What Do Aquaporin Knockout Studies Tell Us about Fluid Transport in Epithelia?

The investigation of near-isosmotic water transport in epithelia goes back over 100 years; however, debates over mechanism and pathway remain. Aquaporin (AQP) knockouts have been used by various research groups to test the hypothesis of an osmotic mechanism as well as to explore the paracellular versus transcellular pathway debate. Nonproportional reductions in the water permeability of a water-transporting epithelial cell (e.g., a reduction of around 80–90 %) compared to the reduction in overall water transport rate in the knockout animal (e.g., a reduction of 50–60 %) are commonly found. This nonproportionality has led to controversy over whether AQP knockout studies support or contradict the osmotic mechanism. Arguments raised for and against an interpretation supporting the osmotic mechanism typically have partially specified, implicit, or incorrect assumptions. We present a simple mathematical model of the osmotic mechanism with clear assumptions and, for models based on this mechanism, establish a baseline prediction of AQP knockout studies. We allow for deviations from isotonic/isosmotic conditions and utilize dimensional analysis to reduce the number of parameters that must be considered independently. This enables a single prediction curve to be used for multiple epithelial systems. We find that a simple, transcellular-only osmotic mechanism sufficiently predicts the results of knockout studies and find criticisms of this mechanism to be overstated. We note, however, that AQP knockout studies do not give sufficient information to definitively rule out an additional paracellular pathway.     

Isosmotic volume reabsorption in rat proximal tubule

A theoretical model incorporation both active and passive forces has been developed for fluid reabsorption from split oil droplets in rat intermediate and late proximal tubule. Of necessity, simplifying assumptions have been introduced; we have assumed that the epithelium can be treated as a single membrane and that the membrane "effective" HCO3 permeability is near zero. Based on this model with its underlying assumptions, the following conclusions are drawn. Regardless of the presence or absence of active NaCl transport, fluid reabsorption from the split oil droplet is isosmotic. The reabsorbate osmolarity can be affected by changes in tubular permeability parameters and applied forces but is not readily altered from an osmolarity essentially equal to that of plasma. In a split droplet, isosmotic flow need not be a special consequence of active Na transport, is not the result of a particular set of permeability properties, and is not merely a trivial consequence of a very high hydraulic conductivity; isosmotic flow can be obtained with hydraulic conductivity nearly an order of magnitude lower than that previously measured in the rat proximal convoluted tubule. Isosmotic reabsorption is, in part, the result of the interdependence of salt and water flows, their changing in parallel, and thus their ratio, the reabsorbate concentration being relatively invariant. Active NaCl transport can cause osmotic water flow by reducing the luminal fluid osmolarity. In the presence of passive forces the luminal fluid can be hypertonic to plasma, and active NaCl transport can still exert its osmotic effect on volume flow. There are two passive forces for volume flow: the Cl gradient and the difference in effective osmotic pressure; they have an approximately equivalent effect on volume flow. Experimentally, we have measured volume changes in a droplet made hyperosmotic by the addition of 50 mM NaCl; the experimental results are predicted reasonably well by our theoretical model.     

Defective secretion of saliva in transgenic mice lacking aquaporin-5 water channels.

Aquaporin-5 (AQP5) is a water-selective transporting protein expressed in epithelial cells of serous acini in salivary gland. We generated AQP5 null mice by targeted gene disruption. The genotype distribution from intercross of founder AQP5 heterozygous mice was 70:69:29 wild-type:heterozygote:knockout, indicating impaired prenatal survival of the null mice. The knockout mice had grossly normal appearance, but grew approximately 20% slower than litter-matched wild-type mice when placed on solid food after weaning. Pilocarpine-stimulated saliva production was reduced by more than 60% in AQP5 knockout mice. Compared with the saliva from wild-type mice, the saliva from knockout mice was hypertonic (420 mosM) and dramatically more viscous. Amylase and protein secretion, functions of salivary mucous cells, were not affected by AQP5 deletion. Water channels AQP1 and AQP4 have also been localized to salivary gland; however, pilocarpine stimulation studies showed no defect in the volume or composition of saliva in AQP1 and AQP4 knockout mice. These results implicate a key role for AQP5 in saliva fluid secretion and provide direct evidence that high epithelial cell membrane water permeability is required for active, near-isosmolar fluid transport.     

Salivary acinar cells from aquaporin 5-deficient mice have decreased membrane water permeability and altered cell volume regulation

Aquaporins (AQPs) are channel proteins that regulate the movement of water through the plasma membrane of secretory and absorptive cells in response to osmotic gradients. In the salivary gland, AQP5 is the major aquaporin expressed on the apical membrane of acinar cells. Previous studies have shown that the volume of saliva secreted by AQP5-deficient mice is decreased, indicating a role for AQP5 in saliva secretion; however, the mechanism by which AQP5 regulates water transport in salivary acinar cells remains to be determined. Here we show that the decreased salivary flow rate and increased tonicity of the saliva secreted by Aqp5(-)/- mice in response to pilocarpine stimulation are not caused by changes in whole body fluid homeostasis, indicated by similar blood gas and electrolyte concentrations in urine and blood in wild-type and AQP5-deficient mice. In contrast, the water permeability in parotid and sublingual acinar cells isolated from Aqp5(-)/- mice is decreased significantly. Water permeability decreased by 65% in parotid and 77% in sublingual acinar cells from Aqp5(-)/- mice in response to hypertonicity-induced cell shrinkage and hypotonicity-induced cell swelling. These data show that AQP5 is the major pathway for regulating the water permeability in acinar cells, a critical property of the plasma membrane which determines the flow rate and ionic composition of secreted saliva.     

Structural and functional aspects of salivary fluid section in Calliphora

Calliphora salivary glands are described, emphasizing correlations between structure and physiology. In vitro studies show that the distal part of each gland produces a potassium-rich primary saliva when stimulated with 5-hydroxytryptamine or cyclic AMP. The secretory cells have elaborate canaliculi opening into the lumen. Stimulation of the secretory region causes a 60-fold increase in fluid secretion rate without affecting cell structure. The proximal part of the gland reabsorbs potassium when stimulated with cyclic AMP, but 5-HT has no effect. Potassium reabsorption from the primary saliva results in formation of a dilute saliva. The structure of the secretory and reabsorptive cells is discussed with regard to the functional role of long narrow channels in transport.     

Real time measurements of water flow in amphibian gastric glands: modulation via the extracellular Ca2+-sensing receptor

The mechanisms for the formation of the osmotic gradient driving water movements in the gastric gland and its modulation via the extracellular Ca(2+)-sensing receptor (CaR) were investigated. Real time measurements of net water flux in the lumen of single gastric glands of the intact amphibian stomach were performed using ion-selective double-barreled microelectrodes. Water movement was measured by recording changes in the concentration of impermeant TEA(+) ions ([TEA(+)](gl)) with TEA(+)-sensitive microelectrodes inserted in the lumen of individual gastric glands. Glandular K(+) (K(+)(gl)) and H(+) (pH(gl)) were also measured by using K(+)- and H(+)-sensitive microelectrodes, respectively. Stimulation with histamine significantly decreased [TEA](gl), indicating net water flow toward the gland lumen. This response was inhibited by the H(+)/K(+)-ATPase inhibitor, SCH 28080. Histamine also elicited a significant and reversible increase in [K(+)](gl) that was blocked by chromanol 293B, a blocker of KCQN1 K(+) channels. Histamine failed to induce net water flow in the presence of chromanol 293B. In the "resting state," stimulation of CaR with diverse agonists resulted in significant increase in [TEA](gl). CaR activation also significantly reduced histamine-induced water secretion and apical K(+) transport. Our data validate the strong link between histamine-stimulated acid secretion and water transport. We also show that cAMP-dependent [K(+)](gl) elevation prior to the onset of acid secretion generates the osmotic gradient initially driving water into the gastric glands and that CaR activation inhibits this process, probably through reduction of intracellular cAMP levels.     

Hormonal control of renal functions in insects

Insect Malpighian tubules secrete an isosmotic, KCl-rich primary urine containing low concentrations of most other blood solutes. Neuropeptide diuretic hormones (DH), possibly related to vasopressin, stimulate tubular fluid secretion by 2- to 200-fold in response to water loading, e.g., feeding. DH acts on tubules through cyclic AMP (cAMP) to stimulate salt transport without measurable change in osmotic permeability. Changes in composition of tubular secretion after stimulation and the possible control of DH release are discussed. Most of the water, ions, and metabolites in tubular secretion are normally reabsorbed by active mechanisms in the rectum, where the urine may finally become either hyposmotic or strongly hyperosmotic to the blood. A newly discovered neuropeptide, chloride transport-stimulating hormone, controls (via cAMP) reabsorption of the principal salt by stimulating K-dependent, electrogenic transport of Cl- across the apical cell border. Passive net absorption of K+ is thereby enhanced. Diuretic and antidiuretic factors may control osmotic permeability of the rectal wall and thereby influence the osmotic concentrations of the rectal absorbate and final urine. The increased recycling of a KCl-rich fluid through the Malpighian tubule-rectal system after feeding probably serves to clear the body of unwanted substances ingested with, and produced by, metabolism of the meal.     

An osmotic mechanism for exocytosis from dissociated chromaffin cells

Dissociated chromaffin cells from bovine adrenal medulla were stimulated to secrete epinephrine and dopamine beta-hydroxylase with a variety of secretagogues in a study designed to test the hypothesis that the chemiosmotic lysis reaction of isolated chromaffin granules might in some way be related to the mechanism of release during exocytosis. Increasing the osmotic strength of the incubation medium with either NaCl or sucrose led to suppression of secretion of epinephrine from the cells regardless of whether secretion was induced with veratridine or acetylcholine. Suppression of secretion was approximately exponential with respect to osmotic strength. Epinephrine secretion occurred only if the medium contained a permeant anion such as chloride, and secretion induced by veratridine was suppressed when Na isethionate replaced NaCl in the medium. In an extensive study with different monovalent anions veratridine supported epinephrine secretion according to the following activity series: Br-, I-, NO3- greater than methylsulfate, SCN- greater than Cl greater than acetate much greater than isethionate. A similar series, except for the potency of NO3-, was observed with A23187 as agonist. In general, the anion series for granule lysis was analogous. However, there was a poor quantitative correlation between the anion dependence of chemiosmotic granule lysis and the anion dependence of cell secretion. Anion transport inhibitors such as probenecid and pyridoxal phosphate also inhibited secretion while the stilbene disulfonates were inactive. The ineffectiveness of the stilbene disulfonates further distinguished chemiosmotic granule lysis from cell secretion. Secretion of catecholamines, induced by veratridine or nicotine, a cholinergic agonist, was suppressed when NaCl in the medium was replaced by isosmotic sucrose and unexpectedly low levels of dopamine beta-hydroxylase were observed in some cases. In sum, these properties of secreting chromaffin cells resembled some properties of isolated chromaffin granules incubated in ATP and Cl-, but were different in a number of instances. We, therefore, have interpreted our data to indicate that while some mechanistic relationships may indeed exist between the release event in exocytosis from chromaffin cells and the chemiosmotic lysis reaction characteristic of isolated chromaffin granules, an understanding of the energetics of exocytosis awaits the discovery of reasons for the quantitative differences between the two systems.     

Na+ Recirculation and Isosmotic Transport

The Na(+) recirculation theory for solute-coupled fluid absorption is an expansion of the local osmosis concept introduced by Curran and analyzed by Diamond & Bossert. Based on studies on small intestine the theory assumes that the observed recirculation of Na(+) serves regulation of the osmolarity of the absorbate. Mathematical modeling reproducing bioelectric and hydrosmotic properties of small intestine and proximal tubule, respectively, predicts a significant range of observations such as isosmotic transport, hyposmotic transport, solvent drag, anomalous solvent drag, the residual hydraulic permeability in proximal tubule of AQP1 (-/-) mice, and the inverse relationship between hydraulic permeability and the concentration difference needed to reverse transepithelial water flow. The model reproduces the volume responses of cells and lateral intercellular space (lis) following replacement of luminal NaCl by sucrose as well as the linear dependence of volume absorption on luminal NaCl concentration. Analysis of solvent drag on Na(+) in tight junctions provides explanation for the surprisingly high metabolic efficiency of Na(+) reabsorption. The model predicts and explains low metabolic efficiency in diluted external baths. Hyperosmolarity of lis is governed by the hydraulic permeability of the apical plasma membrane and tight junction with 6-7 mOsm in small intestine and < or = 1 mOsm in proximal tubule. Truly isosmotic transport demands a Na(+) recirculation of 50-70% in small intestine but might be barely measurable in proximal tubule. The model fails to reproduce a certain type of observations: The reduced volume absorption at transepithelial osmotic equilibrium in AQP1 knockout mice, and the stimulated water absorption by gallbladder in diluted external solutions. Thus, it indicates cellular regulation of apical Na(+) uptake, which is not included in the mathematical treatment.     

The physico-chemical mechanism of mediated transport. II. Osmotic and isosmotic volume flow

The process of volume change of cells subject to osmotic shocks or isosmotic entrance of permeant solute is formulated on the basis of the accepted structure for the plasma membrane and a physico-chemical approach similar to that recently developed. The effect of relevant parameters is discussed and theoretical equilibrium values for the variables are calculated in connection with water and permeant solute permeability determinations. Although a sorption-diffusional mechanism for solute and/or water volume flow within the membrane is assumed in both cases, the kinetics of volume change is shown to be totally different between them. In the isosmotic process a fixed relationship, given by the total solute concentration, is shown to exist between the permeant solute and volume fluxes to the cell, thereby implying a definite value for the volume fraction of water in the migration pathway, higher than 90%. The bi-phase osmotic regulatory response caused by permeant solute is simulated on the basis of an osmotic and isosmotic processes in series, showing good agreement with general behavior. Finally, an explanation to the problem of volume flow and forces in connection with a diffusional mechanism in biological and artificial membranes, is presented.     

Genetic control of single lumen formation in the zebrafish gut

Most organs consist of networks of interconnected tubes that serve as conduits to transport fluid and cells and act as physiological barriers between compartments. Biological tubes are assembled through very diverse developmental processes that generate structures of different shapes and sizes. Nevertheless, all biological tubes invariably possess one single lumen. The mechanisms responsible for single lumen specification are not known. Here we show that zebrafish mutants for the MODY5 and familial GCKD gene tcf2 (also known as vhnf1) fail to specify a single lumen in their gut tube and instead develop multiple lumens. We show that Tcf2 controls single lumen formation by regulating claudin15 and Na+/K+-ATPase expression. Our in vivo and in vitro results indicate that Claudin15 functions in paracellular ion transport to specify single lumen formation. This work shows that single lumen formation is genetically controlled and appears to be driven by the accumulation of fluid.     

Morphology,fine structure,and functional aspects of the labial gland reservoirs of the subterranean termite Reticulitermes santonensis de Feytaud (Isoptera: Rhinotermitidae)

The morphology and fine structure of the labial gland reservoirs in the subterranean termite Reticulitermes santonensis (Isoptera: Rhinotermitidae) was studied by light and transmission electron microscopy. The reservoir wall consists of a single epithelial cell layer and a cuticular intima. The reservoir ducts are formed by a flat epithelial matrix with cuticular ridges lining the duct lumen. Measurements of the ionic concentrations of reservoir fluids and haemolymph show that the osmolality of reservoir fluid ranges from 7 to 28 mosmol kg⁻¹; the haemolymph osmotic pressure was 201 ± 31 mosmol kg⁻. The reservoir lumen is effectively separated from the haemolymph compartment; a net water flow through the reservoir wall could not be induced in physiological experiments. Moreover, typical epithelial structures associated with a fluid transport against an osmotic gradient are lacking. Thus, our fine structural and physiological data support the view that a water transfer from the haemolymph through the reservoir wall into the reservoir lumen does not occur.     

Channels and transporters in salivary glands

According to the two-stage hypothesis, primary saliva, a NaCl-rich plasma-like isotonic fluid is secreted by salivary acinar cells and its ionic composition becomes modified in the duct sytem. The ducts secrete K⁺ and HCO₃^- and reabsorb Na⁺ and Cl^- without any water movement, thus establishing a hypotonic final saliva. Salivary secretion depends on the coordinated action of several channels and transporters localized in the apical and basolateral membrane of acinar and duct cells. Early functional studies in perfused glands, followed by the molecular cloning of several transport proteins and the subsequent analysis of mutant mice, have greatly contributed to our understanding of salivary fluid and the electrolyte secretion process. With a few exceptions, most of the key channels and transporters involved in salivary secretion have now been identified and characterized. However, the picture that has emerged from all these studies is one of a complex molecular network characterized by redundancy for several transport proteins, compensatory mechanisms, and adaptive changes in health and disease. Current research is directed to the molecular interactions between the determinants and the ways in which they are regulated by extracellular signals and intracellular mediators. This review focuses on the functionally and molecularly best-characterized channels and transporters that are considered to be involved in transepithelial fluid and electrolyte transport in salivary glands.     

An ultrastructural analysis of the salivary system of the terrestrial mollusc, Limax maximus.

The bilateral salivary glands, ducts, and nerves of the giant garden slug Limax maximus control the secretion of saliva and its transport to the buccal mass. Each salivary nerve, which originates at the buccal ganglion, contains over 3000 axon profiles. The axons innervate the musculature of the duct and branch within the gland. The salivary duct is composed of several muscular layers surrounding an epithelial layer which lines the duct lumen. The morphology of the duct epithelium indicates that it may function in ion or water balance. The salivary gland contains four major types of secretory cells. The secretory products are released from vacuoles in the gland cells, and are presumably transported by cilia in the collecting ducts of the gland into the larger muscular ducts.     

Dye-coupling between cells of the salivary glands in the cockroach Periplaneta americana

Cockroaches have acinar salivary glands. The acini consist of peripheral cells specialized for electrolyte and water transport and central cells contributing proteinaceous components to the saliva. Salivary duct cells probably modify the primary saliva. The acinar cells in Nauphoeta cinerea had been shown to be electrically coupled and dye-coupled. Since intercellular communication via gap junctions between acinar cells is difficult to reconcile with previous findings that dopamine and serotonin selectively stimulate the secretion of either protein-free or protein-rich saliva in Periplaneta americana, we investigated whether dye-coupling occurs between both acinar cell types and between duct cells. We iontophoretically loaded Lucifer yellow into impaled cells and tested for dye-coupling by confocal laser scanning microscopy. We found that: (1) peripheral and central cells within an acinar lobulus of the gland in P. americana are dye-coupled; and (2) salivary duct cells are dye-coupled.     

UT-B1 urea transporter plays a noble role as active water transporter in C6 glial cells

Since our experimental results suggest that UT-B1 functions as active water transporter against osmotic gradient in C6 glial cells, we report here for the first time the evidence for the active water transport. Exposure of C6 cells to a hyperosmotic solution containing glycerol or sucrose produced cell shrinkage due to water efflux according to osmotic gradient for water movement. On the other hand, C6 cells show cell swelling against osmotic gradient for water movement just after exposure to a hyperosmotic solution containing urea, indicating that water influx against osmotic gradient for water movement is accelerated by urea; i.e., urea performs active water transport. A specific inhibitor of UT-B, pCMBS, blocked the urea-induced swelling. The urea-induced cell swelling was significantly suppressed in the siRNA-induced UT-B1-knockdown C6 cells. Taken together, these observations indicate that UT-B1 acts as an active water transporter, providing a new model on active water transport.     

Excretion in insects: function of gut and rectum in concentrating and diluting the urine.

The diverse excretory systems of insects exhibit several features that appear unusual when comparisons are made with the mammalian kidney. Secretion by the Malpighian tubules of a fluid that is unlike the blood in composition, substitutes for glomerular filtration. Various reabsorptive functions, such as volume reduction, regulation of individual electrolytes, adjustment of osmotic concentration and pH regulation, which are associated with distinct renal segments in the mammalian kidney, all occur simultaneously in the rectum of terrestrial insects. Involvement of an extracellular molecular sieve in selective reabsorption is novel. As far as water transport is concerned, the rectal pads of the cockroach and locust appear to accomplish, across a single layer of cells, the same function as the countercurrent multiplier system of the mammalian kidney with its several epithelial layers. Direct absorption of water vapor in the rectum of some insects from atmospheres of low relative humidity, clearly involves quite different and unknown mechanisms. Finally, saline-water insect larvae produce hyperosmotic excreta by direct secretion of ions into the rectal lumen. They can adjust individual transport processes to form various secretions, which are appropriate to the natural waters of diverse chemical types in which these larvae thrive.     

A gastric acid secretion model.

A theory of gastric acid production and self-protection is formulated mathematically and examined for clinical and experimental correlations, implications, and predictions using analytic and numerical techniques. In our model, gastric acid secretion in the stomach, as represented by an archetypal gastron, consists of two chambers, circulatory and luminal, connected by two different regions of ion exchange. The capillary circulation of the gastric mucosa is arranged in arterial-venous arcades which pass from the gastric glands up to the surface epithelial lining of the lumen; therefore the upstream region of the capillary chamber communicates with oxyntic cells, while the downstream region communicates with epithelial cells. Both cell types abut the gastric lumen. Ion currents across the upstream region are calculated from a steady-state oxyntic cell model with active ion transport, while the downstream ion fluxes are (facilitated) diffusion driven or secondarily active. Water transport is considered iso-osmotic. The steady-state model is solved in closed form for low gastric lumen pH. A wide variety of previously performed static and dynamic experiments on ion and CO2 transport in the gastric lumen and gastric blood supply are for the first time correlated with each other for an (at least) semiquantitative test of current concepts of gastric acid secretion and for the purpose of model verification. Agreement with the data is reported with a few outstanding and instructive exceptions. Model predictions and implications are also discussed.     

Unimpaired osmotic water permeability and fluid secretion in bile duct epithelia of AQP1 null mice

The mechanisms by which fluid moves across the luminal membrane of cholangiocyte epithelia are uncertain. Previous studies suggested that aquaporin-1 (AQP1) is an important determinant of water movement in rat cholangiocytes and that cyclic AMP mediates the movement of these water channels from cytoplasm to apical membrane, thereby increasing the osmotic water permeability. To test this possibility we measured agonist-stimulated fluid secretion and osmotically driven water transport in isolated bile duct units (IBDUs) from AQP1 wild-type (+/+) and null (-/-) mice. AQP1 expression was confirmed in a mouse cholangiocyte cell line and +/+ liver. Forskolin-induced fluid secretion, measured from the kinetics of IBDU luminal expansion, was 0.05 fl/min and was not impaired in -/- mice. Osmotic water permeability (P(f)), measured from the initial rate of IBDU swelling in response to a 70-mosM osmotic gradient, was 11.1 x 10(-4) cm/s in +/+ mice and 11.5 x 10(-4) cm/s in -/- mice. P(f) values increased by approximately 50% in both +/+ and -/- mice following preincubation with forskolin. These findings provide direct evidence that AQP1 is not rate limiting for water movement in mouse cholangiocytes and does not appear to be regulated by cyclic AMP in this species.     

Optimizing automated characterization of liver fibrosis histological images by investigating color spaces at different resolutions

Background

7D-cadherins like LI-cadherin are cell adhesion molecules and represent exceptional members of the cadherin superfamily. Although LI-cadherin was shown to act as a functional Ca²⁺-dependent adhesion molecule, linking neighboring cells together, and to be dysregulated in a variety of diseases, the physiological role is still enigmatic. Interestingly 7D-cadherins occur only in the lateral plasma membranes of cells from epithelia of water transporting tissues like the gut, the liver or the kidney. Furthermore LI-cadherin was shown to exhibit a highly cooperative Ca²⁺-dependency of the binding activity. Thus it is tempting to assume that LI-cadherin regulates the water transport through the epithelium in a passive fashion by changing its binding activity in dependence on the extracellular Ca²⁺.

Results

We developed a simple mathematical model describing the epithelial lining of a lumen with a content of variable osmolarity covering an interstitium of constant osmolarity. The width of the lateral intercellular cleft was found to influence the water transport significantly. In the case of hypertonic luminal content a narrow cleft is necessary to further increase concentration of the luminal content. If the cleft is too wide, the water flux will change direction and water is transported into the lumen. Electron microscopic images show that in fact areas of the gut can be found where the lateral intercellular cleft is narrow throughout the lateral cell border whereas in other areas the lateral intercellular cleft is widened.

Conclusions

Our simple model clearly predicts that changes of the width of the lateral intercellular cleft can regulate the direction and efficiency of water transport through a simple epithelium. In a narrow cleft the cells can increase the concentration of osmotic active substances easily by active transport whereas if the cleft is wide, friction is reduced but the cells can hardly build up high osmotic gradients. It is now tempting to speculate that 7D-cadherins, owing to their location and their Ca²⁺-dependence, will adapt their binding activity and thereby the width of the lateral intercellular cleft automatically as the Ca²⁺-concentration is coupled to the overall electrolyte concentration in the lateral intercellular cleft. This could provide a way to regulate the water resorption in a passive manner adapting to different osmotic conditions.