Aquaporins : The Secret Highways For Water Transport

Hitesh Chavda

Aquaporins are a class of integral membrane proteins or major intrinsic proteins (MIP) that form pores in the membrane of biological cells.

Living organisms are composed mainly of water, whose vital function as a ‘universal solvent’ enables the body to maintain its osmotic, metabolic, thermal, and electrochemical balance, as well as transporting nutrients and catabolites within the cell. It is readily apparent that the movement of water into and out of the cell is essential to adapt to its environment and, therefore, fundamental to life. However, although simple diffusion of water across the lipid bilayer occurs through all biological membranes, its low velocity and finite extent soon became apparent, suggesting the existence of additional pathways for water movement through the membrane. The precise and complete answer only came relatively recently with the discovery of the aquaporins (AQPs), proteinaceous water channels making the membrane 10 to 100 fold more permeable to water than membranes lacking such channels¹. Table 1 shows water content of different adult organs.

Table 1: Water content of different adult organs²

Transmembrane Movement of Water³

Ion fluxes generate osmotic forces, which primarily drive water movement in a physiological system. There are two major routes for osmotically driven water movement across cell layers. Paracellular fluid movement allows the passage of water between the cells, whereas in the transcellular pathway, water traverses through the cells. The latter involves the passage of water through the apical and basal membranes of the cells. There are essentially three major transmembrane pathways for water transport as shown in figure 1. Simple diffusion will allow the passage of water across the cell membrane. All the cell membranes are not very permeable to water; there is a relatively high-energy barrier to diffusion. Therefore, water passage by simple diffusion through the membrane is a slow, non-regulated, temperature-dependent process and is not an efficient means of direct, rapid, or regulated water flux. In second transmembrane pathways, passive co-transport, addition to simple diffusion, in some cells, water can be passively transported in association with the active co-transport of other ions or solutes. This passive co-transport mechanism, where water simply ‘goes along for the ride’ with other solutes, provides a route for the bulk transport of water - even against an osmotic gradient in some cases. This mode of water transport is not regulated in a manner that will contribute to the maintenance of body water homeostasis. Furthermore, simple diffusion or passive co-transport cannot account for the rapid, regulated, and selective transmembrane water movement that occurs in several tissues, like kidney tubules, secretory cells of glands, and red blood cells, during normal physiological processes. While some of these processes are driven by osmotic gradients, others do occur at near isoosmotic conditions. The presence of specific pores (channels) in the cell membrane has long been predicted but the proteins involved in these water channels have only recently been characterized. Only during the past decade has the molecular evidence for the existence of water channels emerged. The first water channel protein was isolated from red blood cell membranes and was named CHIP28 for ‘channel-like integral membrane protein of 28 kDa. This is the landmark discovery that starts the era of molecular water channels, which has permeated virtually every branch of biology and medicine. CHIP28, which is renamed as Aquaporin-1, showed significant amino acid homology to the major intrinsic protein of the lens fibre. Subsequently more and more members of the MIP family were identified through amino acid comparison and functional analysis; the nomenclature was revised to aquaporins (AQPs), water pores, or water channels to reflect the functional nature of this group of proteins.

Figure 1: Movement of water across membranes

Aquaporin Family ²

The existence of canals or channels that allowed the rapid flow of large quantities of water through certain tissues had long been suspected, and this process had to be extremely selective to retain only tiny essential molecules within the cells. In 1988, intent on their work on red blood cells, young scientists of Peter Agre’s team came across a strange protein lodged in the cells’ membrane, i.e. the envelope that surrounds each cell. A few more years’ study revealed that this protein forms in effect a kind of pore through which water flows rapidly in and out of the red blood cells. The first water canal had finally been identified and was named Aquaporin number 1. Since date over 200 different aquaporins have been identified in all kinds of tissues, in mammals, invertebrates, plants, and microorganisms. In humans, 13 aquaporins have been identified to date (Table 2), although this number pales in comparison to the greater than 100 related proteins found in plants and bacteria ⁴.

Table 2: Characteristics of human aquaporins ^{4, 5}

Molecular Structure of Aquaporins ^{3, 6, 7}

A unique architectural motif of aquaporins is their hourglass pseudosymmetric structure shown in Figure 2. Aquaporins are made up of six transmembrane α-helices arranged in a right-handed bundle, with the amino and the carboxyl termini located on the cytoplasmic surface of the membrane. The amino and carboxyl halves of the sequence show similarity to each other, in what appears to be a tandem repeat. There are also five interhelical loop regions A to E that form the extracellular and cytoplasmic vestibules. A highly conserved three amino acid motif, Aspargine-Proline-Alanine (NPA) is present in the B and E loops of almost all AQPs. The NPA motifs, although located in the non-membrane spanning helices are, however, inserted into the membrane. The intracellular loop B and extracellular loop E fold into the membrane and interact with one another, forming a 3D hourglass model, characterized by wide external openings to the channel with a narrow central constriction where the NPA motifs interact in such a way, forming the functional water pore, where the water flows through. Although each individual AQP is a functional water pore, they assemble in groups of four identical protein channels, tetramers, in the membranes. Water movement through each pore can be bidirectional, though the actual direction of water flow in a physiological system is determined by the osmotic gradient or bulk absorption of solutes.

Figure 2: ‘Hour-glass’ model of AQP subunit

Movement of Water through Aquaporin ^{2, 8}

In liquid state of water, the molecules are loosely bound together by chemical links between the oxygen and hydrogen atoms. When the water molecules reach a pore of aquaporins, they rush inside in a single file with the oxygen atom head first towards the pore. Once they have reached the canal’s center, they are literally snatched aside via the interaction of hydrogen bonds between the oxygen of the water molecule and the asparagines in the two NPA motifs, which line the canal. This causes the molecules to perform a pirouette (whirl) and as a result the single file is broken and the molecules proceed to the opposite end of the pore, with the hydrogen atoms head first. This molecular ballet is even more ingenious ballet than it seems at first glance. Aquaporins also prevent protons, the solitary hydrogen atoms, from passing through. Normally, protons use water molecules as a means for movement from one place to another. A file of water molecules is known as a “proton path” where protons literally piggyback from one molecule to another. If the path is disturbed, the protons cannot proceed and they return to where they came from. This is exactly what happens inside an aquaporin: the molecules’ little pirouette disturbs the “proton path” and the protons are sent back while the water molecules carry on. Such a stratagem is simply because cells need protons to charge, if cells lose protons, they will also lose their energy. Therefore, they must not let them escape through their pores, at all costs. This method is devilishly clever, where one second is sufficient for one billion water molecules to pass through a cell’s membrane without a single proton escaping². Transport of water can be induced either by osmotic pressure differences, giv­ing rise to the osmotic permeability (P_f), or by a differ­ence in tracer concentrations, giving rise to the diffusion permeability (P_d)⁸. The diffusion permeability accounts for water molecules transported through the entire channel, whereas osmotic permeability accounts for wa­ter entering and leaving the channel at its exits such that P_f/P_d is greater than one. The ratio P_f/P_d measures the number of essential steps that water molecules need to pass the channel. Aquaporins, in some cases, also facilitate the transport other small-uncharged solutes, such as glycerol, carbon dioxide, ammonia and urea across the membrane depending on the size of the pore. The different aquaporins contain differences in their peptide sequence, which allows for the size of the pore in the protein to differ between aquaporins. The resultant size of the pore directly affects what molecules are able to pass through the pore, with small pore sizes only allowing small molecules like water to pass through the pore.

Measuring Water Channels⁶

Difficulty in developing the techniques for the measurement of water channels is due to two characteristics of water channels in native membranes. The first is that, when water flowing across the membrane through an aquaporin molecule, it does not does not carry any charge, so electrical current cannot be detected. The second is that the background water per­meability of the lipid bilayer is relatively high. In general, water channels only confer an increase in membrane water permeability relative to the lipid bilayer of one to two orders of magnitude, compared with an increase of over three orders of magnitude in the case of ion channels. This makes it difficult to detect minor changes in net water flow resulting from the opening of a small number of water channels against an already high background flow. For a water channel with an osmotic permeability (P_f) of 6 x 10^-20 m³s^-1 per channel subunit, an equivalent hydraulic conductivity of 4.4 x 10^-22 m³s^-1MPa^-1 per channel subunit can be calculated. Therefore, for a gradient of 0.1 MPa (1 bar) in osmotic pressure, which is physiologically realistic, 1.5 x 10⁶ water molecules will flow through each channel subunit per second. This is similar to the flux through an ion channel for a typical driving force. Water flow in walled plant cells can be assayed by measuring pressure changes over time using the pressure probe. For membrane vesicles or wall-less cells, it is necessary to rely on monitoring changes in volume generally measured on a large population of cells or vesicles using optical methods, like light scattering. Unfortunately, in both of these cases, it is not possible to resolve flow to the level of 10⁶ water molecules per second. In addition to this, for relatively large changes in the volume of cells, the volume change is necessarily accom­panied by a significant change in membrane surface area. Since the lipid membrane is not very extensible, water flow must necessarily be accompanied by incorporation or deletion of material into the bounding membrane of the cell by exocytosis or endocytosis, respectively. Another important osmotic parameter that measures the select­ivity of aquaporins is the reflection coefficient (a), which represents a quantitative index of the inter­action between water and solutes as they traverse the membrane. Interactions should be very strong in pores, and the knowledge of reflection coefficients could help indicate the slippage of certain test solutes through aqua­porins. Titrating water channels with mercurial reagents and measuring the overall hydraulic conductivity of the cells may measure reflection coefficients of aquaporins. However, this method allows some passage of water as well as solutes. To avoid the latter difficulty is to use selective inhibitors that specifically block these transporters without affecting water channel activity.

Characteristics of Water Channel Activity in Membranes ⁶

High water permeability

A high value of osmotic water permeability is the first indication of water-channel activity, i.e. P_f much greater than 10x10^-6m s^‑1, which is representative of reported water permeability across lipid vesicles. Typical values for membranes with water channels are between 100 and 200x10^-6 m s^‑1. The erythrocyte membrane, studded with AQP1 has a value of 200x10^-6m s^‑1.

Low activation energy for water transport

If the flow of water through water channels is essentially a viscous flow across a pore, the temperature dependence of P_f should be comparable to that for the self-diffusion of water or for the viscosity of water. This has been found for plant membranes in the presence of water channels. The reason for the low activation energy is that water moving across a channel does not have to overcome a large energy barrier. On the other hand, water movement through the membrane would need to surmount the high-energy barrier of water partitioning into hydrophobic lipid phase. It should be noted that temperature effects on the membrane lipid or temperature effects on the water channels via gating might conceal the actual temperature dependence of water flow through the channel. Ratio of osmotic water permeability to diffusional water permeability

The P_f/P_d ratio characterizes the predominant pathway of water movement. A ratio of unity is typical for independent diffusion into phospholipid vesicles, while membranes containing water channels show ratios greater than unity. This ratio has been used to calculate the number of water molecules that line up in a file across the pore or the effective radius of the pore. Using stopped-flow, a P_f/P_d ratio of 7 for tonoplast-enriched vesicles from wheat roots was founded.

Block of water channels by mercurial reagents^{9, 10}

Mercurial and sulphydryl reagents in general block aquaporins, with some exceptions. Mercurials act by binding to SH-groups of proteins. Block can usually be reversed with a scav­enging agent such as the reducing SH-compound 2-mercaptoethanol. Different mercurials have different effects depending on the ability of the substance to diffuse across the membrane, which may be influenced by pH and temperature. Mercuric chloride, HgCl₂, is able to attack intramembraneous sites because it forms a relatively lipophilic ion pair that can diffuse across the membrane. The mercury-sensitive sites on aquaporins have been probed by molecular techniques. The sensitive cysteine residue in animal aquaporins is somewhat removed from the sensitive residues in plant aquaporins.

Gating of water channels

Currently, it is not known if water channels are gated, i.e. whether they open and close stochastically in the same manner as ion channels. The nodulin-26¹¹ and bovine lens¹² MIPs form ion channels in lipid bilayers and these channels show gating. Nodulin-26 is also water permeable, but the osmotic permeability is 3.2x10^-6 ms^‑1 per channel subunit, over an order of magnitude lower than that for AQP1. It is not known if the water and ions that move through nodulin-26 go via the same pathway. There are some examples in the literature showing that P_f can be a function of turgor pressure or osmotic pressure, components of the gradients that determine water flow analogous to ion concentration and voltage for ion flow.

Reverse genetics

Reverse genetics may provide another route to studying aquaporin function in plants. The task to be achieved is indeed enormous and phenotypic compensation in single-gene mutants by close homologues may occur. Nevertheless, genetic evidence that aquaporins play a crucial role in plant growth and development has started to emerge over the last two years. The mod mutation, which confers a recessive loss of self-incompatibility in Brassica, is the result of a mutation in an aquaporin-like gene¹³. A PIPlb antisense transgene is used to reduce the expression of PIP1 genes in Arabidopsis¹⁴. The result was a decrease in water permeabil­ity of isolated leaf protoplasts and an increase in root mass, which could reflect a compensatory mechanism resulting from reduced root hydraulic conductivity.

Transport Capacity and Selectivity¹⁵

AQP4 has the highest water permeability, whereas AQP0 and AQP3 have the lowest. The permeability of AQP1 is 6x10^-14 cm³s^-1. In addition to water transportation capability, AQP1 also transports CO₂, AQP (3, 7, and 9) transport glycerol and urea and AQP9 transports carbamides, polyols, purines, arsenite, and pyramidines.

Physiology and Pathophysiology of Aquaporins ⁴

Kidney

The kidney is primarily responsible for filtering and eliminating toxic substances from the blood. This function is achieved by the filtration of blood in nephrons, which have important functions in the reabsorption of water, active solute transport, and acid-base balance. To maintain water balance, much of the water in the filtrate (>99%) must be reabsorbed before it leaves the kidney. Approximately 80% of the water is reabsorbed back in to the blood by the epithehum lining, the proximal tubule, and thin descending limb of the nephron. The remaining 19% is reabsorbed across the epithelia that line the collecting duct. The proximal tubule, thin descending limb, and collecting duct of the nephron express the higher numbers of AQPs. AQP1 is expressed in the apical and basolateral membranes of the epithelia lining the proximal tubule and thin descending limb, as well as the endothelium lining the descending vasa recta blood vessel.

Respiratory tract

Water transport in the airways of humans and other mammals is an essential component of important physiological processes. To facilitate the transport of water, AQPs are expressed throughout the respiratory tract. In the ciliated epithelia that line the upper airway, AQP5 is expressed in the apical membrane, whereas AQP4 is expressed in the basolateral membrane. AQP5 is also expressed in secretory cells within the submucosal glands, and in the apical membrane of type 1 pneumocytes in the alveoli within the lower airway. AQP3 is expressed in basal epithelia in the upper airway, whereas AQP1 is found predominantly in the vascular endothelia within the venule and capillary beds surrounding the airways and alveoli.

Nervous system

Maintenance of the ionic and osmotic composition and volume of intestinal, glial, and neuronal compartments within the brain and spinal cord is essential for normal functioning of nervous system. Small changes in ion or solute composition can dramatically alter neuronal signaling and information processing. AQPs are important in central nervous system water homeostasis and neuronal signaling. AQP1 is expressed in the apical membrane of the choroid plexus epithelium and could be involved in the production of cerebral spinal fluid. Moreover, AQP1 is expressed in the endothelia of tumors but not in normal brain endothelium, which suggests the possible involvement of AQP1 in tumor growth. AQP4 and AQP9 expressed in astrocytes in the brain and spinal cord. AQP4 expression is localized to the membrane of astrocytic end-feet adjacent to vascular endothelium, suggesting a role in vascular-glial water transport.

Eye

Apart from the production of tears in lacrimal glands, water movement across endothelium and epithehum is essential for maintaining corneal and lens transparency. The continuous formation of aqueous humor in the non-pigmented epithelia of the ciliary body and its drainage through the trabecular meshwork and canals of Schlemm is required to maintain intraocular pressure. Transport of water across retinal epithelium helps to maintain retinal adhesion and integrity. Five AQPs are expressed in human eye:  AQP0 in the lens; AQP1 in corneal endothelium, ciliary and lens epithelia and trabecular meshwork; AQP3 in conjunctiva; AQP4 in ciliary epithelium and retinal Müller cells; and AQP5 in corneal and lacrimal gland epithelia.

References:

1. Agre P. Aquaporin water channels (Nobel Lecture). Angew. Chem. Int. Ed. Engl. 2004, 43:4278–4290.
2. Prolune, Aquaporins: A Nobel Prize for the Water Ferries, December 11, 2003.
3. Krane CM, Kishore BK. Aquaporins: the membrane water channels of the biological world. Biologist, 2003, 50 (2): 81-86.
4. Castle NA. Aquaporins as targets for drug delivery. Drug Discovery Today, 2005, 10(7): 485-493.
5. Itoh T, Rai T, Kuwahara M, Ko SB, Uchida S, Sasaki S, Ishibashi K. Identification of a novel aquaporin, AQP12, expressed in pancreatic acinar cells Biochem Biophys Res Commun. 2005, 330(3): 832-838.
6. Tyerman SD, Bohnert HJ, Maurel C, Steudle E, Smith JAC. Plant Aquaporins: their molecular Biology, Biophysics and significance for plant water relations. Journal of Experimental Botany, 1999, 50:1055-1071.
7. Chakrabarti N, Tajkhorshid E, Roux B, Pomes R. Molecular Basis of Proton Blockage in Aquaporins. Structure, 2004, 12:65–74.
8. Benoit R, Klaus S. Computational Studies of Membrane Channels. Structure, 2004, 12: 1343–1351.
9. Zhang WH, Tyerman SD. Inhibition of Water Channels by HgCl₂ in Intact Wheat Root Cells. Plant Physiology, 1999, 120: 849-858.
10. Hoch BS, Gorfien PC, Linzer D, Fusco MJ, Levine SD. Mercurial reagents inhibit flow through ADH-induced water channels in toad bladder, Am J Physiol Renal Physiol., 1989,  256: F948-F953.
11. Lee JW, Zhang Y, Weaver CD, Shomer NH, Louis CF, Roberts DM. Phosphorylation of nodulin 26 on serine 262 affects its voltage-sensitive channel activity in planar lipid bilayers. Journal of Biological Chemistry. 1995, 270: 27051-27057
12. Ehring GR, Zampighi G, Horwitz J, Bok D, Hall JE. Properties of channels reconstituted from the major intrinsic protein of lens fiber membranes. Journal of General Physiology. 1990, 96:631-664.
13. Ikeda S, Nasrallah JB, Dixit R, Preiss S, Nasrallah ME. An aquaporin-like gene required for the Brassica self-incompatibility response. Science. 1997, 276:1564-1566.
14. Kaldenhoff R, Grote K, Zhu JJ, Zimmermann U. Significance of plasmalemma aquaporins for water-transport in Arabidopsis thaliana. The Plant Journal. 1998, 14, 121-128.
15. Harjanto. Aquaporin Water Channels: Physiology and Medicine. Folia Medica Indonesiana, 2004, 40(1): 3-9.

About Authors:

Hitesh Chavda is working as a Lecturer at Shri Sarvajanik Pharmacy College, Mehsana, India. He has completed his M. Pharm from Hemchandracharya North Gujarat University, Patan, India. He works on the technologies for solubility enhancement of drugs with poor water solubility and hydrogel based drug delivery systems.

Dr.C.N.Patel is a Principal and Professor of Pharmaceutical Chemistry at Shri Sarvajanik Pharmacy College ,India . He has published numbers of publications at national and international level.  He is deeply involved in national social services (NSS) activities.