Introduction

The genetic backgrounds of animals influence their morphological and behavioral characteristics, physiological functions, and pathological processes^1,2,3,4,5,6. Skeletal development is affected by the animal genotypes. Linkhart et al. reported two inbred mouse strains (C57BL/6J and C3H/HeJ) with large differences in femoral bone density and medullary cavity volume. The lower density and larger medullary cavity volume in C57BL/6J mice result from either decreased bone formation or increased bone resorption or both. The C57BL/6J mice have lower numbers of osteoblast progenitor cells and higher numbers of osteoclasts compared with C3H/HeJ mice⁷. The greater bone area in the high-density C3H mice compared to the low-density B6 mice was, in part, due to the greater periosteal and endosteal bone formation rates during growth in the C3H mice⁸. Shultz et al. have shown that some quantitative trait locus controlled the femoral bone mineral density in congenic strains of mice⁹. Mori et al. reported that a deficit in Fas-mediated apoptosis might enhance ectopic bone formation in mice¹⁰. Significant phenotypic variation at the femoral neck site exists between the two inbred rat strains Copenhagen 2331 (COP) and Dark Agouti (DA), and COP rats appear to have genes that specifically enhance the femoral neck structural properties and strength¹¹. Wergedal et al. studied the relationships of bone size, bone strength, and bone formation in two strains of mice, NZB/B1NJ and RF/J¹². Overall, previous studies indicate that the genetic backgrounds influence bone quality in animals.

As recombinant replication-deficient adenoviral vectors may infect a wide variety of dividing and quiescent cells and possess a large packaging capacity, they have been widely used in basic studies and clinical trials^{13,14,15,16,17,18,19}. However, one of the major limitations in the use of these vectors for gene therapy is the immune response initiated by adenoviral vectors. The immune response may interrupt the osteogenic target gene functions via two mechanisms. First, the viral antigens plus those derived from transgene expression in transduced cells contribute to immune responses leading to the destruction of these cells²⁰. Production of anti-adenovirus antibodies, as well as the cellular immune response and the early nonspecific clearance of the vectors, constitutes a barrier to successful gene therapy²¹. Second, bone formation is under the control of cytokines as well as growth factors such as bone morphogenetic proteins (BMP). This suggests the possibility that osteogenesis might be modulated by factors that also modulate the immune system. Most BMP genes delivered with first-generation E1- and E3-deleted adenoviral vectors have been shown to have very high osteogenic potential in immune-deficient animals. However, only some BMP adenoviral vectors, which were injected directly into muscle, have induced bone formation in immunocompetent animals^22,23,24,25. Although recombinant BMP adenoviral vectors with E2a or all viral protein-encoding genes deleted have been tested, they did not improve the osteogenic potential of BMP vectors in immunocompetent Sprague-Dawley rats^26,27. As the immune system may play a role in bone formation, the function of BMP genes delivered with adenoviral vectors may be affected by the different genetic backgrounds of animals.

In this experiment, our first aim was to test whether recombinant first-generation BMP adenoviral vectors have different osteogenic potential among different immunocompetent rat strains. The second aim was to test the levels of factors that may affect efficiency of adenoviral infection and expression in immunocompetent rats. These factors include the percentages of CD4⁺, CD8⁺, and IL-2⁺ T cells; the titers of adenoviral neutralizing antibodies; and the efficiency of target gene expression in primary muscle tissue culture of different rat strains. The recombinant human BMP9 (ADCMVBMP9) and BMP4 (ADCMVBMP4) adenoviral vectors were respectively used as positive and negative vectors based on their activities in Sprague-Dawley rats. Three outbred rat strains (Wistar, Long-Evans, and Sprague-Dawley) and three inbred rat strains (ACI, PVG, and Fischer344) were chosen for this experiment because of their different genetic backgrounds³. The size of ectopic bone was used to evaluate the osteogenic potential of recombinant human BMP adenoviral vectors.

Results

The basic T cell levels in different rat strains

We measured the basic T cell differentiation levels in five rats per group before viral injection. Fig. 1 shows basic T cell levels before viral injection in different rat strains. The mean percentages of CD4⁺, CD8⁺, and IL-2⁺ in total white blood cells are respectively 36.0, 20.7, and 2.3 in Wistar; 43.6, 11.5, and 2.3 in Long-Evans; 47.8, 32.7, and 2.4 in Sprague-Dawley; 42.9, 27.4, and 3.0 in ACI; 53.3, 7.76, and 3.1 in PVG; and 43.7, 11.3, and 3.7 in Fischer344 rats. The levels of CD4⁺ and CD8⁺ T cells are all significantly different among the different rat strains (P < 0.01 among groups, based on one-way ANOVA). The levels of IL-2⁺ T cells did not differ significantly among the strains (P > 0.05). This result indicates that different rat strains have different basic T cell levels.

Figure 1.

The basic T cell levels in different rat strains. All of the data come from 2-month-old rats. W, Wistar rats; LE, Long-Evans rats; SD, Sprague-Dawley rats; ACI, ACI rats; PVG, PVG rats; F, Fischer 344. Error bars represent standard error. N = 5 for each rat strain. The T cell levels of CD4⁺, CD8⁺, and IL2⁺ cells are shown.

Full figure and legend (114K)

The titers of adenoviral neutralizing antibody in different rat strains

The titer of adenoviral neutralizing antibody is defined as the serum dilution at 97% adenovirus inhibition. All of the titers of adenoviral neutralizing antibody in different rat strains before viral injection were lower than 40 times dilution (the lowest serum dilution). The mean titers of adenoviral neutralizing antibody in different rat strains 30 days post-viral injection were 79.6 16.1 in Wistar, 287.8 80.6 in Long-Evans, 199.4 44.3 in Sprague-Dawley, 230.8 44.4 in ACI, 142.4 9.65 in PVG, and 106.2 18.36 in Fischer344 rats (Fig. 2). The P value is 0.019 among groups. The result shows that rat strains have different potential to produce adenoviral neutralizing antibody after viral injection.

Figure 2.

The titers of adenoviral neutralizing antibody in different rat strains 30 days post-viral injection. The titer of adenoviral neutralizing antibody represents 97% inhibition of ADCMVGFP. W, Wistar rats; LE, Long-Evans rats; SD, Sprague-Dawley rats; ACI, ACI rats; PVG, PVG rats; F, Fischer344 rats. Error bars represent standard error.

Full figure and legend (38K)

The efficiency of target gene expression in primary rat muscle cell culture

The muscle tissue sensitivity to adenoviral infection and CMV promoter expression efficiency in the muscle may affect the functions of recombinant ADCMVBMPs. The primary muscle tissue cell culture might closely reflect the muscle tissue condition in vivo. We infected these cells with recombinant luciferase adenovirus (ADCMVLUC). The luciferase activity in primary muscle cell culture may represent the index that includes adenoviral infection sensitivity and CMV promoter expression efficiency. The relative luciferase activity in primary muscle cell culture 5 days post-viral infection was (15.8 4.7) 10⁵ relative light units (RLU)/g in Wistar, (1.4 0.3) 10⁵ RLU/g in Long-Evans, (9.0 2.6) 10⁵ RLU/g in Sprague-Dawley, (1.0 0.2) 10⁵ RLU/g in ACI, (8.5 2.5) 10⁵ RLU/g in PVG, and (8.7 1.7) 10⁵ RLU/g in Fischer344 (P < 0.01, Fig. 3). The results indicate that the expressions of target genes delivered with adenoviral vectors into primary muscle cell culture are significantly different among rat strains.

Figure 3.

The relative luciferase activities in primary muscle cell cultures of different rat strains post-ADCMVLUC infection. The luciferase activity in primary muscle cell culture is expressed as RLU/g. W, Wistar rats; LE, Long-Evans rats; SD, Sprague-Dawley rats; ACI, ACI rats; PVG, PVG rats; F, Fischer344 rats. Error bars represent standard error.

Full figure and legend (35K)

Bone volumes induced by recombinant BMP4 and BMP9 adenoviruses in different rat strains

None of the rats showed any bone formation in computerized tomography (CT) scans of ADCMVBMP4 injection sites. However, all rats had bone formation in ADCMVBMP9 injection sites 30 days post-viral injection. Fig. 4 shows 2D and 3D CT pictures of ectopic bone formation in different rat strains. Fig. 5 shows ADCMVBMP9-induced bone sizes in different rat strains. The mean volumes of bone induced with ADCMVBMP9 were 0.87 0.2 cm³ in Wistar, 0.26 0.1 cm³ in Long-Evans, 0.34 0.2 cm³ in Sprague-Dawley, 0.44 0.1 cm³ in ACI, 0.66 0.2 cm³ in PVG, and 0.58 0.1 cm³ in Fischer344 rats. The P value was 0.02 among groups using ANOVA single-factor statistical analysis. Multiple comparisons at = 0.05 using Duncan's multiple-range test²⁸ found that the Wistar bone volume was significantly different from that of ACI, Sprague-Dawley, and Long-Evans. Long-Evans also had volumes significantly different from those of PVG. This indicates that the different immunocompetent rat strains have different bone formation potential.

Figure 4.

ADCMVBMP9-induced ectopic bone (arrow) in different rat strains on CT scans. (A) Reconstructed 3D image from Wistar rat. (B) Reconstructed 3D image from Sprague-Dawley rat. (C) Reconstructed 3D image from Long-Evans rat. (D) 2D image from ACI rat. (E) 2D image from PVG rat. (F) 2D image from Fischer344 rat.

Full figure and legend (211K)

Figure 5.

The mean new bone volumes induced with ADCMVBMP9 in different rat strains. The volumes of new bone in different rat strains were induced on day 30 after thigh musculature injection of 3 10⁷ PFU of ADCMVBMP9. W, Wistar rats; LE, Long-Evans rats; SD, Sprague-Dawley rats; ACI, ACI rats; PVG, PVG rats; F, Fischer344 rats. Error bars represent standard error.

Full figure and legend (45K)

Histology of ADCMVBMP9-induced bone formation

Slides from the different rat strains show similar histology. A large number of cells are surrounded by matrix that was shown as blue color in Masson's trichrome staining and has spaces known as lacunae. Histologically the cells appear to be osteocytes. Fig. 6 shows the typical bone tissue formed in ADCMVBMP9-injected muscle sites.

Figure 6.

ADCMVBMP9-induced bone formation in muscle viral injection sites. Tissues from the different rat strains show similar histology. All slides have a large amount of osteocytes and collagen. (A) Wistar rat (H&E staining). (B) Sprague-Dawley rat (Masson's trichrome staining). (C) ACI rat (H&E staining). (D) PVG rat (H&E staining). Original magnification: 12 20. Arrows show osteocytes.

Full figure and legend (181K)

Discussion

The BMP9 adenoviral vector had different osteogenesis potential in the different rat strains. No difference was observed with ADCMVBMP4 because it failed to induce bone formation in any strain. The immune factors and efficiency of viral infection and target gene expression may affect ectopic bone formation. The immune factors include the cellular immune response and humoral immune response. Because the replication-deficient adenovirus induces a strong immune response in immunocompetent animals, the immune response may affect the target gene expression and function. T cells are important mediators of cytotoxicity and help to coordinate cell-mediated and humoral immune responses. CD4⁺ T cells are helper cells in the immune response. CD8⁺ T cells play a very important role among the cytotoxic T lymphocytes. CD4⁺ and CD8⁺ T cells all play important roles in adenovirus-induced immune response²⁹. Interleukin-2 is secreted by activated T cells and one of the critical cytokines that control the proliferation and differentiation of cells of the immune system^30,31. The percentages of these three cell types were measured in this experiment to assess the animals' basic immune ability. The levels of IL-2⁺ T cells were very similar among rat strains. According to the results, the basic levels of CD4⁺ and CD8⁺ T cells were significantly different among rat strains. Levels of these cell types in Sprague-Dawley rats were higher than in Wistar rats. However, the cell levels in other rat strains varied for each cell type. The titers of adenoviral neutralizing antibody were increased in all rat strains 30 days post-viral injection. The Wistar rats had the lowest antibody titers, and the Long-Evans rats had the highest antibody titers. The adenoviral neutralizing antibodies may include antibodies to the three major components of the adenoviral capsid, hexon, penton base, and fiber^29,32,33.

The amount of foreign gene expression in vivo is a critical factor impacting its function. Both the viral receptor distribution and promoter activity in target cells may have large affects on expression of genes delivered by adenoviral vector. The coxsackievirus and adenovirus receptor (CAR) and the integrins _v3 and _v5 are considered to be the adenoviral receptors³⁴. The number of CAR in mature skeletal muscle fibers is lower than in immature muscle tissue^35,36,37,38 and does not actually reflect their susceptibility to viral transduction. The viral concentration in local sites may also significantly impact the cell's transduction ability. Measuring the promoter activity in vivo, which is not affected by other factors, is a technical challenge. In this experiment, the rat primary muscle tissue cultures were used directly to measure the adenoviral vector expression efficiency. There are three obvious advantages in this method. First, the primary muscle tissue cultures include all cell types in muscle tissue, such as muscle fibers, satellite cells, and different connective tissue cells. Second, the viral vector expression efficiency includes the viral transduction efficiency and promoter activity in this tissue. Third, the immune factors will not interrupt the result. The result should more accurately reflect the muscle tissue susceptibility in vivo than other methods. This method is a practical protocol to measure the efficiency of viral infection and target gene expression in the tissue.

The recombinant BMP4 adenoviral vector did not induce bone formation in any immunocompetent rat strain. In fact, the osteogenetic potential of BMP4 in animals may be affected by its forms and delivery vectors. Purified BMP4 combined with collagen or other vectors may induce bone in mice or rats^39,40,41. Recombinant BMP4 baculoviral vector may induce alkaline phosphatase activity in vitro⁴². A recombinant BMP4 retroviral vector stimulated bone formation in syngeneic rats using transduced stromal cells in a gelatin matrix⁴³. Direct recombinant BMP4 adenovirus muscle injection induced ectopic bone formation in athymic nude rats²³, but not in immunocompetent rats. However, cells infected ex vivo with ADCMVBMP4 may induce bone formation in Sprague-Dawley rats (data not shown). It is known that BMP4 through the Smad signaling pathway can inhibit TNF-mediated apoptosis, independent of the prosurvival activity of NF-B⁴⁴. Some adenoviral proteins may block responses to interferons, intrinsic cellular apoptosis, killing by CD8⁺ cytotoxic T lymphocytes, and killing by the death ligands TNF, Fas ligand, and TRAIL^16,45. Croxford et al. reported that mouse IL-10 has a different inhibitory effect in treating experimental allergic encephalomyelitis when delivered by recombinant IL-10 retroviral vector or adenoviral vector⁴⁶. This may indicate that the signal transduction pathway of BMP4 inducing bone formation in the immunocompetent animal was blocked by some factor stimulated by adenoviral vector.

The osteogenetic potential of recombinant BMP adenovirus in animals is determined by many factors. However, the efficiency of target gene expression is a critical factor. All of the indexes chosen in this experiment may affect the target gene expression delivered with adenoviral vectors. We ranked each index among the rat strains; indexes that may enhance bone formation induced with recombinant BMP adenoviral vectors (such as cell expression) were assigned positive numbers, and indexes that may decrease bone formation (such as CD4⁺ and CD8⁺ T cells and neutralizing antibody) were assigned negative numbers. Since six rat strains were used, the rank numbers 1–6 were used for each group. The mean numbers were used to express similar values. Table 1 represents the rank numbers of each index. All of the numbers in the same rat strain were added together to get the "overall" number. The overall number in each strain may reflect the capability of target gene expression in that rat strain. According to these overall numbers, the rat strains may be roughly divided in three groups. The first group includes the Wistar rats. The second group includes Fischer344 and PVG rats. The third group includes the Long-Evans, Sprague-Dawley, and ACI rats. If the predictive score is compared with the experimentally measured bone volumes, we find that most rat strains with a higher overall predictive rank formed larger volumes of ectopic bone. Of course, many refinements should be made to this method. For example, should we use a level of each index to replace the relative number to rank the index? Should the index weight a factor to distinguish the relative role in the process of bone formation? A number of studies are necessary to establish a digital diagnosis to assess the osteogenetic potential of adenoviral vector-delivered BMP gene in animals.

Table 1 - The rank orders of rat strains in each index.

Full table

Among these six rat strains, there are different osteogenetic potentials in response to BMP9 adenoviral vector injection. The basic levels of CD4⁺ and CD8⁺ cells, humoral immune response, and adenoviral expressive efficiency are significantly different among rat strains and may contribute to the differences in bone volume induced by the BMP9 adenoviral vector. Since BMP4 adenoviral vectors did not induce bone in any rat strain, it is unknown how these factors impact BMP4 adenoviral vectors.

Materials and methods

Rat strains and viral injection

Six immunocompetent rat strains including three outbred and three inbred strains were used in this experiment. The outbred rat strains include Wistar, Long-Evans, and Sprague-Dawley. The inbred strains include ACI, PVG, and Fischer344. All rats were purchased from Harlan (Harlan Sprague–Dawley, Inc., Indianapolis, IN) and five 2-month-old rats were used in each group. The animal protocol was approved by the University of Virginia Animal Research Committee and conformed to National Institutes of Health guidelines. The rats were anesthetized with ketamine HCl and xylazine (Penn Veterinary Supply, Inc., Lancaster, PA). All rats were bled by tail vein before viral injection and injected with 50 l ADCMVBMP4 (10⁸ PFU) in the left thigh musculature and 50 l ADCMVBMP9 (3.0 10⁷ PFU) in the right thigh musculature.

T cell differentiation analysis using flow cytometry

All rats were bled by tail vein before viral injection. One hundred microliters of blood was suspended in 1 ml of PBS. Eight milliliters of distilled water was added into the blood cell suspension to lyse the red blood cells. Nine hundred microliters of 10-fold concentrated PBS was immediately added into the cell suspension. The white blood cells were harvested after centrifugation and resuspended in 200 l of staining buffer that contained labeled mouse anti-rat CD4 (APC) and CD8 (TC) (Caltag Laboratories, Burlingame, CA) antibodies. The anti-rat IL-2 (R-PE) antibody (Biosource International, Camarillo, CA) was used to do intracellular cytokine staining. The cell suspension was washed after a 1-h reaction and analyzed by flow cytometry. The percentages of CD cells were based on the total white blood cells.

Adenoviral neutralization assay

Rat sera collected before and 30 days after viral injections were tested for adenoviral neutralizing antibody. The sera were serially diluted in 250 l DMEM with 10% FBS at a dilution factor of 2 from 20- to 640-fold and mixed with equal volumes of viral solution containing 500 green cell forming unit (GFU) of recombinant GFP adenovirus (ADCMVGFP). The mixture was incubated at 37°C for 1 h and added into 96-well plates containing confluent 293 cells. Four wells were used for each concentration with 100 l/100 GFU/well. The plates were incubated at 37°C and 5% CO₂ for 24 h. The total number of green cells in 4 wells of each concentration was counted for the three serum dilutions with the lowest numbers of green cells. The resulting green cell numbers were used in regression analyses. The curves and equations are used to calculate the neutralizing antibody titer in sera. The titer of adenoviral neutralizing antibody was defined as the serum dilution that resulted in 97% inhibition of ADCMVGFP. Comparison of titers among strains 30 days post-viral injection was done using one-way ANOVA.

The efficiency of target gene expression delivered with adenoviral vector in rat primary muscle cell culture

Rat calf muscles were collected from each rat in this experiment and the muscle tissue was placed into cell culture after mechanical and collagenase dissociation of the tissues⁴⁷. In brief, muscle tissue was washed three times in PBS and cut into small pieces. The tissue suspensions were transferred into tubes and centrifuged at low speed. The pellets were resuspended in 10 ml 2.5 mg/ml collagenase type 1 and incubated in a 37°C water bath for 30 min with shaking every 10 min. The collagenase-treated tissues were gently triturated after centrifugation at low speed to dissociate cells with minimal fragmentation of muscle fibers. The cell suspensions were split into 24-well plates and grown in DMEM with 10% FBS and 10⁷ PFU/ml ADCMVLUC. The cell culture was harvested 5 days after infection and lysed using 100 l lysis buffer (Promega, Madison, WI). The lysate was measured for luciferase activity (Promega) and protein concentration (g/ml) using the Bio-Rad protein assay kit (Bio-Rad, Hercules, CA). The luciferase activity in each sample was expressed as relative light units per microgram protein.

Measuring bone volumes with CT scans

All rats were scanned by CT on a Picker PQ 6000 scanner 30 days after viral injection. Axial images with a 1-mm collimation and 1-mm table increment were obtained with the standard algorithm with 130 kV, 100 mA, 2 s scan time, and a 40-mm image size. Three-dimensional reconstruction and bone volumes were performed with a Voxel Q workstation.

Histology

The rats were bled and euthanized 30 days after viral injection. The injection sites of thigh muscles were removed and fixed in 10% neutral-buffered formalin. The thigh muscle samples were decalcified and embedded in paraffin and sectioned. The slides were stained with hematoxylin and eosin and Masson's trichrome.

References

1. Melo, M. E., Gabaglia, C. R., Moudgil, K. D., Sercarz, E. E., and Quinn, A. (2002). Strain-dependent effect of nasal instillation of antigen on the immune response in mice. Isr. Med. Assoc. J 4: 902–907. | PubMed | ChemPort |
2. Buller, R. E., et al. (2002). A phase I/II trial of rAd/p53 (SCH 58500) gene replacement in recurrent ovarian cancer. Cancer Gene Ther. 9: 553–566. | Article | PubMed | ISI | ChemPort |
3. Troy Harker, K., and Whishaw, I. Q. (2002). Place and matching-to-place spatial learning affected by rat inbreeding (dark-agouti, Fischer 344) and albinism (Wistar, Sprague-Dawley) but not domestication (wild rat vs. Long-Evans, Fischer-Norway). Behav. Brain Res. 134: 467–477. | Article | PubMed |
4. Butler, N. S., Monick, M. M., Yarovinsky, T. O., Powers, L. S., and Hunninghake, G. W. (2002). Altered IL-4 mRNA stability correlates with Th1 and Th2 bias and susceptibility to hypersensitivity pneumonitis in two inbred strains of mice. J. ol. 169: 3700–3709. | ChemPort |
5. Drolet, G., Proulx, K., Pearson, D., Rochford, J., and Deschepper, C. F. (2002). Comparisons of behavioral and neurochemical characteristics between WKY, WKHA, and Wistar rat strains. Neuropsychopharmacology 27: 400–409. | Article | PubMed | ChemPort |
6. Tejani-Butt, S., Kluczynski, J., and Pare, W. P. (2003). Strain-dependent modification of behavior following antidepressant treatment. Prog. Neuropsychopharmacol. Biol. Psychiatry 27: 7–14. | Article | PubMed | ChemPort |
7. Linkhart, T. A., et al. (1999). Osteoclast formation in bone marrow cultures from two inbred strains of mice with different bone densities. J. Bone Miner. Res. 14: 39–46. | Article | PubMed | ISI | ChemPort |
8. Sheng, M. H., et al. (1999). Histomorphometric studies show that bone formation and bone mineral apposition rates are greater in C3H/HeJ (high-density) than C57BL/6J (low-density) mice during growth. Bone 25: 421–429. | Article | PubMed | ChemPort |
9. Shultz, K. L., et al. (2003). Congenic strains of mice for verification and genetic decomposition of quantitative trait loci for femoral bone mineral density. J. Bone Miner. Res. 18: 175–185. | Article | PubMed | ChemPort |
10. Mori, S., et al. (1997). Enhancement of ectopic bone formation in mice with a deficit in Fas-mediated apoptosis. Pathol. Int 47: 112–116. | PubMed | ChemPort |
11. Sun, Q., and Turner, C. H. (2003). Two inbred rat strains that differ substantially in hip fragility. Calcif. Tissue Int 15: 15.
12. Wergedal, J. E., Sheng, M. H., Ackert-Bicknell, C. L., Beamer, W. G., and Baylink, D. J. (2002). Mouse genetic model for bone strength and size phenotypes: NZB/B1NJ and RF/J inbred strains. Bone 31: 670–674. | Article | PubMed | ChemPort |
13. Breyer, B., et al. (2001). Adenoviral vector-mediated gene transfer for human gene therapy. Curr. Gene Ther. 1: 149–162. | Article | PubMed | ChemPort |
14. Ferber, D. (2001). Gene therapy. Safer and virus-free? Science 294: 1638–1642. | Article | PubMed | ISI | ChemPort |
15. Rajagopalan, S., et al. (2002). Phase I study of direct administration of a replication deficient adenovirus vector containing the vascular endothelial growth factor cDNA (CI-1023) to patients with claudication. Am. J. Cardiol. 90: 512–516. | Article | PubMed | ChemPort |
16. Wold, W. S., et al. (1999). Immune responses to adenoviruses: viral evasion mechanisms and their implications for the clinic. Curr. Opin. ol. 11: 380–386. | ChemPort |
17. Freytag, S. O., et al. (2002). Phase I study of replication-competent adenovirus-mediated double suicide gene therapy for the treatment of locally recurrent prostate cancer. Cancer Res. 62: 4968–4976. | PubMed | ISI | ChemPort |
18. Kuball, J., et al. (2002). Generating p53-specific cytotoxic T lymphocytes by recombinant adenoviral vector-based vaccination in mice, but not man. Gene Ther. 9: 833–843. | Article | PubMed | ChemPort |
19. Habib, N. A., et al. (2002). Adenovirus replication-competent vectors (KD1, KD3) complement the cytotoxicity and transgene expression from replication-defective vectors (Ad-GFP, Ad-Luc). Cancer Gene Ther. 9: 651–654. | Article | PubMed | ChemPort |
20. Ghazizadeh, S., Kalish, R. S., and Taichman, L. B. (2003). Immune-mediated loss of transgene expression in skin: implications for cutaneous gene therapy. Mol. Ther. 7: 296–303. | Article | PubMed | ChemPort |
21. Christ, M., et al. (1997). Gene therapy with recombinant adenovirus vectors: evaluation of the host immune response. Immunol. Lett 57: 19–25. | Article | PubMed | ChemPort |
22. Alden, T. D., et al. (1999). In vivo endochondral bone formation using a bone morphogenetic protein 2 adenoviral vector. Hum. Gene Ther. 10: 2245–2253. | Article | PubMed | ISI | ChemPort |
23. Jane, J. A. J., et al. (2002). Ectopic osteogenesis using adenoviral bone morphogenetic protein-4 and -6 gene therapy. Mol. Ther. 6: 464–470. | Article | PubMed | ISI | ChemPort |
24. Helm, G. A., et al. (2000). Use of bone morphogenetic protein-9 gene therapy to induce spinal arthrodesis in the rodent. J. Neurosurg 92: 191–196. | PubMed | ISI | ChemPort |
25. Varady, P., et al. (2001). Morphologic analysis of BMP-9 gene therapy-induced osteogenesis. Hum. Gene Ther. 12: 697–710. | Article | PubMed | ISI | ChemPort |
26. Li, J. Z., et al. (2001). In vivo activity of recombinant bone morphogenetic protein 9 gutless adenovirus. Mol. Ther. 3: S228.
27. Li, H. W., Li, J. Z., Amalfitano, A., Hankins, G. R., and Helm, G. A. (2002). Construction of [E1-, E2b-] adenoviral vectors with human bone morphogenetic protein-4 and -6 and their ability to induce ectopic bone formation in rats. Mol. Ther. 5: S373.
28. Duncan, D. B. (1955). Multiple range and multiple F tests. Biometrics 11: 1–42. | Article | ISI |
29. Van Ginkel, F. W., et al. (1997). Adenoviral gene delivery elicits distinct pulmonary-associated T helper cell responses to the vector and to its transgene. J. ol. 159: 685–693. | ChemPort |
30. Warren, H. S. (1996). NK cell proliferation and inflammation. Immunol. Cell Biol. 74: 473–480. | PubMed | ISI | ChemPort |
31. Gesbert, F., Delespine-Carmagnat, M., and Bertoglio, J. (1998). Recent advances in the understanding of interleukin-2 signal transduction. J. Clin. ol. 18: 307–320. | ChemPort |
32. Molinier-Frenkel, V., et al. (2000). Immune response to recombinant adenovirus in humans: capsid components from viral input are targets for vector-specific cytotoxic T lymphocytes. J. Virol. 74: 7678–7682. | Article | PubMed | ChemPort |
33. Gahery-Segard, H., et al. (1997). Humoral immune response to the capsid components of recombinant adenoviruses: routes of immunization modulate virus-induced Ig subclass shifts. Eur. J. ol. 27: 653–659. | ChemPort |
34. Cao, B., et al. (2001). The role of receptors in the maturation-dependent adenoviral transduction of myofibers. Gene Ther. 8: 627–637. | Article | PubMed | ChemPort |
35. Kimura, E., et al. (2001). Efficient repetitive gene delivery to skeletal muscle using recombinant adenovirus vector containing the coxsackievirus and adenovirus receptor cDNA. Gene Ther. 8: 20–27. | Article | PubMed | ChemPort |
36. Van Deutekom, J. C., et al. (1999). Extended tropism of an adenoviral vector does not circumvent the maturation-dependent transducibility of mouse skeletal muscle. J. Gene Med 1: 393–399. | Article | PubMed | ChemPort |
37. Nalbantoglu, J., et al. (2001). Muscle-specific overexpression of the adenovirus primary receptor CAR overcomes low efficiency of gene transfer to mature skeletal muscle. J. Virol. 75: 4276–4282. | Article | PubMed | ChemPort |
38. Nalbantoglu, J., Pari, G., Karpati, G., and Holland, P. C. (1999). Expression of the primary coxsackie and adenovirus receptor is downregulated during skeletal muscle maturation and limits the efficacy of adenovirus-mediated gene delivery to muscle cells. Hum. Gene Ther. 10: 1009–1019. | Article | PubMed | ChemPort |
39. Boden, S. D., et al. (1996). Differential effects and glucocorticoid potentiation of bone morphogenetic protein action during rat osteoblast differentiation in vitro. Endocrinology 137: 3401–3407. | Article | PubMed | ChemPort |
40. Hughes, F. J., Collyer, J., Stanfield, M., and Goodman, S. A. (1995). The effects of bone morphogenetic protein-2, -4, and -6 on differentiation of rat osteoblast cells in vitro. Endocrinology 136: 2671–2677. | Article | PubMed | ISI | ChemPort |
41. Yamaguchi, A., et al. (1996). Effects of BMP-2, BMP-4, and BMP-6 on osteoblastic differentiation of bone marrow-derived stromal cell lines, ST2 and MC3T3-G2/PA6. Biochem. Biophys. Res. Commun 220: 366–371. | Article | PubMed | ChemPort |
42. Hazama, M., Aono, A., Ueno, N., and Fujisawa, Y. (1995). Efficient expression of a heterodimer of bone morphogenetic protein subunits using a baculovirus expression system. Biochem. Biophys. Res. Commun 209: 859–866. | Article | PubMed | ISI | ChemPort |
43. Gysin, R., et al. (2002). Ex vivo gene therapy with stromal cells transduced with a retroviral vector containing the BMP4 gene completely heals critical size calvarial defect in rats. Gene Ther. 9: 991–999. | Article | PubMed | ChemPort |
44. Chen, S., et al. (2001). Suppression of tumor necrosis factor-mediated apoptosis by nuclear factor kappaB-independent bone morphogenetic protein/Smad signaling. J. Biol. Chem. 276: 39259–39263. | Article | PubMed | ISI | ChemPort |
45. Liu, Q., and Muruve, D. A. (2003). Molecular basis of the inflammatory response to adenovirus vectors. Gene Ther. 10: 935–940. | Article | PubMed | ChemPort |
46. Croxford, J. L., Feldmann, M., Chernajovsky, Y., and Baker, D. (2001). Different therapeutic outcomes in experimental allergic encephalomyelitis dependent upon the mode of delivery of IL-10: a comparison of the effects of protein, adenoviral or retroviral IL-10 delivery into the central nervous system. J. ol. 166: 4124–4130. | ChemPort |
47. Farr-Jones, M. A., Parney, I. F., and Petruk, K. C. (1999). Improved technique for establishing short term human brain tumor cultures. J. Neurooncol. 43: 1–10. | Article | PubMed | ChemPort |

Acknowledgements

We thank the Genetics Institute for providing BMP4 and BMP9 cDNAs. This work was supported in part by grants from Medtronic–Sofamor Danek, Inc., and the National Institutes of Health (Grant R01 AR46488-01A2, G.A.H.).