Aquaporin-1: The Original Water Channel
Sarah Nelson
Biochemistry and Molecular Biophysics
The University of Arizona
Abstract
Prior
to the discovery of the aquaporin-1 water channel, all cross-membrane
water transport was attributed to the passive process of simple
diffusion (Borgnia et al., 1999). However, this model fails to
accommodate the large variability in water permeability observed in a
variety of cells (Solomon, 1968; Knepper et al., 1996). After Peter
Agre’s chance discovery of aquaporin-1 these apparent
contradictions were resolved by subsequent findings related to the
structure of the protein and the mechanism associated with water
transport. Aquaporin-1 is a tetrameric membrane protein comprised of
four identical and independently functioning subunits. The monomeric
unit consists of six transmembrane alpha-helices
and two non-membrane-spanning alpha-helices—each
containing the NPA (Asn-Pro-Ala) motif (Walz et al., 1997). The
mechanism of water transport and proton exclusion (critical to the
maintenance of the established proton gradient) relies both on size
exclusion (Kong and Ma, 2001) and hydrogen bond interactions between
water molecules and the two asparagine residues (76 and 192) found in
each instance of the NPA motif (Murata et al., 2000)—which serve to
break the hydrogen bonded chain of the water molecules, inhibiting
the movement of protons. The effects of two known aquaporin
inhibitors are also discussed: mercury, which serves to sterically
block water transport through the aqueous pore (Savage and Stroud,
2007), and tetraethylammonium, which has been shown to interact with
loop E of the protein (Brooks et al., 2000). Inhibition of
aquaporin-1 has many potentially clinical implications (King and
Agre, 1996), thus it is crucial to continue searching for clinically
feasible compounds that serve as inhibitors of the aquaporin-1 water
channel.
Introduction
Water is abundantly present in all living organisms; its role in the maintenance of life cannot be understated. Water can easily cross most biological membranes via the process of simple diffusion or osmosis (Borgnia et al., 1999) and up until the later part of the twentieth century, osmosis was assumed to account for all cross membrane water movement (Borgnia et al., 1999). However, scientific experimentation suggested that the process of simple diffusion could not account for the variability in water permeability observed in different types of cells. Some biological membranes, particularly those of erythrocytes (Solomon, 1968) and cells found in the descending limbs of Henle’s loop and the renal proximal tubules (Knepper et al., 1996) exhibit significantly higher water permeability when compared with the water permeability of other types of cells.
Further experiments propelled the notion that a protein may be responsible for the non-diffusion movement of water across cellular membranes. HgCl2 was shown to act as an inhibitor of the increased water permeability of several highly water permeable cells (Macey, 1984). Also, changes in the water permeability of epithelial cells were shown to occur at a rate much higher than the rates generally associated with changes in membrane composition (Borgnia et al., 1999). Finally, the activation energy associated with increased water permeability is no higher than that associated with the diffusion of water through a solution (Solomon, 1968)—indicating that whatever was responsible for this marked increase in water permeability was no more complex than a simple cross-membrane protein channel (Solomon, 1968). Despite unanimous agreement about the nature of the mechanism responsible for increased water permeability, all attempts to attribute this phenomenon to a previously studied protein were unsuccessful.
The initial discovery of the protein responsible for water transport took place in 1992 and was purely a matter of chance. In the process of attempting to isolate a 32 kDa membrane spanning component of the red blood cell antigen Rh, Peter Agre stumbled upon a 28 kDa polypeptide assumed to be part of the component in question. Further study of the component revealed that this was not in fact part of the Rh antigen, but instead a polypeptide composed primarily of hydrophobic amino acids found in two forms, N-glycosylated and nonglycosylated (Borgnia et al., 1999). The polypeptide exhibited high expression levels in the membranes of erythrocytes, renal proximal tubules and descending limbs of Henle’s loop—all locations known for their high water permeability (Knepper et al, 1996)—and was initially named CHIP28 (standing for Channel-like integral protein of 28 kDa) (Borgnia et al., 1999).
Although Peter Agre has been credited with the discovery of the first water channel protein (and was later awarded the 2003 Nobel Prize in Chemistry in recognition of the vast implications of his discovery), it was John C. Parker who initially suggested that CHIP28 might explain the variable water permeability observed in different types of cell membranes (Borgnia et al., 1999). This assertion was later confirmed after the RNA of the polypeptide was transfected into Xenopus oocytes and a dramatic increase in osmotic swelling was observed (Preston et al., 1992). The 28 kDa polypeptide was initially named CHIP28 (standing for Channel-like integral protein of 28 kDa); after several similar water-channel proteins were subsequently discovered and the need for a functionally significant name became apparent, CHIP28 was renamed aquaporin-1 (AQUAPORIN-1) and made a member of the newly emerging aquaporin family of water channel proteins (Borgnia et al., 1999).
In the fifteen years since the initial discovery of aquaporin-1, research regarding the structure and function of the protein has surged. While there are still contentions regarding the mechanism of the function of the protein and methods of its inhibition, research on the topic continues to flourish. The following paper will discuss, in detail, the structure and function both of water transport and inhibition in the water channel protein, aquaporin-1.
Structure of Aquaporin-1 Water Channel Protein
Aquaporin-1 is a tetrameric protein comprised of four identical and independently functioning monomer subunits. The aquaporin-1 monomer consists of six membrane-spanning alpha-helices and two intra-membrane alpha-helices tilted 21 degrees with respect to the normal of the membrane bilayer (Walz et al., 1997), with both the amino and carboxy termini found on the cytoplasmic side of the membrane (Walz et al., 1997) (Figure 1a). The polypeptide is also comprised of many small amino acids, including glycine and alanine (Ren et al., 2000).
The structural motif seen in aquaporin-1 (six transmembrane helices forming a barrel shape) is a classic identifier of a much larger group of cross-membrane channel proteins to which aquaporin-1 belongs—the ubiquitous major intrinsic protein (MIP) family (Park and Saier, 1996). The MIP family is approximately 2.5 – 3 billion years old and is characterized as a family of membrane spanning proteins that serve as selective transporters of water, small solutes and ions (Park and Saier, 1996). Among the MIP family, the NPA motif, an Asn-Pro-Ala repeat present twice in the polypeptide chain, is highly conserved (de Groot et al., 2003). Given the high conservation of this repeat, it has long been suspected to perform an important role in the mechanism responsible for the selective channel properties exhibited by the MIP family of proteins (Reizer et al., 1993).
In aquaporin-1, the NPA motif can be found on non-membrane crossing helices B and E (HB and HE, respectively) oriented 180 degrees to each other near the center of the protein (Figure 2) (Jung et al., 1994). The two asparagines residues (76 and 192) found in each of the NPA motifs exhibit two fold symmetry with respect to each other (Kong and Ma, 2001) and form the central constriction of the aqueous channel. Moving away from this central constriction towards both the cytoplasmic and extracellular sides of the protein the diameter of the channel increases, creating a pore with an hourglass-like structure (Jung et al., 1994).
Each aquaporin-1 tetramer consists of four independently functioning monomeric water channels (Cheng et al., 1997) and exhibits four-fold symmetry (Figure 1c). At the center of the protein, a central pore is formed by the interior edges of the four monomeric subunits (Figure 1c). At its widest, this pore is 8.5 Å in diameter and is lined with hydrophobic residues eliminating it as a possible pathway for selective water transport (Murata et al., 2000).
Each monomer contributes equally to the stability of the tetramer; each monomer interacts with the two adjacent monomers via residues found in the membrane spanning alpha-helices (Murata et al., 2000). These interactions are stabilized by hydrogen bonds between the Ser-59, Thr-62, and Gln-65 residues of one monomer and the Gln-148, Cys-152 and Thr-156 residues of the neighboring monomer (Murata et al., 2000). Interactions between the four loop A regions (on the extracellular side) of each monomer and the four loop D (on the cytoplasmic side) of each monomer might also serve to stabilize the tetramer (Murata et al., 2000). Aquaporin-1 is an insoluble in water, suggesting that the tetramer might be further stabilized by interactions between the polypeptide and the surrounding lipid molecules. Candidates for such interactions can be found on helices 3 and 6, as well as loop B and loop E (Murata et al., 2000).
Water Transport Mechanism and Proton Exclusion
The aqueous pore of the aquaporin-1 monomer is the site of water transport and is, at its narrowest, 3 Å in diameter—wide enough for 2.8 Å wide water molecules to traverse the constricted region, but effective at providing a physical barrier to larger solutes and ions (Murata et al., 2000). Water molecules in the vicinity of the central pore do not exhibit thermodynamic differences from the bulk movement of water molecules in solution (Ren et al., 2000). The activation energies observed for the movement of water across the central pore are low and correspond approximately to the energy associated with hydrogen bonds. This suggests that the breaking of one or several hydrogen bonds plays an essential role in the selective transport of water through the central pore of the aquaporin-1 monomer (Murata et al., 2000). It is important to note that water movement via aquaporin-1 is bidirectional and dictated by the osmotic gradient present in the environment (Borgnia et al., 1999).
The transport of water by aquaporin-1 is not accompanied by a change in the proton gradient across the lipid bilayer. In the presence of hydrogen bonding, protons can easily move from one water molecule to another via the network of hydrogen bonds present. If there were no mechanism in place to regulate or prevent the movement of protons through the aqueous pore of aquaporin-1, the proton gradient across the cell membrane necessary for normal cell function would be compromised. However, the proton flux through aquaporin-1 is restricted to a rate one-thousand times slower than the rate at which water is transported through the monomeric pore—clearly indicating that a mechanism is in place to prevent protons from leaking through the water channel of the aquaporin-1 protein (de Groot et al., 2003).
Molecular dynamics simulations of proton transport with quantum
mechanically derived proton hopping rates (Q-HOP MD) (Lill and Helms, 2001) performed on the aquaporin-1 monomer have clearly identified the NPA region as the location of the proton exclusion mechanism (de Groot et al., 2003). Protons placed in the aqueous pore of the aquaporin-1 monomer were observed quickly leaving the area; almost all of the protons involved in the simulation were observed leaving the pore via the side of the NPA motif region that they were originally placed—very rarely was a proton observed crossing the NPA motif in the process of leaving the pore region (de Groot et al., 2003). A free energy profile of the aqueous pore created with the data collected in the Q-HOP MD simulations demonstrates that the NPA region exhibits the highest free energy barrier to protons found in the aqueous pathway (de Groot et al., 2003).
The evidence for the NPA region as the site of proton exclusion, as well as the inherent association between protons and the hydrogen bonding of water molecules suggests that the water selectivity mechanism and proton exclusion mechanism are woven together at the site of the NPA motif. Currently, the most widely accepted mechanism for the movement of water molecules through the aqueous core also provides an account for the observed proton exclusion.
Water molecules are limited to passing through the constricted region of the channel pore one molecular at a time due to the small size of the pore (3 Å) relative to the size of one water molecule (2.8 Å) (Murata et al., 2000). The wider regions of the aqueous pore are lined with hydrophobic residues (Ile-60, Phe-24, Leu-149 and Val-176) (Figure 3), inhibiting the formation of hydrogen bonds between water molecules and residues lining the surface of the pathway—allowing the water molecules to flow freely and uninhibited through the wider regions of the aqueous pore (Kong and Ma, 2001). The carboxyl terminus of both loop B and loop E face outward (toward the extracellular and cytoplasmic regions, respectively), creating a positive electrostatic field generated by the dipole moment of the two helices (Murata et al., 2000). As water molecules move towards this electrostatic field they are oriented toward the sides of the NPA motifs facing the surface of the constriction region, such that they are primed to hydrogen bond with either Asn-192 or Asn-76 (of the NPA motif)(Murata et al., 2000). In the process of doing so, the hydrogen bond between the two neighboring water molecules is broken—creating a barrier to proton transfer across the aqueous pore of the aquaporin-1 monomer (Figure 4) (Kong and Ma, 2001).
Results from site-directed mutagenesis studies of Asn-192 and Asn-76 have confirmed the importance of these two residues with regard to aquaporin-1 function. Substituting the two Asparagine residues with non-polar leucine residues led to the total malfunction of the aqueous pathway (Kong and Ma, 2001). This was attributed to both the increased hydrophobicity that the two Leu residues introduced to the aqueous pore and the subsequent decrease in water density in the constriction region following the mutation (Kong and Ma, 2001), making it impossible for the water channel to function in any capacity. However, while unlikely, the possibility of a collapse in the pore following the introduction of the two leucine residues has not been ruled out as a possible explanation for the mutant protein’s defunct aqueous pathway.
The His-180 residue may also play an important, as yet undiscovered, role in the water channel mechanism (Murata et al., 2000). His-180 is present in all members of the MIP family that function as selective water channels, but is, however, supplanted by glycine in MIP proteins that function as glycerol transporters—indicating that its function, whatever it may be, is specific to the water transport mechanism of aquaporin-1 (Murata et al., 2000).
Mercurial Inhibition of Aquaporin-1
By the time Agre discovered the Aquaporin-1 water channel protein in 1992, it had long been acknowledged that mercury, in particular HgCl2, functioned as an inhibitor of water channel function. Shortly after the discovery of the aquaporin-1 water channel, researchers identified Cys-189 as the sole site of mercurial inhibition of aquaporin-1 (Preston et al., 1993).
The location of the cysteine residue is inconsequential (Kuang et al., 2001). C189S mutants are susceptible to mercury inhibition when residues Ser-71, Gly-72 or Ala-73 residues on the inner surface of the aqueous pore were replaced with cysteine (Kuang et al., 2001). This finding indicated not only that the availability of a cysteine was the only requirement for mercurial inhibition of the aquaporin-1 protein, but that also residues 71-73 can easily be reached by extracellular materials, despite their location deep within the aqueous channel.
Recently, the structural characteristics of aquaporin (specifically, aquaporin-1) mercury inhibition were determined via x-ray crystallography using aquaporin-Z, the aquaporin-1 homologue found in bacteria (Savage et al., 2007). While the aquaporin-Z protein contains all other residues critical to the water transport mechanism (those contained in the NPA motifs), it does lack the Cys-189 residue responsible for mercury inhibition (Preston et al., 1993). However, the Thr-183 residue found in aquaporin-Z is homologous to the Cys-189 residue responsible for mercury inhibition in aquaporin-1. A T183C mutant was produced via site-directed mutagenesis, such that mercurial inhibition, as it occurs in the aquaporin-1 monomer, could be adequately observed and studied (Savage et al., 2007).
When the T183C mutant was exposed to mercury, two atoms bound to the aquaporin-1 protein (Hg1 and Hg2). Hg1 was located in the central pore and Hg2 in the interstitial cavity just outside the pore (Figure 5) (Savage et al., 2007). The location of Hg1 was unexpected, given the large distance (5.6 Å) between it and the Cys-183 residue (Savage et al., 2007). Hg1 did, however, form electrostatic interactions with neighboring residues Ser-184 and His-174 (Figure 5) (Savage et al., 2007). Conversely, Hg2 bound directly to the T183C residue (Savage et al., 2007).
Despite the interstitial location of Hg2 no evidence for conformational or structural change within the protein was found (Savage et al., 2007). Consequently, the inhibition of water transport through the central pore must be due to the binding of Hg1 in the pore and is likely sterically induced (Figure 5) (Savage et al., 2007). This steric mechanism of inhibition will allow for mercury to be used in further studies of alternative functions of aquaporin-1, by allowing researchers to distinguish between the effects of the water channel found in each monomeric subunit and the controversial functions recently proposed (Yu et al., 2006) for the tetrameric pore at the center of the aquaporin-1 tetramer (Savage et al., 2007).
Tetraethylammonium Inhibition of Aquaporin-1
Tetraethylammonium chloride (TEA) was initially evaluated as a potential inhibitor of the aquaporin-1 water channel, given its established role as an ion channel inhibitor (MacKinnon and Yellen, 1990). In studies using aquaporin-1 expressing Xenopus oocytes, kidney cells and cells from the MDCK cell-line, TEA was shown to selectively inhibit the water channel function of aquaporin-1 in a concentration dependent and reversible manner (Brooks et al., 2000; Yool et al., 2002).
Exposure of the aquaporin-1 mutant, Y186F (found on the loop E region of the protein and chosen for the close proximity to the mercury binding residue Cys-183), to TEA failed to inhibit aquaporin-1 water permeability even when large concentrations of TEA were used (Brooks et al., 2000). Consequently, loop E was identified as the region of the protein structurally responsible for TEA inhibition of aquaporin-1 (Brooks et al., 2000).
Discussion
The aquaporin-1 water channel has been demonstrated to be a relatively simple protein that has played a major role in the development cellular life. The barrel-like structure consisting of six cross membrane alpha-helices and two intra membrane alpha-helices is far from complex when compared to the structures of other biologically important proteins. Similarly, the mechanism for the transport of water and exclusion of protons is also relatively straightforward.
The importance of the aquaporin-1 water channel in the context of the living cell and on an even larger scale, in an organism, cannot be understated. In recent years, new light has been shed on the vast clinical importance that these small water channels hold. Aquaporin-1 can be found in every type of cell that relies upon cross-membrane water transport (Yang et al., 2006). Studies of aquaporin-1 knockout mice have revealed that the influence of aquaporin-1 reaches far beyond the domain of water transport and influences an incredibly diverse set of other biological events, including but not limited to, tumor angiogenesis (Saadoun et al., 2005), urine concentrating ability (Ma et al., 1998; Schnermann et al., 1998) and cell migration (Verkman, 2005). Further, data now suggests that there may be an emerging role for the clinical use of aquaporin-1 inhibitors pharmaceutically as diuretics (Yang et al., 2006) and as potential therapies for disorders caused by problematic aquapoprin function such as pulmonary edema, glaucoma and polycystic kidney disease (Brooks et al., 2000). Considering the widespread clinical implications of the aquaporin-1 water channel protein, it is important for research on the protein and its inhibition to continue forward—particularly in the search for novel aquaporin-1 inhibitors. Mercury, while demonstrated to be a very effective aquaporin inhibitor and useful in the study of the mechanism of water transport, has very little use in a clinical setting given its lack of specificity for the aquaporin-1 water channel and its inherent toxicity to cellular life (Yang et al., 2006). Tetraethylammonium has shown potential as a clinically applicable inhibitor and hopefully will be the first of many new compounds identified as aquaporin-1 inhibitors.
Acknowledgements
The author would like to thank her peers for their patience and help with Pymol and Dr. Park and Dr. Osterhout for their guidance through this project.
References
Borgnia, M., Nielson, S., Engel, A. and Agre, P. (1999) Cellular and molecular biology of the aquaporin water channels. Annu. Rev. Biochem. 68, 425-458.
Introduction
Water is abundantly present in all living organisms; its role in the maintenance of life cannot be understated. Water can easily cross most biological membranes via the process of simple diffusion or osmosis (Borgnia et al., 1999) and up until the later part of the twentieth century, osmosis was assumed to account for all cross membrane water movement (Borgnia et al., 1999). However, scientific experimentation suggested that the process of simple diffusion could not account for the variability in water permeability observed in different types of cells. Some biological membranes, particularly those of erythrocytes (Solomon, 1968) and cells found in the descending limbs of Henle’s loop and the renal proximal tubules (Knepper et al., 1996) exhibit significantly higher water permeability when compared with the water permeability of other types of cells.
Further experiments propelled the notion that a protein may be responsible for the non-diffusion movement of water across cellular membranes. HgCl2 was shown to act as an inhibitor of the increased water permeability of several highly water permeable cells (Macey, 1984). Also, changes in the water permeability of epithelial cells were shown to occur at a rate much higher than the rates generally associated with changes in membrane composition (Borgnia et al., 1999). Finally, the activation energy associated with increased water permeability is no higher than that associated with the diffusion of water through a solution (Solomon, 1968)—indicating that whatever was responsible for this marked increase in water permeability was no more complex than a simple cross-membrane protein channel (Solomon, 1968). Despite unanimous agreement about the nature of the mechanism responsible for increased water permeability, all attempts to attribute this phenomenon to a previously studied protein were unsuccessful.
The initial discovery of the protein responsible for water transport took place in 1992 and was purely a matter of chance. In the process of attempting to isolate a 32 kDa membrane spanning component of the red blood cell antigen Rh, Peter Agre stumbled upon a 28 kDa polypeptide assumed to be part of the component in question. Further study of the component revealed that this was not in fact part of the Rh antigen, but instead a polypeptide composed primarily of hydrophobic amino acids found in two forms, N-glycosylated and nonglycosylated (Borgnia et al., 1999). The polypeptide exhibited high expression levels in the membranes of erythrocytes, renal proximal tubules and descending limbs of Henle’s loop—all locations known for their high water permeability (Knepper et al, 1996)—and was initially named CHIP28 (standing for Channel-like integral protein of 28 kDa) (Borgnia et al., 1999).
Although Peter Agre has been credited with the discovery of the first water channel protein (and was later awarded the 2003 Nobel Prize in Chemistry in recognition of the vast implications of his discovery), it was John C. Parker who initially suggested that CHIP28 might explain the variable water permeability observed in different types of cell membranes (Borgnia et al., 1999). This assertion was later confirmed after the RNA of the polypeptide was transfected into Xenopus oocytes and a dramatic increase in osmotic swelling was observed (Preston et al., 1992). The 28 kDa polypeptide was initially named CHIP28 (standing for Channel-like integral protein of 28 kDa); after several similar water-channel proteins were subsequently discovered and the need for a functionally significant name became apparent, CHIP28 was renamed aquaporin-1 (AQUAPORIN-1) and made a member of the newly emerging aquaporin family of water channel proteins (Borgnia et al., 1999).
In the fifteen years since the initial discovery of aquaporin-1, research regarding the structure and function of the protein has surged. While there are still contentions regarding the mechanism of the function of the protein and methods of its inhibition, research on the topic continues to flourish. The following paper will discuss, in detail, the structure and function both of water transport and inhibition in the water channel protein, aquaporin-1.
Structure of Aquaporin-1 Water Channel Protein
Aquaporin-1 is a tetrameric protein comprised of four identical and independently functioning monomer subunits. The aquaporin-1 monomer consists of six membrane-spanning alpha-helices and two intra-membrane alpha-helices tilted 21 degrees with respect to the normal of the membrane bilayer (Walz et al., 1997), with both the amino and carboxy termini found on the cytoplasmic side of the membrane (Walz et al., 1997) (Figure 1a). The polypeptide is also comprised of many small amino acids, including glycine and alanine (Ren et al., 2000).
 |
Figure 1. a) Side view of the aquaporin-1 monomer (accession code 1FQY) with trans-membrane helices H1, H2, H3, H4, H5 and H6, non-membrane-crossing helices HE and HB and loops LA, LB and LE labeled. b) End-on view of the aquaporin-1 monomer (accession code 1FQY) from the extracellular surface. The location of the aqueous pore can be seen running through the middle of the monomer. c) End-on view of the aquaporin-1 tetramer (accession code 2B5F) from the extracellular surface. (Larger view.) |
The structural motif seen in aquaporin-1 (six transmembrane helices forming a barrel shape) is a classic identifier of a much larger group of cross-membrane channel proteins to which aquaporin-1 belongs—the ubiquitous major intrinsic protein (MIP) family (Park and Saier, 1996). The MIP family is approximately 2.5 – 3 billion years old and is characterized as a family of membrane spanning proteins that serve as selective transporters of water, small solutes and ions (Park and Saier, 1996). Among the MIP family, the NPA motif, an Asn-Pro-Ala repeat present twice in the polypeptide chain, is highly conserved (de Groot et al., 2003). Given the high conservation of this repeat, it has long been suspected to perform an important role in the mechanism responsible for the selective channel properties exhibited by the MIP family of proteins (Reizer et al., 1993).
In aquaporin-1, the NPA motif can be found on non-membrane crossing helices B and E (HB and HE, respectively) oriented 180 degrees to each other near the center of the protein (Figure 2) (Jung et al., 1994). The two asparagines residues (76 and 192) found in each of the NPA motifs exhibit two fold symmetry with respect to each other (Kong and Ma, 2001) and form the central constriction of the aqueous channel. Moving away from this central constriction towards both the cytoplasmic and extracellular sides of the protein the diameter of the channel increases, creating a pore with an hourglass-like structure (Jung et al., 1994).
 |
Figure 2. Side view of the aquaporin-1 monomer (accession code 1FQY). Residues shown in red are members of the Asn-Pro-Ala residues of the NPA motif found on HB and HE (blue). (Larger view.) |
Each aquaporin-1 tetramer consists of four independently functioning monomeric water channels (Cheng et al., 1997) and exhibits four-fold symmetry (Figure 1c). At the center of the protein, a central pore is formed by the interior edges of the four monomeric subunits (Figure 1c). At its widest, this pore is 8.5 Å in diameter and is lined with hydrophobic residues eliminating it as a possible pathway for selective water transport (Murata et al., 2000).
Each monomer contributes equally to the stability of the tetramer; each monomer interacts with the two adjacent monomers via residues found in the membrane spanning alpha-helices (Murata et al., 2000). These interactions are stabilized by hydrogen bonds between the Ser-59, Thr-62, and Gln-65 residues of one monomer and the Gln-148, Cys-152 and Thr-156 residues of the neighboring monomer (Murata et al., 2000). Interactions between the four loop A regions (on the extracellular side) of each monomer and the four loop D (on the cytoplasmic side) of each monomer might also serve to stabilize the tetramer (Murata et al., 2000). Aquaporin-1 is an insoluble in water, suggesting that the tetramer might be further stabilized by interactions between the polypeptide and the surrounding lipid molecules. Candidates for such interactions can be found on helices 3 and 6, as well as loop B and loop E (Murata et al., 2000).
Water Transport Mechanism and Proton Exclusion
The aqueous pore of the aquaporin-1 monomer is the site of water transport and is, at its narrowest, 3 Å in diameter—wide enough for 2.8 Å wide water molecules to traverse the constricted region, but effective at providing a physical barrier to larger solutes and ions (Murata et al., 2000). Water molecules in the vicinity of the central pore do not exhibit thermodynamic differences from the bulk movement of water molecules in solution (Ren et al., 2000). The activation energies observed for the movement of water across the central pore are low and correspond approximately to the energy associated with hydrogen bonds. This suggests that the breaking of one or several hydrogen bonds plays an essential role in the selective transport of water through the central pore of the aquaporin-1 monomer (Murata et al., 2000). It is important to note that water movement via aquaporin-1 is bidirectional and dictated by the osmotic gradient present in the environment (Borgnia et al., 1999).
The transport of water by aquaporin-1 is not accompanied by a change in the proton gradient across the lipid bilayer. In the presence of hydrogen bonding, protons can easily move from one water molecule to another via the network of hydrogen bonds present. If there were no mechanism in place to regulate or prevent the movement of protons through the aqueous pore of aquaporin-1, the proton gradient across the cell membrane necessary for normal cell function would be compromised. However, the proton flux through aquaporin-1 is restricted to a rate one-thousand times slower than the rate at which water is transported through the monomeric pore—clearly indicating that a mechanism is in place to prevent protons from leaking through the water channel of the aquaporin-1 protein (de Groot et al., 2003).
Molecular dynamics simulations of proton transport with quantum
mechanically derived proton hopping rates (Q-HOP MD) (Lill and Helms, 2001) performed on the aquaporin-1 monomer have clearly identified the NPA region as the location of the proton exclusion mechanism (de Groot et al., 2003). Protons placed in the aqueous pore of the aquaporin-1 monomer were observed quickly leaving the area; almost all of the protons involved in the simulation were observed leaving the pore via the side of the NPA motif region that they were originally placed—very rarely was a proton observed crossing the NPA motif in the process of leaving the pore region (de Groot et al., 2003). A free energy profile of the aqueous pore created with the data collected in the Q-HOP MD simulations demonstrates that the NPA region exhibits the highest free energy barrier to protons found in the aqueous pathway (de Groot et al., 2003).
The evidence for the NPA region as the site of proton exclusion, as well as the inherent association between protons and the hydrogen bonding of water molecules suggests that the water selectivity mechanism and proton exclusion mechanism are woven together at the site of the NPA motif. Currently, the most widely accepted mechanism for the movement of water molecules through the aqueous core also provides an account for the observed proton exclusion.
Water molecules are limited to passing through the constricted region of the channel pore one molecular at a time due to the small size of the pore (3 Å) relative to the size of one water molecule (2.8 Å) (Murata et al., 2000). The wider regions of the aqueous pore are lined with hydrophobic residues (Ile-60, Phe-24, Leu-149 and Val-176) (Figure 3), inhibiting the formation of hydrogen bonds between water molecules and residues lining the surface of the pathway—allowing the water molecules to flow freely and uninhibited through the wider regions of the aqueous pore (Kong and Ma, 2001). The carboxyl terminus of both loop B and loop E face outward (toward the extracellular and cytoplasmic regions, respectively), creating a positive electrostatic field generated by the dipole moment of the two helices (Murata et al., 2000). As water molecules move towards this electrostatic field they are oriented toward the sides of the NPA motifs facing the surface of the constriction region, such that they are primed to hydrogen bond with either Asn-192 or Asn-76 (of the NPA motif)(Murata et al., 2000). In the process of doing so, the hydrogen bond between the two neighboring water molecules is broken—creating a barrier to proton transfer across the aqueous pore of the aquaporin-1 monomer (Figure 4) (Kong and Ma, 2001).
 |
Figure
3:
a) Stereo side view of the aquaporin-1 monomer (accession code 1FQY)
cut
such that the surface of the aqueous pore can clearly be seen in
relation to the hydrophobic residues lining the aqueous pore (red)
and the Asn-76 and Asn-192 residues (green) responsible for forming
hydrogen bonds with passing water molecules can be seen.
b) Stereo end-on view of the surface of the aquaporin-1 monomer (accession code 1FQY) as seen from the extracellular surface in stereo. Asn-76 and Asn-192 are shown in green and the remaining members of each NPA motif are shown in blue (Pro-77, Ala-78, Pro-193 and Ala-184). The hydrophobic residues lining the aqueous pore (Phe-24, Ile-60, Leu-149 and Val-176) are shown in red. The constriction region of the pore is clearly defined by the interior surface of the depicted residues. (Larger view.) |
 |
Figure
4:
Water molecules (blue and grey) can be seen forming a single-file
line as they pass by the NPA motif. Hydrogen bonds formed between
Asn-76 and Asn-192 and the passing water molecules clearly disrupt
the hydrogen bonding chain holding the chain of water together making
it impossible for protons to travel through the aqueous pathway in
concert with water. (Accession code 1J4N). (Larger
view.) |
Results from site-directed mutagenesis studies of Asn-192 and Asn-76 have confirmed the importance of these two residues with regard to aquaporin-1 function. Substituting the two Asparagine residues with non-polar leucine residues led to the total malfunction of the aqueous pathway (Kong and Ma, 2001). This was attributed to both the increased hydrophobicity that the two Leu residues introduced to the aqueous pore and the subsequent decrease in water density in the constriction region following the mutation (Kong and Ma, 2001), making it impossible for the water channel to function in any capacity. However, while unlikely, the possibility of a collapse in the pore following the introduction of the two leucine residues has not been ruled out as a possible explanation for the mutant protein’s defunct aqueous pathway.
The His-180 residue may also play an important, as yet undiscovered, role in the water channel mechanism (Murata et al., 2000). His-180 is present in all members of the MIP family that function as selective water channels, but is, however, supplanted by glycine in MIP proteins that function as glycerol transporters—indicating that its function, whatever it may be, is specific to the water transport mechanism of aquaporin-1 (Murata et al., 2000).
Mercurial Inhibition of Aquaporin-1
By the time Agre discovered the Aquaporin-1 water channel protein in 1992, it had long been acknowledged that mercury, in particular HgCl2, functioned as an inhibitor of water channel function. Shortly after the discovery of the aquaporin-1 water channel, researchers identified Cys-189 as the sole site of mercurial inhibition of aquaporin-1 (Preston et al., 1993).
The location of the cysteine residue is inconsequential (Kuang et al., 2001). C189S mutants are susceptible to mercury inhibition when residues Ser-71, Gly-72 or Ala-73 residues on the inner surface of the aqueous pore were replaced with cysteine (Kuang et al., 2001). This finding indicated not only that the availability of a cysteine was the only requirement for mercurial inhibition of the aquaporin-1 protein, but that also residues 71-73 can easily be reached by extracellular materials, despite their location deep within the aqueous channel.
Recently, the structural characteristics of aquaporin (specifically, aquaporin-1) mercury inhibition were determined via x-ray crystallography using aquaporin-Z, the aquaporin-1 homologue found in bacteria (Savage et al., 2007). While the aquaporin-Z protein contains all other residues critical to the water transport mechanism (those contained in the NPA motifs), it does lack the Cys-189 residue responsible for mercury inhibition (Preston et al., 1993). However, the Thr-183 residue found in aquaporin-Z is homologous to the Cys-189 residue responsible for mercury inhibition in aquaporin-1. A T183C mutant was produced via site-directed mutagenesis, such that mercurial inhibition, as it occurs in the aquaporin-1 monomer, could be adequately observed and studied (Savage et al., 2007).
When the T183C mutant was exposed to mercury, two atoms bound to the aquaporin-1 protein (Hg1 and Hg2). Hg1 was located in the central pore and Hg2 in the interstitial cavity just outside the pore (Figure 5) (Savage et al., 2007). The location of Hg1 was unexpected, given the large distance (5.6 Å) between it and the Cys-183 residue (Savage et al., 2007). Hg1 did, however, form electrostatic interactions with neighboring residues Ser-184 and His-174 (Figure 5) (Savage et al., 2007). Conversely, Hg2 bound directly to the T183C residue (Savage et al., 2007).
 |
Figure 5: Mercury inhibition of the aquaporin-Z monomer in stereo. The surface of the protein and its aqueous pathway can clearly be seen in conjunction with the location of Hg1 and Hg2 as well as pertinent residues contributing to their stability. Hg1 can clearly be seen (in red) sitting in the aqueous pathway, sterically blocking the movement of water. His-174 (orange) and Ser-184 (yellow), which electrostaticly bond with Hg1 are also present. Hg2 (also in red) can be seen on the interior of the protein bonded to the T183C mutant (Accession code 209E). (Larger view.) |
Despite the interstitial location of Hg2 no evidence for conformational or structural change within the protein was found (Savage et al., 2007). Consequently, the inhibition of water transport through the central pore must be due to the binding of Hg1 in the pore and is likely sterically induced (Figure 5) (Savage et al., 2007). This steric mechanism of inhibition will allow for mercury to be used in further studies of alternative functions of aquaporin-1, by allowing researchers to distinguish between the effects of the water channel found in each monomeric subunit and the controversial functions recently proposed (Yu et al., 2006) for the tetrameric pore at the center of the aquaporin-1 tetramer (Savage et al., 2007).
Tetraethylammonium Inhibition of Aquaporin-1
Tetraethylammonium chloride (TEA) was initially evaluated as a potential inhibitor of the aquaporin-1 water channel, given its established role as an ion channel inhibitor (MacKinnon and Yellen, 1990). In studies using aquaporin-1 expressing Xenopus oocytes, kidney cells and cells from the MDCK cell-line, TEA was shown to selectively inhibit the water channel function of aquaporin-1 in a concentration dependent and reversible manner (Brooks et al., 2000; Yool et al., 2002).
Exposure of the aquaporin-1 mutant, Y186F (found on the loop E region of the protein and chosen for the close proximity to the mercury binding residue Cys-183), to TEA failed to inhibit aquaporin-1 water permeability even when large concentrations of TEA were used (Brooks et al., 2000). Consequently, loop E was identified as the region of the protein structurally responsible for TEA inhibition of aquaporin-1 (Brooks et al., 2000).
Discussion
The aquaporin-1 water channel has been demonstrated to be a relatively simple protein that has played a major role in the development cellular life. The barrel-like structure consisting of six cross membrane alpha-helices and two intra membrane alpha-helices is far from complex when compared to the structures of other biologically important proteins. Similarly, the mechanism for the transport of water and exclusion of protons is also relatively straightforward.
The importance of the aquaporin-1 water channel in the context of the living cell and on an even larger scale, in an organism, cannot be understated. In recent years, new light has been shed on the vast clinical importance that these small water channels hold. Aquaporin-1 can be found in every type of cell that relies upon cross-membrane water transport (Yang et al., 2006). Studies of aquaporin-1 knockout mice have revealed that the influence of aquaporin-1 reaches far beyond the domain of water transport and influences an incredibly diverse set of other biological events, including but not limited to, tumor angiogenesis (Saadoun et al., 2005), urine concentrating ability (Ma et al., 1998; Schnermann et al., 1998) and cell migration (Verkman, 2005). Further, data now suggests that there may be an emerging role for the clinical use of aquaporin-1 inhibitors pharmaceutically as diuretics (Yang et al., 2006) and as potential therapies for disorders caused by problematic aquapoprin function such as pulmonary edema, glaucoma and polycystic kidney disease (Brooks et al., 2000). Considering the widespread clinical implications of the aquaporin-1 water channel protein, it is important for research on the protein and its inhibition to continue forward—particularly in the search for novel aquaporin-1 inhibitors. Mercury, while demonstrated to be a very effective aquaporin inhibitor and useful in the study of the mechanism of water transport, has very little use in a clinical setting given its lack of specificity for the aquaporin-1 water channel and its inherent toxicity to cellular life (Yang et al., 2006). Tetraethylammonium has shown potential as a clinically applicable inhibitor and hopefully will be the first of many new compounds identified as aquaporin-1 inhibitors.
Acknowledgements
The author would like to thank her peers for their patience and help with Pymol and Dr. Park and Dr. Osterhout for their guidance through this project.
References
Borgnia, M., Nielson, S., Engel, A. and Agre, P. (1999) Cellular and molecular biology of the aquaporin water channels. Annu. Rev. Biochem. 68, 425-458.
Brooks, H.L., Regan, J. W., and Yool, A. J. (2000) Inhibition of aquaporin-1 water permeability by tetraethylammonium: involvement of the loop E pore region. Mol Pharmocol. 57, 1021-1026.
Cheng, A., van Hoek, A. N., Yeager, M., Verkman, A. S. and Mitra, A. K. (1997) Three-dimensional organization of a human water channel. Nature. 387, 627-630.
de Groot, B. L., Frigato, T., Helms, V. and Grubmüller, H. (2003) The mechanism of proton exclusion in the aquaporin-1 water channel. J. Mol. Biol. 333, 279-293.
Heymann, J. B., and Engel, A. (2000) Structural clues in the sequences of aquaporins. J. Mol. Biol. 295, 1039-1053.
Jung, J. S., Preston, G. M., Smith, B. L., Guggino, W. B. and Agre, P. (1994) Molecular structure of the water channel through aquaporin CHIP. J. Biol. Chem. 269, 14648-14654.
King, L. S., and Agre, P. (1996) Pathophysiology of the aquaporin water channels. Annu. Rev. Physiol. 58, 618-648.
Knepper, M. A., Wade, J. B., Terris, J., Ecelbarger, C. A., Marples, D., Mandon, B., Chou, C. L., Kishore, B. K. and Nielson, S. (1996) Renal aquaporins. Kidney Int. 49, 1712-1717.
Kong, Y. and Ma, J. (2001) Dynamic mechanisms of the membrane water channel aquaporin-1 (AQUAPORIN-1). PNAS. 98, 14345-14349.
Kuang, K., Haller, J. F., Shi, G., Kang, F., Cheung, M., Iserovich, P. and Fischbarg, J. (2001) Mercurial sensitivity of aquaporin 1 endofacial loop B residues. Protein Science. 10, 1627-1634.
Lill, M.A. and Helms, V. (2001). Molecular dynamics simulation of proton transport with quantum mechanically derived proton hopping rates (Q-HOP MD). J. Chem. Phys. 115, 7993-8005.
Meinild, A. K., Klaerke, D. A. and Zeuthen, T. (1998) Bidirectional water fluxes and specificity for small hydrophilic molecules in aquaporins 0-5. J. Biol. Chem. 273, 32446-32451.
Ma, T., Yang, B., Gillespie, A., Carlson, E. J., Epstein, C. J. and Verkman, A. S. (1998) Severely impaired urinary concentrating ability in transgenic mice lacking aquaporin-1 water channels. J. Biol. Chem. 273, 4296-4299.
MacKinnon, R., and Yellen, G. (1990) Mutations affecting TEA blockade and ion permeation in voltage-activated K+ channels. Science. 250, 276-279.
Macey, R. (1984) Transport of water and urea in red blood cells. Am. J. Physiol. 256, C195-203.
Murata, K., Mitsuoka, K., Hirai, T., Walz, T., Agre, P., Heymann, J. B., Engel, A. and Fujiyoshi, Y. (2000) Structural determinants of water permeation through aquaporin-1. Nature. 407, 599-605.
Park, J. H., and Saier, M. H., Jr.(1996) Phylogenetic characterization of the MIP family of transmembrane channel proteins. J. Membrane Biol. 153, 171-180.
Preston, G. M., Carroll, T. P., Guggino, W. B. and Agre. P. (1992) Appearance of water channels in Xenopus oocytes expressing red cell CHIP28 protein. Science. 256, 385-387.
Preston, G. M., Jung, J. S., Guggino, W. B. and Agre, P. (1993) The mercury-sensitive residue at cystein 189 in the CHIP28 water channel. J. Biol. Chem. 268, 17-20.
Reizer, J., Reizer, A., and Saier, M. H., Jr. (1993) The MIP family of integral membrane channel proteins: sequence comparisons, evolutionary relationships, reconstructed pathway of evolution, and proposed functional differentiation of the two repeated halves of the proteins. Crit. Rev. Biochem. Mol. Biol. 28, 235-257.
Ren, G., Cheng, A., Reddy, V., Melnyk, P. and Mitra, A. K. (2000) Three-dimensional fold of the human AQUAPORIN-1 water channel determined at 4 Å resolution by electron crystallography of two-dimensional crystals embedded in ice. J. Mol. Biol. 301, 369-387.
Saadoun, S., Papadopoulos, M.C., Hara-Chikuma, M. and Verkman, A.S. (2005) Impairment of angiogenesis and cell migration by targeted aquaporin-1 gene disruption. Naure. 434, 786-792.
Savage, D. F. and Stroud, R. M. (2007) Structural basis of aquaporin inhibition by mercury. J. Mol. Biol. 368, 607-617.
Schnermann, J., Chou, C. L., Ma, T., Traynor, T., Knepper, M. A. and Verckman, A. S. (1998) Defective proximal tubular fluid reabsorption in transgenic aquaporin-1 null mice. Proc. Natl. Acad. Sci. USA 95, 9660-9664.
Solomon, A. K. (1968) Characterization of biological membranes by equivalent pores. J. Gen. Physiol. 51, S335-364.
Stroud, R. M., Savage, D., Miercke, L. J. W., Lee, J. K., Khademi, S. and Harries, W. (2003) Selectivity and conductance among the glycerol and water conducting aquaporin family of channels. FEBS. 555, 79-84.
Sui, H., Han, B. G., Lee, J. K., Walian, P. and Jap, B. K. (2001) Structural basis of water-specific transport through the AQUAPORIN-1 water channel. Nature. 414, 872-878.
Verkman, A. S. (1992) Water channels in cell membranes. Annu. Rev. Physiol. 54, 97-108.
Verkman, A. S. (2005) More than just water channels: unexpected cellular roles of aquaporins. J. Cell Sci. 118, 3225-3232.
Walz, T., Hirai, T., Murata, K., Heymann, J. B., Mitsuoka, K., Fujiyoshi, Y., Smith, B. L., Agre, P. and Engel, A. (1997) The three-dimensional structure of aquaporin-1. Nature. 387, 624-626.
Yang,
B., Kim, J. K. and Verkman, A. S. (2006) Comparative efficacy of
HgCl2
with candidate
aquaporin-1 inhibitors DMSO, gold, TEA +
and acetazolamide. FEBS. 580, 6679-6684.
Yool, A. J., Brokl, O. H., Pannabecker, T. L., Dantzler, W. H. and Stamer, D. W. (2002) Tetraethylammonium block of water flux in Aquaporin-1 channels expressedin kidney thin limbs of Henle’s loop and a kidney-derived cell line. BMC Physiology. 2, 4.
Yu, J., Yool, A. J., Schulten, K. and Tajkhorshid, E. (2006) Mechanism of gating and ion conductivity of a possible tetrameric pore in aquaporin-1. Structure. 14, 1411-1423.
Zhu, F., Tajkhorshid, E. and Schulten, K. (2004) Theory and simulation of water permeation in aquaporin-1. Biophysical Journal. 86, 50-57.
Biochemistry
462a
Department of Biochemistry and Molecular Biophysics
The University of Arizona
johno@email.arizona.edu
All contents copyright © 1998-2007. All rights reserved.
Last revision fall 2007
Department of Biochemistry and Molecular Biophysics
The University of Arizona
johno@email.arizona.edu
All contents copyright © 1998-2007. All rights reserved.
Last revision fall 2007