The Endocochlear Potential

The endocochlear potential (EP) is the positive voltage of 80 - 100 mV seen in the endolymphatic space of the cochlea.
The EP is only found in the cochlear portion of the inner ear. The endolymphatic spaces of the saccule, utricle and semi-circular canals show much smaller resting potentials of a few mV.
Even within the cochlea the EP varies in magnitude along the length of the cochlea from base to apex. EP is highest in the basal turn and decreases in magnitude towards the apex. In the guinea pig, the EP is typically 93 mV in turn I (the basal turn), 88 mV in turn II, 82 mV in turn III and 74 mV in turn IV (the apical turn).

The EP is highly dependent on metabolism and ion transport. In the event of anoxia, an occluded blood supply or treatment with ion transport inhibitors such as ouabain, ethacrynic acid or furosemide, the EP rapidly falls and becomes negative. In the example shown (from Konishi 1979, Acta Otolaryngol 87, 506), the respirator was stopped at time zero. The EP rapidly falls from its normal value of approximately 80 mV, reaching -40 mV within approximately 10 minutes. This is a markedly different behavior than the resting potential of a cell, which may be maintained for some time when active ion transport is blocked.

How the EP is generated

The source of the EP was originally demonstrated by Tasaki & Spiropolis (1959) J. Neurophysiol 22, 149. They opened scala media at the apex and touched an electrode to different tissues that bounded it. They found that an electrode showed the highest positive voltage when contacting stria vascularis, the highly vascular tissue in the lateral wall of the cochlea.

It was concluded that the site of EP generation was the stria vascularis.

This diagram shows the more-detailed anatomy of stria vacularis. (from Kuijpers, 1969). (Key : BC=basal cell; CAP= blood capillary; IC= intermediate cell; MC=marginal cell).The stria vascularis is a complex, multilayer structure. Facing the endolymphatic space (top) is a layer of marginal cells, which are characterized by their highly folded basolateral membranes which are rich in mitochonidria. In the middle of stria vascularis are the capillaries and intermediate cells. Facing the spiral ligament (the perilymphatic side) are multiple layers of flat, interleaved basal cells. Between the adjacent marginal cells and basal cells are zonnulae occludentes (tight junctions) which are believed to limit ion movements through the intercellular spaces. There are also widespread gap junctions between basal cells and between basal cells and intermediate cells, suggesting that these cells are coupled together as a syncitium, allowing exchange of intracellular contents.
The EP is clearly an unusual potential, with characteristics unlike that of the cell resting potential or of other extracellular potentials. In the mid 1980's, there were a number of explanations for how EP was generated. All explanations centered on the EP being generated by the highly metabolically-active marginal cell. However, to explain the generation of such a high positive potential necessitated the speculation of a unique "electrogenic" transport process, i.e. some form of K transport which generated a large potential. Conventional K transporters such as the Na/K ATPase and Na/K/2Cl cotransport do not generate such large potentials. There is still no evidence that a highly electrogenic transporter is expressed in the mammalian cochlea. The two common models to explain how EP is generated can be described as the one-pump and the two-pump models

The One-pump model

In this model the positive voltage was generated by electrogenic Na/K exchange at the basolateral membrane of the marginal cell. This transport was believed to generate a positive intracellular potential and high K level within the marginal cell. Na and K permeabilities at the apical membrane were believed to allow the potential and ion content to be transmitted to the endolymphatic space.

 

The Two-pump model

This model addressed the concern that it was known that Na/K ATPases did not generate potentials of more than a few mV. The characteristics of Na/K ATPase at the basolateral membrane were well known and it was unlikely that the EP was generated there. Instead, it was postulated that a unique, electrogenic K transporter was present in the apical membrane of the cell. In this model, the cell would have high intracellular K content with a negative resting potential and the large positive voltage would be generated across the apical membrane

One major difference between the two models above concerned the resting potential of the marginal cell. In the one-pump model a positive potential would be expected, while in the two-pump model a negative potential would be expected. As a result there was considerable interest in measuring the marginal cell resting potential

Cell potentials were measured by passing microelectrodes into endolymph through the stria vascularis. Initial potential measurements indicated that the marginal cell resting potential was highly positive, which initially supported the one-pump model and excluded the two-pump model

In a study of this type performed in our laboratory (Salt et al., 1987, Laryngoscope 97, 984), we passed a double-barreled K-selective electrode through stria vascularis. With these electrodes we can simultaneously measure potential and K concentration in all the compartments the electrode passes through.

This show an example electrode track as an electrode is passed from the spiral liagament into endolymph through stria vascularis. Initially the electrode encounted 3 cells with low resting potentials and high internal K level (indicated by *). These are presumed to be basal cells. However, immediately following a basal cell, the electrode registered a high positive potential (the EP) but a low, perilymph-like K level. This region is marked by the pale-purple band. This potential was obviously the EP, but this could not be an intracellular potential because of the low K level. This was the first demonstration that the EP was present in the intrastrial space (the small extracellular space between the basal and the marginal cells, where the capillaries pass through).

On the basis of this finding, an explanation was developed in which the EP was generated not by the marginal cells, but instead by the basal cells of stria vascularis in an analogous way to that of the electroplaque in the electric eel. In this diagram, regions of high K level are shown shaded. It was suggested that a highly K-permeable apical membrane of the basal cells would generate large potential across it, which could either make the cell potential highly negative, or the extracellular space highly positive. The marginal cells then exist within this positively-polarized extracellular space. Thus the potential of the marginal cells is highly positive, but that potential is not generated across either the basolateral or apical membranes of the marginal cell.

Numerous subsequent in vitro studies of isolated marginal cells have shown that the ion transport properties of the marginal cells are extremely similar to those of the dark cells of the vestibular system (Wangemann, 1995, Hearing Research 90, 149). Many of their ion transport properties have been documented with pharmacological blocking agents. These cells maintain the high K level of the endolymph but they do not generate large potentials

On the left is shown a vestibular dark cell. Intracellular K is maintained by Na/K transport and by the Na/K/2Cl cotransporter in the basolateral membrane. A K permeability at the apical membrane allows a K-current to be driven into the endolymph. The marginal cell at the right side has identical properties to the vestibular dark cell. The positive EP is generated by an apical K conductance of the basal and intermediate cells. Basal cells, intermediate cells and fibrocytes of the spiral ligament are joined by gap junctions, which are believed to be required to facilitate K movements towards the endolymph.


Thus, the reason endolymph of the cochlea has a high potential, while endolymph of the vestibular system does not, is because of the presence of the multi-layered structure of stria vascularis, compared to the single cell layer bounding vestibular endolymph.