Dwkwcdiuwnc(o

G

T

A

0©

S

Vasopressin Receptor Mutations inNephrogenic Diabetes Insipidus

Daniel G. Bichet, MD

Summary: The purpose of this review is first to describe the importance of early detection ofvasopressin receptor mutations responsible for X-linked nephrogenic diabetes insipidus(NDI). We have proposed that all families with hereditary diabetes insipidus should have theirmolecular defect identified because early diagnosis and treatment of affected infants can avertthe physical and mental retardation that results from repeated episodes of dehydration.Secondly, 95 published missense mutations responsible for X-linked NDI are likely to result inmisfolded arginine-vasopressin V2 receptors that are trapped in the endoplasmic reticulum.These misfolded receptors are unable to reach the plasma membrane in principal collectingduct cells and to engage the circulating antidiuretic hormone, arginine-vasopressin. Thesemisfolded proteins potentially could be rescued with pharmacologic chaperones, an activearea of research pertinent to other hereditary protein misfolding diseases such as cysticfibrosis, phenylketonuria, and Anderson-Fabry disease among many others. Finally, a long-term careful surveillance of all patients with hereditary NDI should be performed to preventchronic renal failure likely caused by the long-term functional tract obstruction with reflux.Semin Nephrol 28:245-251 © 2008 Elsevier Inc. All rights reserved.Keywords: Nephrogenic diabetes insipidus, AVPR2 mutations, mutational analysis, mis-folded mutants, pharmacological chaperones

ttatwisd

TAPTOC

A

Lrmscr

iabetes insipidus is a disorder character-ized by the excretion of abnormallylarge volumes (�30 mL/kg/d of body

eight for adults) of dilute urine (�250 mmol/g). This definition excludes osmotic diuresis,hich occurs when excess solute is being ex-

reted, for example, glucose in the polyuria ofiabetes mellitus. Four basic defects can be

nvolved: (1) deficient secretion of the antidi-retic hormone arginine vasopressin (AVP),hich is the most common and is referred to aseurohypophyseal (also known as neurogenic,entral, or hypothalamic) diabetes insipidus;2) renal insensitivity to the antidiuretic effectf AVP, which is known as nephrogenic diabe-

enetics of Renal Diseases, Groupe d’Étude des Protéines, Membranaires, Mon-tréal, Québec, Canada. Department of Medicine and Physiology, Université deMontréal, Research Centre, Montréal, Québec, Canada. Nephrology Service,Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada.

his study was supported by the Canadian Institutes of Health Research, and bythe Kidney Foundation of Canada. Dr. Bichet holds a Canada Research Chairin Genetics of Renal Diseases.

ddress reprint requests to Daniel G. Bichet, MD, Research Centre, Hôpital duSacré-Coeur de Montréal, 5400 Boulevard Gouin West, Montréal, QuébecH4J 1C5, Canada. E-mail: Daniel.bichet@umontreal.ca

270-9295/08/$ - see front matter

t2008 Elsevier Inc. All rights reserved. doi:10.1016/j.semnephrol.2008.03.005

eminars in Nephrology, Vol 28, No 3, May 2008, pp 245-251

es insipidus (NDI); (3) excessive water intakehat can result in polyuria, which is referred tos primary polydipsia; and (4) increased me-abolism of vasopressin during pregnancy,hich is referred to as gestational diabetes

nsipidus. The hereditary forms of diabetes in-ipidus account for less than 10% of the cases ofiabetes insipidus seen in clinical practice.

HE BASOLATERAL AVPR2 RECEPTORND THE LUMINAL AQUAPORIN-2ROTEINS ARE REQUIRED FOR THERANSEPITHELIAL WATER PERMEABILITYF THE PRINCIPAL CELLS OF THEOLLECTING DUCT

VPR2 is a GPCR

oss-of-function of either of these proteins willesult in NDI. The V2 receptor is one of 701embers of the rhodopsin family within the

uperfamily of guanine-nucleotide (G) protein-oupled receptors (GPCRs) (Fig. 1).1,2 GPCRsepresent the largest family of membrane pro-

eins in the human genome. They are remark-

245

assashi

GXu

Anpota

T

T

FiGA�ictlootctcpcmarc memb

246 D.G. Bichet

bly versatile signaling molecules that are re-ponsible for the majority of transmembraneignal transduction in response to hormonesnd neurotransmitters. GPCRs share a commontructural signature of 7 membrane-spanningelices with an extracellular N terminus and an

ntracellular C terminus (Fig. 2).3

Understanding how AVPR2 receptor, thePCR with loss-of-function responsible for-linked NDI, operates is a major goal to provide

igure 1. Effect of vasopressin (AVP) to increase waters bound to the V2 receptor (a G-protein–linked rece-protein–coupled receptor signaling consists of 3 stepVP) in the extracellular milieu, a G-protein (G�s) that dis� subunits after interaction with the ligand-bound re

nteracts with dissociated G-protein subunits to generatyclase, increasing the intracellular concentration of cAandem repeats of 6 hydrophobic transmembrane domaarge intracellular tail. The dimeric structure (C1 and C2)f ATP to cAMP takes place at the dimer interface. Two ar Mn2� represented here as 2 small black circles), whiche large open circle and the 3 phosphate groups (ATP)AMP. The binding of cAMP to the regulatory subunits ofo dissociate from the catalytic subunits. These activatontaining endocytic vesicle via an AKAP. The local conchosphodiesterases (PDEs). Cytoplasmic vesicles carryingomplexes) are fused to the luminal membrane in respoembrane. The dissociation of AKAP from the endocyti

re necessary for vesicle movement toward the membetrieved by an endocytic process, and water permeabhannels are expressed constitutively at the basolateral

nderstanding and treatment of this rare disease. (

high-resolution crystal structure of an engi-eered human �-2 adrenergic GPCR recently wasublished3–5 and it is hoped that structure ofther GPCRs, including the human AVPR2 recep-or, can guide the development of specific drugs,ccelerating drug discovery to treat X-linked NDI.

he cAMP-PKA Pathway

he human gene that codes for the V2 receptor

ability in the principal cells of the collecting duct. AVPon the basolateral membrane. The basic process ofpta-helical receptor that detects a ligand (in this case,s into � subunits bound to guanosine triphosphate and

r, and an effector (in this case, adenylyl cyclase) thatll-molecule second messengers. AVP activates adenylylhe topology of adenylyl cyclase is characterized by 2

parated by a large cytoplasmic loop and terminates in acatalytic domains is represented (see text). Conversionte residues (in C1) coordinate 2 metal co-factors (Mg2�

le the catalytic function of the enzyme.29 Adenosine is3 small open circles. PKA is the target of the generatedduces a conformational change, causing these subunits

bunits (C) as shown here are anchored to an AQP2-tion and distribution of the cAMP gradient is limited byater channel proteins (represented as homotetramericAVP, thereby increasing the water permeability of this

le is not represented. Microtubules and actin filamentsWhen AVP is not available, AQP2 water channels areeturns to its original low rate. AQP3 and AQP4 waterrane.

permeptor)s: a hesociateceptoe smaMP. Tins seof thespartah enabare thePKA ined suentrathe w

nse toc vesicrane.ility r

AVPR2) is located in chromosome region

Xsamdocp(npsmiittcrsc

pstnmo

ckompmtaofl(um

Fgmtnm .medi

Vasopressin receptor mutations in NDI 247

q28 and has 3 exons and 2 small introns. Theequence of the complementary DNA predictspolypeptide of 371 amino acids with 7 trans-embrane, 4 extracellular, and 4 cytoplasmic

omains (Fig. 2). The activation of V2 receptorn renal collecting tubules stimulates adenylylyclase via the stimulatory G protein (Gs) andromotes the cyclic adenosine monophosphatecAMP)-mediated incorporation of water chan-els into the luminal surface of these cells. Thisrocess is the molecular basis of the vasopres-in-induced increase of the osmotic water per-eability of the apical membrane of the collect-

ng tubule.6 In the collecting duct, the first stepn the antidiuretic action of AVP is its binding tohe vasopressin V2 receptor (Fig. 1) located onhe basolateral membrane of collecting ductells. This step initiates a cascade of events—eceptor-linked activation of the cholera toxin–ensitive G-protein (Gs), activation of adenylyl

igure 2. V2 receptor and identification of 193 putativeiven as the one-letter amino acid code. Solid symbolsore than one mutation in the same codon; other types

ransmembrane, and cytoplasmic domains are definedonsense, 46 frameshift deletion or insertion, 7 inframutations, and one complex mutation. See http://www

yclase, production of cAMP, and stimulation of m

rotein kinase A (PKA)—that leads to the finaltep in the antidiuretic action of AVP; that is,he exocytic insertion of specific water chan-els, aquaporin-2 (AQP2), into the luminalembrane, thereby increasing the permeability

f the luminal membrane.The cAMP-PKA pathway is one of the most

ommon and versatile signal pathways in eu-aryotic cells and is involved in the regulationf cellular functions in almost all tissues inammals, including regulation of cell cycle,roliferation, differentiation, and regulation oficrotubule dynamics, chromatin condensa-

ion and decondensation, nuclear envelope dis-ssembly and reassembly, as well as regulationf intracellular transport mechanisms and ionuxes.7 Because this single second messengerthe cAMP-PKA pathway) is involved in the reg-lation of so many diverse cellular processes, itust be highly regulated at several levels to

se-causing AVPR2 mutations. Predicted amino acids aree missense or nonsense mutations; a number indicatestations are not indicated on the figure. The extracellular,rding to Mouillac et al.30 There are 95 missense, 18etion or insertion, 4 splice-site, and 22 large deletioncine.mcgill.ca/nephros for a list of mutations.

diseaindicatof mu

accoe del

aintain specificity. AVP-induced changes in

cp

ttgaadnspddrpidm

ctmtauspfstupuvtlo

XFFP

Xttcdav

cQmmdt(mtaeAotmtcdfcpmrtbfisalambcfittidrtptasCwse

248 D.G. Bichet

AMP concentration may vary in duration, am-litude, and extension in the principal cells.A-kinase anchoring proteins (AKAPs) con-

ribute to the specificity of cAMP signaling byargeting PKA in proximity to cAMP gradientsenerated by the counterbalancing activities ofdenylyl cyclases and phosphodiesterases.8 Thectivation of AKAP-anchored PKA by cAMP iniscrete microdomains has been visualized ineonatal cardiomyocytes,9 and AKAP 18� (aplice variant of AKAP 18) has been found inrincipal cells of the inner medullary collectinguct in a distribution closely resembling theistribution of AQP2,10 which could imply aole for AKAP 18� in the AQP2 shuttle exocyticrocess. Chen et al11 recently showed that an

nherited AKAP9 mutation caused long-QT syn-rome but there is no description of AKAPutations responsible for hereditary NDI.AVP also increases the water reabsorptive

apacity of the kidney by regulating the urearansporter A1/3, which is present in the inneredullary collecting duct, predominantly in its

erminal part.12,13 AVP also increases the perme-bility of principal collecting duct cells to sodi-m.14 Vasopressin regulates the apical expres-ion of the epithelial sodium channel byreventing ubiquitin-dependent endocytosis

rom the cell surface.15 In summary, in the ab-ence of AVP stimulation, collecting duct epi-helia show very low permeabilities to sodiumrea and water. These specialized permeabilityroperties permit the excretion of large vol-mes of hypotonic urine formed during inter-als of water diuresis. In contrast, AVP stimula-ion of the principal cells of the collecting ductseads to selective increases in the permeabilityf the apical membrane to water, urea, and Na.

-LINKED NDI (OMIM 304800): LOSS-OF-UNCTION OF AVPR2, MISFOLDING, ANDUNCTIONAL RESCUE WITHHARMACOLOGIC CHAPERONES

-linked NDI is generally a rare disease in whichhe affected male patients do not concentrateheir urine after administration of AVP.16 Be-ause this form is a rare, recessive, X-linkedisease, female individuals are unlikely to beffected, but heterozygous females can show

ariable degrees of polyuria and polydipsia be- t

ause of skewed X-chromosome inactivation. Inuebec, the incidence of this disease amongale individuals was estimated to be approxi-ately 8.8 in 1,000,000 male live births.17 To

ate, 193 putative disease-causing AVPR2 mu-ations have been published in 307 NDI familiesFig. 2). Ninety-five of these 193 mutations areissense mutations likely to be misfolded,

rapped in the endoplasmic reticulum, and un-ble to reach the basolateral cell surface tongage the circulating antidiuretic hormone,VP. Most of the AVPR2 mutants that we andther investigators have tested are type 2 mu-ant receptors. They did not reach the cellembrane and were trapped in the interior of

he cell.18 Other mutant G-protein–coupled re-eptors19 and gene products causing geneticisorders also are characterized by protein mis-olding. Mutations that affect the folding of se-retory proteins; integral plasma membraneroteins; or enzymes destined to the endoplas-ic reticulum, Golgi complex, and lysosomes

esult in loss-of-function phenotypes irrespec-ive of their direct impact on protein functionecause these mutant proteins are preventedrom reaching their final destination.20 Foldingn the endoplasmic reticulum is the limitingtep: mutant proteins that fail to fold correctlyre retained initially in the endoplasmic reticu-um and subsequently often degraded. For ex-mple, cystic fibrosis is caused mainly by pointutations in the gene encoding an apical mem-

rane adenosine triphosphate (ATP)-regulatedhloride channel, which is known as the cysticbrosis transmembrane conductance regula-or (CFTR). The main disease-associated muta-ion, �F508 (deletion of the phenylalanine res-due at position 508 of the wild-type protein),isrupts the folding of CFTR in the endoplasmiceticulum, leading to almost complete degrada-ion of this channel. Interestingly, however,roperly folded CFTR with this mutation canraffic to the plasma membrane, where it forms

functional chloride channel. These findingsuggest that rescuing the folding of �F508-FTR could be used eventually to treat patientsith cystic fibrosis.21 Wang et al22 used the

hotgun liquid chromatography mass spectrom-try (LC-MC) method of multidimensional pro-

ein-identification technology to analyze cells

eCtttpnwbtoHi

dwA

mcrtbm

ietttfmea

F5pt .24

Vasopressin receptor mutations in NDI 249

xpressing the gene encoding the wild-typeFTR and �F508-CFTR. Multidimensional pro-

ein-identification technology analysis of wild-ype CFTR and �F508-CFTR immunoprecipi-ates identified nearly 200 CFTR-associatedroteins (compared with controls in whichonspecific antibodies or cells lacking CFTRere used). Collectively, these proteins haveeen named the CFTR interactome. These pro-eins included known CFTR-binding chaper-nes, such as calnexin, HSP40-HSP-70, andSP90, as well as many previously unknown

nteractors.RNA interference (RNAi)-mediated knock-

own of the HSP90 co-chaperone present inild-type and �F508-CFTR immunoprecipitates,

igure 3. (A) Urine volume, urine osmolality, and wateadult male patients with X-linked NDI. (B) The same vam) when the effect of SR49059 was suspected to be mtest. Data reprinted with permission from Bernier et al

HA1, corrected the amount of both endoplas- a

ic reticulum–associated and cell surface–asso-iated �F508-CFTR. These data suggest that dis-uption of AHA1 facilitates a folding pathwayhat favors not only the stability of the channel,ut also coupling to the endoplasmic exportachinery.If the misfolded protein/traffic problem that

s responsible for so many human genetic dis-ases can be overcome and the mutant proteinransported out of the endoplasmic reticulumo its final destination, then these mutant pro-eins could be sufficiently functional.23 There-ore, using pharmacologic chaperones or phar-acoperones to promote escape from the

ndoplasmic reticulum is a possible therapeuticpproach. We used selective nonpeptide V2-

e on day 1 and after SR49059 administration (day 3) inre described for the afternoon period (2:00 pm to 8:00l. Mean values (�SEM) are presented. *P � .05, paired

r intaklues aaxima

nd V1-receptor antagonists to rescue the cell-

socaattaribtcV

pRitwttasi

LTR

ArtbdAirafubssptp

ATa

R

Fap(c

250 D.G. Bichet

urface expression and function of naturallyccurring misfolded human V2 receptors.24 Be-ause the beneficial effect of nonpeptide V2

ntagonists could be secondary to preventionnd interference with endocytosis, we studiedhe R137H mutant previously reported to leado constitutive endocytosis. We found that thentagonist did not prevent the constitutive �-ar-estin–promoted endocytosis.25 These resultsndicate that as for other AVPR2 mutants, theeneficial effects of the treatment result fromhe action of the pharmacologic chaperones. Inlinical studies, we administered a nonpeptide

igure 4. Dilatation of the (A) kidney calices and ureternd (B) lower urinary tract (bladder) in a 22-year-oldatient bearing an AVPR2 mutation. The bladder residuepostvoiding) was 2 L and improved considerably afteratheterization, then with double micturition.

1a vasopressin antagonist SR49059 to 5 adult

atients who have NDI and bear the del62-64,137H, and W164S mutations. SR49059 signif-

cantly decreased urine volume and water in-ake and increased urine osmolality (Fig. 3),hereas sodium, potassium, creatinine excre-

ion, and plasma sodium were constanthroughout the study.24 This new therapeuticpproach could be applied to the treatment ofeveral hereditary diseases resulting from errorsn protein folding and kinesis.

ARGE DILATATION OF THE URINARYRACT AND BLADDER COULD LEAD TOEFLUX AND CHRONIC RENAL FAILURE

ll polyuric states (whether neurogenic, neph-ogenic, or psychogenic) can induce large dila-ations of the urinary tract and bladder, andladder function impairment has been wellocumented in patients who bear AVPR2 orQP2 mutations (Fig. 4).26,27 Of interest, an

nducible mouse model of NDI was producedecently by floxed Aqp2 gene deletion,28 whichlso showed evidence of structural damagerom the sustained polyuria. Chronic renal fail-re secondary to bilateral hydronephrosis haseen observed as a long-term complication inome rare patients. Renal and abdominal ultra-ound should be performed annually, and sim-le recommendations, including frequent urina-ion and double voiding, could be important torevent these consequences.

cknowledgmentshe author thanks Ivana Arapovic for secretarialssistance.

EFERENCES1. Fredriksson R, Lagerstrom MC, Lundin LG, Schioth

HB. The G-protein-coupled receptors in the humangenome form five main families. Phylogenetic analy-sis, paralogon groups, and fingerprints. Mol Pharma-col. 2003;63:1256-72.

2. Perez DM. The evolutionarily triumphant G-protein-coupled receptor. Mol Pharmacol. 2003;63:1202-5.

3. Rasmussen SG, Choi HJ, Rosenbaum DM, Kobilka TS,Thian FS, Edwards PC, et al. Crystal structure of thehuman beta2 adrenergic G-protein-coupled receptor.Nature. 2007;450:383-7.

4. Cherezov V, Rosenbaum DM, Hanson MA, RasmussenSG, Thian FS, Kobilka TS, et al. High-resolution crystalstructure of an engineered human beta2-adrenergic G

protein-coupled receptor. Science. 2007;318:1258-65.

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

Vasopressin receptor mutations in NDI 251

5. Rosenbaum DM, Cherezov V, Hanson MA, RasmussenSG, Thian FS, Kobilka TS, et al. GPCR engineering yieldshigh-resolution structural insights into beta2-adrenergicreceptor function. Science. 2007;318:1266-73.

6. Nielsen S, Frokiaer J, Marples D, Kwon TH, Agre P,Knepper MA. Aquaporins in the kidney: from mole-cules to medicine. Physiol Rev. 2002;82:205-44.

7. Tasken K, Aandahl EM. Localized effects of cAMPmediated by distinct routes of protein kinase A.Physiol Rev. 2004;84:137-67.

8. McConnachie G, Langeberg LK, Scott JD. AKAP sig-naling complexes: getting to the heart of the matter.Trends Mol Med. 2006;12:317-23.

9. Zaccolo M, Pozzan T. Discrete microdomains withhigh concentration of cAMP in stimulated rat neona-tal cardiac myocytes. Science. 2002;295:1711-5.

0. Henn V, Edemir B, Stefan E, Wiesner B, Lorenz D,Theilig F, et al. Identification of a novel A-kinaseanchoring protein 18 isoform and evidence for itsrole in the vasopressin-induced aquaporin-2 shuttle inrenal principal cells. J Biol Chem. 2004;279:26654-65.

1. Chen L, Marquardt ML, Tester DJ, Sampson KJ, Ack-erman MJ, Kass RS. Mutation of an A-kinase-anchoringprotein causes long-QT syndrome. Proc Natl Acad SciU S A. 2007;104:20990-5.

2. Yang B, Bankir L, Gillespie A, Epstein CJ, Verkman AS.Urea-selective concentrating defect in transgenicmice lacking urea transporter UT-B. J Biol Chem.2002;277:10633-7.

3. Yang B, Bankir L. Urea and urine concentrating abil-ity: new insights from studies in mice. Am J PhysiolRenal Physiol. 2005;288:F881-96.

4. Bankir L, Fernandes S, Bardoux P, Bouby N, BichetDG. Vasopressin-V2 receptor stimulation reduces so-dium excretion in healthy humans. J Am Soc Nephrol.2005;16:1920-8.

5. Snyder PM. Minireview: regulation of epithelial Na�channel trafficking. Endocrinology. 2005;146:5079-85.

6. Fujiwara TM, Bichet DG. Molecular biology of hered-itary diabetes insipidus. J Am Soc Nephrol. 2005;16:2836-46.

7. Arthus MF, Lonergan M, Crumley MJ, Naumova AK,Morin D, De Marco LA, et al. Report of 33 novelAVPR2 mutations and analysis of 117 families withX-linked nephrogenic diabetes insipidus. J Am SocNephrol. 2000;11:1044-54.

8. Morello JP, Salahpour A, Laperrière A, Bernier V,Arthus M-F, Lonergan M, et al. Pharmacological chap-

erones rescue cell-surface expression and function of

misfolded V2 vasopressin receptor mutants. J ClinInvest. 2000;105:887-95.

9. Conn PM, Ulloa-Aguirre A, Ito J, Janovick JA. G pro-tein-coupled receptor trafficking in health and dis-ease: lessons learned to prepare for therapeutic mu-tant rescue in vivo. Pharmacol Rev. 2007;59:225-50.

0. Romisch K. A cure for traffic jams: small moleculechaperones in the endoplasmic reticulum. Traffic.2004;5:815-20.

1. Cravatt BF, Simon GM, Yates JR 3rd. The biologicalimpact of mass-spectrometry-based proteomics. Na-ture. 2007;450:991-1000.

2. Wang X, Venable J, LaPointe P, Hutt DM, Koulov AV,Coppinger J, et al. Hsp90 cochaperone Aha1 down-regulation rescues misfolding of CFTR in cystic fibro-sis. Cell. 2006;127:803-15.

3. Cohen FE, Kelly JW. Therapeutic approaches to pro-tein-misfolding diseases. Nature. 2003;426:905-9.

4. Bernier V, Morello JP, Zarruk A, Debrand N, Salah-pour A, Lonergan M, et al. Pharmacologic chaperonesas a potential treatment for X-linked nephrogenicdiabetes insipidus. J Am Soc Nephrol. 2006;17:232-43.

5. Bernier V, Lagace M, Lonergan M, Arthus MF, BichetDG, Bouvier M. Functional rescue of the constitu-tively internalized V2 vasopressin receptor mutantR137H by the pharmacological chaperone action ofSR49059. Mol Endocrinol. 2004;18:2074-84.

6. Ulinski T, Grapin C, Forin V, Vargas-Poussou R, De-schenes G, Bensman A. Severe bladder dysfunction ina family with ADH receptor gene mutation responsi-ble for X-linked nephrogenic diabetes insipidus.Nephrol Dial Transplant. 2004;19:2928-9.

7. Shalev H, Romanovsky I, Knoers NV, Lupa S, LandauD. Bladder function impairment in aquaporin-2 defec-tive nephrogenic diabetes insipidus. Nephrol DialTransplant. 2004;19:608-13.

8. Yang B, Zhao D, Qian L, Verkman AS. Mouse model ofinducible nephrogenic diabetes insipidus producedby floxed aquaporin-2 gene deletion. Am J PhysiolRenal Physiol. 2006;291:F465-72.

9. Tesmer JJ, Sunahara RK, Gilman AG, Sprang SR. Crys-tal structure of the catalytic domains of adenylyl cy-clase in a complex with Gsalpha.GTPgammaS. Sci-ence. 1997;278:1907-16.

0. Mouillac B, Chini B, Balestre MN, Elands J, Trumpp-Kallmeyer S, Hoflack J, et al. The binding site ofneuropeptide vasopressin V1a receptor. Evidence fora major localization within transmembrane regions.

J Biol Chem. 1995;270:25771-7.