Membrane-bound protein channels that allow only urea to pass through are vital to the kidney's ability to conserve water. Crystal structures show that the channels select urea molecules by passing them through thin slots.
Coin-operated vending machines must reliably accept only valid coins of the correct denomination. Modern machines recognize coins on the basis of their size, shape and even their chemical composition (determined by measuring the coins' electromagnetic properties). On page 757 of this issue, Levin et al.1 describe the crystal structure of a bacterial urea-conducting channel that acts like a molecular version of a coin-operated machine — it selectively allows planar urea molecules to pass through on the basis of their size, shape and electrical-charge distribution. The authors' results thereby provide insight into the function of a class of channel that is vital to the function of the human kidney.
Urea is a small, nitrogen-containing organic molecule. Although uncharged, it is highly polar and adept at hydrogen bonding; indeed, urea's ability to form hydrogen bonds with similarly polar water molecules accounts for its astonishingly high solubility in water (more than 8 M). Urea has a special place in the history of science, because it was the first organic molecule to be synthesized by non-biological means from inorganic starting materials2. It is also a key molecule in human physiology: in the liver, excess nitrogen from the normal breakdown of proteins is incorporated into urea, which is subsequently released into the bloodstream, ultimately to be excreted by the kidneys. Because of its high solubility and low toxicity, large amounts of urea can be excreted in small amounts of water, allowing the kidney to conserve water even when urea excretion is high.
But urea excretion presents a challenge for the kidney by virtue of the osmotic forces that the compound generates. The basic structural and functional unit of the kidney is the nephron. Each nephron filters water and small molecules (including urea) from the blood, creating a flow of fluid that passes through a long, narrow renal tubule. These tubules modify the filtered fluid using myriad transport processes, and what is left becomes urine. The final part of the renal tubule is called the collecting duct. Left uncontrolled, the osmotic force of the highly concentrated urea in collecting ducts would suck water from the kidney interstitium (the space between renal tubules), thus undesirably increasing water excretion, a process called osmotic diuresis (Fig. 1a, overleaf). To avoid this, the kidney must balance urea concentrations inside and outside the urinary space by allowing urea to move rapidly across the membranes of the epithelial cells that line the collecting duct (Fig. 1b).
The polarity of urea molecules prevents them from readily penetrating nonpolar lipid membranes. Kidney cells therefore use specialized channel proteins to move the molecule rapidly into and through cells. These channels — called UT-A1 and UT-A3 — allow urea concentrations to equilibrate rapidly between the urine and interstitium. As a result, urea accrues to extremely high concentrations in the interstitium, osmotically balancing the large amount of urea in the urine and making the urinary urea osmotically 'invisible' — thereby allowing the kidney to conserve water. Consistent with this posited role of urea channels, genetically modified mice lacking the gene that encodes collecting-duct urea channels cannot efficiently conserve water unless fed a low-protein diet to reduce urinary urea3. Inhibitors of urea channels have therefore been proposed as a new class of drug for treating clinical disorders that involve water retention4.
The channel characterized by Levin et al.1, dvUT, is a bacterial protein whose amino-acid sequence resembles that of the kidney urea channels. Because these bacterial proteins don't necessarily have the same function as their mammalian cousins, the authors first tested whether dvUT can carry urea. Gratifyingly, they found that, when expressed in cell membranes, dvUT does allow urea to permeate cells. Furthermore, the authors could inhibit urea movement using phloretin, a blocker of mammalian urea channels.
Levin and colleagues' crystal structure1 of dvUT reveals that it is similar to other proteins that move small neutral molecules across membranes, such as aquaporins (water channels)5 and amtB (an ammonia channel)6. Like those proteins, dvUT forms an oligomeric complex in the membrane (a trimer, in this case). Also like those proteins, each monomer in the dvUT assembly seems to contain an independent pathway for the movement of its substrate. Completing the family resemblance, all three classes of channel have an 'inverted repeat' motif, in which both halves of the molecule have similar amino-acid sequences but adopt opposite orientations with respect to the plane of the membrane.
The crystal structure1 also reveals much about the mechanism of dvUT action. First, there is an unoccluded pathway for urea to travel through the protein. Although urea itself could not be visualized when incorporated into the crystallizing protein, when the authors incorporated a urea analogue, dimethylurea, they observed that it binds at two sites within this pathway. The presence of a clear conduit is the hallmark of proteins that operate by a channel mechanism, in which substrates pass through without requiring extensive changes to the conformation of the protein. Many in the field had assumed that a 'carrier' model — which typically involves large conformational changes — would be needed to explain the kinetics of urea transport7. Other evidence, however, had already suggested channel-like behaviour8,9, as strongly implied by Levin and colleagues' structure1.
The second insight obtained from Levin and colleagues' work is the 'molecular coin-slot' mechanism that allows urea to pass through dvUT in preference to other small molecules. At the entrance to the protein's pore, the planar, aromatic side chains of two phenylalanine amino-acid residues form a hydrophobic slot just wide enough to permit the coin-shaped urea molecule to enter (see Fig. 4 on page 760). This fixes the orientation of the urea molecules, setting them up to be grabbed by a line of oxygen atoms — dubbed the “oxygen ladder” by the authors1 — lying along one side of the slot. The oxygen ladder provides electrostatic interactions for the urea molecules, helping to extract them from the water environment outside the channel.
The whole dvUT pore actually consists of two slots in series, separated by a gap that is lined by hydrophobic amino-acid side chains. The authors' crystal structure of a dvUT–dimethylurea complex reveals two dimethylurea molecules, each one bound to the oxygen ladders at opposite ends of the pore and snuggled between phenylalanine residues at the pore's entrance. The molecules seem to be stabilized in these positions by interactions between their delocalized electrons and the aromatic electrons in the phenylalanine side chains. Thus, the molecular coin slot of dvUT might detect not only urea's size and planar shape, but also its unique electronic configuration.
Levin and colleagues' study1 opens the door to further investigation of the properties of this family of urea channels. This should shed light on such issues as the oligomeric organization of mammalian urea channels; the energetics of the competition between the pore's lining and water outside the channel for binding to urea; and the regulation of the channels by gating or trafficking mechanisms. Such work will provide a treasure trove of information — you can put money on it.
Levin, E. J., Quick, M. & Zhou, M. Nature 462, 757–761 (2009).
Wöhler, F. Ann. Phys. Chem. (Leipz.) 88, 253–256 (1828).
Fenton, R. A., Chou, C. L., Stewart, G. S., Smith, C. P. & Knepper, M. A. Proc. Natl Acad. Sci. USA 101, 7469–7474 (2004).
Levin, M. H., de la Fuente, R. & Verkman, A. S. FASEB J. 21, 551–563 (2007).
Murata, K. et al. Nature 407, 599–605 (2000).
Khademi, S. et al. Science 305, 1587–1594 (2004).
Mayrand, R. R. & Levitt, D. G. J. Gen. Physiol. 81, 221–237 (1983).
Levine, S. D. & Worthington, R. E. J. Membr. Biol. 26, 91–107 (1976).
Finkelstein, A. Water Movement Through Lipid Bilayers, Pores and Plasma Membranes: Theory and Reality (Wiley, 1987).
About this article
Urea‐modulated UT‐B urea transporter internalization is clathrin‐ and caveolae‐dependent in infantile hemangioma‐derived vascular endothelial cells
Journal of Cellular Biochemistry (2019)
Clinical and Translational Oncology (2017)
Computation and Simulation of the Structural Characteristics of the Kidney Urea Transporter and Behaviors of Urea Transport
The Journal of Physical Chemistry B (2015)
Physicochemical Evolution and Molecular Adaptation of the Cetacean Osmoregulation-related Gene UT-A2 and Implications for Functional Studies
Scientific Reports (2015)
Concentrations of osmotically related constituents in plasma and urine of finless porpoise (Neophocaena asiaeorientalis): implications for osmoregulatory strategies for marine mammals living in freshwater
Zoological Studies (2014)