Hearing Research 330 (2015) 106e112
Contents lists available at ScienceDirect
Hearing Research journal homepage: www.elsevier.com/locate/heares
Development of an electrode for the artificial cochlear sensory epithelium Yosuke Tona a, *, Takatoshi Inaoka a, b, Juichi Ito a, Satoyuki Kawano c, Takayuki Nakagawa a a
Department of Otolaryngology, Head and Neck Surgery, Graduate School of Medicine, Kyoto University, Kawaharacho 54, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan b Inaoka ENT Clinic, 2-1-14-201, Neyaminami, Neyagawa 572-0855, Japan c Department of Mechanical Science and Bioengineering, Graduate School of Engineering Science, Osaka University, 1-3 Machikaneyama, Toyonaka, Osaka 560-8531, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 February 2015 Received in revised form 11 August 2015 Accepted 17 August 2015 Available online 20 August 2015
An artificial cochlear sensory epithelium has been developed on the basis of a new concept that the piezoelectric membrane, which converts mechanical distortion into electricity, can mimic the function of the inner hair cell and basilar membrane of the mammalian cochlea. Our previous research demonstrated that the piezoelectric membrane generated electrical outputs in response to the sound stimulation after implantation into the guinea pig cochlea, whereas electrodes for the stimulation of spiral ganglion neurons have not been fabricated, and a method to fix the device in the cochlea is also required to show proof-of-concept. In the present study, to achieve proof-of-concept of hearing recovery by implantation of the artificial cochlear sensory epithelium, we fabricated new electrodes that stick into the cochlear modiolus, which also play a role in the fixation of the device in the cochlea. The efficacy of new electrodes for fixation of the device in the cochlea and for the stimulation of spiral ganglion neurons was estimated in guinea pigs. Four weeks after implantation, we confirmed that the devices were in place. Histological analysis of the implanted cochleae revealed inconspicuous fibrosis and scar formation compared with the sham-operated specimens (n ¼ 5 for each). The terminal deoxynucleotidyl transferase dUTP nick end labeling method was used to assess cell death due to surgical procedures in the cochleae that were harvested after 1 day (n ¼ 6) and 7 days (n ¼ 6) of implantation; there was no significant increase in apoptotic cell death in the implanted cochleae compared with sham-operated cochleae. In seven animals, serial measurements of electrically evoked auditory brainstem responses were obtained, with the electrode positioned in the scala tympani and with the electrode inserted into the cochlear modiolus. With the insertion of electrodes into the cochlear modiolus, significant reduction was achieved in the thresholds of electrically evoked auditory brainstem responses compared with those placed in the scala tympani (p ¼ 0.028). These findings indicated that the new electrodes efficiently fixed the device in the cochlea and were able to stimulate spiral ganglion neurons. This article is part of a Special Issue entitled .
Keywords: Piezoelectric membrane Mechanoelectrical stimulation Modiolus stimulation
© 2015 Elsevier B.V. All rights reserved.
1. Introduction The cochlear sensory hair cells of mammals are vulnerable to ototoxic damage, and mammalian cochlear hair cells do not regenerate spontaneously (Warchol et al., 1993; Forge et al., 1993). Recent studies indicate that cell-based therapy could be used as a
* Corresponding author. E-mail addresses: [email protected] (Y. Tona), [email protected]. kyoto-u.ac.jp (T. Inaoka), [email protected] (J. Ito), [email protected] (S. Kawano), [email protected] (T. Nakagawa). http://dx.doi.org/10.1016/j.heares.2015.08.007 0378-5955/© 2015 Elsevier B.V. All rights reserved.
therapeutic option for inner ear disorders (Nakagawa and Ito, 2005; Coleman et al., 2007; Li et al., 2004); however, this therapy has yet to be applied clinically. Some agents are effective for treating sensorineural hearing loss (SNHL) in mammals (Romand and Chardin, 1999; Chen et al., 2003; Malgrange et al., 1999) and in humans (Chen et al., 2003; Nakagawa et al., 2010, 2014); however, they are limited to the acute phase of SNHL. The use of cochlear implants to treat congenital or chronic severe SNHL is widespread in many countries. Cochlear implants are composed of an implantable device and an external device that contains a microphone, speech processor, and transmitter.
Y. Tona et al. / Hearing Research 330 (2015) 106e112
Fluid oscillations transmitted from ossicles travel from the base to the apex of the scala vestibuli, and reach the helicotrema. They finally reach the round window. The fluid oscillations induce tonotopic vibration of the basilar membrane, leading to depolarization of hair cells at distinct frequencies, namely mechanical tonotopy of mammalian cochleae. We focused on the fact that the damaged cochleae maintain the capacity of mechanical tonotopy. We considered that implantation of a material that is capable of converting vibrations into electrical signals may compensate for hearing dysfunction (Shintaku et al., 2010a, 2013; Inaoka et al., 2011). Piezoelectric materials, which are capable of generating electrical output from mechanical distortion, were considered highly applicable for our concept. A thin membrane made of piezoelectric materials was utilized to convert vibrations into electrical signals (Shintaku et al., 2010a, 2013; Inaoka et al., 2011). A piezoelectric membrane positioned in the scala tympani near the basilar membrane would theoretically convert vibrations into an electrical output that sequentially stimulates the spiral ganglion neurons (SGNs). Our previous research demonstrated that a piezoelectric membrane generated electrical potentials in response to sound stimuli that were in turn able to induce auditory brainstem responses in deafened guinea pigs (Inaoka et al., 2011; Shintaku et al., 2013). In addition, the piezoelectric membrane exhibited tuning for sound frequency similar to the basilar membrane, indicating that the piezoelectric membrane could reproduce the mechanical tonotopy of the basilar membrane (Shintaku et al., 2010a, 2013). However, the electrical output from our device may be insufficient in comparison to that from conventional cochlear implants. In addition, the device required to be fixed in the scala tympani of the cochlea. For proof-of-concept of the device with the piezoelectric membrane, its electrical output needs to be increased, or its required output needs to be reduced for stimulating SGNs. To fix the device in the scala tympani and to reduce demanded outputs for stimulation of SGNs, a new electrode with needles penetrating into the cochlear modiolus was developed. In this paper, we examined whether the device would remain fixed in place after implantation without undergoing severe damage in cochleae. For the functional analysis, the alteration in the evoked auditory brainstem response (eABR) threshold was measured depending on the difference in the site of the electrode in the cochlea, in the scala tympani, or inserted into the cochlear modiolus. 2. Materials and methods 2.1. Experimental animals Thirty-nine Hartley strain guinea pigs (age, 4e10 weeks; weight, 300e600 g) were purchased from Japan SLC, Inc. (Hamamatsu, Japan). Animal care was performed under the supervision of the Institute of Laboratory Animals, Graduate School of Medicine, Kyoto University, (Kyoto, Japan). All experimental procedures were performed in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals.
107
was constructed to evaluate the efficacy for fixation in the scala tympani of cochleae and histological damage after implantation (Fig. 1b). The outer shape of the silicon frame was based on a previous model (Inaoka et al., 2011). The length of the attached needle electrode was 250 mm. A tab was added on the lateral side of the device for easier handling with forceps. Another was the device with one needle, which was used as an electrode in the cochlea for eABR recording (Fig. 1c). 2.3. Implantation surgery Five Hartley guinea pigs were used for the evaluation of efficacy of fixation in the scala tympani and histological damage due to implantation. The animals were anesthetized with an intramuscular injection of midazolam [10 mg/kg, (Astellas Pharma, Tokyo, Japan)] and xylazine [10 mg/kg, (Bayer, Cologne, Germany)]. Supplemental doses were administered every 2 h (or more often if the animal withdrew its leg in response to applied pressure). The conventional retroauricular approach in the lateral recumbent position was used. The bony wall of the otic bulla was removed to expose the basal turn of the cochlea. Cochleostomy was performed on the basal portion of the cochlea in order to visualize the scala tympani. The tab of the device was grasped with forceps and the device was implanted onto the scala tympani with the needle electrodes inserted into the cochlear modiolus. Under the surgical microscope, the needle was cautiously inserted until the root with the device was tightly fixed. The cochleostomy site was finally closed with homogeneous bony powders and fibrin glue. For the comparison of fibrous tissue densities, sham operations were performed for guinea pigs (n ¼ 5) with no device insertion from the cochleostomy site, only to be closed identically. 2.4. eABR recording Measurements of eABRs were modified from previously described methods (Ogita et al., 2009). Monophasic voltage pulses were generated by computer control using a real-time processor (TuckereDavis Technologies, Florida, US). The eABRs were measured from the left cochlea of guinea pigs with normal hearing in a sound-attenuated and electrically shielded room (n ¼ 7). A device with a single needle was inserted into the scala tympani through the cochleostomy site. Another electrode was fixed at the mastoid bulla where it served as the return electrode. The pulse width of the electrical stimulation for eABR was maintained at 0.2 ms biphasic current pulse throughout all experiments. Bioelectrical signals were digitally amplified, averaged for 500 repetitions, and recorded by subdermal stainless steel needle electrodes. The thresholds for eABRs were defined as the smallest current amplitude that was required to evoke a response with a latency of 2e3 ms after the stimulus. The serial measurements of eABR were performed with two sites of the electrode, in the scala tympani (Fig. 1d) or inserted into the cochlear modiolus (Fig. 1e). Thresholds of eABRs were presented as individual values and the mean ± standard error. Unilateral student's paired t-test was used to examine the reduction of the threshold.
2.2. Device and electrodes 2.5. Histological analyses We previously reported the design of an implantable device that was specialized for the guinea pig cochlea [Fig. 1a (Inaoka et al., 2011)]. The device contained a piezoelectric membrane composed of poly(vinylidenefluoride-co-trifluoroethylene) [p(VDF-TrFE)], which was surrounded by a fan-shaped silicon frame. In this experiment, we prepared two different types of devices based on the design of our previous study (Inaoka et al., 2011). One was the device, with two needles for the fixation in the cochlea. This device
On day 28 (n ¼ 5) after device implantation, the temporal bones were collected and immersed in 4% paraformaldehyde in 0.01 M phosphate buffered saline at 4 C for 12 h. After the removal of the devices, the specimens were decalcified with 10% ethylenediaminetetra-acetic acid for 14 days at 4 C. The samples were subsequently embedded and frozen in Tissue-Tec OCT compound (Sakura Finetek, Tokyo, Japan). When samples were embedded in OCT compound,
108
Y. Tona et al. / Hearing Research 330 (2015) 106e112
Fig. 1. The schematic of our device. (a) Design of the previous piezoelectric device. The fan-shaped frame is made of silicone. The piezoelectric membrane is in green. (b) The electrode containing the double needle for assessing the surgical procedure and histology. The needle is made of platinum (Pt) and silicone. A tab is added on the lateral side of the device for easier handling with forceps. The electrode does not contain the piezoelectric material. (c) The electrode containing a single needle for measuring the evoked auditory brainstem response (eABR) thresholds. The electric wire is connected for electrical stimulation. (d) The schema showing the position of the device in the scala tympani when measuring the eABR threshold. (e) The schema showing the position of the device with insertion of the needle into the cochlear modiolus.
we paid special attention to the position of the site where the electrode had been inserted. The puncture site was set to be included in serial sections. The cryosections were obtained at a thickness of 10 mm. For samples obtained 28 days after device implantation, all the cryosections were collected within 500 mm range centered upon the midmodiolar sections for the puncture sites to be included. The cryosections of the every other slice were then stained with hematoxylin and eosin (HE) and viewed under an optical microscope (DP70-WPCXP; Olympus, Tokyo, Japan). The histological findings from the contralateral cochleae that did not undergo surgical treatment (n ¼ 5) served as the control for SGN densities and HC counts. For the comparison of fibrous tissue densities, guinea pigs (n ¼ 5) underwent sham operations with no device implantation after cochleostomy and were closed identically. Sham-operated cochleae were collected on day 28 after operation.
2.6. TdT-mediated dUTP nick end labeling staining On days 1 (n ¼ 6) and 7 (n ¼ 6) after implantation, acute damage due to implantation of the device was assessed by TdT-mediated dUTP Nick End Labeling (TUNEL) staining. An in situ apoptosis detection kit (Takara Bio Inc, Otsu, Japan) following the manufacturer's protocol was used. In brief, fragmented DNA in the apoptotic cells was marked by the TdT enzyme with a FITC tag and then incubated with an anti-FITC HRP conjugate. Staining with DAB (Vector Laboratories, Burlingame, CA, USA) to detect HRP was performed. For nuclear counterstaining, 3% methyl green was used.
To prepare a positive control, DNA fragmentation was performed by exposing tissue sections to deoxyribonuclease 1 (DNase1). TUNEL assay was performed using four randomly selected sections around the site of needle insertion, including the sections with the puncture site. Apoptosis index was determined as the ratio of cells labeled by DAB in Rosenthal canal. For comparison, sham-operated cochleae, for which cochleostomy site was only to be closed identically without device implantation, were harvested at day 1 (n ¼ 5) and day 7 (n ¼ 5) after surgery, followed by TUNEL assays to compare the ratio of apoptosis in Rosenthal canal. 2.7. Immunohistochemistry Immunohistochemistry for b3 tubulin (Tuj1) was performed with the alternate sections to identify SGNs in Tuj1 for samples obtained 28 days after device implantation. The primary antibody was mouse monoclonal anti-b3 tubulin antibody (Covance, Princeton, NJ) used at 1:500 dilution. Secondary antibodies used were Alexa Fluor-488 conjugated goat anti-mouse immunoglobulin G at 1:500 dilution, followed by nuclear staining with 40 ,6diamidino,2-phenylindole dihydrochloride (DAPI; 1 mg/mL in PBS; Life Technologies, California, US). Images of sections were captured on an optical microscope (DP70-WPCXP; Olympus, Tokyo, Japan). 2.8. SGN count The number of SGNs was counted using immunohistochemical sections stained with b3 tubulin from the cochleae that underwent
Y. Tona et al. / Hearing Research 330 (2015) 106e112
109
surgery and from the contralateral cochleae that received no surgical treatment. The cross-sectional areas of Rosenthal canal were measured from HE sections adjacent to those used for immunohistochemistry by using Image J software (http://www.rsb.info.nist. gov/ij). The SGN densities were measured from the average of all the sections collected for histology. Number of hair cells with nuclei detectable on HE sections was also counted within twenty-five slices centered upon the midmodiolar sections to compare the operation side with the contralateral side. Student t-test was used. A value of p < 0.05 was considered statistically significant.
On day 28 after device implantation, inner ears were collected from five guinea pigs. Inflammation of the middle ear was absent in three of five animals and granulation tissue around the cochlea was present in two animals. The cochleostomy site remained closed with ossification in all animals. We fenestrated the ossified site with a surgical drill and confirmed that the device remained fixed in the scala tympani with the needle electrode inserted into the modiolus.
2.9. Fibrous tissue thickness
To examine acute damage due to needle insertion at the implantation site, TUNEL assay was performed on the harvested cochleae 1 (n ¼ 6) and 7 (n ¼ 6) days after device implantation. It was possible to observe the exact site of the electrode insertion on both days 1 and 7 after implantation (Fig. 2aed). In the cochleae harvested on day 1, the small bony fracture around the bony wall covering Rosenthal canal was detected in all cases (Fig. 2a). In the cochleae harvested on day 7, the similar bony fracture was seen in the corresponding portion surrounded by local fibrosis limited to the site of puncture (Fig. 2c). Few TUNEL-positive cells were observed around the puncture site on both day 1 and day 7 (Fig. 2b, d). The DNase-treated section of Rosenthal canal, which served as a positive control, demonstrated a brown nucleus, indicating the existence of fragmented DNA seen in the apoptotic cells (Fig. 2e). The ratios of apoptotic cells in Rosenthal canal were 2.3 ± 3.1% on day 1 and 1.7 ± 1.4% on day 7 after device implantation,
The collagen deposition inside the scala tympani was defined as the maximum perpendicular length of fibrous capsule surrounding the implanted electrode at 1 month after implantation. For statistical analysis, student t-test was used. 3. Results 3.1. Efficacy for fixation of the device The double-electrode devices were implanted onto the modiolus of left cochleae in five guinea pigs, which were allowed to survive until day 28 after surgery. Needle electrodes of the devices were successfully inserted into the modiolus without destruction of the basilar membrane.
3.2. Acute damage due to device implantation
Fig. 2. Representative sections showing electrode-neural interface on day 1 and 7. The acute damage around the site of needle penetration was assessed with TdT-mediated dUTP Nick End Labeling on day 1 (a), (b) and day 7 (c), (d). (a) The representative section of the harvested cochleae on day 1 after device implantation. The arrow head corresponds to the site of electrode insertion. The portion surrounded by the black rectangle was magnified in (b). (b) Higher magnification of the section shown in (a). The small numbers of cells stained in brown were TUNEL-positive cells shown by the arrows. (c) The representative section containing the site of puncture marked by the arrow head, obtained on day 7 after device implantation. (d) Higher magnification of the section shown in (c) marked by the black rectangle. The TUNEL-positive cells were shown by the arrows. (e) DNase-treated cochlear section of Rosenthal canal for positive control. Brown cells indicate the existence of fragmented cDNA seen in apoptotic cells. (f) Apoptotic cells were counted in Rosenthal canal on both day 1 and 7 after device implantation. Bars indicate standard deviation. Scale bars: 200 mm in (a), (c), 40 mm in (b), (d), and (e).
110
Y. Tona et al. / Hearing Research 330 (2015) 106e112
respectively, which were similar to those of the sham-operated cochleae (2.1 ± 2.5% on day 1 and 2.3 ± 0.83% on day 7) (Fig. 2f). 3.3. Chronic damage due to device implantation The histological findings of both cochleae from all the animals are shown in Fig. 3aej. In every animal, the cochleae with device implantation had limited fibrous tissues and cell infiltration around the rim of the scala tympani (Fig. 3a, c, e, g, i), which indicates no severe influence on the scala tympani due to device implantation.
Each cochlear morphology harvested after device implantation was almost identical to the right cochlea without operation, except the thin fibrous tissues and cells around the periphery of the scala tympani (Fig. 3aej). Immunostaining performed for b3 tubulin and nuclear staining with DAPI revealed conserved SGNs (Fig. 3k, l). There were no significant differences in the SGN density between the cochleae with implantation surgery and those without surgical treatment (Fig. 3o). The site of electroeneural interface revealed conserved neural cell bodies around the insertion site with limited inflammatory change (Fig. 3m). The higher magnification of the
Fig. 3. Histology and immunohistochemistry at 1 month after implantation. The histological sections at the mid-basal portion on the operated side (a, c, e, g, i) and the contralateral side (b, d, f, h, j). for animal #1 (a, b), #2 (c, d), #3 (e, f), #4 (g, h), and #5 (i, j) 1 month after surgery (Scale ¼ 100 mm). Arrows indicate the fibrous tissues in the scala tympani. (k) The spiral ganglion neurons (SGNs) stained with Tuj1 on animal #1 for SGN density measurement. (l) The same section as in panel (k), but stained with 40 ,6-diamidino,2-phenylindole dihydrochloride (DAPI). (m) The site of electrode-neural interface from animal #5. The arrow head indicates the electrode-neural interface. Limited inflammation was observed around the site of needle penetration (Scale ¼ 40 mm) (n) Higher magnification of the organ of Corti at the mid-basal portion from the left cochlea of animal #1. The inner (I) and outer (O) hair cells are indicated. (o) The SGN density in Rosenthal canal on the operation side (Lt Op) and on the contralateral side (Rt) at the mid-basal portion. There are no significant differences in the SGN density between the cochleae with and without implantation surgery (p ¼ 0.66, Student paired t-test). (p) Number of inner hair cells (IHCs) and outer hair cells (OHCs) are counted for comparison between the cochleae with and without implantation surgery. There are no significant differences in hair cell numbers (p ¼ 0.43 for IHCs, p ¼ 0.19 for OHCs). Bars indicate the standard error. (q) Thickness of the fibrous formation of the implanted cochleae compared with sham operation. There were no significant differences in the thickness of the fibrous formation (p ¼ 0.70). Bars indicate standard deviation.
Y. Tona et al. / Hearing Research 330 (2015) 106e112
organ of Corti showed both inner and outer hair cells without apparent degeneration (Fig. 3n). There were no significant differences in hair cell numbers between the operation side and the contralateral side (Fig. 3p). The capsular thickness of the fibrous tissue was measured on day 28 after device implantation (Fig. 3q). The maximum fibrous tissue thickness in the scala tympani was almost identical to that in the sham-operated specimens (Fig. 3q). 3.4. Evoked auditory brainstem response measurement The thresholds of eABR were measured and compared between two sites of the electrode, in the scala tympani or inserted into the cochlear modiolus. Fig. 4 shows representative eABR waveforms with two conditions. The mean values of the eABR thresholds of the scala tympani measurement and the modiolus insertion were 200.0 ± 40.8 mA and 110.7 ± 17.1 mA, respectively (Table 1). The eABR threshold of the modiolus insertion was significantly smaller than that of the scala tympani (p ¼ 0.028). 4. Discussion Our previous research demonstrated that a device consisting of a piezoelectric membrane and silicon frame generated electrical potentials in response to sound stimuli after implantation into guinea pig cochleae (Inaoka et al., 2011). This indicated the capacity of the piezoelectric material to mimic the function of inner hair cells and basilar membranes. To achieve hearing recovery using a piezoelectric device, several issues including the equipment to fix the device in the scala tympani and electrodes to transmit electric signals derived from a piezoelectric membrane needed to be addressed. In the current study, we fabricated electrodes for the stimulation of SGNs, which can simultaneously fix a device in the scala tympani by penetrating the cochlear modiolus. Macroscopic findings demonstrated successful fixation of the implanted device after 1 month, indicating the efficacy of penetration of double electrodes into the modiolus adjacent to the SGNs. Acute damage to cochlear tissues due to device implantation was assessed by TUNEL staining. Results demonstrated apoptosis in few cells of the cochleae seven days after device implantation, indicating that the penetration of double electrodes into the cochlear modiolus did not induce severe acute phase damage to cochlear cells, in particular, to SGNs. In the chronic phase, massive infiltration of inflammatory cells into the scala tympani was not identified. Quantitative analysis
111
revealed no decrease in the number of SGNs at 1 month after device implantation. In addition, fibrosis in the scala tympani, which is a common finding after the insertion of cochlear implant electrodes, was not obvious in the implanted cochleae, when compared with sham-operated cochleae. Altogether, only minimal histological changes in the cochleae were caused by device implantation. However, the current model did not contain the piezoelectric membrane. Therefore, in vivo experiments using devices that carry p (VDFeTrFE) films should be performed to confirm their biocompatibility, although an in vitro study has already been reported (Shintaku et al., 2010b). The needle electrode used in the current study was designed to be 250 mm to reach the surface of SGNs. The present results on eABR recording revealed that insertion of needle electrodes into the cochlear modiolus reduced the electric demands for stimulation of SGNs from 200 mA to 110 mA. Previous reports have proven lower stimulation thresholds with direct stimulation of nerve fibers, compared to intrascalar cochlear stimulation (Middlebrooks and Snyder, 2007; Badi et al., 2002). Middlebrooks and Snyder monitored frequency-specific activation of the auditory pathway after implantation of a multielectrode array that was inserted directly into the auditory nerve (Middlebrooks and Snyder, 2007). The threshold induced by intraneural stimulation was 24.5 dB lower than that by monopolar scala tympani stimulation. The 24.5-dB difference corresponds to an approximately 16-fold lower threshold current for intraneural stimulation. Badi et al. measured the serial eABR thresholds in cats with three conditions for the position of the electrodes, in the scala tympani, on the surface of the modiolar nerve, and deep into the modiolar nerve (Badi et al., 2002). Their results demonstrated that eABR threshold of intrascalar stimulation and stimulation on the surface of modiolar nerve were 300 mA and 150 mA, respectively, which is almost identical to changes in eABR thresholds observed in the present study. Notably, the electrode in the modiolar nerve elicited an eABR threshold of 8 mA, indicating that use of longer electrodes than those used in the present study could reduce electric demands for SGN stimulation. On the other hand, use of longer electrodes can cause serious damage to SGNs. To determine the optimal length of electrodes, further investigations are required. Recently, electro-acoustic stimulation has been applied to conventional cochlear implants. For low frequency sounds, acoustic stimulation is used, and for high frequency sounds, electric stimulation via the electrodes of cochlear implants is used. Clinical outcomes of devices receiving electro-acoustic stimulation
Fig. 4. Representative evoked auditory brainstem response (eABR) waveforms with the position of the electrode in the scala tympani or inserted into the cochlear modiolus. The arrows indicate the positive waves in response to the external electrical current. In this animal, the eABR threshold with intrascalar measurement was 150 mA. With the insertion into OSL, the eABR threshold decreased to 100 mA.
112
Y. Tona et al. / Hearing Research 330 (2015) 106e112
Table 1 The evoked auditory brainstem response (eABR) thresholds with the position of the electrode in the scala tympani or inserted into the cochlear modiolus. SE ¼ standard error. Animal
In scala tympani
Modiolus insertion
Reduction of eABR thresholds
#6 #7 #8 #9 #10 #11 #12 Mean SE
325 175 375 150 150 150 75 200 40.8
200 125 75 75 100 125 75 110.7 17.1
125 50 300 75 50 25 0 89.3 38.1 (mA)
demonstrated well preserved hearing at low frequencies, which is obtained via acoustic stimulation (Gantz and Turner, 2003). Hence, the presence of an electrode in the scala tympani may not affect the oscillation of the basilar membrane in the middle and apical portions of cochleae and the mechanical tonotopy in these regions. We, therefore, expect that our device implanted in the scala tympani will not affect the original cochlear tonotopy. However, we need to further investigate these findings more thoroughly. For proof-of-concept of our device, an increase in electric outputs derived from the piezoelectric membrane is included in critical issues. Our previous data suggested that the electrical output should be 105-fold higher to elicit effective stimulation of the auditory primary neurons when the device is positioned in the scala tympani (Inaoka et al., 2011). To increase electric outputs of a device, we are investigating new designs of a piezoelectric membrane and other potential piezoelectric materials. 5. Conclusions The needle electrodes of thexamination demonstrated that device implantation into the OSL did not induce severe inflammatory responses or degeneration of cochlear cells including SGNs. Further improvements of the piezoelectric device are required to validate our results. Acknowledgement We thank Panasonic Corporation for the fabrication of devices with needle electrodes This study was supported by grants of the
Research on Sensory and Communicative Disorders from the Japan Ministry of Health, Labor and Welfare. References Badi, A.N., Hillman, T., Shelton, C., Normann, R.A., 2002. A technique for implantation of a 3-dimensional penetrating electrode array in the modiolar nerve of cats and humans. Arch. Otolaryngol. Head Neck Surg. 128 (9), 1019e1025. Chen, C.Y., Halpin, C., Rauch, S.D., 2003. Oral steroid treatment of sudden sensorineural hearing loss: a ten year retrospective analysis. Otol. Neurotol. 24 (5), 728e733. Coleman, B., de Silva, M.G., Shepherd, R.K., 2007. Concise review: the potential of stem cells for auditory neuron generation and replacement. Stem Cells 25 (11), 2685e2694. Forge, A., Li, L., Corwin, J.T., Nevill, G., 1993. Ultrastructural evidence for hair cell regeneration in the mammalian inner ear. Science 259 (5101), 1616e1619. Gantz, B.J., Turner, C.W., 2003. Combining acoustic and electrical hearing. Laryngoscope 113 (10), 1726e1730. Inaoka, T., Shintaku, H., Nakagawa, T., Kawano, S., Ogita, H., Sakamoto, T., et al., 2011. Piezoelectric materials mimic the function of the cochlear sensory epithelium. Proc. Natl. Acad. Sci. U. S. A. 108 (45), 18390e18395. Li, H., Corrales, C.E., Edge, A., Heller, S., 2004. Stem cells as therapy for hearing loss. Trends Mol. Med. 10 (7), 309e315. Malgrange, B., Rigo, J.M., Van de Water, T.R., Staecker, H., Moonen, G., Lefebvre, P.P., 1999. Growth factor therapy to the damaged inner ear: clinical prospects. Int. J. Pediatr. Otorhinolaryngol. 49 (Suppl. 1), S19eS25. Middlebrooks, J.C., Snyder, R.L., 2007. Auditory prosthesis with a penetrating nerve array. J. Assoc. Res. Otolaryngol. 8 (2), 258e279. Nakagawa, T., Ito, J., 2005. Cell therapy for inner ear diseases. Curr. Pharm. Des. 11 (9), 1203e1207. Nakagawa, T., Sakamoto, T., Hiraumi, H., Kikkawa, Y.S., Yamamoto, N., Hamaguchi, K., et al., 2010. Topical insulin-like growth factor 1 treatment using gelatin hydrogels for glucocorticoid-resistant sudden sensorineural hearing loss: a prospective clinical trial. BMC Med. 8, 76. Nakagawa, T., Kumakawa, K., Usami, S., Hato, N., Tabuchi, K., Takahashi, M., et al., 2014. A randomized controlled clinical trial of topical insulin-like growth factor1 therapy for sudden deafness refractory to systemic corticosteroid treatment. BMC Med. 12, 219. Ogita, H., Nakagawa, T., Lee, K.Y., Inaoka, T., Okano, T., Kikkawa, Y.S., et al., 2009. Surgical invasiveness of cell transplantation into the guinea pig cochlear modiolus. ORL J. Otorhinolaryngol. Relat. Spec. 71 (1), 32e39. Romand, R., Chardin, S., 1999. Effects of growth factors on the hair cells after ototoxic treatment of the neonatal mammalian cochlea in vitro. Brain Res. 825 (1e2), 46e58. Shintaku, H., Kobayashi, H., Zusho, K., Kotera, H., Kawano, S., 2013. Wide-range frequency selectivity in an acoustic sensor fabricated using a microbeam array with non-uniform thickness. J. Micromech. Microeng. 23, 115014. Shintaku, H., Nakagawa, T., Kitagawa, D., Tanujaya, H., Kawano, S., Ito, J., 2010a. Development of piezoelectric acoustic sensor with frequency selectivity for artificial cochlea. Sensors Actuators A Phys. 158 (2), 183e192. Shintaku, H., Tateno, T., Tsuchioka, N., Tanujaya, H., Nakagawa, T., Ito, J., et al., 2010b. Culturing neurons on MEMS fabricated P(VDF-TrFE) films for implantable artificial cochlea. J. Biomech. Sci. Eng. 5 (3), 229e235. Warchol, M.E., Lambert, P.R., Goldstein, B.J., Forge, A., Corwin, J.T., 1993. Regenerative proliferation in inner ear sensory epithelia from adult guinea pigs and humans. Science 259 (5101), 1619e1622.
Contents lists available at ScienceDirect
Hearing Research journal homepage: www.elsevier.com/locate/heares
Development of an electrode for the artificial cochlear sensory epithelium Yosuke Tona a, *, Takatoshi Inaoka a, b, Juichi Ito a, Satoyuki Kawano c, Takayuki Nakagawa a a
Department of Otolaryngology, Head and Neck Surgery, Graduate School of Medicine, Kyoto University, Kawaharacho 54, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan b Inaoka ENT Clinic, 2-1-14-201, Neyaminami, Neyagawa 572-0855, Japan c Department of Mechanical Science and Bioengineering, Graduate School of Engineering Science, Osaka University, 1-3 Machikaneyama, Toyonaka, Osaka 560-8531, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 February 2015 Received in revised form 11 August 2015 Accepted 17 August 2015 Available online 20 August 2015
An artificial cochlear sensory epithelium has been developed on the basis of a new concept that the piezoelectric membrane, which converts mechanical distortion into electricity, can mimic the function of the inner hair cell and basilar membrane of the mammalian cochlea. Our previous research demonstrated that the piezoelectric membrane generated electrical outputs in response to the sound stimulation after implantation into the guinea pig cochlea, whereas electrodes for the stimulation of spiral ganglion neurons have not been fabricated, and a method to fix the device in the cochlea is also required to show proof-of-concept. In the present study, to achieve proof-of-concept of hearing recovery by implantation of the artificial cochlear sensory epithelium, we fabricated new electrodes that stick into the cochlear modiolus, which also play a role in the fixation of the device in the cochlea. The efficacy of new electrodes for fixation of the device in the cochlea and for the stimulation of spiral ganglion neurons was estimated in guinea pigs. Four weeks after implantation, we confirmed that the devices were in place. Histological analysis of the implanted cochleae revealed inconspicuous fibrosis and scar formation compared with the sham-operated specimens (n ¼ 5 for each). The terminal deoxynucleotidyl transferase dUTP nick end labeling method was used to assess cell death due to surgical procedures in the cochleae that were harvested after 1 day (n ¼ 6) and 7 days (n ¼ 6) of implantation; there was no significant increase in apoptotic cell death in the implanted cochleae compared with sham-operated cochleae. In seven animals, serial measurements of electrically evoked auditory brainstem responses were obtained, with the electrode positioned in the scala tympani and with the electrode inserted into the cochlear modiolus. With the insertion of electrodes into the cochlear modiolus, significant reduction was achieved in the thresholds of electrically evoked auditory brainstem responses compared with those placed in the scala tympani (p ¼ 0.028). These findings indicated that the new electrodes efficiently fixed the device in the cochlea and were able to stimulate spiral ganglion neurons. This article is part of a Special Issue entitled .
Keywords: Piezoelectric membrane Mechanoelectrical stimulation Modiolus stimulation
© 2015 Elsevier B.V. All rights reserved.
1. Introduction The cochlear sensory hair cells of mammals are vulnerable to ototoxic damage, and mammalian cochlear hair cells do not regenerate spontaneously (Warchol et al., 1993; Forge et al., 1993). Recent studies indicate that cell-based therapy could be used as a
* Corresponding author. E-mail addresses: [email protected] (Y. Tona), [email protected]. kyoto-u.ac.jp (T. Inaoka), [email protected] (J. Ito), [email protected] (S. Kawano), [email protected] (T. Nakagawa). http://dx.doi.org/10.1016/j.heares.2015.08.007 0378-5955/© 2015 Elsevier B.V. All rights reserved.
therapeutic option for inner ear disorders (Nakagawa and Ito, 2005; Coleman et al., 2007; Li et al., 2004); however, this therapy has yet to be applied clinically. Some agents are effective for treating sensorineural hearing loss (SNHL) in mammals (Romand and Chardin, 1999; Chen et al., 2003; Malgrange et al., 1999) and in humans (Chen et al., 2003; Nakagawa et al., 2010, 2014); however, they are limited to the acute phase of SNHL. The use of cochlear implants to treat congenital or chronic severe SNHL is widespread in many countries. Cochlear implants are composed of an implantable device and an external device that contains a microphone, speech processor, and transmitter.
Y. Tona et al. / Hearing Research 330 (2015) 106e112
Fluid oscillations transmitted from ossicles travel from the base to the apex of the scala vestibuli, and reach the helicotrema. They finally reach the round window. The fluid oscillations induce tonotopic vibration of the basilar membrane, leading to depolarization of hair cells at distinct frequencies, namely mechanical tonotopy of mammalian cochleae. We focused on the fact that the damaged cochleae maintain the capacity of mechanical tonotopy. We considered that implantation of a material that is capable of converting vibrations into electrical signals may compensate for hearing dysfunction (Shintaku et al., 2010a, 2013; Inaoka et al., 2011). Piezoelectric materials, which are capable of generating electrical output from mechanical distortion, were considered highly applicable for our concept. A thin membrane made of piezoelectric materials was utilized to convert vibrations into electrical signals (Shintaku et al., 2010a, 2013; Inaoka et al., 2011). A piezoelectric membrane positioned in the scala tympani near the basilar membrane would theoretically convert vibrations into an electrical output that sequentially stimulates the spiral ganglion neurons (SGNs). Our previous research demonstrated that a piezoelectric membrane generated electrical potentials in response to sound stimuli that were in turn able to induce auditory brainstem responses in deafened guinea pigs (Inaoka et al., 2011; Shintaku et al., 2013). In addition, the piezoelectric membrane exhibited tuning for sound frequency similar to the basilar membrane, indicating that the piezoelectric membrane could reproduce the mechanical tonotopy of the basilar membrane (Shintaku et al., 2010a, 2013). However, the electrical output from our device may be insufficient in comparison to that from conventional cochlear implants. In addition, the device required to be fixed in the scala tympani of the cochlea. For proof-of-concept of the device with the piezoelectric membrane, its electrical output needs to be increased, or its required output needs to be reduced for stimulating SGNs. To fix the device in the scala tympani and to reduce demanded outputs for stimulation of SGNs, a new electrode with needles penetrating into the cochlear modiolus was developed. In this paper, we examined whether the device would remain fixed in place after implantation without undergoing severe damage in cochleae. For the functional analysis, the alteration in the evoked auditory brainstem response (eABR) threshold was measured depending on the difference in the site of the electrode in the cochlea, in the scala tympani, or inserted into the cochlear modiolus. 2. Materials and methods 2.1. Experimental animals Thirty-nine Hartley strain guinea pigs (age, 4e10 weeks; weight, 300e600 g) were purchased from Japan SLC, Inc. (Hamamatsu, Japan). Animal care was performed under the supervision of the Institute of Laboratory Animals, Graduate School of Medicine, Kyoto University, (Kyoto, Japan). All experimental procedures were performed in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals.
107
was constructed to evaluate the efficacy for fixation in the scala tympani of cochleae and histological damage after implantation (Fig. 1b). The outer shape of the silicon frame was based on a previous model (Inaoka et al., 2011). The length of the attached needle electrode was 250 mm. A tab was added on the lateral side of the device for easier handling with forceps. Another was the device with one needle, which was used as an electrode in the cochlea for eABR recording (Fig. 1c). 2.3. Implantation surgery Five Hartley guinea pigs were used for the evaluation of efficacy of fixation in the scala tympani and histological damage due to implantation. The animals were anesthetized with an intramuscular injection of midazolam [10 mg/kg, (Astellas Pharma, Tokyo, Japan)] and xylazine [10 mg/kg, (Bayer, Cologne, Germany)]. Supplemental doses were administered every 2 h (or more often if the animal withdrew its leg in response to applied pressure). The conventional retroauricular approach in the lateral recumbent position was used. The bony wall of the otic bulla was removed to expose the basal turn of the cochlea. Cochleostomy was performed on the basal portion of the cochlea in order to visualize the scala tympani. The tab of the device was grasped with forceps and the device was implanted onto the scala tympani with the needle electrodes inserted into the cochlear modiolus. Under the surgical microscope, the needle was cautiously inserted until the root with the device was tightly fixed. The cochleostomy site was finally closed with homogeneous bony powders and fibrin glue. For the comparison of fibrous tissue densities, sham operations were performed for guinea pigs (n ¼ 5) with no device insertion from the cochleostomy site, only to be closed identically. 2.4. eABR recording Measurements of eABRs were modified from previously described methods (Ogita et al., 2009). Monophasic voltage pulses were generated by computer control using a real-time processor (TuckereDavis Technologies, Florida, US). The eABRs were measured from the left cochlea of guinea pigs with normal hearing in a sound-attenuated and electrically shielded room (n ¼ 7). A device with a single needle was inserted into the scala tympani through the cochleostomy site. Another electrode was fixed at the mastoid bulla where it served as the return electrode. The pulse width of the electrical stimulation for eABR was maintained at 0.2 ms biphasic current pulse throughout all experiments. Bioelectrical signals were digitally amplified, averaged for 500 repetitions, and recorded by subdermal stainless steel needle electrodes. The thresholds for eABRs were defined as the smallest current amplitude that was required to evoke a response with a latency of 2e3 ms after the stimulus. The serial measurements of eABR were performed with two sites of the electrode, in the scala tympani (Fig. 1d) or inserted into the cochlear modiolus (Fig. 1e). Thresholds of eABRs were presented as individual values and the mean ± standard error. Unilateral student's paired t-test was used to examine the reduction of the threshold.
2.2. Device and electrodes 2.5. Histological analyses We previously reported the design of an implantable device that was specialized for the guinea pig cochlea [Fig. 1a (Inaoka et al., 2011)]. The device contained a piezoelectric membrane composed of poly(vinylidenefluoride-co-trifluoroethylene) [p(VDF-TrFE)], which was surrounded by a fan-shaped silicon frame. In this experiment, we prepared two different types of devices based on the design of our previous study (Inaoka et al., 2011). One was the device, with two needles for the fixation in the cochlea. This device
On day 28 (n ¼ 5) after device implantation, the temporal bones were collected and immersed in 4% paraformaldehyde in 0.01 M phosphate buffered saline at 4 C for 12 h. After the removal of the devices, the specimens were decalcified with 10% ethylenediaminetetra-acetic acid for 14 days at 4 C. The samples were subsequently embedded and frozen in Tissue-Tec OCT compound (Sakura Finetek, Tokyo, Japan). When samples were embedded in OCT compound,
108
Y. Tona et al. / Hearing Research 330 (2015) 106e112
Fig. 1. The schematic of our device. (a) Design of the previous piezoelectric device. The fan-shaped frame is made of silicone. The piezoelectric membrane is in green. (b) The electrode containing the double needle for assessing the surgical procedure and histology. The needle is made of platinum (Pt) and silicone. A tab is added on the lateral side of the device for easier handling with forceps. The electrode does not contain the piezoelectric material. (c) The electrode containing a single needle for measuring the evoked auditory brainstem response (eABR) thresholds. The electric wire is connected for electrical stimulation. (d) The schema showing the position of the device in the scala tympani when measuring the eABR threshold. (e) The schema showing the position of the device with insertion of the needle into the cochlear modiolus.
we paid special attention to the position of the site where the electrode had been inserted. The puncture site was set to be included in serial sections. The cryosections were obtained at a thickness of 10 mm. For samples obtained 28 days after device implantation, all the cryosections were collected within 500 mm range centered upon the midmodiolar sections for the puncture sites to be included. The cryosections of the every other slice were then stained with hematoxylin and eosin (HE) and viewed under an optical microscope (DP70-WPCXP; Olympus, Tokyo, Japan). The histological findings from the contralateral cochleae that did not undergo surgical treatment (n ¼ 5) served as the control for SGN densities and HC counts. For the comparison of fibrous tissue densities, guinea pigs (n ¼ 5) underwent sham operations with no device implantation after cochleostomy and were closed identically. Sham-operated cochleae were collected on day 28 after operation.
2.6. TdT-mediated dUTP nick end labeling staining On days 1 (n ¼ 6) and 7 (n ¼ 6) after implantation, acute damage due to implantation of the device was assessed by TdT-mediated dUTP Nick End Labeling (TUNEL) staining. An in situ apoptosis detection kit (Takara Bio Inc, Otsu, Japan) following the manufacturer's protocol was used. In brief, fragmented DNA in the apoptotic cells was marked by the TdT enzyme with a FITC tag and then incubated with an anti-FITC HRP conjugate. Staining with DAB (Vector Laboratories, Burlingame, CA, USA) to detect HRP was performed. For nuclear counterstaining, 3% methyl green was used.
To prepare a positive control, DNA fragmentation was performed by exposing tissue sections to deoxyribonuclease 1 (DNase1). TUNEL assay was performed using four randomly selected sections around the site of needle insertion, including the sections with the puncture site. Apoptosis index was determined as the ratio of cells labeled by DAB in Rosenthal canal. For comparison, sham-operated cochleae, for which cochleostomy site was only to be closed identically without device implantation, were harvested at day 1 (n ¼ 5) and day 7 (n ¼ 5) after surgery, followed by TUNEL assays to compare the ratio of apoptosis in Rosenthal canal. 2.7. Immunohistochemistry Immunohistochemistry for b3 tubulin (Tuj1) was performed with the alternate sections to identify SGNs in Tuj1 for samples obtained 28 days after device implantation. The primary antibody was mouse monoclonal anti-b3 tubulin antibody (Covance, Princeton, NJ) used at 1:500 dilution. Secondary antibodies used were Alexa Fluor-488 conjugated goat anti-mouse immunoglobulin G at 1:500 dilution, followed by nuclear staining with 40 ,6diamidino,2-phenylindole dihydrochloride (DAPI; 1 mg/mL in PBS; Life Technologies, California, US). Images of sections were captured on an optical microscope (DP70-WPCXP; Olympus, Tokyo, Japan). 2.8. SGN count The number of SGNs was counted using immunohistochemical sections stained with b3 tubulin from the cochleae that underwent
Y. Tona et al. / Hearing Research 330 (2015) 106e112
109
surgery and from the contralateral cochleae that received no surgical treatment. The cross-sectional areas of Rosenthal canal were measured from HE sections adjacent to those used for immunohistochemistry by using Image J software (http://www.rsb.info.nist. gov/ij). The SGN densities were measured from the average of all the sections collected for histology. Number of hair cells with nuclei detectable on HE sections was also counted within twenty-five slices centered upon the midmodiolar sections to compare the operation side with the contralateral side. Student t-test was used. A value of p < 0.05 was considered statistically significant.
On day 28 after device implantation, inner ears were collected from five guinea pigs. Inflammation of the middle ear was absent in three of five animals and granulation tissue around the cochlea was present in two animals. The cochleostomy site remained closed with ossification in all animals. We fenestrated the ossified site with a surgical drill and confirmed that the device remained fixed in the scala tympani with the needle electrode inserted into the modiolus.
2.9. Fibrous tissue thickness
To examine acute damage due to needle insertion at the implantation site, TUNEL assay was performed on the harvested cochleae 1 (n ¼ 6) and 7 (n ¼ 6) days after device implantation. It was possible to observe the exact site of the electrode insertion on both days 1 and 7 after implantation (Fig. 2aed). In the cochleae harvested on day 1, the small bony fracture around the bony wall covering Rosenthal canal was detected in all cases (Fig. 2a). In the cochleae harvested on day 7, the similar bony fracture was seen in the corresponding portion surrounded by local fibrosis limited to the site of puncture (Fig. 2c). Few TUNEL-positive cells were observed around the puncture site on both day 1 and day 7 (Fig. 2b, d). The DNase-treated section of Rosenthal canal, which served as a positive control, demonstrated a brown nucleus, indicating the existence of fragmented DNA seen in the apoptotic cells (Fig. 2e). The ratios of apoptotic cells in Rosenthal canal were 2.3 ± 3.1% on day 1 and 1.7 ± 1.4% on day 7 after device implantation,
The collagen deposition inside the scala tympani was defined as the maximum perpendicular length of fibrous capsule surrounding the implanted electrode at 1 month after implantation. For statistical analysis, student t-test was used. 3. Results 3.1. Efficacy for fixation of the device The double-electrode devices were implanted onto the modiolus of left cochleae in five guinea pigs, which were allowed to survive until day 28 after surgery. Needle electrodes of the devices were successfully inserted into the modiolus without destruction of the basilar membrane.
3.2. Acute damage due to device implantation
Fig. 2. Representative sections showing electrode-neural interface on day 1 and 7. The acute damage around the site of needle penetration was assessed with TdT-mediated dUTP Nick End Labeling on day 1 (a), (b) and day 7 (c), (d). (a) The representative section of the harvested cochleae on day 1 after device implantation. The arrow head corresponds to the site of electrode insertion. The portion surrounded by the black rectangle was magnified in (b). (b) Higher magnification of the section shown in (a). The small numbers of cells stained in brown were TUNEL-positive cells shown by the arrows. (c) The representative section containing the site of puncture marked by the arrow head, obtained on day 7 after device implantation. (d) Higher magnification of the section shown in (c) marked by the black rectangle. The TUNEL-positive cells were shown by the arrows. (e) DNase-treated cochlear section of Rosenthal canal for positive control. Brown cells indicate the existence of fragmented cDNA seen in apoptotic cells. (f) Apoptotic cells were counted in Rosenthal canal on both day 1 and 7 after device implantation. Bars indicate standard deviation. Scale bars: 200 mm in (a), (c), 40 mm in (b), (d), and (e).
110
Y. Tona et al. / Hearing Research 330 (2015) 106e112
respectively, which were similar to those of the sham-operated cochleae (2.1 ± 2.5% on day 1 and 2.3 ± 0.83% on day 7) (Fig. 2f). 3.3. Chronic damage due to device implantation The histological findings of both cochleae from all the animals are shown in Fig. 3aej. In every animal, the cochleae with device implantation had limited fibrous tissues and cell infiltration around the rim of the scala tympani (Fig. 3a, c, e, g, i), which indicates no severe influence on the scala tympani due to device implantation.
Each cochlear morphology harvested after device implantation was almost identical to the right cochlea without operation, except the thin fibrous tissues and cells around the periphery of the scala tympani (Fig. 3aej). Immunostaining performed for b3 tubulin and nuclear staining with DAPI revealed conserved SGNs (Fig. 3k, l). There were no significant differences in the SGN density between the cochleae with implantation surgery and those without surgical treatment (Fig. 3o). The site of electroeneural interface revealed conserved neural cell bodies around the insertion site with limited inflammatory change (Fig. 3m). The higher magnification of the
Fig. 3. Histology and immunohistochemistry at 1 month after implantation. The histological sections at the mid-basal portion on the operated side (a, c, e, g, i) and the contralateral side (b, d, f, h, j). for animal #1 (a, b), #2 (c, d), #3 (e, f), #4 (g, h), and #5 (i, j) 1 month after surgery (Scale ¼ 100 mm). Arrows indicate the fibrous tissues in the scala tympani. (k) The spiral ganglion neurons (SGNs) stained with Tuj1 on animal #1 for SGN density measurement. (l) The same section as in panel (k), but stained with 40 ,6-diamidino,2-phenylindole dihydrochloride (DAPI). (m) The site of electrode-neural interface from animal #5. The arrow head indicates the electrode-neural interface. Limited inflammation was observed around the site of needle penetration (Scale ¼ 40 mm) (n) Higher magnification of the organ of Corti at the mid-basal portion from the left cochlea of animal #1. The inner (I) and outer (O) hair cells are indicated. (o) The SGN density in Rosenthal canal on the operation side (Lt Op) and on the contralateral side (Rt) at the mid-basal portion. There are no significant differences in the SGN density between the cochleae with and without implantation surgery (p ¼ 0.66, Student paired t-test). (p) Number of inner hair cells (IHCs) and outer hair cells (OHCs) are counted for comparison between the cochleae with and without implantation surgery. There are no significant differences in hair cell numbers (p ¼ 0.43 for IHCs, p ¼ 0.19 for OHCs). Bars indicate the standard error. (q) Thickness of the fibrous formation of the implanted cochleae compared with sham operation. There were no significant differences in the thickness of the fibrous formation (p ¼ 0.70). Bars indicate standard deviation.
Y. Tona et al. / Hearing Research 330 (2015) 106e112
organ of Corti showed both inner and outer hair cells without apparent degeneration (Fig. 3n). There were no significant differences in hair cell numbers between the operation side and the contralateral side (Fig. 3p). The capsular thickness of the fibrous tissue was measured on day 28 after device implantation (Fig. 3q). The maximum fibrous tissue thickness in the scala tympani was almost identical to that in the sham-operated specimens (Fig. 3q). 3.4. Evoked auditory brainstem response measurement The thresholds of eABR were measured and compared between two sites of the electrode, in the scala tympani or inserted into the cochlear modiolus. Fig. 4 shows representative eABR waveforms with two conditions. The mean values of the eABR thresholds of the scala tympani measurement and the modiolus insertion were 200.0 ± 40.8 mA and 110.7 ± 17.1 mA, respectively (Table 1). The eABR threshold of the modiolus insertion was significantly smaller than that of the scala tympani (p ¼ 0.028). 4. Discussion Our previous research demonstrated that a device consisting of a piezoelectric membrane and silicon frame generated electrical potentials in response to sound stimuli after implantation into guinea pig cochleae (Inaoka et al., 2011). This indicated the capacity of the piezoelectric material to mimic the function of inner hair cells and basilar membranes. To achieve hearing recovery using a piezoelectric device, several issues including the equipment to fix the device in the scala tympani and electrodes to transmit electric signals derived from a piezoelectric membrane needed to be addressed. In the current study, we fabricated electrodes for the stimulation of SGNs, which can simultaneously fix a device in the scala tympani by penetrating the cochlear modiolus. Macroscopic findings demonstrated successful fixation of the implanted device after 1 month, indicating the efficacy of penetration of double electrodes into the modiolus adjacent to the SGNs. Acute damage to cochlear tissues due to device implantation was assessed by TUNEL staining. Results demonstrated apoptosis in few cells of the cochleae seven days after device implantation, indicating that the penetration of double electrodes into the cochlear modiolus did not induce severe acute phase damage to cochlear cells, in particular, to SGNs. In the chronic phase, massive infiltration of inflammatory cells into the scala tympani was not identified. Quantitative analysis
111
revealed no decrease in the number of SGNs at 1 month after device implantation. In addition, fibrosis in the scala tympani, which is a common finding after the insertion of cochlear implant electrodes, was not obvious in the implanted cochleae, when compared with sham-operated cochleae. Altogether, only minimal histological changes in the cochleae were caused by device implantation. However, the current model did not contain the piezoelectric membrane. Therefore, in vivo experiments using devices that carry p (VDFeTrFE) films should be performed to confirm their biocompatibility, although an in vitro study has already been reported (Shintaku et al., 2010b). The needle electrode used in the current study was designed to be 250 mm to reach the surface of SGNs. The present results on eABR recording revealed that insertion of needle electrodes into the cochlear modiolus reduced the electric demands for stimulation of SGNs from 200 mA to 110 mA. Previous reports have proven lower stimulation thresholds with direct stimulation of nerve fibers, compared to intrascalar cochlear stimulation (Middlebrooks and Snyder, 2007; Badi et al., 2002). Middlebrooks and Snyder monitored frequency-specific activation of the auditory pathway after implantation of a multielectrode array that was inserted directly into the auditory nerve (Middlebrooks and Snyder, 2007). The threshold induced by intraneural stimulation was 24.5 dB lower than that by monopolar scala tympani stimulation. The 24.5-dB difference corresponds to an approximately 16-fold lower threshold current for intraneural stimulation. Badi et al. measured the serial eABR thresholds in cats with three conditions for the position of the electrodes, in the scala tympani, on the surface of the modiolar nerve, and deep into the modiolar nerve (Badi et al., 2002). Their results demonstrated that eABR threshold of intrascalar stimulation and stimulation on the surface of modiolar nerve were 300 mA and 150 mA, respectively, which is almost identical to changes in eABR thresholds observed in the present study. Notably, the electrode in the modiolar nerve elicited an eABR threshold of 8 mA, indicating that use of longer electrodes than those used in the present study could reduce electric demands for SGN stimulation. On the other hand, use of longer electrodes can cause serious damage to SGNs. To determine the optimal length of electrodes, further investigations are required. Recently, electro-acoustic stimulation has been applied to conventional cochlear implants. For low frequency sounds, acoustic stimulation is used, and for high frequency sounds, electric stimulation via the electrodes of cochlear implants is used. Clinical outcomes of devices receiving electro-acoustic stimulation
Fig. 4. Representative evoked auditory brainstem response (eABR) waveforms with the position of the electrode in the scala tympani or inserted into the cochlear modiolus. The arrows indicate the positive waves in response to the external electrical current. In this animal, the eABR threshold with intrascalar measurement was 150 mA. With the insertion into OSL, the eABR threshold decreased to 100 mA.
112
Y. Tona et al. / Hearing Research 330 (2015) 106e112
Table 1 The evoked auditory brainstem response (eABR) thresholds with the position of the electrode in the scala tympani or inserted into the cochlear modiolus. SE ¼ standard error. Animal
In scala tympani
Modiolus insertion
Reduction of eABR thresholds
#6 #7 #8 #9 #10 #11 #12 Mean SE
325 175 375 150 150 150 75 200 40.8
200 125 75 75 100 125 75 110.7 17.1
125 50 300 75 50 25 0 89.3 38.1 (mA)
demonstrated well preserved hearing at low frequencies, which is obtained via acoustic stimulation (Gantz and Turner, 2003). Hence, the presence of an electrode in the scala tympani may not affect the oscillation of the basilar membrane in the middle and apical portions of cochleae and the mechanical tonotopy in these regions. We, therefore, expect that our device implanted in the scala tympani will not affect the original cochlear tonotopy. However, we need to further investigate these findings more thoroughly. For proof-of-concept of our device, an increase in electric outputs derived from the piezoelectric membrane is included in critical issues. Our previous data suggested that the electrical output should be 105-fold higher to elicit effective stimulation of the auditory primary neurons when the device is positioned in the scala tympani (Inaoka et al., 2011). To increase electric outputs of a device, we are investigating new designs of a piezoelectric membrane and other potential piezoelectric materials. 5. Conclusions The needle electrodes of thexamination demonstrated that device implantation into the OSL did not induce severe inflammatory responses or degeneration of cochlear cells including SGNs. Further improvements of the piezoelectric device are required to validate our results. Acknowledgement We thank Panasonic Corporation for the fabrication of devices with needle electrodes This study was supported by grants of the
Research on Sensory and Communicative Disorders from the Japan Ministry of Health, Labor and Welfare. References Badi, A.N., Hillman, T., Shelton, C., Normann, R.A., 2002. A technique for implantation of a 3-dimensional penetrating electrode array in the modiolar nerve of cats and humans. Arch. Otolaryngol. Head Neck Surg. 128 (9), 1019e1025. Chen, C.Y., Halpin, C., Rauch, S.D., 2003. Oral steroid treatment of sudden sensorineural hearing loss: a ten year retrospective analysis. Otol. Neurotol. 24 (5), 728e733. Coleman, B., de Silva, M.G., Shepherd, R.K., 2007. Concise review: the potential of stem cells for auditory neuron generation and replacement. Stem Cells 25 (11), 2685e2694. Forge, A., Li, L., Corwin, J.T., Nevill, G., 1993. Ultrastructural evidence for hair cell regeneration in the mammalian inner ear. Science 259 (5101), 1616e1619. Gantz, B.J., Turner, C.W., 2003. Combining acoustic and electrical hearing. Laryngoscope 113 (10), 1726e1730. Inaoka, T., Shintaku, H., Nakagawa, T., Kawano, S., Ogita, H., Sakamoto, T., et al., 2011. Piezoelectric materials mimic the function of the cochlear sensory epithelium. Proc. Natl. Acad. Sci. U. S. A. 108 (45), 18390e18395. Li, H., Corrales, C.E., Edge, A., Heller, S., 2004. Stem cells as therapy for hearing loss. Trends Mol. Med. 10 (7), 309e315. Malgrange, B., Rigo, J.M., Van de Water, T.R., Staecker, H., Moonen, G., Lefebvre, P.P., 1999. Growth factor therapy to the damaged inner ear: clinical prospects. Int. J. Pediatr. Otorhinolaryngol. 49 (Suppl. 1), S19eS25. Middlebrooks, J.C., Snyder, R.L., 2007. Auditory prosthesis with a penetrating nerve array. J. Assoc. Res. Otolaryngol. 8 (2), 258e279. Nakagawa, T., Ito, J., 2005. Cell therapy for inner ear diseases. Curr. Pharm. Des. 11 (9), 1203e1207. Nakagawa, T., Sakamoto, T., Hiraumi, H., Kikkawa, Y.S., Yamamoto, N., Hamaguchi, K., et al., 2010. Topical insulin-like growth factor 1 treatment using gelatin hydrogels for glucocorticoid-resistant sudden sensorineural hearing loss: a prospective clinical trial. BMC Med. 8, 76. Nakagawa, T., Kumakawa, K., Usami, S., Hato, N., Tabuchi, K., Takahashi, M., et al., 2014. A randomized controlled clinical trial of topical insulin-like growth factor1 therapy for sudden deafness refractory to systemic corticosteroid treatment. BMC Med. 12, 219. Ogita, H., Nakagawa, T., Lee, K.Y., Inaoka, T., Okano, T., Kikkawa, Y.S., et al., 2009. Surgical invasiveness of cell transplantation into the guinea pig cochlear modiolus. ORL J. Otorhinolaryngol. Relat. Spec. 71 (1), 32e39. Romand, R., Chardin, S., 1999. Effects of growth factors on the hair cells after ototoxic treatment of the neonatal mammalian cochlea in vitro. Brain Res. 825 (1e2), 46e58. Shintaku, H., Kobayashi, H., Zusho, K., Kotera, H., Kawano, S., 2013. Wide-range frequency selectivity in an acoustic sensor fabricated using a microbeam array with non-uniform thickness. J. Micromech. Microeng. 23, 115014. Shintaku, H., Nakagawa, T., Kitagawa, D., Tanujaya, H., Kawano, S., Ito, J., 2010a. Development of piezoelectric acoustic sensor with frequency selectivity for artificial cochlea. Sensors Actuators A Phys. 158 (2), 183e192. Shintaku, H., Tateno, T., Tsuchioka, N., Tanujaya, H., Nakagawa, T., Ito, J., et al., 2010b. Culturing neurons on MEMS fabricated P(VDF-TrFE) films for implantable artificial cochlea. J. Biomech. Sci. Eng. 5 (3), 229e235. Warchol, M.E., Lambert, P.R., Goldstein, B.J., Forge, A., Corwin, J.T., 1993. Regenerative proliferation in inner ear sensory epithelia from adult guinea pigs and humans. Science 259 (5101), 1619e1622.
