A water-soluble fraction from adult bone stimulates the differentiation of cartilage in explants of embryonic muscle

Effects of TGF-β1 and VEGF-A transgenes on the osteogenic potential of bone marrow stromal cells in vitro and in vivo

An exogenous supply of growth factors and bioreplaceable scaffolds may help bone regeneration. The aim of this study was to examine the effects of TGF-β1 and VEGF-A transgenes on the osteogenic potential of bone marrow stromal cells. Rat bone marrow stromal cells were transfected with plasmids encoding mouse TGF-β1 and/or VEGF-A complementary DNAs and cultured for up to 28 days. Furthermore, collagen scaffolds carrying combinations of the plasmids-transfected cells were implanted subcutaneously in rats. The transgenes increased alkaline phosphatase activity, enhanced mineralized nodule formation, and elevated osteogenic gene expressions in vitro. In vivo, messenger RNA expression of osteogenic genes such as BMPs and Runx2 elevated higher by the transgenes. The data indicate that exogenous TGF-β1 and VEGF-A acted synergistically and could induce osteoblastic differentiation of bone marrow stromal cells in both cell culture and an animal model. The results may provide valuable information to optimize protocols for transgene-and-cell-based tissue engineering.

Review: mesenchymal stem cells: cell-based reconstructive therapy in orthopedics

Adult stem cells provide replacement and repair descendants for normal turnover or injured tissues. These cells have been isolated and expanded in culture, and their use for therapeutic strategies requires technologies not yet perfected. In the 1970s, the embryonic chick limb bud mesenchymal cell culture system provided data on the differentiation of cartilage, bone, and muscle. In the 1980s, we used this limb bud cell system as an assay for the purification of inductive factors in bone. In the 1990s, we used the expertise gained with embryonic mesenchymal progenitor cells in culture to develop the technology for isolating, expanding, and preserving the stem cell capacity of adult bone marrow-derived mesenchymal stem cells (MSCs). The 1990s brought us into the new field of tissue engineering, where we used MSCs with site-specific delivery vehicles to repair cartilage, bone, tendon, marrow stroma, muscle, and other connective tissues. In the beginning of the 21st century, we have made substantial advances: the most important is the development of a cell-coating technology, called painting, that allows us to introduce informational proteins to the outer surface of cells. These paints can serve as targeting addresses to specifically dock MSCs or other reparative cells to unique tissue addresses. The scientific and clinical challenge remains: to perfect cell-based tissue-engineering protocols to utilize the body's own rejuvenation capabilities by managing surgical implantations of scaffolds, bioactive factors, and reparative cells to regenerate damaged or diseased skeletal tissues.

Chondroinduction of mouse mesenchymal stem cells in three-dimensional highly porous matrix scaffolds

Porous polyvinyl formal (PVF) resin and poly(lactide-caprolactone) [P(LA/CL)] sponges were examined as three-dimensional matrices for chondroinduction of cultured bone marrow mesenchymal stem cells (MSCs). Approximately 5 x 10(5) mouse MSCs were seeded in porous PVF resin or P(LA/CL) sponges and were cultured for up to 1 month in serum-free high-glucose Dulbecco's modified Eagle's medium containing 10 ng/mL transforming growth factor-beta3 and 100 nM dexamethasone for chondroinduction. After the 1-month culture period, the PVF resin and P(LA/CL) sponges contained approximately twice the amount of glycosaminoglycans compared with the control pellet. Safranin-O staining of PVF and P(LA/CL) after 1 month of culture revealed a cartilage-like extracellular matrix containing glycosaminoglycans and collagen. When implanted into nude mice, PVF and P(LA/CL) seeded with MSCs were found to be both biocompatible and chondroinductive. These highly porous scaffolds can maintain a large number of cells in a three-dimensional structure. Both are potentially promising for the chondroinduction of bone marrow MSCs for research and clinical applications.

Osteoprogenitor cells of mature human skeletal muscle tissue: an in vitro study

The presence of osteogenic progenitors in human skeletal muscle is suggested by the formation of ectopic bone in clinical and experimental conditions, but their direct identification has not yet been demonstrated. The aims of this study were to identify osteogenic progenitor cells in human skeletal muscle tissue and to expand and characterize them in culture. Specimens of gracilis and semitendinosus muscle were obtained from young adults and digested to separate the connective tissue and satellite cell fractions. The cells were cultured and characterized morphologically and immunohistochemically using antibodies known to be reactive with primitive osteoprogenitor cells, pericytes, intermediate filaments, and endothelial cells. Alkaline phosphatase activity and osteocalcin gene expression were also determined. In the early stages of culture, the connective tissue cells obtained were highly positive for primitive osteoprogenitor cell and for pericyte markers. Alkaline phosphatase activity was detectable at early stages of culture and rose as a function of time, whereas primitive osteoprogenitor cell markers declined and osteocalcin mRNA expression became detectable by reverse transcriptase-polymerase chain reaction (RT-PCR). It is shown that human skeletal muscle connective tissue contains osteogenic progenitor cells. Their identification as pericytes, perivascular cells with established osteogenic potential, suggests a cellular link between angiogenesis and bone formation in muscle tissue. These cells are easily cultured and expanded in vitro by standard techniques, providing an alternative source of osteogenic progenitor cells for possible cell-based therapeutic use in certain conditions.

Osteochondral progenitor cells in acute and chronic canine nonunions

This study examined the ability of cells isolated from early healing segmental defects and from tissue from chronic nonunions to support bone and cartilage formation in vivo and their response to transforming growth factor-beta1 in vitro. Ostectomies (3 mm) were created in the radial diaphysis of four dogs. The dogs were splinted 3-5 days postoperatively and then allowed to bear full weight. At 7 days, tissue in the defect was removed and any periosteum was discarded; cells in the defect tissue were released by enzymatic digestion. The dogs were splinted again and allowed to bear full weight for 12 weeks. Radiographs confirmed a persistent nonunion in each dog. Defect tissue was again removed, any periosteum was discarded, and cells were isolated. Cells were also obtained from the defect tissue by nonenzymatic means with use of explant cultures. One-half of the tissue and one-half of any preconfluent, first-passage cultures were shipped to Cleveland by overnight carrier. At second passage, cells were loaded into ceramic cubes and implanted into immunocompromised mice for 3 or 6 weeks. Harvested cubes were examined histologically for cartilage and bone with use of a semiquantitative scoring system. Confluent fourth-passage cultures of 7 and 84-day defect tissue cells were cultured with 0.03-0.88 ng/ml transforming growth factor-beta1 for 24 hours, and [3H]thymidine incorporation and alkaline phosphatase specific activity were determined. Donor-dependent differences were noted in the rate at which defect cells achieved confluence; in general, cells from 7-day tissue divided most rapidly. Seven-day defect cells formed less bone and at a slower rate than was seen in the ceramic cubes containing samples from day 84. Cells derived enzymatically behaved similarly to those from explant cultures. Ceramic cubes contained fibrous connective tissue, cartilage, bone, and fat, indicating that multipotent cells were present. Stimulation of [3H]thymidine incorporation in response to transforming growth factor-beta1 was donor dependent and variable; only two of six separate isolates of cells exposed to it had measurable alkaline phosphatase activity (which was relatively low), and none of the cultures exhibited an increase in response to transforming growth factor-beta1 for 24 hours. This indicates that mesenchymal progenitor cells are present in the healing defect tissue at 7 and 84 days and that the relative proportion of osteochondroprogenitor cells is greater at the later time. The response to transforming growth factor-beta1 is typical of multipotent mesenchymal cells but not of committed chondrocytes or osteoblasts, indicating that these committed and differentiated cells are not present in early stages of healing and suggesting that their differentiation is inhibited in chronic nonunion.

Osteoprogenitor cells as targets for ex vivo gene transfer

We transduced osteoprogenitor cells with recombinant retrovirus and analyzed proviral integration patterns into chromosomal DNA to detect for the first time the clonal and cellular fate of osteoprogenitor-derived progeny cells. Metaphyseal bone cells and diaphyseal stromal cells were isolated from the distal femurs of young rats, transduced with the vM5neolacZ recombinant retrovirus, and selected in the neomycin analog, G418. Following surgical marrow ablation of a femur in one leg of mature rats, retroviral-transduced metaphyseal or diaphyseal cells were injected into the ablated site. These rats were killed 5-6 days later. Metaphyseal and diaphyseal cells were isolated from distal femurs, selected in G418, and stained for beta-galactosidase (beta-gal+). The number and clonal origin of transduced progenitor cells were determined. High numbers of beta-galactosidase colonies with an osteoblast phenotype were obtained following metaphyseal transplants and detected in 100% of metaphyseal and none of diaphyseal specimens. In contrast, beta-galactosidase colonies derived from diaphyseal transplants were detected in 50% of specimens in both the metaphysis and diaphysis, and the absolute number of progenitor cell colonies was 60-fold less than metaphyseal transplants. Provirus was only detected in the ablated bones and not in the contralateral bone or other tissues. Proviral integration fragment analysis showed a single integration site for recovered metaphyseal cell clones, consistent with their origination from a common single progenitor. This is one of the first demonstrations of successful transplantation of clonal osteoprogenitors to their site of origin in bone. It may be possible to use these cells to target genes to bone for therapeutic use in skeletal and hematopoietic diseases.

[Osteoinduction and -reparation]

Traumata, diseases, developmental deformities, and tumor resections frequently cause bone defects and atrophies. In general, three different mechanisms exist by which bone restoration can be achieved: (1) osteogenesis initiated by vital, osteoblastic cells of autografts; (2) osteoconduction (or creeping substitution); and (3) osteoinduction. The latter mechanism means the differentiation of pluripotent, mesenchymal-type cells (located in a recipient bed with strong regenerative capacity) into cartilage- and bone-forming progenitor cells under the influence of inductive bone morphogenetic proteins (BMPs). Some BMPs are physiologically included in low concentrations as organic components in bone tissue. They can diffuse from demineralized bone implants into the recipient bed and induce a differentiation into new bone tissue. Nine different BMPs have been isolated, characterized, and cloned. Some of these possess inductive properties and can initiate new bone formation in muscle tissue or in bone defects. In the future recombinant BMPs will be available in unlimited quantities. This will lead to completely new therapeutic concepts in reconstructive bone surgery.

Repair of calvarial defects with flap tissue: role of bone morphogenetic proteins and competent responding tissues

Bone morphogenetic proteins 2 through 8 have the ability to induce the in vivo transformation of extraskeletal mesenchymal tissue into bone. The aims of this investigation were to determine the optimal responding tissue and the specificity of the inductive effect of bone morphogenetic protein 3. The optimal responding tissue was found to be skeletal muscle. The specificity of this response to bone morphogenetic protein 3 was compared with that of recombinant human basic fibroblast growth factor, recombinant platelet-derived growth factor, and recombinant insulin-like growth factor. Bone morphogenetic protein 3 was the only factor that induced de novo bone formation. This ability to transform muscle into bone was tested in 7 x 7 mm irradiated skull defects in the rat. After 1500 rads of exposure, these defects showed no significant signs of healing by 8 months. When these defects were treated with the microvascular transfer of a nonirradiated muscle flap, they had 8 percent healing at 4 months and 37 percent healing by 8 months. Defects treated with 30 micrograms bone morphogenetic protein 3 (without the muscle flap) achieved 50 percent healing by 4 months and 64 percent healing by 8 months. When the defects were treated with both the muscle flap and bone morphogenetic protein 3, there was 96 percent healing by 4 months and 100 percent healing by 8 months (p < 0.015, compared with bone morphogenetic protein 3 alone at both time points). At 8 months, the transplanted muscle was entirely transformed into bone and healed the skull defect with newly generated bone indistinguishable from the surrounding calvarial tissue. These findings suggest a potential clinical utility of bone morphogenetic protein 3-induced bone formation in skeletal reconstructions. Furthermore, they also show that there is a collaborative requirement for both the osteoinductive factor bone morphogenetic protein 3 and the presence of competent responsive cells in the well-perfused muscle.

Pluripotent mesenchymal stem cells reside within avian connective tissue matrices

Recent studies have noted the presence of putative stem cells derived from the connective tissues associated with skeletal muscle, heart, and dermis. Long-term continuous cultures of these cells from each tissue demonstrated five distinct phenotypes of mesodermal origin, i.e. muscle, fat, cartilage, bone, and connective tissue. Clonal analysis was performed to determine whether these morphologies were the result of a mixed population of lineage-committed stem cells or the differentiation of pluripotent stem cells or both. Putative stem cells from four tissues (skeletal muscle, dermis, atria, and ventricle) were isolated and cloned. Combined, 1158 clones were generated from the initial cloning and two subsequent subclonings. Plating efficiency approximated 5.8%. Approximately 70% of the 1158 clones displayed a pure stellate morphology, while the remaining clones contained a mixture of stellate, chondrogenic- or osteogenic-like morphologies or both. When cultured in the presence of dexamethasone, cells from all clones differentiated in a time- and concentration-dependent manner into muscle, fat, cartilage, and bone. These results suggest that pluripotent mesenchymal stem cells are present within the connective tissues of skeletal muscle, dermis, and heart and may prove useful for studies concerning the regulation of stem cell differentiation, wound healing, and tissue restoration, replacement and repair.

Bone induction by demineralized bone particles: long-term observations of the implant-connective tissue interface

Allogeneic demineralized bone (DB) powder was applied to subcutaneous pockets and cranial defects of rats and histological, histomorphometrical, and radiological evaluation was performed one to 52 weeks after implantation. In both type of implants cartilage formation was observed after 1 week between DB particles and in former vascular channels and cervices within the particles. Foci of bone formation were observed after 2 weeks in the center of the implant. Remineralization of DB particles only occurred in close contact with the new bone tissue by which they became incorporated, indicating that remineralization of DB is a continuation of a process which has started in live bone. At all times, the interface between implant and connective tissue of the host consisted of a layer of non-remineralized DB particles embedded in fibrous connective tissue. This rim was thinner at the dura-mater-side in the cranial implants, compared with the skin-side of the same implants as well as all sides of the subcutaneous implants. The rim became thinner with time but never disappeared completely. The outer contour of the mineralized implant sometimes ran right through a DB particle. A critical concentration of bone-inducing agent leaking from DB particles seems to be necessary for bone induction. We suggest that at the interface of the implant and loose connective tissue this concentration is never reached, which results in maintenance of the rim.

Mesenchymal stem cells

Bone and cartilage formation in the embryo and repair and turnover in the adult involve the progeny of a small number of cells called mesenchymal stem cells. These cells divide, and their progeny become committed to a specific and distinctive phenotypic pathway, a lineage with discrete steps and, finally, end-stage cells involved with fabrication of a unique tissue type, e.g., cartilage or bone. Local cuing (extrinsic factors) and the genomic potential (intrinsic factors) interact at each lineage step to control the rate and characteristic phenotype of the cells in the emerging tissue. The study of these mesenchymal stem cells, whether isolated from embryos or adults, provides the basis for the emergence of a new therapeutic technology of self-cell repair. The isolation, mitotic expansion, and site-directed delivery of autologous stem cells can govern the rapid and specific repair of skeletal tissues.

Ectopic induction of cartilage and bone by water-soluble proteins from bovine bone using a polyanhydride delivery vehicle

Controlled release delivery vehicles for water-soluble osteogenic proteins from demineralized bovine bone matrix were constructed using polyanhydride polymers. The water-soluble proteins were isolated from a 4 M guanidine hydrochloride extract of bone matrix. The water-soluble proteins possessed Chondrogenic Stimulating Activity (CSA) when tested in stage 24 chick limb bud cell cultures, but were incapable of inducing cartilage or bone in vivo when implanted intramuscularly into mice by themselves. The polyanhydride polymers alone were also incapable of inducing ectopic cartilage or bone. However, when the water-soluble proteins were incorporated into the polymeric delivery vehicle, the combination was capable of inducing cartilage and bone up to 50% of the time. These results demonstrate that it is possible to use polyanhydride polymers as controlled-release delivery vehicles for soluble bioactive factors that interact with a local cell population.

Ectopic induction of cartilage and bone by water-soluble proteins from bovine bone using a collagenous delivery vehicle

A controlled-release delivery vehicle for water-soluble osteogenic proteins from demineralized bone matrix was constructed using purified type I collagen. The water-soluble proteins were isolated from a 4 M GdnHCl extract of bone matrix. Although the water-soluble proteins were capable of inducing cartilage formation in vitro, they were incapable of inducing cartilage or bone in vivo when implanted intramuscularly into mice in the absence of an appropriate delivery vehicle. The collagen-based delivery vehicle alone was also incapable of inducing osteogenesis in vivo. However, when the water-soluble proteins were incorporated into the delivery vehicle, the combination was capable of inducing cartilage and bone 76% of the time. These results demonstrate that it is possible to formulate a controlled-release delivery vehicles for soluble bioactive factors which upon release interact with local responsive cells.