Local immunomodulation with CD4 and CD8 antibodies, but ... - Nature

1 downloads 0 Views 235KB Size Report
Jul 7, 2005 - This study was designed to see if immunosuppression achieved using local application of cyclosporine A (Cs. A) or CD4 and CD8 antibodies ...
Gene Therapy (2005) 12, 1235–1241 & 2005 Nature Publishing Group All rights reserved 0969-7128/05 $30.00 www.nature.com/gt

RESEARCH ARTICLE

Local immunomodulation with CD4 and CD8 antibodies, but not cyclosporine A, improves osteogenesis induced by ADhBMP9 gene therapy JZ Li1, H Li1,3, GR Hankins1, B Dunford1 and GA Helm1,2 1

Department of Neurological Surgery, University of Virginia Health System, Charlottesville, VA 22908, USA; and Department of Biomedical Engineering, University of Virginia Health System, Charlottesville, VA 22908, USA

2

This study was designed to see if immunosuppression achieved using local application of cyclosporine A (Cs. A) or CD4 and CD8 antibodies would improve bone formation following intramuscular injections of human BMP-4 and BMP-9 adenoviral vectors (ADhBMP4 and ADhBMP9) in Sprague–Dawley rats. Cs. A was injected into the thigh muscle. After 2 days, ADhBMP4, ADhBMP9, and the antibodies were separately injected into the left and right rear legs. At this time, the number of CD4+/CD3+ cells was significantly lower and the number of CD8+/CD3+ cells higher in the Cs. A group than in the control group (Po0.01). The total number of white blood cells 3 days following

injection of CD4 and CD8 antibodies was significantly lower than that before the injection (Po0.01). At 4 weeks after the viral and antibody injections, mean bone volumes at the ADhBMP9 treatment sites were 0.2970.01 cm3 in the viral control group, 0.1770.03 cm3 in the Cs. A-ADhBMPs group, and 0.5970.07 cm3 in the antibodies-ADhBMPs group. ADhBMP4 did not induce new bone formation in any group. This study demonstrates that local immunomodulation may improve the osteogenic potential of bone morphogenetic protein gene therapy in the clinical setting. Gene Therapy (2005) 12, 1235–1241. doi:10.1038/ sj.gt.3302502; published online 7 July 2005

Keywords: CD4 and CD8 antibodies; cyclosporine A; immunomulation; osteogenic; adenovirus; bone morphogenetic proteins

Introduction Bone morphogenetic proteins (BMPs) are members of the transforming growth factor b (TGF-b) superfamily. At present, more than 30 BMPs have been discovered.1 Some members of the BMP family are also known as osteogenic proteins (OPs), growth/differentiation factors (GDFs), and cartilage-derived morphogenetic proteins (CDMPs).2 Some BMPs have significant osteogenic potential in vitro and in vivo; indeed, BMP-2 has recently been approved by the Food and Drug Administration for use in spinal fusion procedures in humans.3,4 Nevertheless, many factors, such as the need for high doses and specialized carrier systems, have limited the use of BMPs in the daily practice of orthopedic and neurological surgery.5 Gene therapy techniques are being developed to enhance local expression of osteogenic BMPs and to induce local bone formation. The BMP genes that have been used in previous gene therapy studies include BMP-2, -4, -6, -7, and -9.6,7 The viral vectors used to deliver these genes, either directly or through ex vivo techniques, include adenoviruses, retroviruses, and Correspondence: Dr GA Helm, Department of Neurological Surgery, University of Virginia Health System, PO Box 800212, Charlottesville, VA 22908, USA 3 Current address: Department of Physiology and Functional Genomics, College of Medicine and McKnight Brain Institute, University of Florida, Gainesville, FL 32610, USA Received 13 May 2004; accepted 28 December 2004; published online 7 July 2005

adeno-associated viruses.8,9Adenoviral vectors containing any of the aforementioned BMP genes have been shown to have significant osteogenic potential in immunodeficient animals. In immunocompetent animals, however, the osteogenic activity of BMP adenoviral vectors is significantly reduced.6,10,11 Efforts are currently being directed toward improving the osteogenic potential of BMP gene therapy in immunocompetent animals. The development of immunosuppressive agents for organ transplantation may provide novel tools to improve the efficacy of BMP gene therapy.11–14 The use of short-term, local immunosuppression may improve the biologic activity of BMP vectors and reduce the side effects of immunosuppression because the bone formation process can be initiated by very brief exposures to BMP.15,16 From an immunological point of view, the T-cell defect in athymic nude animals is the major factor separating these animals from immunocompetent ones. Therefore, T-cell functions are critical factors that determine the different osteogenic potentials of BMP adenoviral vectors in immunodeficient and immunocompetent animals. This suggests that immunosuppressive agents that have an inhibitory effect on T cells may improve the efficacy of BMP gene therapy. Cyclosporine A (Cs. A) is a cyclic undecapeptide that was originally isolated from Tolypocladium inflatum Gams and is commonly used as an immunomodulatory agent in the treatment of patients receiving allogeneic organ transplants. The major pharmacological effect of Cs. A is the inhibition of early events in the activation of T cells by

Local immunomodulation improves osteogenesis JZ Li et al

1236

antigens. These T cells regulate B cells, macrophages, and other cells through the secretion of cytokines and chemokines.13,14,17,18 Therefore, Cs. A may be an ideal agent to improve bone formation induced by BMP adenoviral vectors. Cell surface markers are basic characteristics used to distinguish the various types of immune cells. CD4 is a surface marker that is mainly expressed by helper T cells, but is also expressed by macrophages, natural killer cells, and dendritic cells. CD8 is a surface marker of the cytotoxic T lymphocyte, one of the major cell types involved in the clearance of foreign agent-transduced cells. Local delivery of CD4 and CD8 antibodies may temporarily interrupt host immune responses to adenoviral vectors, increase the length and level of transgene expression, and lead to increased osteogenesis without significantly affecting the immune capabilities of the entire system. Our previous studies of human BMP-4 and BMP-9 adenoviral vectors (ADhBMP4 and ADhBMP9) have demonstrated that these vectors have similar osteogenic potentials in athymic nude rats, but their activities are significantly different in immunocompetent animals.6,19 ADhBMP4 does not induce bone formation in immunocompetent rats. ADhBMP9 can produce significant osteogenesis in these animals, but the new bone volumes are smaller than those generated in immunodeficient animals. The reasons for the difference in in vivo osteogenic activity between these two BMP vectors is not clearly known, but may be related to different types of receptors and signal transduction pathways.20–23 In the present study, we tested the effects of Cs. A and those of CD4 and CD8 antibodies on the osteogenic activity

Table 1

of ADhBMP4 and ADhBMP9 following a direct intramuscular injection in Sprague–Dawley rats.

Results To assess the activities of local delivery of immunosuppressive agents in rats, total numbers of white blood cells (WBCs), percentages of different T cells, and titers of adenovirus-neutralizing antibody were measured. Bone volumes were quantified using computerized tomography (CT) to assess the effects of two types of immunosuppressive agents on the process of bone formation induced by ADhBMP4 and ADhBMP9 in immunocompetent rats. A total of twenty-five 2-month-old Sprague–Dawley rats were used in this study. The rats were randomly separated into five groups with five rats in each group. Table 1 shows that the rats were treated and tested at different time points.

Analysis of WBCs The total number of WBCs was determined at different time points (Table 2). On day 6, the number of WBCs was significantly lower in the CD4 and CD8 antibodies group (Group 4; 58857760/ml) and in the antibodies plus ADhBMPs group (Group 5; 60507389/ml) than that determined on day 1 prior to treatment (84557603/ml in Group 4 (P ¼ 0.04) and 89457623/ml in Group 5 (Po0.01)). In the other groups and at different time points, the numbers were not significantly different and lay within the normal range. This result indicates that local intramuscular delivery of mouse anti-rat CD4 and

Animal groups, treatments, and tests at different time points

Animal group

Day 1

Day 3

Day 6

Day 10

Day 16

ADhBMP4 and ADhBMP9 (Group 1)

WBCsa PBSb Anti-ADe

WBCs CDsc ADhBMPsf

WBCs CDs

WBCs

WBCs

Cyclosporine A (Group 2)

WBCs Cs. Ag Anti-AD

WBCs CDs PBS

WBCs CDs

Cyclosporine A+ ADhBMP4 and ADhBMP9 (Group3)

WBCs Cs. A Anti-AD

WBCs CDs ADhBMPs

WBCs CDs

CD4 and CD8 antibodies (Group 4)

WBCs Anti-AD

CD4 and CD8 Antibodiesh

WBCs CDs

WBCs

Antibodies+ ADhBMP4 and ADhBMP9 (Group 5)

WBCs Anti-AD

CD4 andCD8 Antibodies +ADhBMPsi

WBCs CDs

WBCs

a

Anti-AD WBCs

CTd Anti-AD

WBCs Anti-AD

WBCs

Day 30

Anti-AD

WBCs Anti-AD

CT Anti-AD

WBCs Anti-AD

Anti-AD

WBCs Anti-AD

CT Anti-AD

WBCs: analysis of white blood cells. PBS: PBS was injected into the thigh musculature on both sides. c CDs: determination of levels of CD3+, CD4+/CD3+, and CD8+/CD3+. d CT: computerized tomography imaging. e Anti-AD: measurement of adenovirus-neutralizing antibody titers. f ADhBMPs: ADhBMP4 and ADhBMP9 were injected into the thigh musculature on the left and right sides, respectively. g Cs. A: cyclosporine A was injected into the thigh musculature on both sides. h CD4 and CD8 antibodies: the mixture of CD4 and CD8 antibodies was injected into the thigh musculature on both sides. i CD4 and CD8 Antibodies + ADhBMPs: the mixture of CD4 and CD8 antibodies was added to the ADhBMP4 or ADhBMP9 solution and injected into the thigh musculature on the left and right sides, respectively. b

Gene Therapy

Local immunomodulation improves osteogenesis JZ Li et al

1237

Table 2 Total numbers of WBCs (  100) per microliter in different groups at various time points Animal group ADhBMP4 and ADhBMP9 Cs. A Cs. A+ ADhBMP4 and ADhBMP9 CD4 and CD8 antibodies CD4 and CD8 antibodies+ ADhBMP4 and ADhBMP9

Day 1

Day 3

Day 6

Day 10

Day 16

113716 9778

97712 9074

100710 122711

143722 85710

115714 114720

98710 5978

108710 7973

10878 6574

6174

8171

7572

8372 8576

11477

9076

CD8 antibodies temporarily reduces the systemic number of WBCs and that the doses of adenoviral vectors used in this study do not induce a statistically significant change in the number of WBCs.

Percentages of CD3+, CD4+/CD3+, and CD8+/CD3+ cells The number of T cells was determined using antibodies against CD3, CD4, and CD8. On day 3, the percentages of CD3-positive (CD3+) cells in all lymphocytes were similar in the control (Group 1; 20.072.0%) and Cs. A (Group 2 & 3; 16.371.8%) groups. The percentage of CD4+/CD3+ cells was significantly lower in the Cs. A group than in the control group (Po0.01), but the percentage of CD8+/CD3+ cells was significantly higher in the Cs. A group than in the control group (Po0.01) (Figure 1a). On day 6, there was no statistical difference in the percentages of CD3+, CD4+/CD3+, and CD8+/ CD3+ cells among the five groups (Figure 1b). Treatment with CD4 and CD8 antibodies produced no significant effect on the percentages of lymphocyte subtypes. Titers of adenovirus-neutralizing antibody in different groups The titer of adenovirus-neutralizing antibody was defined as the serum dilution level at which there was a 97% inhibition of the adenovirus. All titers of adenovirus-neutralizing antibody measured in animals before viral injection and in animals after they were given Cs. A (Group 2) or antibodies (Group 4) were lower than 20 times dilution (the lowest serum dilution). The mean titers of adenovirus-neutralizing antibody on days 16 and 30 were, respectively, 289792 and 4347119 in the viral control group (Group 1), 244750 and 364751 in the Cs. A with ADhBMPs group (Group 3), and 103720 and 124724 in the CD antibodies with ADhBMPs group (Group 5) (Figure 2). On day 30, the titers of adenovirus-neutralizing antibodies were significantly lower in the group also given CD4 and CD8 antibodies (Group 5) than in the viral control group (Group 1). Bone volumes induced by ADhBMP4 and ADhBMP9 in different groups No bone formation in the thigh muscle was demonstrated in any rat in the negative control groups (Groups 2 and 4). In addition, there was no evidence of bone formation at the ADhBMP4 injection site in the rats in Groups 1, 3, and 5. Nevertheless, on day 30, bone formation was found at the ADhBMP9 injection site in all the rats in Groups 1, 3, and 5. Figure 3 shows two-

Figure 1 Effects of Cs. A and ADhBMPs on percentages of T cells in Sprague–Dawley rats. The percentage of CD3+ cells was determined by counting FITC+ cells and comparing their number with the total number of lymphocytes. The percentages of CD4+/CD3+ cells and CD8+/CD3+ cells were obtained by taking the FITC+ cells and counting the cells that were also APC (CD4) or TC (CD8)+. (a) On day 3, the percentages of T cells were affected by Cs. A (*shows a significant difference, Po0.01 (single-factor ANOVA)). (b) Different levels of T cells on day 6 (P40.05 among groups). There were five groups (five rats/group): Group 1, ADhBMPs alone; Group 2, Cs. A alone; Group 3, Cs. A with ADhBMPs; Group 4, CD4 and CD8 antibodies; and Group 5, CD4 and CD8 antibodies with ADhBMPs. Bars represent standard errors.

dimensional CT scans and three-dimensional reconstructions of ectopic bone formation in different groups. Figure 4 shows the volumes of ADhBMP9-induced bone in the various groups. The mean volumes of bone induced by ADhBMP9 were 0.2970.01 cm3 in animals given only the BMP vector (Group 1), 0.1770.03 cm3 in animals given Cs. A and the BMP vector (Group 3), and 0.5970.07 cm3 in animals given CD4 and CD8 antibodies and the BMP vector (Group 5). When bone volumes in the groups that received immunosuppressive agents were compared with those in the viral control group, the probability value was 0.03 for the Cs. A group and less than 0.01 for the antibodies group (Scheffe´’s method applied after single-factor analysis of variance Gene Therapy

Local immunomodulation improves osteogenesis JZ Li et al

1238

(ANOVA)). These results indicate that Cs. A has an inhibitory effect on bone formation induced by ADhBMP9 and that local immunomodulation with CD4 and CD8 antibodies significantly improves the osteogenic potential of adenovirus-mediated BMP-9 gene therapy. The impact these two kinds of immunosuppressive agents have on the induction of bone formation by ADhBMP4, however, is not yet clear because no bone formation was observed at the ADhBMP4 injection site in any group.

Histological analysis of the ADhBMP4 and ADhBMP9 injection sites Tissue sections obtained from the ADhBMP4 injection sites showed no evidence of bone formation on day 32. Histological changes were found in sections obtained at the ADhBMP9 injection sites, however, and these changes appeared similar in different groups. Many cells typical of osteocytes were surrounded by an osteoid matrix, similar to our findings in previous BMP gene therapy studies. Figure 5 demonstrates normal muscle tissue at the ADhBMP4 injection sites and typical bone tissue formed at the ADhBMP9 injection sites.

Figure 2 Time curves of adenovirus-neutralizing antibodies among the different groups. The lowest serum dilution is 20 in all groups (*shows a difference between the viral control group and this group, P ¼ 0.04).

Discussion In this study, CD4 and CD8 antibody treatment was clearly shown to increase the volume of bone induced following an intramuscular injection of ADhBMP9. CD4 and CD8 antibody treatment also temporarily reduced the number of circulating WBCs and decreased the production of adenovirus-neutralizing antibody, which occurred in response to administration of BMP adenoviral vectors. The sizes of bone induced by ADhBMP9 were significantly larger in animals given antibodies than in animals in the viral control group. These results show that CD4 and CD8 antibodies changed the host immune response to ADhBMPs, virus-transduced cells, and the process of bone formation. Although details on the mechanisms of CD4 and CD8 antibodies in the process of bone formation induced by ADhBMP9 are not clear, it is evident that T cells and the immune response may play important roles in the process of bone formation. The roles of CD4, CD8, other types of immune cells, and their secreted cytokines in the process of bone formation should continue to be studied in the future. Cs. A treatment significantly changed the percentages of doubly positive cells (CD4+/CD3+ and CD8+/CD3+

Figure 4 Bone volume was induced by ADhBMP9. There were five groups (five rats/group): Group 1, ADhBMPs alone; Group 2, Cs. A alone; Group 3, Cs. A with ADhBMPs; Group 4, CD4 and CD8 antibodies; and Group 5, CD4 and CD8 antibodies with ADhBMPs.

Figure 3 Ectopic bone formation induced by ADhBMP9 in different groups on day 30. Panels a–c show two-dimensional CT images and panels d–f show three-dimensional reconstructions of CT images. The images in panels a and d were obtained in the same rat as were b and e, and c and f. The arrows indicate ectopic bone. Gene Therapy

Local immunomodulation improves osteogenesis JZ Li et al

Figure 5 Typical histological characteristics of sites injected with ADhBMP4 and ADhBMP9. Upper Row: hematoxylin and eosin staining. Lower Row: Masson trichrome staining.

cells) 2 days after injection and reduced the volume of bone induced by an intramuscular injection of ADhBMP9. The mechanisms involved in cyclosporine’s inhibition of osteogenesis are currently unknown. Previous studies have shown that Cs. A is a selective inhibitor in the production of several cytokines involved in the regulation of immune cell activity. Cs. A inhibits T-cell secretion of several cytokines, including interleukin (IL)-2, IL-3, IL-4, and interferon gamma (IFN-g). Cs. A may also partially inhibit the expression of cell surface receptors for IL-2 on T cells. Additional targets of Cs. A probably include macrophages, monocytes, dendritic cells, and B cells.17,18 These cytokines play an important role in the regulation of bone metabolism.24–26 For example, IL-1, tumor necrosis factor (TNF)-a, and TNF-b are potent stimulators of bone resorption and also affect osteoblast metabolism. INF-g preferentially inhibits cytokine-stimulated resorption of bone.27–31 The role that cytokines play in the process of BMP-induced bone formation is currently not clear. Unpublished results from our laboratory have shown that some of the components of serum, especially following adenoviral vaccination, can strongly inhibit the alkaline phosphatase activities in C2C12 cells treated with BMP-2, BMP-6, and BMP-9. Some reports have also shown that high dose (415 mg/kg) and long-term Cs. A treatment significantly increases bone resorption in rats.24,32,33 This may be relevant to the Cs. A-induced decrease in ADhBMP9 bone induction because a dose of 40 mg/kg Cs. A was used in this study. The mechanisms of action of Cs. A have been shown to occur through several binding proteins, including cyclophilins, serine/threonine phosphatase, and calcineurin. Evidence suggests that cyclophilins regulate NFAT, AP-3, and NF-kB. Furthermore, the activity of NF-kB is regulated in a complex fashion that includes phosphorylation and dephosphorylation events. There may be crosstalk between these signaling pathways and the signal transduction pathways activated by BMPs.34,35 Both Type I and Type II BMP receptors are serine/

threonine kinases. The Type II receptors are constitutive active serine/threonine kinase receptors. The Type I receptor is activated by the Type II receptor by phosporylation at the GS box, a juxtamembrane domain rich in glycines and serines.20 Calcineurin usually functions as a Ca2+- and calmodulin-dependent complex, but calmodulin may not be involved in complexes with cyclophilin. Calmodulin can bind Smads 1 through 4 in a calcium-dependent manner. Data from some studies indicate that calmodulin inhibits the activin signals and stimulates the BMP signal. Other studies have shown that calmodulin inhibits Smad activities.36,37 In summary, there are most likely complex interactions between BMP-induced bone formation pathways and the molecular pathways involved in the activity of Cs. A. Immunosuppressive agents can be used not only to increase the osteogenic activity of BMP gene therapy but also to elucidate the complex relationship between the BMP signaling pathways and the host immune response. One research group has reported that both cyclophosphamide and fujimycin (FK-506) enhanced bone formation induced by BMP-2 and by ADhBMP2.11,12,38 Although FK-506 is an immunosuppressive agent closely related to Cs. A, the different effects of Cs. A on ADhBMP4 and ADhBMP9 gene therapy and those of FK-506 on ADhBMP2 gene therapy may involve different BMP receptors and signal transduction pathways or the different binding proteins of the two immunosuppressive agents. Recent advances in immunomodulation provide a wide choice of immunosuppressive agents for both direct and ex vivo gene therapy. The ideal immunosuppressive agent used in this field should exhibit temporary, specific, and local immunosuppressive effects. As bone induction can be initiated in 2–7 days in animals following BMP delivery, the potential risks of the immunosuppressive approach should be minimal and this approach may be practical in the clinical setting. Since there is a very complex relationship between the process of bone formation and the animal immune system, it should be paid more attention in future studies. In conclusion, in this study, we demonstrated that Cs. A and CD4 and CD8 antibodies have immunosuppressive functions. Cs. A inhibits bone formation induced by ADhBMP9, whereas the antibodies significantly improve the volume of bone formation induced by this BMP vector. The impact of the two kinds of immunosuppressive agents on ADhBMP4 remains unclear, however, because no new bone was observed in any animal after administration of ADhBMP4.

1239

Materials and methods Immunosuppressive agents and adenoviral vectors Mouse anti-rat CD4 and CD8 antibodies were purchased from Research Diagnostics. Inc. (Flanders, NJ, USA), and Cs. A was obtained from Sigma Chemical Co. (St Louis, MO, USA). To simplify the experimental design and reduce the number of animals that were used, relatively high doses of immunosuppressive agents were chosen: 40 mg/kg for each CD4 and CD8 antibody and 40 mg/kg for Cs. A.39,40 ADhBMP4 and ADhBMP9 were constructed, produced, purified, and titrated in a manner Gene Therapy

Local immunomodulation improves osteogenesis JZ Li et al

1240

previously reported.6,16 The measured titer of ADhBMP4 was 4  109 plaque-forming units (PFU)/ml (2.12  1012 particles/ml) and the titer of ADhBMP9 was 1.5  109 PFU/ml (4.38  1011 particles/ml). Based on our published data and previous experimental results, a dose of 107 PFU for both ADhBMP4 and ADhBMP9 may induce a large size of ectopic bone in athymic nude rats. Doses of ADhBMP9 between 107 and 5  107 PFU can induce a large size of bone in immunocompetent rats, but doses of ADhBMP4 between 107 and 5  108 PFU cannot induce any bone formation in immunocompetent animals.6,15,16,19,23 A medium dose of ADhBMP9 (1.5  107 PFU) or ADhBMP4 (1.2  108 PFU) was used in this study.

Animal groups and injections In total, twenty-five 2-month-old Sprague–Dawley rats were used in this study. The rats were randomly separated into the following five groups with five rats in each group (Table 1): Group 1, ADhBMP4 and ADhBMP9; Group 2, Cs. A; Group 3, Cs. A with ADhBMP4 and ADhBMP9; Group 4, mouse anti-rat CD4 and CD8 antibodies; and Group 5, mouse anti-rat CD4 and CD8 antibodies with ADhBMP4 and ADhBMP9. On day 1, blood was collected from the tail vein of each animal and immediately mixed with heparin. In the rats in Groups 2 and 3, Cs. A (40 mg/ kg) was injected into the thigh musculature on both sides. On day 3, blood samples were first collected from all animals and the rats received the following injections in the front thigh muscles with the needle touching the femur: Group 1 and 3, ADhBMP4 in the left thigh (1.2  108 PFU/70 ml) and ADhBMP9 in the right thigh (1.5  107 PFU/70 ml); Group 2, phosphate-buffered saline (PBS; 70 ml) in each thigh; Group 4, mouse anti-rat CD4 and CD8 antibodies in each thigh (10 mg/10 ml CD4 plus 10 mg/10 ml CD8 with 50 ml PBS); and Group 5, mouse anti-rat CD4 and CD8 antibodies plus ADhBMP4 in the left thigh and ADhBMP9 in the right thigh (5 mg/ 5 ml CD4 plus 5 mg/5 ml CD8 plus 50 ml ADhBMP4 or ADhBMP9). Blood samples were again collected on days 6, 10, 16, and 30. Analysis of WBCs Total numbers of WBCs were measured on days 1, 3, 6, 10, and 16 as a standard procedure. Briefly, 10 ml of blood was added to 1 ml red blood cell (RBC) lysis buffer (ACK buffer: 150 mmol/l NH4Cl, 61 mmol/l KHCO3, and 1 mmol/l Na2EDTA, pH 7.3).41 The total number of WBCs was counted on a hemacytometer and the number of cells per microliter was calculated. Determining levels of CD3+, CD4+, and CD8+ cells The procedures used to isolate and stain WBCs were identical to those previously reported.16 Briefly, 100 ml of blood was suspended in 1 ml PBS. In total, 9 ml of distilled water was added to the blood to lyse the RBCs. After 30 s, 1 ml of 10-fold concentrated PBS was added to each tube. The WBCs were isolated after centrifugation and stained with mixed antibodies against CD3 (fluorescein isothiocyanate (FITC)), CD4 (allophycocyanin (APC)), and CD8 (tricolor (TC)) (Caltag Laboratories, Burlingame, CA, USA) in 200 ml PBS for 12 h at 41C. The stained WBCs were washed with PBS and measured using flow cytometry. The percentages of CD3+, CD4+/ Gene Therapy

CD3+, and CD8+/CD3+ cells were calculated and compared among the various animal groups on days 3 and 6 by using the one-way ANOVA multiple-comparison method.

Measuring adenovirus-neutralizing antibody titers Rat sera collected on days 1, 16, and 30 were tested for adenovirus-neutralizing antibody.16 The sera were serially diluted in 250 ml Dulbecco minimal Eagle medium with 10% fetal bovine serum at a dilution factor of 2 from 20- to 640-fold and mixed with equal volumes of viral solution containing 500 green cell-forming units (GFU) of recombinant green fluorescent protein adenovirus (ADCMVGFP). The mixture was incubated at 371C for 1 h and added to 96-well plates containing confluent 293 cells. Four wells were used for each concentration with 100 ml/100 GFU/well. The plates were incubated at 371C in an atmosphere of 95% air/5% CO2 for 24 h. The total numbers of green cells in four wells of each concentration were counted to obtain 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 were used to calculate the neutralizing antibody titer in sera. The titer of adenovirus-neutralizing antibody was defined as the serum dilution that resulted in a 97% inhibition of ADCMVGFP. A comparison of titers among groups on days 16 and 30 was accomplished using the one-way ANOVA multiplecomparison method. CT imaging analysis of ectopic bone formation All rats were scanned by CT using a Picker PQ 6000 scanner on day 30. Axial images with a 1-mm collimation and a 1-mm table increment were obtained using the standard algorithm with 130 kV, 100 mA, a 2-s scan time, and a 40-mm image size. Three-dimensional reconstructions and evaluations of bone volumes were performed with the aid of a Voxel Q workstation. Bone volumes on day 30 were compared using one-way ANOVA followed by the Scheffe´ method of comparing contrasts. Histological analysis The rats were euthanized on day 32. The injection sites in the thigh muscles were removed and fixed in 10% neutral-buffered formalin. The thigh muscle samples were decalcified, embedded in paraffin, and cut into sections, after which the sections were placed on slides. The slides were then stained with hematoxylin and eosin and Masson trichrome.

Acknowledgements This project was supported by National Institutes of Health Grant No. R01 AR046488-02, ‘Tissue Engineering Utilizing BMP Gene Therapy’ (Gregory A Helm, MD, PhD, Principle Investigator).

References 1 Wozney JM. The bone morphogenetic protein family: multifunctional cellular regulators in the embryo and adult. Eur J Oral Sci 1998; 106 (Suppl 1): 160–166.

Local immunomodulation improves osteogenesis JZ Li et al

1241 2 Ducy P, Karsenty G. The family of bone morphogenetic proteins. Kidney Int 2000; 57: 2207–2214. 3 Nakashima M, Reddi AH. The application of bone morphogenetic proteins to dental tissue engineering. Nat Biotechnol 2003; 21: 1025–1032. 4 Sheehan JP et al. The safety and utility of recombinant human bone morphogenetic protein-2 for cranial procedures in a nonhuman primate model. J Neurosurg 2003; 98: 125–130. 5 Boden SD. Clinical application of the BMPs. J Bone Joint Surg Am 2001; 83-A (Suppl 1): S161. 6 Li JZ et al. Osteogenic potential of five different recombinant human bone morphogenetic protein adenoviral vectors in the rat. Gene Therapy 2003; 10: 1735–1743. 7 Cheng H et al. Osteogenic activity of the fourteen types of human bone morphogenetic proteins (BMPs). J Bone Joint Surg Am 2003; 85-A: 1544–1552. 8 Chen Y et al. Gene therapy for new bone formation using adenoassociated viral bone morphogenetic protein-2 vectors. Gene Therapy 2003; 10: 1345–1353. 9 Musgrave DS et al. Human skeletal muscle cells in ex vivo gene therapy to deliver bone morphogenetic protein-2. J Bone Joint Surg Br 2002; 84: 120–127. 10 Alden TD et al. In vivo endochondral bone formation using a bone morphogenetic protein 2 adenoviral vector. Hum Gene Ther 1999; 10: 2245–2253. 11 Kaihara S et al. Simple and effective osteoinductive gene therapy by local injection of a bone morphogenetic protein-2-expressing recombinant adenoviral vector and FK506 mixture in rats. Gene Therapy 2004; 11: 439–447. 12 Okubo Y et al. The time course study of osteoinduction by bone morphogenetic protein-2 via adenoviral vector. Life Sci 2001; 70: 325–336. 13 Russell RG, Graveley R, Skjodt H. The effects of cyclosporin A on bone and cartilage. Br J Rheumatol 1993; 32 (Suppl 1): 42–46. 14 Hong JC, Kahan BD. Immunosuppressive agents in organ transplantation: past, present, and future. Semin Nephrol 2000; 20: 108–125. 15 Jane Jr JA et al. Ectopic osteogenesis using adenoviral bone morphogenetic protein (BMP)-4 and BMP-6 gene transfer. Mol Ther 2002; 6: 464–470. 16 Li JZ et al. Osteogenesis in rats induced by a novel recombinant helper-dependent bone morphogenetic protein-9 (BMP-9) adenovirus. J Gene Med 2003; 5: 748–756. 17 Russell G et al. Mechanisms of action of cyclosporine and effects on connective tissues. Semin Arthritis Rheum 1992; 21: 16–22. 18 Russell RG et al. Cyclosporin A. Mode of action and effects on bone and joint tissues. Scand J Rheumatol Suppl 1992; 95: 9–18. 19 Li JZ et al. Rat strain differences in the ectopic osteogenic potential of recombinant human BMP adenoviruses. Mol Ther 2003; 8: 822–829. 20 Nohe A, Keating E, Knaus P, Petersen NO. Signal transduction of bone morphogenetic protein receptors. Cell Signal 2004; 16: 291–299. 21 Miller AF, Harvey SA, Thies RS, Olson MS. Bone morphogenetic protein-9. An autocrine/paracrine cytokine in the liver. J Biol Chem 2000; 275: 17937–17945. 22 Lopez-Coviella I, Berse B, Thies RS, Krzysztof Blusztajn J. Upregulation of acetylcholine synthesis by bone morphogenetic protein 9 in a murine septal cell line. J Physiol Paris 2002; 96: 53–59. 23 Varady P et al. Morphologic analysis of BMP-9 gene therapyinduced osteogenesis. Hum Gene Ther 2001; 12: 697–710.

24 Goodman GR et al. Transforming growth factor-beta administration modifies cyclosporine A-induced bone loss. Bone 2001; 28: 583–588. 25 Goodman GR et al. Interferon-alpha, unlike interferon-gamma, does not cause bone loss in the rat. Bone 1999; 25: 459–463. 26 Mann GN et al. Interferon-gamma causes loss of bone volume in vivo and fails to ameliorate cyclosporin A-induced osteopenia. Endocrinology 1994; 135: 1077–1083. 27 Kvien TK et al. Long term efficacy and safety of cyclosporin versus parenteral gold in early rheumatoid arthritis: a three year study of radiographic progression, renal function, and arterial hypertension. Ann Rheum Dis 2002; 61: 511–516. 28 Gowen M, MacDonald BR, Russell RG. Actions of recombinant human gamma-interferon and tumor necrosis factor alpha on the proliferation and osteoblastic characteristics of human trabecular bone cells in vitro. Arthritis Rheum 1988; 31: 1500–1507. 29 Evans DB, Bunning RA, Van Damme J, Russell RG. Natural human IL-1 beta exhibits regulatory actions on human bonederived cells in vitro. Biochem Biophys Res Commun 1989; 159: 1242–1248. 30 Evans DB, Bunning RA, Russell RG. The effects of recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) on human osteoblast-like cells. Biochem Biophys Res Commun 1989; 160: 588–595. 31 Evans DB, Bunning RA, Russell RG. The effects of recombinant human interleukin-1 beta on cellular proliferation and the production of prostaglandin E2, plasminogen activator, osteocalcin and alkaline phosphatase by osteoblast-like cells derived from human bone. Biochem Biophys Res Commun 1990; 166: 208–216. 32 Epstein S et al. Effect of the interaction of parathyroid hormone and cyclosporine a on bone mineral metabolism in the rat. Calcif Tissue Int 2001; 68: 240–247. 33 Awumey EM et al. Molecular and functional evidence for calcineurin-A alpha and beta isoforms in the osteoclast: novel insights into cyclosporin A action on bone resorption. Biochem Biophys Res Commun 1999; 254: 248–252. 34 Ulloa L, Doody J, Massague J. Inhibition of transforming growth factor-beta/SMAD signalling by the interferon-gamma/STAT pathway. Nature 1999; 397: 710–713. 35 Miyazawa K et al. Two major Smad pathways in TGF-beta superfamily signalling. Genes Cells 2002; 7: 1191–1204. 36 Funaba M, Zimmerman CM, Mathews LS. Modulation of Smad2-mediated signaling by extracellular signal-regulated kinase. J Biol Chem 2002; 277: 41361–41368. 37 Zimmerman CM, Kariapper MS, Mathews LS. Smad proteins physically interact with calmodulin. J Biol Chem 1998; 273: 677–680. 38 Kaihara S et al. Effect of FK506 on osteoinduction by recombinant human bone morphogenetic protein-2. Life Sci 2002; 72: 247–256. 39 Nassar CA et al. Effect of cyclosporin A on alveolar bone homeostasis in a rat periodontitis model. J Periodontal Res 2004; 39: 143–148. 40 Lallemand F et al. Cyclosporine A delivery to the eye: a pharmaceutical challenge. Eur J Pharm Biopharm 2003; 56: 307–318. 41 Shi W et al. Effect of macrophage-derived apolipoprotein E on established atherosclerosis in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol 2000; 20: 2261–2266.

Gene Therapy