The Role of Potassium Recirculation in Cochlear Amplification

Home , Bartter syndrome, Endolymph, GJB6, Jonathan Ashmore, KCNJ10, KCNQ4

The role of potassium recirculation in cochlear ampliﬁcation Pavel Mistrika and Jonathan Ashmorea,b

aUCL Ear Institute and bDepartment of Neuroscience, Purpose of review Physiology and Pharmacology, UCL, London, UK Normal cochlear function depends on maintaining the correct ionic environment for the Correspondence to Jonathan Ashmore, Department of sensory hair cells. Here we review recent literature on the cellular distribution of Neuroscience, Physiology and Pharmacology, UCL, Gower Street, London WC1E 6BT, UK potassium transport-related molecules in the cochlea. Tel: +44 20 7679 8937; fax: +44 20 7679 8990; Recent findings e-mail: [email protected] Transgenic animal models have identified novel molecules essential for normal hearing Current Opinion in Otolaryngology & Head and and support the idea that potassium is recycled in the cochlea. The findings indicate that Neck Surgery 2009, 17:394–399 extracellular potassium released by outer hair cells into the space of Nuel is taken up by supporting cells, that the gap junction system in the organ of Corti is involved in potassium handling in the cochlea, that the gap junction system in stria vascularis is essential for the generation of the endocochlear potential, and that computational models can assist in the interpretation of the systems biology of hearing and integrate the molecular, electrical, and mechanical networks of the cochlear partition. Such models suggest that outer hair cell electromotility can amplify over a much broader frequency range than expected from isolated cell studies. Summary These new findings clarify the role of endolymphatic potassium in normal cochlear function. They also help current understanding of the mechanisms of certain forms of hereditary hearing loss.

Keywords cochlea, cochlear ampliﬁer, computational modeling, endolymph, gap junctions, hair cells, potassium, stria vascularis, transporters

Curr Opin Otolaryngol Head Neck Surg 17:394–399 ß 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins 1068-9508

Energy demands are made on stria vascularis instead. An Introduction additional explanation may be that the great diversity of Potassium (Kþ) is the major ion of the inner ear endo- potassium channel proteins means that hair cells can lymph and fills the entire scala media of the cochlea. The employ multiple strategies to extrude Kþ.Kþ channels rest of the cochlea contains perilymph, a conventional are certainly used in ionic pathways in hair cells [2,3]. extracellular fluid. Why does scala media contain high However, when Kþ does enter the cell, the resulting Kþ? With low levels of sodium and calcium, endolymph depolarization leads to force generation by the outer hair resembles intracellular medium. Its composition is main- cells (OHCs; reviewed in [4–6]) and activates transmitter tained by the correct operation of the stria vascularis release form inner hair cells onto the afferent nerve. We along the full length of the cochlea. It has long been a shall concentrate here on the Kþ flow and OHC mech- puzzle why scala media contains this high Kþ solution. In anisms. this article, we show that the cochlear design takes advantage of many different possible Kþ permeant path- Cochlear amplification is shorthand for the processes that ways in order to achieve its sensitivity to sound. inject power into the partition and increase auditory sensitivity by a factor of more than 100 (40 dB). The mechanism can be traced to the correct operation of the Hair cells and cochlear amplification OHCs and, in turn, to the insertion into the OHC The mechanotransducing surfaces of hair cells face the basolateral membrane of multiple copies of a protein same high Kþ composition fluids in both vertebrate and, SLC26A5 (prestin). There are arguments that these probably, in invertebrate systems too [1]. During sound forces could originate from the stereociliary bundle itself stimulation, Kþ becomes the charge carrier entering [6–8], but such mechanisms depend primarily on the through the transducer channels. A partial explanation entry of calcium, rather than Kþ, through the transducer for a high endolymphatic Kþ is that transduction does not channels. Prestin, acting as a voltage-gated mechano- thereby place a high metabolic load on the hair cells. enzyme, depends on the potentials across the OHC

1068-9508 ß 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins DOI:10.1097/MOO.0b013e328330366f Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Potassium recycling in cochlear amplification Mistrik and Ashmore 395 basolateral membrane. As a result, both intracellular and The permeabilities of connexin hemichannels in normal extracellular potentials determine the operation of the physiological conditions are unclear. amplifier. Both of these potentials depend on the distribution of Kþ within the organ of Corti. Despite uncertainty about the exact Kþ pathway, the candidates for its transport into stria vascularis are poorly Much of what we understand of the amplification process understood. The presence of Kþ/Naþ ATPase activity and is informed by computer models of the cochlea. Despite aKþ/Naþ/2Cl co-transporter in type II fibrocytes of the many decades of effort, there is no complete agreement spiral ligament suggests that Kþ is actively transported. about the ‘best’ model, as many of the processes are Immunolocalization of other gap junctional proteins nonlinear. There is agreement that amplification depends (Cx30) between II fibrocytes with basal and intermediate on feedback operating close to instability regimes. In cells of stria vascularis [24] suggests that Kþ diffusion these cases, the precise parameters and regulation of the occurs through the connective tissue gap junction system parameters become critical. Kþ recirculation plays a into stria vascularis. The stria vascularis itself forms a significant part in this homeostasis. system of electrically isolated compartments characterized by tight junctions forming barrier between the compartments [25] and two Kþ-diffusion potentials [26]. In this R Molecular recycling pathways for K in the system, the absence of the underlying transport proteins cochlea can be responsible for a hereditary hearing loss [27]. The Several crucial potassium transport-related molecules absence of Cx30 in the stria vascularis is also reported to have been found in very specific cochlear compartments lead to hearing losses [28]. In this latter case, it has been [9,10]. The compartments include hair cells and sup- proposed that disruption of this gap junction protein leads porting cells in the organ of Corti, type II fibrocytes in the to the endothelial barrier of the capillaries supplying the spiral ligament, and the intermediate and marginal cells stria vascularis to becoming leaky; the resulting intrastrial of stria vascularis. This expression pattern strongly sup- electric shunt would then be sufficient to reduce any ports the idea that potassium is moved between cochlear epithelial potentials significantly during development. compartments during sound transduction (reviewed in [11,12]). The endolymphatic compartment is held at a positive potential of at least þ80 mV. This endocochlear potential The recycling hypothesis proposes that Kþ enters the is generated by a large Kþ-diffusion potential generated OHCs and IHCs through the apical mechanotransducer across the apical membrane of intermediate cells of the channels. Intracellular Kþ buildup is prevented by the stria vascularis (reviewed in [12]). The major molecular activation of Kþ conductances (KCNQ4 [2], SK, and BK component responsible is inward rectifier channel [13] channels) localized around the OHC basal pole KCNJ10 (Kir4.1) [29,30]. The low Kþ levels (5 mmol/ [2,14]. The subsequent pathway for Kþ ions is less clear. l) in the intrastrial space between the marginal and inter- In one classical scenario, Kþ ions diffuse extracellularly mediate cells is maintained by the action of a Kþ/Naþ through the tunnel of Corti to reach perilymph [15]. The ATPase, a Kþ/Naþ/2Cl co-transporter (SLC12A2) and a existence of this low-impedance pathway is supported by Cl return path at the base of the marginal cells [31]. classical measurements of current loops in the cochlea Finally, Kþ permeates into scala media through a further in vivo [16] and from current injection experiments [17]. potassium channel complex, KCNQ1/KCNE1 in the apical membrane of marginal cells [32]. An alternative proposal is that Kþ ions are taken up by neighboring Deiters’ cells. Mice lacking the K-Cl co- The resulting 150 mmol/l Kþ concentration in scala transporters Kcc4 or Kcc3, both normally expressed in media is a necessary (but not sufficient) condition for Deiters’ cells, are deaf [18,19]. From the Deiters’ cells, the þ80 mV of endocochlear potential. Indeed, during Kþ ions pass through the epithelial tissue gap junction development, high Kþ in scala media precedes the estab- system coupling supporting cells in the organ of Corti lishment of endocochlear potential by 2–3 days [33]. [20]. This gap junction system is implicated by the Even when hearing matures, normally at day P14 in extensive evidence from human GJB gene mutations the mouse, the endocochlear potential still has not and in mouse models in which targeted ablation of con- reached its final adult value of nearly 100 mV. The nexin 26 (Cx26) in the organ of Corti leads to deafness difference in the timing may be due to experimental [21]. The gap junction system represents a potentially techniques, but it seems likely that high values of endo- high-impedance pathway between cells’ in contrast to the cochlear potential require full maturation of tight elec- low-impedance extracellular pathway through the tunnel trical junctions and maturity of the pumps in the stria of Corti. It is also not known how Kþ ions are released vascularis itself. The precise estimate of the necessary from the gap junction system into perilymph, although balance can benefit from relatively simple computer connexin hemichannels may form one route [22,23]. models of stria vascularis [34].

þ R [35]. There are no significant falls of endolymphatic K Reduced K recycling and cochlear during these short experimental periods. amplification þ Although the components for K recycling are in place, Second, the current flowing through the cochlea depends direct evidence for this proposal is still missing. In the on the effective electrical impedance of the stria vascu- absence of high-resolution imaging methods sensitive laris source. Recent evidence indicates that endocochlear to ion fluxes, there are severe technical difficulties in potential can also be reduced when chloride ion flow in approaching the question experimentally. It seems stria vascularis is altered [31]. To counteract Cl trans- unlikely that electrophysiological evidence for recy- port into the marginal cells of stria vascularis through the þ cling can be bettered [16]. However, assuming K SLC12A1 co-transporter, a small b-subunit protein, bart- recycling hypothesis does open alternative ways for tin, is required to permit anion exit from the marginal þ thinking about how K flow affects the cochlear ampli- cells. Barttin is a subunit of the two types of chloride fication mechanism. There are several indirect effects channel (ClC-Ka, ClC-Kb) found in these cells. In some (see Fig. 1). mouse models of Bartter syndrome, mutations of barttin compromise both Cl re-exit and the positive endo- First, any reduction in endocochlear potential would cochlear potential, but the high endolymph Kþ is not affect the magnitude of the electrical driving force of affected. Cochlear sensitivity is reduced, but the outer the OHC mechanotransducer. The consequent reduction hair cells only degenerate over several months. in OHC depolarization and forces during a sound stimulus would affect cochlear amplification. Such The barttin-deficient mouse [31] has no distortion reductions in endocochlear potential should arise from product otoacoustic emission (DPOAE). DPOAEs are þ mutations in any of the genes encoding K transport- believed to be a good measure of the OHC function related proteins in stria vascularis, or equally from [36]. This is clearly not due to the loss of OHCs. The reduction in the energy supply for the pumps in stria driving force for Kþ entry through the mechanotransdu- vascularis (as occurs in anoxia). The magnitude of the cer channel in the absence of endocochlear potential is effect is difficult to predict. It is known that relatively reduced by less than half as the resting potentials of short (2–3 min) periods of anoxia can reduce endoco- OHCs seem to be still maintained by the Kþ diffusion chlear potential by over 30 mV and affect cochlear tuning potential across the OHC basolateral membrane. One possible explanation may instead be that there is an Figure 1 Cochlear KR flow and amplification interlock increased resistance to Kþ flow from stria vascularis, effectively increasing in the resistance in stria vascularis sourcing the current for the OHCs. In turn, there would Endolymph be decreased current through OHCs, resulting in smaller K flow Cochlear amplifier depolarizations and less electromotile force. The mouse model of Bartter syndrome IV [31] is thus consistent KCNQ1/KCNE1 Stereocilial displacement with the hypothesis that potassium is recycled in the KCNJ10 MC cochlea. Na/K ATPase SLC12A2 þ Basilar CIC-Ka/Kb IC SV Third, irrespective of the K concentration in scala membrane þ barttin OHC media, the outflow of K from the hair cells is likely BC mechanics to be a critical determinant of OHC function and long- Na/K ATPase Cx26 term survival. Loss of OHCs may arise either from supporting cells Cx30 intracellular potassium accumulation [37,38], in which case, the cells would swell to maintain osmotic balance, or Cx26 + Cx30 + Kcc3/4 Force f eedback from the depolarization when Kþ channels are lost. In the Perilymph latter case, intracellular calcium would rise, producing a loss of transduction sensitivity and cell death. A simplified schematic of the Kþ recirculation (left hand sectiion) from endolymph through the outer hair cells (OHCs) to perilymph. Displace- Fourth, clearance of Kþ from around OHCs [15] is likely ment of the OHC hair bundle gates current flow and generates force þ feedback (right hand section) to the cochlear partition. Each slice of the to be critical. Often described as ‘K intoxication’ of the organ of Corti serves as a recycling and force feedback section, although organ of Corti (an inappropriate anthropomorphism), Kþ further longitudinal current flow can occur through a gap junction net- buildup would depolarize the OHCs. The consequences work or the spaces of Nuel adjacent to the OHCs through supporting þ cells. The components of the stria vascularis are shown with the should be similar to K channel mutations in OHCs. The appropriate transport elements at the boundaries of the marginal cells clearance mechanisms for Kþ, however, may also involve (MC), intermediate cells (IC), and border cells (BC) [26 ,34 ]. Fibrocytes other permeable ions. As a final hypothesis, some forms of responsible for BC uptake are shown only as Na/K ATPase. deafness and sensitivity loss could result from altered

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Potassium recycling in cochlear amplification Mistrik and Ashmore 397 intercellular flow of metabolites and secondary messen- computational biology using in-silico cochlear models. gers between cells during the embryogenesis of the Indeed, the first attempt to model computationally a cochlea. It has been reported that some of these com- three-dimensional ionic circulation (i.e., the electrical pounds (in particular, ATP and IP3) are not transported current) between distinct cochlear compartments was well through the gap junction protein Cx26 containing made more than 30 years ago [47]. The computational mutations of the same type as found in human deafness power now available is much greater. This allows the mutations [23,39,40]. consideration of relatively complex circuits (although still based on models of semi-isolated hair cells [48–50]) with reactive elements. The cochlear molecular network can Extracellular current flow in the organ of Corti then be represented in sufficient detail to compute the Defects in Kþ circulation are expected to affect prestin- time dependence of the response to transient sounds and driven OHC electromotility if the potential across the tone bursts [51,52]. OHC lateral membrane were to be altered. Prestin is driven not simply by the intracellular receptor potential As an example, a recent model of cochlear current flow but by the transmembrane potential (the difference based on guinea pig cochlear data predicts a low attenu- between intracellular and extracellular potentials). Such ation of the OHC transmembrane potential in the 0.25– a scenario seems very likely if connexin mutations were 30 kHz range (6 dB per decade) [51], provided an to reduce the conductivity of gap junctions in the organ of experimentally observed gradient in the OHCs conduc- Corti (reviewed recently in [41,42]). In this case, the Kþ tances from cochlear base to apex is included. In other ion becomes simply a charge carrier and determines the words, larger currents in basal (high-frequency) OHCs extracellular field around the cells. are important for balancing the low-pass filtering property of the OHC basolateral membrane [53,54]. Low levels of In support of this proposal, it has been inferred that OHC attenuation with frequency suggests that the prestin- electromotility is altered in humans with Cx26 mutations. driven amplification process could operate in much Homozygous individuals carrying the 30delG mutation in broader frequency range than has been inferred from GJB2 gene (encoding human Cx26) lack a DPOAE [43]. isolated OHCs studies. Similarly, the 167delT allele of the GJB2 gene results in an alteration of DPOAE [44]. Thus, at least some con- Such models are sufficiently detailed, allowing the phys- nexin mutations may interfere with the OHC amplifi- iological effects of selected mutations of transport and cation processes. gap junction proteins on the Kþ currents to be studied. Computational analysis can indicate whether such Connexin mutations do not necessarily change only the mutations do significantly reduce the OHC transmem- gap junction conductance: they might also have their brane potential and hence reduce electromotility. How- effect by altering the mechanics of the cochlear partition. ever, there remain disagreements about how outer hair The human mutation R75W in the Cx26 gene causes a cell forces affect cochlear mechanics (see Gopfert paper, malformation of the tunnel of Corti in an animal model this issue). Some of the features may account for the [45]. In addition to an increased impedance of the absence of DPOAEs in mice and humans. A compu- electrical junctions between cells in the organ of Corti, tational approach is, therefore, a fruitful way of integrat- these cochleas thus lack any extracellular (low-impe- ing a growing body of experimental data on individual Kþ dance) pathway through the tunnel of Corti. A prelimi- transport-related molecules into a systemic description of nary report suggests that the OHCs in this mutant are Kþ homeostasis in the cochlea. intact, with unaffected cellular physiology, but the DPOAEs are much reduced [46]. The observations also suggest that the R75W Cx26 mutation could be associ- Conclusion ated with reduced OHC electromotility. However, it is The idea of recycling of Kþ in the cochlea is based on equally likely that the DPOAE is affected because there the cellular localization of critical Kþ transport-related is an altered potential across the OHC lateral membrane molecules in different cochlear compartments. This or there is mechanical hindrance in the collapsed organ recycling is almost certainly critical for the efficiency of Corti. of cochlear amplification based on OHC somatic electromotility but still lacking are dynamic measurements of this flow. However, mutations in genes encoding proteins R Modeling K circulation in the cochlea involved in the Kþ recycling appear to reduce the pot- The picture painted above of global potassium recircula- assium flow through the network. The OHC transmem- tion in the cochlea is difficult to investigate experimen- brane potential is always affected. Thus, the mechanical tally due to complexity of the underlying molecular feedback implied in cochlear amplification is perturbed. network. However, it can be approached by means of OHC operation can be seen as a sensitive functional

indicator of mutations in Kþ transport-related molecules 22 Zhao HB, Yu N, Fleming CR. Gap junctional hemichannel-mediated ATP release and hearing controls in the inner ear. Proc Natl Acad Sci U S A 2005; in the cochlea, of the absence of endocochlear potential, 102:18724–18729. þ and of K buildup in the organ of Corti itself. 23 Anselmi F, Hernandez VH, Crispino G, et al. ATP release through connexin hemichannels and gap junction transfer of second messengers propagate Ca2þ signals across the inner ear. Proc Natl Acad Sci U S A 2008; 105: 18770–18775. Demonstrates that connexin hemichannels can promote ATP release and sus- Acknowledgement 2þ The study was supported by EU FP6 integrated project LSHG-CT- tained long-range intercellular Ca propagation. Using imaging methods com- 20054-512063, EuroHear. bined with focal delivery of ATP or photostimulation with caged IP(3), calcium signals are studied in organotypic cochlear cultures. 24 Teubner B, Michel V, Pesch J, et al. Connexin30 (Gjb6)-deficiency causes severe hearing impairment and lack of endocochlear potential. Hum Mol References and recommended reading Genet 2003; 12:13–21. Papers of particular interest, published within the annual period of review, have 25 Gow A, Davies C, Southwood CM, et al. Deafness in Claudin 11-null mice been highlighted as: reveals the critical contribution of basal cell tight junctions to stria vascularis of special interest function. J Neurosci 2004; 24:7051–7062. of outstanding interest 26 Nin F, Hibino H, Doi K, et al. The endocochlear potential depends on two Kþ Additional references related to this topic can also be found in the Current diffusion potentials and an electrical barrier in the stria vascularis of the inner World Literature section in this issue (p. 416). ear. Proc Natl Acad Sci U S A 2008; 105:1751–1756. A study using multibarrel electrodes to demonstrate the resistance and Kþ content 1 Steinbrecht RA. Ions and mucoid substances in sensory organs: microana- of stria vascularis compartments. Highlights the role of the tight junctions between lytical data from insect sensilla. Symp Soc Exp Biol 1989; 43:131–138. stria vascularis cells in the generation of the endocochlear potential and proposes 2 Kharkovets T, Dedek K, Maier H, et al. Mice with altered KCNQ4 Kþ channels an origin for endocochlear potential based on the Kþ diffusion potentials. implicate sensory outer hair cells in human progressive deafness. EMBO J 27 Jin Z, Ulfendahl M, Jarlebark L. Spatiotemporal loss of Kþ transport proteins in 2006; 25:642–652. the developing cochlear lateral wall of guinea pigs with hereditary deafness. 3 Marcotti W, Johnson SL, Holley MC, Kros CJ. Developmental changes in the Eur J Neurosci 2008; 27:145–154. expression of potassium currents of embryonic, neonatal and mature mouse 28 Cohen-Salmon M, Regnault B, Cayet N, et al. Connexin30 deficiency causes inner hair cells. J Physiol 2003; 548:383–400. intrastrial fluid-blood barrier disruption within the cochlear stria vascularis. 4 Dallos P. Cochlear amplification, outer hair cells and prestin. Curr Opin Proc Natl Acad Sci U S A 2007; 104:6229–6234. Neurobiol 2008; 18:370–376. 29 Marcus DC, Wu T, Wangemann P, Kofuji P. KCNJ10 (Kir4.1) potassium 5 Ashmore J. Cochlear outer hair cell motility. Physiol Rev 2008; 88:173–210. channel knockout abolishes endocochlear potential. Am J Physiol Cell Physiol 6 Hudspeth AJ. Making an effort to listen: mechanical amplification in the ear. 2002; 282:C403–C407. Neuron 2008; 59:530–545. 30 Wangemann P, Itza EM, Albrecht B, et al. Loss of KCNJ10 protein expression 7 Chan DK, Hudspeth AJ. Ca2þ current-driven nonlinear amplification by the abolishes endocochlear potential and causes deafness in Pendred syndrome mammalian cochlea in vitro. Nat Neurosci 2005; 8:149–155. mouse model. BMC Med 2004; 2:30. 8 Martin P, Hudspeth AJ. Active hair-bundle movements can amplify a hair cell’s 31 Rickheit G, Maier H, Strenzke N, et al. Endocochlear potential depends on Cl- response to oscillatory mechanical stimuli. Proc Natl Acad Sci U S A 1999; channels: mechanism underlying deafness in Bartter syndrome IV. EMBO J 96:14306–14311. 2008; 27:2907–2917. Identifies the role of barttin, a subunit of expressed ClC-Ka and ClC-Kb channels in 9 Brown SD, Hardisty-Hughes RE, Mburu P. Quiet as a mouse: dissecting the the marginal cells of stria vascularis, as being essential for the generation of the molecular and genetic basis of hearing. Nat Rev Genet 2008; 9:277–290. endocochlear potential. Without this chloride return path, the reduction of en- A comprehensive and up-to-date review of the gene mutations affecting cochlear docochlear potential leads to deafness. function in mouse models of hearing and deafness. 32 Knipper M, Claussen C, Ruttiger L, et al. Deafness in LIMP2-deficient mice 10 Lang F, Vallon V, Knipper M, Wangemann P. Functional significance of due to early loss of the potassium channel KCNQ1/KCNE1 in marginal cells channels and transporters expressed in the inner ear and kidney. Am J Physiol of the stria vascularis. J Physiol 2006; 576:73–86. Cell Physiol 2007; 293:C1187–C1208. 33 Yamasaki M, Komune S, Shimozono M, et al. Development of monovalent ions 11 Hibino H, Kurachi Y. Molecular and physiological bases of the Kþ circulation in the endolymph in mouse cochlea. ORL J Otorhinolaryngol Relat Spec in the mammalian inner ear. Physiology (Bethesda) 2006; 21:336–345. 2000; 62:241–246. 12 Wangemann P. Supporting sensory transduction: cochlear fluid homeostasis 34 Quraishi IH, Raphael RM. Generation of the endocochlear potential: a and the endocochlear potential. J Physiol 2006; 576:11–21. biophysical model. Biophys J 2008; 94:L64–L66. 13 Marcotti W, Johnson SL, Kros CJ. Effects of intracellular stores and extra- Develops a simple mathematical model of the generation of the endocochlear cellular Ca2þ on Ca2þ-activated Kþ currents in mature mouse inner hair cells. potential in stria vascularis. The model captures critical biophysical interactions J Physiol 2004; 557:613–633. between distinct stria vascularis compartments. 14 Winter H, Bragg C, Zimmerman U, et al. Thyroid hormone receptor alpha1 is a 35 Russell IJ, Cody AR, Richardson GP. The responses of inner and outer hair critical regulator for the expression of ion channels during final differentiation cells in the basal turn of the guinea-pig cochlea and in the mouse cochlea of outer hair cells. Histochem Cell Biol 2007; 128:65–75. grown in vitro. Hear Res 1986; 22:199–216. 15 Johnstone BM, Patuzzi R, Syka J, Sykova E. Stimulus-related potassium 36 Kemp DT. Otoacoustic emissions, their origin in cochlear function, and use. Br changes in the organ of Corti of guinea-pig. J Physiol 1989; 408:77–92. Med Bull 2002; 63:223–241. þ 16 Zidanic M, Brownell WE. Fine structure of the intracochlear potential field. I: 37 Kharkovets T, Hardelin JP, Safieddine S, et al. KCNQ4, a K channel mutated The silent current. Biophys J 1990; 57:1253–1268. in a form of dominant deafness, is expressed in the inner ear and the central auditory pathway. Proc Natl Acad Sci U S A 2000; 97:4333–4338. 17 Mountain DC, Cody AR. Multiple modes of inner hair cell stimulation. Hear Res 1999; 132:1–14. 38 Mustapha M, Fang Q, Gong TW, et al. Deafness and permanently reduced potassium channel gene expression and function in hypothyroid Pit1dw 18 Boettger T, Rust MB, Maier H, et al. Loss of K-Cl co-transporter KCC3 causes mutants. J Neurosci 2009; 29:1212–1223. deafness, neurodegeneration and reduced seizure threshold. EMBO J 2003; An animal model of secondary hypothyroidism revealing novel defects of thyroid 22:5422–5434. hormone deficiency related to deafness. The defects include impaired expression 19 Boettger T, Hubner CA, Maier H, et al. Deafness and renal tubular acidosis of KCNJ10 and KCNQ4 channels and sensory and strial cells deterioration all in mice lacking the K-Cl co-transporter Kcc4. Nature 2002; 416:874– involved in K recirculation. 878. 39 Hernandez VH, Bortolozzi M, Pertegato V, et al. Unitary permeability of gap 20 Jagger DJ, Forge A. Compartmentalized and signal-selective gap junctional junction channels to second messengers measured by FRET microscopy. Nat coupling in the hearing cochlea. J Neurosci 2006; 26:1260–1268. Methods 2007; 4:353–358. 21 Cohen-Salmon M, Ott T, Michel V, et al. Targeted ablation of connexin26 in 40 Ortolano S, Di PG, Crispino G, et al. Coordinated control of connexin 26 and the inner ear epithelial gap junction network causes hearing impairment and connexin 30 at the regulatory and functional level in the inner ear. Proc Natl cell death. Curr Biol 2002; 12:1106–1111. Acad Sci U S A 2008; 105:18776–18781.

41 Nickel R, Forge A. Gap junctions and connexins in the inner ear: their roles in 48 Dallos P. Some electrical circuit properties of the organ of Corti. I: Analysis homeostasis and deafness. Curr Opin Otolaryngol Head Neck Surg 2008; without reactive elements. Hear Res 1983; 12:89–119. 16:452–457. 49 Mountain DC, Hubbard AE. A piezoelectric model of outer hair cell function. 42 Zhao HB, Kikuchi T, Ngezahayo A, White TW. Gap junctions and cochlear J Acoust Soc Am 1994; 95:350–354. homeostasis. J Membr Biol 2006; 209:177–186. 50 O’Beirne GA, Patuzzi RB. Mathematical model of outer hair cell regula- 43 Engel-Yeger B, Zaaroura S, Zlotogora J, et al. The effects of a connexin 26 tion including ion transport and cell motility. Hear Res 2007; 234:29– mutation – 35delG – on oto-acoustic emissions and brainstem evoked 51. potentials: homozygotes and carriers. Hear Res 2002; 163:93–100. 51 Mistrik P, Mullaley C, Mammano F, Ashmore J. Three-dimensional current flow 44 Morell RJ, Kim HJ, Hood LJ, et al. Mutations in the connexin 26 gene (GJB2) in a large-scale model of the cochlea and the mechanism of amplification of among Ashkenazi Jews with nonsyndromic recessive deafness. N Engl J Med sound. J R Soc Interface 2009; 6:279–291. 1998; 339:1500–1505. An in-silico model of the cochlear partition, exploring the flow of current between 45 Inoshita A, Iizuka T, Okamura HO, et al. Postnatal development of the organ of different compartments when the cochlea is stimulated by sound. By including Corti in dominant-negative Gjb2 transgenic mice. Neuroscience 2008; realistic physiological intercellular flow, the model also addresses the question of 156:1039–1047. high-frequency attenuation of OHC receptor potentials. Demonstrates the effect of human mutations in the Cx26 gene on the development 52 Ramamoorthy S, Deo NV, Grosh K. A mechano-electro-acoustical model for and mechanics of the cochlear partition. This morphological study of the organ of the cochlea: response to acoustic stimuli. J Acoust Soc Am 2007; 121: Corti in mice carrying the human GJB2 mutation R75Wþ exhibits incomplete 2758–2773. supporting cell maturation and a malformation of the spaces of Nuel. 46 Minekawa A, Narui Y, Inoshita A, et al. Cochlear outer hair cells in a dominant- 53 Mammano F, Ashmore JF. Differential expression of outer hair cell potassium negative Connexin26 mutant mouse preserved non-linear capacitance with currents in the isolated cochlea of the guinea-pig. J Physiol 1996; 496 (Pt 3): impaired distortion product otoacoustic emission. Assoc Res Otolaryngol 639–646. Abs 2009; 32:107. 54 Raybould NP, Jagger DJ, Housley GD. Positional analysis of guinea pig inner 47 Strelioff D. A computer simulation of the generation and distribution of hair cell membrane conductances: implications for regulation of the mem- cochlear potentials. J Acoust Soc Am 1973; 54:620–629. brane filter. J Assoc Res Otolaryngol 2001; 2:362–376.