作者
Dmitry A. Ovchinnikov,Jian Min Deng,George Ogunrinu,Richard R. Behringer
摘要
Type II collagen is one of the principal markers of chondrocyte differentiation (Mayne, 1990) and is encoded by the proa1(II) collagen gene (Col2a1) (Cheah et al., 1985). In the mouse, Col2a1 transcripts are first detected at 9.5 days post-coitum (dpc) in the sclerotome of the differentiating somites and the cranial mesenchyme destined to give rise to the cartilage. Expression is also observed in the notochord and later in some neural tissues, including regions of the rhombencephalic basal plate and the ventricular layer of the hindbrain (Cheah et al., 1991). The regulatory elements that direct chondrocyte-specific expression in tissue culture cells and transgenic mice are located within the first intron of the Col2a1 gene (Mukhopadhyay et al., 1995; Zhou et al., 1995). To direct the expression of Cre recombinase to developing chondrocytes, we generated transgenic mice expressing a Col2a1-Cre gene construct (Fig. 1). The gene construct consisted of 3 kb of the Col2a1 promoter region, the first exon with a mutated initiation codon, and a 3.02 kb fragment of intron 1 ligated to a splice acceptor sequence (Zhou et al., 1995) followed by an internal ribosome-entry site (IRES), Cre recombinase coding region, and the SV40 large T antigen polyadenylation signal. The 8.4 kb gene construct was purified from vector sequences and microinjected into the pronuclei of fertilized C57BL/6 3 SJL F2 hybrid eggs to generate transgenic mice (Brinster et al., 1985). To analyze the expression pattern of Cre recombinase in these Col2-Cre transgenic mice, we utilized the ROSA26 Cre reporter mouse strain, R26R (Soriano, 1999). Col2-Cre transgenic males from line A were bred with R26R/1 females to establish timed matings. Embryos were stained with X-gal to detect b-galactosidase (b-gal) activity (Hogan et al., 1994). Reciprocal crosses, using Col2-Cre females, yielded identical patterns. Another Col2-Cre transgenic mouse line (B) demonstrated essentially identical patterns of Cre activity. b-gal activity was first detected between 8.75 and 9.0 dpc in the notochord and cranial mesenchyme of Col2Cre, R26R compound heterozygotes (Fig. 2A). In somites, b-gal activity was first detected at 9.5 dpc in the sclerotomes (Fig. 2b). b-gal activity was strongest in the posterior portion of the sclerotome. Additionally, strong b-gal activity was observed in the otic vesicle region. By 11.5–12.0 dpc, intense b-gal activity was observed in the notochord and the surrounding sclerotomal cells of the vertebral anlagen undergoing chondrocytic differentiation (Fig. 2C,D). In the limb buds, b-gal activity was observed in the developing cartilaginous anlagen of the long bones (Fig. 2C,D). At 15 dpc, b-gal activity was detected in virtually all of the existing cartilaginous primordia of the bones of the axial and appendicular skeleton, temporal and basioccipital bones, and the other elements of the base of the skull developing by endochondral bone formation. b-gal activity was also observed in the submandibular glands (Fig. 2E). Osteoblasts are also derived from the sclerotome of somites (Aubin, 1998). Therefore, Col2a1-directed expression of Cre may also lead to the activation of the R26R locus in the osteoblast lineage. To address this question, we analyzed the long bones of Col2-Cre mice for Cre activity. Longitudinal sections of the X-gal stained hindlimbs of neonatal Col2-Cre/R26R compound heterozygotes show specific staining in the chondrocytes of the epiphysis of the bone but not in the osteoblasts or perichondrial fibroblasts (Fig. 2F,G). Some mosaicism in b-gal activity was observed in the cartilage, with approximately 5% of chondrocytes being b-gal negative. The Col2-Cre transgenic mice described here should be a useful resource for analysis of gene function, employing conditional genetics approaches in differentiating chondrocytes, notochord, and submandibular glands.