DEVELOPMENTAL DYNAMICS 218:394 – 400 (2000) BRIEF COMMUNICATION Col2-GFP Reporter Marks Chondrocyte Lineage and Chondrogenesis During Mouse Skeletal Development T. DAWN GRANT,1,2 JAY CHO,1,2 KILEY S. ARIAIL,1 NICOLE B. WEKSLER,1 RANDALL W. SMITH,3 1,2* AND WILLIAM A. HORTON 1 Research Center, Shriners Hospital for Children, Portland, Oregon 2 Department of Molecular and Medical Genetics, Oregon Health Sciences University, Portland, Oregon 3 Department of Pathology, Oregon Health Sciences University, Portland, Oregon ABSTRACT Mice were generated in which a Col2-GFP transgene serves as a reporter for the chondrocyte lineage and for chondrogenesis in live embryos and newborn pups. Cells actively engaged in chondrogenesis were identified by confocal optical sectioning within their native environments in embryos and in thick tissue slices. Chondrocytes exhibiting GFP fluorescence were purified from rib cages by high-speed cell sorting of crude cell suspensions. Intensity of fluorescence correlated with biosynthesis of procollagen II in these cells. The use of these mice and their cells provides a novel approach for studying chondrocyte differentiation and chondrogenesis during skeletal development. Dev Dyn 2000;218:394 – 400. © 2000 Wiley-Liss, Inc. Key words: transgenic mice; reporter mice; green fluorescent protein; type II collagen; chondrocytes; chondrogenesis; skeletal development; confocal microscopy; cell sorting INTRODUCTION Chondrocytes play a key role in formation and maintenance of the vertebrate skeleton (Tickle and Eichele, 1994; Erlebacher et al., 1995; Johnson and Tabin, 1997; Beier et al., 1999; Neame et al., 1999; St-Jacques et al., 1999; Stevens and Williams, 1999). They differentiate in response to signals that determine where the future skeleton will form. They proliferate, terminally differentiate and die in the growth plate in response to signals that regulate linear bone growth. They synthesize and maintain cartilage matrices that serve as templates for endochondral ossification, account for the functional properties of permanent cartilages and contribute to the repair of cartilage injuries. In short, chondrocytes exhibit a number of adaptive behaviors in response to changing environments in skeletal tissues. Despite their importance to skeletal development and homeostasis, chondrocyte behaviors are difficult to study because the microenvironments responsible for them are difficult to reproduce experimentally. Indeed, © 2000 WILEY-LISS, INC. loss of typical chondrocyte characteristics, i.e., dedifferentiation, has been an obstacle for study of chondrocytes, especially mammalian chondrocytes, in vitro (Benya et al., 1978; Benya and Shaffer, 1982; Aulthouse et al., 1989; Ramdi et al., 1993). Even if chondrocyte phenotypes can be produced in vitro, it is difficult to be certain that in vitro behaviors accurately reflect in vivo behaviors given the substantial differences in microenvironment. Consequently, investigators often turn to in situ approaches in which experiments are carried out in living animals or in organ cultures that maintain native environments. These tissues are examined after fixation with probes that can be detected microscopically. Such analyses provide snapshots of relevant behaviors, but dynamic aspects of chondrocyte behavior can not be studied directly because preparation of tissues for conventional microscopy kills cells. We have developed an approach to study chondrocyte biology during skeletal development that allows analysis of live cells within tissues with the potential for real-time studies of adaptive behaviors relevant to skeletal biology. Transgenic mice were generated in which expression of enhanced Green Fluorescent Protein (GFP) is controlled by the murine type II collagen promoter/enhancer (Col2-GFP) to produce a fluorescent reporter for chondrocytes. Expression of type II collagen is widely accepted as a chondrocyte marker and its promoter/enhancer has been well characterized (von der Mark, 1980; Horton et al., 1987; Mendler et al., 1989; Vikkula et al., 1992; Ala-Kokko et al., 1995; Metsaranta et al., 1995). GFP has been utilized as a lineage marker in many cell types (Doevendans et al., 1996; Zhuo et al., 1997; Chen et al., 1998; Fleischmann et al., 1998; van den Pol and Ghosh, 1998; Minegishi et Grant sponsor: Shriners Hospitals for Children; Grant number: 8650. *Correspondence to: William A. Horton, M.D., Research Center, Shriners Hospital for Children, 3101 S.W. Sam Jackson Park Road, Portland, OR 97201. E-mail: firstname.lastname@example.org Received 21 December 1999; Accepted 17 February 2000 Col2-GFP REPORTER MICE Fig. 1. Composite images of cartilaginous skeletons of live embryos transgenic for Col2-GFP (A, E14.5; B, E17.5). Many structures that exhibit fluorescence at E14.5 no longer fluoresce at E17.5, including 395 diaphyses of radius and ulna (arrow) and femur and fibula (double arrows), posterior ribs and vertebral bodies (arrowhead) except for those in the tail. Double arrowheads mark external ear in A. Scale bars ⫽ 1 mm. Fig. 2. Distal femur from transgenic embryo (E 17.5) imaged by three different modes. A is a paraffin section stained with alcian blue and H&E. B is a paraffin section immunostained with antibody to collagen II. C is a thick section imaged by confocal microscopy. Note that fluorescence is restricted to matrix in B and to cells in C. Scale bars ⫽ 200 . al., 1999; Ono et al., 1999), but not for chondrocytes to our knowledge. Previous studies have utilized ␤-galactosidase (␤gal) as a reporter protein for the Col2 promoter. The present study documents that Col2-GFP protein identifies chondrocytes throughout development in a pattern comparable to that of Col2-␤gal reporter protein (Metsaranta et al., 1995). However, the Col2-GFP reporter system is more versatile since it can be visualized directly in living tissues by conventional fluorescence or by laser-scanning confocal microscopy. The latter technique minimizes background fluorescence and offers a potential means to investigate chondrocytic behaviors relevant to skeletal development in intact tissues and in real time. GFP fluorescence provides an in situ assay of type II collagen synthesis during processes, such as bone growth and cartilage repair. Finally, the presence of the Col2-GFP protein also provides an endogenous label for high-speed cell sorting that allows quick and efficient isolation of chondrocytes for in vitro studies. We report here the generation of the transgenic mouse, document the expression of the reporter during mouse development, correlate GFP fluorescence with biosynthesis of type II collagen and describe a protocol developed to isolate chondrocytes by cell sorting. RESULTS AND DISCUSSION Transgenic Mice Eight founder mice were positive for the Col2-GFP transgene by PCR and Southern blot analysis of tail gDNA. Transgenic newborn progeny of three of these founders exhibited fluorescence of skeletal elements of the paw when examined using epifluorescence microscopy. No fluorescence was observed for wild-type offspring. These three strains were expanded based on fluorescence phenotyping and PCR genotyping. All observations were made on embryos and newborn mice heterozygous for the transgene. No differences were noted in general development, health, and fertility of transgenic versus wild-type mice. Crown-rump lengths and weights from E12.5– 396 GRANT ET AL. E19.5 were comparable, and no differences were specifically observed in skeletal development. Imaging of Whole Embryos by Epifluorescence Fluorescence of the cartilaginous skeleton was visualized in whole live embryos by epifluorescence microscopy. It was not diminished by paraformaldehyde fixation, which made subsequent dissection much easier for embryos younger than E15.5. Wild-type embryos exhibited no fluorescence. The expected progression of skeletal development is shown at low magnification in composites of E14.5 and E17.5 transgenic embryos in Figure 1. Confocal Imaging of Skeletal Elements The distribution of GFP-positive cells matched closely the distribution collagen II detected by immunostaining (Fig. 2). However, fluorescence in the former case was restricted to cells, whereas it localized to matrix in the latter case. The distribution closely resembled the pattern observed in previous studies using in situ hybridization and ␤gal reporter transgene strategies to demonstrate expression of Col2a1 (Cheah et al., 1991; Metsaranta et al., 1995). This was expected since we used the same murine Col2a1 promoter/enhancer to drive GFP expression that had been used by Metsaranta et al. to direct the ␤gal expression (Metsaranta et al., 1995). GFP fluorescence was first detected at E10.5 (Fig. 3A,B). This is one day earlier than that observed in the Col2-␤gal mice (Metsaranta et al., 1995). However, it corresponds to the earliest detection of Col2a1 transcripts by in situ hybridization (Cheah et al., 1991) and is consistent with other reports that GFP is a more sensitive reporter than ␤gal in transgenic mice (Chiocchetti et al., 1997). GFP fluorescence was localized in E10.5 embryos to axial structures, including the otic capsule, chondrogenic regions of developing maxillary and mandibular processes and sclerotome condensations of pre-vertebrae. Figure 3B shows high magnification of the sclerotomal cells in a pre-vertebra. By E14.5, most of the endochondral skeleton shows GFP fluorescence (Fig 1A, 3C, 3D) as do nonskeletal cartilages, such as nasal and external ear cartilages (Fig 1A). Less GFP fluorescence was present by E17.5 as many bones are mostly or partially ossified (Fig. 1B). This was most notable for the spine (except for tail), posterior ribs and diaphyses of long bones of the limbs. Figures 3E and 3F demonstrate this phenomenon in the lower extremity at E16.5, where the tibia is partially ossified while the bones of the feet are primarily cartilaginous. The proximal tibial growth plate is seen at higher magnification in Figure 3F. There are fewer fluorescing cells in the hypertrophic zone than in the resting and proliferating zones of the growth plate. GFP fluorescence tended to be more intense in the nucleus than in the cytoplasm of cells. The intensity of GFP fluorescence was not the same in all chondrocytes within a given cartilaginous struc- ture. This variability was consistently more evident in cartilaginous structures from older embryos, especially beyond E16.5. Image analysis of z-series of confocal images showed that this was not an artifact of optical sectioning. Cell Sorting Figure 4A shows a confocal image of a rib cage from a transgenic newborn prior to collagenase digestion. The results of typical high-speed sorting runs of cells derived from wild-type (left) and transgenic (right) rib cages are shown in Figure 4B. The threshold established from sorting the wild-type cells was used to isolate GFP fluorescence-positive cells in this experiment. The characteristics of cells isolated by sorting were compared to cells from a comparable cell suspension that had not been sorted after 48 hr of routine culture on plastic (Fig. 5). Most of the sorted cells displayed a rounded morphology and GFP fluorescence typical of chondrocytes, whereas most of the non-sorted cells exhibited a spread-out “fibroblastic” morphology and were not fluorescent. Fluorescence was restricted to rounded cells in both instances. Thus, sorting produced considerable enrichment for round, fluorescing cells. To further demonstrate that GFP fluorescence marks expression of the chondrocyte phenotype and indicates expression of endogenous Col2a1, crude cell suspensions from ribs of transgenic newborn mice were sorted twice. The first sort separated non-fluorescing from fluorescing cells; the second sort separated high from low-moderate fluorescing cells. The three cell populations were compared by western blot analysis of procollagen II. Figure 6 shows that intensity of fluorescence correlated well with the level of procollagen II synthesis. Thus, fluorescence derived from the Col2-GFP transgene provides an in situ cellular assay for type II collagen synthesis. One can assess the level of type II collagen synthesis of one or a group of cells relative to other cells within intact skeletal structures. This capability would be useful for studying biologic phenomena in which synthesis of type II collagen is considered important, such as bone development and growth and repair of bone fractures and cartilage injuries. However, one must be aware that GFP is a stable protein that may survive in cells for up to 24 hours making the assay most useful for assessing collagen synthesis over many hours to days rather than for shorter periods (Li et al., 1998). The mouse is becoming the experimental model of choice to study mammalian skeletal development. However, a significant drawback is the difficulty isolating chondrocytes for in vitro studies. The small size of embryonic and newborn mouse bones makes it almost impossible when isolating chondrocytes to avoid substantial contamination of nonchondrocytic cells, which in turn, may compromise analyses being carried out. High-speed flow cytometric cell sorting of GFPpositive cells circumvents this problem. Indeed, our results demonstrate that an almost pure population of Col2-GFP REPORTER MICE chondrocytes can be obtained by sorting of crude collagenase cell suspensions of ribs in less than 24 hr. A similar strategy has been used successfully to isolate dermal papilla cells from the skin of newborn mice transgenic for a versican-GFP transgene (Kishimoto et al., 1999). Confocal imaging of endogenous fluorescence in skeletal elements has many advantages over conventional microscopy. Artifacts related to processing and staining are avoided because the tissues are sectioned optically rather than mechanically, and probes or treatments with other agents are not needed. Background fluorescence of non-cartilaginous tissues, such as bone and perchondrium, is minimal in optical sectioning of even fixed tissues. This presumably reflects the much thinner optical sections (0.5) imaged by confocal microscopy compared to thicker sections (8) imaged conventionally, the relatively high intensity of GFP fluorescence compared to the intensity of background fluorescence and the avoidance of paraffin embedment. Most important is the ability to image single cells in intact, live tissues such as individual bones or limbs in culture or in whole embryos. At low magnification, images resemble X-rays, except that cartilage rather than bone is identified. A potential use of this strategy will be to study the effects of various mutations on skeletal development in mice and to isolate chondrocytes harboring these mutations. In this case the relevant mutants would be genetically crossed with the Col2-GFP reporter mice to produce mice that carry both the mutant gene of interest and the reporter transgene. Such mice could be studied microscopically and cell sorting would yield chondrocytes bearing the mutation for analyses. We have begun to exploit this strategy to study the consequences of receptor mutations on chondrocyte signaling pathways. The use of the Col2-GFP reporter mice as reported here provides a new tool to study murine chondrocyte biology. It offers a potential window through which to observe the dynamic events in which chondrocytes participate and a quick access to murine chondrocytes. EXPERIMENTAL PROCEDURES Generation of Transgenic Mice The pCol2a1-GFP vector was constructed using standard cloning procedures. The Col2a1 promoter was excised from the plasmid p3000i3020Col2a1 by Cla I-Not I-Asp 718 digestion (Zhou et al., 1995). After filling in the ends with Klenow, the 6.3 kb Col2a1 promoter was blunt-end ligated into the filled-in BamHI site of pEGFP-1 (Clontech, Palo Alto, CA). Orientation was verified by Not I-Xba I digestion. The entire 7.35 kb Col2a1-GFP transgene was excised by Asp718-Afl II digestion and gel-purified. Microinjection and other methods used to generate the transgenic mice have been described previously (Garofalo et al., 1991; Metsaranta et al., 1995). 397 Detection of Transgenic Mice and Embryos Pregnant mice were sacrificed following anesthesia using CO2 inhalation. Embryos were removed by Csection at 9.5–18.5 days p.c. Transgenic embryos and neonatates were identified by fluorescence microscopy and confirmed by PCR and Southern blot analysis of gDNA using probes and primers for GFP. Tissue Preparation and Staining The whole skeletons of some embryos were cleared and stained for cartilage and bone using Alizarin red S and alcian blue (McLeod, 1980). For alcian blue/hematoxylin and eosin (H&E) staining, tissues were fixed 24 hr in 4% paraformaldehyde, embedded in paraffin, cut into 8 m sections and stained using standard protocols. Immunostaining for collagen II was done on 8-m paraffin sections fixed according to the AMeX method (Sato et al., 1986). Deparaffinized sections were treated with chondroitinase ABC (0.25 units/ml, Sigma, St. Louis, MO) for 30 min at RT, washed in PBS and incubated in primary antibody (Collagen Type II Ab-2, Clone 2B1.5, Neomarkers, Inc., Fremont, CA) diluted 1:20 in PBS overnight in a humidified chamber at 4°C. Secondary antibody (Alexa 546, 1:50, Molecular Probes, Inc., Eugene, OR) was applied at RT for 30 min. Finally, sections were washed with PBS, mounted with glycerol and imaged by epifluorescence. For confocal microscopy, whole embryos and embryonic tissues were embedded in 7% agarose (Gibco-BRL, Gaithersburg, MD) either unfixed or after overnight fixation in 4% paraformaldehyde before embedment. For newborn pups, bones were dissected free, cut in half and in some cases fixed overnight in 4% paraformaldehyde. In both instances skeletal tissues were cut into 200 –300 m sections using a D.S.K. DTK-2000 Microslicer (Ted Pella, Inc., Irvine, CA), Imaging GFP fluorescence was visualized in whole animals using a Nikon E800 microscope equipped with epifluorescence, a SenSys digital camera (Photometrics, Tucson, AZ), and Metamorph imaging system (Universal Imaging, West Chester, PA). Confocal images were captured using a Nikon Diaphot 300, attached to a Biorad MRC 1024 confocal system. The digitized images were assembled in Photoshop 5.02 (Adobe Systems, Seattle, WA). Isolation of Chondrocytes and Cell Sorting One- to two-day-old mice were anesthetized by CO2 inhalation. Rib cages were removed and in a few cases imaged confocally to demonstrate GFP fluorescence. The rib cages were treated with pronase (Sigma, 2 mg/ml in PBS) at 37°C for 30 min washed in PBS washes and transferred to a bacteriological dish. Collagenase B was added (Boehringer Mannheim; 2 mg/ml in cell culture medium: DMEM without phenol red supplemented with 10% fetal calf serum, 100 units/ml 398 GRANT ET AL. Fig. 3. Confocal images of chondrogenic/cartilaginous structures of transgenic embryos E10.5 (A, B), E14.5 (C, D), and E16.5 (E, F). A shows otic vesicle (arrow), chondrogenic regions of developing maxillary and mandibular processes (double arrows) and sclerotome condensations of pre-vertebrae (arrowhead). B is a high magnification of the sclerotomal cells in a pre-vertebra. C shows thoracic vertebral bodies. D shows vertebral bodies in the lower lumbar area. E and F show the middle segment of the lower limb at low and higher magnification. Arrow points to fibula, double arrows to ossified portion of tibia and arrowhead to the proximal tibial growth plate. Note the cellular distribution of fluorescence in B–F. Scale bars ⫽ 100 for A, 10 for B–D, 1 mm for E and 500 for F. penicillin, 100 ug/ml streptomycin, and 2 mM sodium pyruvate; Gibco-BRL) and incubation was carried out overnight at 37°C with gentle shaking to dissociate the ribs. The cell suspension formed by pipeting with PBS was strained through a 70 filter. The dishes were rinsed with an equal volume of PBS and the solution also filtered. After the filtrate was centrifuged for 5 min at 1,000 rpm, the supernatant was discarded and the pellet was resuspended in cell culture medium. Viable cells were counted using trypan blue solution, and the cells were resuspended to 5 ⫻ 105 cells/ml in cell culture medium with 0.5% Collagenase-B to prevent clumping. Cells were then sorted into fluorescing and non-fluorescing populations. In some cases the fluorescing cell population was sorted again into low-tomoderate and high fluorescing cell subpopulations. Col2-GFP REPORTER MICE 399 Fig. 4. Isolation of chondrocytes by high-speed flow cytometry. A shows ribs from transgenic newborn prior to digestion. Cellular distribution of fluorescence is indicated in inset. B shows results of a sorting experiment for cells from a wild-type mouse on left and from a transgenic mouse on the right. Fig. 5. Cells cultured on plastic chamber slides 48 hr after isolation from transgenic newborn mouse ribs. Cells in A and B were not sorted; fluorescence-positive cells in C and D were selected by sorting. A and C were imaged by DIC to show morphology; B and D by epifluorescence. Sorted cells exhibit predominantly a rounded morphology and most cells fluoresce. In contrast, many of the non-sorted cells have a spread-out appearance and there are many fewer fluorescing cells. Fluorescence is restricted to non-spread out, rounded cells in both instances. Cell Culture Some cells were cultured on Permanox plastic chamber slides (NUNC, Naperville, IL) and photographed using DIC optics to show morphology and fluorescence. After sorting, each population of cells was plated in cell culture medium at a density of 1 ⫻ 106 cells per 60 mm dish. After 24 hr, the cell layer was washed with PBS and incubated for 24 hr in DMEM supplemented with 0.05% fetal calf serum and 50 g/ml ascorbic acid phosphate (Wako Biochemicals, Richmond, VA). The collected media were diluted in 5X SDS sample buffer (325 mM Tris pH6.8, 10% SDS, 25% glycerol, and 5% bromophenol blue) and used for immunoblot analysis. Immunoblot Analysis Equal volumes of media samples were electrophoresed on 7.5% SDS polyacrylamide gels and proteins were transferred to PVDF membranes (Millipore, Bedford, MA). After 1 hr incubation at RT in blocking buffer (1X Tris-buffered saline with 0.05% Tween-20 400 GRANT ET AL. Fig. 6. GFP fluorescence intensity correlates with procollagen II synthesis. Cells were sorted into populations exhibiting no (lane 1), lowmoderate (lane 2) and high intensity (lane 3) fluorescence and cultured as described in the Experimental Procedures. Equivalent aliquots of media from these cells were immunoblotted for procollagen II. Pepsindigested collagen II served as a control (C). Increasing amounts of procollagen II were detected in lanes 2 and 3. The slightly smaller band of much lower intensity in lane 3 is probably partially processed pC collagen II. and 4% BSA), the membranes were incubated in primary antibody diluted in blocking buffer overnight at 4°C. They were washed 3 times in blocking buffer and incubated in secondary antibody (anti-mouse IgG linked to horseradish peroxidase; Amersham) diluted in blocking buffer without BSA for 1 hr at RT. After 3 more washes, membranes were developed for chemiluminescence using SuperSignal (Pierce, Rockford, IL). ACKNOWLEDGMENTS The work was supported by a grant from the Shriners Hospitals Research Program (WAH). We wish to thank Silvio Garofalo for plasmid p3000i3020Col2a1, Megan Nguyen for technical assistance and Peter Hurlin and Gregory Lunstrum for helpful suggestions. REFERENCES Ala-Kokko L, Kvist AP, Metsaranta M, Kivirikko KI, de Crombrugghe B, Prockop DJ, Vuorio E. 1995. Conservation of the sizes of 53 introns and over 100 intronic sequences for the binding of common transcription factors in the human and mouse genes for type II procollagen (COL2A1). Biochem J 308:923–929. Aulthouse AL, Beck M, Griffey E, Sanford J, Arden K, Machado MA, Horton WA. 1989. Expression of the human chondrocyte phenotype in vitro. In Vitro Cell Dev Biol 25:659 – 668. Beier F, Leask TA, Haque S, Chow C, Taylor AC, Lee RJ, Pestell RG, Ballock RT, LuValle P. 1999. Cell cycle genes in chondrocyte proliferation and differentiation. Matrix Biol 18:109 –120. Benya PD, Shaffer JD. 1982. Dedifferentiated chondrocytes reexpress the differentiated collagen phenotype when cultured in agarose gels. Cell 30:215–224. Benya PD, Padilla SR, Nimni ME. 1978. Independent regulation of collagen types by chondrocytes during the loss of differentiated function in culture. Cell 15:1313–1321. Cheah KS, Lau ET, Au PK, Tam PP. 1991. Expression of the mouse alpha 1(II) collagen gene is not restricted to cartilage during development. Development 111:945–953. Chen H, McCarty DM, Bruce AT, Suzuki K. 1998. Gene transfer and expression in oligodendrocytes under the control of myelin basic protein transcriptional control region mediated by adeno-associated virus. Gene Ther 5:50 –58. Chiocchetti A, Tolosano E, Hirsch E, Silengo L, Altruda F. 1997. Green fluorescent protein as a reporter of gene expression in transgenic mice. Biochim Biophys Acta 1352:193–202. Doevendans PA, Becker KD, An RH, Kass RS. 1996. The utility of fluorescent in vivo reporter genes in molecular cardiology. Biochem Biophys Res Commun 222:352–358. Erlebacher A, Filvaroff EH, Gitelman SE, Derynck R. 1995. Toward a molecular understanding of skeletal development [comment]. Cell 80:371–378. Fleischmann M, Bloch W, Kolossov E, Andressen C, Muller M, Brem G, Hescheler J, Addicks K, Fleischmann BK. 1998. Cardiac specific expression of the green fluorescent protein during early murine embryonic development. FEBS Lett 440:370 –376. Garofalo S, Vuorio E, Metsaranta M, Rosati R, Toman D, Vaughan J, Lozano G, Mayne R, Ellard J, Horton W, et al. 1991. Reduced amounts of cartilage collagen fibrils and growth plate anomalies in transgenic mice harboring a glycine-to-cysteine mutation in the mouse type II procollagen alpha 1-chain gene. Proc Natl Acad Sci USA 88:9648 –9652. Horton W, Miyashita T, Kohno K, Hassell JR, Yamada Y. 1987. Identification of a phenotype-specific enhancer in the first intron of the rat collagen II gene. Proc Natl Acad Sci USA 84:8864 – 8868. Johnson RL, Tabin CJ. 1997. Molecular models for vertebrate limb development. Cell 90:979 –990. Kishimoto J, Ehama R, Wu L, Jiang S, Jiang N, Burgeson RE. 1999. Selective activation of the versican promoter by epithelial-mesenchymal interactions during hair follicle development. Proc Natl Acad Sci USA 96:7336 –7341. Li X, Zhao X, Fang Y, Jiang X, Duong T, Fan C, Huang CC, Kain SR. 1998. Generation of destabilized green fluorescent protein as a transcription reporter. J Biol Chem 273:34970 –34975. McLeod MJ. 1980. Differential staining of cartilage and bone in whole mouse fetuses by alcian blue and alizarin red S. Teratology 22:299 – 301. Mendler M, Eich-Bender SG, Vaughan L, Winterhalter KH, Bruckner P. 1989. Cartilage contains mixed fibrils of collagen types II, IX, and XI. J Cell Biol 108:191–197. Metsaranta M, Garofalo S, Smith C, Niederreither K, de Crombrugghe B, Vuorio E. 1995. Developmental expression of a type II collagen/beta-galactosidase fusion gene in transgenic mice. Dev Dyn 204:202–210. Minegishi N, Ohta J, Yamagiwa H, Suzuki N, Kawauchi S, Zhou Y, Takahashi S, Hayashi N, Engel JD, Yamamoto M. 1999. The mouse GATA-2 gene is expressed in the para-aortic splanchnopleura and aorta-gonads and mesonephros region. Blood 93:4196 – 4207. Neame PJ, Tapp H, Azizan, A. 1999. Noncollagenous, nonproteoglycan macromolecules of cartilage. Cell Mol Life Sci 55:1327–1340. Ono K, Takii T, Onozaki K, Ikawa M, Okabe M, Sawada M. 1999. Migration of exogenous immature hematopoietic cells into adult mouse brain parenchyma under GFP-expressing bone marrow chimera. Biochem Biophys Res Commun 262:610 – 614. Ramdi H, Legay C, Lievremont M. 1993. Influence of matricial molecules on growth and differentiation of entrapped chondrocytes. Exp Cell Res 207:449 – 454. Sato Y, Mukai K, Watanabe S, Goto M, Shimosato Y. (1986) The AMeX method. A simplified technique of tissue processing and paraffin embedding with improved preservation of antigens for immunostaining. Am J Pathol 125:431– 435. St-Jacques B, Hammerschmidt M, McMahon AP. 1999. Indian hedgehog signaling regulates proliferation and differentiation of chondrocytes and is essential for bone formation. Genes Dev 13:2072–2086. Stevens DA, Williams GR. 1999. Hormone regulation of chondrocyte differentiation and endochondral bone formation. Mol Cell Endocrinol 151:195–204. Tickle C, Eichele G. 1994. Vertebrate limb development. Annu Rev Cell Biol 10:121–152. van den Pol AN, Ghosh PK. 1998. Selective neuronal expression of green fluorescent protein with cytomegalovirus promoter reveals entire neuronal arbor in transgenic mice. J Neurosci 18:10640–10651. Vikkula M, Metsaranta M, Syvanen AC, Ala-Kokko L, Vuorio E, Peltonen L. 1992. Structural analysis of the regulatory elements of the type-II procollagen gene. Conservation of promoter and first intron sequences between human and mouse. Biochem J 285:287–294. von der Mark K. 1980. Immunological studies on collagen type transition in chondrogenesis. Curr Top Dev Biol 14:199 –225. Zhou G, Garofalo S, Mukhopadhyay K, Lefebvre V, Smith CN, Eberspaecher H, de Crombrugghe B. 1995. A 182 bp fragment of the mouse pro alpha 1(II) collagen gene is sufficient to direct chondrocyte expression in transgenic mice. J Cell Sci 108:3677–3684. Zhuo L, Sun B, Zhang CL, Fine A, Chiu SY, Messing A. 1997. Live astrocytes visualized by green fluorescent protein in transgenic mice. Dev Biol 187:36 – 42.