- Email: [email protected]
Escape from antidiuresis: A good story
Kidney International, Vol. 60 (2001), pp. 1608–1610
EDITORIAL
Escape from antidiuresis: A good story Good stories often share common characteristics. They address topics of interest and/or significance, they are played out over extended periods of time, and they have elements of controversy, if not drama. But perhaps most important, they generally include a definable beginning, a middle, and an end. The story of escape from vasopressin [arginine vasopressin (AVP)]-induced antidiuresis certainly fulfills most of these criteria. The phenomenon of AVP escape has arguably been one of the most interesting unresolved aspects of renal function that has been studied over the last half century. How the kidney is able to escape AVP-induced antidiuresis is also of considerable significance because this process allows survival of the organism by allowing free water excretion, despite inappropriate secretion of AVP, thus effectively antagonizing the effects of one of the most powerful hormones involved in body fluid homeostasis. The story of escape has continued to evolve over the last half century since this phenomenon was first recognized in 1959 with the landmark report of Levinsky, Davidson, and Berliner [1]. And, as for controversy, each decade has witnessed a new potential explanation for this phenomenon, including decreased water permeability of collecting duct principal cells [1], dissipation of the cortico-medullary osmotic gradient [2], increased generation of renal prostaglandin E2 [3], and renal hemodynamic changes mediated by increased renal artery perfusion pressure leading to a pressure diuresis [4]. Although the various mechanisms proposed to underlie escape have differed, all studies to date are in agreement that expansion of the extracellular fluid (ECF) space by AVP-induced water retention is crucial for the onset and maintenance of AVP escape. Thus, we have the makings of an interesting, and controversial, story with a clear beginning: ECF volume expansion. But now we also know the ending of this story. In this issue of Kidney International, Saito et al [5] confirm the recent findings of Ecelbarger et al [6] that AVP escape is associated with a marked down-regulation of expression of the AVP-sensitive water channel, aquaporin-2 (AQP-2), in the kidney. Because AQP-2 insertion into the apical membrane of collecting tubule principal cells represents the final step of facilitated water transport from the colKey words: vasopressin, extracellular fluid, aquaporin-2, hypo-osmolality.
2001 by the International Society of Nephrology
lecting tubule lumen into the medullary interstitium [7], a regulatory mechanism that impacts at this stage of the urinary concentrating mechanism would be maximally effective at limiting AVP-induced antidiuresis. Using similar animal models of dilutional hyponatremia [8], the results of the two studies differ quantitatively, but not qualitatively. Ecelbarger et al found 70 to 80% decreases in l-deamino8-d-arginine vasopressin (dDAVP)-stimulated AQP-2 protein and mRNA expression in the kidney that correlated temporally with the decrease in urine osmolality and increase in urine volume characteristic of escape. Similarly, Saito et al found 35 to 55% decreases in AQP-2 protein and mRNA expression after 7 days of dDAVPinduced hyponatremia. In both cases, this marked downregulation of AQP-2 expression occurred despite pharmacological dDAVP infusion. Therefore, both studies clearly indicate an AVP-independent regulation of AQP-2 expression. Although a causal rather than correlational association between the AQP-2 down-regulation and AVP escape has not yet been definitely established, given the strong association between AQP-2 membrane insertion and collecting tubule water permeability, this amount of change in AQP-2 expression is clearly sufficient to alter the degree of antidiuresis produced [9]. Thus, it seems likely that this is, in fact, a very plausible ending to our story. However, a beginning and an end do not by themselves make a satisfying story. There needs to be a middle that ties these ends together and fleshes out the relationship between them. What are the events that occur between AVP-induced volume expansion and AQP-2 down-regulation, and how are they related? The results of Saito et al and previous studies differ substantially and further fuel the controversy that has surrounded this area for the last half century. Ecelbarger et al found a 40 to 45% down-regulation of dDAVP-stimulated cyclic adenosine monophosphate (cAMP) generation in inner medullary collecting duct cell suspensions prepared from escaped rats compared to controls [10]. Because cAMP mediates signal transduction between the AVP V2 receptor and its effects on both AQP-2 membrane insertion and AQP-2 transcription [9], resistance to AVP-stimulated adenylate cyclase activity would represent a prime candidate for modulation of AVP effects at a cellular level. Consistent with this idea, Tian et al found a marked down-regulation of the AVP V2 receptor number (Bmax) in escaped rats that was 70% below control levels and 35% below the
1608
Editorial
levels of nonescaped dDAVP-treated rats [11], which are known to undergo ligand-induced receptor desensitization. In contrast, Saito et al found no differences in cAMP generation between the escaped and control rats and observed a 40% reduction in V2 Bmax secondary to dDAVP treatment but not any further decreases in the escaped rats [5]. Such discrepancies between studies are usually attributable to differences in the animal model, or in the methodologies used to make the measurements. The former explanation seems unlikely in this situation given the similarities of the animal models. Instead, it is much more likely that the significant differences in the methodologies employed explain these discrepancies. Although both studies measured AVP or dDAVP-stimulated cAMP generation from medullary suspensions, the cAMP generation in the presence of 3-isobutyl-l-methylxanthine (IBMX) was 3-fold (basally) to 8-fold (stimulated) higher in the studies of Ecelbarger et al compared to Saito et al. This difference in cAMP generation may reflect the differences in tissue used to prepare the suspensions, as Saito et al used outer and inner medullary tissue, with a resulting greater admixture of thick ascending limb cells with collecting duct cells, whereas Ecelbarger et al used only inner medullary tissue, illustrating the difficulty with interpretation of adenylate cyclase activity measurements when using mixed cell tissue extracts. Even greater differences exist with regard to the radioligand binding assays. Saito et al employed a standard [3H]AVP binding assay, whereas Tian et al developed a new binding assay employing an AVP V2 receptor antagonist with an iodinatable tyrosine residue that does not interfere with receptor binding, thus allowing higher specific activity [11]. A review of previously reported values for Bmax of the AVP V2 receptor measured under basal conditions from inner medullary membrane preparations of Sprague-Dawley rats via saturation binding analyses shows means of 272 ⫾ 109 fmol/mg protein when iodinated radioligands were employed (four studies, including the results of Tian et al) and 200 ⫾ 69 fmol/mg protein when tritiated radioligands were used (four studies). In contrast, the Bmax found by Saito et al using a competition binding analysis was much higher (11,800 fmol/mg protein [5]), raising the possibility that differences in assay sensitivity could account for the discordant results regarding changes in Bmax with escape. Additional studies will therefore be required to resolve the apparent discrepancies between the present study and those done previously and, specifically, to address the most accurate methods for measuring receptor number and adenylate cyclase activity in this system. However, even after these issues are resolved, there are still more basic questions that demand answers. Whether the AQP-2 down-regulation with escape occurs via changes in AVP-V2 receptor expression and signal transduction, or whether it is regulated at other sites such as directly at
1609
the level of AQP-2 transcription, an inevitable question remains: What mediates these effects? Somehow, ECF volume expansion must be sensed by the collecting duct cells. Although changes in osmolality appear to be an attractive candidate for such a mediator, down-regulation of AQP-2 with escape occurs throughout the kidney, including in the cortex where osmotic perturbations are much less marked [12]. Furthermore, escape can be reversed even in the presence of continued hypo-osmolality [12], consistent with previous studies in AVP-infused dogs in which maintenance of constant renal perfusion pressure via servo-control prevented the occurrence of escape despite severe hypo-osmolality [4]. Taken together, these combined studies indicate that hypo-osmolality by itself is not a prerequisite for escape. Future studies will need to consider additional potential mediators of escape and particularly those endocrine, paracrine and mechanical factors involved with both sensing and controlling intrarenal hemodynamics and inner medullary blood flow [13]. Consequently, while we understand much more about escape today as a result of the studies of Ecelbarger et al, Murase et al, Tian et al, and Saito et al, escape from AVP-induced antidiuresis very much remains a work in progress. There are still more characters to be implicated, the likelihood of intricate subplots, and even the possibility of promiscuous hormonal-receptor relationships left to be written. Readers can therefore look forward to more interesting chapters before this story is completed. Joseph G. Verbalis Washington, DC, USA Correspondence to Joseph G. Verbalis, M.D., Professor of Medicine and Physiology, Georgetown University, 4000 Reservoir Rd., NW, 232 Building D, Washington, DC 20007, USA. E-mail: [email protected]
REFERENCES 1. Levinsky NG, Davidson DG, Berliner RW: Changes in urine concentration during prolonged administration of vasopressin and water. Am J Physiol 196:451–456, 1959 2. Chan WY: A study of the mechanism of vasopressin escape: Effects of chronic vasopressin and overhydration on renal tissue osmolality and electrolytes in dogs. J Pharmacol Exp Ther 184:244–252, 1973 3. Gross PA, Kim JK, Anderson RJ: Mechanisms of escape from desmopressin in the rat. Circ Res 53:794–804, 1983 4. Hall JE, Montani JP, Woods LL, Mizelle HL: Renal escape from vasopressin: Role of pressure diuresis. Am J Physiol 250: F907–F916, 1986 5. Saito T, Higashiyama M, Nagasaka S, et al: Role of aquaporin-2 gene expression in hyponatremic rats with chronic vasopressininduced antidiuresis. Kidney Int 60:1266–1276, 2001 6. Ecelbarger CA, Nielsen S, Murase T, et al: Role of renal aquaporins in escape from vasopressin-induced antidiuresis. J Clin Invest 99:1852–1863, 1997 7. Knepper MA: Molecular physiology of urinary concentrating mechanism: Regulation of aquaporin water channels by vasopressin. Am J Physiol 272:F3–F12, 1997 8. Verbalis JG, Drutarosky MD: Adaptation to chronic hypoosmolality in rats. Kidney Int 34:351–360, 1988 9. Nielsen S, Marples D, Frokiaer J, et al: The aquaporin family of water channels in kidney: An update on physiology and pathophysiology of aquaporin-2. Kidney Int 49:1718–1723, 1996
1610
Editorial
10. Ecelbarger CA, Chou CL, Lee AJ, et al: Escape from vasopressininduced antidiuresis: Role of vasopressin resistance of the collecting duct. Am J Physiol 274:F1161–F1166, 1998 11. Tian Y, Sandberg K, Murase T, et al: Vasopressin V2 receptor binding is down-regulated during renal escape from vasopressininduced antidiuresis. Endocrinology 141:307–314, 2000
12. Murase T, Ecelbarger CA, Baker EA, et al: Kidney aquaporin-2 expression during escape from antidiuresis is not related to plasma or tissue osmolality. J Am Soc Nephrol 10:2067–2075, 1999 13. Cowley AW Jr: Control of the renal medullary circulation by vasopressin V1 and V2 receptors in the rat. Exp Physiol 86:223S– 231S, 2000
EDITORIAL
Escape from antidiuresis: A good story Good stories often share common characteristics. They address topics of interest and/or significance, they are played out over extended periods of time, and they have elements of controversy, if not drama. But perhaps most important, they generally include a definable beginning, a middle, and an end. The story of escape from vasopressin [arginine vasopressin (AVP)]-induced antidiuresis certainly fulfills most of these criteria. The phenomenon of AVP escape has arguably been one of the most interesting unresolved aspects of renal function that has been studied over the last half century. How the kidney is able to escape AVP-induced antidiuresis is also of considerable significance because this process allows survival of the organism by allowing free water excretion, despite inappropriate secretion of AVP, thus effectively antagonizing the effects of one of the most powerful hormones involved in body fluid homeostasis. The story of escape has continued to evolve over the last half century since this phenomenon was first recognized in 1959 with the landmark report of Levinsky, Davidson, and Berliner [1]. And, as for controversy, each decade has witnessed a new potential explanation for this phenomenon, including decreased water permeability of collecting duct principal cells [1], dissipation of the cortico-medullary osmotic gradient [2], increased generation of renal prostaglandin E2 [3], and renal hemodynamic changes mediated by increased renal artery perfusion pressure leading to a pressure diuresis [4]. Although the various mechanisms proposed to underlie escape have differed, all studies to date are in agreement that expansion of the extracellular fluid (ECF) space by AVP-induced water retention is crucial for the onset and maintenance of AVP escape. Thus, we have the makings of an interesting, and controversial, story with a clear beginning: ECF volume expansion. But now we also know the ending of this story. In this issue of Kidney International, Saito et al [5] confirm the recent findings of Ecelbarger et al [6] that AVP escape is associated with a marked down-regulation of expression of the AVP-sensitive water channel, aquaporin-2 (AQP-2), in the kidney. Because AQP-2 insertion into the apical membrane of collecting tubule principal cells represents the final step of facilitated water transport from the colKey words: vasopressin, extracellular fluid, aquaporin-2, hypo-osmolality.
2001 by the International Society of Nephrology
lecting tubule lumen into the medullary interstitium [7], a regulatory mechanism that impacts at this stage of the urinary concentrating mechanism would be maximally effective at limiting AVP-induced antidiuresis. Using similar animal models of dilutional hyponatremia [8], the results of the two studies differ quantitatively, but not qualitatively. Ecelbarger et al found 70 to 80% decreases in l-deamino8-d-arginine vasopressin (dDAVP)-stimulated AQP-2 protein and mRNA expression in the kidney that correlated temporally with the decrease in urine osmolality and increase in urine volume characteristic of escape. Similarly, Saito et al found 35 to 55% decreases in AQP-2 protein and mRNA expression after 7 days of dDAVPinduced hyponatremia. In both cases, this marked downregulation of AQP-2 expression occurred despite pharmacological dDAVP infusion. Therefore, both studies clearly indicate an AVP-independent regulation of AQP-2 expression. Although a causal rather than correlational association between the AQP-2 down-regulation and AVP escape has not yet been definitely established, given the strong association between AQP-2 membrane insertion and collecting tubule water permeability, this amount of change in AQP-2 expression is clearly sufficient to alter the degree of antidiuresis produced [9]. Thus, it seems likely that this is, in fact, a very plausible ending to our story. However, a beginning and an end do not by themselves make a satisfying story. There needs to be a middle that ties these ends together and fleshes out the relationship between them. What are the events that occur between AVP-induced volume expansion and AQP-2 down-regulation, and how are they related? The results of Saito et al and previous studies differ substantially and further fuel the controversy that has surrounded this area for the last half century. Ecelbarger et al found a 40 to 45% down-regulation of dDAVP-stimulated cyclic adenosine monophosphate (cAMP) generation in inner medullary collecting duct cell suspensions prepared from escaped rats compared to controls [10]. Because cAMP mediates signal transduction between the AVP V2 receptor and its effects on both AQP-2 membrane insertion and AQP-2 transcription [9], resistance to AVP-stimulated adenylate cyclase activity would represent a prime candidate for modulation of AVP effects at a cellular level. Consistent with this idea, Tian et al found a marked down-regulation of the AVP V2 receptor number (Bmax) in escaped rats that was 70% below control levels and 35% below the
1608
Editorial
levels of nonescaped dDAVP-treated rats [11], which are known to undergo ligand-induced receptor desensitization. In contrast, Saito et al found no differences in cAMP generation between the escaped and control rats and observed a 40% reduction in V2 Bmax secondary to dDAVP treatment but not any further decreases in the escaped rats [5]. Such discrepancies between studies are usually attributable to differences in the animal model, or in the methodologies used to make the measurements. The former explanation seems unlikely in this situation given the similarities of the animal models. Instead, it is much more likely that the significant differences in the methodologies employed explain these discrepancies. Although both studies measured AVP or dDAVP-stimulated cAMP generation from medullary suspensions, the cAMP generation in the presence of 3-isobutyl-l-methylxanthine (IBMX) was 3-fold (basally) to 8-fold (stimulated) higher in the studies of Ecelbarger et al compared to Saito et al. This difference in cAMP generation may reflect the differences in tissue used to prepare the suspensions, as Saito et al used outer and inner medullary tissue, with a resulting greater admixture of thick ascending limb cells with collecting duct cells, whereas Ecelbarger et al used only inner medullary tissue, illustrating the difficulty with interpretation of adenylate cyclase activity measurements when using mixed cell tissue extracts. Even greater differences exist with regard to the radioligand binding assays. Saito et al employed a standard [3H]AVP binding assay, whereas Tian et al developed a new binding assay employing an AVP V2 receptor antagonist with an iodinatable tyrosine residue that does not interfere with receptor binding, thus allowing higher specific activity [11]. A review of previously reported values for Bmax of the AVP V2 receptor measured under basal conditions from inner medullary membrane preparations of Sprague-Dawley rats via saturation binding analyses shows means of 272 ⫾ 109 fmol/mg protein when iodinated radioligands were employed (four studies, including the results of Tian et al) and 200 ⫾ 69 fmol/mg protein when tritiated radioligands were used (four studies). In contrast, the Bmax found by Saito et al using a competition binding analysis was much higher (11,800 fmol/mg protein [5]), raising the possibility that differences in assay sensitivity could account for the discordant results regarding changes in Bmax with escape. Additional studies will therefore be required to resolve the apparent discrepancies between the present study and those done previously and, specifically, to address the most accurate methods for measuring receptor number and adenylate cyclase activity in this system. However, even after these issues are resolved, there are still more basic questions that demand answers. Whether the AQP-2 down-regulation with escape occurs via changes in AVP-V2 receptor expression and signal transduction, or whether it is regulated at other sites such as directly at
1609
the level of AQP-2 transcription, an inevitable question remains: What mediates these effects? Somehow, ECF volume expansion must be sensed by the collecting duct cells. Although changes in osmolality appear to be an attractive candidate for such a mediator, down-regulation of AQP-2 with escape occurs throughout the kidney, including in the cortex where osmotic perturbations are much less marked [12]. Furthermore, escape can be reversed even in the presence of continued hypo-osmolality [12], consistent with previous studies in AVP-infused dogs in which maintenance of constant renal perfusion pressure via servo-control prevented the occurrence of escape despite severe hypo-osmolality [4]. Taken together, these combined studies indicate that hypo-osmolality by itself is not a prerequisite for escape. Future studies will need to consider additional potential mediators of escape and particularly those endocrine, paracrine and mechanical factors involved with both sensing and controlling intrarenal hemodynamics and inner medullary blood flow [13]. Consequently, while we understand much more about escape today as a result of the studies of Ecelbarger et al, Murase et al, Tian et al, and Saito et al, escape from AVP-induced antidiuresis very much remains a work in progress. There are still more characters to be implicated, the likelihood of intricate subplots, and even the possibility of promiscuous hormonal-receptor relationships left to be written. Readers can therefore look forward to more interesting chapters before this story is completed. Joseph G. Verbalis Washington, DC, USA Correspondence to Joseph G. Verbalis, M.D., Professor of Medicine and Physiology, Georgetown University, 4000 Reservoir Rd., NW, 232 Building D, Washington, DC 20007, USA. E-mail: [email protected]
REFERENCES 1. Levinsky NG, Davidson DG, Berliner RW: Changes in urine concentration during prolonged administration of vasopressin and water. Am J Physiol 196:451–456, 1959 2. Chan WY: A study of the mechanism of vasopressin escape: Effects of chronic vasopressin and overhydration on renal tissue osmolality and electrolytes in dogs. J Pharmacol Exp Ther 184:244–252, 1973 3. Gross PA, Kim JK, Anderson RJ: Mechanisms of escape from desmopressin in the rat. Circ Res 53:794–804, 1983 4. Hall JE, Montani JP, Woods LL, Mizelle HL: Renal escape from vasopressin: Role of pressure diuresis. Am J Physiol 250: F907–F916, 1986 5. Saito T, Higashiyama M, Nagasaka S, et al: Role of aquaporin-2 gene expression in hyponatremic rats with chronic vasopressininduced antidiuresis. Kidney Int 60:1266–1276, 2001 6. Ecelbarger CA, Nielsen S, Murase T, et al: Role of renal aquaporins in escape from vasopressin-induced antidiuresis. J Clin Invest 99:1852–1863, 1997 7. Knepper MA: Molecular physiology of urinary concentrating mechanism: Regulation of aquaporin water channels by vasopressin. Am J Physiol 272:F3–F12, 1997 8. Verbalis JG, Drutarosky MD: Adaptation to chronic hypoosmolality in rats. Kidney Int 34:351–360, 1988 9. Nielsen S, Marples D, Frokiaer J, et al: The aquaporin family of water channels in kidney: An update on physiology and pathophysiology of aquaporin-2. Kidney Int 49:1718–1723, 1996
1610
Editorial
10. Ecelbarger CA, Chou CL, Lee AJ, et al: Escape from vasopressininduced antidiuresis: Role of vasopressin resistance of the collecting duct. Am J Physiol 274:F1161–F1166, 1998 11. Tian Y, Sandberg K, Murase T, et al: Vasopressin V2 receptor binding is down-regulated during renal escape from vasopressininduced antidiuresis. Endocrinology 141:307–314, 2000
12. Murase T, Ecelbarger CA, Baker EA, et al: Kidney aquaporin-2 expression during escape from antidiuresis is not related to plasma or tissue osmolality. J Am Soc Nephrol 10:2067–2075, 1999 13. Cowley AW Jr: Control of the renal medullary circulation by vasopressin V1 and V2 receptors in the rat. Exp Physiol 86:223S– 231S, 2000