JBJS | Effect of Anti‐Tumor Necrosis Factor-α Gene Therapy on Wear Debris‐Induced Osteolysis

The Journal of Bone & Joint Surgery, Volume 83, Issue 12

Scientific Article | December 01, 2001

Effect of Anti‐Tumor Necrosis Factor-α Gene Therapy on Wear Debris‐Induced Osteolysis

Lisa M. Childs, MS; J. Jeffrey Goater, MS; Regis J. O'Keefe, MD, PhD; Edward M. Schwarz, PhD

Investigation performed at the University of Rochester Medical Center, Rochester, New York

Lisa M. Childs, MS
Department of Microbiology and Immunology, University of Rochester Medical Center, 601 Elmwood Avenue, Box 672, Rochester, NY 14642. E-mail address: lisachilds@mail.com

J. Jeffrey Goater, MS
Regis J. O’Keefe, MD, PhD
Edward M. Schwarz, PhD
Department of Orthopaedics, University of Rochester Medical Center, 601 Elmwood Avenue, Box 665, Rochester, NY 14642. E-mail address for J.J. Goater: jeff_goater@urmc.rochester.edu. E-mail address for R.J. O’Keefe: regis_okeefe@urmc.rochester.edu. E-mail address for E.M. Schwarz: edward_schwarz@urmc.rochester.edu

In support of their research or preparation of this manuscript, one or more of the authors received grants or outside funding from the National Institutes of Health (Public Health Service Grants AR45971 and AR46545), the National Institute of Arthritis and Musculoskeletal and Skin Diseases, and the Orthopaedic Research and Education Foundation. None of the authors received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.

The Journal of Bone & Joint Surgery. 2001; 83:1789-1797

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

Background: Particle phagocytosis by macrophages induces the secretion of tumor necrosis factor-a, which is involved in the development of an osteolytic response. Therefore, we aimed to determine whether gene delivery of a soluble inhibitor of tumor necrosis factor-a (sTNFR:Fc) could prevent wear debris-induced osteolysis in a mouse model. sTNFR:Fc is a fusion protein containing the extracellular domain of the human type-I tumor necrosis factor receptor fused to the Fc region of mouse immunoglobulin. It acts by binding to tumor necrosis factor-a and preventing signaling through the membrane-bound tumor necrosis factor receptors.

Methods: An adenoviral vector encoding the LacZ gene (Ad.CMV-NlacZ) was propagated and was tested for its ability to transduce calvarial tissue. Ad.CMV-TNFR:Fc (encoding sTNFR:Fc) or Ad.CMV-NlacZ was administered to CBAxB6 mice in the presence or absence of titanium particles implanted onto the calvaria. Serum levels of sTNFR:Fc were measured with enzyme-linked immunosorbent assay, and the mice were killed on the tenth postoperative day for histological analysis. The experiments were repeated in athymic nude mice to avoid complications associated with the adenovirus-specific immune response.

Results: Administration of the control virus (Ad.CMV-NlacZ) transduced 10% of the cells in the periosteum. Ad.CMV-NlacZ treatment of sham-treated or titanium-treated animals induced significant bone resorption and osteoclastogenesis above control levels (that is, those in animals not treated with a virus). Treatment with the sTNFR:Fc virus did not reduce bone resorption or osteoclast numbers below control levels in CBAxB6 mice. In the athymic mice, no increase in the midline sagittal suture area or osteoclastogenesis was observed after treatment with the control vector and sTNFR:Fc gene therapy reduced the suture area to background levels.

Conclusions: An immunologic response to Ad.CMV-NlacZ was most likely responsible for the increase in bone resorption and osteoclastogenesis in the animals treated with the control vector alone. In the athymic mice, in the absence of this immune response, sTNFR:Fc gene therapy reduced bone resorption in the midline sagittal suture area but had no effect on osteoclastogenesis.

Clinical Relevance: These results indicate that adenoviral vectors should not be used for gene therapy for the prevention of osteolysis. The results of adenovirus-mediated gene therapy in other systems should also be questioned because of the strong immune response to this vector. However, administration of the sTNFR:Fc gene in a different, nonimmunogenic vector may be useful for the treatment or prevention of wear debris-induced osteolysis.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

Wear debris-induced osteolysis is a common cause of orthopaedic implant failure for which there is no proven drug therapy. Osteolysis occurs most often as a result of particle phagocytosis by macrophages, which leads to the secretion of tumor necrosis factor-a (TNF-a). Tumor necrosis factor-a is involved in the pathogenesis of the osteolytic response. Macrophages produce high levels of TNF-a following ingestion of wear debris particles¹, and TNF-a has also been detected in the fluid and in the inflamed tissue surrounding loosened implants^2-4. Tumor necrosis factor-a stimulates osteoclast formation and activity in vitro^5,6 through mechanisms that may be both dependent and independent of osteoclast differentiation factor⁷. Recently, we⁸ and others⁹ reported that TNF-a signaling was crucial to the development of the inflammatory osteolytic response to wear debris in a mouse calvaria model in which type-I TNF-a receptor (TNFRI) knockout mice failed to respond to polymethylmethacrylate or titanium particles. Thus, TNF-a signaling appears to be an appropriate target for therapeutic agents intended to treat or prevent wear debris-induced osteolysis. Tumor necrosis factor-a is also a good target for gene therapy, since this cytokine is involved early in the response to particles and controls the release of other osteoclastogenic factors, including interleukin-1 and 6^2,6,10,11.

Tumor necrosis factor-a antagonists (e.g., etanercept) have been effective for the treatment of inflammatory diseases and have produced impressive results as a treatment of rheumatoid arthritis¹², a disease in which the generation of high levels of TNF-a is thought to play an important role in joint damage¹¹. More recent clinical data have indicated that anti-TNF-a therapy also helps to prevent bone erosions in rheumatoid arthritis^13,14. Given the emerging evidence that TNF-a plays a major role in both rheumatoid arthritis and prosthetic loosening, we¹⁵ and others¹⁶ evaluated the efficacy of etanercept in a mouse calvaria model of wear debris-induced osteolysis and found it to be effective in inhibiting osteolysis. To determine whether de novo synthesis of this molecule is as effective, we investigated gene therapy with a soluble inhibitor of TNF-a to inhibit wear debris-induced osteolysis in vivo. The TNF-a inhibitor used in the current study, TNFR:Fc, is a fusion protein encoded by a replication-deficient adenovirus and is composed of the extracellular domain of the human type-I (p55) TNF receptor fused to the Fc region of a mouse IgG1 immunoglobulin molecule¹⁷. This fusion protein is capable of neutralizing TNF-a by binding to the soluble cytokine, thus blocking signaling through the membrane-bound TNF receptors¹⁸. This vector has been shown to be protective in animal models of arthritis^19,20.

Although TNFR:Fc has efficacy, the use of recombinant biological agents is not without problems. Drug delivery requires repeated injection, is costly, and involves systemic therapy for a localized disease process. Also, after an arthroplasty, particulate debris is generated continuously and patients would therefore require lifelong therapy to prevent the activation of macrophages and the initiation of pathological conditions. Gene therapy offers the potential for more efficient, localized delivery of large recombinant proteins. The objective of this study was to determine whether adenovirus-mediated gene delivery of a soluble inhibitor of TNF-a (sTNFR:Fc) can prevent wear debris-induced osteolysis.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1:: Calvarial tissue is transduced with local administration of Ad.CMV-NlacZ. 10⁸ infectious units of Ad.CMV-NlacZ (A) or phosphate-buffered saline solution vehicle control (B) was injected subcutaneously over the calvaria, which were harvested after forty-eight hours and were stained with X-gal. The calvaria in Figure 1,A were sectioned and stained with eosin. The midline sagittal suture area is shown at 100 original magnification (C). The blue staining indicates cell transduction.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-A:: Figs. 2-A, 2-B, and 2-C Ad.CMV-NlacZ increases osteoclast activity and osteoclastogenesis in normal mice. Ad.CMV-NlacZ or Ad.CMV-TNFR:Fc was administered locally or systemically to CBAxB6 mice (five per group) at the time of titanium implantation or in the absence of titanium particles. Fig. 2-A Sera were collected, and the concentration of sTNFR:Fc was analyzed with enzyme-linked immunosorbent assay. The data are given as the mean and the standard error of the mean.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-B:: Figs. 2-B and 2-C The sagittal suture area (Fig. 2-B) and the number of osteoclasts (Fig. 2-C) were calculated as described in the Materials and Methods section. The data are given as the mean and the standard error of the mean for each group. The adenovirus groups were treated either locally (striped bars) or systemically (white bars). The sham-treated control group received surgery without titanium and no adenovirus, the sham-treated Ad.CMV-NlacZ group received surgery and adenovirus but no titanium, the titanium-treated control group received surgery with titanium but no adenovirus, and the titanium-treated Ad.CMV-NlacZ and Ad.CMV-TNFR:Fc groups received surgery with titanium and the respective adenovirus. A single footnote symbol indicates that p < 0.05 and a double symbol indicates that p < 0.005 *compared with the sham-treated controls that did not receive Ad.CMV-NlacZ, †compared with the titanium-treated controls, ‡compared with the titanium-treated group that received Ad.CMV-NlacZ, or §compared with local administration.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-C:: Figs. 2-B and 2-C The sagittal suture area (Fig. 2-B) and the number of osteoclasts (Fig. 2-C) were calculated as described in the Materials and Methods section. The data are given as the mean and the standard error of the mean for each group. The adenovirus groups were treated either locally (striped bars) or systemically (white bars). The sham-treated control group received surgery without titanium and no adenovirus, the sham-treated Ad.CMV-NlacZ group received surgery and adenovirus but no titanium, the titanium-treated control group received surgery with titanium but no adenovirus, and the titanium-treated Ad.CMV-NlacZ and Ad.CMV-TNFR:Fc groups received surgery with titanium and the respective adenovirus. A single footnote symbol indicates that p < 0.05 and a double symbol indicates that p < 0.005 *compared with the sham-treated controls that did not receive Ad.CMV-NlacZ, †compared with the titanium-treated controls, ‡compared with the titanium-treated group that received Ad.CMV-NlacZ, or §compared with local administration.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3:: Ad.CMV-NlacZ increases osteoclast activity above control levels in normal mice. Representative trichrome-stained sections demonstrating the midline sagittal suture are shown at 40 original magnification. CBAxB6 mice were treated with sham surgery but no adenovirus (A), sham surgery with Ad.CMVNlacZ (B), surgery with titanium but no adenovirus (C), or surgery with titanium and Ad.CMV-TNFR:Fc (D).

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-A:: Figs. 4-A, 4-B, and 4-C Ad.CMV-NlacZ does not affect osteoclast activity or osteoclastogenesis in athymic nude mice, and sTNFR:Fc gene therapy decreases the midline sagittal suture area. Adenoviruses were administered locally or systemically on day 0 to athymic nude mice (four per group). Fig. 4-A Sera were collected over the course of ten days for analysis by enzyme-linked immunosorbent assay. The data are given as the mean and the standard error of the mean.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-B:: Figs. 4-B and 4-C The sagittal suture area (Fig. 4-B) and the number of osteoclasts (Fig. 4-C) were calculated as described in the Materials and Methods section. The data are presented as the mean and the standard error of the mean for each group. The adenovirus groups were treated either locally (striped bars) or systemically (white bars). The sham-treated control group received surgery without titanium and no adenovirus, the sham-treated Ad.CMV-NlacZ group received surgery and adenovirus but no titanium, the titanium-treated control group received surgery with titanium but no adenovirus, and the titanium-treated Ad.CMV-NlacZ and Ad.CMV-TNFR:Fc groups received surgery with titanium and the respective adenovirus. *p < 0.05 compared with the Ad.CMV-NlacZ groups.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-C:: Figs. 4-B and 4-C The sagittal suture area (Fig. 4-B) and the number of osteoclasts (Fig. 4-C) were calculated as described in the Materials and Methods section. The data are presented as the mean and the standard error of the mean for each group. The adenovirus groups were treated either locally (striped bars) or systemically (white bars). The sham-treated control group received surgery without titanium and no adenovirus, the sham-treated Ad.CMV-NlacZ group received surgery and adenovirus but no titanium, the titanium-treated control group received surgery with titanium but no adenovirus, and the titanium-treated Ad.CMV-NlacZ and Ad.CMV-TNFR:Fc groups received surgery with titanium and the respective adenovirus. *p < 0.05 compared with the Ad.CMV-NlacZ groups.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 5:: Osteolysis is not induced by Ad.CMV-NlacZ alone in athymic mice, and it is decreased by administration of Ad.CMV-TNFR:Fc to athymic nude mice treated with titanium. Representative trichrome-stained sections from Ad.CMV-NlacZ-treated athymic nude mice in the absence (A) and presence (B) of titanium particles and from Ad.CMV-TNFR:Fc-treated athymic mice in the presence of titanium (C) are shown at 40 original magnification.

View Larger

TABLE I: Experimental Groups

Titanium Particles	Treatment	Mouse Strain	No. of Animals
—	None	CBAxB6	5
—	Ad.CMV-NlacZ (local)	CBAxB6	5
—	Ad.CMV-NlacZ (intraperitoneal)	CBAxB6	5
+	Phosphate-buffered saline solution (intraperitoneal)	CBAxB6	5
+	Ad.CMV-NlacZ (local)	CBAxB6	5
+	Ad.CMV-NlacZ (intraperitoneal)	CBAxB6	5
+	Ad.CMV-TNFR:Fc (local)	CBAxB6	5
+	Ad.CMV-TNFR:Fc (intraperitoneal)	CBAxB6	5
—	None	Athymic nude	4
—	Ad.CMV-NlacZ (local)	Athymic nude	4
—	Ad.CMV-NlacZ (intraperitoneal)	Athymic nude	4
+	Phosphate-buffered saline solution (intraperitoneal)	Athymic nude	4
+	Ad.CMV-NlacZ (local)	Athymic nude	4
+	Ad.CMV-NlacZ (intraperitoneal)	Athymic nude	4
+	Ad.CMV-TNFR:Fc (local)	Athymic nude	4
+	Ad.CMV-TNFR:Fc (intraperitoneal)	Athymic nude	4

Add to My JBJS

TABLE I: Experimental Groups

Titanium Particles	Treatment	Mouse Strain	No. of Animals
—	None	CBAxB6	5
—	Ad.CMV-NlacZ (local)	CBAxB6	5
—	Ad.CMV-NlacZ (intraperitoneal)	CBAxB6	5
+	Phosphate-buffered saline solution (intraperitoneal)	CBAxB6	5
+	Ad.CMV-NlacZ (local)	CBAxB6	5
+	Ad.CMV-NlacZ (intraperitoneal)	CBAxB6	5
+	Ad.CMV-TNFR:Fc (local)	CBAxB6	5
+	Ad.CMV-TNFR:Fc (intraperitoneal)	CBAxB6	5
—	None	Athymic nude	4
—	Ad.CMV-NlacZ (local)	Athymic nude	4
—	Ad.CMV-NlacZ (intraperitoneal)	Athymic nude	4
+	Phosphate-buffered saline solution (intraperitoneal)	Athymic nude	4
+	Ad.CMV-NlacZ (local)	Athymic nude	4
+	Ad.CMV-NlacZ (intraperitoneal)	Athymic nude	4
+	Ad.CMV-TNFR:Fc (local)	Athymic nude	4
+	Ad.CMV-TNFR:Fc (intraperitoneal)	Athymic nude	4

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

Titanium Particles

Pure titanium particles were obtained from Johnson Matthey Chemicals (Ward Hill, Massachusetts) and were prepared as previously described²¹. Briefly, particles 1 to 3 m in diameter were suspended in phosphate-buffered saline solution at a concentration of 1¥10⁸ particles/mL. Particle size was confirmed with a channelizer (Coulter Electronics, Hialeah, California), which showed 90% of the particles to be >5 m in diameter. A limulus assay (BioWhittaker, Walkersville, Maryland) was used to show that the suspension was free of endotoxin.

Adenoviruses

An adenoviral vector encoding the LacZ gene (Ad.CMV-NlacZ) and another encoding sTNFR:Fc (Ad.CMV-TNFR:Fc) were obtained from Inder M. Verma, PhD (Salk Institute, La Jolla, California), and Bruce Beutler, MD (Scripps Research Institute, La Jolla, California), respectively, and were propagated as previously described^18,22. Briefly, 293T cells were grown in 150-mm culture dishes in Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin (Gibco, Grand Island, New York). The cells were infected after they reached 60% confluence, and they were harvested thirty-six to forty-eight hours later, after the cytopathic effect was complete. Cell lysates were produced by freeze-thawing three times in a dry ice/ethanol bath and removing cell debris by centrifugation at 2500 rpm for five minutes. The lysate was applied to a 1.25 to 1.5 g/mL cesium chloride gradient, and purified viral stocks were obtained with ultracentrifugation at 10Â°C for one hour at 30,000 rpm. The viral band was collected and was spun through a second cesium chloride gradient, was harvested and dialyzed over three changes of buffer (10 mM Tris HCl, pH 7.4, 1 mM MgCl2, and 10% glycerol), and was stored at —80Â°C. Viral stocks were titered with use of a standard protein assay (Bio-Rad, Hercules, California) to determine the concentration of viral particles. Murine fibroblasts were infected at various multiplicities of infection and were examined forty-eight hours later for b-galactosidase activity by staining with X-gal (Gibco) or for sTNFR:Fc production by measuring fusion protein concentrations in the supernatants with an enzyme-linked immunoassay kit (BioSource International, Camarillo, California) as previously described²³. The number of infectious units was calculated by dividing the viral particle concentration by the lowest multiplicities of infection that resulted in 100% of cells stained with X-gal or the maximal sTNFR:Fc concentration for the Ad.CMV-NlacZ and Ad.CMV-TNFR:Fc viruses, respectively.

X-gal Staining of Ad.CMV-NlacZ-Infected Calvaria

We administered 10⁸ infectious units of Ad.CMV-NlacZ or vehicle control (phosphate-buffered saline solution) directly to the calvaria through subcutaneous injection using a 26-gauge syringe. The calvaria were harvested forty-eight hours later and were fixed in 2% formaldehyde/0.2% glutaraldehyde, pH 7.4, for two hours at 4Â°C. Following fixation, the calvaria were removed and were incubated in staining solution (5 mM potassium ferricyanide/5 mM potassium ferrocyanide/2 mM MgCl2) with 1 mg/mL X-gal overnight at 37Â°C. After staining, the calvaria were photographed and then decalcified, embedded in paraffin, sectioned (3 m), and stained with eosin (Fig. 1,A).

In Vivo Mouse Calvaria Resorption Model

Bone resorption and osteoclast numbers were determined in vivo as previously described²¹. Briefly, five healthy female CBAxB6 mice (Jackson Laboratory, Bar Harbor, Maine) or four healthy female athymic nude mice (Harlan Sprague Dawley, Indianapolis, Indiana) were used in each group (Table I). All of the animals were housed and treated according to university-approved guidelines. The mice were anesthetized with 100 mg/kg of ketamine and 50 mg/kg of xylazine by intraperitoneal injection. A 11-cm area of periosteum was exposed by making a midline sagittal incision over the calvaria. Thirty milligrams of titanium particles (Johnson Matthey Chemicals) were spread over the intact periosteum and the incision was closed or, in the sham-treated animals, the incision was closed without implantation of particles. For the gene-transfer experiments, 210⁸ infectious units of either adenovirus was administered at the time of titanium implantation through intraperitoneal injection or by adding the virus directly onto the periosteum with the titanium particles. The mice were bled retro-orbitally each day. After the blood was allowed to clot, the sera were collected by centrifugation for five minutes at 14,000 rpm. The sera were frozen at —80Â°C until they were analyzed for sTNFR:Fc content with an enzyme-linked immunoassay kit according to the manufacturer’s instructions (BioSource).

Ten days after surgery, the mice were killed and the calvaria were harvested. An elliptical plate of bone consisting of parietal bone, sagittal suture, and adjacent periosteum was removed. The tissue was cut in half coronally (transversely), fixed in 70% ethanol, and embedded in glycol methacrylate with the cut surfaces facing down. Coronal (transverse) sections (3 m thick) were made through each specimen. Each specimen was sampled twice, 150 m apart, producing four sections per animal (the titanium-treated area was consistently located over the sagittal suture). The sections were stained with trichrome stain or for tartrate-resistant acid phosphatase with use of the Diagnostics acid phosphatase kit (Sigma Chemical, St. Louis, Missouri). Each section was digitally photographed at 40 original magnification, and the image was oriented with the midline suture in the center of the field. The sagittal suture area in the trichrome-stained sections was determined by tracing the area of soft tissue between the parietal bones and including any resorption pits on the superior surface of the calvaria visible in the same field as the midline suture. These areas were quantified with Scion Image software (Scion, Frederick, Maryland). The number of osteoclasts in the midline suture area was determined by counting the number of tartrate-resistant acid phosphatase-positive cells within the suture area and along the superior calvarial surface visible in a 40¥ field. The results from the four sections for each animal were averaged, and the average sagittal suture area and number of osteoclasts for each group of animals were determined.

Statistical Analysis

Significance was determined with the Student t test. A p value of <0.05 was considered significant.

Results

Introduction | Materials and Methods | Results | Discussion | References

Adenoviral Transduction of Calvaria

The mouse calvaria model of wear debris-induced osteolysis is used to measure bone loss in the midline sagittal suture as a result of titanium implantation²¹. Before examining the ability of sTNFR:Fc gene therapy to prevent this response, we first determined whether the calvarial tissue could be efficiently transduced by an adenoviral construct. Calvaria of normal CBAxB6 mice were treated with Ad.CMV-NlacZ or phosphate-buffered saline solution and were examined forty-eight hours later for b-galactosidase expression. On gross examination of the calvaria following X-gal staining, we observed intense blue staining of the Ad.CMV-NlacZ-treated calvaria, while the control tissue remained negative for active b-galactosidase (Fig. 1, A and B). Figure 1,C shows that Ad.CMV-NlacZ infects approximately 10% of the cells in the periosteum within forty-eight hours after infection, with the positive cells exhibiting a nuclear pattern of b-galactosidase expression.

Effects of sTNFR:Fc Gene Therapy on Wear Debris-Induced Osteoclastogenesis and Osteolysis

Normal Mice

The effect of sTNFR:Fc gene therapy on osteoclast activity in vivo was tested with use of the mouse calvaria model of wear debris-induced osteolysis²¹ as described above. Enzyme-linked immunosorbent analysis showed that the CBAxB6 mice treated with the sTNFR:Fc adenovirus had measurable levels of sTNFR:Fc in the sera beginning on day 1, which declined after day 5 and was undetectable on day 10 (Fig. 2-A). Expression of sTNFR:Fc peaked at 800 ng/mL on day 3 and at 900 ng/mL on day 5 for the systemically and locally injected animals, respectively, although these differences were not significant. Mice treated with Ad.CMV-NlacZ, sham-treated animals, or mice given titanium alone did not have measurable levels of sTNFR:Fc in the sera at any time during the experiment (Fig. 2-A and data not shown).

To evaluate the ability of Ad.CMV-TNFR:Fc gene therapy to prevent wear debris-induced osteolysis, bone resorption was assessed in calvarial sections obtained from the animals ten days after the surgery. Ad.CMV-NlacZ treatment was found to significantly increase the sagittal suture area in the sham-treated animals (p < 0.005 and p < 0.05 for the local and systemic treatments, respectively) (Figs. 2-B and 3,B). Osteoclast activity in the calvaria of the titanium-treated animals given a local injection of the control vector significantly (p < 0.05) increased the midline sagittal suture area to 0.1507 mm² 0.0203 mm²compared with 0.1117 0.0323 mm² in the group that received titanium but no virus (Fig. 2-B). Systemic delivery of Ad.CMV-NlacZ to titanium-treated mice resulted in an overall increase in the sagittal suture area but the increase was not significant. Treatment with Ad.CMV-TNFR:Fc did not result in a decrease in sagittal suture area compared with the area in the titanium-treated control animals (no virus), but it significantly (p < 0.05) reduced bone resorption compared with that in the group treated with Ad.CMV-NlacZ, which in itself exhibited a large increase in sagittal suture area not attributable to titanium-mediated osteolysis (Fig. 2-B).

Direct examination of tartrate-resistant acid phosphatase-stained calvarial sections was performed to determine the effect of adenovirus treatment on osteoclast numbers. Local and systemic gene delivery of Ad.CMV-NlacZ to animals without titanium particles tripled the number of osteoclasts in the midline sagittal suture (Fig. 2-C). Local application of this control vector to titanium-treated mice increased the number of osteoclasts to 63 ± 6, compared with 27 ± 5 in the titanium-treated controls (p < 0.005), and systemic delivery increased the number to 44 ± 6 (p < 0.05 compared with the controls) (Fig. 2-C). Interestingly, the administration of Ad.CMV-TNFR:Fc did not reduce the number of osteoclasts to below background levels. In fact, local delivery of Ad.CMV-TNTR:Fc resulted in the same increase in osteoclast numbers as Ad.CMV-NlacZ (Fig. 2-C).

Athymic Nude Mice

Previously, we²² and others²⁴ showed that adenoviral vectors induce inflammation and that this response is T-cell mediated. Therefore, to eliminate this effect of the Ad.CMV-NlacZ vector, we investigated the effect of systemic and local gene delivery of Ad.CMV-NlacZ and Ad.CMV-TNFR:Fc on osteoclastogenesis and osteolysis in athymic nude mice, which do not have mature T-cells. sTNFR:Fc levels in the sera of these mice were detectable by day 2 and peaked at 700 ng/mL on days 3 and 4 after both the systemic and local administrations of Ad.CMV-TNFR:Fc (Fig. 4-A). In contrast to the findings in the CBAxB6 mice, sTNFR:Fc could still be measured (300 ng/mL) in the sera collected on day 10, suggesting a lack of immunological recognition and clearance of the virus. No detectable levels of sTNFR:Fc were found in any of the control sera examined (Fig. 4-A). Contrary to the effect observed in the normal mice, the midline sagittal suture area and the number of osteoclasts remained unchanged (compared with control values) following local or systemic delivery of Ad.CMV-NlacZ (Figs. 4-B, 4-B, and 5). In the athymic nude mice, treatment with Ad.CMV-TNFR:Fc, with either delivery method, resulted in a significant reduction in bone resorption (to <0.0599 0.0041 mm²) compared with that in the titanium-treated control animals (p < 0.05) (Figs. 4-B and 5,C). Interestingly, the athymic nude mice treated with titanium alone had a smaller sagittal suture area (0.0757 0.0099 mm²) than did the CBAxB6 mice treated with equal amounts of titanium (0.1117 ± 0.0323 mm²), indicating that lymphocytes may contribute to the osteolysis in this model (Fig. 4-B). Osteoclast numbers in the mice treated with the sTNFR:Fc adenovirus were decreased compared with the numbers in the control mice, although this effect was not significant (Fig. 4-C).

Discussion

Introduction | Materials and Methods | Results | Discussion | References

Tumor necrosis factor-a is a pivotal cytokine involved in pathologic bone resorption such as that seen in association with aseptic loosening^2,25,26. TNF-a is the first proinflammatory cytokine produced by macrophages in response to wear-debris particles, and it has been shown to control the release of other proinflammatory mediators (such as granulocyte macrophage colony-stimulating factor, interleukin-1b, interleukin-6, and prostaglandin E2) that are also thought to contribute to aseptic loosening^2,6,10,11. Since these factors in turn induce osteoclastogenesis and activate osteoclasts, TNF-a blockade may be a dominant upstream event that has dramatic consequences on periprosthetic osteolysis. A variety of therapies designed to block TNF-a have already been tested and approved for use in the treatment of inflammatory diseases such as rheumatoid arthritis^12,27. Because the inflamed synovial and periprosthetic membranes are somewhat similar in composition²⁸ and are known to contain large amounts of TNF-a^3,4, we reasoned that therapeutic agents such as the TNF receptor fusion protein sTNFR:Fc would be effective in reducing the wear debris-induced osteolysis seen in patients with aseptic loosening.

Although other types of wear debris (such as polyethylene) are present in the interfacial membranes of failed prostheses, we used titanium particles to induce an osteolytic response in vivo. The particles were within the phagocytosable size range, they have been shown to stimulate monocytes in vitro by several investigators^1,10,29-31, and they provided an identical source of particles to be used in all of the experiments. Similarly, we chose the Ad.CMV-TNFR:Fc vector because high titers of infectious virus can be readily produced, because it is capable of secreting nanograms-per-milliliter quantities of sTNFR:Fc^18,20, and because several reports have already demonstrated its efficacy in ameliorating TNF-a-mediated disease, including septic shock¹⁸ and experimental arthritis^19,20. Additionally, this vector is similar to the one previously used in all of the animal and human studies of gene therapy for musculoskeletal conditions.

To evaluate the effect of local gene delivery in our model, we first tested the ability of Ad.CMV-NlacZ to infect the periosteum forty-eight hours after a subcutaneous injection over the calvaria. Figure 1,A shows that approximately 10% of the cells in the periosteum were infected following the administration of 10⁸ infectious units of Ad.CMV-NlacZ. Injection of a relatively small number of viral particles onto the mouse calvaria resulted in a relatively high rate of infection. Since TNFR:Fc is a secreted protein, we anticipated that a 10% rate of infection would result in adequate protein production, an effect that was confirmed by the detection of elevated systemic levels of sTNFR:Fc. Serum levels of sTNFR:Fc were also elevated following a systemic injection of the vector. It has been previously shown that most of the virus administered with systemic injection migrates to the liver and that secreted target proteins are produced from the transduced hepatocytes^22,24.

The inflammatory response to adenovirus has been well documented³², occasionally with catastrophic effects. The mechanism of the inflammatory response has been related to the expression of viral proteins, which leads to a T-cell-mediated immune response. The subsequent immunological detection of the infected cells results in their clearance and limits the duration of expression of target proteins. Thus, protein induction with use of adenoviral gene transfer is typically limited in duration, although there are reports of target protein expression for periods of several months^22,33,34. In the current study, serum levels of TNFR:Fc fell to undetectable levels by ten days in immunocompetent mice but remained elevated in T-cell-deficient nude mice, findings that support this paradigm of adenoviral gene transfer.

Despite these undesirable effects of adenoviruses, this vector has been successfully used in a number of gene-therapy applications in trials involving the treatment of both animal and human musculoskeletal diseases. Similarly, adenovirus-mediated delivery of cytokines^35-37, cytokine antagonists^19,20, and dominant negative signaling molecules^38,39 has been shown to be effective in animal models of arthritis. The mouse model of wear debris-induced osteolysis used in the studies reported here is a sensitive and rapid model of inflammatory bone loss and thus is ideal for evaluation of adenoviral gene transfer. In the sham-treated wild-type animals, the local and systemic delivery of the control adenoviral vector resulted in increases in both osteoclast numbers and bone resorption (Figs. 2-A, 2-B, and 2-C). In addition, the control virus enhanced the bone-resorption response to titanium particles. The role of the cell-mediated immune response in these effects was further confirmed by the absence of similar findings in athymic nude mice. These findings suggest that the immune response to adenovirus can have a deleterious effect on bone metabolism and can alter the normal balance between bone resorption and bone formation, a process that is ordinarily tightly coupled.

These results have broad orthopaedic implications, since adenovirus has been investigated as a potential delivery system for genes designed to stimulate bone formation and to inhibit periarticular inflammation and tissue catabolism. The dose of vector chosen for our studies falls within the range used by other investigators, and although lower doses of vector may reduce the negative effects of adenoviral gene transfer, it is likely that the therapeutic effects would also decrease.

In the absence of an inflammatory response to adenovirus (in the athymic nude mice), the delivery of TNFR:Fc resulted in a decrease in bone resorption compared with that in the titanium-treated controls, confirming the importance of TNF-a in the osteolytic response and the effectiveness of this blocking agent. Consistent with the results from etanercept experiments^16,40, the Ad.CMV-TNFR:Fc treatment had no effect on osteoclastogenesis in either the wild-type or the nude mice. The inhibitory response of sTNFR:Fc gene therapy in the nude mice was probably due to both the lack of an inflammatory response to adenovirus in these animals (as evidenced by the lack of an effect on bone resorption and osteoclastogenesis with Ad.CMV-NlacZ treatment alone) and to the sustained expression of TNFR:Fc. However, it was also noted that the average sagittal suture area in the titanium-treated nude mice was smaller than that observed in the titanium-treated normal mice. Since T-cells have been shown to synthesize the osteoclast-inducing factor receptor activator of NF-kappaB (RANK) ligand, the difference implicates T-cells as an active participant in particle-mediated bone loss. However, macrophages still play an important role, as evidenced by the increase in suture area above sham values in the titanium-treated nude mice. The effect of sTNFR:Fc gene therapy with a nonimmunogenic vector in an immunocompetent host is unknown, although one might expect that bone resorption would be reduced to background levels, as has been observed in nude mice and in normal mice treated with recombinant protein⁴⁰.

The use of TNF-a antagonists to treat inflammatory diseases such as rheumatoid arthritis has dramatically increased in recent years and shows promise for future applications as well. TNF-a inhibitors may be especially useful for the treatment of aseptic loosening and wear debris-induced osteolysis²⁸, which are characterized by the infiltration of monocytes and the expression of TNF-a that in turn leads to increased osteoclast development and activity. We suggested that TNF-a is involved in the activation of osteoclasts to resorb bone because of the finding that a soluble TNF-a inhibitor (sTNFR:Fc) prevented wear debris-induced osteolysis in a mouse calvaria model. Ad.CMVTNFR:Fc gene therapy prevents wear debris-induced osteolysis; however, the dominant inflammatory response to the adenoviral vector contraindicates its use in human gene therapy for the treatment of aseptic loosening. As gene therapy for the treatment of arthritis has gained momentum in recent years^37,41,42, future studies in this area may allow gene-therapy strategies for preventing aseptic loosening of orthopaedic implants to follow suit. Improved vectors lacking independent immunological effects will likely be more effective for the treatment of this and other musculoskeletal diseases.

References

Introduction | Materials and Methods | Results | Discussion | References

Blaine TA, Rosier RN, Puzas JE, Looney RJ, Reynolds PR, Reynolds SD,O’Keefe RJ. Increased levels of tumor necrosis factor-alpha and interleukin-6 protein and messenger RNA in human peripheral blood monocytes due to titanium particles. J Bone Joint Surg Am,1996;78: 1181-92. 781181 1996 [PubMed]

Horowitz SM,Purdon MA. Mechanisms of cellular recruitment in aseptic loosening of prosthetic joint implants. Calcif Tissue Int,1995;57: 301-5. 57301 1995 [PubMed][CrossRef]

al Saffar N,Revell PA. Interleukin-1 production by activated macrophages surrounding loosened orthopaedic implants: a potential role in osteolysis. Br J Rheumatol,1994;33: 309-16. 33309 1994 [PubMed][CrossRef]

Kadoya Y, Revell PA, al-Saffar N, Kobayashi A, Scott G,Freeman MA. Bone formation and bone resorption in failed total joint arthroplasties: histomorphometric analysis with histochemical and immunohistochemical technique.. J Orthop Res,1996;14: 473-82. 14473 1996 [PubMed][CrossRef]

Bertolini DR, Nedwin GE, Bringman TS, Smith DD,Mundy GR. Stimulation of bone resorption and inhibition of bone formation in vitro by human tumour necrosis factors. Nature,1986;319: 516-8. 319516 1986 [PubMed][CrossRef]

Lader CS,Flanagan AM. Prostaglandin E2, interleukin 1 alpha, and tumor necrosis factor-alpha increase human osteoclast formation and bone resorption in vitro. Endocrinology,1998;139: 3157-64. 1393157 1998 [PubMed][CrossRef]

Kobayashi K, Takahashi N, Jimi E, Udagawa N, Takami M, Kotake S, Nakagawa N, Kinosaki M, Yamaguchi K, Shima N, Yasuda H, Morinaga T, Higashio K, Martin TJ,Suda T. Tumor necrosis factor alpha stimulates osteoclast differentiation by a mechanism independent of the ODF/RANKL-RANK interaction. J Exp Med,2000;191: 275-86. 191275 2000 [PubMed][CrossRef]

Schwarz EM, Lu AP, Goater JJ, Benz EB, Kollias G, Rosier RN, Puzas JE,O’Keefe RJ.. Tumor necrosis factor-alpha/nuclear transcription factor-kappaB signaling in periprosthetic osteolysis.. J Orthop Res,2000;18: 472-80. 18472 2000 [PubMed][CrossRef]

Merkel KD, Erdmann JM, McHugh KP, Abu-Amer Y, Ross FP,Teitelbaum SL. Tumor necrosis factor-alpha mediates orthopedicimplant osteolysis. Am J Pathol,1999;154: 203-10. 154203 1999 [PubMed][CrossRef]

Wang JY, Wicklund BH, Gustilo RB,Tsukayama DT. Titanium, chromium and cobalt ions modulate the release of bone-associated cytokines by human monocytes/macrophages in vitro. Biomaterials,1996;17: 2233-40. 172233 1996 [PubMed][CrossRef]

Feldmann M, Brennan FM,Maini RN. Rheumatoid arthritis. Cell,1996;85: 307-10. 85307 1996 [PubMed][CrossRef]

Moreland LW, Baumgartner SW, Schiff MH, Tindall EA, Fleischmann RM, Weaver AL, Ettlinger RE, Cohen S, Koopman WJ, Mohler K, Widmer MB,Blosch CM. Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein. N Engl J Med,1997;337: 141-7. 337141 1997 [PubMed][CrossRef]

Bathon JM, Martin RW, Fleischmann RM, Tesser JR, Schiff MH, Keystone EC, Genovese MC, Wasko MC, Moreland LW, Weaver AL, Markenson J,Finck BK. A comparison of etanercept and methotrexate in patients with early rheumatoid arthritis. N Engl J Med,2000;343: 1586-93. 3431586 2000 [PubMed][CrossRef]

Lipsky PE, van der Heijde DM, St Clair EW, Furst DE, Breedveld FC, Kalden JR, Smolen JS, Weisman M, Emery P, Feldmann M, Harriman GR,Maini RN. Infliximab and methotrexate in the treatment of rheumatoid arthritis. Anti-Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy Study Group. N Engl J Med,2000;343: 1594-602. 3431594 2000 [PubMed][CrossRef]

Childs LM, Goater JJ, Sanz I, Ritchlin C, Looney RJ, Zwillich SJ,Schwarz EM. Efficacy of the soluble TNFa inhibitor (Enbrel) to prevent prosthetic wear debris-induced osteolysis. Arthritis Rheum,1999;42(Suppl): 154. 42(Suppl)154 1999

Clohisy JC, Teitelbaum SL, Ross FP, Erdmann JM,Abu-Amer Y. Blockade of TNF-activation of NF-kB in osteoclast precursors prevents implant osteolysis. J Bone Miner Res,1999;14(Suppl): 489. 14(Suppl)489 1999

Peppel K, Crawford D,Beutler B. A tumor necrosis factor (TNF) receptor-IgG heavy chain chimeric protein as a bivalent antagonist of TNF activity. J Exp Med,1991;174: 1483-9. 1741483 1991 [PubMed][CrossRef]

Kolls J, Peppel K, Silva M,Beutler B. Prolonged and effective blockade of tumor necrosis factor activity through adenovirus-mediated gene transfer [erratum in 1994;91:1979]. Proc Natl Acad Sci U S A,1994;91: 215-9. 91215 1994 [PubMed][CrossRef]

Ghivizzani SC, Lechman ER, Kang R, Tio C, Kolls J, Evans CH,Robbins PD. Direct adenovirus-mediated gene transfer of interleukin 1 and tumor necrosis factor alpha soluble receptors to rabbit knees with experimental arthritis has local and distal anti-arthritic effects. Proc Natl Acad Sci U S A,1998;95: 4613-8. 954613 1998 [PubMed][CrossRef]

Quattrocchi E, Walmsley M, Browne K, Williams RO, Marinova-Mutafchieva L, Buurman W, Butler DM,Feldmann M. Paradoxical effects of adenovirus mediated blockade of TNF activity in murine collagen-induced arthritis. J Immunol,1999;163: 1000-9.. 1631000 1999 [PubMed]

Schwarz EM, Benz EB, Lu AP, Goater JJ, Mollano AV, Rosier RN, Puzas JE,O’Keefe RJ. Quantitative small-animal surrogate to evaluate drug efficacy in preventing wear debris-induced osteolysis. J Orthop Res,2000;18: 849-55. 18849 2000 [PubMed][CrossRef]

Dai Y, Schwarz EM, Gu D, Zhang WW, Sarvetnick N,Verma IM. Cellular and humoral immune responses to adenoviral vectors containing factor IX gene: tolerization of factor IX and vector antigens allows for long-term expression. Proc Natl Acad Sci U S A,1995;92: 1401-5. 921401 1995 [PubMed][CrossRef]

Le CH, Nicolson AG, Morales A,Sewell KL. Suppression of collagen-induced arthritis through adenovirus-mediated transfer of a modified tumor necrosis factor alpha receptor gene. Arthritis Rheum,1997;40: 1662-9. 401662 1997 [PubMed][CrossRef]

Yang Y, Ertl HC,Wilson JM. MHC class I-restricted cytotoxic T lymphocytes to viral antigens destroy hepatocytes in mice infected with E1-deleted recombinant adenoviruses. Immunity,1994;1: 433-42. 1433 1994 [PubMed][CrossRef]

Ishiguro N, Kojima T, Ito T, Saga S, Anma H, Kurokouchi K, Iwahori Y, Iwase T,Iwata H. Macrophage activation and migration in interface tissue around loosening total hip arthroplasty components. J Biomed Mater Res,1997;35: 399-406. 35399 1997 [PubMed][CrossRef]

Xu JW, Konttinen YT, Lassus J, Natah S, Ceponis A, Solovieva S, Aspenberg P,Santavirta S. Tumor necrosis factor-alpha (TNT-alpha) in loosening of total hip replacement (THR). Clin Exp Rheumatol,1996;14: 643-8. 14643 1996 [PubMed]

Maini RN, Elliott MJ, Brennan FM, Williams RO, Chu CQ, Paleolog E, Charles PJ, Taylor PC,Feldmann M. Monoclonal anti-TNF alpha antibody as a probe of pathogenesis and therapy of rheumatoid disease. Immunol Rev,1995;144: 195-223. 144195 1995 [PubMed][CrossRef]

Schwarz EM, Looney RJ,O’Keefe RJ. Anti-TNF-alpha therapy as a clinical intervention for periprosthetic osteolysis. Arthritis Res,2000;2: 165-8. 2165 2000 [PubMed][CrossRef]

Glant TT,Jacobs JJ. Response of three muring macrophage populations to particulate debris: bone resorption in organ cultures. J Orthop Res,1994;12: 720-31. 12720 1994 [PubMed][CrossRef]

Shanbhag AS, Jacobs JJ, Black J, Galante JO,Glant TT.. Human monocyte response to particulate biomaterials generated in vivo and in vitro. J Orthop Res,1995;13: 792-801. 13792 1995 [PubMed][CrossRef]

Lee SH, Brennan FR, Jacobs JJ, Urban RM, Ragasa DR,Glant TT.. Human monocyte/macrophage response to cobalt-chromium corrosion products and titanium particles in patients with total joint replacements. J Orthop Res,1997;15: 40-9. 1540 1997 [PubMed][CrossRef]

Verma IM,Somia N. Gene therapy—promises, problems and prospects [news]. Nature,1997;389: 239-42. 389239 1997 [PubMed][CrossRef]

Geddes BJ, Harding TC, Hughes DS, Byrnes AP, Lightman SL, Conde G,Uney JB.. Persistent transgene expression in the hypothalamus following stereotaxic delivery of a recombinant adenovirus: suppression of the immune response with cyclosporin. Endocrinology,1996;137: 5166-9. 1375166 1996 [PubMed][CrossRef]

Navarro V, Millecamps S, Geoffroy MC, Robert JJ, Valin A, Mallet J,Gal La Salle GL.. Efficient gene transfer and long-term expression in neurons using a recombinant adenovirus with a neuron-specific promoter. Gene Ther,1999;6: 1884-92.. 61884 1999 [PubMed][CrossRef]

Ma Y, Thornton S, Duwel LE, Boivin GP, Giannini EH, Leiden JM, Bluestone JA,Hirsch R. Inhibition of collagen-induced arthritis in mice by viral IL-10 gene transfer. J Immunol,1998;161: 1516-24. 1611516 1998 [PubMed]

Whalen JD, Lechman EL, Carlos CA, Weiss K, Kovesdi L, Glorioso JC, Robbins PD,Evans CH. Adenoviral transfer of the viral IL-10 gene periarticularly to mouse paws suppresses development of collagen-induced arthritis in both injected and uninjected paws. J Immunol,1999;162: 3625-32. 1623625 1999 [PubMed]

Lubberts E, Joosten LA, Chabaud M, van Den Bersselaar L, Oppers B, Coenen-De Roo CJ, Richards CD, Miossec P,van Den Berg WB. IL-4 gene therapy for collagen arthritis suppresses synovial IL-17 and osteoprotegerin ligand and prevents bone erosion. J Clin Invest,2000;105: 1697-710. 1051697 2000 [PubMed][CrossRef]

Bondeson J, Foxwell B, Brennan F,Feldmann M. Defining therapeutic targets by using adenovirus: blocking NF-kappaB inhibits both inflammatory and destructive mechanisms in rheumatoid synovium but spares anti-inflammatory mediators. Proc Natl Acad Sci U S A,1999;96: 5668-73. 965668 1999 [PubMed][CrossRef]

Takayanagi H, Juji T, Miyazaki T, Iizuka H, Takahashi T, Isshiki M, Okada M, Tanaka Y, Koshihara Y, Oda H, Kurokawa T, Nakamura K,Tanaka S. Suppression of arthritic bone destruction by adenovirus-mediated csk gene transfer to synoviocytes and osteoclasts. J Clin Invest,1999;104: 137-46. 104137 1999 [PubMed][CrossRef]

Childs LM, Goater JJ, O’Keefe RJ,Schwarz EM. Efficacy of etanercept for wear debris-induced osteolysis. J Bone Miner Res,2001;16: 338-4. 16338 2001 [PubMed][CrossRef]

Evans CH, Ghivizzani SC, Kang R, Muzzonigro T, Wasko MC, Herndon JH,Robbins PD. Gene therapy for rheumatic diseases. Arthritis Rheum,1999;42: 1-16. 421 1999 [PubMed][CrossRef]

Kim SH, Evans CH, Kim S, Oligino T, Ghivizzani SC,Robbins PD. Gene therapy for established murine collagen-induced arthritis by local and systemic adenovirus-mediated delivery of interleukin-4. Arthritis Res,2000;2: 293-302. 2293 2000 [PubMed][CrossRef]

Topics

View Large | Download Slide(.ppt)
Add to My JBJS

View Larger

TABLE I: Experimental Groups

Titanium Particles	Treatment	Mouse Strain	No. of Animals
—	None	CBAxB6	5
—	Ad.CMV-NlacZ (local)	CBAxB6	5
—	Ad.CMV-NlacZ (intraperitoneal)	CBAxB6	5
+	Phosphate-buffered saline solution (intraperitoneal)	CBAxB6	5
+	Ad.CMV-NlacZ (local)	CBAxB6	5
+	Ad.CMV-NlacZ (intraperitoneal)	CBAxB6	5
+	Ad.CMV-TNFR:Fc (local)	CBAxB6	5
+	Ad.CMV-TNFR:Fc (intraperitoneal)	CBAxB6	5
—	None	Athymic nude	4
—	Ad.CMV-NlacZ (local)	Athymic nude	4
—	Ad.CMV-NlacZ (intraperitoneal)	Athymic nude	4
+	Phosphate-buffered saline solution (intraperitoneal)	Athymic nude	4
+	Ad.CMV-NlacZ (local)	Athymic nude	4
+	Ad.CMV-NlacZ (intraperitoneal)	Athymic nude	4
+	Ad.CMV-TNFR:Fc (local)	Athymic nude	4
+	Ad.CMV-TNFR:Fc (intraperitoneal)	Athymic nude	4

Add to My JBJS

TABLE I: Experimental Groups

Titanium Particles	Treatment	Mouse Strain	No. of Animals
—	None	CBAxB6	5
—	Ad.CMV-NlacZ (local)	CBAxB6	5
—	Ad.CMV-NlacZ (intraperitoneal)	CBAxB6	5
+	Phosphate-buffered saline solution (intraperitoneal)	CBAxB6	5
+	Ad.CMV-NlacZ (local)	CBAxB6	5
+	Ad.CMV-NlacZ (intraperitoneal)	CBAxB6	5
+	Ad.CMV-TNFR:Fc (local)	CBAxB6	5
+	Ad.CMV-TNFR:Fc (intraperitoneal)	CBAxB6	5
—	None	Athymic nude	4
—	Ad.CMV-NlacZ (local)	Athymic nude	4
—	Ad.CMV-NlacZ (intraperitoneal)	Athymic nude	4
+	Phosphate-buffered saline solution (intraperitoneal)	Athymic nude	4
+	Ad.CMV-NlacZ (local)	Athymic nude	4
+	Ad.CMV-NlacZ (intraperitoneal)	Athymic nude	4
+	Ad.CMV-TNFR:Fc (local)	Athymic nude	4
+	Ad.CMV-TNFR:Fc (intraperitoneal)	Athymic nude	4