Download Aquaporin-2 Water Channel Mutations Causing

!
Vasopressin: Genes, RecePtors, Water Channels, and Antagonists
el
Aquaporin-2 Water Channel Mutations Causing
Nephrogenic Diabetes InsiPidus
Carel H. van Os and Peter M.T. Deen
Department of Cell Physiology, University of Nijmegen, Nijmegen, The Netherlands
I
v
I
Since the discovery of aquaporin water channels, insight into the molecular mechanism by which rapid osmotic
water occurs across cell membranes has greatly improved. Aquaporin-2 is the vasopressin-responsive water
channel in the collecting duct, and vasopressin control of water permeability in the collecting duct occurs in two
ways: a short-term regulation and a long-term adaptation. In congenital nephrogenic diabetes insipidus, the kidney does not respond to vasopressin. Ninety percent of these patients cany a mutation in the gene coding for the
vasopressin V2 receptor located on the X chromosome. Autosomal recessive and dominant forms of nephrogenic
diabetes insipidus that are caused by mutations in the aquaporin-2 gene have now been described. This review
focuses on recent insight in the molecular and cellular defect in autosomal nephrogenic diabetes insipidus.
¡^\ ne of the most important functions of the kidney
(J l, to regulate body water balance' without major
REGULATION OF COLLECTING DUCT
WATER PERMEABILITY
changes in solute excretion. The "antidiuretic horrnone" arginine vasopressin (AVP) plays a key role in
renal water excretion, which is the basis of osmoregulation. The best understood effect of AVP is the increase in the water permeability of the collecting duct
principal cells after binding of AVP to its V2 receptor.
This process allows for rapid osmotic water flow from
tubular lumen to the blood.
The discovery of the aquaporin family of water
channels has provided insight in the molecular mechanism by which rapid osmotic water flow occurs in
kidney and in other organs. Nine mammalian aquaporins have been cloned to date, seven of which are
expressed in kidney (for a review, see l-4). Only one
of these aquaporins, namely aquaporin-2 (AQP2), has
been shown to be essential in AVP-dependent concentration of urine (5). It is now well established that
in antidiuresis AQP2 is abundant in the apical membrane ofcollecting duct principal cells, which is the ratelimiting barrier for transepithelial water transport and
the chief site of AVP action (for a review, see 6-8).
The collecting duct water permeability can be changed
Key words: kidney, water excretion, collecting duct, vasopressln.
Address correspondence and reprint requests to: Carol H. van Os,
M.D., Department of Cell Physiology, University of Nijrnegen,
P.O. Box 9101, 6500 HB Nijmegen, The Netherlands.
Received 15 January 1998; Accepted l2 March 1998.
Copyright
@
in two ways: a shorl-term (min) regulation and a longterm (hr) adaptation. The short-term regulation is a
consequence of AVP binding to its V2 receptor and a
subsequent adenosine 3:5-cyclic phosphate (cAMP)dependent insertion of AQP2-containing vesicles into
the apical membrane (9). This process is rapidly reversible after dissociation of AVP from its receptor.
The molecular machinery for exocytic insertion and en-
docytic retrieval of AQP2-bearing vesicles is largely
unknown, but recent studies hint at a similar mechanism as proposed for regulated exocytosis of synaptic
vesicles. Members of an important group of proteins
that mediate docking and fusion of a vesicle to its acceptor membrane have been identified in collecting
duct principle cells (8,10,1
l).
Long-term adaptation to circulating AVP levels
increases collecting duct water permeability by increasing the expression level of AQP2 (12). In the
promotor region of the AQP2 gene, different cisacting elements have been reported that are supposed
to be involved in AVP-induced expression of a reporter gene (13,14). The main action of AVP seems to
be a transcriptional regulation, which is mediated by
phosphorylation of a cAMP response element-binding
protein and binding of phosphorylated cAMP response
element-binding protein to the cAMP response element in the promoter region of the AQP2 gene (15).
lggS,Proceedings of the Association of American Physicians, Volume 110, Number 5' pp. 395-400
396
Proceedings
ofthe Association ofAmerican Physicians 110:5 September/October 1998
DIABETES INSIPIDUS
Because AVP is the hormone that controls serum osmolality by decreasing free water clearance, any condition
that interferes with AVP production, secretion, and
binding to V2 receptors or with AQP2 synthesis and
trafficking will result in loss of the ability to concentrate
urine. In patients suffering from familial central diabetes insipidus, numerous mutations within the AVP gene
have recently been identified (for a review, see 16). This
disorder is transmitted in an autosomal dominant manner and typically presents in early childhood but can,
however, be treated by 1-desamino-8-D-arginine vasopressin (dDAVP) administration.
In nephrogenic diabetes insipidus (NDI), the kidney
does not respond to AVP, and here two forms are recognized: acquired NDI and congenital NDI. The congenital form of NDI is relatively rare and is estimated to be
present in approximately four in 1 million newborns
(17). In most families, NDI shows a Xlinked recessive
mode of inheritance. The diseased Xlinked NDI gene
was identified in 1992 when the V2 receptor was cloned
and the first mutations in this gene were reported (1821). Many investigators have reported on the heterogeneity in congenital NDI. In addition, some NDI patients
showed a normal extrarenal V, receptor response after
dDAVP infusion, which is release of coagulation and fi-
brinolytic factors and vasodilatory response (22). Both
autosomal recessive NDI and dominant NDI have now
been reported, and mutations in the AQP2 gene are the
cause (5,23,24). In the first year of life, patients suffering from congenital NDI are at risk of severe dehydration, and at this age, symptoms such as vomiting, anorexia, fever, failure to thrive, and mental retardation are
predominant. Early recognition of congenital NDI with
an abundant intake of water allows a normal life span
,25)' Hence, the
of
NDI has imporgenetic
causes
of
different
discovery
especially in
genetic
counseling,
for
tant implications
with normal mental development
(11
major intrinsic protein is the first cloned member.
Hydrophobicity plots predicted that major intrinsic
protein family members have six hydrophobic transmembrane domains or ct helices with intracellular
amino- and carboxy-termini. The six bilayer-spanning
domains are connected by five loops (Fig. 1, A-E).
The molecule consists of two repeats of three ct helices, which are 18O-degree miffor images of each other
(Fig. l). Each repeat contains the highly conserved
family characteristic asparagine-proline-alanine sequence in loops B and E. Loops B and E are postulated to fold back into the membrane and form the water pore. Recently, the three-dimensional structure of
AQÞl was determined at 6 Å resolution by cryoelectron microscopy (27). AQPI is a homotetramer containing four independent aqueous channels' It was
shown that each monomer is composed of six tilted c¿
helices that form a barrel that encloses a central density, which is attributed to the functionally important
asparagine-proline-alanine boxes in loops B and E.
Given the highly homologous primary structures of
AQPI and AQP2, there is no reason to postulate a
completely different three-dimensional structure for
I
À
the AVP-responsive water channel AQP2, and the
naturally occurring mutations in the AQP2 gene provide support for the structural similarity of AQPI and
AQP2 (28).
The less-conserved region among the aquaporins
is the carboxy-terminus, and this property has been
exploited to generate highly specific polyclonal antibodies. The tail of AQP2 contains a protein kinase A
(PKA) phosphorylation site, Serine 256, which plays
an important role in PKA-induced exocytic insertion
of AQP2-containing vesicles into the apical
brane.
mem-
It has been shown that phosphorylation of
1f
v
À
Ser256 did not alter the water permeability of AQP2
(29), but when 5256 was replaced with alanine
(52564) and LLC-PK, cells were stably transfected
those families in which only one patient is affected.
Acquired NDI is much more common than congenital NDI and often occurs as a side effect in humans subjected to lithium therapy or as a secondary
phenomenon in low-protein diet, hypercalciuria, hypokalemia, ureteral obstruction, and puromycin aminonucleotide-induced nephrosis (for a review, see
6,24,26). Decreased AQP2 abundance is a striking observation in acquired NDI (6,26).
In this review, we focus on the most seldom form
of NDI, the autosomal recessive and dominant forms
in which AQP2 gene mutations are the cause.
MOLECULAR STRUCTURE OF AQUAPORINS
Aquaporins are members of a large family of membrane-intrinsic proteins of which the lens fiber cell
c
A
Outside
E
Ë
B
lnside
cooH
Figure 1, Proposed functional model of the aquaporin-2 water
channel.
The molecule consists of six transmembrane sggments, connected
by loops A to E, with cytoplasmic amino- and carboxy-termini' Indicated are the highly conserved asparagine-proline-alanine rnotifs
(NPA) ancl the serine, phosphorylated by protein kinase A (S256)'
\
I
van Os and Deen: Vasopressin and Water Channels
397
-
32kD, which proved to be sensitive to deglycosylation by endoglycosidase H (endo H; 36) (Fig.
3). In general, plasma membrane integral proteins un-
band of
with an AQP2S256A expression construct, the AVPinduced exocytosis seen in wild{ype AQP2-expressing
cells was no longer observed (30).
dergo several types ofposttranslational processes dur-
ing their passage from the endoplasmic reticulum
(ER) through the Golgi apparatus. After synthesis,
AUTOSOMAL RECESSIVE NDI
In l}Vo of families, NDI shows a non-X-linked
they are first N-glycosylated with high mannose residues, which catr be trimmed by endo H. During passage through the Golgi complex, deglycosylation and
often reglycosylation with different sugar residues to
a complex structure take place that cannot be removed
by endo H (37; Fig. 3). Therefore, the appearance of
endo H-sensitive bands on immunoblots represents a
high mannose ER-retarded form of AQP2, and immunocytochemistry further corroborated this (36). Oo-
pat-
tern of inheritance. In one patient from such a family,
a normal extrarenal response to dDAVP had been observed, and sequencing of the AQP2 gene in this patient revealed two point mutations (5). Since this first
identification of AQP2 gene mutations, 14 mutations
have now been reported in autosomal-recessive NDI
(Fig. 2). Three patients were compound heterozygotes
(5,31 ,32), whereas 10 patients were siblings from consan guineous parents and homozyg otes (28,3 1,33,3 4).
Of these mutations in the AQP2 gene, one consists of
a nucleotide deletion (463delC;35) leading to a truncated protein. One is a nonsense mutation (G100x;
'/
cytes injected with wild-type AQP2 cRNA always
show intense staining of the plasma membrane,
whereas mutant AQP2 cRNA injections result in intense intracellular staining and only weak staining of
the oolemma (28,36). Although protein routing in Xenopus ooeytes and mammalian cells can be quite different, impaired cellular routing of two AQP2 mutants
was confirmed in transiently transfected Chinese
hamster ovary (CHO) cells (32).
Expression in oocytes and in CHO cells also re-
34), and the remaining are missense mutations.
To test whether identified mutations in AQP2 are
causal for NDI and not polymorphisms with no significant effects on AQP2 function, we started to test encoded missense mutants in heterologous expression
systems. Xenopus oocytes have been very useful in
aquaporin expression studies because of the simple
osmotic swelling test to quantitate osmotic water permeability conferred by expressed aquaporins (1-3).
Expression in oocytes revealed that all missense
mutations studied were impaired in their routing to
the plasma membrane. This was concluded from immunoblots of oocytes expressing mutant AQP2 proteins. These immunoblots showed besides the wildtype AQP2 band of : 29 kD always an additional
R187C
vealed that three mutant AQP2 proteins (L22Y,
TI26M, Al4lT;Fig.2) did confer a significant, albeit
smaller than wild-type, water permeability to the
plasma membrane. All other mutant proteins studied
do not confer any water permeability after expression.
The explanation for these observations is most likely
that mutations outside the functionally important B
and E loops (Fig. 1) induce minor or subtle changes in
the three-dimensional structure that do not impair the
water pore but are nevertheless noticed by the sorting
A1 9OT
P1854
c18'tw
Tr28¡il
w202c
E)dracêllular
GlOOX
s21
A.147f
lntracellular
G64R
\
2. Proposed model of the aqua'
porin-2 water channel.
Figure
N883
E258K
wlM
Mútations detected in patients with autosomalrecessive or dominant nephrogenic diabetes
insipidus are indicated by text (missense/
nonsense) or bars (nucleotide insertions or
deletions resulting in a frameshift).
Y-
398
Proceedings ofthe Association ofAmerican Physicians
@
@
@
tt-Rcetylglucosamine
@
@
Galactose
110:5
September/October 1998
Mannose
Glucose
Endo H resistant
Endo H sensitive
s¡alic nc¡o
l.:-'.
. i".1r.;,,'
ta'
glucosidase
and ll
Mannosidase
I
:'::l'ij.t,,:j
@
'fl'
@
@
I
7
Æ
cis
Endoplasmic
-;:._....111-:.r::..
Reticulum
@@
@
medial
trans and TGN
Golgi Complex
3. Glycosylation states of proteins in different subcellular compartments.
prðteins are synthesizecl in the endopiasmic reticulum and traverse via the (cls, medial, and /rans) Golgi complex and trzrs-Golgi network (TGN) to
their.final destination. In these different compafiments, ploteins have a different state of glycosylation. Using specific glycosidases [e.g., endoglycosican be detetmined.
dase H (endo H)ì, the localization of aquapoi'in-2 proteins in nephrogenic diabetes insipidus, which are in.rpaired in their tt'ansport,
Figure
machinery of the ER, resulting in retarded routing.
The impaired routing of these functional AQP2 mu-
tion resulted in routing impairment in rat and human
AQP2, whereas the same mutation C1894 in AQP1
tants most likely explains the cause of NDI in these
patients. In addition, these results support the idea that
AQP1 and AQP2 have identical three-dimensional
had no effect on routing (41). This suggests that mutations in AQPI are better tolerated than in AQP2 and
hints at subtle differences in tertiary structures, making AQP2 more susceptible to mutations.
structufes.
Mutations in proteins can, in principle, interfere
with proper folding in the ER, which then leads to
degradation, and this so-called quality control of the
ER thus results in a lower stability of the mutant proteins (38). Surprisingly, only two mutant AQP2 proteins (S216P, A141T;Fig.2) were found unstable after expression in oocytes (28,36). Because both
mutations are located in a transmembrane domain,
misfolding may cause the exposure of hydrophobic
regions on the surface of the molecule, which is an
important signal in the quality control of the ER, and
this could result in a higher rate of degradation (39).
One mutation (C18lW) deserves special attention. The Cys181 mutated in AQP2 and the Cysl89 in
AQPl are the sites of inhibition of water permeation
by mercurial compounds (1-3). Bai et al. (40) described that a Cl8lA mutation in AQP2 results in a
functional water channel that is mercurial insensitive
and that a Cl8lW mutation gives a nonfunctional water channel. In our hands, a Cl8lA and Cl81S muta-
AUTOSOMAL DOMINANT NDI
Recently, three missense mutations that were the
cause of autosomal dominant NDI have been found in
the AQP2 gene (31). In contrast to the AQP2 mutations in recessive NDI, which are all located in between the first and last transmembrane domain, the
dominant mutations are predicted to affect the Cterminus of AQP2. A dominant form of inheritance,
as shown for other diseases, occurs when a mutant
protein oligomerizes with other subunits of a functional complex and disturbs the intracellular routing
or function of the complex. As described earlier here,
AQPl functional subunits oligomerize into homotetramers, and other aquaporins most likely do the same.
Therefore, the dominant action of mutant AQP2 proteins in NDI can only be explained if the mutant protein oligomerizes with wild-type AQP2 and that any
van Os and Deen: Vasopressin and Water Channels
tetramer containing one or more mutant monomers is
impaired in its routing after oligomerization. It is yet
unknown in which organelle AQP2 oligomerization
occurs, but likely compartments are the ER and the
trans-Golgí network, which are known to assemble
multiprotein complexes as the T-cell antigen receptor
and the gapjunction protein connexin-43 (42,43).
One dominant AQP2 mutation has now been ana-
lyzed in detail (44). A point mutation (G8664) in
only one allele causes a substitution of a lysine for a
glutamic acid in the C-terminus of AQP2 at position
258 (E258K; Fig.2). The E258K is only two residues
downstream from 5256, the residue which becomes
phosphorylated by PKA after V, receptor stimulation
by AVP. It was demonstrated rhat AQP2E258K was
phosphorylated as wild-type AQP2. Expression in oocytes revealed that AQPE258K was retarded in the
Golgi or post-Golgi compartment and did not reach
the plasma membrane. In coexpression experiments,
expression of low levels of AQP2-E258K, but not
AQP2-RI87C (a mutant in recessive NDI), appeared
to interfere with the function of wt-AQP2. Therefore,
since AQPs form tetramers and AQP2-8258K is retained in the Golgi apparatus, oligomerization of AQP2E258K with wt-AQP2 and the subsequent impaired
routing of the hetero-tetramers likely explain dominant NDI in this particular family. Although oocytes
provide valuable data on AQP2 mutants in NDI, in
contrast to the principal cells of collecting ducts, they
are not polarized and do not express V, receptors. Several mammalian epithelial cell lines do have these features, and preliminary results showed that AvP-induced
AQP2 shuttling can be mimicked after stable transfection of such cell lines with AQP2 constructs (45-47).
Therefore, these stably transfected cell lines provide
excellent cell models for studying both short-term
regulation of AQP2 and routing defects in autosomal
dominant NDL
REFERENCES
1. Nielsen S. and Agre P. The aquaporin family of water
channels in kidney. Kidney Int.48 1057-1068, 1995.
2. King L.S. and Agre P. Pathophysiology of the aquaporin water channels. Annu. Rev. Physiol. 58 619-648,
1996.
3. Verkman 4.S., Van Hoek 4.N., Ma T., et al. Water
transport across mammalian cell membranes. An. "/.
P hy s io I. 27 0: C12*C30, 1996.
4. Mulders S.M., Van Lieburg 4.F., Monnens L.A.H., et
al. Physiology and pathophysiology of aquaporins. Eur.
J. Clin. Invest. 26 104l-1050, 1996.
5. Deen P.M.T., Verdijk M.A.J., Knoers N.V.A.M., et al.
Requirement of human renal water channel aquaporin-2
for vasopressin-dependent concentration of urine. ,Scl-
ence
399
264:92*95, 1994.
6. Nielsen S., Marples D., Trøkiaer J., et al. The aquaporin
family of water channels in the kidney: An update on
physiology and pathophysiology of aquaporin-2. Kidney Int. 49: l'7 18-1723, 1996.
7. Knepper M.A. Molecular physiology of urinary concentrating mechanism: Regulation of aquaporin wate¡ channels by vasopressin. A¡2. J. Physiol.272:F3-F12,1997.
8. Knepper M.A. and Inoue T. Regulation of aquaporin-2
water channel trafficking by vasopressin. Curr. Opin.
Cell Biol.9: 560-564, 1997.
9. Nielsen S., Chou C.-L., Marples D., et al. Vasopressin
increases water permeability of kidney collecting duct
by inducing translocation of aquaporin-CD water channels to plasma membrane. Proc. Natl. Acad. Scí. U.S.A.
92: 1013^1017,1995.
10. Nielsen S., Marples D., Mohtashami M., et al. Expression of VAMP2-like protein in kidney collecting duct
intracellular vesicles: Colocalization with aquaporin-2
water channels . J. Clín. Invest. 96: 1834-1844, 1995.
11. Mandon 8., Chou C.-L., Nielsen S., and Knepper M.A.
Syntaxin-4 is localized to the apical plasma membrane
. of rat renal collecting duct cells: Possible role in aquaporin-2 trafficking. l. C I i n. I nv e s t. 98: 906-9 13, 199 6.
12.
DiGiovanni S.R., Nielsen S., Chlistensen E.L, and
Knepper M.A. Regulation of collecting duct water
channel expression by vasopressin in Brattleboro rat.
Proc. Natl. Acad. Scí. U.S.A.9l 8984-8988, 1994.
13. Hozawa S., Holtzman E.J., and
Ausiello D.A. cAMP
motifs regulating transcription in the aquaporin-2 gene.
Am. J. PhysioL 39: C1695-C1702,1996.
14. Yazú M., Zelenin S.M., Celsi G., and Aperia A. Adenylate cyclase-coupled vasopressin receptor activâtes
AQP2 promotor via a dual effect on CRE and APl elements. Am. J. Physíol. 4l: F443-F450, 1997.
15. Matsumura Y., Uchida S., Rai T., et al. Transcription
regulation of aquaporin-2 water channel gene by cAMP.
J. Am. Soc. Nephrol.
16. Robertson G.L. and
S:861-867,1997.
Berl T. Pathophysiology of water
metabolism. In: Brenner B.M. and Rector R.C. (eds).
Disturbances ín Control of Body Fluid Volume and Compositíon. Philadelphia: Saunders, 1995. Pp. 873-928.
17. Bichet D.G. Vasopressin receptors in health and disease. Kidney Int. 49 l7 06-1'7 11, 1996.
18. Lolait S.J., O'Canoll 4.M., McBride O.W., et al. Cloning and characterization of a vasopressin V2 receptor
and possible link to nephrogenic diabetes insipidus. Nature 357'. 336-339, 1992.
19. Pan Y., Metzenberg 4., Das S., Jing 8., and Gitschier J.
Mutations in the V2 vasopressin receptor gene are associated with X-linked nephrogenic diabetes insipidus.
N at. G ene t. 2(2) : 103 -106, 1992.
20. Van den Ouweland 4.M., Dreesen J.C., Verdijk M., and
Knoers N.V. Mutations in the vasopressin type 2 receptor gene (AVPR2) associated with nephrogenic diabetes
insipidus. Nat. Genet. 2: 99-102, 1992.
21. Rosenthal W., Seibold 4., Antaramian 4., et al. Molecul¿u
identification of the gene responsible for congenital nephrogenic diabetes insipidus. Nara re 359 233235, 1992.
22. Knoers N.A.V.M. and Monnens L.A. Nephrogenic dia-
ProceedingsoftheÄssociationofAmericanPhysiciansll0:5September/october1998
400
betes insipidus: Clinical symptoms, pathogenesis, genetics and treatment. Pedíat. Nephrol.6:476482, 1992'
23. Knoers N.A.V.M. and van Os C.H. Molecular and cel-
lular defects in nephrogenic diabetes insipidus' Carr'
Opit'r. Nephrol. Hypertens.5: 353-358, 1996'
24. Deen P.M.T. and Knoers N.V.A.M' Physiology and
pathophysiology of the aquaporin-2 water channel'
Curr. Opitt. Nephrol. Hypertens. 7 : 3'7 42' 1998'
25. Hoekstra J.4., van Lieburg 4.F., Monnens L'A'H', et al'
Cognitive and psychosocial functioning of patients with
congenital nephrogenic diabetes insipidus' Am' J' Med'
Genet. 6l'. 8 1-88, 1996.
26. Knepper M.4., Verbalis J'G', and Nielsen S' Role of
aquaporins in water balance disorders' Curr' Opin'
Nephrol. Hypertens. 6: 367 -3'7 l, 199"1 .
2'7 . Wahz'].,Hirai T., Mulata K., et al. The three-dimensional
strucntre of aquaporin- l. Nature 387:.624-627 ' 1997 '
28. Mulders S.M., Knoers N.V.A'M', Van Lieburg A'F'' et
al. New mutations in the AQP2 gene in nephrogenic diabetes insipidus resulting in functional but misrouted
water channels . J. Am' Soc. Nephrol. S:242-248, 199"1 '
29.Lande M.8., Jo I',Zeidel M.L., et al. Phosphorylation
of aquaporin-2 does not alter the membrane water permeability of rat papillary water channel containing vesicles. J.
Biol. Chem.27l: 5552-5557,1996'
30. Katsura T., Gustafson C.E., Ausiello D A', and Brown
D. Protein kinase A phosphorylation is involved in regulated exocytosis of aquaporin-2 in transfected LLCPK 1 cells. Am. J. Physiol. 272: F8l6-F822, 199'7 '
31. Bichet D.G., Arthus M.-F., Lonergan M', et al' Autosomal dominant and autosomal recessive nephrogenic diabetes insipidus: Novel mutations in the AQP2 gene [abstractl. An. Soc. Nephrol. 6:'ll7 ' 1995.
32. Canfield M.C., Tamarappoo 8.K., Moses A'M', Verkman
4.S., and Holtzman E.J. Identification and chatacterization of aquaporin-2 water channel mutations catrsing
nephrogenic diabetes insipidus with partial vasopressin response. Hum. Mol. Genet.6: 1865-1871' 1997 '
33. Oksche 4., Moller 4., Dickson J., et al. Two novel mutations in the aquaporin-2 and the vasopressin V2 receptor genes in patients with congenital nephrogenic diabe-
587-589' 1996'
Knoers N'V'A'M', et al'
M'4.,
Verdrjk
4.F.,
34. Van Lieburg
tes insipidus. Hum. Genet.98:
Patients with autosomal nephrogenic diabetes insipidus
homozygous for mutations in the aquaporin 2 waterchannel gene. Am. J' Hum. Genet.55'.648-652, 1994'
35. Hochberg Z.,Yanlieburg 4., Even L., et al' Autosomal
recessive nephrogenic diabetes insipidus caused by an
aquaporin-2 mutation. J. Clin. Ettdocrinol. Metab' 82l'
686-689,1997.
36. Deen P.M.T., Croes H., Van Aubel R.A', Ginsel L'A''
and van Os C.H. Water channels encoded by mutant
aquaporin-2 genes in nephrogenic diåbetes insipidus are
impaired in their cellular routing. J. Clin. Invest' 95"
229r-2296, 1995.
37. Halban P.A. and Irminger J'C. Sorting and processing
of secretory proteins. Biochem. J' 299: t-18,1994'
38. Brodsky J.L. and McCracken A.A. ER-associated and
proteasome-mediated protein degradation: How two topologically restricted events came together' Trends Cell
Biol.7: 15l-156,1997.
39. Hammond C. and Helenius A. Quality control in the secrctory pathway. Curr. Opin. Cell Biol.1: 523-529,1995'
40. Bai L.Q., Fushimi K., Sasaki S', and Marumo F' Structure of aquaporin-2 vasopressin water channel' 'l' Biol'
Chem. 27 l: 5 l'7 l-5 1'l 6, 1996.
41. Mulders S.M.A., Rijss J.P.L., Hartog A', et al' Importance of the mercury-sensitive cysteine on function and
routing of AQPI and AQP2 in oocytes. Am' J' Physiol'
273: F451-F456, 1997.
42. Da Silva 4., Braakman I., and Helenius A' Posttranslational folding of vesicular stomatitis virus G protein in
the ER: Involvement of noncovalent and covalent complexes. J. Celt Biol. 120 647-655, 1993.
43. Musil L.S. and Goodenough D.A' Multisubunit assembly of an integral plasma membrane channel protein,
gap junction connexin43, occurs after exit from the ER'
Cell 74: 1065-10'77, 1993.
44. Mulders S.M.A., Bichet D.G., Rijss J'P'L', et al' An
aquaporin-2 water channel mutant which causes autosomal dominant nephrogenic diabetes insipidus is retained
in the Golgi complex. J. Clitt. Invest. 102l.57-66' 1998'
45. Katsura T., Ausiello D.4., Brown D. Dilect demonstration of aquaporin-2 water channel recycling in stably
transfected LLC-PK, epithelial cells. Am. J' Physiol' 39:
F548-F553, 1996.
46. Valenti G., Frigeri 4., Ronco P'M., et al' Expression
and functional analysis of water channels in a stably
AQP2-transfected human collecting duct cell line' J'
Biol. Chem. 27 l: 24365-2437 0' 1996.
47. Deen P.M.T., Rijss J'P'L., Mulders S'M'A', et al' Aquaporin-2 transfection of MDCK cells reconstitutes vasopressin-regulated transcellular osmotic water transport'
J. Am. Soc. Nephrol.S: 1493-1501' 199'7 '
/