In vivo adenoviral transduction of the neonatal rat cochlea andmiddle ear

Stefan Dazert a, Christoph Aletsee a;b, Dominik Brors a, Claude Gravel d,Michael Sendtner c, Allen Ryan b;*

a Department of Otolaryngology-Head and Neck Surgery, Julius-Maximilians-Universita«t, 97080 Wu«rzburg, Germanyb Department of Surgery/Otolaryngology and Department of Neurosciences, UCSD School of Medicine and Veterans A¡airs Medical Center,

9500 Gilman Drive, La Jolla, CA 92093-0666, USAc Department of Neurology, Julius-Maximilians-Universita«t, 97080 Wu«rzburg, Germany

d Laboratoire de transfert de genes, Centre de recherche Universite Laval Robert-Gi¡ard, Beauport, Que., Canada

Received 21 April 2000; accepted 9 August 2000

Abstract

Virally mediated gene transfer to the adult mammalian ear appears to be a powerful strategy to investigate gene function in theauditory system and to develop new therapeutic treatment for hearing impaired patients. However, there has been little work done inthe neonatal middle and inner ear. In this study, a recombinant adenoviral (AdV) vector was used for gene transfer of a L-galactosidase (L-gal) reporter gene to the neonatal middle ear and cochlea of 5 day old rats. For transduction of middle ear, AdV wasinjected through the tympanic membrane into the tympanic cavity. Three and 7 days later, strong expression of L-gal was observed inepithelial cells of the mucosa, but not in the underlying stroma or mesenchyme. There was little or no infiltration of leukocytes. Noexpression of L-gal was detected inside the cochlea or vestibular system. When AdV was injected into the basal turn of the cochlea,high levels of L-gal expression were observed in cells lining the perilymphatic space and in parts of the spiral ligament 3, 7 and 21days later. Spiral ganglion cells did not express L-gal. However, virally mediated gene transfer was observed in some cells of theorgan of Corti. A moderate infiltration of leukocytes into the labyrinth was observed, but no vestibular or auditory dysfunction.These results demonstrate that neonatal middle ear and cochlear cells can be successfully transduced with an AdV vector in vivo,without obvious morphological signs of inflammation or cellular damage. AdV vectors provide a tool for investigation of the role ofgenes in influencing the development of middle and inner ear structures. Virally mediated expression of protective genes could alsobe used to rescue hair cells or spiral ganglion cells from congenital degeneration or damage. ß 2001 Elsevier Science B.V. All rightsreserved.

Key words: Adenovirus vector; Gene transfer; Cochlea; Middle ear; Neonatal rat

1. Introduction

During the last 10 years a variety of gene defectshave been identi¢ed that are responsible for deafnessor genetic predisposition for postnatal loss of hearing.Potential treatment of congenital disorders may there-fore require the delivery of genetic information to cellsof the middle and inner ear. The introduction of foreign

genes into target cells has been performed by severalmethods such as transfection of cells with expressionplasmids, implantation of stably transfected cell lines,and the production of transgenic animals with appro-priate promotors. Among the most promising methodsfor gene therapy are viral vectors, which have beenwidely used to introduce new genetic information intoa variety of cell types.

The application of adenovirus (AdV) vectors is wellestablished for experimental and clinical purposes, sinceAdVs can infect a variety of cell types and do not re-quire cell division for transduction. In addition, theAdV system introduces a stable genome into target cells

0378-5955 / 01 / $ ^ see front matter ß 2001 Elsevier Science B.V. All rights reserved.PII: S 0 3 7 8 - 5 9 5 5 ( 0 0 ) 0 0 1 8 9 - 1

* Corresponding author. Tel. : +1 (858) 534 4594;Fax: +1 (858) 534 5319; E-mail: afryan@ucsd.edu

HEARES 3557 12-12-00

Hearing Research 151 (2001) 30^40

www.elsevier.com/locate/heares

and can induce relatively long-term expression of anintroduced gene at high levels (Graham and Prevec,1991). Successful gene transfer has been achieved withAdV vectors into hepatocellular carcinoma cells (Kane-ko et al., 1995), prostate cancer cell lines (Eastham etal., 1995), mammary epithelial cells (Yang et al., 1995),cells of blood vessels (Rios et al., 1995), hematopoieticcells (Tani et al., 1995), and skeletal muscle (Haase etal., 1997). AdV-mediated gene transfer has also beenused to infect post-mitotic cells of the central nervoussystem (Draghia et al., 1995).

AdV gene transfer of foreign DNA sequences intotissues of the adult mammalian middle ear (Mondainet al., 1998) and inner ear (e.g. Raphael et al., 1996;Yamasoba et al., 1999) is increasingly used for scienti¢cinvestigations with animal models and may eventuallylead to the development of new methods of treatmentfor dysfunction of hearing processes. Besides AdV vec-tors, retroviruses (Kiernan and Fekete, 1997; Zheng etal., 1998; Homburger and Fekete, 1996), herpes simplexviruses (Garrido et al., 1998; Staecker et al., 1998; Der-by et al., 1999), adeno-associated viruses (Lalwani etal., 1996) and liposomes (Wareing et al., 1999) havebeen used for gene transfer into inner ear tissues. In-vestigators have also used viral vectors to transducecells in the early fetal inner ear to study the early devel-opment of the labyrinth (Homburger and Fekete, 1996).However, later development is also an important periodfor scienti¢c investigation and for potential gene ther-apy, and relatively little work has been performed ontransducing genetic information into cells of the neo-natal mammalian cochlea. In an earlier in vitro inves-tigation with cultured cells, we demonstrated that neo-natal rat cochlear cells can be transfected successfullywith an AdV vector, at viral titers that do not lead tocellular damage or dysfunction. Various cell types ofthe neonatal rat organ of Corti and the spiral ganglionshowed strong expression of the introduced L-galacto-sidase (L-gal) reporter gene (Dazert et al., 1997). Morerecently, Holt et al. (1999) used AdV vectors to success-fully transduce neonatal mouse inner ear cells in vitro.To our knowledge, AdV transduction has not been ap-plied so far for gene transfer to the mammalian neo-natal middle and inner ear in vivo.

The present study was designed to investigate thee¤cacy, practicability and the potential side e¡ects ofin vivo transduction of the rat middle and inner earwith recombinant AdVs. Several neurotrophic factorshave been shown to protect and/or in£uence the growthpattern of rat cochlear cells during their neonatal ter-minal development (Low et al., 1996; Staecker et al.,1997; Dazert et al., 1998). Therefore, we conclude thatAdV gene transfer could be a promising strategy for theapplication of protective and/or neurotrophic genesunder pathophysiological conditions, such as congenital

deafness, when cell death of neurons and other cochlearcells occurs during the perinatal period.

2. Materials and methods

2.1. Adenoviral vector

In the AdV vector employed in these experiments(AdV-CMV-L-gal), expression of the L-gal reportergene is driven by the immediate/early cytomegalovirus(CMV) gene promotor. The expression cassette replacesthe E1 region of the adenoviral genome between nu-cleotides 452 and 3328, rendering the virus replicationincompetent. Construction, production, puri¢cationand titration of the vector has been described elsewhere(Vilquin et al., 1995).

2.2. Application of the virus

Animal investigation procedures were reviewed andapproved by the animal use committee of the Univer-sity of Wu«rzburg (Germany), in adherence to the guide-lines of the Declaration of Helsinki. The study wascarried out in postnatal day 5 Sprague^Dawley rats(day of birth being considered day 0). Prior to the sur-gical procedures, animals were anesthetized by halo-thane inhalation as described earlier (Dazert et al.,2000). Two sites of virus application were tested. First,the tympanic membrane was exposed using a postaur-icular incision and the virus suspension (2 Wl, 0.5U107

pfu/ml) was injected through the tympanic membraneinto the middle ear space over a period of 15 s (n = 8).This was done to explore middle ear transduction andto determine whether inner ear transduction would oc-cur via the round window membrane, or by penetrationof AdV through the cartilaginous cochlear capsule. In asecond approach, 2 Wl of virus suspension were trans-ferred directly into the cochlea, after opening of theeardrum and exposure of the otic capsule (n = 13).The viral suspension was introduced into the basalturn of the cochlea by perforating the capsule withthe tip of a sterile glass micropipette and injectingover a period of 15 s. Control animals were injectedwith saline (n = 2 for middle ear, n = 5 for inner ear)or were untreated (n = 2). Injection pressures were notmeasured. All surgical procedures were performedunder the operating microscope using sterile instru-ments.

2.3. Tissue preparation and X-gal staining

Either 3 or 7 days after application of the vector,animals were anesthetized by halothane inhalation,and perfused transcardially with warm saline followed

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^40 31

by 4% paraformaldehyde. After removal of the skin, theskull was dissected to obtain a portion containing bothtemporal bones and the adjacent brain tissue and thenleft in 4% paraformaldehyde overnight. The sampleswere embedded in Tissue-Tec and 20 Wm frozen serialsections were prepared. The sections were mounted ongelatin coated slides and processed for X-gal staining.

Brie£y, the frozen sections were incubated overnightat 37³C in fresh X-gal solution that was ¢ltered prior touse and made in phosphate-bu¡ered saline (PBS) with1 mg/ml X-gal, 5 mM K-ferricyanide, 5 mM K-ferro-cyanide, and 2 mM MgCl2. After incubation, the sec-tions were rinsed in PBS, counterstained with 0.1% neu-

tral red and mounted in Canada balsam. The sectionswere then evaluated by light microscopy, comparing thetransfected side with the contralateral, untransfectedside and with control animals injected with saline.

To assess transduction after a longer period of sur-vival ¢ve additional animals were injected with virus inthe cochlea as above. After 21 days, the animals wereperfused and the cochleas were ¢xed overnight asabove. The cochlear capsules were then widely openedand incubated en bloc in fresh X-gal solution overnightat 37³C. They were then thoroughly rinsed in PBS andevaluated by stereomicroscopy. One animal was in-jected with saline as a control.

Fig. 1. Nonspeci¢c X-gal reaction product in the organ of Corti of an untreated control animal (A), and in the organ of Corti of the contralat-eral, uninjected ear of an animal injected with AdV in the opposite cochlea (B). The same pattern of X-gal staining was observed in both cases.Light staining was seen near the inner hair cells (arrowheads). Cells in bone marrow (BM) and perivascular cells (arrows) were also nonspeci¢-cally labeled.

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^4032

Quantitative evaluation of transduction, either by in-tensity of reaction product or by the number of cellstransduced, was not performed. The variability ob-served between subjects within a survival interval, pre-sumably related to the e¤ciency of viral application,appeared to be greater than the variability observedacross survival intervals. This would tend to obscureany trend in transduction over time.

3. Results

After completion of the surgical procedure and re-covery from anesthesia, the treated rat pups were re-turned to the mother with their untreated littermates.The transduced animals showed normal behavior with-out any signs of vertigo or indisposition. The animalsallowed to survive for 3 weeks showed a Preyer re£exindistinguishable from those of their untreated litter-mates. The histological sections of the temporal boneswith surrounding brain tissue were analyzed by lightmicroscopy. Control animals that were injected withsaline or were not treated showed nonspeci¢c reactionproduct in bone marrow and perivascular cells, and

faint labeling of cells in the organ of Corti near theinner hair cells (Fig. 1).

3.1. Virus application into the tympanic cavity

Three days after application of AdV through thetympanic membrane into the neonatal middle ear, theepithelium of the mucosa covering the tympanic mem-brane and subepithelial stroma inside the developingtympanic cavity showed strong expression of L-gal(Figs. 2 and 3). Staining was sporadic, with some epi-thelial cells transduced and others apparently unaf-fected. There was little or no transduction of cells be-neath the mucosal epithelium, except in the region ofthe surgical incision where tissues external to the middleear were sometimes transduced. Reporter gene expres-sion could be detected close to the promontory, but nostaining was seen inside the cochlea or the vestibularinner ear. No leukocytes were observed in the middleear of most subjects. Seven days after AdV application,a similar transduction pattern of middle ear mucosawas observed (Fig. 4). In the middle ear contralateralto the injection site, no staining was observed in anyanimals.

Fig. 2. Section of neonatal rat middle ear stained for L-gal 3 days after application of AdV into the tympanic cavity. Scattered cells of the mu-cosal epithelium (arrowheads indicate examples) covering remnants of mesenchymal tissue show strong expression of L-gal. There was little orno transduction of cells beneath the mucosal epithelium. TC = tympanic cavity.

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^40 33

3.2. Virus application into the cochlea

AdV transfer into the basal turn of the cochlea re-sulted in high levels of L-gal expression in cochlear cells3 and 7 days later. Staining was observed primarily incells lining the perilymphatic spaces including the peri-lymphatic cells of Reissner's membrane (Figs. 5 and 6).Staining was observed in cells lining the scala tympani,scala vestibuli or both, presumably depending upon theinjection site. No staining was observed in cells liningthe scala media. Strong transduction of cells of thespiral ligament was observed in some areas (Fig. 6).The organ of Corti was typically not transduced, how-ever, in some areas sporadic but intense transductionwas observed in cells of the organ (Fig. 7). The reactionproduct in these cells far exceeded the nonspeci¢c label-ing seen in the organ of Corti of control inner ears. Atthe level of resolution provided by para¤n sections, theidentity of the cells could not be determined, althoughthey were observed primarily in the area of the outerhair cells and Deiter's cells. In the spiral ganglion, noexpression of L-gal by neurons or Schwann cells wasseen in any of the animals. A moderate in¢ltration of

leukocytes into the labyrinth was observed in most ani-mals.

Adenoviral gene transfer to the inner ear using viraltiters of 0.5U107 pfu/ml produced very few negativee¡ects on cochlear tissues after 3 or 7 days that wereapparent histologically. The sensory epithelium as wellas spiral ganglion cells and their processes forming theauditory nerve presented with normal morphologyand orientation, and no evidence of hair cell or neuro-nal loss. Some vacuoles were occasionally notedin cells of the spiral ligament (Fig. 6). Thus in-fection with AdV vector at the doses employed hadno major e¡ect on the histological pattern of inner earcells.

Twenty-one days after cochlear injection, animalsshowed strong or moderate X-gal staining in cells liningthe perilymphatic scalae, and in the spiral ligament(Fig. 8). The degree of staining observed in these enbloc preparations could not be directly compared tothat observed in tissue sections at 3 and 7 days. How-ever, the staining appeared to be approximately as in-tense, and the distribution of transduced cells was sim-ilar. The animals exhibited no vestibular abnormalities

Fig. 3. Section of neonatal rat middle ear stained for L-gal 72 h after application of AdV into the tympanic cavity (TC). Epithelial cells of themucosa (arrowheads indicate examples) on the inside of the tympanic membrane (TM) are strongly transduced. Staining was sporadic, withsome cells transduced and others apparently una¡ected. OEC = outer ear canal.

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^4034

and normal Preyer re£exes. No sign of in£ammationwas visible by stereomicroscopy.

The inner ears contralateral to the injected inner earof all animals showed X-gal staining in bone marrowand perivascular cells, and faint staining of cells nearthe inner hair cells, comparable to that seen in un-treated control animals and animals injected only withsaline. This labeling was consistent with non-speci¢cbackground staining (Fig. 1).

4. Discussion

The results of the present investigation indicate thatsubsets of cells in the neonatal rat middle ear mucosaand cochlea can be successfully transduced in vivo us-ing an AdV vector. The treated animals showed normalvestibular and overall behavior, normal Preyer re£exes,and well-preserved morphology of the middle ear andcochlea including the organ of Corti and the spiralganglion, compared to their untreated littermates.Therefore neonatal transduction of cochlear cells didnot interfere with the development and structure ofthe middle and inner ears, at least over a 3 week period.The in£ammatory response induced by the vector in the

inner and to a lesser extent the middle ear did notappear to cause major damage to the tissues of thesestructures. Minor damage was observed, but this didnot appear to increase with longer survival times, upto 3 weeks. Of course, it is possible that even longersurvival times might have resulted in increased damage.It is also possible that introduction of virus into theinner ear of neonates is less damaging than in adults.The immune response to viruses is a major cause ofdamage to the inner ear in viral labyrinthitis (Darm-stadt et al., 1990). This would presumably be evenmore true of replication-incompetent viral vectors.However, exposure to antigens during the early postna-tal period is well known to induce tolerance (e.g. Phil-lips-Quagliata, 1972). If neonatal exposure to AdV vec-tors results in tolerance to viral antigens, they couldresult in less damage than in adults. It should also benoted that new generations of AdV vectors that mayproduce less in£ammatory response are becoming avail-able.

Three and 7 days after application of the AdV vector,the middle ear mucosa showed transduction limited tothe epithelial layer. The virus thus did not appear topenetrate the mucosal surface, and presumably was ex-cluded from the subepithelial compartment by the in-

Fig. 4. Section of neonatal rat middle ear similar to Fig. 2, 1 week after application of the AdV into the tympanic cavity (TC). TM = tympanicmembrane, OEC = outer ear canal.

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^40 35

tercellular junctions of the epithelium. This is consistentwith the lack of penetration of virus through the roundwindow and cochlear capsule, into the inner ear.

After injection of AdV into the cochlea, the cells lin-ing the scala tympani and scala vestibuli, including theperilymphatic cells of Reissner's membrane, producedhigh levels of L-gal expression, as did cells in the spiralligament. Few organ of Corti cells produced L-gal ex-pression, and spiral ganglion cells were not found to betransduced. The virus therefore seems to have distrib-uted readily via the perilymph, transducing primarilycells that have ready access to this £uid. Transductionlimited primarily to the perilymphatic space is consis-tent with results described in the adult guinea pig innerear after perilymphatic AdV injection (Raphael et al.,1996). Since little or no L-gal expression was detectedinside the organ of Corti, stria vascularis and spiralganglion, the virus either did not easily migrate intothe cochlear duct and Rosenthal's canal, or perhapswas unable to transduce cells at these locations. Yama-soba et al. (1999) injected AdV into the endolymphaticsac resulting in gene expression in cells of the endolym-

phatic compartment including cells of the organ of Cor-ti. In addition, we previously found that cells of spiralganglion and organ of Corti explants could be trans-duced in vitro (Dazert et al., 1997). These data implythat the lack of transduction in spiral ganglion andorgan of Corti seen in the present study was due topoor access of the virus to these structures. It is ofcourse possible that higher AdV titers would have en-hanced transduction of these tissues. Transduction ofcochlear cells persisted at high levels for at least 3weeks, the longest time examined.

No transduction of the middle or inner ear contralat-eral to the injection site was observed in any animals inthe present study. Viral vectors injected into the innerear of adult guinea pigs have been shown to spread tothe contralateral inner ear, primarily via the cerebrospi-nal £uid (Lalwani et al., 1996; Sto«ver et al., 2000). Thelack of contralateral transduction observed in thepresent study may relate to the volume of inoculumused, since Sto«ver et al. (2000) observed contralateraltransduction with a 25 Wl injection, but not with 5 Wl.Our volume of 2 Wl may have been below the threshold

Fig. 5. Section of neonatal rat cochlea stained for L-gal, 3 days after AdV transfer into the basal turn of the cochlea resulted. High levels of L-gal expression are observed in cells (arrowheads indicate examples) lining the scala tympani (ST) and the spiral ligament (SL).

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^4036

for contralateral transduction, although the inner ear ofthe rat is smaller than that of the guinea pig. The im-maturity of the structures involved, or di¡erences be-tween the two species, might also be involved. For ex-ample, the cochlear aqueduct of the guinea pig is largerthan that of the rat, allowing greater communicationbetween perilymph and cerebrospinal £uid.

Gene transfer to the neonatal middle and inner earusing AdV is a potentially useful tool for research into

factors controlling terminal development. It should bepossible to induce expression of genes potentially a¡ect-ing the maturation of the mucosal epithelium of themiddle ear. Induced expression of secreted proteinsmight also a¡ect underlying structures within the mes-enchyme or mucosal stroma. Similarly, maturation ofcells lining the perilymphatic compartment could bemanipulated directly, while that of other structuresmight be a¡ected by secretion of substances that act

Fig. 6. Section of neonatal rat inner ear, 3 days after injection of AdV into the perilymphatic compartment. L-gal expression is observed in cellsof Reissner's membrane (arrowheads indicate examples). Transduction of the upper portion of the spiral ligament accompanied by vessiculationis also apparent (arrow). SV = scala vestibuli, StV = stria vascularis, oC = organ of Corti.

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^40 37

at a distance. For example, neurotrophins are involvedin the support and survival of spiral ganglion neuronsas well as neurite targeting during the early postnatalperiod in rodents. Induction of ectopic neurotrophinexpression during this period could in£uence the courseof ganglion neuron development.

In addition, AdV vectors may have clinical applica-tions in the developing middle and inner ears. Otherinvestigators have demonstrated that viral vectors canprotect the hair cells and neurons of the adult cochleafrom a variety of insults, including aminoglycoside tox-icity (Staecker et al., 1997). Treatment during develop-ment could potentially be even more e¡ective. In manyregions of the nervous system, neurons appear morevulnerable to various types of injury during the perina-tal period compared to later stages of postnatal devel-opment. For example, axotomy of spinal or facialmotoneurons in newborn rodents leads to massive celldeath of the corresponding cell bodies resulting in irre-versible functional de¢cits (Sendtner et al., 1990). Axo-nal lesions in adult animals do not result in cell deathbut allow successful functional regeneration. A similarmanner, damage to hair cells, spiral ganglion cells or

epithelial cells of the middle ear mucosa may have moreserious consequences during neonatal development.Therapeutic strategies with the potential to protect cellsduring this critical developmental period are thereforeof clinical interest. Of course, clinical application to theinner ear would require a less traumatic method of in-troducing the vector, such as injection through theround window.

A potential strength of AdV vectors for the treatmentof developmental disorders is the temporary nature ofgene expression induced by this method. AdV vectorsare e¡ective tools for transient expression of foreigngenes in various types of cells. The expression of report-er genes or neurotrophic factors under the control ofthe CMV promoter has been shown to be very highduring the ¢rst weeks after in vivo application ofAdV. However, at longer times expression of thesetransduced genes decreases (Gravel et al., 1997). Thiscould be a disadvantage for treatment of disorders thatrequire long-term expression of genes. However, whentransient expression of genes is desired, this methodwould appear to be appropriate. For example, to pro-tect spiral ganglion cells from transient injury such as

Fig. 7. Section of neonatal rat inner ear 3 days after injection of AdV into the perilymphatic compartment. Many perilymphatic lining cellsand possibly in£ammatory cells beneath the organ of Corti (oC) are transduced (arrow). While the organ of Corti (oC) was not transduced inmost animals, in this case strong transduction was observed in scattered cells, especially in the region of the outer hair cells (arrowheads indi-cate examples). It could not be determined whether these were outer hair cells or Deiter's cells. ST = Scala tympani, SM = Scala media.

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^4038

during placement of a cochlear implant electrode thistechnique might be used to provide expression of pro-tective factors such as neurotrophic molecules especiallyduring the ¢rst 2 weeks after implantation. Develop-mental disorders are another category that could bene¢tfrom temporary gene expression.

The frequency of childhood deafness is estimated at1/500 with about 50% of genetic origin. The remainderof congenital hearing impairment is related to transientenvironmental in£uences, including infection and trau-ma. In many instances only brief gene expression mightbe required to correct such disorders, such as protectionof hair cells and/or neurons during a period of congen-ital infection. Similarly, many developmental genes areexpressed for a relatively short period of time, and cor-rection of defects involving such genes would presum-ably be best accomplished during this brief period.

Viral gene transfer might be best suited to cases ofrecessive inherited deafness, since these conditions ofteninvolve null mutations. Many dominant mutations indeafness produce abnormal proteins with mutant func-tion, and act as dominant negative mutations. In suchcases, additional normal protein would not necessarilybe bene¢cial. Mutations that involve loss of function orhaploinsu¤ciency would be more likely to bene¢t fromreplacement of the normal protein by gene therapy.Under these conditions, gene replacement therapy using

viral vectors may provide new strategies for the treat-ment of hereditary disorders in children, during fetaland neonatal development.

Acknowledgements

Supported by the Bayerische Sonderforschungsfo«r-derung, the Deutsche Forschungsgemeinschaft, grants# DC00129 and DC00139 from the NIH/NIDCD,and by the Research Service of the VA. The authorswish to thank Mrs. Petra Joa for helpful support inhistological preparation and evaluation.

References

Darmstadt, G.L., Keithley, E.M., Harris, J.P., 1990. E¡ects of cyclo-phosphamide on the pathogenesis of cytomegalovirus-induced lab-yrinthitis. Ann. Otol. Rhinol. Laryngol. 99, 960^968.

Dazert, S., Battaglia, A., Ryan, A.F., 1997. Transfection of neonatalrat cochlear cells in vitro with an adenovirus vector. Int. J. Dev.Neurosci. 15, 595^600.

Dazert, S., Kim, D., Luo, L., Aletsee, C., Garfunkel, S., Maciag, T.,Baird, A., Ryan, A.F., 1998. Focal delivery of ¢broblast growthfactor-1 by transfected cells induces spiral ganglion neurite target-ing in vitro. J. Cell. Physiol. 177, 123^129.

Dazert, S., Scho«mig, P., Shehata-Dieler, W.E., Aletsee, C., Dieler, R.,

Fig. 8. Moderate transduction of the inner ear 3 weeks after injection of AdV into the neonatal cochlea. In this view, transduced cells can beseen in Reissner's membrane (RM) and in the lining of the scala vestibuli (SV) of the basal turn. In the middle turn, transduction of cells inthe apical portion of the spiral ligament (SL) is apparent.

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^40 39

2000. General anesthesia in new born rodents for experimentalmicrosurgery of the ear. Laryngo-Rhino-Otol. 79, 26^29.

Derby, M.L., Sena-Esteves, M., Breake¢eld, X.O., Corey, D.P., 1999.Gene transfer into the mammalian inner ear using HSV-1 andvaccinia virus vectors. Hear. Res. 134, 1^8.

Draghia, R., Caillaud, C., Manicom, R., Pavirani, A., Kahn, A.,Poenaru, L., 1995. Gene delivery into the central nervous systemby nasal instillation in rats. Gene Ther. 2, 418^423.

Eastham, J.A., Hall, S.J., Sehgal, I., Wang, J., Timme, T.L., Yang,G., Connel-Crowley, L., Elledge, S.J., Zhang, W.W., Harper,J.W., 1995. In vivo gene therapy with p53 or p21 adenovirus forprostate cancer. Cancer Res. 55, 5151^5155.

Garrido, J.J., Alonso, M.T., Lim, F., Carnicero, E., Giraldez, F.,Schimmang, T., 1998. De¢ning responsiveness of avian cochlearneurons to brain-derived neurotrophic factor and nerve growthfactor by HSV-1-mediated gene transfer. J. Neurochem. 70,2336^2346.

Graham, F.L., Prevec, L., 1991. Manipulation of adenovirus vectors.In: Murray, E.J. (Ed.), Methods of Molecular Biology. HumanaPress, Clifton, NJ, pp. 109^132.

Gravel, C., Go«tz, R., Lorrain, A., Sendtner, M., 1997. Adenoviralgene trasfer of ciliary neurotrophic factor and brain-derived neu-rotrophic factor leads to long-term survival of axotomized motorneurons. Nat. Med. 3, 765^770.

Haase, G., Kennel, P., Pettmann, B., Vigne, E., Akli, S., Revah, F.,Schmalbruch, H., Kahn, A., 1997. Gene therapy of murine motorneuron disease using adenoviral vectors for neurotrophic factors(see comments). Nature Medicine 3(4), 429^436.

Holt, J.R., Johns, D.C., Wang, S., Chen, Z.Y., Dunn, R.J., Marban,E., Corey, D.P., 1999. Functional expression of exogenous pro-teins in mammalian sensory hair cells infected with adenoviralvectors. J. Neurophysiol. 81, 1881^1888.

Homburger, S.A., Fekete, D.M., 1996. high e¤ciency gene transferinto the embryonic chicken CNS using B-subgroup retroviruses.Dev. Dyn. 206, 112^120.

Kaneko, S., Hallenbeck, P., Kotani, T., Nakabayashi, H., McGarrity,G., Tamaoki, T., Anderson, W.F., Chiang, Y.L., 1995. Adenovi-rus-mediated gene therapy of hepatocellular carcinoma using can-cer-speci¢c gene expression. Cancer Res. 55, 5283^5287.

Kiernan, A.E., Fekete, D.M., 1997. In vivo gene transfer into theembryonic inner ear using retroviral vectors. Audiol. Neurootol.2, 12^24.

Lalwani, A.K., Walsh, B.J., Reilly, P.G., Muzyczka, N., Mhatre,A.N., 1996. Development of in vivo gene therapy for hearing dis-orders: introduction of adeno-associated virus into the cochlea ofthe guinea pig. Gene Ther. 3, 588^592.

Low, W., Dazert, S., Baird, A., Ryan, A.F., 1996. Basic ¢broblastgrowth factor (FGF-2) protects rat cochlear hair cells in organo-typical culture from aminoglycoside injury. J. Cell. Physiol. 167,443^450.

Mondain, M., Restituito, S., Vincenti, V., Gardiner, Q., Uziel, A.,Delabre, A., Mathieu, M., Bousquet, J., Demoly, P., 1998. Ad-

enovirus-mediated in vivo gene transfer in guinea pig middle earmucosa. Hum. Gene Ther. 9, 1217^1221.

Phillips-Quagliata, J.M., 1972. Cellular events in tolerance. I. Induc-tion and loss of tolerance in neonatal and adult rats. Cell. Immu-nol. 4, 116^133.

Raphael, Y., Frisancho, J.C., Roessler, B.J., 1996. Adenoviral-medi-ated gene transfer into guinea pig cochlear cells in vivo. Neurosci.Lett. 207, 137^141.

Rios, C.D., Ooboshi, H., Piegors, D., Davidson, B.L., Heistad, D.D.,1995. Adenovirus-mediated gene transfer to normal and athero-sclerotic arteries. A novel approach. Arterioscler. Thromb. Vasc.Biol. 15, 2241^2245.

Sendtner, M., Kreutzberg, G.W., Thoenen, H., 1990. Ciliary neuro-trophic factor prevents the degeneration of motor neurons afteraxotomy. Nature 345, 440^441.

Staecker, H., Dazert, S., Malgrange, B., Lefebvre, P.P., Ryan, A.F.,Van De Water, T.R., 1997. Transforming growth factor alphatreatment alters intracellular calcium levels in hair cells and pro-tects them from ototoxic damage in vitro. Int. J. Dev. Neurosci.15, 553^562.

Staecker, H., Gabaizadeh, R., Federo¡, H., Van De Water, T.R.,1998. Brain-derived neurotrophic factor gene therapy prevents spi-ral ganglion degeneration after hair cell loss. Otolaryngol. HeadNeck Surg. 119, 7^13.

Sto«ver, T., Yagi, M., Raphael, Y., 2000. Transduction of the contra-lateral ear after adenovirus-mediated cochlear gene transfer. GeneTher. 7, 377^383.

Tani, K., Lin, T., Hibino, H., Takahashi, K., Nakazaki, Y., Takaha-shi, S., Nagayama, H., Ozawa, K., Saitho, I., Mulligan, R., 1995.Transduction of LacZ gene into leukemia cells using viral vectorsof retrovirus and adenovirus. Leukemia 1 ((Suppl.)), 64^65.

Vilquin, J.T., Guerette, B., Kinoshita, I., Roy, B., Goulet, M., Gravel,C., Roy, R., Tremblay, J.P., 1995. FK506 immunosuppression tocontrol the immune reactions triggered by ¢rst-generation adeno-virus-mediated gene transfer. Human Gene Therapy 6(11), 1391^1401.

Wareing, M., Mhatre, A.N., Pettis, R., Han, J.J., P¢ster, M.H., Hong,K., Zheng, W.W., Lalwani, A.K., 1999. Cationic liposome medi-ated transgene expression in the guinea pig cochlea. Hear. Res.128, 61^69.

Yamasoba, T., Yagi, M., Roessler, B.J., Miller, J.M., Raphael, Y.,1999. Inner ear transgene expression after adenoviral vector inoc-ulation in the endolymphatic sac. Hum. Gen. Ther. 10, 769^774.

Yang, J., Tsukamoto, T., Popnikolov, N., Guzman, R.C., Chen, X.,Yang, J.H., Nandi, S., 1995. Adenoviral-mediated gene transferinto primary human and mouse mammary epithelial cells in vitroand in vivo. Cancer Lett. 98, 9^17.

Zheng, J.L., Lewis, A.K., Gao, W.Q., 1998. Establishment of con-ditionally immortalized rat utricular epithelial cell lines using aretrovirus-mediated gene transfer technique. Hear. Res. 117,13^23.

HEARES 3557 12-12-00

S. Dazert et al. / Hearing Research 151 (2001) 30^4040