Sandell Lab

Linda Sandell

Linda Sandell, Ph.D

Mildred B. Simon Research Professor and Director of Research

Department of Orthopaedic Surgery, Washington University

sandelll@wustl.edu
BJC - Institute of Health
11th floor - RM 11617
Phone: (314) 454-7800

Projects

1. CD-RAP Gene Regulation
2. Transcription of Chondrocyte Genes
3. PIIBNP and Bone Resorption
4. Molecular Regulation of Chondrogenesis and Osteoarthritis
5. Chondrogenesis from Stem Cells
6. Inhibition of Angiogenesis
7. Site-1 Protease and Endoplasmic Reticulum Stress
8. Genetics of Cartilage Regeneration and Degeneration

 

1. CD-RAP Gene Regulation

CD-RAP, also known as MIA (melanoma inhibiting activity), and the CD-RAP gene provides an excellent model for chondrocyte gene expression for two primary reasons: it is the most cartilage-specific protein, and the gene has proven very amenable to study. Even though CD-RAP is expressed in some other tissues at low levels, it is less widespread than COL2A1 and aggrecan (ACAN) in the embryo. CD-RAP expression is not found in many of the tissues known to synthesize COL2A1, namely skin, aorta, intervertebral disc, and vitreous. Secondly, and most importantly, of all the cartilage-characteristic genes, the CD-RAP gene (Fig. 1) is the most compact gene and has proven to be easily amenable to study; our discoveries in this gene have been used to direct new studies on chondrocyte gene regulation. While little progress has been made recently in the regulation of many other cartilage-characteristic genes such as ACAN and COL2A1, our studies with CD-RAP continue to be productive.

In non-neoplastic tissues MIA/CD-RAP expression is activated from the beginning of chondrogenesis throughout cartilage development and in vitro it is a specific marker for chondroid differentiation. Cartilage damage due to rheumatoid arthritis releases MIA/CD-RAP form the chondroid matrix and this can be monitored clinically by enhanced CD-RAP serum levels. Based on its highly restricted activity, the CD-RAP promoter was used to study transcriptional mechanisms mediating chondrocyte differentiation. Our laboratory has shown that a 2251 bp fragment of the murine CD-RAP 5’ flanking sequence contains all known functionally important transcriptional regulatory elements sufficient to confer tissue-specific expression in vivo. Expression of a LacZ-reporter under control of the 2251 bp CD-RAP promoter was exclusively observed in developing cartilage and transiently in embryonic breast buds. Recent studies using a longer promoter (3.3 kb) have shown that a specific region of the gene also contains the necessary elements to restrict gene expression to cartilage, and, when removed, permits inappropriate gene expression in liver, lung and muscle (Okazaki et al., 2006). Cell culture studies and unpublished in vivo studies in mice indicate that binding site for C/EBPß is at least partially responsible for this repression in non-cartilaginous tissues (Okazaki et al., 2006).

CD Rap

Figure: Schematic drawing of the CD-RAP gene. snRNP and Rab4b are flanking genes.

2. Transcription of Chondrocyte Genes

Transcription Factors Involved in the Regulation of Cartilage-Specific Gene. During the past two decades, many TFs that control the expression of cartilage-specific genes have been discovered and characterized by in vitro and in vivo studies. Both positive and negative TFs have been found to determine developmental events during chondrogenesis, the process by which mesenchymal condensations form the cartilage anlagen, which eventually forms the cartilage of the articular joint or undergoes hypertrophy and endochondral ossification to form bone. (Fig. 1) These events are controlled by cascades of both activators and repressors that interact with the promoter or enhanced regions of chondrocyte-specific genes, including those encoding type II, type IX and type XI collagens, aggrecan (ACAN), and CD-RAP. The high mobility group (HMG) protein Sox9 plays a key role in cartilage formation and maintenance by permitting transcription of cartilage-specific genes such as type II and type IX collagens, aggrecan and CD-RAP. Sox9 activates COL2A1 transcription by binding to the first intron enhancer in the COL2A1 gene (the first intron also doubles as an enhancer) through its HMG DNA-binding domain and acts cooperatively with L-Sox5 and Sox6 to regulate chondrogenesis in vivo. These and other Sox genes are regulated in a dynamic fashion during chondrogenesis by members of the BMP-TGFß family. Other extracellular mediators that control chondrocyte differentiation include Indian Hedgehog (Ihh) via PTHrP, Wnt proteins via ß-catenin, and fibroblast growth factors (FGFs) via specific receptors that promote or suppress proliferation. The anabolic effects of IGF-1, BMP-2 and FGF-2 on differentiated chondrocytes appear to be mediated, at least in part, by Sox9. The TFs involved in positive and negative regulation of chondrocyte differentiation are shown in Figure 1 showing the general steps in chondrogenesis regulated by TFs.

Figure: Positive and negative transcriptional regulation during cartilage development.

Negative Regulatory Factors. While Sox binding is necessary for up-regulation of chondrocyte genes, removal of positive factors is not sufficient for repressed expression. Indeed, tissue specificity necessitates repressed expression in most all other tissues and down-regulation during skeletal development. Negative TF activity on chondrocyte-specific genes is necessary for two reasons: (1) when required to down-regulate genes expressed in the chondrocyte and (2) to repress transcription in “non-chondrocytes”. The expression of most cartilage genes is very high during maturation of the growth plate and expansion of articular and hyaline cartilages. However, expression is much lower in hypertrophic chondrocytes and in mature cartilage tissue. In fact the negative regulators, ?EF-1 and AP-2, are detected by immunohistochemistry in mature tissues and hypertrophic cartilage while the positive factor, Sox9, is greatly reduced. E-box motifs, which are consensus-binding sites for bHLH proteins, are present in promoter and enhancer regions of COL2A1 and CD-RAP genes from different species. The interaction of ?EF-1 with conserved E-box sites containing CACCTG, or Snail family members Snail and Slug with CAGGTG, represses constitutive activity of the COL2A1 promoter. The bHLH protein, Scleraxis, can dimerize with other E-box-binding proteins and is expressed at early stages of chondrogenesis in regions surrounding Sox9. The TF DEC1 promotes chondrocyte differentiation at early and late stages in response to PTHrP and cAMP. Differential expression of the TFs Id1, 2, 3 and 4 may influence chondrogenesis and phenotypic expression in mature chondrocytes and chondrosarcoma cells. The nuclear factor of activated T cells NFATp(c2) suppresses chondrogenesis and inhibits aggrecan and COL2A1 gene expression in adult chondrocytes, whereas NFAT4 induces chondrogenesis by stimulating BMP expression.

3. PIIBNP and Bone Resorption

200 million people worldwide suffer from osteoporosis. Osteoporosis causes low bone mass and increases bone fragility that leads to osteoporotic fracture, which is largely a problem of the elderly, particularly women. Osteoclasts are cells of hematopoietic origin with the unique capacity to resorb bone. Consequently, these cells play a major role in several common skeletal diseases including osteoporosis. Several new drugs have been recently developed to inhibit bone resorption and are now currently available to treat and prevent diseases expressing high bone resorption rate.

We have recently found that chondrostatin (PIIBNP) inhibits cell survival and function of osteoclasts via integrin-mediated signal in vitro. Therefore, PIIBNP might be a new therapeutic strategy for osteoporosis.

Integrins are heterodimeric adhesion receptors that mediate cell-matrix interaction. Osteoclasts exhibit high expression levels of alpha v beta 3 integrin that binds to various extracellular matrix proteins. Arg-Gly-Asp (RGD) containing peptides bind to the alpha v beta 3 integrin receptor, and inhibit cell survival and function. The N-propeptide of both type IIA and type IIB procollagen contains the integrin binding sequence RGDRGD encoded by exon 6. This sequence is conserved across species, suggesting a potential function. We have shown that type IIB procollagen N-propeptide (PIIBNP) mediates cell adhesion via beta 3 and beta 5 integrin.

4. Molecular Regulation of Chondrogenesis and Osteoarthritis

Osteoarthritis will affect 50 million Americans by 2020. The disease process is characterized by the aberrant gene expression and the inability of cartilage chondrocytes to repair the extracellular matrix leading to cartilage degeneration. Computational and experimental methods have been combined to analyze groups of co-expressed genes and define common regulatory domains. These analyses to define transcriptional regulatory networks that predominantly operate in and define cartilage development (as opposed to the growth plate which results in bone formation). As new transcription factors arise, specific mechanisms of activity will be deciphered using our CD-RAP and COL2A1 gene models. We have four areas of research (1) mechanism of transcriptional regulation of the CD-RAP gene by the E-box proteins ?EF-1/USF1/USF2, C/EBPß, and the forkhead transcription factor, HNF-3ß (2) the role of transcriptional control in IL-1ß induced genes such as chemokines (3) the role of transcription factors in differentiation of cartilage from stem cells, and (4) the role of transcriptional regulation in osteoarthritis. The experimental approach focuses on (1) using genome-wide expression analyses to determine co-regulated genes, (2) screening for expression of transcription factors using a 1700 oligonucleotide array, (3) computational analyses for transcription factor binding domains of the promoters of co-regulated genes, and (4) confirmation of mechanistic function on the specific cartilage genes, CD-RAP and COL2A1. These studies will apply powerful techniques combining experimental and computational approaches to look beyond candidate genes to elucidate regulatory mechanisms critical for developing strategies to control chondrogenesis during development, in tissue engineering and repair, and ultimately to help find biological methods to control osteoarthritis.

5. Chondrogenesis from Stem Cells

The creation of cartilage from mesenchymal stem cells (stromal cells) has proven difficult primarily because, after differentiating into chondrocytes, the cells continue to differentiate into hypertrophic chondrocytes and eventually disintegrate or are replaced by bone as in the typical endochondral bone cascade. Pluripotent mesenchymal stem cell-like population that have the capacity to differentiate toward cells of connective tissue origins including bone, cartilage, adipose and muscle, have been isolated from a number of different sources including adult bone marrow aspirates, adipose, synovium and skeletal muscle. TGFß is commonly added to induction media to promote chondrocyte differentiation in vitro. Confirmation of proper chondrogenesis, osteogenesis, and adipogenesis is usually determined by analysis of “end-stage” differentiation markers such as type II collagen and aggrecan for chondrocytes and type I collagen and alkaline phosphatase for osteoblasts. However, it is also necessary to understand the molecular mechanisms that regulate expression of these differentiation markers in order to generate a more complete picture of the intracellular pathways being activated in response to exogenous induction factors in vitro. We have undertaken studies to analyze TF expression as well as end-stage differentiation markers in human adipose-derived stromal cells (HADSCs) and bone marrow-derived stromal cells (BMSCs) during in vitro chondrogenesis, osteogenesis, and adipogenesis at early day (day 3) and late time points (day 28) of differentiation. We examined expression levels of Sox TFs (Sox5, Sox6 and Sox9) that are essential for cartilage formation two osteogenesis-promoting TFs (Runx2 and Osterix) and two adipogenic regulatory nuclear proteins (C/EBP and PPAR-2). Recent studies in collaboration with Dr. Audrey McAlinden indicate that TGF-ß3 chondrogenic induction of HADSCs and BMSCs do not induce an articular chondrocyte population but rather TFs and collagen indicative of a hypertrophic phenotype, and, as such, alternative methods are required to benefit the field of cartilage tissue regeneration (Rich et al., 2008).

To solve the problem of induction of bone-promoting TFs under chondrogenic conditions, we investigated alternative methods of inducing a chondrogenic phenotype in BMSCs that avoided the use of TGFß. This method uses transduction of cells with adenovirus construct of Sox9 and the Sox-9 co-regulatory factor peroxisome proliferator-activated receptor gamma co-activator 1 alpha (PGC-1a). Using this system, we have been able to induce chondrogenesis without hypertrophy.

6. Inhibition of Angiogenesis

Angiogenesis is a natural process in the body that involves the growth of new blood vessels by spouting of preexisting vessels. It is a main process of vascularization during embryonic development, growth, regeneration and wound healing. A substantial and persuasive body of data support the view that angiogenesis plays a crucial role in the invasive tumor growth and metastasis, as well as the control of cancer progression. The microvascular endothelial cells, which are recruited by tumors, has become an important target in tumor therapy. We are interested in the searching of the angiogenesis inhibitors in cartilage and trying to understand the molecular mechanism by which the cartilage remains avascular. We applied several methods to perform angiogenesis assays. Tube formation assay is a in vitro method that is performed by culturing human umbilical vein endothelial cells (HUVEC) in three-dimensional matrigel and quantifying the total length or branch points of the tube formed (Fig A). Rat aortic ring assay uses the culture of aortic rings from 5 to 10-week rat in three-dimensional collagen gel and measurement of the total length of the micro vessel outgrowth (Fig B). Mouse corneal assay is a in vivo assay method that apply implantation of slow release polymer matrix containing angiogenesis activator, label of the micro vessels by fluorescence, and quantification of the fluorescently-labeled micro vessels (Fig C).

angiogenesis

7. Site-1 Protease and Endoplasmic Reticulum Stress

We are interested in the dynamics of protein synthesis by the chondrocytes, the developmental program that allows for the synthesis of the extracellular matrix production that is cartilage. Chondrocytes are secretory cells that like other secretory cells in the body such the differentiating plasma B cells and the cells of exocrine tissues such as the pancreas produce large amounts of protein which mandates the application of special pathways and cellular apparatus allowing for normal matrix production. Very little is however known about these pathways in the chondrocytes. Our recent study of the properties of the matrix in the S1Pcko mutant underscores the importance of endoplasmic reticulum (ER) stress response in the chondrocyte (Journal of Cell Biology, vol. 179, pages 687 -700, 2008). Site-1 Protease (S1P) is a golgi-resident, proprotein convertase that processes endoplasmic reticulum (ER) membrane-bound, latent transcription factors such as SREBPs and ATF6 to their free and active form. While SREBPs play a role in fatty acid and cholesterol homeostasis, ATF6 is involved in ER stress signaling (ERSS) to alleviate ER stress. When we created a cartilage-specific S1P knockout mouse (S1Pcko), we found that the mutant matrix shows a drastic reduction of the collagen type II B (Col IIB) protein, which is a major component of cartilage. Ultrastructural analysis shows that chondrocytes in S1Pcko suffer from ER stress and perhaps an inability to respond to ER stress. Currently we are exploring the nature of ER stress and ER stress response in chondrocytes and its requirement for cartilage production through the use of S1Pcko mice.

The figure shows double-labeled immunofluorescence studies with antibodies to Col IIA (green) and Col IIB (red) demonstrating the abnormal Col IIB fibril deposition. A, D: matrix surrounding early immature chondrocytes in E18.5 humerus; B, E: matrix surrounding mature chondrocytes in E18.5 humerus. Notice the absence of a Col IIB lattice network in the mutant. The arrows show the entrapment of Col IIB inside the cell. Electron microscopy of E15.5 tibial cartilage showing ER stress in S1Pcko mice (F) mice as compared to wild type (C).

8. Genetics of Cartilage Regeneration and Degeneration

Studies have been recently inititated on the genetics of cartilage regeneration and degeneration. In collaboration with Dr. Jim Cheverud, we are studying 10 recombinant inbred lines of large and small mice in addition to the quantitative phenotypes of the F44 generation.

The figure shows the analysis of articular cartilage lesions in selected strains of mice. Representative sagittal sections of full-thickness cartilage defects stained with toluidine blue for proteoglycan are shown.