97 Ann. N.Y. Acad. Sci. 1049: 97���106 (2005). �� 2005 New York Academy of Sciences. doi: 10.1196/annals.1334.010 How Wnt Signaling Affects Bone Repair by Mesenchymal Stem Cells from the Bone Marrow CARL A. GREGORY, WILLIAM G. GUNN, EMIGDIO REYES, ANGELA J. SMOLARZ, JAMES MUNOZ, JEFFREY L. SPEES, AND DARWIN J. PROCKOP Center for Gene Therapy, Tulane University Health Sciences Center, New Orleans. Louisiana 70112, USA ABSTRACT: Human mesenchymal stem cells (hMSCs) from bone marrow are a source of osteoblast progenitors in vivo, and under appropriate conditions they differentiate into osteoblasts ex vivo. The cells provide a convenient cell culture model for the study of osteogenic tissue repair in an experimentally accessible system. Recent advances in the field of skeletal development and osteogenesis have demonstrated that signaling through the canonical wingless (Wnt) path- way is critical for the differentiation of progenitor cell lines into osteoblasts. In- hibition of such signals can predispose hMSCs to cell cycle entry and prevent osteogenesis. Our investigation of the role of Wnt signaling in osteogenesis by hMSCs ex vivo has demonstrated that osteogenesis proceeds in response to bone morphogenic protein 2 stimulation and is sustained by Wnt signaling. In the presence of Dkk-1, an inhibitor of Wnt signaling, the cascade is disrupted, resulting in inhibition of osteogenesis. Peptide mapping studies have provided peptide Dkk-1 agonists and the opportunity for the production of blocking an- tibodies. Anti-Dkk-1 strategies are clinically relevant since high serum levels of Dkk-1 are thought to contribute to osteolytic lesion formation in multiple my- eloma and possibly some forms of osteosarcoma. Specific inhibitors of glycogen synthetase kinase 3 (GSK3 ), which mimic Wnt signaling, may also have a therapeutic benefit by enhancing in vitro osteogenesis despite the presence of Dkk-1. Antibodies that block Dkk-1 and GSK3 inhibitors may provide novel opportunities for the enhancement of bone repair in a variety of human diseas- es such as multiple myeloma and osteosarcoma. KEYWORDS: mesenchymal stem cells Wnt Dkk-1 bone osteogenesis cancer MESENCHYMAL STEM CELLS FROM BONE MARROW The first non-hematopoietic mesenchymal stem cells (MSCs) were discovered by Friedenstein,1 who described clonal, plastic adherent cells from bone marrow capa- ble of differentiating into osteoblasts, adipocytes, and chondrocytes.2���5 These cells Address for correspondence: Carl A. Gregory, Center for Gene Therapy, Tulane University Health Sciences Center, 1430 Tulane Avenue, New Orleans, LA 70112. Voice: 504-988-7716 fax: 504-988-7710. ca_gregory@hotmail.com

98 ANNALS NEW YORK ACADEMY OF SCIENCES were later found to differentiate into ���stromal��� cells, structural components of the bone marrow that supported ex vivo culture of hematopoiesis by providing extracel- lular matrix components, cytokines, and growth factors.6���10 Numerous laboratories around the world have now demonstrated that multipotent MSCs can be recovered from a variety of adult tissues and can differentiate into various tissue lineages. In particular, Verfaille and colleagues report that a specific type of murine MSC isolat- ed from bone marrow, muscle or brain termed multipotential adult progenitor cells (MAPCs) differentiate into a variety of tissue lineages including myoblasts, hepato- cytes, and even neural tissue.11���14 Although it is clear that single cell���derived colonies of MSCs can transdifferen- tiate into multiple tissue lineages in culture, there have been indications that they can also undergo cell fusion. Evidence supporting the case for cell fusion by MSCs comes from a set of experiments conducted by Spees and colleagues15 in which MSCs, labeled with green fluorescent protein, were co-cultured with pulmonary small airway epithelial cells (SAECs) after a brief heat shock to mimic tissue dam- age. After heat shock, a good proportion of the confluent SAEC monolayer died by apoptosis. When introduced to the damaged SAEC cultures, the green MSCs ad- hered to the tissue culture plastic, resulting in a confluent mixture of cells. Over a few days, some of the MSCs differentiated into epithelial-like cells but others fused with SAECs, producing chimeras. These investigators suggested that cell fusion may be an acute response to tissue damage and may rescue dying cells by supplying ad- ditional cytosolic metabolites and organellar components. Because multipotent MSCs are easily expanded in culture, there has been much interest in their clinical potential for tissue repair and gene therapy.16 As a result, nu- merous studies have been carried out demonstrating the migration and multi-organ en- graftment potential of MSCs in animal models and in human clinical trials.13,17���25 One trial in particular utilized hMSCs from compatible healthy donors to treat indi- viduals with the brittle bone disease osteogenesis imperfecta, yielding encouraging results.18���20 In addition to their ability to home, engraft, and differentiate into damaged tis- sues, MSCs secrete cytokines and trophic factors that provide beneficial effects to surrounding tissue. This characteristic has been exploited in the field of regenerative neuroscience, where MSCs, when injected into the brain or damaged spinal cord, en- graft and provide a permissive stromal microenvironment for the growth and repair of existing neural tissue.22,26���29 In our laboratory, MSCs have been shown to ex- press and secrete a variety of neurotrophic and neuroprotective proteins in culture and when implanted into the brains of test animals (Munoz et al., unpublished ma- terial). Expression of these beneficial factors persists from stably engrafted MSCs for a number of weeks post implantation, and it is hoped that this discovery will lead to a promising treatment for neural lesions caused by ischemia and spinal cord inju- ry. Although the MSCs secrete their own extensive repertoire of cytokines, MSCs can be genetically modified in a stable manner by a variety of retroviral and lentiviral vectors and therefore could be induced to secrete agents at sites of tissue injury or tumors. It is noteworthy that MSCs have been shown to preferentially migrate to tu- mors and, when genetically modified to secrete interferon-��, they reduce the size of the lesions.36 Graft and host compatibility is always an issue when transplanting heterologous tissue into an immunocompetent host, and immunotolerance of implanted MSCs has