MT1-MMP and Type II Collagen Specify Skeletal Stem Cells and Their Bone and Cartilage Progeny. Based on our characterization of MT1-MMP deficiency in skeletal development, we have identified that MT1-MMP is expressed in both the cartilage and bone compartment of the skeleton where it is an important collagenase required for cell function. The severe postnatal developmental defects observed in MT1-MMP deficient mice are therefore a compound phenotype caused by a defect in several cell populations, yet the severity of the disease associated with loss of the gene function makes analysis of a tissue specific function complicated, if not impossible, in a background with complete loss of MT1-MMP function. To overcome this obstacle, we undertook selective reconstitution of the MT1-MMP expression in one tissue (cartilage) to assess the role of expression in that particular tissue. The selection of cartilage was based, in part, upon our finding of a previously uncharacterized program for cartilage remodeling distinct from the osteoclast dependent pathway found in endochondral ossification. This non-endochondral ossification process is strictly dependent on MT1-MMP for apparent chondrocytes-specific matrix remodeling and particularly prominent in craniofacial cartilage remodeling. For selective reconstitution, we generated a mouse MT1-MMP transgene under control of the type II collagen promoter/enhancer. Mice carrying this construct were generated and expression of mRNA and protein was verified before the transgene was bred into an MT1-MMP deficient background. The resulting mice were deficient for MT1-MMP in all other cells than those expressing type II collagen (henceforth MT1-MMP-/-;T+), and displayed a marked improvement in survival with rescue of the 33% pre-weaning demise normally associated with MT1-MMP deficiency. This observation in itself provided significant information about the role of MT1-MMP in development, by demonstrating that reconstitution of MT1-MMP in a part of the skeleton was sufficient to reduce mortality and thus highlighting the importance of skeletal development for survival. Interestingly the most prominent finding in MT1-MMP-/-;T+ mice is the increase in bone formation observed in both craniofacial bones and long bones. To reconcile this finding with the expression of type II collagen, which ordinarily is considered a cartilage-specific collagen, we analyzed wildtype mouse tissue for bone-associated type II collagen expression. We documented abundant expression of type II collagen in periosteal tissues of wildtype mice. In light of this, we discounted the notion that MT1-MMP-/-;T+ mice formed bone due to ectopic transgene expression, rather the expression of the transgene reflected the true expression of endogenous type II collagen and thereby endowed the bone cells with MT1-MMP activity which is essential for osteogenic cell function. To reproduce this finding in an in vitro/in vivo experimental setting, we isolated bone marrow stromal cells (BMSCs) from wildtype mice, MT1-MMP deficient littermates and MT1-MMP-/-;T+ siblings. First, we demonstrated that wildtype BMSCs expressed type II collagen and MT1-MMP as suggested and we next evaluated the ability of the cells to generate bone using an in vivo transplantation assay. In this assay, wildtype cells generated abundant bone, fat and stroma supporting myelopoiesis, which is considered the hallmark of skeletal stem cells. MT1-MMP-deficient cells on the other hand only generated scant amounts of woven bone without adipogenesis and myelopoiesis and as such fail the test of stem cell characteristics. Cells derived from MT1-MMP-/-;T+ mice, in contrast, were indistinguishable from wildtype cells, demonstrating the requirement for MT1-MMP expression in osteogenic cells and retention of skeletal stem cell properties. Based on these findings we assign the expression of type II collagen and MT1-MMP to skeletal stem cells and demonstrate that type II collagen is expressed in at least a subset of skeletal stem cells in vivo. Additionally, we observed that cartilage is not susceptible to MT1-MMP mediated proteolysis under any circumstance. Indeed, despite our intentions to characterize cartilage dissolution we did not observe spontaneous loss of cartilage matrix in MT1-MMP;T+ mice. Our interpretation of this observation, combined with our other work, suggests that cartilage is only susceptible to MT1-MMP activity under special conditions when inhibitor levels, matrix synthesis rate and proteoglycan content are permissive.
Showing the most recent 10 out of 20 publications
