Marvin L. Sears
Aqueous humor provides optical transparency, structural integrity, and nutrition for the eye without the need for blood vessels that would be opaque. Ordinarily, no blood vessels are found in the cornea, lens, vitreous, or trabecular meshwork. Indeed, proportionately, the eye contains the largest avascular mass found in an organ anywhere in the body. Nutrition of these avascular structures can be accomplished perhaps by the formation and delivery of the aqueous humor. The production of the aqueous humor also generates a pressure within the eye to maintain its structural integrity and the position of the refractive surfaces of the eye relative to each other. It is largely the energy available from the cellular metabolism of the epithelia of the ciliary processes that drives the extraction and formation of the aqueous humor.
HISTORY
Questions about the intraocular aqueous fluid were asked at the turn of the nineteenth century.[1] The issue of the origin of the fluid was closely linked to whether it had a circulation. Three different theories were proposed. One view was that intraocular fluid did not flow through the eye but that the fluid was renewed by a metabolic interchange throughout the tissues of the eye. This view, derived from Ulrich's hypothesis about the seclusion of the pupil (a condition that would have precluded any flow of fluid through the pupillary aperture), was described by Hamburger between 1898 and 1924. It was elaborated by Magitot[2] in 1917 and restated by him in 1928:
|
We shall conclude then today, as in 1916, that the aqueous humor. is [not] animated by a current in the real sense of the word. It is, in its normal state, a dormant water, like that of a well, which renews itself only extremely slowly if one does not cause any withholding [of the source]. Its production and elimination are two balanced phenomena, resulting from the work of dialysis from the capillaries of the uvea and from the retina. The same membranes, those of the capillaries, produce the liquid, then reabsorb it and reproduce it. |
Seeing no readily visible outflow vessels, Magitot opposed a theory of circulation. Duke-Elder's view[3] was that there was a metabolic interchange between the intraocular fluid and blood through the capillary walls throughout all the tissues of the eye, much as Magitot said, but superimposed on this interchange was a secondary pressure-derived circulation conditioned by the pulse beat, respiratory variations, and muscular activity. He also noted a thermal circulation that added to the constant internal movement of the aqueous humor. Duke-Elder chose the best of both worlds; no circulatory thruway and yet a secondary circulation. The idea that there was no circulation of aqueous and that no secretory mechanism was required fits with the concept that aqueous is a dialysate of the plasma and that the two fluids were in a Gibbs-Donnan equilibrium.
The concept of aqueous humor as a stagnant fluid was rejected promptly when experiments using dyes and other markers proved that there was a through-and-through circulation. This view was first fully elaborated by Leber[4] in 1913, but it had been noted in experiments Lauber performed earlier with dogs. In 1901, Lauber[5] published a comparison of red blood cell counts of blood taken from the anterior ciliary veins and from the paw of the same dog. The blood from the paw contained 3.61 million red blood cells per cubic millimeter, and that taken from the ciliary veins contained between 2.68 and 2.9 million red blood cells per cubic millimeter. Lauber explained this difference between paw and ciliary vein blood cell counts by assuming a dilution of the blood in the ciliary veins by the outflow of aqueous humor into the sclerocorneal vessels that emptied into the anterior ciliary veins.
In 1921, Troncoso[6] was able to document such an outflow of a clear fluid from the eyes of living rabbits. An eye was proptosed, the sclera was cleansed, and the globe was immersed in a glass cup filled with olive oil. Minute clear droplets of aqueous collected around the cornea, and 5-10 min later new droplets appeared along the limbus. A rate of aqueous flow was calculated by this method. Seidel[7] also concluded from dye experiments, done in 1923, that a continuous outflow of fluid takes place from the anterior chamber of the living animal into the veins situated in the scleral tissue near the chamber angle. He believed that the driving force was the hydrostatic pressure difference between the anterior chamber and the veins. In any event, evidence for a circulation of aqueous was established and the theory of stagnant fluid was discarded. In human eyes, in 1942, vessels at the limbus carrying translucent fluid were observed and reported.[7a] Then Goldmann showed that these vessels carried aqueous. Fluoroscein applied topically entered these veins but when fluorescein was injected intravenously the blood vessels at the limbus were colored but not so the aqueous veins.[7b]
Clinical observations of anatomic and pressure changes in the condition of pupillary block strongly suggested that the ciliary body might be the source of the intraocular fluid. Experiments included Seidel's observations[8] that correlated mitochondrial numbers with secretory activity, whereas Ma and Pillat[8] showed the resemblance of ciliary processes to intestinal villi. Enzymes related to secretory activity were demonstrated in the ciliary epithelial cells between 1925 and 1954 by Schmelzer[8] and Noda[8] and by Friedenwald and Stiehler,[9] who showed selective staining affiities of the secretory cells. New investigations of the distribution ratio of ions and of osmotic pressure (between aqueous and plasma) further doomed the dialysis theory for the origin of aqueous humor. The way was clear for Friedenwald's hypothesis[10] that the ciliary body exhibits a one-way permeability to water and that at the membrane separating ciliary stroma from epithelium an energy source generated a transepithelial potential difference that drove the secretion of salt and water into the eye. Finally, biochemical studies supported the secretory theory of aqueous humor formation. The effect of aqueous outflow on solute concentrations, the rates of entry into and turnover of radioactive tracers into the aqueous humor, the high concentration of ascorbic acid in the posterior chamber of the eye compared with the plasma, and the low concentration of certain substances such as urea in the aqueous humor as compared with the plasma all substantiated the necessity for an active secretion from the ciliary processes to explain the fidings see Chapter 298.
THE CILIARY PROCESSES
The ciliary body is a specialized structure of the uveal tract. It has at least four functions other than the formation of aqueous humor. It (1) secretes hyaluronic acid to the vitreous, (2) plays the dominant role in accommodation, (3) influences trabecular outflow facility of aqueous by way of its meridional smooth muscle, and (4) constitutes in large part the blood-aqueous barrier.
The ciliary body extends from the ora serrata posteriorly to the scleral spur anteriorly. On sagittal section the ciliary body is seen as a triangular structure with its shortest side facing anteriorly. This anterior surface gives rise to the uveal portion of the trabecular meshwork in the angle of the anterior chamber. Also, the iris originates from its middle portion. The outer surface of the ciliary body lies against the sclera. The potential supraciliary space is continuous posteriorly with the potential suprachoroidal space (see the section on Plasmoid Aqueous and Hypotony). The ciliary body is held against the sclera by the intraocular pressure. The inner surface of the ciliary body can be divided into two regions: (1) the posterior two-thirds appears grossly smooth and is called the pars plana ciliaris, and (2) the anterior third is called the pars plicata ciliaris because ?70 radiating villiform ridges project inward (mesially) from it. These ridges are the ciliary processes (Fig. 190.1). All ridges are not the same, and in many of the individual ridges there are differences between the anterior and posterior regions (Fig. 190.2). In particular, certain more anteriorly located plicae have surfaces that are enormously elaborated, suggesting the sites for secretion and/or absorption. The anteriorly located plicated processes are complex (Fig. 190.3) and constitute an impressive secretory structure (Figs 190.4 and 190.5).
|
|
|
|
FIGURE 190.1 The ciliary processes in partial silhouette. |
|
|
|
|
FIGURE 190.2 Scan of the posterior structure of the rabbit iris shows that the physiognomy of the ciliary processes varies. Regional differences include the elaborations of the surface anteriorly, which is undoubtedly the site of secretory activity and absorptive functions. |
|
|
|
|
FIGURE 190.3 Scanning electron micrograph of ciliary processes after fracture. Note blood vessels in stroma and the epithelial elaborations. |
|
|
|
|
FIGURE 190.4 From inside the posterior chamber one can see thousands of microvilli, which are figer-like projections extending from the NPE into the posterior chamber. |
|
|
|
|
FIGURE 190.5 Scanning electron micrograph shows a double cell layer with extensive basolateral infoldings. |
The layers of the ciliary body are (1) ciliary muscle, (2) a layer of vessels and the ciliary processes, (3) basal lamina, (4) ciliary epithelium, and (5) internal limiting membrane.
The ciliary muscle resembles a right-angled triangle in anteroposterior section with the right angle facing inward. The hypotenuse of the triangle runs parallel with the sclera, and the acute posterior angle points toward the choroid. The muscle that gives shape to the ciliary body is made up of flat, unstriated muscle cells. These are organized into tightly packed bundles with each bundle containing its own nerve supply. The bundles, with interweaving connections, tend to be organized into three masses, the meridional one of which has the scleral spur as its locus of insertion. The meridional bundles are likely the most involved in the regulation of outflow of aqueous humor.
The stroma of the ciliary body is separated from the ciliary epithelium by the forward continuation of Bruch's membrane. This basal lamina, as it approaches the ora serrata, splits into an outer elastic and an inner cuticular layer separated by a layer of avascular connective tissue.
NERVE SUPPLY
The ciliary body is innervated by the long and short ciliary nerves that accompany the similarly named arteries. Anatomic studies have demonstrated cholinergic fibers along the ciliary arteries and a few in the stroma of the processes, but the physiologic significance of these is not yet proved. Scattered throughout the ciliary body are sensory nerve fibers that can be recognized by club-shaped nerve endings. It is unlikely that these are pressure-sensitive nerve endings.
In the ciliary processes, numerous nonmedullated nerve fibers can be found surrounding the smaller vessels. These are largely noradrenergic[11] and probably subserve vasomotion. Unfortunately, any effects of this dense noradrenergic innervation (Fig. 190.6) to the ciliary processes on either the vascular supply (Fig. 190.7) or the epithelia or both (Fig. 190.8) that are meaningful to the process of aqueous humor formation have not been completely disentangled from each other. Yamada[12] did demonstrate the occasional presence of intraepithelial nerve fibers within the iridial ciliary epithelium of rabbits. The fibers were found at the level of the pigmented epithelium and almost reached the apical surface. All fibers observed contain both small and large core vesicles. A predominance of dense core vesicles was found. Although few fibers were encountered, the potential neurohumoral effect of these on epithelial function may spread over a wide area because of the presence of innumerable gap junctions between the two ciliary epithelial cell layers.
|
|
|
|
FIGURE 190.6 Formaldehyde-induced fluorescence of adrenergic nerves in ciliary processes and ciliary body in the vervet monkey (Cercopithecus ethiops). Adrenergic neurons appear as bright strands and spots in stroma of processes and in plexus under epithelium of ciliary body itself. V, blood vessels; M, ciliary muscle. ×157. |
|
|
|
|
FIGURE 190.7 Ciliary stromal perivascular stromal nerve by transmission electron microscopy. |
|
|
|
|
FIGURE 190.8 Occasional intraepithelial pigmented epithelium noradrenergic nerve by transmission electron microscopy. |
VASCULAR SUPPLY
The major arterial circle of the iris forms from the long posterior ciliary arteries after their bifurcation in the anterior choroid. A functional contribution from the anterior ciliary arteries must occur at times because there are anastomoses between these arteries and the long posterior ciliary arteries. These anastomoses may be under some regulatory control. The arteries of the ciliary processes originate from the major arterial circle of the iris. In the subhuman primate and other species there appear to be three different vascular territories in the ciliary body that correspond both to regional structural differences and to the major and minor ciliary processes. In particular, the anterior parts of the processes (the plicated portion) seem to be most responsiveto vasoactive agents. The capillarity of these processes tends to be dense in the midportion of the processes while toward the apex a marginal efferent venule forms a vasculardrainage channel (Fig. 190.9). The blood vessels in the ciliary processes themselves are embedded in a loose connective tissue. The capillaries in this loose stroma are fenestrated, like those in the kidney (Fig. 190.10).
|
|
|
|
FIGURE 190.9 Ciliary process, plicated portion, in cross section, shows the double cell layer and stromal capillaries. ×320. |
|
|
|
|
FIGURE 190.10 View from inside a ciliary process. Fenestrations of a ciliary capillary (see text). |
The fenestrations may permit proteins as large as 150 000 Da to enter the stromal compartment. This anatomic feature significantly influences the process of aqueous humor formation because it permits entry of protein into the ciliary stroma, influencing the transepithelial pressure relationships by way of the effective colloid osmotic pressure.
ULTRAFILTRATION AND BLOOD FLOW
Transciliary epithelial filtration by hydrostatic pressure does not occur because the value for the transciliary epithelial difference in hydrostatic pressure is less than the oncotic pressure of the stromal tissue fluid.[13] If anything, a net movement of fluid out of the eye from the posterior chamber into the stroma would occur.[14] Aqueous humor is formed in subhuman primates at very low arterial pressures. The implication for drug therapy in elevated intraocular pressure and other ocular conditions is that no therapy can logically be considered if the cytoarchitecture and delivery of a proper aqueous are disturbed.[18]
PLASMOID AQUEOUS AND HYPOTONY
Any increase in transmural pressure across ocular blood vessels, particularly if abrupt, may lead to the production of a suprachoroidal fluid that is relatively protein free. The increase in hydrostatic pressure usually from vasodilation and/or increased blood flow may then disrupt the blood-aqueous barrier, forcing a plasmoid aqueous to enter the posterior chamber and cause a transient hypertensive state.[19] At the same time, the increase in water in the ciliary stroma will decrease the colloid osmotic pressure of that tissue relative to the blood, prompting the transient exit of water from the stroma. In addition, because the tight junctions have been ruptured by the abrupt increase in hydrostatic pressure in the ciliary stroma, the solute gradient across the epithelia will be lost. A decrease in pH of the cell may occur as well. These three conditions - a change in the colloid osmotic pressure, the loss of solute gradient, and a possible decrease in cell pH - all act to reduce secretion. The barrier may be restored at variable rates (tight junctions resynthesized) depending on the underlying pathophysiology.[20]
THE SECRETORY EPITHELIA
ANATOMY AND EMBRYOGENESIS
The ciliary epithelia as glandular epithelia are unusual (Fig. 190.11). The two layers of secretory cells are apposed apex to apex, caused by the invagination of the optic cup during embryogenesis. The pigmented cells represent the forward continuation of the retinal pigment epithelium. The nonpigmented cells represent the forward continuation of the retina. Both the double layer of ciliary epithelium and its highly convoluted geometry complicate the approach to and analysis of the basic mechanisms of aqueous humor formation. Although arguments have been proposed that the nonpigmented epithelium (NPE) is largely responsible for aqueous humor formation, it is clear that the separate layers act as a functional unit, joined apex to apex by innumerable gap junctions. Indeed, the pigmented epithelium may serve several purposes, not the least of which is as a signal transducer. From a structural point of view, the pigmented epithelium has characteristic morphologic features for 'active' cellular functions. For example, its basal as well as lateral cell surface shows complicated infoldings, thus increasing the surface area that faces the capillary bed. Freeze-fracture study of this membrane revealed closely packed intramembrane particles, which suggests the presence of multiple enzyme proteins. The cytoplasm contains abundant mitochondria and distinct Golgi apparatus. The apical cell surface extends numerous microvilli into the ciliary channels (Fig. 190.12). In addition, the apposing apical cell membranes of pigmented and nonpigmented epithelia are joined together by many gap junctions (fasciae communicantes), which indicates that these epithelia are intimately related in their function. These bridging membranous structures make the bilayer a unified functional syncytium and can permit or discourage molecular (less than ?900 kDa) and electrical communication between the cell layers dependent on their size, number, and charge (Figs 190.13 to 190.16). The gap junctions provide for (1) speed in the transmission of signal; (2) a short circuit that ensures maximum responsiveness and buffers transient signals; (3) a switch to move from a low to a high resistance state isolating any cell during metabolic damage, intense secretory activity, or perhaps during differentiation; and (4) a continuous aqueous compartment establishing the bilayer as a metabolic unit under integrated control. Changes in the permeability or conductance of the ciliary gap junctional area in the bilayer have not yet been studied. Compounds generally increasing junctional permeability, for example, cyclic adenosine monophosphate (cAMP), vasoactive intestinal peptide (VIP), protein kinase A, inositol, phosphates, and phosphodiesterase inhibitors increase junctional permeability, while substances generally decreasing junctional permeability such as anoxia, low pH, high calcium level, protein kinase C, diacylglycerol, and others do have important effects on the transciliary epithelial potential of ciliary tissue and on its transporters. The gap junction could be a site mediating some actions. It is an area for future investigation. In any event, the pigmented and nonpigmented epithelia function together as a bilayer and not separately. Indeed, electrical coupling between the two layers has been reported in the rabbit ciliary epithelium.[21,22] Anatomic as well as functional studies indicate regional variation in secretory patterns, i.e., tips versus bases, anterior processes versus posterior ones.
|
|
|
|
FIGURE 190.11 Fine structural detail of ciliary process of vervet monkey. Section is slightly tangential to the process, grazing a pigment epithelial cell, and shows a capillary within the stroma (bottom). In isotonically fixed tissue such as this, intercellular clefts of epithelium are 200-300 Å wide. ×7500. |
|
|
|
|
FIGURE 190.12 Interdigitating processes of the pigmented and NPE 'forming' a ciliary channel (space homologous to cerebrospinal fluid space). |
|
|
|
|
FIGURE 190.13 In the rabbit, the zonula occludens frequently occurs near the edge (arrowheads) between apical (AS) and lateral (LS) surfaces of the nonpigmented cells. In this instance, one or two strands sit on the cell edge; these give rise to numerous strands directed toward the cell base, which run a short course on the lateral cell surface and soon merge with a second set of circumferentially oriented strands. On the apical surface of the nonpigmented cells, gap junctions (gj) and tight junctional strands (tj) connect the nonpigmented cells to a pigmented cell. ×67 000. |
|
|
|
|
FIGURE 190.14 In the monkey (Macaca mulatta), the fracture process has exposed a broad expanse of the inner leaflet of the plasmalemma at the apex of a nonpigmented cell. Gap junctions (arrows) are numerous and appear as aggregates of intramembrane particles of varying size and shape. In places, they are associated with tight junctional strands (arrowheads). Note that the aggregates are often compartmentalized into linear domains of hexagonally packed particles, alternating with aisles of undifferentiated membrane matrix. ×49 000. Inset, The junctional particles on the inner leaflet (PF) are complemented by a hexagonal array of pits on the outer leaflet (EF) of the cell membrane. ×105 000. |
|
|
|
|
FIGURE 190.15 In the monkey (Macaca mulatta), the crystalline lattice of the junctional subunits in the gap junctions between pigmented and nonpigmented epithelial cells is often disrupted by islands of unspecialized membrane matrix (arrows). The grooves (arrowheads) at the periphery of two gap junctions represent the complementation of tight junctional strands that remain associated with the inner leaflet of the plasmalemma. ×83 400. |
|
|
|
|
FIGURE 190.16 Distribution of the intercellular junctions in the ciliary epithelium of the rhesus monkey. (a) Gap junction. (b) Punctum adherens. (c) Junctional complex consisting of zonula occludens, gap junctions, and zonula adherens. (d) Desmosome. PC, pigmented cell; NPC, nonpigmented cell. |
The actual formation of fluid by these transporting cells is perhaps best described by the standing gradient osmotic flow model originally proposed by Diamond and Bossert.[23] A standing osmotic gradient is maintained in the lateral intercellular channel beyond the tight junction, with the greatest concentration of solute occurring immediately adjacent to the tight junction. The hypertonic fluid in the proximal region of the channel forces an osmotic flux of water into the channel. As water enters, the solute concentration decreases from the proximal to the distal end of the channel when the fluid becomes isosmotic with the fluid in the posterior chamber.
METABOLISM
The active transport of solute depends on the metabolic activity of the ciliary epithelia. Active ion transport across the cell membrane utilizes the energy derived from the hydrolysis of adenosine triphosphate (ATP) to establish an electrochemical gradient. In the ciliary epithelium there is ample ATP to support Na+ secretion. Interference with oxidative phosphorylation, either by inhibiting electron transport (cyanine dyes[24]) or by directly preventing the formation of ATP (oligomycin) or by uncoupling oxidative phosphorylation from electron transfer by the use of lipid-soluble weak acids such as dinitrophenol or with cyclic polypeptide antibiotics, will reduce aqueous humor formation. Glycolysis plays only a small role.[25,26]
There is no evidence that pigmented or nonpigmented epithelia differ in their energy-yielding metabolism, although the overall activity is greater in the NPE. This fact correlates with the observation that the NPE contains a major fraction of the Na+-K+-ATPase. Finally, the rates of respiration and glycolysis of the NPE are more sensitive to variations in substrate and oxygen supply than those of pigmented epithelium.[27] These fidings taken together lead to the conclusion that the NPE plays the major role in the provision of energy for sodium transport by the ciliary epithelium.
ION AND WATER TRANSPORT
The entry rate of Na+ and HCO3? into the aqueous humor from the blood and its modification by suppressants or inhibitors has been well examined in vivo. Ouabain[28] and/or carbonic anhydrase inhibitors[29-31] inhibit the transport of these ions and water into the posterior chamber, suggesting that the ciliary epithelium can transport Na+ and HCO3? utilizing the membrane-bound enzyme systems, Na+-K+-ATPase and carbonic anhydrase. Furthermore, timolol, a potent l3-adrenergic antagonist, decreases the rate of aqueous inflow.[32-34] This particular decrease in inflow may occur without a change in the rate of Na+entry into the posterior chamber.[35] From this and many other experiments, it has become clear that the rate of formation of aqueous can be an ambiguous concept because different components are turned over at different rates. Different defiitions of 'the rate of formation of aqueous' have been used, depending on which components of the process are under examination and whether it is the secretory processper se or the issue of pressure and flow that compels the interest. The interactions are complex and are more easily addressed in vitro. The problem with in vitro studies is, with the sole exception of the demonstration of the functionally coupled ciliary epithelial beta adrenergic receptor,[36-37] identification of a signal transduction loop for ion and water movement has not been done. The same could be said of aquaporin water channels identified to regulate membrane permeability.[38] Demonstration of a functionally coupled receptor is required for such a mechanism.
IN VITRO STUDIES
Isolated preparations of ciliary processes were first made by Friedenwald.[39] In this way, vascular and other indirect effects on secretion are eliminated. Ciliary processes studied in this manner can yield significant information about receptors for secretion of aqueous humor. Planimetric measurements of photographs of optical sections of the processes may permit an estimate of the rate of shrinkage of the processes.[40,41] Shrinkage is assumed to be a measure of transport across the processes (Fig. 190.17). Another approach has been to study the uptake and accumulation of substances by the freshly excised iris and/or ciliary body and its processes.[42] Here it is hoped that the accumulation of substances by the ciliary processes in vitro represents secretion of these substances in vivo. Becker demonstrated the Na+-, K+-, Ca2+-, and glucose-dependent uptake of ascorbic acid by the ciliary process, a process that operates by saturation kinetics. Of course, accumulation does not prove that substances are transported from the plasma to the aqueous. Other substances can be moved in the opposite direction, and these may also be accumulated by the ciliary processes. Besides, the cellular and molecular sites of transfer are not known. Attempts to learn about such vectorial transport across the ciliary epithelium have been accomplished mainly with irisciliary body preparations, mounted in an Ussing type chamber.[28,43,44] A net Na+ transport occurs that is ouabain sensitive. A net Cl? flux is found in some species.[45,46] These ion fluxes represent the major source of the short circuit current across the preparation.[47] The short circuit current correlates closely with the Na+ pump and is remarkably HCO3-dependent.[43] The value of unidirectional fluxes for Na+ and Cl? is larger than that expected from the values for conductances, suggesting that electrically silent transport mechanisms are operative.[48] Again, with the iris-ciliary body preparation it turns out that the short circuit current is insufficient to explain the net movement of ions from plasma to aqueous humor. The preparation also does not transfer enough ascorbic acid to explain the in vivo fiding of 20- to 30-fold the concentration of ascorbic acid in the posterior chamber.[49]
|
|
|
|
FIGURE 190.17 Ciliary processes taken from right and left eyes of rabbits several hours after treatment with close (lingual) arterial infusion of 2 mg cholera toxin (right) and normal saline solution (left). The pigmented and nonpigmented cell layers are intact, but prominent right stromal edema indicates inhibition of net aqueous humor transport. |
In cell culture, reconstituted epithelia provide a simple system, grown as a monolayer.[50] The extent to which information derived from cultured cells can address the secretory process is not yet clear. Nonpigmented epithelial cells cannot be passed in culture unless immortalized by transformation with oncogenic viruses,[51] and then they may lose certain functions and gain others as a result of transformation. Even in primary explants, transport characteristics may be different. For example, certainly for the NPE, polarity is changed because cells in the monolayer rest on their base with the apex free, rather than the reverse. For the pigmented epithelium, the apex is exposed while the cells rest on their base, the side normally exposed to the stroma, so the asymmetries necessary for transport may not be maintained. This latter objection may be more relevant for the NPE because it has occluding junctions. Many individual transporters that have been characterized have been studied only in cultured pigmented epithelial cells. The actual participation and plasma membrane loci of the transporters in transepithelial transport is therefore speculative. Some studies use cells from eyes that don't actively form aqueous in vivo.In spite of these limitations it is important to give a short account of transporters as learned from the activity of these cultured cells.[52]
Uptake Processes in Pigmented Epithelium
In cultured bovine pigment epithelium a stilbene-sensitive, Cl?-dependent, Na+-HCO3? cotransporter has been found.[53] These cells also express a Cl?-HCO3? exchanger that may be coupled to the NPE, causing a net Cl? flux into the posterior chamber that is HCO3?-dependent.[54-56]
In the cultured bovine pigment epithelium there are at least three mechanisms for Na+ uptake: An Na+-2Cl?-K+-symport sensitive to loop diuretics, an Na+-H+ antiport (and Cl?-HCO3 exchange) system, and an electrogenic-ascorbic acid system.[57]
Uptake and Release or Secretion in NPE
Ascorbic acid probably enters the posterior chamber by facilitated diffusion, and it is the diffusion or exit rate from the NPE that is rate limiting for the accumulation of ascorbic acid by the ciliary epithelium.
K+ channels of several types are present.[58,59] They are opened by the mobilization of calcium from intracellular stores sensitive to several second-messenger systems such as acetylcholine, the phosphatidyl inositol pathway,[60] and perhaps others. These pathways may regulate sodium transport by modulation of potassium channels.
The ciliary epithelial bilayer would function as a unit in this manner.[61] It is the stromal substrate from which the pigment epithelium takes up substances. Movement across the pigment epithelium, then through gap junctions, into the NPE would occur where the fial transport into the posterior chamber would occur. Na+ entry seems coupled to other ions (a conclusion Chu and Candia[48] drew from their studies with iris-ciliary body preparations), and the Na+ gradient serves as an energy source for other ions (Cl?, HCO3?, ascorbic acid).
MEMBRANE-BOUND ENZYME SYSTEMS
ADENYLATE CYCLASE RECEPTOR COMPLEX
At the turn of the twentieth-century clinical observations indicated that shortly after either cervical sympathectomy or a conduction block of the stellate ganglion a drop in ipsilateral intraocular pressure ensued. In 1949, Friedenwald[10] enunciated further his idea that the energy of cellular metabolism is made available for the work of water transport: secretion. He asked the question, "Is there a humoral activator to the secretion of the ciliary processes, apart from vasomotor influences in ciliary capillaries and osmotic forces?" A qualitative description of his answer follows: Epinephrine, administered supravitally produced defiite, although slight, changes in tissue redox potential. Adrenalectomy was done in hopes of enhancing this effect. A dramatic reactivation was caused by epinephrine. Friedenwald introduced acidic and basic dyes into the tissues of normal animals. They were distributed with the ionic electric current; basic dyes accumulate in the epithelium and acidic dyes accumulate in the stroma. Epinephrine enhanced this process. Friedenwald concluded that epinephrine can activate intercellular oxidative exchange in the ciliary body as demonstrated by (1) transport of dyes, (2) oxidative exchange, and (3) rate of regeneration of fluid.
After a lengthy interval during which sympatholytic and mimetic agents were used in glaucoma, physiologic studies of the influence of the adrenergic nervous system on intraocular pressure began in the 1950s, including electrical stimulation of the cervical sympathetic nerve, denervation, degeneration release after denervation, and the effects of exogenously administered adrenergic agonists and blocking agents (reviewed by Sears[62]).
Study of the chemistry of the ciliary epithelial system did not begin until 1971, when a catecholamine-stimulated adenylate cyclase in ciliary processes of rabbits was demonstrated.[63] Then in vitro accumulation of cAMP in the rabbit ciliary body, stimulated by catecholamines,[64] was found. The presence of ?-adrenergic receptors in the isolated ciliary epithelia was shown,[65] among which there was a predominance of ?2-adrenergic receptors.[36] High-affiity binding sites of[125] I-hydroxybenzylpindolol were found in the same particulate membrane fractions of homogenized ciliary processes as adenyl ate cyclase activity.[37] Stimulation of adenylate cyclase activity by catecholamines was completely blocked by ?-adrenergic antagonists but not by phenoxybenzamine, an ?-adrenergic blocker. The Kd is comparable to that for ?-adrenergic receptors of other tissues. Kact, for stimulation of enzyme activity, was of the order expected for a ?-adrenergic receptor-linked adenylate cyclase. Kifor inhibition of levoepinephrine stimulation was similar to binding constants for these ?-adrenergic antagonists in other systems. Similar results were obtained in membrane preparations from other species and from human eyes.[66] The order of potency of agonist activation confirmed that the ciliary processes contain a predominance of ?2-adrenergic receptors. Finally, binding constants determined by the direct ligand-binding technique and by the assay for adenylate cyclase agree (Fig. 190.18). The agreement argues that the two techniques measure the interaction between the ?-adrenergic ligand and the ?-adrenergic receptor of the ciliary processes. Therefore, a functionally coupled ?-adrenergic receptor is located in the ciliary processes.
|
|
|
|
FIGURE 190.18 Correlation of Ki from inhibition of[125] I-hydroxybenzylpindolol binding with Kact and Ki from adenylcyclase assays. Line is line of identity. |
Cytochemical and electrophysiologic studies indicate that the adenylate cyclase receptor complex may be located in either cell layer of the ciliary processes.[67-69] Some studies indicated a predominance of adrenergic receptors in the NPE, while one electrophysiologic study indicates that such receptors are located in the pigmented epithelium but not in the NPE.[44] The precise localization of the adrenoreceptors will need to be established by biochemical studies of the separated layers of freshly isolated ciliary epithelial cells.[70]
It turns out that cAMP production by excised whole-rabbit ciliary processes is relatively insensitive to epinephrine and norepinephrine and is stimulated only at low isoproterenol concentrations.[71] These results can be explained by the interaction between ?2- and ?2-adrenergic receptors of ciliary processes and the relative affiities of isoproterenol, epinephrine, and norepinephrine for the two adrenergic receptors (Figs 190.19 to 190.21).[72] Epinephrine and norepinephrine both stimulate the inhibitory ?2-adrenergic receptors at concentrations lower than they stimulate the ?-adrenergic receptors of rabbit ciliary processes and, therefore, inhibit stimulation of cAMP production. Isoproterenol, on the other hand, stimulates ?2-adrenergic receptors at lower concentrations than it interacts with inhibitory ?2-adrenergic receptors and, therefore, stimulates cAMP production (see Fig. 190.22). The effects of catecholamines on aqueous flow in rabbits are therefore unlikely to be mediated only by the ?2-adrenergic receptors of the ciliary processes.
|
|
|
|
FIGURE 190.19 Specificity of inhibition of isoproterenol-stimulated cAMP production in ciliary processes of rabbit by the ?2-agonist clonidine. |
|
|
|
|
FIGURE 190.20 Dose-response curves for several ?-adrenergic agonists depicting relative ability to inhibit vasoactive intestinal polypeptide (VIP)-stimulated cAMP production in rabbit ciliary processes. |
|
|
|
|
FIGURE 190.21 The ?2-adrenergic specific inhibition of VIP-stimulated cAMP production in the presence of a ?-adrenergic blocker, timolol. |
|
|
|
|
FIGURE 190.22 Relationship between ?2-adrenergic stimulation and ?2-adrenergic inhibition of cAMP production in rabbit ciliary processes. The ordinate represents percent maximum adenylate cyclase (AC) activation or percent maximum stimulated cAMP production. The lines are theoretical curves generated from Kact for isoproterenol (ISO) (1.0 ?m), epinephrine (EPI) (0.4 ?m), and norepinephrine (NE) (10 ?m) stimulation of AC activity (solid lines) by the ?-adrenergic receptors in crude membrane preparations from rabbit ciliary processes. We observe no inhibition by the ?2-adrenergic receptors or from I50 for ISO (3 ?m), EPI (0.06 ?m), and NE (0.8 ?m). Inhibition of cAMP production (dotted lines) by ?2-adrenergic receptors in intact, excised ciliary processes in the presence of 3 ?m forskolin (FSK) and 10 ?m timolol (to block stimulation of ?-adrenergic receptors). |
It should be noted that comparisons between expectations from in vitro pharmacologic experiments and in vivo measurements of aqueous humor formation are not infrequently at odds.[73] Some discrepancies relate to the relatively small clinical effects, especially in the presence of a substantial error of the method (e.g., flow measurements), others to the presence of different types of receptors located near enough to each other that each is affected by tissue concentrations produced after topically administered drugs (e.g., epinephrine is both an ?2- agonist and a ?2-agonist). A third important factor is the level or 'tone' of the system at the time the experiments are accomplished. For example, it is possible that under the conditions of low tone, activation of the adenylate cyclase receptor complex stimulates aqueous humor formation, whereas at a time of high tone, either decreases (inhibition) or no effects occur after activation of the adenylate cyclase receptor complex.[74] In this connection, Brubaker and associates[75] found that timolol (for ophthalmologists, the first and classic nonselective ?-adrenergic blocker[76] decreases aqueous humor in humans only during daytime conditions when flow is relatively higher (2.5 ?L/min) than at night when flow is at an ebb (1.5 ?L/min).
How does one put together (1) the relative insensitivity of the ?-adrenergic receptor, (2) the disparate experimental results reported on aqueous humor formation with the same adrenergic compounds, and (3) the different effects of the same adrenergic compound on flow within the diurnal cycle? On a molecular level, adrenergic inhibitory effects are a function of receptor-mediated nucleotide-binding protein Gi input negatively coupled to adenylate cyclase for compounds such as ?2-agonists, ?1-adenosine agonists, and opiates or somatostatin or neuropeptide Y. Stimulatory effects of cAMP generating systems are mediated by G5 input, positively coupled to adenylate cyclase for epinephrine and its analogs. This duality, taken together with the circadian time at which experimentation is done, may explain the mixed effects reported for a given drug. It is clear from the foregoing that a significant interaction exists between the ?2 receptor and the ?2 receptor in ciliary processes.[71,77] The response of the ?-adrenergic receptor to yield increased production of cAMP is modulated by endogenous mechanisms operating through Gi and the ?2 receptor (Fig. 190.23). Other examples of ?2-adrenergic interactions with the ? receptor are discussed below.
|
|
|
|
FIGURE 190.23 Generalized adrenergic receptor schema; agonist-receptor link in ciliary epithelial adenylate cyclase receptor complex. |
VIP is a peptide with a prominent histochemical localization to the anterior uvea.[78] Indeed, VIP-stimulated cAMP formation is linked to adenylate cyclase because its action in this regard is inhibited by ?2-agonists and neuropeptide Y and somatostatin. Furthermore, binding sites specific for noradrenergic fibers to the eye have effects that are coupled negatively to the production of cAMP through Gi. The demonstrated inhibitory effect of neuropeptide Y on cAMP production by the ciliary epithelium suggests that this neuropeptide may have a dampening effect on aqueous humor formation.[80-82]
Thus, one part of the answer to the puzzle of whether aqueous humor formation is increased or decreased by cAMP is suggested by the ?2-adrenergic receptor modulation (through Gi) of the ? receptor in the ciliary processes. What about the considerable body of data supporting the idea that increased ciliary epithelial cAMP decreases aqueous humor formation? This idea was based on the powerful nonadrenergic activation of adenylate cyclase with either forskolin[83] or cholera toxin.[84] It is possible that certain of these effects are not related to changesin cAMP production but rather to prostaglandin release, yielding the thought that non-cAMP pathways may exist for these agonists ordinarily acting through cAMP, a hypothesis not inconsistent with the observations detailed earlier, namely, that ?2-adrenergic agonists decrease aqueous flow while decreasing cAMP production (Fig. 190.23).[71]
Although it is unlikely that the cholinergic or parasympathetic nervous system directly influences flow, muscarinic cholinergic inhibition of adenylate cyclase occurs in the rabbit ciliary body and ciliary epithelium.[85] Perhaps the cholinergic system modulates aqueous humor secretion in this manner. Consonant with the traditional juxtaposition of the cholinergic and adrenergic systems, the endogenous interaction of these systems could be further explored within the framework of the diurnal cycle (e.g., during sleep) when parasympathetic tone is dominant and aqueous flow is low.
In addition to these modulations of the ?-adrenergic receptor, other dynamic and complex factors are present. Modifying adrenergic effects on the eye are the rates of desensitization and uncoupling of the receptor, alterations in the rates of synthesis or degradation of receptor protein and its mRNA, and the regulation of the membrane-bound protein receptor. Downregulation and/or loss of receptors and response to long-term agonist exposure is particularly relevant to the clinical use of glaucoma agents that are adrenergic agonists.
Important evidence has accumulated to implicate the ciliary adenylate cyclase receptor complex in diurnal rhythms. Gregory and co-workers[86-90] used the circadian rhythms of intraocular pressure and aqueous flow in entrained rabbits to study the control of intraocular pressure and aqueous flow (Figs 190.24 to 190.28). It turns out that intraocular pressure, aqueous flow, and the concentrations of norepinephrine and cAMP in the aqueous increase during the dark in rabbits entrained to 12 h of light and 12 h of dark. Depriving the eye of sympathetic input by excision of the superior cervical ganglion or preganglionic section of the cervical sympathetic trunk blunted the dark phase increases of intraocular pressure, flow, and aqueous norepinephrine and cAMP. Blockade of ocular ?-adrenergic receptors with timolol decreased intraocular pressure and aqueous flow in the dark but not during the light phase, did not lower intraocular pressure during the dark phase in rabbits previously subjected to excision of the superior cervical ganglion or section of the cervical sympathetic trunk, and abolished the dark phase increase of aqueous cAMP. These data are consistent with the idea that increased sympathetic input to the rabbit eye during the dark phase of the circadian cycle increases aqueous flow and intraocular pressure and that these increases are mediated in part by stimulations of ocular ?-adrenergic receptors.
|
|
|
|
FIGURE 190.24 (a) Intraocular pressure (IOP) versus circadian time in rabbits housed in 12 h of light to 12 h of dark (L:D) and in constant dark (D:D). (b) Intraocular pressure versus time in rabbits housed in 12 h of light to 12 h of dark (L:D) and in the same animals 2 weeks after reversal of the phase of the light:dark cycle (D:L)( |
|
|
|
|
FIGURE 190.25 (a) Aqueous flow versus circadian time in rabbits housed in 12 h of light to 12 h of dark (L:D) and in constant dark (D:D). (b) Aqueous flow versus time in rabbits housed in 12 h of light to 12 h of dark (L:D) and in a second group of rabbits housed in a lighting schedule with reversed phase (D:L). |
|
|
|
|
FIGURE 190.26 Intraocular pressure versus circadian time in rabbits more than 2 weeks after unilateral superior cervical ganglionectomy (a) or unilateral section of the cervical sympathetic trunk decentralization (DX) (b). The asterisk indicates a statistically significant difference from control eye. P <.05. |
|
|
|
|
FIGURE 190.27 Aqueous flow versus circadian time in rabbits more than 2 weeks after unilateral superior cervical ganglionectomy (a) or unilateral section of the cervical sympathetic trunk, decentralization (b). The asterisk indicates a statistically significant difference from control eye. P < 0.05. |
|
|
|
|
FIGURE 190.28 Concentration of norepinephrine (NE) and cAMP versus circadian time in rabbits housed in 12 h of light to 12 h of dark. |
Finally, perhaps the most stimulating and productive idea in the field of aqueous formation is that the circadian rhythm of intraocular pressure and aqueous flow could hold the key to molecular regulatory mechanisms for aqueous formation. Indeed, within the functional framework of the circadian cycle of aqueous flow, it was found that the protein, beta arrestin, required for a full quench of the adrenergic beta receptor,[90a] was differientially expressed by the ciliary epithelium. During the circadian cycle a decrease in beta arrestin expression occurred when maximal activation of the beta adrenergic receptor increases flow.[90b,][90c]
By what cellular mechanism could the adenylate cyclase receptor complex influence aqueous humor formation? Intracellular mediators contribute to the opening of ion-selective channels. Intracellular mediators can alter ion transport in part by activating protein kinases that are mediator and substrate specific and either phosphorylate ion channels or carrier proteins directly or phosphorylate regulatory proteins that may be associated with membrane transport proteins. cAMP as a second messenger may modify the Na+ pump (see the section on Na+-K+-ATPase), or cAMP may activate (open) a Cl?channel. Beta adrenoceptor stimulation in isolated dog NPE yields increased cAMP formation to cause increased chloride efflux along with increase in the short circuit current. These events are blocked by timolol. Possiblesites of action for cAMP in this system would be on the Cl? channel ordinarily existing in the basolateral cell membranes of absorptive epithelia (sensitive to stilbenes) or on the Cl? channel that is apical in location in secretory epithelia (sensitive to loop diuretics, e.g., bumetanide). Positions for these transporters could be reversed. The transmembrane protein mediating the transport of NaCl and, secondarily, water is electroneutral, which means that this transporter uncouples the movement of ions from the electric potential of the cell. Should it turn out that Cl? is involved in the ciliary epithelial secretory process, its transport would be important for an understanding of aqueous humor formation because it might account for some of the solute movement across the ciliary epithelium that is not electrogenic.
CHLORIDE SECRETION
Although Cl? has never been considered a primary secretate coupled to cell metabolism, evidence for Cl? transport across the ciliary epithelia continues to be published.[91-94] Some of the short circuit current and transepithelial potential difference across isolated iris ciliary bodies has been shown to be Cl?-dependent. Net Cl? flux toward the aqueous has been reported.[42,91,92] The presence of Cl? channels in ciliary epithelium has been demonstrated by measurements of intracellular potential. Intracellular Cl? ciliary epithelial activity has been shown to be greater than that predicted from considerations of electrochemical equilibrium.[23] Single channel activity is present.[95] Cl? transport across other epithelia has been reported in these ways: (1) passive, driven by concentration differences and electrical potential; (2) Cl?-HCO3? exchange, electrically silent; and (3) Cl? absorption or secretion as described earlier, coupled to cations, in an electrically neutral manner.
In intact ciliary epithelium, it was shown that 4,4-didisothiocyanostilbene-2,2'-disulfonic acid (DIDS) affected the membrane potential and the equilibrium distribution of Cl?.[23] In addition, in cultured bovine pigmented epithelium, a Cl?-HCO3? exchanger was expressed.[52] In an earlier investigation,[42] in an intact iris ciliary body, DIDS decreased the short circuit current, further suggesting a Cl? conductance in NPE as a mechanism for the release of Cl? into the posterior chamber. With the use of the isolated bilayer, the stilbene-sensitive exchanger was localized in the basolateral membranes of the NPE cells.[96]The demonstrated effect of mercury in an in vitro system is undoubtedly on a Cl? channel. Recently, electron probe micro analysis (EPMA)[97a] has corroborated the thought that paired activity of Na+/H+ and Cl?/HCO3? antiports is the mechanism supporting the first step in secretion, i.e., uptake of stromal NaCl by PE cells.[97b]
CARBONIC ANHYDRASE
It is generally believed that transport of HCO3? ions across the ciliary epithelium is crucial to the formation of aqueous humor. The precise mechanism for HCO3? transport across the ciliary epithelium is unknown, and not very clear either is the localization of carbonic anhydrase in this process (i.e., cytosolic vs membranal). Friedenwald[98] first suggested that the movement of solute and solvent from the plasma into the posterior chamber involved a redox pump that provided a direct link between ion transport and electron transfer. Cell metabolism results in the production of unbalanced hydroxyl ions. By reaction with CO2, OH? ions are converted to HCO3? ions, in turn electrically neutralized by the entrance of Na+ from the blood. Studies of oxygen consumption in the ciliary processes indicated that the stoichiometric requirement for the movement of Na+ was not met by the redox theory.
Nevertheless, the splitting of water into OH? and H+ ions is the first step. The catalysis of OH? and CO2 to form HCO3? is the second step. It is at this stage that the enzyme carbonic anhydrase comes into play.[99,100] The observation that HCO3? is in excess in the posterior chamber of rabbits appeared to lend support to the direct involvement of catalyzed HCO3? formation in the production of aqueous humor. Later, however, it became known that, in primates, Cl? rather than HCO3? was in excess in the posterior chamber. Zimmerman and co-workers[101] resolved the apparent discrepancy presented by these 'irrelevant' observations by studying the entry rate of newly formed bicarbonate (H[14] CO3?) into the posterior chamber aqueous from plasma[14] CO2, rather than by analyzing steady-state levels of HCO3?. Kinsey's[102] and Kinsey and Reddy's work[103] and later Maren's studies[99] showed that HCO3? formation in newly formed aqueous humor does occur and can be slowed by inhibitors of the enzyme carbonic anhydrase. HCO3? formation in the dog and monkey is not as rapid as in the rabbit, and the half time to equilibrium can be measured at ?7 min and, under conditions of carbonic anhydrase inhibition, at 18 min. The molar transports of Na+ and HCO3? decline approximately equivalently by 1.9 and 1.6 mmol/min, respectively. Under these conditions inflow falls by 50%. Thus, in a few well-chosen experiments in the dog and in the monkey, carbonic anhydrase inhibitors were shown to reduce the accession rate of labeled CO2 and Na+ from the blood to the aqueous humor, the latter by ?29%.
Cole,[104] among others, reasoned earlier with respect to the HCO3? system that the action of acetazolamide, a carbonic anhydrase inhibitor, was to reduce the bioelectric potential across an in vitro ciliary body preparation owing to a fall in Na+ transport. This could have been related either to a reduced intracellular pH changing the polarity of the Na+-K+-ATPase pump or, possibly, to a reduction in the amount of Na+ entering the cell by way of the antiporter H+-Na+ system. The exchange of cellular H+ for Na+ from the stroma eventuating in the movement of Na+ and HCO3? into the posterior chamber is not unreasonable.
In a separate series of experiments, done in a novel preparation of the isolated intact ciliary epithelial bilayer in which the transepithelial potential is virtually completely maintained by the Na+ pump,[105] the HCO3? dependency of this potential can be shown, verifying work done in the iris-ciliary body preparations.[41] Furthermore, a 30% decrease in its magnitude, similar in amount to the reduction in the entry rate of Na+ into the posterior chamber found by Maren and co-workers,[101] occurs after the use of several different carbonic anhydrase inhibitors. Assuming that the reduction in the short circuit current is related to the decrease in bicarbonate transport from pigmented epithelium through the NPE into the aqueous, the effect of carbonic anhydrase inhibitors is consistent with a reduction in aqueous humor formation.[42,92] These coincident observations certainly support a coupling of Na+ and HCO3? transport across the ciliary epithelium.
The action of carbonic anhydrase has been included here among the membrane-bound enzyme complexes, but there are at least two carbonic anhydrases in the ciliary epithelium: cytosolic and membrane bound. Histochemical localization of the enzyme done with the cobalt method of Hanson[106,107] suffers from a low level of resolution. Several authors have reservations about immunocytochemical methods for carbonic anhydrase localization,[108] although in the kidney it appears that a type 4 membrane-associated carbonic anhydrase is present in luminal membranes of the proximal tubules and other portions of the kidney of lower vertebrates.[109,110] It is clear for the kidney that it is the activity of carbonic anhydrase bound to luminal membranes that is critical to the normal reabsorption of HCO3?.[111] Experiments were performed on the ciliary bilayer using highly polar molecules that are inhibitors of carbonic anhydrase; these molecules are impermeant or penetrate into the cell only very slowly and therefore remain extracellular. Results of these experiments show that a membrane-bound carbonic anhydrase is associated with the basolateral membranes of the NPE.[112] This fiding fits with an operative Cl?-HCO3?exchanger and/or an electrogenic NalHCO3 cotransporter found in this same membrane.
Thus, the hypothesis with regard to the action of carbonic anhydrase in the formation of aqueous humor would be as follows: Uptake of HCO3? into the pigment epithelium occurs and is probably both Na+and Cl? dependent. Disposal of OH? within the cell requires the catalyzed hydroxylation of CO2 and the secretion of H+ ions. Cytosolic carbonic anhydrase inhibitors may reduce the potential generated by the ciliary epithelium indirectly by lowering intracellular pH (OH? substrate becomes unavailable) to change the polarity of the Na+ pump or to activate pH-sensitive potassium conductance channels. The membrane-bound enzyme in the NPE converts extracellular HCO3 to CO2 and creates continuous extrusion of HCO3 in exchange for Cl?.[108] The function of the membrane-bound enzyme would be to separate the products of the enzyme catalysis: H+ and HCO3?. H+ would move to the cytoplasm in accord with the negative membrane potential inside the cell, and HCO3? would enter the aqueous humor. The mechanism of entry of HCO3? into the posterior chamber of the eye is not known but could occur by one or more of these general mechanisms[113]: (1) through a GABA-activated Cl? channel; (2) through an Na+-independent Cl?-HCO3? efflux (band 3); (3) influenced by an Na+-dependent Cl?-HCO3? exchanger; or (4) through an electrogenic Na+/HCO3cotransporter, very probably(3).
Na+-K+-ATPase
Active transport processes are coupled to the breakdown of high-energy PO4 bonds. The energy of the PO4 bond is used to drive the transport process. In the case of the linked Na+-K+ transport system, about one-third of the ATP expended by a cell is used to maintain the Na+-K+ gradient through the Na+ pump, the membrane-bound enzyme complex Na+-K+-ATPase. Intracellular
Na+ binds to the transport protein, which is phosphorylated by ATP. The phosphorylation process causes the bound Na+ to translocate extracellularly and be released. K+ then binds to the transport protein, which becomes dephosphorylated and releases K+ inside the cell. Ouabain competes with K+ for its binding site. Thus, the catalyzed hydrolysis of ATP to adenosine diphosphate (ADP) and PO4 by Na+-K+-ATPase is linked to Na+ transport.
Active transport is directional; hence, at least one component of the system must be asymmetric with respect to the cell membrane (see the section on Localization of Proteins). Directionality is provided by the energy coupling system. When energy coupling or production is inhibited, active transport systems lose their directionality and then simple facilitated diffusion may occur. Active transport of many ions is coupled directly to ATP hydrolysis, but the coupling mechanism is not completely understood. In principle, the oxidative phosphorylation of ADP to ATP is coupled to electron transport by a chemiosmotic mechanism. Oxidative phosphorylation can be blocked by (1) inhibiting electron transport with cyanine dyes; (2) preventing ATP formation directly with oligomycin; or (3) uncoupling oxidative phosphorylation by allowing electron transfer to occur without ATP formation, with lipid-soluble weak acids such as dinitrophenol or cyclic polypeptide antibiotics. Each of these compounds has been shown to interfere with aqueous humor formation.
In the ocular ciliary epithelium the membrane-bound enzyme complex Na+-K+-ATPase, an energy-dependent active transport system, is present. Cole[29] first demonstrated that the transport of Na+ into the eye could be halved by pretreatment of an in vitro ciliary body preparation with dinitrophenol. Later he showed that ouabain reduced the influx of Na+ into the anterior chamber by 50%. Still later he calculated the fall in intraocular pressure, related to reduced rate of aqueous humor secretion, to the reduction in Na+ influx after ouabain injection.[114] The rate constant for the entry of Na+ from the plasma to the posterior chamber was measured at ?0.013/min in the rabbit. This value is virtually the same as the value for the rate constant for fluid formation, 0.013/min. After ouabain, the decrease in Na+ entry and the reduction in fluid formation is equivalent.[104] This is the reason why Na+ is considered the primary secretate, and, since Na+ and water move together, the aqueous humor produced is isotonic with the plasma. Further experiments[115,116] showed ouabain-inhibitable Na+-K+-ATPase activity in the ciliary body. In the work of Bonting and Becker[117] there were significant correlations between flow decreases and Na+-K+-ATPase inhibition shortly after intravitreal administration of ouabain. In addition, the shrinkage of the 'hydrated' stroma of an isolated ciliary process can be inhibited by the action of ouabain on the ciliary epithelium.[38]
Rb+ is taken up by cells through the mediation of Na+-K+-ATPase in the same manner as K+. Ouabain prevents the binding of K+ to the enzyme-bound complex at its extracellular exposure. Thus, the accumulation of Rb+ in isolated iris ciliary body preparations can be largely prevented by preincubation with the digitalis glycosides,[118] further substantiating the functional activity of the Na+-K+-activated ATPase in the ciliary epithelium.
Electrophysiologic experiments by Krupin,[41] Candia and co-workers,[42] and others working with mounted iris-ciliary body specimens showed that the 'transpreparation' potential difference can be markedly reduced by ouabain. It was clear from these bioelectrical measurements that the Na+ pump generates a transepithelial potential difference. Also, in the isolated intact ciliary epithelial bilayer, free of vascular or connective tissue components and blood but in which the tight junctions and gap junctions of the epithelial bilayer are preserved, ouabain similarly reduced the transepithelial potential difference virtually to zero,[105] confirming the presence of an electrogenic Na+ pump.
The distribution of ion-stimulated ATPases of the ciliary epithelium has been examined biochemically in tissues from bovine and rabbit eyes.[28] Na+-K+-stimulated enzyme activities were found in homogenates of tissues from both species. Separation of pigmented and nonpigmented layers of the bovine ciliary epithelium and isolation of the two cell types of density gradients showed higher activities of Na+-K+-ATPases in the nonpigmented cells. Subcellular fractionation of a mixed population of cells showed that the plasma membrane fraction contained more than 80% of the Na+-K+-ATPase activity and that the NPE had two to three times the biochemical activity of the pigmented epithelium.
The localization of Na+-K+-ATPase in the ciliary epithelium from both light and electron microscopic studies indicates that the membrane-bound complex can be best seen on the basolateral membranes of the NPE and to a lesser degree on the basolateral membranes of the pigmented epithelium (Fig. 190.29).[119,120] Coca-Prados and coworkers,[121,122] using immunocytochemistry, indicated that the a1-, a2-,and a3-subunit isoforms are expressed together in the NPE cells, whereas only the a1 subunit isoform is expressed in the pigmented epithelial cells (Fig. 190.30). In rat adipocytes, Lytton and associates[123] have reported that the a2-subunit isoform of the enzyme may be hormonally regulated. How might Na+-K+-ATPase of the ciliary epithelium be regulated? Could the enzyme complex or one of its isoforms be modulated by a cAMP-dependent protein kinase, perhaps protein kinase A? Experiments testing this speculation showed that rabbit iris-ciliary body Na+-K+-ATPase activity is reduced by a commercial (nonendogenous) preparation of a cAMP-dependent kinase.[124] In addition, the ouabain-inhibitable activity of the Na+ pump was reduced by either cAMP or its kinase. These studies suggest that increases in cAMP could affect the rate of Na+ transport and thus alter aqueous humor formation. In other tissues the activation of other pathways, like the phosphoinositol pathway, may affect Na+ transport through another protein kinase, namely, protein kinase C. The speculation that the Na+-K+-ATPase membrane-bound protein complex can indeed be phosphorylated or dephosphorylated so that its action to promote water and salt secretion is altered; is not an unattractive hypothesis but requires further evidence. The link between the adenylate cyclase receptor complex and the Na+ pump would be established.
|
|
|
|
FIGURE 190.29 Fresh tissue immediately incubated en bloc in ATP medium containing 0.33 M sucrose. Incubation was carried out at room temperature for 30 min. Large deposits are dispersed in a spotty fashion and fill the space between apposing membranes. Note relatively dense and irregular mode of precipitation along the internal limiting membrane (ILM), whereas the free surface of this epithelium shows no reaction product. PC, posterior chamber; NPE, nonpigmented epithelium. ×23 900. |
|
|
|
|
FIGURE 190.30 Immunolocalization of a1 (a and b), a2 (c and d), and a3 (e and f) subunit isoforms of the Na+, K+-ATPase in 0.5-?m cryostat sections of the pars plicata region of the bovine ciliary epithelium. a1-Antiserum (polyclonal antibodies to purified canine kidney Na+, K+-ATPase) stained the basolateral plasma membrane domains of nonpigmented (NPE) and pigmented (PE) epithelial cells (arrows). a2-Antiserum (mcB2) stained only the basolateral plasma membrane of nonpigmented cells, not that of the pigmented cells. a3-Antiserum (McB-X3.1) also stained the basolateral plasma membrane domain, composed solely of nonpigmented cells, not of pigmented cells. Specific bound antibodies were visualized with rhodamine-conjugated secondary antibodies. On the right are phase-contrast photographs corresponding to the immunomicrographs on the left. |
COMMENTS ABOUT THE RELATIONSHIP BETWEEN RATES OF FLOW AND RATES OF TRANSPORT IN VIVO
Measurements of aqueous flow rates have been done using the following:
|
1. |
Clearance rates. After systemic administration of compounds such as para aminohippuric acid (PAHA),[131] I Rayopake, or[131] I Diodrast, (1) these substances are maintained at a constant plasma concentration and the time course of their intraocular accumulation measured[125-127]; or (2) the steady-state distribution of compounds between the intraocular fluids and plasma is determined[128]; or (3) a bolus of a test substance is given at zero time, the resulting plasma function is determined, and then the ocular function is derived.[129,130] |
|
|
2. |
Static chemistries using ratios among posterior chamber-anterior chamber/plasma. |
|
|
3. |
Turnover of test substances such as ascorbic acid and labeled ions (24Na, 36Cl) before and after aqueous humor suppressants. |
|
|
4. |
Clearance rates by noninvasive methods employing topically applied tracers such as fluorescein or use of the flare decline curve or by invasive methods that perfuse or inject labeled tracers, such as[131] I or nonmetabolized molecules. |
The second and third techniques require simultaneous sampling of the plasma, posterior chamber, and anterior chambers. A review of these different methods has been given in several works.[132,133]
Manometric determinations of aqueous flow rate are done by tonography, by perfusion at constant pressure, constant rate, and constant volume, and by obstruction of the outflow channels, as with the use of the perilimbal suction cup. As discussed, the correlation between flow rates and ion transport has been made by comparing the rates of penetration of ions with rates of water penetration before and after the effects of inhibitors. Of course, the effect of drugs to inhibit or stimulate the membrane-bound protein complexes of adenylate cyclase, carbonic anhydrase, and Na+-K+-ATPase and to affect the rate of aqueous flow support the role of these enzymes in aqueous humor production.
Clearance or dye-dilution methods do not measure aqueous secretion directly but only the net flow of aqueous from the posterior chamber into the anterior chamber. In this connection it should be noted that aqueous humor formation (i.e., secretion) is not necessarily the same as aqueous flow, measured in vivo either in animals or humans, namely, with clearance of topically applied fluorescein. From the viewpoint of ocular pressure (because of glaucoma) only that amount of fluid that must exit through pathways that comprise the outflow resistance is relevant. Aqueous secreted, however, either under normal conditions or under the influence of drugs, and, then, perhaps reabsorbed by pressure-independent mechanisms so that it never reaches nor passes through resistance pathways 'has never been formed'. Thus, pressure-independent pathways (uveoscleral flow) affect chemical and clearance methods for the measurement of flow because they contribute to the turnover of substances in the aqueous. These latter pathways are 'negative' secretion; that is, F = c (Pa - Pe) + U, where the units are F = ?L/min; c = ?L/min/mmHg; Pa = mmHg intraocular pressure and Pe = mmHg episcleral venous pressure; and U = uveoscleral flow. In Goldmann and Schmidt's experiments,[134] no significant negative secretion in normal human eyes was found but after cyclodialysis it became prominent. In human experiments there is very little information about the normal rate of uveoscleral flow.[135] Bill did fid that uveoscleral flow ranged from a value of 0-27% of the total flow of aqueous, depending on the drug treatment used (i.e., pilocarpine shut it down, atropine increased it). In two untreated eyes of the 12 human eyes studied uveoscleral flow represented 4% and 14% of the total outflow. However, in Bill's series a retinal detachment was frequently present, suggesting that only with large exposures of the choroid, as after cyclodialysis or after ciliary body detachment, or perhaps with drugs, does significant reabsorption of secreted aqueous by this pathway occur.[136] In freshly enucleated human eyes with intact chamber angles and choroid only 4% uveoscleral outflow could be found.[136a] Drugs, such as analogs of the prostaglandins, especially, of the F2-alpha class, may influence the pressure-independent outflow of water through the uveoscleral pathway itself and change the access of water to the pathway.
OUTWARD TRANSPORT
There is an active transport out of the eye of iodopyracetate and other related organic anions.[137-140] This system, called the kidney system, has properties that parallel the accumulation and transport of the same anions by renal tubules. So the eye is an extrarenal kidney. In addition, Bárány found a hippurate-insensitive iodipamide system.[141,142] Although this system is complex, the presence of a liver-like system has probably been established for the ciliary epithelium. These systems need further characterization with regard to their rate-limiting steps, metabolic dependencies, molecular mechanisms, and polarity (or direction), especially of the liver-like system. The results may have important implications for understanding detoxification and other protective functions performed by the ciliary epithelium. Apart from the movement (transport) of metabolites and toxins, the enzymes for detoxification have been identified in the ciliary epithelium and studied. The role of such metabolite and drug-metabolizing detoxifying systems has been addressed,[142] but progress has been slow.[143,144]
COMPOSITION OF AQUEOUS
It is beyond the scope of this chapter dealing with the process of aqueous humor production to describe in detail the composition of the aqueous humor.[42,49,128,130,132,145-147]
Three tables have been taken for inclusion here from among numerous articles[145] describing the aqueous components (Tables 190.1 to 190.3). Forvirtually all substances, carrier mechanisms exist that provide for an accumulation within one or both ciliary epithelial layers, usually by saturation kinetics, after which release by diffusion into the posterior chamber from the NPE becomes a rate-limiting factor for the process. Continued exploration of aqueous composition is required to decide what are the essential substrates for the nutrition of the avascular structures of the eye.
TABLE 190.1 -- Concentrations of Inorganic Substances in the Aqueous Humor
|
Concentration |
|||||
|
Substance |
Anterior Aqueous |
Posterior Aqueous |
Plasma |
Species |
Investigator |
|
Bicarbonate (?mol/mL) |
27.7 |
34.1 |
24.0 |
Rabbit |
Kinsey, 1953 |
|
33.6 |
27.4 |
Rabbit |
Davson, 1962 |
||
|
20.2 |
27.5 |
Human |
DeBernadinis et al, 1965 |
||
|
22.5 |
18.8 |
Monkey |
Gaasterland et al, 1979 |
||
|
Chloride (?mol/mL) |
131.6 |
124.0 |
Human |
Remky, 1956, 1957 |
|
|
105.7 |
106.2 |
Rabbit |
Cole, 1959 |
||
|
131.0 |
107.0 |
Human |
DeBernadinis et al, 1965 |
||
|
105.1 |
111.8 |
Rabbit |
Kinsey and Reddy, 1964 |
||
|
124.8 |
100.0 |
107.3 |
Human |
Gaasterland et al, 1979 |
|
|
Calcium (?mol/mL) |
1.7 |
2.5 |
2.6 |
Rabbit |
Davson, 1962 |
|
2.5 |
4.9 |
Monkey |
Bito, 1970 |
||
|
Hydrogen ion (pH) |
7.60 |
7.57 |
7.40 |
Rabbit |
Kinsey and Reddy, 1964 |
|
7.49 |
Monkey |
Gaasterland et al, 1979 |
|||
|
Magnesium (?mol/mL) |
0.8 |
1.0 |
Rabbit |
Davson, 1962 |
|
|
1.2 |
1.3 |
1.2 |
Monkey |
Bito, 1970 |
|
|
0.8 |
0.7 |
Monkey |
Gaasterland et al, 1979 |
||
|
Oxygen (mm Hg) |
55 |
100-150 |
Rabbit |
Heald and Langham, 1956 |
|
|
53 |
[*] |
Human |
Kleifeld and Neumann, 1959 |
||
|
30 |
77 |
Rabbit |
Wegener and Moller, 1971 |
||
|
32 |
[*] |
Rabbit |
Stefansson et al, 1983 |
||
|
Phosphate (?mol/mL) |
0.62 |
1.11 |
Human |
Walker, 1933 |
|
|
0.86 |
0.57 |
1.11 |
Rabbit |
Constant and Falch, 1963 |
|
|
0.89 |
0.52 |
1.49 |
Rabbit |
Kinsey and Reddy, 1964 |
|
|
0.14 |
0.68 |
Monkey |
Gaasterland et al, 1979 |
||
|
Potassium (?mol/mL) |
5.1 |
5.6 |
5.6 |
Rabbit |
Reddy and Kinsey, 1960 |
|
5.2 |
5.5 |
Rabbit |
Davson, 1962 |
||
|
3.6 |
4.1 |
4.2 |
Monkey |
Bito, 1970 |
|
|
3.9 |
4.0 |
Monkey |
Gaasterland et al, 1979 |
||
|
Sodium (?mol/mL) |
146 |
144 |
Rabbit |
Kinsey and Reddy, 1964 |
|
|
143 |
146 |
Rabbit |
Kinsey and Reddy, 1964 |
||
|
145 |
150 |
144 |
Sheep |
Cole, 1970 |
|
|
153 |
153 |
152 |
Dog |
Maren, 1976 |
|
|
152 |
148 |
Monkey |
Gaasterland, 1979 |
||
|
* |
Breathing room air. |
TABLE 190.2 -- Concentrations of Organic Substances in Aqueous Humor
|
Concentration |
|||||
|
Substance |
Anterior Aqueous |
Posterior Aqueous |
Plasma |
Species |
Investigator |
|
Ascorbate (?mol/mL) |
0.96 |
1.30 |
0.02 |
Rabbit |
Kinsey, 1953 |
|
1.06 |
0.04 |
Human |
DeBernadinis et al, 1965 |
||
|
1.18 |
0.02 |
Monkey |
Gaasterland et al, 1979 |
||
|
Citrate (?mol/ml) |
0.38-0.46 |
Rabbit |
Granwall, 1937 |
||
|
0.12 |
Human |
Granwall, 1937 |
|||
|
Creatinine (?mol/mL) |
0.18 |
0.18 |
Horse |
Duke Elder, 1927 |
|
|
0.11 |
Rabbit |
Furuichi, 1961 |
|||
|
0.04 |
0.03 |
Monkey |
Gaasterland et al, 1979 |
||
|
Glucose (?mol/mL) |
4.9 |
5.3 |
Rabbit |
Reddy and Kinsey, 1960 |
|
|
2.8 |
5.9 |
Human |
DeBernadinis et al, 1965 |
||
|
6.9 |
7.2 |
Rabbit |
Reim et al, 1967 |
||
|
3.0 |
4.1 |
Monkey |
Gaasterland et al, 1979 |
||
|
Hyaluronate (mg/mL) |
4.0 |
Ox |
Duke-Elder and Goldsmith, 1951 |
||
|
4.4 |
Cattle |
Laurent, 1981 |
|||
|
1.1 |
Human |
Laurent, 1981 |
|||
|
Lactate (?mol/mL) |
12.1 |
11.2 |
8.2 |
Rabbit |
Kinsey, 1953 |
|
9.3 |
9.9 |
10.3 |
Rabbit |
Reddy and Kinsey, 1960 |
|
|
4.5 |
1.9 |
Human |
DeBernadinis et al, 1965 |
||
|
9.9 |
9.0 |
5.6 |
Rabbit |
Riley, 1972 |
|
|
4.3 |
3.0 |
Monkey |
Gaasterland et al, 1979 |
||
|
Protein (mg/dL) |
13.5 |
Human |
Krause and Raunio, 1969 |
||
|
100.3 |
Rat |
Stjernschantz et al, 1973 |
|||
|
33.3 |
Monkey |
Gaasterland et al, 1979 |
|||
|
25.9 |
Rabbit |
Dernouchamps, 1982 |
|||
|
23.7 |
Human |
Dernouchamps, 1982 |
|||
|
Urea (?mol/mL) |
6.3 |
5.8 |
7.3 |
Rabbit |
Kinsey, 1953 |
|
7.0 |
9.1 |
Rabbit |
Davson, 1962 |
||
|
6.1 |
7.3 |
Monkey |
Gaasterland et al, 1979 |
||
TABLE 190.3 -- Concentrations of Amino Acids in Aqueous Humor and Aqueous:Plasma Ratio in Rabbits, Monkeys, and Humans
|
Rabbit[*] |
Monkey[?] |
Human[?][§] |
||||
|
Amino Acid (?mol/kg H2O) |
Aqueous |
Aqueous: Plasma |
Aqueous |
Aqueous: Plasma |
Aqueous |
Aqueous: Plasma |
|
Alanine |
480 |
1.59 |
208.5 |
0.76 |
306 |
0.94 |
|
Arginine |
272 |
2.83 |
51.0 |
0.49 |
105 |
1.50 |
|
Aspartate |
55 |
1.90 |
Trace |
- |
- |
- |
|
Citrulline |
- |
- |
4.0 |
0.12 |
- |
- |
|
Cysteine |
- |
- |
217.0 |
1.01 |
- |
- |
|
Glutamate |
295 |
1.66 |
13.0 |
0.18 |
9 |
0.19 |
|
Glycine |
614 |
0.52 |
44.5 |
0.12 |
24 |
0.11 |
|
Histidine |
210 |
1.81 |
40.5 |
0.45 |
67 |
0.85 |
|
Isoleucine |
116 |
1.04 |
47.5 |
0.82 |
65 |
1.30 |
|
Leucine |
174 |
1.07 |
110.5 |
1.16 |
139 |
1.42 |
|
Lysine |
423 |
2.00 |
85.0 |
0.53 |
159 |
0.64 |
|
Methionine |
23 |
1.64 |
46.5 |
1.42 |
44 |
2.54 |
|
Methylhistidine |
- |
- |
8.5 |
0.27 |
- |
- |
|
Ornithine |
- |
- |
9.0 |
0.25 |
- |
- |
|
Phenylalanine |
97 |
1.00 |
73.0 |
1.45 |
93 |
2.01 |
|
Proline |
267 |
0.83 |
27.0 |
0.16 |
44 |
0.19 |
|
Serine |
585 |
1.39 |
718.0 |
1.38 |
- |
- |
|
Taurine |
- |
- |
11.5 |
0.09 |
66 |
1.02 |
|
Threonine |
138 |
0.84 |
61.5 |
0.83 |
128 |
1.17 |
|
Tryptophan |
24 |
1.85 |
11.5 |
0.62 |
- |
- |
|
Tyrosine |
101 |
1.74 |
65.0 |
1.26 |
91 |
1.84 |
|
Valine |
230 |
1.18 |
167.0 |
1.02 |
285 |
1.35 |
Courtesy of J. Caprioli.
|
* |
From Reddy DVN, Rosenburg C, Kinsey VE: Exp Eye Res 1961; 1:175, 1961. |
|
? |
From Gaasterland DF, et al: Invest Ophthalmol Vis Sci 1979; 18:1139, 1979. |
|
? |
From Ehlers N, Schonheyder F: Acta Ophthalmol (Suppl) 1974; 123:179, 1974. |
|
§ |
Aqueous samples were taken just before cataract extraction in 30 patients. |
ACKNOWLEDGEMENT
The author gratefully acknowledges the work of his friend and colleague, Professor Eichi Yamada, who contributed original photographs to this chapter.
REFERENCES
1. Knutson SL, Sears ML: Herman Boerhaave and the history of vessels carrying aqueous humor from the eye. Am S Ophthalmol 1973; 76:648-654.
2. Magitot A: Sur les sources multiples de I'humeur aqueuse. Ann Oct 1928; 11165:481-507.
3. Duke-Elder S: Textbook of ophthalmology, London: Kimpton; 1932:455.
4. Leber T: Die Zirkulations-und Ertlahrungsverhaltnisse des Auges. Graefe-Saemiach's Handbuch der Gesellshaft fur Augenheilk, Leipzig: W. Engelmann; 1903:63.
5. Lauber H: Beitrag zur Anatomic des vorderen Augen-abbachnittes der Wirbeithiere Anat I-lent 1901; 18:430.
6. Troncoso MU: The physiologic nature of the Schlemm canal. Am J Ophthalmol 1921; 4:321-326.
7. Seidel E: Mikroskopische Beobachtungen uber den Mechanismus des Abflusses aus der Vorderkammer des Lebenden Tieres bie physiologimchem Augendrucke. Graefes Arch ophthalniol 1923; 111:167.
7a. Ascher KW: Aqueous veins. Am J Ophthalmol 1942; 25:31.
7b. Goldmann H: Weitere Mitteilung uber den Abfluss des Kammerwassers beim menschen. Ophthalmologica 1946; 112:344.
8. Davson H: Physiology of the Ocular and Cerebrospinal Fluid, Boston: Little, Brown; 1956:7-11.
9. Friedenwald JS, Stichler RD: Circulation of the aqueous: VII. A mechanism of secretion of the intraocular fluid. Arch Ophthalmol 1938; 20:761-786.
10. Friedenwald JS: The formation of the intraocular fluid. Am J Ophthalmol 1949; 32:9-27.
11. Ehinger B: Ocular and orbital vegetative nerves. Acta Physiol Scand 1966; 67(Suppl 2681):1-35.
12. Yamada E: Further observation on the intraepithelial nerve fibers of rabbit ocular ciliary epithclium. Arch Histol Cytol 1989; 52:191-195.
13. Bill A: The role of ciliary blood flow and ultrafiltration in aqueous humor formation. Exp Eye Res 1973; 16:287-295.
14. Bill A: Blood circulation and fluid dynamics in the eye. Phytiol Rev 1971; 55:383-417.
15. Alm A: Microcirculution of the eye. In: Mortillaro NA, ed. The Physiology and pharmacology of the microcirculation, New York: Academic Press; 1983:299-359.
16. Brodwall J, Fishbarg J: The hydraulic conductivity of rabbit ciliary epithelium in vitro. Exp Eye Res 1982; 34:121-129.
17. Linner E: Ascorbic acid is a test substance for measuring relative changes in the rabbit of plasma flow through the ciliary processes. Acta Physiol Scand 1952; 26:57-103.
18. Sears ML: Physiology and pharmacology of aqueous humor formation: implications with respect to treatment. Ophthalmol Clin North Am 1991; 4:781-802.
19. NeufeId AH, Sears ML: The site of action of prostaglandin E on the disruption of the blood/aqueous barrier in the rabbit eye. Exp Eye Res 1973; 17:445-448.
20. Vegge T, Neufeld AH, Sears ML: Morphology of the breakdown of the blood-aqueous barrier in the ciliary processes of the rabbit eye after prostaglandin E. Invest Ophthalmol 1975; 14:33-40.
21. Green K Bountra C, Georgiou P, et al: Invest Ophthalmol Vis Sci 1985; 26:371-381.
22. Wiederholt M, Zadunaisky JA: Membrane potentials and intra cellular chloride activity in the ciliary body of the shark. Pflugers Arch 1986; 407(Suppl 2):5112-5115.
23. Diamond JR, Bossert WH: Standing gradient osmotic flow: a mechanism for coupling of water and solute transport in epithelia. J Gen Physiol 1967; 50:2061-2083.
24. Ballintine ET, Peters L: Effects of intracarotid injection of a basic dye on the ciliary body. Am J Ophthalmol 1954; 38:153-163.
25. deRoetth Jr A: Glycolytic activity of ciliary process. Arch Ophthalmol 1954; 51:599-608.
26. Shimizu H, Riley MV, Cole DF: The isolation of whole cell from the ciliary epithelium together with some observations on the metabolism of the two cell types. Exp Eye Res 1967; 6:141-151.
27. Riley MV, Kishida K: ATPases of ciliary epithelium: cellular and subcellular distribution and probable role in secretion of aqueous humor. Exp Eye Res 1986; 42:559-568.
28. Cole DF: Aqueous humor formation. Doc Ophthalmol 1966; 21:116-238.
29. Becker B: Decrease in intraocular pressure in man by a carbonic anhydrase inhibitor, Diamos. Am J Ophthalmol 1954; 37:13-15.
30. Grant WM, Trotter RR: Diamox (acetazolamide) in the treatment of glaucoma. Arch Ophthalmol 1954; 51:735-739.
31. Breinin GM, Gortz H: Carbonic anhydrase inhibitor acetazelamide (Diamiox): a new approach to the therapy of glaucoma. Arch Ophthalmol 1954; 52:333-348.
32. Vareilles P, Silverstone D, Pazonnet B, et al: Comparison of the effects of timolol and other adrenergic agents on intraocular pressure in the rabbit. Invest Ophthalmol Vis Sci 1977; 16:987-996.
33. Katz IM, Hubbard WA, Getson AJ, et al: Intraocular pressure decrease in normal volunteers following timolol ophthalmic solution. Invest Ophthalmol Vis Sci 1976; 15:489-492.
34. Zimmerman TJ, Kaufman HE: Timolol: a beta adrenergic blocking agent for the treatment of glaucoma. Arch Ophthalmol 1977; 95:601-604.
35. Brechue WF, Maren TH: Correlation of drug accession with 10P reduction following local and intravenous carbonic anhydrase inhihitors. Invest Ophthalmol Vis Sci 1991; 32:156.
36. Nathanson JA: Adrenergic regulation of intraocular pressure: Identification of beta, adrcncrgic stimulated adenylate cyclase in ciliary process epithelium. Proc Natl Acad Sci USA 1980; 77:7420-7424.
37. Gregory OS, Bausher TP, Bromberg BB, et al: The beta adrenergic receptor and adenyl cyclase of rabbit ciliary processes. In: Sears ML, ed. New directions in ophthalmic research, New Haven: CT. Yale University Press; 1981:127-148.
38. King LS, Agre P: Pathophysiology of the aquaporin channels. Ann Rev Physiol 1996; 58:619-648.
39. Friedenwald JS, Herrmann H, Moses R: The distribution of certain oxidative enzymes in tbc ciliary body. Johns Hopkins Bull 1943; 73:421-434.
40. Berggren L: Direct observation of secrctory pumping in vitro of the rabbit eye ciliary processes: influence of ion milieu and carbonic anhydrase inhibition. Invest Ophthalmol 1964; 3:266-272.
41. Berggren L: Effect of a mercurial diuretic, Mersalyl, on in vitro secretory activity of the rabbit eye ciliary processes. Acta Ophthalmol 1970; 48:253-275.
42. Becker B: Ascorbate transport in guinea pig eyes. Invest Ophthalmol 1967; 6:410-415.
43. Krupin T, Reinach PS, Candia OA, et al: Transepithelial electrical measurements on the isolated rabbit iris-ciliary body. Exp Eye Res 1984; 311:115-123.
44. Pesin SR, Candia OA: Na+ and Cl? fluxes and effects of pharmacological agents on the short circuit current of the isolated rabbit iris-ciliary body. Curr Eye Res 1982/83; 2:815-827.
45. Holland MG, Gipson CC: Chloride ion transport in the isolated ciliary body. Invest Ophthalmol 1970; 9:20-29.
46. Cole DF: Evidence for acrive transport of chloride in ciliary epithelium of the rabbit. Exp Eye Res 1969; 8:5-15.
47. Cole DF: Secretion of the aqueous humor. Exp Eye Res 1977; 25:161-176.
48. Chu TC, Candia OA: Electrically silent Na+ and Cl? fluxes across the rabbit cilia, epithelium. Invest Ophthalmol Vis Sci 1987; 28:445-450.
49. Chu TC, Candia OA: Active transport of ascorbate across the isolated rabbit ciliary epishelium. Invest Ophthalmol Vis Sci 1988; 29:594-599.
50. Kondo K, Coca-Prados M, Sears ML: Human ciliary epithelia in monolayer culture. Exp Eye Res 1984; 38:423-433.
51. Coca Prados M, Wax MB: Transformation of human ciliary epithelial cells by simian virus 40: induction of cell proliferation and retention of beta adrenergic receptors. Proc Natl Acad Sci USA 1986; 83:8754-8758.
52. Wiederholt M, Helbig H, Kormacher C: Ion transport across the ciliary epithelium: lessons from cultured cells and proposed role of the carbonic anhydrase. In: Botre F, Gross C, Storey UT, ed. : Carbonic anhydrase, New York: VCH Publications; 1991:232-244.
53. Helbig H, Kormacher C, Nawrath M, et al: Sndinm bicarbonate cotransport in cultured pigmented ciliary epithelial cells. Curr Eye Res 1989; 8:595-598.
54. Helbig H, Korbmacher C, Kuhner D, et al: Characterization of Cl/HCO; - exchange in cultured bovine pigmented ciliary epithelium. Exp Eye Res 1988; 47:515-523.
55. Helbig H, Korbmacher C, Berweck S, et al: Kinetic properties of Na'/H exchanged in cultured bovine pigmented ciliary epithelial cells. Eur J Physiol 1988; 412:80-85.
56. Helbig H, Korbmacher C, Stumpff F, et al: Na/H exchange regulated intracellular pH in a cell done derived from bovine pigmented ciliary epithelium. J Cell Physiol 1988; 137:384-389.
57. Helbig H, Korbmacher C, Wohlfarth I, et al: Electrogenic Na ascorbate cotransport in cultured bovine pigmented diliatry epitheliai cells. Am J Physiol 1989; 256:44-49.
58. Helbig H, Korhmacher C, Wohlfarth J, et al: Effect of acetylcholine on membrane potential of cultured human nonpigmensed ciliary epithclial cells. Invest Ophthalmol Vis Sci 1989; 30:890-896.
59. Barros F, Lopez-Briones LU, Coca-Prados M, et al: Detection and characterization of Ca2+-activated K+ channels in transformed cells of human non-pigmented ciliary epithelium. Curr Eye Res 1991; 10:731-738.
60. Wax MB, Coca-Prados M: Receptor-mediated phosphoinositide hydrolysis in human ocular cillary epithelial cells. Invest Ophthalmol Vis Sci 1989; 30:1675-1679.
61. Sears ML: The aqueous. In: Moses RA, ed. Adler's physiology of the eye, St Louis: CV Mosby; 1980:204-226.
62. Sears ML: Catecholamines in relation to the eye. In: Astwood B, Creep R, ed. Handbook of physiology, section on endocrinology, Washington, DC: American Physiological Society; 1975:553-590.
63. Waitzman MB, Woods WD: Some characteristics of an adenyl cyclase preparation from rabbit ciliary process tissue. Exp Eye Res 1971; 12:99-111.
64. Neufeld AH, Sears ML: Cyclic AMP in the ocular tissues of the rabbit, monkey, and human. Invest Ophthalmol Vis Sci 1974; 13:475-477.
65. Bromberg BB, Gregory DS, Sears ML: Beta adrenergie receptors in ciliary processes of the rabbit. Invest Ophthalmol Vis Sci 1980; 19:203-207.
66. Wax MB, Molinoff PB: Distribution and properties of beta adrenergic receptors in human iris-ciliary body. Invest Ophthalmol Vis Sci 1987; 28:420-430.
67. Elena PP, Fredj-Reydgrobellet D, Moulin G, et al: Pharmacological characteristics of ?-adrenergic-sensitive adenylate cyclase in nonpigmented and in pigmented cells of bovine diary process. Curr Eye Res 1984; 3:1383-1389.
68. Palkama A, Kaufman H, Uusitalo R, et al: Histochemistry of adenylate cyclase activity in the anterior segment of the eye: a methodological evaluation with biochemical background. Exp Eye Res 1986; 43:1043-1056.
69. Tsukahara S, Maezewa N: Cytochemical localization of adenyl cyclase in the rabbit ciliary body. Exp Eye Res 1978; 26:99-106.
70. Polansky JR, Alvarado JA: Isolation and evaluation of target cells in glaucoma research: hormone receptors and drug responses. Curr Eye Res 1985; 4:267-279.
71. Bausher LP, Gregory DS, Sears ML: Interaction between alpha and beta,-adrenergic receptors in rabbit ciliary processes. Curr Eye Res 1987; 6:497-505.
72. Bausher LP, Gregory DS, Sears ML: Alpha-adrenergic and VIP receptors in rabbit ciliary processes interact. Curr Eye Res 1989; 8:47-54.
73. Sears ML: Autonomic nervous system: adrenergic agonists. In: Sears ML, ed. Handbook of experimental pharmacology, pharmacology of the eye, Berlin: Springer; 1984:193-233.
74. Sears ML: Regulation of aqueous flow by the adenylate cyclase receptor complex in the ciliary epithelium. Am J Ophthalmol 1985; 1053:194-198.
75. Topper RE, Brubaker RF: Effects of timolol, epinephrine, and acetacolamide on aqueous flow during sleep. Invest Ophthalmol Vis Sci 1985; 26:1315-1319.
76. Vareilles P, Silveratone D, Plazonnet B, et al: The effect of timolol, a beta-adrenergic blocking agent, on 10P in rabbits. Pharmacologist 1976; 18:139.
77. Mittag TW, Torstasy A, Podos SM: Vascotive intestinal peptide and intraocular pressure: adenylate cyclase activation and binding sites for vasoactive intestinal peptide in membranes of ocular ciliary processes. J Pharmacol Exp Ther 1987; 241:230-235.
78. Uddman R, Alumets J, Ehinger B, et al: Vasoactive intestinal peptide nerves in ocular and orbital structures of the cat. Invest Ophthalmol Vis Sci 1980; 19:875-885.
79. Nilsson SFE, Macpea O, Samuelsson M, et al: Effects of timolol on terbutaline- and VIP-stimulated aqueous humor flow in the cynomolgus monkey. Curt Eye Res 1990; 9:863-872.
80. Cepelik J, Hynie S: Inhibitory effects of neuropeptide Y on adenylate cyclase of rabbit diary processes. Curr Eye Res 1990; 9:121-128.
81. Bausher LP, Horio B: Neuropeptide Y and somatostatin inhibited stimulated cyclic AMP production in rabbit ciliary processes. Curr Eye Res 1990; 9:371-378.
82. Junblatt JE, Gooch JM: Neuropeptide y modulates adenylate cyclase in the rabbit iris ciliary body and ciliary epithelium. Exp Eye Rca 1990; 51:229-231.
83. Caprioli J, Sears M, Bausher L, et al: Forskolin lowers intraocular pressure by reducing aqueous inflow. Invest Ophthalmol Vis Sci 1984; 25:268-277.
84. Gregory D, Sears ML, Bausher L, et al: Intraocular pressure and aqueous flow are decreased by cholera toxin. Invest Ophthalmol Vis Sci 1981; 20:371-381.
85. Jumblatt JE, North GT, Hackmiller RC: Muscarinic cholinergic inhibition of adnylate cyclase in the rabbit iris-ciliary body and ciliary epithclium. Invest Ophthalmol Vis Sci 1990; 31:1103-1108.
86. Gregory D: Timolol reduces lOP in normal NZW rabbits during the dark only. Invest Ophthalmol Vis Sci 1990; 31:715-721.
87. Gregory D, Aviado DG, Sears ML: Cervical ganglionectomy alters the circadian rhythm of intraocular pressure Sn New Zealand White rabbits. Curr Eye Res 1985; 4:1273-1279.
88. Smith SD, Gregory D: A circadian rhythm of aqueous flow underlies the circadian rhythm of lOP in NZW rabbits. Invest Ophthalmol Vis Sci 1989; 30:775.
89. Yoshitomi T, Gregory DS: Ocular adrenergic nerves control the circadian rhythm of aqueous flow in rabbits. Invest Ophthalmol Vis Sci 1991; 32:523-528.
90. Yoshitomi T, Horn B, Gregory DS: Changes in aqueous norepinephrine and cyclic adenosine monophosphate during the circadian cycle in rabbits. Invest Ophthalmol Vis Sci 1991; 32:1609-1613.
90a. Lohse MJ, Benovic JL, Codina J, et al: Beta arrestin: a protein that regulates beta receptor function. Science 1990; 248:1547-1550.
90b. Sears J, Sears M: Circadian Rhythms in Aqueous Humor Formation. In: Civan MM, ed. The eye's aqueous humor: From secretion to glaucomaed by Civan MM. Current topics in membranes, San Diego, CA: Harcourt Brace & Co; 1998.
90c. Chen S, Inoue R, Inomata H, Ito Y: Role of cAMP induced Cl conductance in aqueous humor formation by the dog ciliary epithelium. Br J Pharmacol 1994; 112:1137-1145.
91. Saito Y, Watanabe T: Relationship between short-circuit current and unidirectional fluxes of Na and Cl across the diary epitheliam of the load: demonstration of active Cl transport. Exp Eye Res 1979; 28:71-79.
92. Kishida K, Sasabe T, Manabe R, et al: Electric characteristics of the isolated rabbit cilial body. Jpn J Ophthalmol 1983; 25:407-416.
93. Yamashita H, Yamamoto T: Histochemical observation of distribution of chloride ion around ciliary epithelium. Jpn J Ophthalmol 1989; 33:41-47.
94. Holland MG, Gipson CC: Chloride ion transport in the isolated ciliary body. Invest Ophthalmol Vis Sci 1970; 90:20-29.
95. Yantorno RE, Krupin T, Civan MM: Selective and non-selective ion channels in intact rabbit ciliary epithelium. Invest Ophthalmol Vis Sci 1987; 28(suppl):374.
96. Murakami M, Sears M, Mori N, et al: The loci of carbonic anhydrase activity in the ciliary epithelium of the rabbit eye: an electrophysiological study using isolated ciliary epithelial bilayer. Acta Histochem Cytochem 1992; 25:77-85.
97. Berggren L: Effect of a mercurial diuretic, mersalyl, on in vitro secretory activity of she rabbit eye ciliary processes. Acta Ophthalmol 1970; 48:275-283.
97a. Civan MM: The ins and outs of aqueous secretion. Exp Eye Res 2004; 78:625-631.
97b. Civan MM: Basis of chloride transport in ciliary epithelium. J Membrane Biol 2004; 200:1-13.
98. Friedenwald JS: Carbonic anhydrase inhibition and aqueous flow. Am J Ophthalmol 1955; 39:59-64.
99. Maren TH: HCO3 formation in aqueous humor: mechanism and relation to the treatment of glaucoma. Invest Ophthalmol 1974; 3:479-484.
100. Friedland BR, Maren TH: Carbonic anhydrate: pharmacology of inhibitors and treatment of glaucoma. In: Sears ML, ed. Handbook of experimental pharmacology, pharmacology of the eye, Berlin: Springer; 1984:279-303.
101. Zimmerman TJ, Garg LC, Vogh BP, et al: The effect of acetazolamide on the movement of sodium into the posterior chamber of the dog eye. J Pharmacol Exp Ther 1976; 199:510.
102. Kinsey VE: Comparative chemistry of aqueous humor in posterior and anterior chambers of rabbit eye: Its physiologic significance. Arch Ophthalmol 1953; 40:401-417.
103. Kinsey VE, Reddy DVN: Turnover of carbon dioxide in the aqueous humor and the effect thereon of acetasolamide. Arch Ophthalmol 1959; 62:78-83.
104. Cole DF: Electrochemical changes associated with the formation of the aqueous humor. Br J Ophthalmol 1961; 45:202-217.
105. Sears ML, Yamada E, Cummins D, et al: The isolated ciliary epithelial bilayer is useful for in vitro studies of aqueous humor formation. Trans Am Ophthalmol Soc 1991; 89:131-154.
106. Lutjen-Drecoll E, Lonnerholm G: Carbonic anhydrase in the rabbit eye by light and electron microscopy. Invest Ophthalmol Vis Sci 1981; 21:782-797.
107. Lutjen-Drecoll E, Lonnerholm G, Eichhorn M: Carbonic anhydrase distribution in the human and monkey eye by light and electron microscopy. Graefes Arch Clin Exp Ophthalmol 1983; 220:285-291.
108. Wistrand PJ: Properties of membrane-bound carbonic anhydrase. Ann N Y Acad Sci 1984; 429:195-206.
109. Brown D, Zhu XL, Sly WS: Localization of membrane-associated carbonic anhydrase type IV in kidney epithelial cells. Proc Natl Acad Sci USA 1990; 87:7457-7461.
110. Lonnerholm G, Ridderstrale Y: Intracellular distribution of carbonic anhydrase in the rat kidney. Kidney mt 1980; 17:162-174.
111. Maren TH: The kinetics of HCO synthesis related to fluid secretion, pH control, and CO2 elimination. Annu Rev Physiol 1988; 50:695-717.
112. Murakami M, Sears M, Mori N, et al: The loci of carbonic anhydrase activity in the ciliary epithelium of the rabbit eye: an eleetrophysiological study using isolated eiliary epithelial bilayer. Acta Histochem Cytochem 1992; 25:77-85.
113. Boron WF, Boulpaep EL: The electrogenic Na/BCE dotrans. porter. Kidney Ins 1989; 36:392-402.
114. Cole DF: Rate of entrance of sodium into the aqueous humour of the rabbit. Br J Ophthalmol 1960; 44:225-242.
115. Simon KA, Bonting SL, Hawkins NM: Studies on sodium-potassium activated adenosinetriphosphatase II. Exp Eye Res 1962; 1:253-261.
116. Bonting SL, Caravaggio L, Hawkins NM: Studies on sodium-potassium-activated adenosine triphosphatase: IV. Correlation with cation transport sensitive to cardiac glycosides. Arch Bio Chem Biophys 1962; 98:413-419.
117. Bonting SL, Reeker B: Studies on sodium-potassium-activated adenosine triphosphasase: IV. Inhibition of enzyme activity and aqueous humor flow in the rabbit eye after intravitreal injections of ouabain. Invest Ophthalmol 1964; 3:523-533.
118. Krupin T, Fritz C: Becker B: Maturation of Rb and PAH accumulation by rabbit anterior uvea and choroid plexus. Invest Ophthalmol Vis Sci 1985; 26:159-162.
119. Shiose Y, Sears ML: Localization and other aspects of nucleoside phosphatase activity in the ciliary epithelium of albino rabbits. Invest Ophthalmol 1965; 4:64-75.
120. Mori N, Yamada E, Sears ML: Immunocytochemical localixation of Na/K-ATPa5e in the isolated eiliary epithelial bilayer of the rabbit. Arch Histol Cytol 1991; 54:259-265.
121. Ghosh S, Freitag A, Martin-Vasallo P, et al: Cellular distribution and differential gene expression of the three subunit isoforms of the Na, K-ATPase in the ocular ciliary. J Biol Chem 1990; 265:2935.
122. Martin-Vasallo F, Ghosh S, Coca-Prados M: Expression of Na, K-ATPase subunit isoforms in the human ciliary body and cultured epithelial cells. J Cell Physiol 1989; 141:243-252.
123. Lytton J, Lin JC, Guidotti G: Identification of two molecular forms of (Na+ K+)-ATPase in rat adipocytes. J Bid Chem 1985; 260:1177-1184.
124. Delamere NA, Socci RR, King KL: Alteration of sodium, potassium-adenosine triphosphatase activity in rabbit ciliary processes by cyclic adenosine monophosphate-dependent protein kinase. Invest Ophthalmol Vis Sci 1990; 31:2164-2170.
125. Barany E, Kinsey VE: The rate of flow of aqueous humor: I. The rate of disappearance of para-aminohippuric acid, radioactive Rayopake. and radioactive Diodrast from the aqueous humor of rabbits. Am J Ophthalmol 1949; 32:171-177.78.
126. Reddy DVN: Distribution of free amino acids and related compounds in ocular fluids, lens, and plasma of various mammalian species. Invest Ophthalmol 1967; 6:478-483.
127. Reddy DVN, Thompson MR: Chakrapani B: Amino acid transport across blood-aqueous barrier of mammalian species. Exp Eye Res 1977; 25:555-562.
128. Reddy VN: Dynamics of transport systems in the eye: Friedenwald Lecture. Invest Ophthalmol Vis Sci 1979; 18:1000-1018.
129. Davson H, Matchett PA: The kinetics of penetration of the blood aqueous barier. J Physiol 1953; 122:11-32.
130. DiMattio J, Zadunaisky JA: Glucose transport into the ocular compartments of the rat. Exp Eye Res 1981; 32:517-532.
131. Linner E: A method of determination of time-concentration curves using one single sample and several test substances. Acta Soc Med Ups 1953; 59(3-4):241-242.
132. Kinsey VE, Reddy DVN: Chemistry and dynamics of aqueous humor. In: Prince JH, ed. The rabbit in eye research, Springfield, IL: Charles C Thomas; 1964:218-319.
133. Brubaker K: Flow of aqueous humor in humans. Invest Ophthalmol Vis Sci 1991; 32:3145-3166.
134. Goldmann H, Schmidt T: Studien mittels Applanationstonographie. Doc Ophthalmol 1966; 20:184-213.
135. Bill A, Phillips CI: Uveoscleral drainage of aqueous humour in human eyes. Exp Eyc Res 1971; 12:275-281.
136. Barany EH: Pseudofacility and uveo-scleral outflow routes: some nontechnical difficulties in the determination of outflow facility and rate of formation of aqueous humor. Glaucoma Symposium, Tutzing Castle, 1966, Basel: Karger; 1967:27-51.
136a. Jocson VL, Sears M: Experimental aqueous perfusion in enucleated human eyes after obstruction of Schlemm's canal. AMA Arch Ophthalmol 1971; 86:65.
137. Becker B: The transport of organic anions by the rabbit eye: I. In vitro iodopyracet (Diodrast) accumulation by ciliary body-iris preparations. Am J Ophthalmol 1960; 50:862-867.
138. Becker B: Iodide transport by the rabbit eye. Am J Physiol 1961; 200:804-806.
139. Becker B, Forbes M: Iodopyracet (Diodratt) transport by the rabbit eye. Am J Physiol 1961; 200:461-464.
140. Barany EH: Inhibition by hippurate and probenecid of in vitro uptake of iodipamide and o-iodohippurate: a composite uptake system for iodipamide in choroid plexus,kidney cortex and anterior uvea of several species. Acta Physiol Scand 1972; 86:12-27.
141. Barany EH: The liver-like anion transport system in rabbit kidney, uvea and ehoroid plexus: I. Selectivity of some inhibitors, direction of transport, possible physiological substrates. Acta Physiol Scand 1973; 88:412-429.
142. Barany EH: Organic cation uptake in vitro by the rabbit irisciliary body, renal cortex, and choroid plexus. Invest Oplithslmol 1976; 15:341-348.
143. Shichi H: Biotransformation and drug metabolism. In: Sears ML, ed. Pharmacology of the Eye, Berlin: Springer; 1914:117-148.
144. Shichi H: Glutathione-dependent detoxification of peroxide in bovine ciliary body. Exp Eye Res 1990; 50:813-818.
145. Caprioli J: The ciliary epithelia and aqueous humor. In: Moses K, Hart WM, ed. Adler's Physiology of the Eye, 8th edn, St Louis: CV Mosby; 1987:204-222.
146. DiMattio J, Degnan KJ, Zadunaisky IA: A model for transepithelial ion transport across the isolated retinal pigment epithelium of the frog. Exp Eye Res 1983; 37:409-420.
147. Kinsey VE, Reddy DVN: Transport of glucose across blood-aqueous barriers as affected by insulin. Physiol 1961.1516-1568.
