Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Vasopressin in the Kidney—Historical Aspects
1
Introduction
A century ago it was well established that the infusion of pituitary extract into animals could cause antidiuresis and a concentrated urine. The structure of this neurohypophyseal hormone, mammalian vasopressin (AVP) or antidiuretic hormone (ADH) was discovered in 1954 by Acher and Chavet. In 1957, two patients were diagnosed with bronchogenic cancer and subsequently developed hyponatremia. Both patients were observed to have an elevated urine osmolality in the presence of a low serum osmolality, elevated excretion of sodium, and weight gain. The authors concluded that the presence of both a low serum sodium and a hypertonic urine with normal renal function constituted prima facie evidence of the presence of antidiuretic hormone. This coincided with a new syndrome coined the syndrome of inappropriate antidiuretic hormone (SIADH) or Schwartz-Bartter syndrome.
Water homeostasis is maintained by the interaction between the hypothalamus, kidney, and the thirst response to maintain the serum osmolality within 280–295 mOsm/kg. Arginine 8 vasopressin, whose gene is located on chromosome 20, is a mammalian cyclic nonapeptide hormone and is synthesized in the magnocellular neurons (MCN) of supraoptic (SON) and paraventricular (PVN) nuclei of the hypothalamus in a precursor form composed of vasopressin, neurophysin II, and glycopeptide (copeptin) known as prepro-AVP-NPII. The precursor form is packaged into vesicles and transported down the axon to the posterior lobe of the pituitary gland where it is stored. During transport, the vesicles are converted by proteolysis into three proteins and await release into the circulation.
The secretion of AVP can be from osmotic or nonosmotic stimuli, with direct activation of the primary osmoreceptors located in the organum vasculosum of the lamina terminalis (OVLT), near the third ventricle, by transient receptor potential vanilloid channels or by peripheral baroreceptors that innervate brainstem neurons that project to the hypothalamus, respectively. Thirst has an important role in maintaining water balance and protection against hypernatremia. Behavioral changes with infusion of hypertonic solutes have been observed with an increase in fluid intake. Conversely, a hypotonic environment or destruction of the OVLT causes inhibition of thirst.
In the kidney, the proximal tubule accounts for the reabsorption of a large percentage of the isotonic glomerular filtrate. The descending and ascending limbs of the loops of Henle are responsible for the generation and maintenance of the hypertonicity of the medullary interstitium. The collecting ducts are responsible for water reabsorption via water channels (aquaporins) resulting in concentrated urine in the presence of vasopressin and a hypertonic interstitium.
Over the past 50 years, there has been a vast amount of information inferred and discovered about osmoreceptors, vasopressin, vasopressin receptors, aquaporins, urea transporters, and the treatment of hyponatremia, polycystic kidney disease (PCKD), and nephrogenic diabetes insipidus. Some of the highlights of the past 50 or so years will be reviewed that have allowed us to understand the physiology and pathophysiology of water balance in the following monograph.
2
Hypothalamus
In 1947, Verney proposed that osmotic pressure resulted in secretion of antidiuretic hormone from the posterior pituitary gland. A decade later, the osmoreceptors were located in the anterior hypothalamus adjacent to the third ventricle. However, the exact osmotic stimulus for vasopressin release, hypertonicity versus sodium salt, was uncertain and the development of a reliable radioimmunoassay for ADH in 1973 permitted studies to substantiate Verney’s observations.
In the late 1970s and early 1980s, several animal studies looked at the effect of hypertonic solutions, such as mannitol, sucrose, urea, glucose, and sodium to discern the type of osmoreceptors, the location, and the mechanism for release of AVP from the hypothalamus. In one study, conscious rhesus monkeys were studied to determine the effect of various hypertonic solutions infused into hypothalamus, anterior third ventricle, and the carotid artery. It was observed that the anterior ventricle appeared to be “salt-sensitive” with a greater release of AVP with exposure to hypertonic sodium than to sucrose or mannitol. While direct infusion into the carotid artery gave rise to AVP with all hypertonic solutions including saline, sucrose, and mannitol, there was greater ADH release with hypertonic saline. This went along with the assumption that the central osmosensory cells were “salt-sensitive” rather than tonic responsive. Conversely, infusion of various hypertonic and isotonic solutions in cerebrospinal fluid into the anterior third ventricle in six mongrel dogs revealed an increase in the release of AVP and an increase in thirst with exposure only to hypertonic saline or sucrose but not to normal saline, urea, or glucose. The osmoreceptor appeared to respond more to impermeable hypertonic solutes than to permeable hypertonic solutes, suggesting that the osmoreceptor’s location is outside the blood–brain barrier (BBB).
Intravenous infusion of hypertonic solutes in both humans and animals has shown similar AVP responses. The peripheral injection of hypertonic saline and mannitol showed positive correlation with increases in plasma osmolality, plasma AVP, and thirst, while urea showed a blunted response and no response to glucose or normal saline infusion. This suggested the existence of osmoreceptors outside the BBB.
Then the next step was to determine the location of these osmoreceptors. A total of 11 dogs were used to determine the site of osmosensory receptors and the role of OVLT in hyperosmolality. Electrodes were implanted in the OVLT with 90% of this site destroyed in four dogs and none in the seven controlled dogs. There was a loss of the positive correlation between plasma AVP and changes in osmolality in the dogs with destruction of the OVLT, while not in the controls. In addition, there was a significant attenuation of osmotic-induced intake of water in those with a destroyed OVTL and a higher requirement for an osmotic threshold to stimulate drinking. This further supported the theory that the primary osmosensors are located outside the BBB and primarily located in the OVLT. Other central osmoreceptors sites have been identified including the medial preoptic area, median preoptic nucleus, subfornical organ, and SON, as well as peripheral osmorecptors.
Using whole cell patch clamping technique, isolated mouse hypothalamus explants and isolated OVLT neurons exposed to extracellular mannitol solution showed an increased conductance (depolarization) with increased action potential firing rate correlating with osmotic exposure. The increase in conductance of the OVLT neurons was from activation of calcium permeable nonselective cation channel belonging to the transient receptor potential vanilloid (TRPV) family. The activation of these channels by osmotic stimuli into electrical impulses was not present in genetically engineered TRVP1 −/− mice.
The manipulation of cell size can change conductance regardless of the extracellular tonicity suggesting that part of the activation of the TRVP1 channel is from mechanical changes in these cells from extracellular tonicity. This mechanical activation was lost in TRVP1 −/− mice, but not in TRVP4 −/− or control mice. The OVLT and MCN neuron cells are “intrinsically” osmosensitive and observed to be under mechanical regulation of the nonselective cation channels by volume changes from tonic changes in the surrounding environment.
In addition to the osmotic stimulus, hypovolemia, a nonosmotic stimulus, results in an exponentially delayed release in vasopressin through baroreceptors. The contribution of the osmotic and volume effect on release of vasopressin from the hypothalamus was evaluated in Sprague-Dawley rats. A linear relationship between serum osmolality and vasopressin secretion was noted, while an exponential increase of vasopressin secretion was found with a decrease in blood volume of more than 8%. However, they concluded that AVP secretion is regulated principally by blood osmolality, but a modified response occurs with volume changes. A study of denervation of the sympathetic innervation of baroreceptors in rats suggested a basal inhibitory effect of baroreceptors on neurophyseal vasopressin and oxytocin secretion, with a greater effect on oxytocin, in addition to their role in hypovolemic stimulation.
3
Vasopressin Receptors
In 1979, it was proposed that there existed two types of vasopressin receptors, a V1 and a V2, based on differences in signaling pathways and pharmacological studies with synthetic peptide agonists and antagonists. This was confirmed by the discovery of different vasopressin receptors in the rat renal medulla and liver with similar molecular weights, 83,000 and 80,000, respectively, but possessing different functions. Subsequently, the V1 receptors were further subdivided into V1a and V1b (or V3) by different pharmacological profiles with similar signaling pathways. The V1a and V1b receptors both stimulate the heterotrimeric G-protein subtype Gq resulting in activation of phospholipase C with increasing inositol phosphate production and intracellular calcium. The V1b receptor is located primarily in the anterior pituitary gland and is associated with ACTH secretion while the V1a is located in arterioles associated with vasoconstriction.
In contrast, the V2 receptor was shown to stimulate the heterotrimeric G-protein subtype Gs, which activates adenylyl cyclase and in turn increases cAMP followed by activation of PKA. The V2 receptor plays a major role in water transport in the collecting duct and is responsible for concentrating the urine via aquaporin-2 (AQP2) water channels.
Using a rat liver cDNA library, molecular cloning and expression of rat V1a vasopressin was performed in 1992 by Morel and colleagues. They noted that the rat V1a receptor cDNA encodes 394 amino acids with the translated protein having seven transmembrane spanning domains corresponding to hydrophobic clusters connected by three intracellular and three extracellular loops with structural homology to the family of G protein-coupled receptors. In 1994, the human V1a vasopressin receptor was cloned. The human V1a receptor encodes 418 amino acids with the putative seven transmembrane domains found in G protein-coupled receptors. The homology of the human V1a receptor to rat liver V1a receptor, human and rat V2 receptors, and human oxytocin receptor was 72%, 36%, 37%, and 45%, respectively.
In 1992, rat and human V2 receptors were cloned and characterized. Two putative transmembrane domains derived from cDNA templates of the V1a receptors were used for PCR to identify the V2 receptors. Using the oligonucleotide complimentary to one of the PCR clones, a rat kidney cDNA library was screened and a positive clone isolated. Analysis of this translated protein revealed seven putative transmembrane domains that were characteristic of G protein-coupled receptors. Northern blot analysis identified a 2.2 kb mRNA solely from the rat kidney. In situ hybridization showed high levels of mRNA in the collecting tubules, medullary thick ascending tubules, and periglomerular tubules.
In 1994, the human V1b receptor was cloned and characterized. A V1a cDNA probe was used to screen a corticotrophin pituitary adenoma library and a 424 amino acid protein was found that was consistent with a G protein-coupled receptor. The human V1b has a relatively high degree of amino acid homology with human V1a (45.5%), human oxytocin (44.8%), human V2 (37.3%), rat V1a (46.2%), and rat V2 (44.3%).
3.1
Localization of Vasopressin Receptors in the Kidney
The presence of basolateral plasma membrane V2 receptors in the kidney and their role in water reabsorption in the principal cells of the collecting duct are well established. In 1993, the RT-PCR technique was a new method of measuring the relative levels of mRNA and was used to measure the V2 and V1a receptor mRNAs in rat microdissected tubules. The V1a receptor RT-PCR showed a single band of 598 bp in the glomerulus, initial cortical collecting duct, cortical collecting duct, outer medullary collecting duct, inner medullary collecting duct, and arcuate artery with the greatest signal from the glomerulus. There was a faint signal in the proximal convoluted and straight tubules, inner medullary thin limb, and medullary thick ascending limb. The V2 receptor RT-PCR showed a single band of 625 bp in the inner medullary collecting duct, outer medullary collecting duct, cortical collecting duct, initial cortical collecting duct, medullary thick ascending limb, and the inner medullary thin limb. Large signals from the cortical collecting duct, outer medullary collecting duct, and inner medullary collecting duct correlated with the location and functionality studies demonstrated by others. Interestingly, there was no detection of a V2 receptor in the glomerulus, arcuate artery, and proximal convoluted or straight tubules.
Hermosilla and Schulein transfected a Madin–Darby canine kidney (MDCK) cell line with truncated forms of the cytoplasmic domains of the V2 receptor to determine which domain is responsible for sorting the receptor to the basolateral membrane. The second cytoplasmic loop appeared to promote basolateral membrane expression and contains this sorting signal. The C-terminus was involved in the apical transport of the receptor and the third cytoplasmic loop leads to the retention of the receptor in the endoplasmic reticulum.
The localization and regulation of the V2 receptor was studied in stably transfected polarized MDCK cells. The cells were stably transfected with a construct encoding human V2 receptor, C-terminally tagged with GFP, and confirmed by immunoblot. Immunocytochemistry showed that the V2 receptor mostly colocalized with the basolateral membrane marker E-cadherin with a small percent (11%) in the apical membrane. Once the V2-selective vasopressin agonist, dDAVP, and V2 receptor-GFP became bound, the structure was internalized and transported transiently to early lysosomes and then transported to the late endosomes where the acidic environment led to dissociation of the agonist and the receptor. The V2 receptor was shown to be degraded after internalization via the endosome/lysosomal pathway rather than recycled to the basolateral membrane.
3.2
Action of Vasopressin in the Toad Bladder, Collecting Tubules, and Inner Medulla
The major action of vasopressin in the mammalian kidney is to increase water permeability in the collecting tubules resulting in a concentrated urine. Toad bladder and Brattleboro rats have been studied to evaluate the effect of vasopressin on water permeability in epithelial cells. The homozygous Brattleboro rats, derived from parental Long-Evan rats, have central diabetes insipidus, produce large amounts of urine, and are unable to concentrate urine. They contain no functional vasopressin secondary to a single nucleotide mutation in the C-terminal region of the vasopressin precursor.
Using horseradish perioxidase as a marker, Wade and colleagues studied the effect of an osmotic gradient and vasopressin on the movement of water through the urinary bladder of Bufo marinus . Woodhill and Tisher determined the histological and functional heterogeneity of the distal convoluted, connecting and collecting tubules, and late distal tubules via micropuncture of the Brattleboro rat. Morphological changes were found in the rats exposed to vasopressin under both light and electron microscopy. There was marked dilatation of lateral and basilar intercellular spaces of the entire late distal tubule, consistent with vasopressin-sensitive morphological changes as water entered the cells. This was the first in vivo documentation of vasopressin “responsiveness” by this segment of renal tubules.
The technique of freeze-fracture electron microscopy, which resulted in membrane fracture within its hydrophobic interior, allowed evaluation of membrane structure. Two fractured surfaces are produced: an inner and outer membrane face. In the urinary bladder of Bufo marinus , freeze-fracture revealed that there was no aggregation of intramembranous particles in the absence of vasopressin, but there was aggregation of particles with exposure to vasopressin. A linear relationship resulted between the amount of aggregates and osmotic water flow following exposure to vasopressin, suggesting that vasopressin altered the distribution of intramembranous particles and induced changes in the water permeability of the membrane. These aggrephores were thought to contain channels for water transport across the bilayer membrane. Biophysical analysis revealed that water transport across the toad bladder or tubules relied upon channels within the membrane rather than simple diffusion through bilayer membranes.
In the late 1970s and early 1980s, studies of isolated perfused collecting ducts revealed accumulation of intramembranous particle clusters induced by vasopressin. The aggrephores, containing vesicles, fused to plasma membranes by exocytosis and detached by endocytosis. The aggrephores were noted to contain vesicles coated with clathrin and were thought to be important for vesicle trafficking transport. This was consistent with the “shuttle hypothesis” proposed by Wade in the early 1980s to describe the shuttling of intracellular vesicles to the apical membrane in presence of AVP and detached from the membrane in the absence of vasopressin.
After identification of the water channels, a study was performed on the urinary bladder of anurean Xenopus laevis . Because it is less responsive to vasopressin than other amphibian bladder, the effect of antidiuretic hormone on aggrephores and apical aggregates can be more easily discerned. Using the volumetric technique, net water flow in response to changes of hydrostatic or osmotic pressure was determined and only occurred with an osmotic gradient across the membrane. This effect was enhanced by the presence of AVP (100 mU/mL). Morphological studies using freeze-fracture electron microscopy revealed aggrephores localized to the subapical membrane. In the presence of AVP, the intramembranous aggregates were inserted into the apical membrane in the granular cells. This process was reversed after a 60 min washout period of the AVP and correlated with the finding on light and electron microscopy of an increase in fluid accumulation in the intercellular space, and suggested that the apical borders are more permeable to water. This finding provided further support to the “shuttle” theory proposed by Wade.
Hayashi and colleagues found an increase in aquaporin 2 (AQP2) water channels in the apical membrane and subapical compartment in response to an increase of endogenous vasopressin in rat inner medulla. An infusion of a V2 receptor antagonist (OPC-31260) decreased both apical and subapical AQP2 immunostaining, while a V1a receptor antagonist (OPC-21268) had no effect on AQP2 water channel distribution. Therefore, the study demonstrated that the location and distribution of AQP2 water channels is regulated by the V2 receptor.
In 1995, Nielsen and collegues perfused rat inner medullary collecting ducts with 100 pM AVP (added to the bath) and noted a fivefold increase in osmotic water permeability (Pf) and a decrease in Pf after washout of AVP. Immunocytochemistry confirmed exocytic insertion of intracellular vesicles containing AQP2 in the apical membrane and endocytosis following removal of vasopressin. This proved that vasopressin increases water permeability of collecting duct cells by inducing a reversible translocation of the aquaporin 2 (AQP2) water channel from intracellular vesicles to the apical membrane by exocytosis.
The presence of apical vasopressin receptors in collecting tubules was discovered in addition to the well-known basolateral V2 receptors in MTAL, CCD, and IMCD. In 1991, Ando and colleagues revealed that the application of luminal AVP at nanomolar concentrations in the presence of basolateral picomolar concentrations of AVP in rabbit CCD cells decreased hydraulic conductivity by 35%, as well as a sustained hyperpolarization of transepithelial voltage. Luminal V1a receptors appeared to be responsible for these changes in the rabbit CCD rather than V2 receptor or oxytoxin. Since nanomolar concentrations of AVP have been found in the urine, AVP may bind to these luminal receptors and result in self-inhibition of the antidiuretic effect of AVP in the CCD. The pathophysiological and clinical importance of the various locations of the V1, oxytocin, and V2 receptors require further investigation in their role in water homeostasis.
3.3
Vasopressin Effects in the Other Renal Tubules
The countercurrent multiplier is required for vasopressin to concentrate the urine and was proposed in the 1950s by Hargitay and Kuhn. They hypothesized that tubular urine is first osmotically concentrated in the descending limb of the loop of Henle, and then diluted in the ascending limb of the loop of Henle, before it is finally concentrated in the collecting tubules. Micropuncture study of the composition of the fluid at the hairpin turn of the loop of Henle in desert rodents and hamsters confirmed this hypothesis. These findings supported the countercurrent hypothesis with different concentrations of osmoles being measured in the loop of Henle versus the collecting tubules. Sodium and urea were the major active osmoles in the loop of Henle while urea was the major active osmole in the collecting tubules (see Section 5 urea transport).
There have been numerous publications on the regulation of thick ascending limb transport by vasopressin (refer to review by Knepper ). Vasopressin V1a and V2 receptors have been localized to different segments of the tubules including the thin and thick ascending limbs, collecting tubules, and proximal and distal convoluting tubules. In addition to the regulation of AQP2 water channels, V2 receptors regulate sodium reabsorption by the Na-K-2Cl bumetanide-sensitive (NKCC2) cotransporter and the Na-Cl thiazide-sensitive (NCC) transporter.
In medullary thick ascending limbs, there is evidence that vasopressin stimulates Na-K ATPase and the NKCC2 cotransporter. The direct effect of endogenous vasopressin on the NKCC2 cotransporter and on Na-K ATPase in rats with normal renal function and with induced renal failure was studied. The rats received an infusion of the V2 antagonist OPC-31260 for 3 days before sacrifice. With the infusion, there was an increase in urine output and a decrease in urine osmolality in both treated groups. Western blot analysis of the outer medulla showed a decrease in the Na-K ATPase alpha subunit and the NKCC2 cotransporter in both treated groups as compared to their controls. Both the cotransporter and ATPase appeared to be regulated by endogenous vasopressin. Similar findings were found in Sprague–Dawley and Brattleboro rats with an elevated vasopressin level induced by either dehydration in the former or dDAVP infusion in the latter for 7 days. Therefore, vasopressin appears to have a dual role: maintaining the concentration gradient in the medullary interstitium while allowing its antidiuretic effect downstream in the collecting ducts.
More recently, AVP has been shown to phosphorylate the NKCC2 and NCC cotransporters at the conserved N-terminal theorine and serine residues. Saritas et al. studied the two homologous Ste20-like kinases, SPS-related proline/alanine-rich kinase (SPAK) and oxidative stress responsive kinase (OSR1) to determine which kinases are part of this AVP-signaling pathway. They observed induced regulatory changes in SPAK and SPAK phosphorylated transporters after exposure to dDAVP. Therefore, AVP appears to activate the NKCC2 and NCC cotransporter through SPAK.
Vasopressin effect on the epithelial sodium channel (ENaC) subunits, α, β, and γ, in the rat MTAL, CCD, MAL, CAL, and IMCD, showed different expression of the three subunits in certain tubules. In the CCD, IMCD, and OMCD, all three ENaC mRNA subunits were present but only α mRNA subunit in the MAL and CAL. After the tubules were acutely exposed to a low vasopressin dose, 10 −12 M, cAMP, and a hyperosmolar environment, ENaC mRNA α increased in the MAL but not in the collecting tubules. Recently, AVP effects on ENaC function have been studied, but this is beyond the scope of this chapter.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here