Immunoselection and adenoviral genetic modulation of human osteoprogenitors: in vivo bone formation on PLA scaffold Daniel Howard, a Kris Partridge, a Xuebin Yang, a Nicholas M.P. Clarke, a Yasunori Okubo, b Kazuhisa Bessho, b Steven M. Howdle, c Kevin M. Shakesheff, d and Richard O.C. Oreffo a, * a University Orthopaedics, University of Southampton, Southampton SO16 6YD, UK b Department of Oral and Maxillofacial Surgery, Graduate School of Medicine, Kyoto University, Sako-ku, Kyoto 606-8507, Japan c School of Chemistry, University of Nottingham, Nottingham NG7 2RD, UK d School of Pharmaceutical Sciences, University of Nottingham, Nottingham NG7 2RD,UK Received 8 October 2002 Abstract The aim of this study was to examine the potential of immunoselected genetically modified human osteoprogenitors to form bone in vivo on porous PLA scaffolds. Human osteoprogenitors from bone marrow were selected using the antibody STRO-1 utilising a magnetically activated cell separation system. The STRO-1 þ fraction isolated 7% of nucleated marrow cells and increased fibro- blastic colony formation by 300% and alkaline phosphatase activity by 190% over unselected marrow cell cultures. To engineer bone tissue, STRO-1 þ culture-expanded cells were transduced with AxCAOBMP-2, an adenovirus carrying the human BMP-2 gene, injected into diffusion chambers containing porous PLA scaffolds, and implanted in vivo. After 11 weeks the presence of bone mineral was observed by X-ray analysis and confirmed for mineral by von Kossa, as well as bone matrix composition by Sirius red staining, birefringence, and type I collagen immunohistochemistry. Bone formation in vivo indicates the potential of using im- munoselected progenitor cells and ex vivo gene transfer with biodegradable scaffolds, for the development of protocols for the treatment of a wide variety of musculo-skeletal disorders. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: Osteoprogenitor; Mesenchymal stem cell; Immunoselection; STRO-1 monoclonal antibody; Adenoviral gene-transfer; Bone morpho- genetic protein-2; Biodegradable polymer; Poly(D,L-lactic acid); Tissue engineering The development of new bone formation strategies has important clinical therapeutic implications in a va- riety of musculoskeletal injuries and diseases. Currently, bone formation stimulating regimes that achieve signif- icant increases in bone density have yet to yield func- tional and mechanically competent bone. Autografts and allografts have been used extensively in bone repair, autografts, taken typically from the iliac crest, are con- strained by anatomical limitations, and are associated with donor-site morbidity, whilst allografts are limited by the potential risks of rejection and infection. In the current study, we have sought to apply tissue engineer- ing principles to the development of bone augmentation strategies through: (i) isolation and expansion of a de- fined osteoprogenitor population, (ii) gene transduction using osteoinductive Bone Morphogenetic Protein-2 (BMP-2), and (iii) appropriately designed biodegradable scaffolding to provide a 3D growth environment. Stem cells from marrow stroma have been shown to differentiate into osteogenic as well as chondrogenic and adipogenic cells [4,9,19,24,36]. The potential to isolate and expand these cells while retaining their dif- ferentiation ability has been well documented [5,18,20,49]. Initial studies from Simmons and Torok- Storb (1991) showed that the heterogeneity of the stromal cell population could be reduced by isolation using the monoclonal antibody STRO-1, which recog- nises a trypsin-resistant cell surface antigen present on a Biochemical and Biophysical Research Communications 299 (2002) 208–215 www.academicpress.com BBRC * Corresponding author. Fax: +44-2380-796141. E-mail address: roco@soton.ac.uk (R.O.C. Oreffo). 0006-291X/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII:S0006-291X(02)02561-5