However, the dimensions of the fetal proteoglycans differed strikingly from those of proteoglycans from mature cow nasal or immature calf nasal cartilage. Specifically, proteoglycan aggregates from bovine fetal epiphyseal cartilage showed: (a) longer hyaluronic acid central filaments; (b) greater numbers of proteoglycan monomers per aggregate; (c) closer spacing of proteoglycan monomers along the hyaluronic acid central filament; and (d) longer proteoglycan monomer core proteins. Proteoglycan monomers bound to hyaluronate consisted of two segments: (1) a peripheral thick segment, composed of the chondroitin sulfate chains condensed along the peripheral portion of the protein core, which corresponds to the chondroitin sulfate-rich region; and, (2) a central thin segment, devoid of visible glycosaminoglycan chains, which attaches directly to the hyaluronic acid central filament and contains the hyaluronic acid binding region and a portion of the keratan sulfate-rich region. The contribution of the thin segment to total monomer length decreased as total monomer length increased. Thus, in longer monomers the thick segment contributed more to total monomer length and the thin segment contributed less. Both the thin and thick segments of monomers from fetal epiphyseal cartilage were longer than the corresponding segments of calf nasal cartilage and mature bovine nasal cartilage monomers.(ABSTRACT TRUNCATED AT 400 WORDS)

Viscoelastic properties of proteoglycan subunits and aggregates in varying solution concentrations

Using a cone-on-plate mechanical spectrometer, we have measured the linear and non-linear rheological properties of cartilage proteoglycan solutions at concentrations similar to those found in situ. Solutions of bovine nasal cartilage proteoglycan subunits (22S) and aggregates (79S) were studied at concentrations ranging from 10 to 50 mg ml-1. We determined: (1) the complex viscoelastic shear modulus G (omega) under small amplitude (0.02 radians) oscillatory excitation at frequencies (omega) ranging from 1.0 to 20.0 Hz, (2) the non-linear shear rate (gamma) dependent apparent viscosity napp (gamma) in continuous shear, and (3) the non-linear shear rate dependent primary normal stress difference sigma 1 (gamma) in continuous shear. Both the apparent viscosity and normal stress difference were measured over four decades of shear rates ranging from 0.25 to 250 s-1. Analysis of the experimental results were performed using a variety of materially objective non-linear viscoelastic constitutive laws. We found that the non-linear, four-coefficient Oldroyd rate-type model was most effective for describing the measured flow characteristics of proteoglycan subunit and aggregate solutions. Values of the relaxation time lambda 1, retardation time lambda 2, zero shear viscosity no, and nonlinear viscosity parameter muo were computed for the aggregate and subunit solutions at all of the solute concentrations used. The four independent material coefficients showed marked dependence on the two different molecular conformations, i.e. aggregate or subunit, of proteoglycans in solution.

Changes in proteoglycan aggregates during cartilage mineralization.

SummaryThe dimensions of proteoglycan aggregates, aggregated monomers, and nonaggregated monomers, and the proportion of aggregated monomers found in the different zones of bovine rib growth plate have been defined by the electron microscopic monolayer technique. Growth plates were divided into the following 1 mm thick transverse slices; the hypertrophic zone, the lower proliferative zone, the upper proliferative zone, a transitional zone, and epiphyseal cartilage. Proteoglycans prepared by associative extraction followed by equilibrium density gradient centrifugation under associative conditions were examined by electron microscopy. Proteoglycan aggregate size decreased sharply in the lower proliferative and hypertrophic zones, as indicated by decreases in hyaluronate filament length and in the number of monomers per aggregate. Aggregated proteoglycan monomers did not show evidence of proteolytic degradation. Nonaggregated monomers were shorter than aggregated monomers, but their mean length did not decrease in the lower proliferative and hypertrophic zones. However, the proportion of nonaggregated monomers increased in these zones. Thus, before the cartilage matrix mineralized in the lower proliferative zone and as the cartilage matrix began to mineralize in the hypertrophic zone, proteoglycan aggregate size decreased and the proportion of aggregated monomers decreased. These changes in matrix proteoglycans may be one of the events that allow cartilage mineralization.

Electron microscopic studies of cartilage proteoglycans

Proteoglycans, molecules consisting of glycosaminoglycan chains bound to protein, form a significant part of the cartilage extracellular matrix. Biochemical and biophysical methods describe the average composition and physical properties of these polydispense molecules. Electron microscopy reveals the structure and dimensions of individual proteoglycans. Examination of individual molecules can confirm or challenge concepts of their structure developed from studies of their chemical composition and physical properties, and may suggest new directions for biochemical investigation. Electron microscopy has confirmed that cartilage proteoglycans exist on two levels of organization: monomers consisting of central protein core filaments with attached glycosaminoglycan chains and aggregates consisting of central hyaluronate filaments with multiple attached monomers. Most aggregated monomers have a thin segment which attaches to the hyaluronate filament and probably represents primarily the keratan sulfate rich region of the protein core, and a peripheral thick segment that represents the chondroitin sulfate rich region and in some monomers part of the keratan sulfate rich region. Proteoglycans vary considerably in size, charge and composition. Direct visualization of proteoglycan aggregates and nonaggregated monomers has helped explain the structural basis of this polydispensity. Monomers vary in protein core length, number of glycosaminoglycan chains and length of the glycosaminoglycan chains. Aggregates vary in hyaluronate filament length, spacing between monomers, number of monomers per aggregate, and aggregated monomer length. In most populations of aggregates, from most tissues, variability in the number of monomers per aggregate produces most of the difference in aggregate size. Link proteins, small proteins that bind to monomers and hyaluronate, help determine aggregate size and the proportion of monomers that aggregate. Experiments in vitro show that link protein can increase aggregate size four fold, make the spacing between aggregated monomers more regular and increase the proportion of monomers that aggregate ten fold. With increasing age, cartilage proteoglycan monomers become shorter, more variable in length, have shorter chondroitin sulfate chain clusters and have a shorter thin segment which may result from an increase in keratan sulfate content. Study of monomers newly synthesized by calf and steer chondrocytes suggests that the age related changes in monomer structure result largely from changes in proteoglycan synthesis or intracellular processing. Aggregates also change with age. They become shorter, have fewer monomers per aggregate and have shorter aggregated monomers. In addition, the proportion of monomers that aggregate decreases. These age related changes in proteoglycan aggregation may result from a decreasing concentration of functional link protein or from accumulation of fragments of the protein core containing the hyaluronic acid binding region.(ABSTRACT TRUNCATED AT 400 WORDS)

Immunity to the G1 globular domain of the cartilage proteoglycan aggrecan can induce inflammatory erosive polyarthritis and spondylitis in BALB/c mice but immunity to G1 is inhibited by covalently bound keratan sulfate in vitro and in vivo.

Earlier work from this laboratory showed that the human proteoglycan aggrecan from fetal cartilages can induce a CD4+ T cell-dependent inflammatory polyarthritis in BALB/c mice when injected after removal of chondroitin sulfate chains. Adult keratan sulfate (KS)-rich aggrecan does not possess this property. We found that two CD4+ T cell hybridomas (TH5 and TH14) isolated from arthritic mice recognize bovine calf aggrecan and the purified G1 domain of this molecule, which also contains a portion of the interglobular domain to which KS is bound. These hybridoma responses to G1 are enhanced by partial removal of KS by the endoglycosidase keratanase or by cyanogen bromide cleavage of core protein. KS removal results in increased cellular uptake by antigen-present cells in vitro. After removal of KS by keratanase, G1 alone can induce a severe erosive polyarthritis and spondylitis in BALB/c mice identifying it as an arthritogenic domain of aggrecan. The presence of KS prevents induction of arthritis presumably as a result of an impaired immune response as observed in vitro. These observations not only identify the arthritogenic properties of G1 but they also point to the importance of glycosylation and proteolysis in determining the arthritogenicity of aggrecan and fragments thereof.

Cellular immunity to the G1 domain of cartilage proteoglycan aggrecan is enhanced in patients with rheumatoid arthritis but only after removal of keratan sulfate.

OBJECTIVE To determine whether patients with rheumatoid arthritis (RA) express cellular immunity to the purified G1 globular domain of cartilage proteoglycan (PG) aggrecan and whether it is influenced by the removal of keratan sulfate (KS) chains from the molecule. METHODS The G1 globular domain of PG was purified from mature bovine articular cartilage, digested with keratanase, and used in proliferation assays with peripheral blood lymphocytes (PBL) isolated from 43 patients with RA, 11 patients with nonarticular rheumatism (NAR), including soft tissue rheumatism and mechanical back pain, and 13 healthy age- and sex-matched control subjects. RESULTS Removal of KS chains from the G1 globular domain resulted in significantly increased prevalence and values of cellular immune responses to G1 in RA patients compared with the control and NAR groups. In the majority of RA patients, KS chains on G1 significantly inhibited its immune recognition by PBL. There was no significant effect of KS removal on the immunity to G1 in patients with NAR and in the healthy control group. CONCLUSION These results reveal that immune reactivity to the G1 globular domain of the cartilage PG aggrecan is enhanced in patients with RA but only when KS chains are removed. Thus, KS chains inhibit immune responses to this domain of aggrecan. Since immunity to the G1 globular domain of aggrecan induces an erosive polyarthritis in BALB/c mice after removal of KS chains, immunity to the G1 globular domain, cleaved by proteases to remove KS chains, may play a role in the pathogenesis of RA.