Osteosarcoma (OS) is the most common primary malignancy of bone. Here, we investigated a possible role of defective osteoblast differentiation in OS tumorigenesis. We found that basal levels of the early osteogenic marker alkaline phosphatase (ALP) activity were low in OS lines. Osteogenic regulators Runx2 and OSX, and the late marker osteopontin (OPN) expressed at low levels in most OS lines, indicating that most OS cells fail to undergo terminal differentiation. Furthermore, OS cells were refractory to osteogenic BMP-induced increases in ALP activity. Osteogenic BMPs were shown to upregulate early target genes, but not late osteogenic markers OPN and osteocalcin (OC). Furthermore, osteogenic BMPs failed to induce bone formation from human OS cells, rather effectively promoted OS tumor growth in an orthotopic OS model. Exogenous expression of early target genes enhanced BMP-stimulated OS tumor growth, whereas osteogenic BMP-promoted OS tumor growth was inhibited by exogenous Runx2 expression. These results suggest that alterations in osteoprogenitors may disrupt osteogenic differentiation pathway. Thus, identifying potential differentiation defects in OS tumors would allow us to reconstruct the tumorigenic events in osteoprogenitors and to develop rational differentiation therapies for clinical OS management.

The most commonly used adenoviral vectors are derived from human adenovirus serotypes 2 and 5. Adenoviruses are non-enveloped DNA viruses whose capsid is composed of pentons and hexons. The viral genome consists of 36-kb double-stranded linear DNA with inverted terminal repeat (ITR) sequences at each end. DNA of length greater than 38 kb cannot be efficiently packaged into competent viral particles. The adenovirus life cycle begins with the attachment of the fiber to cell surface receptors, such as coxsackievirus and adenovirus receptor, and the interaction of the pentons with αvβ3 and αvβ5 integrin proteins. Following receptor-mediated endocytosis, adenovirus escapes from the endosomes to the cytoplasm and translocates into the nucleus, where viral transcription and replication begin. Completion of the virus life cycle triggers cell death and the release of progeny viruses. On the basis of temporal expression relative to the onset of viral DNA replication, viral transcription units are conventionally referred to as early (E1a, E1b, E2, E3 and E4), delayed early (proteins IX and Iva2) or late genes (L1–L5). The early gene products are generally involved in viral gene transcription, DNA replication, host immune suppression and inhibition of host cell apoptosis, whereas the late gene products are required for virion assembly8.The generation and production of recombinant adenoviruses should be performed in a laboratory operating at Biosafety Level 2 (BL2), as approved by the researcher's institutional biosafety committee and following the National Institutes of Health Biosafety guidelines (http://www.cdc.gov/od/ohs/biosfty/bmbl4/bmbl4toc.htm). These requirements include the use of biosafety cabinet hoods, the establishment of proper procedures for decontamination and disposal of liquid and solid wastes and the disinfection of contaminated surfaces and equipment. In addition, users should conduct a regular analysis of adenovirus stocks to exclude the presence of replication-competent adenoviruses in their preparations.The AdEasy system described in this protocol exploits the high efficiency of homologous recombination in a specific bacterial strain coupled with selectable antibiotic resistance markers to simplify recombinant vector production4, 7, 20.Two adenoviral backbone vectors can be used for adenovirus production (Fig. 2b). The commonly used pAdEasy-1 is an E1 and E3 double-deletion vector. Hence, AdEasy-1-derived recombinant adenoviruses can be propagated in E1-expressing packaging cells, such as HEK-293 cells (E3 is not necessary for viral production). The pAdEasy-2 backbone vector is an E1, E3 and E4 triple-deletion vector7, 20. Thus, propagation of AdEasy-2-derived vectors requires the use of packaging cell lines expressing both E1 and E4. As the E4 gene product is toxic to mammalian cells, its expression is usually controlled by inducible promoters. Several such E1/E4 packaging lines are available. However, because of the leakiness of most inducible systems, high E4-expressing cells are often lost after serial passages, and it is difficult to generate high-titer AdEasy-2-derived viruses. Thus, in this protocol we focus on procedures for generating recombinant adenoviruses with the AdEasy-1 system. For investigators wishing to express particularly long (or multiple) transgenes, we recommend gutless systems. Unlike AdEasy vector systems, which can tolerate replacements in the E1 and E4 regions only, gutless systems can tolerate much larger replacements in the adenoviral backbone4, 5, 6, 11. Although the AdEasy system has been made freely available to thousands of academic researchers, the system can also be obtained through commercial sources listed on the AdEasy website (http://www.coloncancer.org/adeasy.htm). Note: As BJ5183 cells have a relatively high frequency of homologous recombination, unwanted or detrimental rearrangements and/or recombinations of the pAdEasy sequences in AdEasier cells can occur. For homemade AdEasier cells, it is thus important to pick individual clones and characterize the clones with extensive restriction digestions, for example with HindIII and/or PstI. The digestion patterns can be compared with the pAdEasy stock DNA made in a strain incapable of homologous recombination (e.g., DH10B or XL-1 blue). Note: All shuttle vectors confer resistance to kanamycin. Note: All cell culture work involved in adenovirus generation and subsequent amplification should be performed in accordance with BL2 guidelines. Note: Special precautions are needed when washing HEK-293 cells because the cells are often weakly adherent to the flask. One wash is usually sufficient. If multiple flasks are used for transfections, wash no more than five flasks at a time to minimize detachment of HEK-293 cells. Note: The transfected cells (and the subsequent cell cultures) should be treated as a BL2 biohazard and handled with care, as the medium may contain recombinant adenoviruses. Note: No obvious plaques or cytopathic effect (CPE) is generally observed by light microscopy up to 2 weeks after transfection. However, GFP plaques are usually observed under fluorescence microscopy from 5–7 d after transfection (Fig. 4a). The shapes of the fluorescent plaques depend on transfection efficiency. Low transfection efficiencies may produce 'comet-like' plaques, whereas high transfection efficiencies may produce an intense 'scattered stars' phenomenon. Note: We have generally been able to generate adenoviruses suitable for experimental use in tissue culture without plaque purification. For more exacting applications, however, it is advisable to perform plaque purification. Homogeneity of adenoviruses can be judged by restriction endonuclease digestion, paying careful attention to the ratio of the fragments obtained (which should be molar if the viruses are homogeneous). We have not seen evidence of recombination within more than 50 adenoviruses produced with the AdEasy system. However, we cannot exclude subtle changes in sequence that may occur during the cloning or viral production steps. If such subtle changes are an issue, the GOI can be amplified from the final viral preparation using PCR and the sequence determined using conventional means. Note: Perform infection under BL2 conditions. Note: The virus-containing waste should be disinfected with chlorine bleach. Note: Titers can be measured at any time, which is particularly easy with AdTrack-based vectors. Simply infect HEK-293 cells with various dilutions of viral supernatant and see how many are GFP-positive cells 24 h later. Without AdTrack, viruses can be plaque titered or titered by limiting dilution using standard methods7. After three rounds of amplification, viral titer should reach 109–1010 infectious particles [or plaque-forming units (pfu)] per milliliter of lysate. Note: The virus-containing waste should be disinfected with chlorine bleach.Step 1, cloning GOI into a shuttle vector: 3–7 dSteps 2–12, generating recombinant adenovirus plasmids: 4–5 dSteps 13–20, generating the initial stocks of adenoviruses: 14–20 dSteps 21–29, stepwise amplification for high-titer adenoviruses: 7–10 dStep 31, determining viral titers: 2–10 dTroubleshooting advice can be found in Table 1.The AdEasy system has proved to be an efficient and robust technology for generating recombinant adenoviruses in thousands of laboratories worldwide. The key step is the generation of recombinants in AdEasier cells. Once these recombinants are obtained, generation of adenoviruses in HEK-293 cells is virtually guaranteed23, 24, 25, 26, 27. Depending on the transfection efficiency of HEK-293 cells, one can expect titers in the initial lysate to be 108– 1010 pfu ml−1. It usually requires two to three rounds of further amplification in HEK-293 cells to achieve viral titers of 1012 or 1013 pfu ml−1. Cloning genes of interest into shuttle vectors entails standard molecular biology techniques. The user's experience in molecular cloning will determine the time required for this step.After subcloning GOI into a shuttle vector, the rate-limiting step is generally the production of the desired recombinants in AdEasier cells. With experience and high-quality competent AdEasier cells, it takes as few as 2 d to obtain and verify the recombinants.If every step works well, one can expect to obtain the initial virus lysates 2–3 weeks thereafter and the subsequent large-scale virus preparation in an additional 1–2 weeks (see TIMING). Further information can be found at the Internet resources listed in Box 1.

Wnt signaling plays an important role in regulating cell proliferation and differentiation. De-regulation of these signaling pathways has been implicated in many human diseases, ranging from cancers to skeletal disorders. Wnt proteins are a large family of secreted factors that bind to the Frizzled receptors and LRP5/6 co-receptors and initiate complex signaling cascades. Over the past two decades, our understanding of Wnt signaling has been significantly improved due to the identification of many key regulators and mediators of these pathways. Given that Wnt signaling is tightly regulated at multiple cellular levels, these pathways themselves offer ample nodal points for targeted therapeutics. Here, we focus on our current understanding of these pathways, the associations of Wnt signaling with human disorders, and the opportunities to target key components of Wnt signaling for rational drug discovery.

RNA interference (RNAi)-mediated gene silencing has become a valuable tool for functional studies, reverse genomics, and drug discoveries. One major challenge of using RNAi is to identify the most effective short interfering RNAs (siRNAs) sites of a given gene. Although several published bioinformatic prediction models have proven useful, the process to select and validate optimal siRNA sites for a given gene remains empirical and laborious. Here, we developed a fluorescence-based selection system using a retroviral vector backbone, namely pSOS, which was based on the premise that candidate siRNAs would knockdown the chimeric transcript between GFP and target gene. The expression of siRNA was driven by the opposing convergent H1 and U6 promoters. This configuration simplifies the cloning of duplex siRNA oligonucleotide cassettes. We demonstrated that GFP signal reduction was closely correlated with siRNA knockdown efficiency of human β-catenin, as well as with the inhibition of β-catenin/Tcf4 signaling activity. The pSOS should not only facilitate the selection and validation of candidate siRNA sites, but also provide efficient delivery tools of siRNAs via viral vectors in mammalian cells. Thus, the pSOS system represents an efficient and user-friendly strategy to select and validate siRNA target sites.