Published: January 9th, 2019
Animal models of cochlear implants can advance knowledge of the technological bases of treating permanent sensorineural hearing loss with electrical stimulation. This study presents a surgical protocol for acute deafening and cochlear implantation of an electrode array in mice as well as the functional assessment with auditory brainstem response.
Cochlear implants (CIs) are neuroprosthetic devices that can provide a sense of hearing to deaf people. However, a CI cannot restore all aspects of hearing. Improvement of the implant technology is needed if CI users are to perceive music and perform in more natural environments, such as hearing out a voice with competing talkers, reflections, and other sounds. Such improvement requires experimental animals to better understand the mechanisms of electric stimulation in the cochlea and its responses in the whole auditory system. The mouse is an increasingly attractive model due to the many genetic models available. However, the limited use of this species as a CI model is mainly due to the difficulty of implanting small electrode arrays. More details about the surgical procedure are therefore of great interest to expand the use of mice in CI research.
In this report, we describe in detail the protocol for acute deafening and cochlear implantation of an electrode array in the C57BL/6 mouse strain. We demonstrate the functional efficacy of this procedure with electrically-evoked auditory brainstem response (eABR) and show examples of facial nerve stimulation. Finally, we also discuss the importance of including a deafening procedure when using a normally hearing animal. This mouse model provides a powerful opportunity to study genetic and neurobiological mechanisms that would be of relevance for CI users.
Cochlear implants (CIs) are electronic devices that can provide a sense of hearing to people with severe and profound hearing loss. It uses electrodes surgically implanted in the cochlea of the inner ear to directly stimulate the auditory nerve. To date, the CI is the most successful sensory prosthesis and has helped more than 600,000 people worldwide1. However, the device has shortcomings. First, the benefits provided by the device vary greatly among recipients. Second, speech in noisy environments and music are still poorly perceived by most CI users.
For many years, animal models have been used to better understand these issues in CI research and to continuously improve safety and efficacy of the devices. The models have given valuable insight into several phenomena, such as plastic changes in the brain taking place following CI implantation2, the effect of applying gene therapy to preserve residual hearing3, and biophysical properties of the electrically stimulated auditory nerve4, among many other examples.
Mice are a powerful model organism due to the large availability of genetic models of deafness. Other advantages include the ability to manipulate the mouse genome (e.g., via the CRISPR-Cas system), the opportunity to use advanced imaging techniques to study mechanisms, particularly in the brain, the high reproduction rate, rapid development and easy breeding and handling. The main technical challenges in performing CI surgeries in mice are the small size of the cochlea and the presence of a large stapedial artery (SA). The SA usually disappears during embryonic development in humans but persists throughout life in a number of rodents, including mice, rats, and gerbils. The SA runs below the round window niche, which complicates access to the cochlea and increases surgical risk.
Previous studies have shown the feasibility of CI implantation in mice5,6,7. Irving et al. demonstrated that chronic intracochlear electrical stimulation can be achieved for up to one month. Acute stimulation was also performed but the recordings were not presented. They showed that cauterizing the stapedial artery had no significant effect on the hearing threshold or the number of spiral ganglion neurons and that topical application of the aminoglycoside neomycin, an ototoxic drug, was an effective deafening procedure in mice5. Soken et al. described a modified dorsal approach to the mouse cochlea through the round window to better preserve hearing status6. Following insertion of a platinum-iridium wire, substantial residual hearing was observed with an increased auditory brainstem response (ABR) threshold of 28 dB. Otoacoustic emissions (OAE) were lost in animals with large ABR threshold shifts6. Mistry et al. tested the functional and histopathological effects of implantation in the absence of electric stimulation7. Even though hearing was preserved in both 3 and 6 months-old implanted mice at low frequencies, implantation resulted in fibrosis-like tissue around the implant and osteoneogenesis around the bullostomy7.
In short, out of the three studies on CIs in mice, only one demonstrates functional recording of CI stimulation. Irving and colleagues performed both acute and chronic eABR recordings but only showed data from chronic CI stimulation5. However, the chronic model with a fully implantable device developed by Irving et al. is technically challenging. It is not yet known if acute CI stimulation, both less challenging and faster, can achieve similar results.
CIs are used by people with severe and profound hearing loss who no longer benefit from hearing aids. Animal models for CI users should therefore include a deafening procedure when normally hearing animals are used. Another reason to deafen hearing animals is that the electrical stimulation of a deaf or hearing cochlea produces different neural responses4,8,9,10,11,12. Electrical stimulation of a deaf cochlea directly activates the auditory nerve fibers and generates an electroneural response (α). It is characterized by short latency and a small dynamic range in the periphery8,10. On the other hand, electrical stimulation of a hearing cochlea also excites the hair cells in an electrophonic response (β) that is characterized by longer latencies and larger dynamic range4,11. The electrophonic response is attributed to normal excitation of nerve fibers by inner hair cells, electrically induced contraction of outer hair cells, and generation of a travelling wave4. Electroneural and electrophonic responses also result in two different activity patterns in the central nervous system9. Sato et al. recorded midbrain neurons of a CI implanted guinea pig before and after deafening with neomycin, which eliminates the electrophonic contribution. They showed that the slope of the rate-level function was steeper and firing rates higher in the deafened condition compared to the hearing condition9. Therefore, depending on the research question stated, it is important to consider including deafening to separate electrophonic and electroneural responses upon electric stimulation of the auditory nerve.
Here, we describe the procedure for acute deafening and the cochlear implantation of an electrode array in a mouse as well as the functional recording of intracochlear electric stimulation with electrically-evoked auditory brainstem response (eABR).
All procedures were carried out according to Basel University, Switzerland, animal care and guidelines. They were licensed by the Veterinary Office of the Canton of Basel, Switzerland.
NOTE: C57BL/6 adult mice, aged 8–12 weeks (weight 20–30 g), were used in this study.
The left ear is used as the experimental ear. The right ear serves as an intra-animal control and is not surgically altered.
1. Preoperative Procedures
2. Pre-deafening Acoustic Auditory Brainstem Response (aABR)
NOTE: aABR is used to measure the status of hearing before and after deafening. Testing is performed on the left ear and in a soundproof electrically shielded booth. We recommend to test and later implant the left ear for a right-handed person. Further details on ABR in mice can be found in13,14. Tucker Davis Technologies (TDT) hardware and software (BioSig) are used to record ABR but other systems can be used.
NOTE: Typical instruments used include a scissor, a scalpel, a pair of metallic forceps with straight or curved tips, a tissue retractor tool, several suction wedges and absorbable paper points. The surgery is performed on the left ear.
4. Round Window Application of Ototoxic Agent
5. Post-deafening Acoustic ABR
6. Insertion of CI Electrode Array
NOTE: The intracochlear electrode array consists of four platinum bands (Ø0.2 mm) with platinum/iridum parylene insulated wire shielded in a silicone tube (Figure 5).
7. Electric Auditory Brainstem Response (eABR)
NOTE: An Animal Stimulator Platform (ASP) is used to electrically stimulate the electrode array. Other current sources and software systems can be used.
8. End of Experiment
The purpose of this study was to describe a reliable model for acute CI stimulation in the deafened mouse. Pre- and post-surgical hearing thresholds served as a functional readout of the deafening procedure. Topical application of 0.05% neomycin in the oval and round window significantly increased click-evoked hearing thresholds by 46 dB ± 6 (pre- vs post-neomycin: 30.0 dB ± 3.8 vs 75.7 dB ± 3.7, p = 0.0003, paired t-test, n= 7) (Figure 3). The mouse-sized electrode array was hereafter inserted into the round window (Figure 4, Figure 5). Electric simulation of an intracochlear electrode could reliably generate eABR activity. (Figure 7). In some cases, CI stimulation activated the facial nerve and produced a high amplitude wave with either short or long latency (Figure 8A and Figure 8B, respectively). The short latency response was characterized by a rapid amplification of wave IV around 3 ms and is likely to be a direct response of the facial nerve. The long latency response appeared around 5–6 ms and is likely to be a non-auditory muscle (myogenic) response evoked indirectly by the facial nerve. Facial nerve responses are rarely reported in animal studies in literature but is a well-known complication in human CI users17,18,19. In Figure 8, facial nerve stimulation appeared at relatively medium current levels (150–200 μA) and in two different animals. In other cases, both responses could appear in the same animal at very high current levels (not shown). We recommend limiting the current level to levels below the appearance of facial nerve stimulation.
Figure 1: Auditory Brainstem Response (ABR) setup. Subdermal electrodes are placed at the vertex (active/channel 1 [Ch1]), behind the ipsilateral ear (reference [Ref]) and at the hind leg (ground [Gnd]) of the anesthetized mouse. Electrode signals are amplified and then recorded by a TDT system. Acoustic and electric stimulation are presented via a microphone and an Animal Stimulator Platform, respectively. Please click here to view a larger version of this figure.
Figure 2: Representative aABR waves to click stimulation from a wild-type mouse before and after deafening with 0.05% neomycin. (A) The normal-hearing aABR pattern is characterized by waves labelled I-V and a low hearing threshold, here 30 dB SPL (arrow). (B) The deafened aABR pattern shows an increased hearing threshold, here 70 dB SPL (arrow). The waves have a longer latency and more temporal jitter. Please click here to view a larger version of this figure.
Figure 3: aABR threshold before and after deafening. Application of neomycin significantly increased aABR thresholds by 46 dB ± 6. Pre- vs post-neomycin: 30.0 dB ± 3.8 vs 75.7 dB ± 3.7, p = 0.0003, paired t-test, n = 7. Errors are standard error of the means. Please click here to view a larger version of this figure.
Figure 4: The surgery. (A) Exposure to the auditory bulla. The bullostomy is performed (white dotted circle) along the ridge on the tympanic bulla (black dotted line). (B) The bullostomy allows visualization of the round window, stapedial artery, and oval window. Neomycin is gently flushed through first the oval window, then the round window. (C) The electrode array is inserted until the 4th electrode is located just inside the round window niche. The electrode wire is coiled inside the bulla to keep the array in place before the incision is closed. CN VII = cranial nerve VII (facial nerve), OW = oval window, RW = round window, SA = stapedial artery, SCM = sternocleidomastoid muscle, TB = tympanic bulla. Please click here to view a larger version of this figure.
Figure 5: The mouse cochlear implant. (A) The intracochlear electrode array consists of four platinum bands spaced at a 0.4 mm interval with a diameter d: 0 [tip] (d=0.21), 1 (d=0.23), 2 (d=0.25), 3 (d=0.27). The width of each electrode is 0.2 mm. The four platinum/iridium (90/10) parylene insulated wires are shielded in a silicone tube. (B) Magnification of the electrode array tip (red dotted square). The electrode array and a platinum reference ball are connected to a print board. Scale bar = 1 mm. Please click here to view a larger version of this figure.
Figure 6: Electrically-evoked ABR (eABR) setup. The CI platinum ground ball (Gnd, red) is placed in a subcutaneous pocket in the neck of the mouse. The line between active (Ch1(+) at the vertex) and reference (Ref (-) at the ipsilateral ear) ABR electrodes is perpendicular to the line between the electrode array and the ground in order to obtain the best eABR response. The eABR ground electrode (Gnd, black) is placed in the hind leg. Please click here to view a larger version of this figure.
Figure 7: Representative eABR waves to CI stimulation in a deafened mouse. A biphasic pulse train is presented to electrode #1 in monopolar configuration at 23.3 pulses per second (pps) with 400 repetitions. Stimuli level 0-175 μA is shown in 25 μA steps (see stimulation details in Step 7.2). Roman numerals denote eABR wave number. The wave amplitudes and latency increase and decrease, respectively, with increasing current level. In this example, wave II appeared around 1 ms, wave III around 2 ms, wave IV around 3 ms, wave V around 4 ms. Please click here to view a larger version of this figure.
Figure 8: Example of facial nerve stimulation. In some cases, CI stimulation can activate the facial nerve and evoke a direct response with short latency (A) (arrow) or indirect response with longer latency (B) (arrow). The examples shown are from two CI-implanted animals stimulated with a biphasic pulse train using 0–300 μA in 50 μA steps (see stimulation details in Step 7.2). Roman numerals denote eABR wave numbers. * denotes clipping of the eABR wave due to saturation of the amplifier. Please click here to view a larger version of this figure.
This manuscript describes the surgical approach for acute deafening and cochlear implantation in the mouse, as well as the functional assessment of CI stimulation with auditory brainstem response. Although the mouse cochlea is small and the surgery challenging, the CI mouse model is feasible and serves as a valuable tool in auditory research.
The stapedial artery is present in the middle ear of the mouse. The artery enters the bulla posterior-medially, and runs inferiorly to the round window niche and then superiorly to the oval window niche. In the initial development of the mouse model, we experienced fatal intra-operative bleeding following trauma to the stapedial artery, mainly while accessing the bulla. As a consequence, we adapted a more limited approach and accessed the bulla in smaller, refined dissection steps. No further complications due to bleeding were thereafter observed. Despite the fact that stapedial artery cauterization has no significant effect on hearing threshold or number of spiral ganglion neurons in mice5, in our opinion, it is unnecessary so long as great care and attention are taken during the surgery. We suggest taking the time needed to develop fine psychomotor skills and reach technical proficiency. The average time from initial incision to closure around the implanted electrode array is typically 1–1.5 h.
The described acute CI surgery in mice is similar to the "ventral" procedure and round window insertion used in other rodents, including rats and gerbils20,21,22. Other rodent studies have used the "dorsal approach" with a basal turn cochleostomy instead of a round window insertion, avoiding the SA entirely and inserting the array more deeply6,23,24. The implantation of a chronic stimulation assembly in mice follows the same steps as described in this protocol with the addition of a Dacron mesh to fix the implant and postoperative care5.
The main technical challenges when performing CI surgeries in mice are the small size of the cochlea compared to the cochlea of rats and gerbils, and the presence of a large SA. The SA is also present in rats but not in gerbils. In addition, since mice are smaller than rats and gerbils, they are more vulnerable to surgical procedures.
To eliminate electrophonic responses in eABR recordings and to mimic hair cell loss found in most CI users, we deafened the animals before CI insertion. Mice are difficult to deafen ototoxically in vivo25 because the systemic concentrations of aminoglycosides required to cause ototoxicity has a narrow dose window: lower doses given over several days results in no hair cell loss whereas a single injection of a higher dose can be lethal26. Also, susceptibility to aminoglycosides is strain dependent26. However, it has been shown that a single dose of aminoglycosides in combination with a loop diuretic can produce excessive outer hair cell loss in CBA/CaJ mice without fatal consequences27. Delayed inner hair cell death was reported in half of all cochleae examined27.
In this manuscript, we used topical application of the aminoglycosides neomycin inspired by the protocol recently established for C57BL/6 mice5. Acute application of neomycin significantly increased the click-evoked hearing threshold by 46 dB ± 6.1. Although this increase is larger than the 35 dB increase reported by Irving et al. (pre- vs post-surgery: 41.6 dB ± 3.3 vs 76.6 dB ± 4.4, p = 0.02, n = 3)5, we achieved the same post-deafening threshold (75.7 dB ± 3.7 vs 76.6 dB ± 4.4). 0.05% neomycin is thought to cause a partial loss of hearing, mainly by rapid outer hair cell death, as inner hair cell loss takes longer to occur27. It is therefore possible that electrophonic response, which is generated both by inner and outer hair cells4,8,9,10,11,12, is only partially eliminated in deafened animals with residual hearing. Even though 0.05% (weight/volume) neomycin does not decrease the number of spiral ganglion neurons 4 weeks post-deafening5, it is yet unknown if neomycin in our acute setup affects the auditory nerve fibers or promotes synaptopathy (loss of synapses between inner hair cells and type I auditory nerve fibers). Another uncertainty is that the topical neomycin treatment may not produce a uniform distribution of hair cell loss along the length of the cochlea. Future studies are required to answer these questions.
In summary, the growing number of genetic models for human deafness and the biochemical tools available make the mouse an attractive animal model for auditory research, including the field of CIs.
No competing financial interests. The authors have nothing to disclose.
The authors would like to thank Pierre Stahl, Oticon Medical, Nice, France, for providing the Animal Stimulation Platform and advice on stimulation paradigms, and James B. Fallon and Andrew K. Wise from Bionics Institute, Melbourne, Australia, for surgical advice. This work was supported by a grant from the Swiss National Science Foundation (ERC transfer grant to T.R.B.).
|IAC Acoustics, Winchester, UK
|Mac-2 Enclosure RF Shielded Box 2A
|Tucker Davis Technologies (TDT), FL, USA
|PCB Piezotronics, Inc, NY, USA
|Low impedance headstage
|TDT, FL, USA
|TDT, FL, USA
|RZ6 auditory processor
|TDT, FL, USA
|Animal Stimulator Platform
|ASP, Oticon Medical, Nice, France
|Closed-loop heating pad
|FHC, Inc. ME, USA
|Otoform Ak, Dreve Otoplastik GmbH
|Disposable subdermal needle electrodes
|Horizon, Rochester Electro-Medical Inc.
|Self-retaining retractor tool (Mini Collibri Retractor)
|Fine Science Tools
|Absorbable paper point (Medium)
|WPI, FL, USA
|Intracochlear electrode array
|Bionics Institute, Melbourne, Australia
|Ferrosan Medical Devices
|Tissue glue. Loctite 4161 Superbond
|Part No 19743
|Animal Stimulator Platform (ASP)
|Oticon Medical, Nice, France
|Provet AG, CH
|Provet AG, CH
|Inj Diss 2%, # 1315
|Bupivacain Sintetica inj Diss 0.5%
|Atropine (Atropinesulfat Amino)
|Amino AG, CH
|Neomycin (Neomycin trisulfate salt)
|TDT, FL, USA
|MathWorks, MA, USA
|Oticon Medical, Nice, France
Copyright © 2024 MyJoVE Corporation. All rights reserved