- Email: [email protected]
The temporal course of adenovirus-mediated gene transfer in the rat larynx
The temporal course of adenovirus-mediated gene transfer in the rat larynx IAN N. JACOBS, MD, RALPH P. TUFANO, MD, DANIELLE S. WALSH, MD, and TIMOTHY CROMBLEHOLME, MD, Philadelphia,
Pennsylvania
OBJECTIVE: Polypeptide growth factors have important influences on wound-healing and scar tissue formation. Specific growth factors or their inhibitors may potentially decrease scar tissue formation and prevent subglottic stenosis. Gene transfer using recombinant adenovirus may be an ideal method to mediate endogenous production of growth factors to inhibit fibrosis. STUDY DESIGN: The study incorporated adenovirusmediated transduction of normal and stenotic rat larynges and histologic analysis of the sequential expression of a β-galactosidase marker gene over time. SETTING: The study was conducted at the animal care facility of an academic children’s hospital. RESULTS: We report successful transduction in normal and injured rat larynx with peak expression of β-galactosidase at 2 days after transduction and almost complete disappearance by 7 days. There appeared to be an early inflammatory response to the viral injection, but at 7 and 14 days after injection (transduction) the uninjured rat larynges From the Division of Pediatric Otolaryngology (Drs Jacobs and Tufano), Children’s Institute for Surgical Science, and the Division of Pediatric and Fetal Surgery (Drs Walsh, Radu, and Crombleholme), The Children’s Hospital of Philadelphia, University of Pennsylvania School of Medicine. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head Neck Surgery, New Orleans, LA, September 28, 1999. This study was supported by the Percy Memorial Award Grant from the American Academy of Otolaryngology–Head and Neck Surgery. Reprint requests: Ian N. Jacobs, MD, The Children’s Hospital of Philadelphia, Division of Pediatric Otolaryngology, 34th St and Civic Center Blvd, 1 Wood Center, Philadelphia, PA 19104; email, [email protected] Copyright © 2002 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. 0194-5998/2002/$35.00 + 0 23/1/122633 doi:10.1067/mhn.2002.122633
resumed a normal histologic appearance. All distant sites stained negative for β-galactosidase. CONCLUSION: Recombinant adenovirus-mediated gene transfer is feasible in the rat larynx with transient duration and limited toxicity. SIGNIFICANCE: Adenovirus-mediated gene transfer has the potential to deliver growth factors that modulate wound healing and inflammation in the larynx by inhibiting fibrosis. (Otolaryngol Head Neck Surg 2002;126:281-9.)
G
ene therapy holds promise for the treatment of inherited genetic diseases such as cystic fibrosis and acquired diseases such as cancer and chronic inflammatory disorders.1 It may also be useful in wound healing; adenovirus may be an effective vector for the transfer of growth factors or cytokines to modulate wound healing and potentially decrease scar tissue formation.1 Adenovirus has been shown in experimental studies to transfect a broad range of human cells at high efficiency, including fibroblasts, keratinocytes, and endothelial cells.2 In addition, the adenovirus has demonstrated gene transfer efficiency in human airway epithelium and has been the subject of numerous studies in animal models of cystic fibrosis.3-5 Subglottic stenosis (SGS), a narrowing of the larynx below the vocal cords, occurs in a significant percentage of premature infants who are ventilated for prolonged periods of time.6 SGS is associated with injury and infection of the cricoid ring. When healing leads to excessive deposition of connective tissue (collagen, fibronectin), permanent stenosis results. Substantial evidence indicates that transforming growth factor β1 (TGF-β1) is a potent mediator of fibrosis and scarring in many organ systems of the human body.7 TGF-β1 induces the formation of fibronectin, collagen, tenascin, and other extracellular matrix proteins. 281
JACOBS et al
➔
282
Otolaryngology– Head and Neck Surgery March 2002
Fig 1. Cross section of an uninjured rat larynx 1 day after viral injection stained for β-galactosidase; light staining is revealed (arrow) in the submucosal space of the posterior wall of the larynx. (Original magnification ×20.)
Blocking antibodies seem to reduce the expression of extracellular matrix proteins. Other TGF-β1 antagonists, including TGF-β3 and decorin, are potential inhibitors of TGFβ1 activity.8 Moreover, a major stumbling block to effective growth factor therapy is the inefficiency of the delivery systems and their limited duration of action. To be effective, specific growth factors or their inhibitors must be administered continuously during critical stages of wound healing at optimal concentrations.9 Virally mediated gene transfer is a potential way to overcome this problem. Furthermore, the transient period of adenovirus expression may be ideal for modulating wound healing. However, there are several potential problems with an adenovirus gene transfer system, including the actual affinity of the virus for the laryngeal tissues, as well as the duration of viral expression in laryngeal cells. In addition, the immune response may limit in vivo expression of the virus in the airway and may in itself elicit a
large inflammatory response, which would be counterproductive to efforts to decrease the degree of scar tissue formation. On the other hand, a brief duration of expression may be all that is necessary to enhance intrinsic wound healing. Because the effects of adenovirus-mediated gene transfer in the larynx have not been previously studied, a β-galactosidase reporter gene (LacZ) was used to evaluate gene transfer in the rat larynx. The first objective of this investigation was to determine the feasibility of adenovirus-mediated gene transfer in both normal and stenotic larynges. The second goal of this study was to determine the nature and extent of the local cellular immune response to the adenovirus vector in the rat larynx. MATERIALS AND METHODS The virus used in this study was a first-generation recombinant human serotype 5 adenovirus (seed stocks were obtained from Vector Core, Institute for Human Gene Therapy, University of
Otolaryngology– Head and Neck Surgery Volume 126 Number 3
JACOBS et al
283
➔ Fig 2. Cross section of uninjured rat larynx 2 days after injection stained for β-galactosidase, showing more intense staining (arrow) in both the luminal and extraluminal submucosal spaces of the posterior laryngeal wall. (Original magnification ×20.)
Pennsylvania). The virus is a double-stranded DNA virus that has been rendered replication deficient through the deletion of the E1 region of its viral genome.10 The Ela and Elb coding sequences have been replaced with an expression cassette consisting of the Escherichia coli β-galactosidase gene coding sequence as a reporter gene driven by a cytomegalovirus promoter (Ad.CMV.LacZ). The replication-deficient adenovirus was propagated in human embryonal kidney cells (293 cells), which complement the viral E1 gene products. Viral titers were determined by plaque-forming assay.11 Viral titers of 1 × 108 plaque-forming units (pfu) were used based on empiric observations with particleto-pfu ratio of 25:1. Forty-five adult male Sprague-Dawley rats were used in this study, which was approved by the Institutional Animal Care and Use Committee of The Children’s Hospital of Philadelphia in accordance with “The Principles of Laboratory Animal Care” formulated by the National Society for
Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH publication No. 80-23, revised 1978). All rats in the study were anesthetized using inhalational isoflurane and intraperitoneal ketamine. Gene transfer was first attempted in 25 uninjured control rats. All of these rats underwent an anterior laryngofissure to gain access to the cricoid. The midline posterior cricoid wall was injected submucosally with 25 µl of virus (1 × 108 pfu in HEPES solution) or HEPES solution alone (controls) with a 33-gauge needle and a 50-µl Hamilton syringe. There were 3 separate adenovirus and 2 HEPES solution injections at each time point. Rats were killed at 1, 2, 4, 7, and 14 days after surgery. The larynges were removed for histologic processing. All the specimens were embedded in paraffin, and the midpoint of the injury site was
284
Otolaryngology– Head and Neck Surgery March 2002
JACOBS et al
Fig 3. Cross section of a more caudal view of the rat larynx stained for β-galactosidase 4 days after injection, showing no expression. (Original magnification ×20.)
sectioned and underwent X-gal staining for βgalactosidase expression to assess gene transfer and immunostaining for fibronectin and type 1 procollagen, using standard immunohistochemical techniques. The stained specimens were mounted onto glass slides and underwent qualitative morphometric analysis. To assess the local inflammatory response, the same specimens underwent staining for hematoxylin and eosin. Tissue samples from the spleen, gonads, liver, brain, lymph nodes, and stomach underwent β-galactosidase staining as well to detect systemic dissemination of the virus. Once a successful gene was established in the uninjured controls, the rats with SGS were transduced as well. To induce SGS, 20 rats underwent a ventral laryngofissure to gain access to the posterior tracheal wall. A 1- × 3-mm area of mucosa and perichondrium was removed from the posterior cricoid wall under an operating microscope with magnification of ×10. To induce a severe grade of SGS, 100 µl of 10N NaOH was applied to the
wound area. The laryngofissure site was then closed, and the animals were reanesthetized 2 days later. The Hamilton syringe was used to infiltrate the entire posterior cricoid submucosa with 1 × 10 8 pfu of recombinant human serotype 5 adenovirus carrying the β-galactosidase reporter gene (LacZ) in HEPES. Three animals were treated with adenovirus and 2 control animals were injected with HEPES solution at each time point. The larynges and organs were harvested at 1, 2, 4, and 7 days after transduction and underwent identical histologic and immunochemical analyses as the uninjured specimens. RESULTS In the normal uninjured rat larynx, β-galactosidase was expressed in the early time periods after injection. β-Galactosidase staining was detected in the perichondrium, epithelium, chondrocytes, and fibroblasts 1 and 2 days after transduction (Figs 1 and 2). Its expression became faint at 4 days in most specimens, and it could not be detected at 7
Otolaryngology– Head and Neck Surgery Volume 126 Number 3
JACOBS et al
285
➔ Fig 4. Cross section of the injured rat larynx 2 days after viral injection and 4 days after injury to the posterior cricoid wall, showing the most intense staining (arrow) of β-galactosidase in the submucosal space. (Original magnification ×20.)
days (Fig 3). The 14-day specimen was negative as well. In the first 2 days after transduction, the staining intensity for β-galactosidase appeared greater in the SGS specimens than in the uninjured control specimens. In the injured specimens, the intensity of β-galactosidase stain peaked at 2 days after injection (Fig 4). At 4 days, the epithelium stained positive but is less intense. At 7 days, staining was negative (Fig 5). On the hematoxylin and eosin stains, there was a moderate inflammatory response in the larynges of the control specimens injected with HEPES solution. Mucous gland hyperplasia and an inflammatory cell infiltration were observed in the posterior and lateral walls of the cricoid (Fig 6A). The inflammatory cell infiltrate appeared to be enhanced at the site of adenoviral injection at 2 days after injection (Fig 6B). However, at 7 and 14 days, the transduced specimens assumed a normal histologic appearance without evidence of SGS (Fig 7A), in stark contrast to the injured specimens
(Fig 7B). In addition, the distant sites (spleen, liver, lymph nodes, brain, gonads, and stomach) stained negative in all of the transduced animals. DISCUSSION This is the first reported demonstration of successful gene transfer to the larynx using an adenovirus vector. Direct injection lead to transduction of the submucosal connective tissue, where collagen and other extracellular matrix proteins are deposited in the evolution of SGS. The duration of transgene expression was transient as β-galactosidase activity peaked at 2 days and essentially disappeared by 7 days. Although this transient period of transgene expression may be a problem for long-term genetic diseases, it may be ideal for wound healing. The critical stages of wound healing and extracellular matrix expression occur in the first several days after injury. We also found an intense inflammatory response to the injection of the HEPES solution at
286
JACOBS et al
Otolaryngology– Head and Neck Surgery March 2002
Fig 5. Cross section of the injured rat larynx 7 days after injection and 9 days after injury, showing significant subglottic stenosis but no β-galactosidase staining. (Original magnification ×20.)
the posterior cricoid and what appeared to be an augmented immune response to the virus itself. However, there did not seem to be any long-lasting effects as the transduced specimens resumed a nearly normal histologic appearance by 7 days after injection. There also was no distal infection or toxicity from the adenovirus. Although a single administration of adenovirus appeared to be well tolerated, repeated injections might create a greater problem with the immune response.1 The inflammatory response after multiple injections will have to be addressed in future studies. Alternatively, nonviral approaches may also be considered. The adenovirus is a double-stranded DNA virus that transfects both dividing and nondividing cells with high efficiency. A wide variety of human cells are transduced by the adenovirus, including epithelial cells, endothelial cells, and keratinocytes.2 The virus remains episomal and is not fully integrated into the genome. Therefore, transient periods of expression are expected (days to weeks).1 In addition, high titers of 1014 can be isolated and puri-
fied.1 The vector may be ideal for modulating airway wound-healing response and has been used extensively in cystic fibrosis research.3-5 A first-generation recombinant adenovirus was used in this investigation, but newer generations of adenoviral vectors may be less immunogenic and have a greater transduction efficiency.12 Specific genomic deletions may increase the cloning capacity with less toxicity. In addition, immune modulations, such as immunosuppressive agents, may block the inflammatory response.13,14 This may prolong the expression of the important cytokines. It also is not clear why the β-galactosidase gene expression appeared to be enhanced in the injured rat laryngeal tissue. Possible explanations include CAR receptor up-regulation, reduced cell-mediated immune response to adenoviral infection, and delayed clearance by inflammatory cell infiltrate. These explanations are purely speculative and need confirmation in future studies. This study sets the background for future approaches to modulate the wound-healing response in the airway. The use of an adenoviral
Otolaryngology– Head and Neck Surgery Volume 126 Number 3
JACOBS et al
287
A
B
➔
C Fig 6. A, Cross section of the cricoid posterior wall injection site 4 days after administration of HEPES solution (Control). The β-galactosidase stain reveals evidence of mucous gland hyperplasia (large arrow) with a mild inflammatory cell infiltration (small arrow). (Original magnification ×50.) B, Cross section of the posterior cricoid wall at the site of adenovirus injection at 4 days later, revealing an intense inflammatory cell infiltration (arrow). (Original magnification ×50.) C, Higher magnification view of Fig 6B revealing an intense inflammatory cell infiltration. (Original magnification x250.)
vector to deliver therapeutic proteins for the treatment of impaired wound healing may represent an effective application of current gene transfer tech-
niques. Future specific applications of the adenovirus in a laryngeal model include gene transfer of transgene-expressing inhibitors of TGF-β1
288
Otolaryngology– Head and Neck Surgery March 2002
JACOBS et al
A
B Fig 7. A, Cross section of the uninjured rat larynx 14 days after Ad.CMV.LacZ transduction, revealing an almost normal histologic appearance with hemotoxylin and eosin stain except for a small area of submucosal fibrosis at the site of the adenovirus injection (arrow). (Original magnification ×10.) B, Cross section of the rat larynx at 21 days after AdCMV.LacZ transduction and 23 days after injury, revealing severe subglottic stenosis. (Original magnification ×10.)
activity such as neutralizing antibodies (anti–TGFβ1) or TGF-β3. These agents have been shown to inhibit fibroplasia and cutaneous scar formation in an animal model.7 Their transient expression in the submucosa may lead to a decrease in collagen and scar tissue in the larynx after direct injection. CONCLUSIONS An effective in vivo model was established to study modulation of airway wound healing using an adenovirus-mediated gene transfer. Successful
adenovirus-mediated tranduction of both injured and uninjured rat larynges was achieved with a first-generation human recombinant adenoviral vector. Injury enhanced β-galactosidase transduction. β-Galactosidase transgene expression was detectable 1 day after injection and was transient as its expression disappeared by 7 days. The background inflammatory response was transient, and there was no systemic dissemination of the virus. Thus, it appears that the human recombinant adenovirus is an effective vector for gene transfer
Otolaryngology– Head and Neck Surgery Volume 126 Number 3
in the rat larynx and may facilitate novel approaches to airway wound healing. Future strategies might include use of multiple injections or advanced-generation adenoviral vectors or nonviral vectors, which may lead to a reduced inflammatory reaction and greater persistence. TGF-β1 inhibitors, transduced using the adenoviral vector, may decrease scar tissue formation in the submucosa of the larynx.
JACOBS et al
6. 7.
8.
REFERENCES
1. Jain KK. Vectors for gene therapy. In: Textbook of gene therapy, ed 1. Seattle, WA: Hogrefe and Huber Publishers; 1998. Chapter 4. 2. Setoguchi Y, Jaffe HA, Danel C, et al. Ex vivo and in vivo gene transfer to skin using replication-deficient recombinant vectors. Soc Invest Dermatol 1994;102:415-21. 3. Goldman MJ, Yang Y, Wilson JM. Gene therapy in a zenograft model of cystic fibrosis lung corrects chloride transport more effectively than sodium defect. Nat Genet 1995;9:126-31. 4. Dupuit F, Zahm J-M, Pierrot D, et al. Regenerating cells in human airway surface epithelium represent preferential targets for recombinant adenovirus. Hum Gene Ther 1995;6;1185-93. 5. Jiang C, Akita GY, Collegel WH, et al. Increased contact
9. 10. 11. 12. 13.
14.
289
time improves adenovirus-mediated CFTR gene transfer to nasal epithelium of CF mice. Hum Gene Ther 1997;8:671-80. Parkin JP, Stevens MH, Jung AL. Acquired and congenital subglottic stenosis in the infant. Ann Otol Rhinol Laryngol 1976;85:573-81. Shah M, Foreman DM, Ferguson MWJ. Neutralisation of TGF-β1 and TGF-β2 or exogenous addition of TGF-β3 to cutaneous rat wounds reduces scarring. J Cell Sci 1995;108:985-1002. Border WA, Noble NA, Yamamoto T, et al. Natural inhibitor of transforming growth factor-B protects against scarring in experimental kidney disease. Nature 1992;9360:361-4. Hom DB. Growth factors and wound-healing in otolaryngology. Otolaryngol Head Neck Surg 1994;110:560-4. Trapnell BC. 1993. Graham FL. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 1973;52:456-67. Wang Q, Finer MH. Second-generation adenovirus vectors. Nat Med 1996;2:714-6. Kaplan JM, Smith AE. Transient immunosuppression with deoxyspergualin improves longevity of transgene expression and ability to re-administer adenoviral vector to the mouse lung. Hum Gene Ther 1997;8:1095-104. Worgall S, Leoplad PL, Wolff G, et al. Role of alveolar macrophages in rapid elimination of adenovirus vectors administered to the epithelial surface of the respiratory tract. Hum Gene Ther 1997;8:1675-84.
290
JACOBS et al
Otolaryngology– Head and Neck Surgery March 2002
Pennsylvania
OBJECTIVE: Polypeptide growth factors have important influences on wound-healing and scar tissue formation. Specific growth factors or their inhibitors may potentially decrease scar tissue formation and prevent subglottic stenosis. Gene transfer using recombinant adenovirus may be an ideal method to mediate endogenous production of growth factors to inhibit fibrosis. STUDY DESIGN: The study incorporated adenovirusmediated transduction of normal and stenotic rat larynges and histologic analysis of the sequential expression of a β-galactosidase marker gene over time. SETTING: The study was conducted at the animal care facility of an academic children’s hospital. RESULTS: We report successful transduction in normal and injured rat larynx with peak expression of β-galactosidase at 2 days after transduction and almost complete disappearance by 7 days. There appeared to be an early inflammatory response to the viral injection, but at 7 and 14 days after injection (transduction) the uninjured rat larynges From the Division of Pediatric Otolaryngology (Drs Jacobs and Tufano), Children’s Institute for Surgical Science, and the Division of Pediatric and Fetal Surgery (Drs Walsh, Radu, and Crombleholme), The Children’s Hospital of Philadelphia, University of Pennsylvania School of Medicine. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head Neck Surgery, New Orleans, LA, September 28, 1999. This study was supported by the Percy Memorial Award Grant from the American Academy of Otolaryngology–Head and Neck Surgery. Reprint requests: Ian N. Jacobs, MD, The Children’s Hospital of Philadelphia, Division of Pediatric Otolaryngology, 34th St and Civic Center Blvd, 1 Wood Center, Philadelphia, PA 19104; email, [email protected] Copyright © 2002 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. 0194-5998/2002/$35.00 + 0 23/1/122633 doi:10.1067/mhn.2002.122633
resumed a normal histologic appearance. All distant sites stained negative for β-galactosidase. CONCLUSION: Recombinant adenovirus-mediated gene transfer is feasible in the rat larynx with transient duration and limited toxicity. SIGNIFICANCE: Adenovirus-mediated gene transfer has the potential to deliver growth factors that modulate wound healing and inflammation in the larynx by inhibiting fibrosis. (Otolaryngol Head Neck Surg 2002;126:281-9.)
G
ene therapy holds promise for the treatment of inherited genetic diseases such as cystic fibrosis and acquired diseases such as cancer and chronic inflammatory disorders.1 It may also be useful in wound healing; adenovirus may be an effective vector for the transfer of growth factors or cytokines to modulate wound healing and potentially decrease scar tissue formation.1 Adenovirus has been shown in experimental studies to transfect a broad range of human cells at high efficiency, including fibroblasts, keratinocytes, and endothelial cells.2 In addition, the adenovirus has demonstrated gene transfer efficiency in human airway epithelium and has been the subject of numerous studies in animal models of cystic fibrosis.3-5 Subglottic stenosis (SGS), a narrowing of the larynx below the vocal cords, occurs in a significant percentage of premature infants who are ventilated for prolonged periods of time.6 SGS is associated with injury and infection of the cricoid ring. When healing leads to excessive deposition of connective tissue (collagen, fibronectin), permanent stenosis results. Substantial evidence indicates that transforming growth factor β1 (TGF-β1) is a potent mediator of fibrosis and scarring in many organ systems of the human body.7 TGF-β1 induces the formation of fibronectin, collagen, tenascin, and other extracellular matrix proteins. 281
JACOBS et al
➔
282
Otolaryngology– Head and Neck Surgery March 2002
Fig 1. Cross section of an uninjured rat larynx 1 day after viral injection stained for β-galactosidase; light staining is revealed (arrow) in the submucosal space of the posterior wall of the larynx. (Original magnification ×20.)
Blocking antibodies seem to reduce the expression of extracellular matrix proteins. Other TGF-β1 antagonists, including TGF-β3 and decorin, are potential inhibitors of TGFβ1 activity.8 Moreover, a major stumbling block to effective growth factor therapy is the inefficiency of the delivery systems and their limited duration of action. To be effective, specific growth factors or their inhibitors must be administered continuously during critical stages of wound healing at optimal concentrations.9 Virally mediated gene transfer is a potential way to overcome this problem. Furthermore, the transient period of adenovirus expression may be ideal for modulating wound healing. However, there are several potential problems with an adenovirus gene transfer system, including the actual affinity of the virus for the laryngeal tissues, as well as the duration of viral expression in laryngeal cells. In addition, the immune response may limit in vivo expression of the virus in the airway and may in itself elicit a
large inflammatory response, which would be counterproductive to efforts to decrease the degree of scar tissue formation. On the other hand, a brief duration of expression may be all that is necessary to enhance intrinsic wound healing. Because the effects of adenovirus-mediated gene transfer in the larynx have not been previously studied, a β-galactosidase reporter gene (LacZ) was used to evaluate gene transfer in the rat larynx. The first objective of this investigation was to determine the feasibility of adenovirus-mediated gene transfer in both normal and stenotic larynges. The second goal of this study was to determine the nature and extent of the local cellular immune response to the adenovirus vector in the rat larynx. MATERIALS AND METHODS The virus used in this study was a first-generation recombinant human serotype 5 adenovirus (seed stocks were obtained from Vector Core, Institute for Human Gene Therapy, University of
Otolaryngology– Head and Neck Surgery Volume 126 Number 3
JACOBS et al
283
➔ Fig 2. Cross section of uninjured rat larynx 2 days after injection stained for β-galactosidase, showing more intense staining (arrow) in both the luminal and extraluminal submucosal spaces of the posterior laryngeal wall. (Original magnification ×20.)
Pennsylvania). The virus is a double-stranded DNA virus that has been rendered replication deficient through the deletion of the E1 region of its viral genome.10 The Ela and Elb coding sequences have been replaced with an expression cassette consisting of the Escherichia coli β-galactosidase gene coding sequence as a reporter gene driven by a cytomegalovirus promoter (Ad.CMV.LacZ). The replication-deficient adenovirus was propagated in human embryonal kidney cells (293 cells), which complement the viral E1 gene products. Viral titers were determined by plaque-forming assay.11 Viral titers of 1 × 108 plaque-forming units (pfu) were used based on empiric observations with particleto-pfu ratio of 25:1. Forty-five adult male Sprague-Dawley rats were used in this study, which was approved by the Institutional Animal Care and Use Committee of The Children’s Hospital of Philadelphia in accordance with “The Principles of Laboratory Animal Care” formulated by the National Society for
Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH publication No. 80-23, revised 1978). All rats in the study were anesthetized using inhalational isoflurane and intraperitoneal ketamine. Gene transfer was first attempted in 25 uninjured control rats. All of these rats underwent an anterior laryngofissure to gain access to the cricoid. The midline posterior cricoid wall was injected submucosally with 25 µl of virus (1 × 108 pfu in HEPES solution) or HEPES solution alone (controls) with a 33-gauge needle and a 50-µl Hamilton syringe. There were 3 separate adenovirus and 2 HEPES solution injections at each time point. Rats were killed at 1, 2, 4, 7, and 14 days after surgery. The larynges were removed for histologic processing. All the specimens were embedded in paraffin, and the midpoint of the injury site was
284
Otolaryngology– Head and Neck Surgery March 2002
JACOBS et al
Fig 3. Cross section of a more caudal view of the rat larynx stained for β-galactosidase 4 days after injection, showing no expression. (Original magnification ×20.)
sectioned and underwent X-gal staining for βgalactosidase expression to assess gene transfer and immunostaining for fibronectin and type 1 procollagen, using standard immunohistochemical techniques. The stained specimens were mounted onto glass slides and underwent qualitative morphometric analysis. To assess the local inflammatory response, the same specimens underwent staining for hematoxylin and eosin. Tissue samples from the spleen, gonads, liver, brain, lymph nodes, and stomach underwent β-galactosidase staining as well to detect systemic dissemination of the virus. Once a successful gene was established in the uninjured controls, the rats with SGS were transduced as well. To induce SGS, 20 rats underwent a ventral laryngofissure to gain access to the posterior tracheal wall. A 1- × 3-mm area of mucosa and perichondrium was removed from the posterior cricoid wall under an operating microscope with magnification of ×10. To induce a severe grade of SGS, 100 µl of 10N NaOH was applied to the
wound area. The laryngofissure site was then closed, and the animals were reanesthetized 2 days later. The Hamilton syringe was used to infiltrate the entire posterior cricoid submucosa with 1 × 10 8 pfu of recombinant human serotype 5 adenovirus carrying the β-galactosidase reporter gene (LacZ) in HEPES. Three animals were treated with adenovirus and 2 control animals were injected with HEPES solution at each time point. The larynges and organs were harvested at 1, 2, 4, and 7 days after transduction and underwent identical histologic and immunochemical analyses as the uninjured specimens. RESULTS In the normal uninjured rat larynx, β-galactosidase was expressed in the early time periods after injection. β-Galactosidase staining was detected in the perichondrium, epithelium, chondrocytes, and fibroblasts 1 and 2 days after transduction (Figs 1 and 2). Its expression became faint at 4 days in most specimens, and it could not be detected at 7
Otolaryngology– Head and Neck Surgery Volume 126 Number 3
JACOBS et al
285
➔ Fig 4. Cross section of the injured rat larynx 2 days after viral injection and 4 days after injury to the posterior cricoid wall, showing the most intense staining (arrow) of β-galactosidase in the submucosal space. (Original magnification ×20.)
days (Fig 3). The 14-day specimen was negative as well. In the first 2 days after transduction, the staining intensity for β-galactosidase appeared greater in the SGS specimens than in the uninjured control specimens. In the injured specimens, the intensity of β-galactosidase stain peaked at 2 days after injection (Fig 4). At 4 days, the epithelium stained positive but is less intense. At 7 days, staining was negative (Fig 5). On the hematoxylin and eosin stains, there was a moderate inflammatory response in the larynges of the control specimens injected with HEPES solution. Mucous gland hyperplasia and an inflammatory cell infiltration were observed in the posterior and lateral walls of the cricoid (Fig 6A). The inflammatory cell infiltrate appeared to be enhanced at the site of adenoviral injection at 2 days after injection (Fig 6B). However, at 7 and 14 days, the transduced specimens assumed a normal histologic appearance without evidence of SGS (Fig 7A), in stark contrast to the injured specimens
(Fig 7B). In addition, the distant sites (spleen, liver, lymph nodes, brain, gonads, and stomach) stained negative in all of the transduced animals. DISCUSSION This is the first reported demonstration of successful gene transfer to the larynx using an adenovirus vector. Direct injection lead to transduction of the submucosal connective tissue, where collagen and other extracellular matrix proteins are deposited in the evolution of SGS. The duration of transgene expression was transient as β-galactosidase activity peaked at 2 days and essentially disappeared by 7 days. Although this transient period of transgene expression may be a problem for long-term genetic diseases, it may be ideal for wound healing. The critical stages of wound healing and extracellular matrix expression occur in the first several days after injury. We also found an intense inflammatory response to the injection of the HEPES solution at
286
JACOBS et al
Otolaryngology– Head and Neck Surgery March 2002
Fig 5. Cross section of the injured rat larynx 7 days after injection and 9 days after injury, showing significant subglottic stenosis but no β-galactosidase staining. (Original magnification ×20.)
the posterior cricoid and what appeared to be an augmented immune response to the virus itself. However, there did not seem to be any long-lasting effects as the transduced specimens resumed a nearly normal histologic appearance by 7 days after injection. There also was no distal infection or toxicity from the adenovirus. Although a single administration of adenovirus appeared to be well tolerated, repeated injections might create a greater problem with the immune response.1 The inflammatory response after multiple injections will have to be addressed in future studies. Alternatively, nonviral approaches may also be considered. The adenovirus is a double-stranded DNA virus that transfects both dividing and nondividing cells with high efficiency. A wide variety of human cells are transduced by the adenovirus, including epithelial cells, endothelial cells, and keratinocytes.2 The virus remains episomal and is not fully integrated into the genome. Therefore, transient periods of expression are expected (days to weeks).1 In addition, high titers of 1014 can be isolated and puri-
fied.1 The vector may be ideal for modulating airway wound-healing response and has been used extensively in cystic fibrosis research.3-5 A first-generation recombinant adenovirus was used in this investigation, but newer generations of adenoviral vectors may be less immunogenic and have a greater transduction efficiency.12 Specific genomic deletions may increase the cloning capacity with less toxicity. In addition, immune modulations, such as immunosuppressive agents, may block the inflammatory response.13,14 This may prolong the expression of the important cytokines. It also is not clear why the β-galactosidase gene expression appeared to be enhanced in the injured rat laryngeal tissue. Possible explanations include CAR receptor up-regulation, reduced cell-mediated immune response to adenoviral infection, and delayed clearance by inflammatory cell infiltrate. These explanations are purely speculative and need confirmation in future studies. This study sets the background for future approaches to modulate the wound-healing response in the airway. The use of an adenoviral
Otolaryngology– Head and Neck Surgery Volume 126 Number 3
JACOBS et al
287
A
B
➔
C Fig 6. A, Cross section of the cricoid posterior wall injection site 4 days after administration of HEPES solution (Control). The β-galactosidase stain reveals evidence of mucous gland hyperplasia (large arrow) with a mild inflammatory cell infiltration (small arrow). (Original magnification ×50.) B, Cross section of the posterior cricoid wall at the site of adenovirus injection at 4 days later, revealing an intense inflammatory cell infiltration (arrow). (Original magnification ×50.) C, Higher magnification view of Fig 6B revealing an intense inflammatory cell infiltration. (Original magnification x250.)
vector to deliver therapeutic proteins for the treatment of impaired wound healing may represent an effective application of current gene transfer tech-
niques. Future specific applications of the adenovirus in a laryngeal model include gene transfer of transgene-expressing inhibitors of TGF-β1
288
Otolaryngology– Head and Neck Surgery March 2002
JACOBS et al
A
B Fig 7. A, Cross section of the uninjured rat larynx 14 days after Ad.CMV.LacZ transduction, revealing an almost normal histologic appearance with hemotoxylin and eosin stain except for a small area of submucosal fibrosis at the site of the adenovirus injection (arrow). (Original magnification ×10.) B, Cross section of the rat larynx at 21 days after AdCMV.LacZ transduction and 23 days after injury, revealing severe subglottic stenosis. (Original magnification ×10.)
activity such as neutralizing antibodies (anti–TGFβ1) or TGF-β3. These agents have been shown to inhibit fibroplasia and cutaneous scar formation in an animal model.7 Their transient expression in the submucosa may lead to a decrease in collagen and scar tissue in the larynx after direct injection. CONCLUSIONS An effective in vivo model was established to study modulation of airway wound healing using an adenovirus-mediated gene transfer. Successful
adenovirus-mediated tranduction of both injured and uninjured rat larynges was achieved with a first-generation human recombinant adenoviral vector. Injury enhanced β-galactosidase transduction. β-Galactosidase transgene expression was detectable 1 day after injection and was transient as its expression disappeared by 7 days. The background inflammatory response was transient, and there was no systemic dissemination of the virus. Thus, it appears that the human recombinant adenovirus is an effective vector for gene transfer
Otolaryngology– Head and Neck Surgery Volume 126 Number 3
in the rat larynx and may facilitate novel approaches to airway wound healing. Future strategies might include use of multiple injections or advanced-generation adenoviral vectors or nonviral vectors, which may lead to a reduced inflammatory reaction and greater persistence. TGF-β1 inhibitors, transduced using the adenoviral vector, may decrease scar tissue formation in the submucosa of the larynx.
JACOBS et al
6. 7.
8.
REFERENCES
1. Jain KK. Vectors for gene therapy. In: Textbook of gene therapy, ed 1. Seattle, WA: Hogrefe and Huber Publishers; 1998. Chapter 4. 2. Setoguchi Y, Jaffe HA, Danel C, et al. Ex vivo and in vivo gene transfer to skin using replication-deficient recombinant vectors. Soc Invest Dermatol 1994;102:415-21. 3. Goldman MJ, Yang Y, Wilson JM. Gene therapy in a zenograft model of cystic fibrosis lung corrects chloride transport more effectively than sodium defect. Nat Genet 1995;9:126-31. 4. Dupuit F, Zahm J-M, Pierrot D, et al. Regenerating cells in human airway surface epithelium represent preferential targets for recombinant adenovirus. Hum Gene Ther 1995;6;1185-93. 5. Jiang C, Akita GY, Collegel WH, et al. Increased contact
9. 10. 11. 12. 13.
14.
289
time improves adenovirus-mediated CFTR gene transfer to nasal epithelium of CF mice. Hum Gene Ther 1997;8:671-80. Parkin JP, Stevens MH, Jung AL. Acquired and congenital subglottic stenosis in the infant. Ann Otol Rhinol Laryngol 1976;85:573-81. Shah M, Foreman DM, Ferguson MWJ. Neutralisation of TGF-β1 and TGF-β2 or exogenous addition of TGF-β3 to cutaneous rat wounds reduces scarring. J Cell Sci 1995;108:985-1002. Border WA, Noble NA, Yamamoto T, et al. Natural inhibitor of transforming growth factor-B protects against scarring in experimental kidney disease. Nature 1992;9360:361-4. Hom DB. Growth factors and wound-healing in otolaryngology. Otolaryngol Head Neck Surg 1994;110:560-4. Trapnell BC. 1993. Graham FL. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 1973;52:456-67. Wang Q, Finer MH. Second-generation adenovirus vectors. Nat Med 1996;2:714-6. Kaplan JM, Smith AE. Transient immunosuppression with deoxyspergualin improves longevity of transgene expression and ability to re-administer adenoviral vector to the mouse lung. Hum Gene Ther 1997;8:1095-104. Worgall S, Leoplad PL, Wolff G, et al. Role of alveolar macrophages in rapid elimination of adenovirus vectors administered to the epithelial surface of the respiratory tract. Hum Gene Ther 1997;8:1675-84.
290
JACOBS et al
Otolaryngology– Head and Neck Surgery March 2002