Secondary Logo

Journal Logo

Feature Article

Water and Kidney Physiology

Bankir, Lise PhD

Author Information
doi: 10.1097/NT.0b013e318297864b
  • Free


The mammalian kidney is extremely well adapted to produce concentrated urine. The specialized loop-shaped nephrons and the vascular organization surrounding the loops favor the accumulation of concentrated solutes in the medulla. The blood vessels supplying and draining the medulla run in parallel and countercurrent and form spatially distinct vascular beds in different cortical and medullary regions. Collecting ducts traverse the whole kidney along its radial axis, from the cortex to the tip of the medulla where solutes (mainly urea and NaCl) are accumulated. This very unique organization allows the kidney to excrete the soluble wastes with an “economy of water” by concentrating them into the urine.1

The hormone vasopressin (VP or AVP), also called antidiuretic hormone, plays a crucial role in the ability of the kidney to concentrate urine. Its main sites of action are the collecting ducts. Vasopressin concentrates urine by allowing water to be reabsorbed through cell membrane water channels called aquaporin 2 when these ducts traverse the hyperosmotic medulla. From the huge amount of fluid that enters the nephrons by glomerular filtration (about 140 L/d for a healthy adult), about 8% to 10% are reabsorbed under the influence of antidiuretic hormone.

Urine Dilution and Concentration

The kidney may also need to excrete dilute urine in order to excrete “solute-free water” when high amounts of fluid have been ingested. Thus, the kidney may either reabsorb water in excess of solutes to produce “hyperosmotic” urine or reabsorb solutes in excess to water to produce “hypo-osmotic” urine. When urine osmolarity is equal to plasma osmolarity (that is around 300 mOsm/L), urine is iso-osmotic. Urine osmolarities observed in healthy humans range from 70 to 1300 mOsm/L, thus from about 4-fold below to 4-fold above that of plasma.2

Actually, when the filtered fluid progresses through the nephrons, it is always first diluted, even when urine needs to be concentrated. This process is already present in lower vertebrates that live in fresh water and need to produce dilute urine all the time. Their nephrons possess a “diluting segment,” which has been conserved through evolution and corresponds in mammals to the thick ascending limb of Henle loops. The diluting segment strongly reabsorbs NaCl and thus dilutes the luminal fluid (down to about 150 mOsm/L) because its epithelium exhibits an extremely low permeability to water. In the absence of VP (and thus, when their permeability to water is very low), urine can be further diluted in the collecting ducts because they also actively reabsorb NaCl. But if VP acts normally on these ducts, urine is reconcentrated up to plasma osmolarity during their course in the cortex, and is further concentrated during their course in the outer and inner medulla. Thus, even when urine is iso-osmotic to plasma, VP has been involved in some water conservation.3

Vasopressin’s effects are not duplicated by any other hormonal system. If (a) VP secretion is impaired, or (b) the action of VP on its receptors in the collecting duct is compromised, the urine flow rate may be 10 times normal (10–14 L/d), a situation called diabetes insipidus (DI). Administration of exogenous hormone or of a hormone analog (desmopressin) can restore a normal urine flow in the first case (a = central DI), but not in the second case (b = nephrogenic DI).4 Thus, VP is indispensable for urine concentration.

The Antidiuretic Hormone Vasopressin

Vasopressin is a small 9-amino-acid peptide (molecular weight = 1084) secreted by the neurohypophysis. It is a very old hormone in evolution (similar peptides exist in worms, mollusks, insects). It is released under the influence of 2 different stimuli: an increase in plasma osmolarity (and especially natremia) or a decrease in blood volume. In usual life, the former is the most sensitive. Because of its short half-life (3–5 minutes), VP effects are quickly reversible (contrary to those of aldosterone).

Three types of VP receptors have been identified, V1a, V1b (also called V3), and V2 (V1aR, V1bR, V2R, respectively). The second messenger of receptor activation is intracellular calcium for V1aR and V1bR, and cyclic AMP for V2R. V1bR are expressed in several organs (pancreas, adrenals, spleen, thymus, anterior hypophysis), but their in vivo effects are still poorly understood. V1aR, expressed in vascular smooth muscle cells, are responsible for the well-known vasoconstrictive action of VP. However, the contribution of this hormone to the regulation of blood pressure via vascular V1aR in usual life is probably not very significant (it may be important in case of hemorrhage or significant reduction in blood volume). The V2R expressed abundantly in the renal collecting ducts are responsible for the antidiuretic action of the hormone. In usual life, these V2R-mediated actions are critical for water conservation and are triggered by extremely low (often undetectable) levels of the hormone.3

Besides these classic localizations, VP receptors are also expressed in other organs or cell types. V2R are expressed in the vascular endothelium (where it stimulates nitric oxide production and may thus exert a vasodilatory action), type II pneumocytes, inner ear, and eye. V1aR are expressed in the luminal membrane of the renal collecting duct, in interstitial cells of the renal medulla, in hepatocytes, blood platelets, certain brain nuclei, and adrenal gland. In this article, we focus only on the antidiuretic effects of VP mediated by the V2R. But we need to remember that this hormone may have multiple other actions, besides its effects on water handling, even if the main stimulus for VP release remains the osmotic stimulus.

In contrast to the wide interest in the renin-angiotensin-aldosterone system, little attention has been paid to VP, thirst, and fluid intake in the 1970–2000 period. This poor interest may be explained by several cumulated factors: (1) no good nonpeptide antagonists were available; (2) vasopressin is difficult to measure in the plasma because its usual concentration is very low; in most situations in physiological life, VP concentration is close to the limit of detection of the immunoassays (0.5 pmol/L; (3) its small size makes it difficult to raise good antibodies for immunoassays; (4) not much attention was paid by physicians in the follow-up of their patients to 24-hour urine volume and even less to urine osmolarity; and (5) most clinical investigations of kidney function involve an initial water load in order to improve urine collection. Because this protocol depresses VP secretion, it prevents the disclosure of any possible contribution of VP to the biological or pathological processes under study.

In the last 3 to 4 years, a renewed interest in VP has emerged, because of 2 main factors. First, very selective nonpeptide, orally active antagonists of the different VP receptors are now available (the vaptans).5 Second, a new assay has been designed for the measurement of copeptin, a surrogate marker of VP.6

Interindividual Variability; Sex- and Ethnicity-Related Differences

An increase in plasma osmolarity (Posm), due primarily to solutes that do not freely cross cell membranes (mainly sodium chloride), is the most significant stimulus for VP release under physiologic conditions. Thirst is also induced by increases in Posm, but VP secretion exhibits a lower threshold of Posm than thirst (Figure 1). Thus, VP is almost constantly present in the plasma even when fluids are ingested in amounts sufficient to quench thirst.7 Vasopressin secretion is abolished only when unusually high amounts of fluid are ingested in excess of thirst.

Plasma vasopressin (VP) and thirst thresholds. Relationship between plasma osmolarity (P osm) and plasma VP concentration or intensity of thirst in normal subjects. P osm was lowered or raised by intravenous infusions of either hypotonic or hypertonic sodium chloride solution. Left, A representative subject separately defines P osm thresholds and slopes for VP release (squares) and for sense of thirst (circles). Right, Thresholds for AVP secretion and thirst (left and right ends of the horizontal bars, respectively) in 15 different subjects. The circles inside the horizontal bars represent the resting P osm for each of 15 different healthy men and women. Note the large interindividual differences in the distances between the 2 thresholds and in the resting P osm. Adapted from Robertson GL (Kidney Int. 1984;25:460–480) and reproduced from Torres et al.2

The threshold and the slope of the rise in plasma VP concentration as a function of Posm vary quite largely from subject to subject but are quite reproducible in a given subject, suggesting a significant genetic influence. Figure 1 shows how much both the resting Posm and the osmolarity difference between VP and thirst thresholds vary from subject to subject.7

Men exhibit higher VP levels and concentrate urine more than do women.8 However, because they eat larger amounts of food (and thus excrete more osmoles), men have approximately similar 24-hour urine volume as do women, with a higher urine osmolarity. This sex difference in urine osmolarity may explain the higher male incidence of urolithiasis. Vasopressin levels and urine osmolarity also show differences among ethnic groups. African Americans have higher plasma VP concentration and a higher urine osmolarity than do white Americans.9

Solute-Specific Concentration in Urine With Respect to Plasma

If water reabsorption was the only way by which solutes were concentrated in the urine, all solutes would be concentrated to the same extent in urine with respect to plasma. However, this is not the case, first because some solutes are selectively secreted in the urine by an active or secondary active mechanism involving energy-consuming processes and specialized membrane transporters, and second because some tubule segments exhibit selective high or low permeability to some solutes, allowing them to be reabsorbed preferentially or to be trapped in the tubule lumen, respectively. Finally, countercurrent multiplication processes and countercurrent exchanges take place in the renal medulla between ascending and descending tubules and vessels that run in parallel. The different solutes also exhibit widely different concentrations in the blood plasma (eg, 50 to 100 μmol/L for NH4+ vs 140 mmol/L for Na+). Thus, the concentration of the different solutes in the urine needs to be expressed with respect to their concentration in plasma to be best appreciated.

Figure 2 shows plasma and urine composition and daily excretion of total osmoles (top) and of the main solutes in representative healthy humans on a Western-type diet. The urine/plasma ratios of concentration for the different solutes vary widely from less than 1 (for Na+) to several hundreds (for NH4+). Urea represents almost half of the osmoles excreted daily, although it exhibits a relatively low level in plasma (4–7 mmol/L). The concentration of urea and of other nitrogenous end-products (ammonia, uric acid, etc) accounts for most of the solute-free water reabsorbed by the kidney. Although not yet formerly established, it is clear also that urea must be actively secreted,10 as are other solutes such as ammonia, uric acid, potassium, and protons.

All solutes are not concentrated equally in the urine. “U/P” indicates the urine/plasma ratio of the concentration of the different solutes. “Total” indicates the sum of all solutes, that is, the total osmoles. “Excretion” represents the excretion observed per day in a normal human on a Western-type diet. Urea is by far the most abundant solute in the urine and accounts for a very significant fraction of the total “concentrating burden” of the kidney. Although it is much less abundant than urea, ammonia also represents a significant “concentrating burden” because of its extremely low level in plasma.

Pathophysiologic Consequences of Vasopressin Action

Besides its influence on water permeability along the entire length of the collecting duct, VP also exerts 2 other actions on selective subsegments of this duct. In its cortical part, VP stimulates sodium reabsorption, whereas in its most terminal part, in the deep inner medulla, VP increases the permeability to urea. These 2 actions improve the urine-concentrating ability at the expense of a less efficient sodium and urea excretion, as shown in Figure 3. Accordingly, some sodium retention may ensue and lead to a rise in blood pressure that restores normal sodium excretion by the “pressure natriuresis” mechanism.11 The less efficient urea excretion is compensated for by 2 additive mechanisms, a rise in plasma urea concentration and an enhanced glomerular filtration rate, as demonstrated in rats chronically infused with desmopressin.12 In healthy humans studied twice at a 2-week interval, glomerular filtration rate was lower when subjects drank a large water load than when they drank only a small amount of water.13 As illustrated in Figure 4, these changes are very similar to those induced by a protein-rich meal or a high-protein diet.14

Vasopressin does not act only on water permeability (aquaporin 2), but it also stimulates ENaC and urea transporters. These additional actions of vasopressin increase urine osmolarity at the expense of a reduced efficiency in Na and urea excretion. Data drawn from Bouby et al12 and Bankir et al.11
A high urinary-concentrating activity induces similar changes in the kidney as does a high-protein diet. The hyperfiltration is beneficial in the short term but may induce adverse effects in the long term (due to increased glomerulosclerosis, extra workload for reabsorbing solutes such as glucose, amino acids, sodium, etc). Reproduced from Bankir et al.14

Adverse Effects of Vasopressin

Patients with chronic kidney disease are usually advised to reduce their protein intake. The end-products of carbohydrate and lipid catabolism are only CO2 and H2O (metabolic water) that are easily excreted in the expired air and urine, respectively. In contrast, protein catabolism produces urea, ammonia, uric acid, phosphates, and sulfates, all solutes that are excreted in the urine at a concentration that is much higher than that in plasma and extracellular fluids. Thus, the excretion of protein waste products imposes a significant concentrating burden on the kidney. A high protein intake has been known for a long time to induce an increase in glomerular filtration rate and a significant hypertrophy of the kidney.14 It has been understood that this “hyperfiltration” may, in the long term, induce adverse effects. The increased pressures and flows through glomerular structures may accelerate glomerulosclerosis, albuminuria, and inflammation.15 The higher filtration rate imposes on the renal tubule the reabsorption of increased amounts of glucose, amino acids, sodium, and so on, thus, an increased metabolic demand. As mentioned in the previous section, the chronic stimulation of VP V2 receptors and/or the resulting increase in urine concentration, even in the absence of any change in protein intake, also results in a sustained hyperfiltration and a hypertrophy of the kidney that shares the same anatomical characteristics as that induced by a high protein intake.14

Hyperfiltration, whether induced by high urine-concentrating activity or diabetes mellitus (a situation in which VP is known to be elevated), has been shown to contribute to progression of chronic kidney disease in several animal models,16,17 as does a high protein intake.15 A 3-fold increase in water intake (which reduced VP secretion) ameliorated proteinuria, blood pressure, and glomerulosclerosis in rats with fifth/sixth nephrectomy (a classic model of progressive renal disease). Infusion of desmopressin increased albuminuria in healthy humans and in rats, whereas a selective antagonist of V2R prevented the rise in albuminuria usually seen in rats with diabetes mellitus. Several recent epidemiological studies in humans now suggest that the decline in renal function observed in diverse cohorts is slower, or albuminuria less intense in subjects who excrete larger volumes of urine, or who have lower copeptin concentration or lower urine osmolarity.6,18

Altogether, these results suggest that the concentration of soluble wastes in the urine (and more so the nitrogenous wastes) represents a burden for the kidney and may accelerate disease progression. Reduced protein intake is a mean to reduce the amount of these waste products and thus the concentrating burden on the kidney. Treatment with vaptans may provide an additional benefit.6,19


In prehistorical times, the need to conserve water was a strong priority because a severe lack of water is lethal in a few days. Life expectancy was short, and thus, the delayed adverse consequences of high urine concentration did not have time to appear. In present times, the need to conserve water has decreased because water is easily available (at least in Western countries). But the mechanisms and the structure of the kidney allowing water conservation have remained powerful. The water conservation mechanism operates at the expense of a less efficient excretion of several solutes. This has to be compensated in ways that aggravate all other problems the kidney may have to face.


1. Bankir L, Bouby N, Trinh-Trang-Tan MM. The role of the kidney in the maintenance of water balance. Baillieres Clin Endocrinol Metab. 1989; 3: 249–311.
2. Torres VE, Bankir L, Grantham JJ. A case for water in the treatment of polycystic kidney disease. Clin J Am Soc Nephrol. 2009; 4: 1140–1150.
3. Bankir L. Antidiuretic action of vasopressin: quantitative aspects and interaction between V1a and V2 receptor–mediated effects. Cardiovasc Res. 2001; 51: 372–390.
4. Fujiwara TM, Bichet DG. Molecular biology of hereditary diabetes insipidus. J Am Soc Nephrol. 2005; 16: 2836–2846.
5. Decaux G, Soupart A, Vassart G. Non-peptide arginine-vasopressin antagonists: the vaptans. Lancet. 2008; 371: 1624–1632.
6. Bankir L, Bouby N, Ritz E. Vasopressin: a novel target for the prevention and retardation of kidney disease? [published online ahead of print February 26, 2013] Nat Rev Nephrol. 2013.
7. Zerbe RL, Robertson GL. Osmoregulation of thirst and vasopressin secretion in human subjects: effect of various solutes. Am J Physiol. 1983; 244: E607–E614.
8. Perucca J, Bouby N, Valeix P, Bankir L. Sex difference in urine concentration across differing ages, sodium intake, and level of kidney disease. Am J Physiol Regul Integr Comp Physiol. 2007; 292: R700–R705.
9. Bankir L, Perucca J, Weinberger MH. Ethnic differences in urine concentration: possible relationship to blood pressure. Clin J Am Soc Nephrol. 2007; 2: 304–312.
10. Bankir L, Yang B. New insights into urea and glucose handling by the kidney, and the urine concentrating mechanism. Kidney Int. 2012; 81: 1179–1198.
11. Bankir L, Bichet DG, Bouby N. Vasopressin V2 receptors, ENaC, and sodium reabsorption: a risk factor for hypertension? Am J Physiol Renal Physiol. 2010; 299: F917–F928.
12. Bouby N, Ahloulay M, Nsegbe E, Dechaux M, Schmitt F, Bankir L. Vasopressin increases glomerular filtration rate in conscious rats through its antidiuretic action. J Am Soc Nephrol. 1996; 7: 842–851.
13. Anastasio P, Cirillo M, Spitali L, Frangiosa A, Pollastro RM, De Santo NG. Level of hydration and renal function in healthy humans. Kidney Int. 2001; 60: 748–756.
14. Bankir L, Bouby N, Trinh-Trang-Tan MM, Ahloulay M, Promeneur D. Direct and indirect cost of urea excretion. Kidney Int. 1996; 49: 1598–1607.
15. Brenner BM. Nephron adaptation to renal injury or ablation. Am J Physiol. 1985; 249: F324–F337.
16. Bardoux P, Bruneval P, Heudes D, Bouby N, Bankir L. Diabetes-induced albuminuria: role of antidiuretic hormone as revealed by chronic V2 receptor antagonism in rats. Nephrol Dial Transplant. 2003; 18: 1755–1763.
17. Bouby N, Bachmann S, Bichet D, Bankir L. Effect of water intake on the progression of chronic renal failure in the 5/6 nephrectomized rat. Am J Physiol. 1990; 258: F973–F979.
18. Clark WF, Sontrop JM, Macnab JJ, et al. Urine volume and change in estimated GFR in a community-based cohort study. Clin J Am Soc Nephrol. 2011; 6: 2634–2641.
19. Torres VE, Chapman AB, Devuyst O, et al. The TEMPO 3:4 Trial Investigators. Tolvaptan in patients with autosomal dominant polycystic kidney disease. N Engl J Med. 2012.
© 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins