Liu Studies
Detecting the Warning Signs of Arthritis
Over 40 million Americans have been diagnosed with osteoarthritis, a condition second only to cardiovascular disease as a cause of disability in Americans over 50 years of age. Nonsurgical treatments, even when they provide temporary relief of symptoms, do not address the underlying disease processes. Ultimately many sufferers will require surgical intervention, such as joint replacement surgery, to achieve meaningful improvement, as reflected by the fact that more than 120,000 artificial hip joints are implanted in the United States annually.
Despite the prevalence of the condition, little is known about how osteoarthritis develops or how it progresses. Many arthritis sufferers are not seen by a doctor until long after they have experienced substantial loss of cartilage and impairment of function. When detection of cartilage degeneration is made so late in the disease process, the body’s natural resources are are no longer effective in stopping or reversing joint degeneration.
Studies in this area have identified a “target” molecule, cartilage oligomeric matrix protein (COMP), which is found in abundance in the extracellular matrix of joint tissues. The significance of COMP derives from the fact that its degradation appears to herald the onset of arthritis.
 |
|
Early studies in this area resulted in purifying COMP from human articular cartilage and delineating its structure and tissue distribution in normal and diseased cartilage. Specific mono- and polycolonal antibodies against COMP have been produced in our laboratories. These reagents have been used to determine the tissue distribution of COMP by immunostaining, the amounts and types of COMP in biological fluids using competitive enzyme-linked immunosorbent assay (ELISA) and western blot analysis.
The key finding in these studies is that the tissue distribution for COMP becomes altered in arthritic states. Analysis of synovial fluid revealed increased amounts of degradation fragments of COMP as well as specific degradation fragments that serve as “flags” for altered cartilage metabolism.
 |
|
Two clinical studies were subsequently conducted to test the utility of COMP as a biological marker of arthritis. The first demonstrated greater levels of COMP in patients with osteoarthritis, rheumatoid arthritis, and other arthritides than in healthy “controls” subjects. The second study focused on a potential prearthritic condition that arises inpatients with traumatic bone bruises that have been detected detected by magnetic resonance imaging (MRI). This study revealed that there is about 10 times higher synovial fluid COMP levels in the injured knees of these patients than in their noninjured knees.
Encouraged by these findings, we have planned future studies to determine the specific enzymes that degrade COMP, to develop more specific reagents toward degradative fragments (neoepitopes), and to conduct longitudinal tracking of COMP synovial levels in patients who exhibit prearthritic symptoms.
Decoding the Formula for Cartilage Growth
A clearer understanding of regulation of cartilage growth and differentiation has potent clinical applications in designing biologically based treatment for several orthopaedic and rheumatologic conditions, including fracture nonunion, large segmental bone or cartilage defects, and osteoarthritis.
Research in this area seeks to identify the regulatory agents—transcription factors—that are responsible for signaling undifferentiated cells to create cartilage. In the human embryo, certain of these mesenchymal stem cells are pluripotential: they have the ability to develop at different locations into connective or supporting tissues, smooth muscle, vascular endothelium, or blood cells. Until recently, the molecular events governing the differentiation of stem cells into chondrocytes (cartilage cells) and the expression of cartilage marker genes have been poorly understood.
The differentiation of stem cells into chondrocytes is a fundamental molecular event essential for growth, repair, and regeneration of cartilage and bone. After commitment to the chondrocyte lineage, these cells undergo condensation, cease expression of type I collagen, and differentiate into a chondrocytic phenotype characterized by expression of collagens type II, IX, XI, the noncollagenous protein cartilage oligomeric matrix protein (COMP), and the proteoglycan aggrecan.
A key component in this process is COMP, a noncollagenous extracellular matrix protein with a relatively cartilage-specific spatial and temporal expression pattern. We have identified the COMP promoter—that portion of the cartilage gene that regulates the expression of COMP. Work is now under way to identify the transcription factors that, by binding to the COMP promoter, direct cartilage cells’ function and phenotype.
Recent studies examining transcriptional regulation of the genes for subunits of types II and XI collagen have identified members of the Sox family of transcription factors (L-Sox-5, Sox-6, Sox-9) as direct regulators of these genes. Mutations in the human Sox-9 gene are associated with the development of camptomelic dysplasia, a human dwarfism condition that affects all cartilage-derived structures. These findings suggest that Sox-9, and members of the Sox gene family, may participate in the activation of a genetic program designed to coordinately regulate the expression of genes responsible for the chondrocytic phenotype. Similar analysis of the transcriptional regulation of other cartilage-specific genes, encoding both collagenous and noncollagenous proteins, provides a useful strategy for identifying additional transcription factors that control chondrocyte specification and differentiation.
The gene for cartilage oligomeric matrix protein encodes a noncollagenous pentameric matrix protein expressed predominantly in articular cartilage. Mutations in the human COMP gene have been linked to the development of pseudoachondroplasia and multiple epiphyseal dysplasia autosomal dominant forms of short-limb dwarfism characterized by short stature and early-onset osteoarthritis. During embryonic and adult stages in several different species, including mouse, rat and human, COMP expression appears to be spatially and temporally restricted primarily to chondrocytes. This relatively cartilage-specific expression pattern, taken together with the observation that COMP mRNA levels are elevated by the chondrogenic factor TGFb-1 suggest that COMP gene expression may be regulated by the chondrocyte differentiation state.
To begin to understand the mechanisms governing cartilage-specific expression of COMP, we cloned the mouse COMP promoter and identified genetic elements necessary for cartilage-specific expression in the chondrocytic cell line. This cell line, established from long-term culture of the transplantable Swarm rat chondrosarcoma, displays a stable chondrocytic phenotype. We have identified two regions in the promoter sequence, an element situated proximally (–125 to –75) and a region located distally (–1925 to –592), that are necessary for COMP expression. Sequences within the proximal region are conserved between the mouse and human promoters. Analysis of nucleotides within this conserved region suggests possible regulation by the high-mobility group class of transcription factors. Identification of cartilage-specific enhancers in the COMP promoter provides genetic probes for cloning transcription factors that regulate chondrocyte-specific gene expression.
 |
|
Our laboratory is in the process of localizing the active element(s) in the distal region. The next step will be to determine the nature of possible DNA-binding proteins by gel mobility shift assays. The functional importance of these elements for transcriptional activity must then be determined by site-directed mutagenesis of these regions within the context of the wild-type promoter. Identification of nucleotides within the element(s) that are required for both protein binding and promoter activity should provide molecular probes for cloning cartilage-specific transcriptional regulators. These experiments promise to further our understanding of how stem cells differentiate into cartilage.
Defining the Role of COMP in Skeletal Development
In the 1990s researchers found that disruption of the cartilage oligomeric matrix protein (COMP) gene is responsible for two types of dwarfism, multiple epiphyseal dysplasia and pseudoachondroplasia. Patients with these disorders are characterized by short stature and the early onset of osteoarthritis.
Discovery of the link between COMP and these conditions has suggested new lines of research involving this key protein. Dr. Liu is conducting studies to determine the underlying mechanism of this phenomenon in the hope of gaining new insights into skeletal development and growth in general. The research now under way has two aims: (1) to characterize the function of COMP in both its normal and mutated manifestations; and (2) to develop a transgenic mouse that exhibits the characteristics of COMP-altered dwarfism and can thus be used to learn how COMP mutations disrupt growth.
Our recent work in this area indicates that COMP is an important biochemical component of extracellular matrix (the “scaffolding” in musculoskeletal tissues), particularly in its interactions with another, important extracellular matrix protein, fibronectin. Laboratory studies undertaken so far have revealed that these protein interactions are of high affinity and occur at specific sites on each protein.
 |
|
We discovered that the interaction is concentration-dependent and saturable and appears to occur under physiologically relevant conditions. Rotary shadowing and molecular electron microscopy and fragment binding analysis using the solid-phase assay revealed a predominant binding site for the COMP C-terminal globular domain to a molecular domain approximately 14 nanometers (a nanometer is a millionth of a millimeter) from the N-terminal domain of fibronectin. The fact that these molecular species bind in vivo was demonstrated by colocalization of both COMP and fibronectin in the chondrocyte pericellular matrix by laser confocal microscopy of chondrocytes grown in agarose culture and by appositional and colocalization of these proteins in the growth plate of primates by immunohistochemistry.
 |
|
It has also been demonstrated that COMP can mediate cell-matrix interactions as well as matrix-matrix interactions. In our laboratory experiments, articular chondrocytes exhibited both a time- and dose-dependent attachment to COMP, fibronectin, and vitronectin. Antibodies to the vitronectin receptor were able to significantly inhibit chondrocyte attachment to both COMP and vitronectin, although they had no inhibitory effect on cell attachment to fibronectin. Iodinated chondrocyte cell membrane receptors were bound to an affinity column of COMP. Once bound material was eluted, autoradiography revealed two prominent bands that were confirmed to be the vitronectin receptor. We were thus able to demonstrate for the first time that COMP is a adhesion molecule for chondrocytes and that this interaction is mediated by the integrin receptor.
 |
|
In preparation for creating a transgenic mouse model, we studied the normal sequence and distribution of COMP in the developing mouse. To begin, we cloned and sequenced mouse COMP cDNA. The encoded mouse product of 755 amino acids shares a high degree of identity to and possesses all the characteristic molecular features of both rat and human COMP. The significance of this finding derives from the concept of conservation of sequence during evolution, implying as it does biological, nonredundant protein function.
 |
|
In adult mouse tissues, COMP was found to be highly expressed in cartilage and tendon, moderately expressed in trachea, bone, skeletal muscle, eye, heart, and placenta, and minimally expressed in testis. Immunohistology revealed that COMP expression began as early as 10 days postcoitus in predifferentiated mouse embryo mesenchyme. COMP was detected in all cartilaginous tissues and the skeletal muscles in the embryo at day 13. As development progressed, expression of COMP was marked in the growth plate. At 19 days postcoitus, COMP was prominently expressed in the hypertrophic zone of the growth plate, perichondrium, and periosteum as well as in the superficial layer of articular cartilage surface, but it was absent in the more central areas of the epiphyseal cartilage. The restricted tissue distribution and expression of COMP in developing as well as adult mouse tissues suggest the regulation of this protein at the transcriptional level. These findings regarding COMP expression during normal mouse skeletal development are critical to determining how its mutation perturbs growth.
Experiments are under way in vitro to determine normal and mutant protein function. Knowledge of the mouse genomic sequence has enabled us to produce a vector to introduce the mutated COMP and develop a transgenic mouse that will express the phenotype seen in human dwarfisms. These animals will be instrumental in a detailed developmental analysis.
