SEARCH

SEARCH BY CITATION

Keywords:

  • bone;
  • transgenic mouse;
  • bone sialoprotein;
  • core binding factor 1 (Cbfa1/Runx2);
  • avian retroviral receptor gene;
  • osteopontin

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

To study bone development in vivo, a transgenic mouse model was established in which an avian retroviral receptor (TVA) gene driven by the BSP promoter was selectively expressed in skeletal tissues. The model was validated by showing suppressed BSP expression and delayed bone and tooth formation after infection with a virus expressing a mutated Cbfa1/Runx2 gene.

Introduction: Tissue-specific expression of the avian retroviral (TVA) receptor can be used to efficiently target ectopic expression of genes in vivo. To determine the use of this approach for studies of osteogenic differentiation and bone formation at specific developmental stages, transgenic mice expressing the TVA receptor under the control of a 5-kb bone sialoprotein (BSP) promoter were generated. The mice were first analyzed for tissue-specific expression of the TVA gene and then, after infection with a viral construct, for the effects of a dominant-negative form of the Cbfa1/Runx2 transcription factor on bone formation.

Materials and Methods: We first generated transgenic mice (BSP/TVA) in which the TVA gene was expressed under the control of a 4.9-kb mouse BSP promoter. The tissue-specific expression of the TVA gene was analyzed by RT-PCR, in situ hybridization, and immunohistochemistry and compared with the expression of the endogenous BSP gene. A 396-bp fragment of mutated Cbfa1/Runx2 (Cbfa1mu) encoding the DNA-binding domain was cloned into a RCASBP (A) viral vector, which was used to infect neonatal BSP/TVA mice.

Results and Conclusion: Expression of the TVA receptor mRNA and protein in the transgenic mice was consistent with the expression of endogenous BSP. Four days after systemic infection with the Cbfa1mu-RCASBP (A) vector, RT-PCR analyses revealed that the expression of BSP mRNA in tibia and mandibles was virtually abolished, whereas a 30% reduction was seen in calvarial bone. After 9 days, BSP expression in the tibia and mandible was reduced by 45% in comparison with control animals infected with an empty RCASBP vector, whereas BSP expression in the membranous bone of calvariae was decreased ∼15%. However, after 4 and 8 weeks, there was almost no change in BSP expression in any of the bone tissues. In comparison, a reduction in osteopontin expression was only observed 9 days after viral transfection in the three bones. Histomorphological examination revealed that bone formation and tooth development were delayed in some of the mice infected with mutated Cbfa1. These studies show that BSP/TVA transgenic mice can be used to target genes to sites of osteogenesis, providing a unique system for studying molecular events associated with bone formation in vivo.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

THE MOLECULAR MECHANISMS that regulate differentiation of osteoblastic cells involve the coordinated expression of genes that guide the formation and remodeling of mineralized bone tissues. Previous studies have identified core binding factor Cbfa1 (recently named Runx2)(1, 2) and osterix (Osx)(3) as “master genes” for osteogenic differentiation. Cbfa1/Runx2 is expressed early in mesenchymal and epithelial tissues destined to form the mineralized tissues of the tooth and periodontium.(4, 5) In Cbfa1/Runx2-deficient mice, the formation of membranous and endochondral bones, but not mineralized cartilage, is blocked,(1, 2) whereas both bone and mineralized cartilage formation are lost in Osx-deficient mice, indicating that Osx acts downstream of Cbfa1/Runx2.(3) Studies in vitro have shown that Cbfa1/Runx2 upregulates osteocalcin (OCN) expression through multiple elements in the proximal gene promoter(6) and also increases osteopontin (OPN) expression through interactions with ETS1.(7) However the effects of Cbfa1/Runx2 on bone sialoprotein (BSP) expression are unclear. Whereas Cbfa1/Runx2 downregulates avian BSP(8) and upregulates BSP in human breast cancer cells,(9) no effect of Cbfa1/Runx2 was observed on the human BSP promoter expressed in several normal or transformed bone cells,(10) suggesting that there are species variations in Cbfa1/Runx2 regulation of BSP expression. Furthermore, although Cbfa1/Runx2 is clearly required for bone formation, overexpression of Cbfa1/Runx2 can inhibit osteogenic differentiation leading to osteopenia.(11) Thus, the precise functions of Cbfa1/Runx2 in bone formation are complex.

BSP is a major protein of the bone matrix,(12, 13) which is extensively modified by post-translational glycosylation,(14, 15) phosphorylation,(16) and sulphation of tyrosines.(17) The sulphated tyrosines flank an RGD cell attachment site near the C terminus that is recognized by the αvβ3 vitronectin receptor.(12, 18) In the amino terminal region, two/three polyglutamic acid regions can mediate binding to hydroxyapatite(18) and may also provide a nucleation site for hydroxyapatite formation.(19)BSP gene expression is associated with the onset of mineralization and is essentially restricted to differentiated osteoblasts, cementoblasts, odontoblasts, and ameloblasts.(20–22) Recently, we have also shown that overexpression of BSP in transgenic mice facilitates wound healing in bone and promotes bone maturation during the repair process. Although generally restricted to mineralizing tissues, expression of BSP has been reported in developing brain(21) and in salivary glands(23) and in human cancer cells that metastasize to bone.(24) Recent studies indicate that BSP may protect cancer cells from complement-mediated lysis(25) and promote the metastasis of breast cancer cells to bone.(26, 27) Although BSP expression provides a specific marker for differentiated cells responsible for mineralized tissue formation,(4, 28) the mechanisms regulating the tissue-specific expression of BSP in these cells are largely unknown. Previous studies on the mouse BSP promoter have suggested that tissue-specific expression resides in the proximal promoter region.(29) However, whereas transgenic mice expressing a 2.7-kb rat BSP promoter construct ligated to a luciferase reporter showed strong expression in mineralizing tissues, ectopic expression was observed in several other tissues,(21) indicating that upstream elements are required to confer absolute tissue specificity. After the characterization of the upstream promoter of the mouse BSP promoter,(30, 31) we recently generated transgenic mice expressing 4.8- and 9.0-kb constructs linked to a luciferase gene, from which preliminary studies have shown tissue-specific expression.

A novel system for targeting genes to specific cells was developed recently by Holland and Varmus.(32) Transgenic mice were engineered to express a transgene encoding the receptor for subgroup A avian leukosis virus (tva) ligated to the astrocyte-specific glial fibrillary acidic protein promoter. The promoter was used to selectively express the viral tva receptor gene, permitting efficient glia-specific transfer of genes carried by subgroup A avian leukosis virus vectors. We have adapted this system to selectively target TVA expression in osteoblasts by ligating the tva gene to the 4.9-kb BSP promoter construct and have tested the use of the system by directing expression of a dominant-negative Cbfa1/Runx2 (Cbfa1-μ) construct to TVA-expressing bone cells.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Preparation of BSP/TVA construct for microinjection

The mRNA transcribed from the avian tva gene is alternatively spliced to produce at least two proteins: a transmembrane isoform and a GPI-anchored isoform. Both have 83 amino acid extracellular domains and are sufficient to permit infection of mammalian cells. To generate the construct of tva driven by the promoter of the mouse BSP gene, we used the pKCR cloning vector (a gift from Dr Jolene Windle, VA Commonwealth University). We used the BSPLuc4.9 construct we recently isolated and characterized(31) and excised the promoter fragment with KpnI and SmaI, the same enzymes used to linearize the pKCR. The BSP promoter was inserted into the corresponding sites of the plasmid to generate a transitional construct named pBSP4.9pKCR. The pCR-tva-800 expression vector was a generous gift from Dr Harold Varmus' laboratory, Memorial Sloan-Kettering Cancer Center, NY. This vector contains 800 bp of tva gene, which was excised from a pSPKE 0.8 originally developed by Drs Paul Bates and Stephen Hughes (NCI/NIH). We removed the tva fragment with EcoRI and subsequently cloned it into the EcoRI site of pBSP4.9pKCR. We named this new construct pBSPtva (Fig. 1). The orientation of both inserts, the BSP promoter and the tva segment, were confirmed by restriction mapping.

thumbnail image

Figure FIG. 1.. Schematic drawing of the BSP/TVA system. (A) Diagram of the pBSPtva construct used in microinjection for generating BSP/TVA transgenic mice. The 4.9-kb mouse BSP promoter is indicated by the open region; the 800-bp tva sequences are indicated by the lattice region. Restriction enzymes used in isolation of the transgene BSP/tva are indicated. (B) In transgenic mice carrying a viral receptor gene, tva, driven by a 4.9-kb BSP promoter, all the BSP expressing cells (e.g., osteoblasts) will bear the viral receptor on their cellular membrane. (C) A viral vector carrying gene of interest (e.g., DN Cbfa1 or Cbfa1 wildtype) will specifically target these cells. This infection does not cause viral spreading or immunity because gag (for a structural protein), pol (for a viral enzyme), or evn (for an envelope glycoprotein) are poorly expressed in mammal. LTR, long terminal repeats, serve as a constitutive promoter resulting in expression of the Cbfa1/Runx2 gene.

Download figure to PowerPoint

Establishment of the BSP/TVA transgenic mouse line

A 5.7-kb fragment of BSPtva was released by KpnI and XmaIII and purified using a Qiagen Maxi prep kit. This pBSPtva construct was used to generate transgenic mice in the Transgenic Mouse Facilities at the University of Texas, San Antonio (UTHSCSA). Transgenic mice were screened for positive founders (BSP/TVA) using Southern analysis and PCR of DNA from tail clips, following the procedure described previously.(21) Briefly, high molecular weight mouse genomic DNA samples were obtained from the tail clips. Ten micrograms of genomic DNA was digested with EcoRV fractionated by agarose gel electrophoresis, transferred to a nylon membrane, and hybridized with BSPtva [32P]labeled by random priming. Four samples of 10 μg genomic DNA from a control mouse were mixed with an appropriate amount of BSPtva construct corresponding to 0, 1, 10, and 50 copies per haploid genome, respectively. These four samples were digested and included on analytical gels to serve as negative and positive controls and also as standards to determine transgene copy numbers.

PCR was used as an alternative to screen for positive animals. Two oligonucleotide primers (GenBank accession L22753) were synthesized (Integrated DNA Technologies, Coralville, IA, USA): tva5 (23 bases: 5′-CTGCTGCCCGGTAACGTGACCGG-3′) and tva3 (21 bases: 5′-GCCCTGGGGAAGGTCCTGCCC-3′). Approximately 1 μg of purified DNA was added to a mixture containing a 1.1× concentration of PCR Supermix (Invitrogen, Carlsbad, CA, USA), 1 μl of each primer (10 μM), and water. As a positive control, we used BSPtva plasmid DNA as template, and a negative control was also included. The total volume was 50 μl. The samples were heated to 94°C for 5 minutes, followed by 30 cycles of 94°C for 30 s, 58°C for 30 s, 72°C for 30 s, with a final extension at 72°C for 10 minutes. The PCR products were electrophoresed on a 1.5% agarose gel.

To generate homozygote mice for the BSP/TVA gene, founders were mated with normal mice, and their offspring were interbred and identified using slot blot hybridization. Briefly, 5 μg of genomic DNA was prepared in 0.5 ml of 0.4 M NaOH, 10 mM EDTA, heated at 100°C for 10 minutes and transferred to a nylon membrane with Bio-Dot SF Microfiltration Apparatus (Bio-Rad, Hercules, CA, USA). The membrane was cross-linked in the UV Stratalinker (Stratagene), baked at 80°C for 2 h. Prehybridization was performed in 0.5 M Na2HPO4 pH 7.2, 1 mM EDTA, and 7% SDS at 65°C for 30 minutes in a hybridization oven. Hybridization was performed overnight at 65°C with a random-primed probe prepared by radioactively labeling of 25 ng of the tva DNA and [32P]dCTP (Perkin-Elmer Life Science, Boston, MA, USA) using a Megaprime DNA labeling systems (Amersham Life Science). The membrane was washed twice in 40 mM Na2HPO4 pH 7.2, 1 mM EDTA, and 5% SDS at 65°C for 30 minutes and exposed to X-ray film at −80°C for 2–4 h.

Analysis of BSP promoter and TVA expression in transgenic mice

To analyze the tissue-specific expression of TVA driven by a 4.9-kb mouse BSP promoter, total RNA was extracted from the muscle, skin, brain, intestine, kidney, liver, and bone tissues, including calvaria, tibia, and mandibles of 4-day-old BSP/TVA mice. After removing DNA with DNA-free kit (Ambion), the total RNA was reverse transcribed and amplified by the OneStep RT-PCR kit (Qiagen). The PCR primers for amplifying TVA were described above. Endogenous BSP expression was analyzed in parallel using mouse BSP primers.(22) GAPDH (GenBank accession M32599) mRNA expression was used as a loading control (20 bases for the sense sequences: 5′-ATCACTGCCACCC AGAAGAC-3′; 20 bases for the antisense sequences: 5′-ATGAGGTCCACCACCCTGTT-3′ for a 443-bp GAPDH fragment). The amplified products were separated on a 1.0% agarose gel, stained with ethidium bromide, and photographed under UV illumination.

To show the tissue-specific expression of TVA, in situ hybridization of bone tissues with a [35S]labeled TVA RNA probe was performed. The detailed description of the process of in situ hybridization in mineralized tissues can be found in previous reports.(22, 33) To further characterize the TVA receptor expression in targeted cells, immunohistochemistry was performed. Affinity-purified rabbit polyclonal anti-TVA antibody(34, 35) was obtained from Dr Andrew Leavitt (UCSF) and used at a concentration of 1 μg/ml. The immunostaining techniques used have been detailed in our previous papers.(22, 27)

Constructs of viral vectors carrying dominant-negative Cbfa1/Runx2 DNA binding domain

The viral vector RCASBP used to target a mutated Cbfa1/Runx2 gene to BSP expressing cells was generously provided by Dr Stephen Hughes (NCI-Fredrick). The Cbfa1/Runx2 DNA binding domain in the mutated Cbfa1/Runx2 construct has a higher affinity for DNA than Cbfa1/Runx2 itself, but has no transcriptional activity on its own, and can act in a dominant negative manner in DNA co-transfection assays and in vivo.(36) The 396-bp NcoI-HindIII fragment of Cbfa1/Runx2 was cloned into the ClaI site in RCASBP (A) through the Cla12 Nco shuttle vector. This viral vector carrying the mutant Cbfa1/Runx2 DNA was named as RCAS-Cbfa1mu.

Production of high titer RCAS-Cbfa1mu stock in DF1 cells

The established chicken fibroblast cell line, DF1 (CL-12203) was purchased from ATCC (Manassas, VA, USA) and propagated in DMEM with 4 mM l-glutamine, 1.5 g/liter sodium bicarbonate, and 4.5 g/liter glucose (ATCC), 10% FBS (GIBCO/BRL), and 1% penicillin/streptomycin at 39°C and 5% CO2. The cells were maintained in DMEM medium. Transfection was performed using calcium phosphate precipitation as described previously.(37) We passaged DF1 cells 1 day before transfection of RCAS-Cbfa1mu using Superfect (Qiagen) so that the cells were ∼80% confluent at the time of transfection. The supernatant was harvested over 3 days and concentrated ∼30-fold by ultracentrifugation at 26,000 rpm at 4°C for 1.5 h. Transfection of the RCAS-Cbfa1mu plasmid DNA into DF1 cells resulted in the production of high titer replication-competent viral stock. The viral pellet was resuspended to a titer of about 108 cfu/ml and stored at −80°C.

Delivery of virus vector of RCAS-Cbfa1mu into BSP/TVA transgenic mice

Mammalian cells engineered to express TVA are highly susceptible to infection by a replication-competent, subgroup A avian leucosis viral (ASLV) vector, RCAS. After entry into mammalian cells, a newly synthesized DNA copy of the viral genome integrates into the host DNA, and viral LTRs (long terminal repeats, as a constitutive promoter) promote a high level transcription of integrated provirus. An artificial acceptor in RCAS, downstream of env, was used for efficient processing of the mRNA representing the insert at the cloning site. The osteoblasts in transgenic mice expressing TVA receptors were readily infected by the viral vector carrying the Cbfa1/Runx2 domain. To target all the cells involved in early bone formation, we delivered the viral vector RCAS-Cbfa1mu into the 5-day-old BSP/TVA mice by intraperitoneal injection.

Sample collection, RT-PCR, histology, and immunohistochemistry

An RNeasy Kit (Qiagen) was used to isolate total RNA from the BSP/TVA mouse calvariae, mandibles, and tibias at 9, 14, 28, and 56 days of age. One microgram of the RNA was used to perform the RT-PCR using SuperScript one-step RT-PCR with platinum Taq (Invitrogen). Specific primers for mouse BSP (GenBank accession L20232) were 5′-AACAATCCGTGCCACTCA-3′ (18 bases); 5′-GGAGGGGGCTTCACTGAT-3′ (18 bases) to amplify a 1048-bp fragment) and for OPN were (GenBank accession AF515708), 5′-CTCCCGGTGAAAGTGACTGA-3′ (20 bases) and 5′-GACCTCAGAAGATGAACTCT-3′ (20 bases) to amplify a 831-bp fragment).(22) Primers were also used to amplify the Cbfa1mu gene and the full-length Cbfa1/Runx2 gene.(36) cDNA was synthesized at 50°C for 30 minutes. PCR amplification were performed with a GeneAMP PCR System 2400 (Applied Biosystems). Thermal cycling conditions for the first round was a 2-minute 94°C denaturing step, followed by 30 cycles of 94°C for 30 s, 58°C for 30 s, 72°C for 1 minute, and a final extension at 72°C for 7 minutes. PCR products were electrophoresed on 1.5% agarose gels containing ethidium bromide, and the gel was photographed and quantitated using UVP Image software and corrected for GAPDH gene amplified from the same samples. At least three to four tissue samples were analyzed for both experimental and control groups. All the RT-PCR data were analyzed by one-way ANOVA using a Statview (Abacus Concepts, Berkeley, CA, USA) software package. When the test indicated significance, the difference between the Cbfa1mu and empty vector groups was determined by the Student-Newman-Keuls test. The test of significance was performed at both 95% and 99% confidence intervals compared with the control group.(22)

To determine the effect of Cbfa1//Runx2 on the development of mineralized tissues, selective tissues were dissected from the BSP/TVA mice 5 days after infection with the Cbfa1mu or empty vector. Calvaria, mandibles, and tibia were collected and processed for histological examination. To observe changes in BSP and OPN, protein immunohistochemical studies were performed using polyclonal antibodies for BSP and OPN (gifts from Dr Larry Fisher, NIH/NIDCR) using previously described methodologies.(22, 27)

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Genotyping of BSP/TVA transgenic mouse line

A DNA construct comprising the 4.9-kb mouse BSP promoter driving a 800-bp tva gene, pBSPtva, (Fig. 1A) was used for microinjection into fertilized mouse eggs. As depicted in the schematic of the BSP/TVA system (Fig. 1), BSP-expressing osteoblasts in the transgenic mice would be expected to present the viral TVA receptor on their cellular membrane (Fig. 1B) and, these receptors can be targeted by a viral vector carrying gene of interest (e.g., the Cbfa1mu in this study; Fig. 1C). Therefore, litters were screened for founders using both Southern hybridization and PCR. Digestion of the mouse DNA with EcoRV cut the gene at nucleotide 3836 and generated 3204- and 4160-bp fragments in positive mice. A founder mouse, revealing the expected major bands, was identified in the first litter (Fig. 2) by Southern hybridization of tail genomic DNA probed after32P-labeling of the DNA used in microinjection. The Southern data were confirmed by PCR. The 500-bp amplified DNA fragment represented the region joining the BSP 5′-flanking sequences and the tva gene (Fig. 2B). One BSP/TVA mouse was identified in Southern hybridization and PCR analyses. To obtain mice homozygous for the tva gene, the genotyped founder mouse was mated with normal mice, and the offspring were interbred. An example of a slot blot identifying the generation of BSP/TVA homozygotes through breeding is shown in Fig. 2C. After genotyping and phenotyping, the BSP/TVA transgenic mouse line was maintained through breeding between male and female homozygotes.

thumbnail image

Figure FIG. 2.. Analysis of genomic DNA from BSP/TVA transgenic mice. (A) Southern hybridization of genomic DNA from BSP/TVA transgenic mice. High molecular weight mouse tail DNA samples were digested with EcoRV, fractionated by agarose gel electrophoresis, transferred to nylon filters, and hybridized with BSP/tva, [32P]labeled by random priming. One transgenic founder was identified (lane 1). Shown in lanes 3, 4, 5, and 6 are BSP/tva constructs corresponding to 0, 1, 10, and 50 copies per haploid genome, respectively. (B) Agarose gel electrophoresis of genomic DNA PCR products for transgenic genotyping. Genomic DNA of three mice was amplified using two sets of primers. The 500-bp amplified DNA fragment represents the joint region of the BSP 5′-flanking sequences and the tva gene. Lane 1, 100-bp DNA ladder (New England Biolab); lanes 2–4, genomic DNA samples from mice tails; lane 1 shows a positive mouse founder; lane 5, positive control (tva plasmid, 100 ng); lane 6, negative control (no DNA included in the PCR reaction). (C) A representative sample of slot blot of agarose gel electrophoresis of genomic DNA for identification of homozygocity of BSP/TVA mice. Tail DNA from 34 mice was analyzed in this group samples. A number of homozygotes were identified judged by the radioactive intensity of bands shown in hybridization with a [32P]labeled probe. Seventeen homozygotic mice for the BSP/tva were shown on this slot blot: 10, 11, 15, 20, and 22–34.

Download figure to PowerPoint

BSP/TVA transgenic mice express TVA viral receptor

Viral spreading and immune responses are minimized by the poor expression of the viral genes in mammalian cells. In contrast, the Cbfa1mu is expressed at high levels by a constitutive LTR promoter. To confirm the targeting of TVA retroviral receptor to bone tissue in the BSP/TVA mice, both RT-PCR and in situ hybridization analyzes were performed. RT-PCR showed that bone tissues, including calvaria, tibia, and mandibles, expressed high levels of the tva gene, whereas in kidney, liver, intestine, muscle, and skin, expression was not detected under the same conditions of RT-PCR (Fig. 3). Notably, however, the transgene was also expressed in brain tissue, consistent with previous observations of BSP expression in embryonic and neonatal rat brain(21) and in mouse brain (unpublished data). In situ hybridization using a tva mRNA probe showed the expression of the BSP/TVA gene in a variety of bone tissues (Fig. 4). In tibia, the hybridization signals were seen in the osteoblasts in the growth plate, whereas in the mandible, the strongest signals were located in the cells in periosteum, indicating sites of active bone formation. This expression pattern for the TVA receptor was consistent with endogenous BSP expression that has been shown in previous studies.(21) The expression of TVA receptor was also analyzed by immunohistochemistry (Fig. 4). Specific staining was restricted to bone tissues as shown in the epiphysis of tibial bones, in calvarial bone, and in mandibular bone. Notably, no staining was evident in other tissues or bones of wildtype mice (Fig. 4).

thumbnail image

Figure FIG. 3.. Agarose gel electrophoresis of RT-PCR products of various tissues from BSP/TVA transgenic mice. Samples are labeled as follows: 1, calvaria; 2, mandible; 3, tibia; 4, brain; 5, intestine; 6, kidney; 7, liver; 8, muscle; 9, skin. M represents a 1-kb DNA ladder. (A) RT-PCR for the tva gene in a variety of tissues. All bone tissues including calvaria, mandible, and tibia expressed the tva gene, and this expressing pattern is consistent with that of endogenous BSP expression shown in B, suggesting a targeting of the retroviral receptor to the bone cells in a tissue-specific manner. Interestingly, both tva and BSP genes were expressed in brain tissue from the BSP/TVA mice (lane 4). (C) Gel for GAPDH expression serving as a control in the RT-PCR analysis and being used to adjust and normalize the sample loading error.

Download figure to PowerPoint

thumbnail image

Figure FIG. 4.. (A-F) In situ hybridization and (G-L) immunohistochemistry of tva in tibial, calvaria, and mandibular bones from 4-day-old BSP/TVA transgenic mice. (A-C) Bright field. (B-D) Corresponding dark field illustration. Strong tva signals can be seen in osteoblastic cells (arrows) in the (A and D) tibial primary spongiosa, (B and E) calvarial bone, and (C and F) mandible. Cells in (B and E) nonmineralized cartilage of tibia and (C and F) tooth pulp cells in an incisor in the mandible showed no hybridization signals. The tva mRNA shown as silver grains in the dark field photos (D-F) seem to concentrate in the cells that are actively involved in bone formation in neonatal mouse skeleton. (G) Immunohistochemistry using a polyclonal anti-TVA antibody showed positive staining in the osteoblasts (arrows) in tibia primary spongiosa. (H) Osteocytes (arrows) in calvarial also reacted with the TVA antibody. (I) In mandibular bone from a BSP/TVA mouse, cells (arrows) in alveolar bone around the developing tooth germs showed strong staining for the TVA antibody. (L) An incisor in the mouse mandible. Whereas the bone cells (arrows) located on both sides of the tooth were positively stained with the TVA antibody, the connective tissue cells showed no immunoreaction. (J) Tibia and (K) calvaria from wildtype mice showed no immunostaining, reflecting the specificity of the TVA antibody. Magnification, ×56.

Download figure to PowerPoint

Expression of Cbfa1mu and endogenous Cbfa1/Runx2 in RCASBP-Cbfa1mu-infected mice

The expression of Cbfa1mu in bone tissues from BSP/TVA mice 4 and 9 days after systemic infection with the RCAS-Cbfa1mu viral construct was determined by RT-PCR using primer pairs that recognized the Cbfa1mu (Fig. 5). It was found that mRNA for the mutated Cbfa1/Runx2 increased dramatically (Figs. 5A and 5B). The increased signal in the infected mice is assumed to reflect the additional expression of the Cbfa1mu, above that of the endogenous Cbfa1. In the control animals where an RCAS empty vector was used for infection, the mRNA amplified by using the same primers shows an expression level similar to the endogenous Cbfa1/Runx2 expression. A similar trend and pattern were seen in the data derived from both 4 (Fig A) and 9 (Fig. 5B) postinfection days. However, it seemed that a stronger effect of Cbfa1mu occurred 4 days after infection. However, because Cbfa1/Runx2 is a powerful regulator of its own gene,(11) the endogenous levels of Cbfa1/Runx2 would be expected to be decreased in the infected animals, so that the differences between the Cbfa1mu and controls are likely to be much greater than appears on the gel. Thus, when using a primer pair that only recognized endogenous Cbfa1/Runx2, it can be seen that infection of Cbfa1mu substantially decreased endogenous Cbfa1/Runx2 expression (Figs. 5C and 5D). Thus, there was an increased expression of nonfunctional Cbfa1mu in combination with decreased endogenous Cbfa1/Runx2 in the infected mice. Again the reduction of the endogenous Cbfa1/Runx2 was more significant at 4 days than 9 days after infection. These data indicate that the infection with mutated Cbfa1/Runx2 elevates the amount of nonfunctional Cbfa1mu and reduces the endogenous Cbfa1/Runx2 expression in the bone cells. Also, these results provided the foundation for the mechanism of how Cbfa1/Runx2 affects the expression patterns of bone matrix proteins.

thumbnail image

Figure FIG. 5.. Expression of Cbfa1mu in bone tissues from BSP/TVA mice. RT-PCR products separated on agarose gels and the corresponding histograms quantitating expression of mutated Cbfa1/Runx2 (Cbfa1mu, A and B) and Cbfa1/Runx2 (C and D) in bone tissues from BSP/TVA mice (A) 4 and (B) 9 days after systemic infection with RCAS-Cbfa1mu viral construct are shown. Samples from animals treated with empty vector (RCAS-EV) served as control. mRNA for the mutated Cbfa1/Runx2 increased dramatically. In the control animals where an RCAS empty vector was used for infection, the mRNA amplified by using the same primers shows an expression level similar to the endogenous Cbfa1/Runx2 expression. (C and D) RT-PCR products and the corresponding histogram quantitating expression of Cbfa1/Runx2 in bone tissues from BSP/TVA mice (C) 4 and (D) 9 days after systemic infection with RCAS-Cbfa1mu viral construct, respectively. When using primer pairs that recognize endogenous Cbfa1/Runx2, it can be seen that infection of Cbfa1mu substantially decreased endogenous Cbfa1/Runx2 expression. There is an increased expression of nonfunctional Cbfa1mu in combination with decreased endogenous Cbfa1/Runx2 in the infected mice. PCR products of GAPDH, which served as internal controls, are shown below for the corresponding samples.

Download figure to PowerPoint

Effects of Cbfa1/Runx2 in regulating BSP and OPN expression in vivo

Four days after infection of 5-day-old mice with the RCASCbfa1mu virus expressing the dominant-negative Cbfa1mu, BSP gene expression in BSP/TVA mice was markedly diminished in calvarial bone and completely ablated in mandibular and tibial bones (Fig. 6). After 9 days, there was still a ∼15% reduction in the BSP expression in calvaria and a 45% reduction in tibia and mandible compared with control animals that received an empty RCASBP vector. At 4 and 8 weeks after viral infection, there was almost no change in BSP expression in any of the three bone tissues (data not shown). In contrast to BSP, no reduction in OPN expression was observed after 4 days of infection (data not shown). However, after 9 days, a reduction of OPN expression was seen in all the three bones (Fig. 6). Statistically significant differences were seen in all groups with confidence levels of p < 0.01, except for the difference in BSP expression between Cbfa1mu and control group in calvarial bones after 9 days (p = 0.02).

thumbnail image

Figure FIG. 6.. Expression of BSP in bone tissues from BSP/TVA mice. RT-PCR products separated on agarose gels and the corresponding histograms quantitating the expression of mouse BSP in bone tissues from BSP/TVA mice (A) 4 and (B) 9 days after systemic infection with RCAS-Cbfa1mu viral construct are shown. Four days after infection, there was a 30% reduction in BSP expression in calvaria, whereas the expression of BSP in tibia and mandibles was abolished. After 9 days, there was still a 45% reduction in BSP expression in tibia and mandibular bones. (C) A significant reduction in OPN expression was only observed 9 days after viral infection in the three bones. However, the decreased level of OPN expression was not as dramatic as BSP.

Download figure to PowerPoint

The expression and location of BSP and OPN proteins were analyzed by immunohistochemistry. A general reduction in the intensity of immunostaining for BSP in the bone and tooth tissues from animals infected with Cbfa1mu was observed (Fig. 7) This finding was consistent with the RT-PCR results, indicating the dominant negative Cbfa1/Runx2 inhibited major bone matrix protein expression at both RNA and protein levels. Immunostaining also indicated reduced OPN expression, particularly in the calvariae and mandibular bone (Figs. 7H and 7N). The effect of Cbfa1/Runx2 on bone formation was assessed from a morphological examination of the tissues collected from BSP/TVA transfected with Cbfa1mu. In the tibia, there was an accumulation of cartilage cells in the growth plate area, and bone marrow formation was delayed (Figs. 7D and 7J). Moreover, whereas the secondary ossification center was smaller compared with the same tissue from the control group, in some cases, there was an absence of formation of secondary ossification centers (Figs. 7D and 7J). Examination of the calvarial bone indicated delayed growth resulting in a bone structure that was thinner than in control calvaria. Cbfa1mu expression also affected the morphology of developing teeth in the mouse mandibles. Thus, tooth germs in Cbfa1mu-infected mouse mandible appeared smaller in size and less mature in shape. Normally, tooth germs are surrounded by bone spicules, which will develop into alveolar bone. In the Cbfa1mu-infected mice, there was an apparent decrease in the amount of bone tissue around the tooth germ (Fig. 7L).

thumbnail image

Figure FIG. 7.. Immunohistochemical staining of BSP and OPN in tissues of infected BSP/TVA mice. Photographs of immunostained tissues from BSP/TVA mice infected with RCAS-Cbfa1mu are presented in the right panels and those from mice tissues infected with an empty vector are listed in the left panels. A-F are stained for BSP, and G-N are stained for OPN. Morphological analysis of tissue sections revealed delayed (B, D, H, and J) bone formation and (F and L) tooth development in response to RCAS-Cbfa1mu. Calvarial bone was thinner, and staining for (H) OPN and particularly (B) BSP (black arrows) was clearly reduced by expression of Cbfa1mu. (D and J) In the growth plate of the tibia, there was an accumulation of cartilage cells in the growth plate area (*), and there were less bone trabeculae formed in primary spongiosa in response to the expression of the mutated Cbfa1. Also, the secondary ossification centers could not be seen, whereas the ossification centers were well developed in the corresponding tibias from the control group (C and I). Immunostaining of (C) BSP and (I) OPN was strong in the primary spongiosa (arrows) of the control tissues, whereas staining of both (D) BSP and (J) OPN was restricted to the reduced primary spongiosa (arrows) in the Cbfa1mu expression tibias. Tooth germs in Cbfa1mu-infected mouse mandible (F and L) were smaller in size and less mature in shape and there was less supporting bone staining for BSP (arrows, F) in the mandible Cbfa1mu infected mice, compared with controls (arrows, E). OPN staining (arrow, L) was also more restricted in the tissues associated with the tooth germs of Cbfa1mu-infected mice compared with control tissues (K). Staining for OPN (M and N) was also markedly reduced in the alveolar bone expressing Cbfa1mu. All the tissue sections were counterstained with hematoxylin. Magnification, ×56.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

BSP is a unique marker for bone cell differentiation during osteogenesis and is believed to play an important role in bone formation and remodeling. In this study, we established that a BSP/TVA construct in a transgenic mouse model can be used to selectively express the viral receptor in bone-forming cells; the tissue-specific expression of the TVA receptor being consistent with the expression of endogenous BSP. The studies also confirm that tissue specificity lies within 4.9 kb of the murine BSP promoter. By showing the tissue-specific expression of a mutated Cbfa1/Runx2 gene, these studies also show that targeted expression of genes packaged in a viral construct can be achieved. This provides a powerful approach for the analysis of biological events occurring during early bone formation. Expression of the Cbfa1mu resulted in a marked downregulation of BSP expression, with a delayed effect on OPN. The effects on bone protein expression were accompanied by delayed bone formation and tooth development.

The RCAS/tva system, which allows cell type-specific gene transfer in mice was initially described to study the effect of overexpression of beta fibroblast growth factor (bFGF) in normal astrocytes using the astrocyte-specific glial fibrillary acidic protein promoter to generate glia specific transfer of the FGF.(32) In this system, a transgenic mouse line (Gtv-a) producing the receptor for the ALV-A receptor (TVA) driven by the astrocyte-specific promoter was developed to allow selective infection by replication-competent avian leukosis virus (ALV) splice acceptor (RCAS) viral vectors, derived from the avian retrovirus, ALV subgroup A. The same approach has also been used to express mutant epidermal growth factor receptor (EGFR) in glial precursors and astrocytes to show that constitutively active EGFR induces glioma-like lesions in mice in combination with the disruption of G1 cell-cycle arrest pathways.(38) The TVA system has also been used to develop a murine ovarian cancer cell model using cells from TVA transgenic mice deficient for p53.(34) Whereas used initially in cancer studies, the use of the system has been shown more recently to study normal biological processes. Thus, transgenic mouse lines that express TVA in SOX10-expressing NC stem cells under the control of the Pax3 promoter were generated to assess genetic hierarchies in NC development,(39) whereas an enhancer 3′ of the scl gene has been used to direct transgene expression to hematopoietic progenitors and stem cells.(40) Transgenic mice with TVA in either nestin-expressing neural precursor cells or dopachrome tautomerase-expressing melanoblasts have been used to overstimulate Wnt signaling,(41) whereas transgenic mice with the dopachrome tautomerase-expressing melanoblasts have also been used to show that the RCAS-TVA system can be used for in utero infection.(42)

In our studies, we have generated a transgenic mouse line in which the TVA receptor is selectively expressed in osteogenic cells using the BSP promoter to determine tissue-specific expression. The bone-specific expression of BSP has been established in previous studies,(17, 20, 43) which have shown that its expression is restricted in a temporo-spatial manner to differentiated osteoblasts. Thus, expression of BSP is absent in osteoblast progenitors but is expressed at high levels by newly formed osteoblasts coincident with the initial formation of the mineralized bone matrix.(20) Recently we have found that, whereas a 2.7-kb promoter sequence provided a good degree of tissue-specific expression,(21) selective expression in osteogenic tissues requires additional upstream sequences encompassed in a 4.9-kb promoter. Consequently, we used the 4.9-kb promoter to drive TVA expression. Although only one BSP/TVA founder was generated, Southern hybridization and PCR analysis confirmed the integration of the BSP/TVA gene into the mouse genome, whereas RT-PCR and in situ hybridization have shown the selective expression of the TVA receptor in osteoblastic cells of mice generated from the single founder. Moreover, because all the animals in this mouse line carry the same copy number of the TVA gene, infection and expression of ectopic genes should be comparable between animals. Consequently, the single mouse line has all the attributes required for the transgenic model. To test the efficacy of the model system for studying bone metabolism, these mice were infected with a construct expressing a mutated form of Cbfa1/Runx2 (Cbfa1mu), which had been used previously to analyze the effects of Cbfa1/Runx2 in differentiated osteoblasts in vivo.(11) Successful expression of the Cbfa1mu was indicated by the increase in Cbfa1/Runx2 levels in bone tissues after transfection. Because Cbfa1mu has a higher affinity for DNA than Cbfa1/Runx2 but has no transcriptional activity, it acts as a dominant-negative regulator of Cbfa1/Runx2.(11) When expressed in mice as a transgene under the control of the OCN promoter in mice, no effects on bone formation are apparent until after birth. Thereafter, an osteopenic phenotype develops because of a marked decrease in bone formation, which was also reflected in a dramatic reduction in the expression of bone proteins. In the Cbfa1mu-infected TVA/BSP mice, we also observed delayed development of bones and teeth and a marked reduction in BSP expression. Because OPN is directly regulated by Cbfa1/Runx2, the modest effect of the Cbfa1mu on OPN was surprising. However, this apparent discrepancy may be caused by temporal differences in the activities of the BSP promoter, which is driving TVA expression, and the OPN promoter. Notably, transcriptional regulation of the BSP gene begins with the onset of mineralization, whereas high expression of OPN occurs after the initial formation of mineral. Additionally, in the TVA system, ectopic expression is restricted to a narrow time frame dictated by the time of infection. In this regard, we were unable to show any marked effects on OCN, which is also a direct target of Cbfa1/Runx2, and is downregulated when the Cbfa1mu is expressed under an OCN promoter.(35) However, whereas OCN, like OPN, is also largely expressed after BSP, further studies are required to more clearly show the basis of these differences, which may reflect advantages of the TVA system in providing a more discrete time period of analysis which can be varied by using different infection times.

Because the molecular mechanisms controlling bone extracellular matrix production and bone formation by differentiated osteoblasts in postnatal animals are unknown, the establishment of transgenic mice with the tva receptor gene controlled by the BSP promoter has considerable potential to facilitate these and other studies. First, it is an in vivo model providing an environment where gene expression and regulation can be analyzed in real time. Second, this transgenic mouse line will facilitate studies of extracellular matrix protein gene expression by individual regulatory molecules and the sequelae of the effects on bone formation in vivo. Third, it will be possible to perform studies that are otherwise compromised by embryonic lethality or premature death of animals deficient in important gene(s). Thus, this system is being used to study the effects of other regulatory factors, including Osx, Dlx5, AJ-18, and TGF-β, on bone development and remodeling at different developmental stages.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

This work was supported by National Institutes of Health Grants DE11088 and DE14537 to JC and CIHR Grant 37785 to JS.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  • 1
    Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu Y, Bronson RT, Gao YH, Inada M, Sato M, Okamoto R, Kitamura Y, Yoshiki S, Kishimoto T 1997 Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89: 755764.
  • 2
    Otto F, Thornell AP, Crompton T, Denzel A, Gilmour KC, Rosewell IR, Stamp GW, Beddington RS, Mundlos S, Olsen BR, Selby PB, Owen MJ 1997 Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89: 765771.
  • 3
    Nakashima K, Zhou X, Kunkel G, Zhang Z, Deng JM, Behringer RR, de Crombrugghe B 2002 The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell 108: 1729.
  • 4
    Jiang H, Sodek J, Karsenty G, Thomas H, Ranly D, Chen J 1999 Expression of core binding factor Osf2/Cbfa-1 and bone sialoprotein in tooth development. Mech Dev 81: 169173.
  • 5
    D'Souza RN, Aberg T, Gaikwad J, Cavender A, Owen M, Karsenty G, Thesleff I 1999 Cbfa1 is required for epithelial-mesenchymal interactions regulating tooth development in mice. Development 126: 29112920.
  • 6
    Frendo JL, Xiao G, Fuchs S, Franceschi RT, Karsenty G, Ducy P 1998 Functional hierarchy between two OSE2 elements in the control of osteocalcin gene expression in vivo. J Biol Chem 273: 3050930516.
  • 7
    Sato M, Morii E, Komori T, Kawahata H, Sugimoto M, Terai K, Shimizu H, Yasui T, Ogihara H, Yasui N, Ochi T, Kitamura Y, Ito Y, Nomura S 1998 Transcriptional regulation of osteopontin gene in vivo by PEBP2alphaA/CBFA1 and ETS1 in the skeletal tissues. Oncogene 17: 15171525.
  • 8
    Javed A, Barnes GL, Jasanya BO, Stein JL, Gerstenfeld L, Lian JB, Stein GS 2001 runt homology domain transcription factors (Runx, Cbfa, and AML) mediate repression of the bone sialoprotein promoter: Evidence for promoter context-dependent activity of Cbfa proteins. Mol Cell Biol 21: 28912905.
  • 9
    Barnes GL, Javed A, Waller SM, Kamal MH, Hebert KE, Hassan MQ, Bellahcene A, Van Wijnen AJ, Young MF, Lian JB, Stein GS, Gerstenfeld LC 2003 Osteoblast-related transcription factors Runx2 (Cbfa1/AML3) and MSX2 mediate the expression of bone sialoprotein in human metastatic breast cancer cells. Cancer Res 63: 26312637.
  • 10
    Kiyoshima T, Yamauchi M, Wong C, Jheon A, Ganss B, Sodek J 2002 An L1 element disrupts human bone sialoprotein promoter: Lack of tissue-specific regulation by distalless5 (Dl × 5) and runt homeodomain protein2 (Runx2)/core binding factor a1 (Cbfa1) elements*1. Gene 299: 205217.
  • 11
    Liu W, Toyosawa S, Furuichi T, Kanatani N, Yoshida C, Liu Y, Himeno M, Narai S, Yamaguchi A, Komori T 2001 Overexpression of Cbfa1 in osteoblasts inhibits osteoblast maturation and causes osteopenia with multiple fractures. J Cell Biol 155: 157166.
  • 12
    Oldberg A, Franzen A, Heinegard D, Pierschbacher M, Ruoslahti E 1988 Identification of a bone sialoprotein receptor in osteosarcoma cells. J Biol Chem 263: 1943319436.
  • 13
    Fisher LW, McBride OW, Termine JD, Young MF 1990 Human bone sialoprotein. Deduced protein sequence and chromosomal localization. J Biol Chem 265: 23472351.
  • 14
    Fisher LW, Robey PG, Young MF, Termine JD 1987 Bone glycoproteins. Methods Enzymol 145: 269289.
  • 15
    Young M, Ibaraki K, Kerr J, Heegard A 1993 Molecular and cellular biology of the major noncollagenous proteins in bone. In: NodaM (ed.) Celluar and Molecular Biology of Bone. Academic Press, San Diego, CA, USA, pp. 191234.
  • 16
    Franzen A, Heinegard D 1985 Isolation and characterization of two sialoprotein present only in bone calcified matrix. Biochem J 232: 715724.
  • 17
    Ecarot-Charrier B, Bouchard F, Delloye C 1989 Bone sialoprotein II synthesized by cultured osteoblasts contains tyrosine sulfate. J Biol Chem 264: 2004920053.
  • 18
    Oldberg A, Franzen A, Heinegard D 1988 The primary structure of a cell-binding bone sialoprotein. J Biol Chem 263: 1943019432.
  • 19
    Hunter GK, Goldberg HA 1993 Nucleation of hydroxyapatite by bone sialoprotein. Proc Natl Acad Sci USA 90: 85628565.
  • 20
    Chen J, Shapiro HS, Sodek J 1992 Development expression of bone sialoprotein mRNA in rat mineralized connective tissues. J Bone Miner Res 7: 987997.
  • 21
    Chen J, Thomas HF, Jin H, Jiang H, Sodek J 1996 Expression of rat bone sialoprotein promoter in transgenic mice. J Bone Miner Res 11: 654664.
  • 22
    Chen JJ, Jin H, Ranly DM, Sodek J, Boyan BD 1999 Altered expression of bone sialoprotein in vitamin D-deficient rBSP2.7Luc transgenic mice. J Bone Miner Res 14: 221229.
  • 23
    Ogbureke KU, Fisher LW 2004 Expression of SIBLINGs and their partner MMPs in salivary glands. J Dent Res 83: 664670.
  • 24
    Bellahcene A, Castronovo V 1995 Increased expression of osteonectin and osteopontin, two bone matrix proteins, in human breast cancer. Am J Pathol 146: 95100.
  • 25
    Fedarko NS, Fohr B, Robey PG, Young MF, Fisher LW 2000 Factor H binding to bone sialoprotein and osteopontin enables tumor cell evasion of complement-mediated attack. J Biol Chem 275: 1666616672.
  • 26
    Zhang JH, Tang J, Wang J, Ma W, Zheng W, Yoneda T, Chen J 2003 Over-expression of bone sialoprotein enhances bone metastasis of human breast cancer cells in a mouse model. Int J Oncol 23: 10431048.
  • 27
    Zhang JH, Wang J, Tang J, Barnett B, Dickson J, Hahsimoto N, Williams P, Ma W, Zheng W, Yoneda T, Pageau S, Chen J 2004 Bone sialoprotein promotes bone metastasis of a non-bone-seeking clone of human breast cancer cells. Anticancer Res 24: 13611368.
  • 28
    Chen J, Jiang H, Sodek J, Karsenty G, Thomas H, Ranly D 1999 Comparison of expression patterns of core binding factor (Osf2/Cbfa1) and bone sialoprotein (BSP). In: GoldbergM, BoskeyA, RobinsonC (eds.) Chemistry and Biology of Mineralized Tissues. American Academy of Orthopaedic Surgeons, Rosemont, IL, USA, pp. 149154.
  • 29
    Benson MD, Aubin JE, Xiao G, Thomas PE, Franceschi RT 1999 Cloning of a 2.5 kb murine bone sialoprotein promoter fragment and functional analysis of putative Osf2 binding sites. J Bone Miner Res 14: 396405.
  • 30
    Yamauchi M, Sasaguri K, Hasegawa E, Sato S, Umezawa A 1999 Transcriptional regulation of bone sialoprotein gene in differentiated osteoblast cell line derived from normal mouse bone marrow and its involvement of Cbfa1/Pebp2αA1. Bull Kanagawa Dent Col 27: 911.
  • 31
    Tu Q, Yamauchi M, Pageau SC, Chen JJ 2004 Autoregulation of bone sialoprotein gene in pre-osteoblastic and non-osteoblastic cells. Biochem Biophys Res Commun 316: 461467.
  • 32
    Holland EC, Varmus HE 1998 Basic fibroblast growth factor induces cell migration and proliferation after glia-specific gene transfer in mice. Proc Natl Acad Sci USA 95: 12181223.
  • 33
    Sasaguri K, Ganss B, Sodek J, Chen JK 2000 Expression of bone sialoprotein in mineralized tissues of tooth and bone and in buccal-pouch carcinomas of Syrian golden hamsters. Arch Oral Biol 45: 551562.
  • 34
    Orsulic S, Li Y, Soslow RA, Vitale-Cross LA, Gutkind JS, Varmus HE 2002 Induction of ovarian cancer by defined multiple genetic changes in a mouse model system. Cancer Cell 1: 5362.
  • 35
    Bates P, Young JA, Varmus HE 1993 A receptor for subgroup A Rous sarcoma virus is related to the low density lipoprotein receptor. Cell 74: 10431051.
  • 36
    Ducy P, Starbuck M, Priemel M, Shen J, Pinero G, Geoffroy V, Amling M, Karsenty G 1999 A Cbfa1-dependent genetic pathway controls bone formation beyond embryonic development. Genes Dev 13: 10251036.
  • 37
    Kingston R, Chen C, Okayama H 1989 Introduction of DNA into eukaryotic cells. In: AusubelF, BrentR, KinstonR, MooreD, SeidmanJ, SmithJ, StruhlK (eds.) Current Protocols in Molecular Biology, vol. 1. Wiley, New York, NY, USA, pp. 911919.
  • 38
    Holland EC, Hively WP, DePinho RA, Varmus HE 1998 A constitutively active epidermal growth factor receptor cooperates with disruption of G1 cell-cycle arrest pathways to induce glioma-like lesions in mice. Genes Dev 12: 36753685.
  • 39
    Hou L, Loftus SK, Incao A, Chen A, Pavan WJ 2004 Complementation of melanocyte development in SOX10 mutant neural crest using lineage-directed gene transfer. Dev Dyn 229: 5462.
  • 40
    Murphy GJ, Gottgens B, Vegiopoulos A, Sanchez MJ, Leavitt AD, Watson SP, Green AR, Frampton J 2003 Manipulation of mouse hematopoietic progenitors by specific retroviral infection. J Biol Chem 278: 4355643563.
  • 41
    Dunn KJ, Williams BO, Li Y, Pavan WJ 2000 Neural crest-directed gene transfer demonstrates Wnt1 role in melanocyte expansion and differentiation during mouse development. Proc Natl Acad Sci USA 97: 1005010055.
  • 42
    Dunn KJ, Incao A, Watkins-Chow D, Li Y, Pavan WJ 2001 In utero complementation of a neural crest-derived melanocyte defect using cell directed gene transfer. Genesis 30: 7076.
  • 43
    Arvilommi AM, Salmi M, Kalimo K, Jalkanen S 1996 Lymphocyte binding to vascular endothelium in inflamed skin revisited: A central role for vascular adhesion protein-1 (VAP-1). Eur J Immunol 26: 825833.