Kidney International, Vol. 52 (1997), pp. 1206—1215

HORMONES - CYTOKINES - SIGNALING

Immunolocalization of V1 vasopressin receptors in the rat kidney

using anti-receptor antibodiesCARLOS B. GONZALEZ, CARLOS D. FIGUEROA, CAiuos E. REYEs, CAIuos E. CA0RsI, SILVIA TRoNcoso,and DORIS MENZEL

Instituto de Fisiologia and Instituto de Histologla y Patologla, Facultad de Medicina, Universidad Austral de Chile, Valdivia, Chile

Immunolocalization of V1 vasopressin receptors in the rat kidney usinganti-receptor antibodies. By using immunocytochemical techniques wehave been able to localize the V1 vasopressin receptor in the rat kidney.Immunoblotting using an antiserum raised against an affinity-purifiedvasopressin receptor showed a 55,000 daltons protein band that has amolecular mass similar to that of the liver V1 vasopressin receptor, asdemonstrated by cross-linking studies. Immunoblotting of the antibodyshowed a band of 55,000 daltons in A-10 cells, which contains the V1subtype, whereas it did not stain LLC-PK1 cells, which possess the V2subtype, showing that the antibody recognizes the V1 vasopressin receptor.The immunostaining of kidney sections with this antiserum showed astrong reaction of the connecting tubules and cortical and medullarycollecting ducts. The immunostaining pattern of connecting tubule andcollecting duct cells was different, that is, the former showed a staining ofboth the apical and basal plasma membrane but also in the cytoplasm,whereas the latter showed a strong reaction mainly in the basolateralmembrane. Immunostaining of consecutive serial sections with an anti-serum raised against tissue kallikrein, an enzyme present exclusively inconnecting tubules, and with the anti-receptor serum allowed us to show,for the first time, the presence of the vasopressin receptor in theconnecting tubule cells and their absence in intercalated cells, the othercell type present in connecting tubules. These findings support experi-ments carried in the eighties on the release of renal tissue kallikrein byAVP.

Vasopressin (AVP) is synthesized in the supraoptic nucleus ofthe hypothalamus, packaged into secretory granules and releasedat the neural lobe into the bloodstream under adequate stimulusby a calcium dependent mechanism [1].

Vasopressin plays an important role in the fluid and electrolytebalance by regulating water and solute reabsorption in the kidney.This role is carried out by the binding of AVP to specificmembrane receptors. The vasopressin receptors have been clas-sified into two types that are functionally coupled to differenttransduction systems: the V1 receptor is coupled to phospholipaseC and therefore produces 1,2,4-inositol triphosphate and diacyl-glycerol [2]; the V2 receptor is associated with adenylate cyclase,which in turn increases the intracellular levels of cyclic AMP [3].The V1 receptor has been further subdivided into the Via for the

vascular and hepatic subtypes and the Vib for the adenohypophy-sial subtype according to their pharmacological properties [4].Recently, the eDNA and the predicted sequences for the rat Viaand V2 receptors have been reported [5, 6]. Both cDNAs encodeproteins with seven transmembrane domains as the other G-protein coupled receptor family.

The V1 receptor has been found in the liver, brain, platelets andvascular smooth muscle cells [7—10] whereas the V2 receptor hasbeen detected in the kidney and in the seminal vesicles [ii, 12].There is also pharmacological, biochemical and autoradiographi-cal evidence that the V1 receptor is present in the distal nephron[13—is]. The presence, in the kidney, of the Via receptor messen-ger RNA has been confirmed by in situ hybridization [16] and byreverse transcription and PCR in isolated rat nephrons [17, 18].However, there are still discrepancies about the exact location andproportion of the Via receptor in the different segments of thenephron.

Several approaches have been used to localize the vasopressinreceptors in the kidney. Autoradiographic techniques or fluores-cent microscopy using radio- or fluorescent-labeled vasopressin orits analogues have been employed to detect AVP binding sites [14,15, 19, 20], but this technique lacks of enough resolution. Immu-nochemical methods using anti-vasopressin or anti-idiotypic anti-bodies to localize the AVP receptors have also been used [21—23],however, these techniques have not provided precise informationon the specific tubular cell types that express the AVP receptor.On the other hand, a detailed immunocytochemical descriptionand localization of the vasopressin V1 and the V2 receptors in thekidney using anti-receptor antibodies has not been carried out.

We have identified the Via receptor from rat and pig liver as amonomeric protein of approximately 60,000 daltons [24]. Thisreceptor was later purified by affinity chromatography and poly-clonal antibodies were raised against it [25]. We have used theseanti-receptor antibodies in conjunction with autoradiographictechniques using a specific ligand to localize the Via vasopressinreceptor in the rat kidney.

Key words: vasopressin receptor, renal kallikrein, immunocytochemistry,receptors, connecting tubules, collecting ducts.

Received for publication March 21, 1996and in revised form June 12, 1997Accepted for publication June 16, 1997

© 1997 by the International Society of Nephrology

METHODS

Production and specificity of antibodies

The anti-receptor antibodies were raised in rabbits against afraction eluted from an affinity column. Rat liver membranes were

1206

Gonzalez et al: Vasopressin receptors in the rat kidney 1207

then solubilized with 20 mi CHAPS (3-[3-cholamidopropyl)-dimethylammonioj-1-propanesulfonate) and the extract passedthrough a lysine-vasopressin Sepharose column. The column wassuccessively washed with Tris-HCI buffer, pH 7.4, and eluted withglycine-HCI, pH 2.5, and 0.5% SDS. Reconstitution of fractionsand assay using a V1 labeled antagonist showed that most of thereceptor binding activity was present in fractions eluting with thebuffer, pH 2.5, and particularly in fractions eluting with 0.5% SDS.Polyclonal antibodies were raised against fractions eluting withpH 2.5 buffer and against fractions eluting with SDS. The anti-bodies were characterized by immunoblotting, immunoprecipita-tion and immunocytochemistry. In addition, these results werecompared with those obtained by cross linking experiments. Thisantibody was used to localize rat liver V1 vasopressin receptor.

Tissue kallikrein antiserum was used as a marker of connectingtubules, which is a nephron segment that is not clearly distin-guished on conventionally stained sections. Tissue kallikrein is aserine-protease that localizes exclusively in this late portion of themammalian nephron [261.

Membrane preparation

Rat liver membranes were obtained as it was previously de-scribed [24] and rat kidney membranes were prepared by densitygradient centrifugation based on the method described by Bock-aert eta! [111. Briefly, lOg of rat kidney were homogenized in 250ml 1 m sodium bicarbonate containing a mixture of the followinginhibitors: 0.2 m'vi PMSF, 1 tg/ml leupeptin, 1 g/ml pepstatin, 1p.g/ml aprotinin and 0.1 jig/mI antipain. The homogenate wasfiltered through a gauze and centrifuged at 1200 X g for 10minutes at 4°C. The pellet was resuspended in 69% wt/wt sucroseto give a final sucrose concentration of 44% wt/wt. Fourteenmilliliters of the mixture were placed in an ultracentrifuge tubeand 7 ml of 42.3% wt/wt sucrose solution were carefully added onthe top. Centrifugation was carried out at 80,000 x g for one hourat 4°C. Membranes were collected from the 42.3% sucrose-airinterface, washed with homogenization buffer and centrifuged at30,000 X g for 30 minutes. Finally, the membranes were resus-pended, by sonication, in 0.25 M sucrose containing the sameinhibitors included in the homogenization buffer. The membraneswere kept frozen at —20°C until use.

ImmunoprecipitationRat liver membranes were solubi!ized with 10 mvi CHAPS for

30 minutes at 4CC, then diluted to a final concentration of Cl-TAPSof 2 m and 50 jil of anti-vasopressin serum were added. Afterovernight incubation at 4°C 60 jd of a protein A-agarose suspen-sion were added and incubated for an additional two hours. Theprotein A-agarose was collected by centrifugation and resus-pended in sample buffer. The protein was separated by electro-phoresis and radioactivity detected by autoradiography.

Immunoblotting

Membrane proteins were separated by polyacrylamide gelelectrophoresis and electrotransferred onto nitrocellulose sheetsthat were incubated with the anti-vasopressin receptor serumdiluted 1:200. The nitrocellulose membranes were then washedand incubated with a peroxidase-labeled second antibody (Bio-Rad) diluted 1:3000. The peroxidase was developed using 4-Chloro-1-Naphthol and H202.

ImmunocytochemistryRat kidney slices 1 to 2 millimeters thick were fixed by

immersion in Bouin's fluid dehydrated in ethanol and embeddedin paraplast plus or methacrylate [27]. The latter embeddingmedium allowed us to perform the immunostaining for the AVPreceptor and for tissue kallikrein in two consecutive 1 jim thicksections. This meant that the same cells were immunostained withtwo different antibodies. Both embedding media were removedbefore immunostaining by immersion in xylene (2 X 15 mm).Paraplast sections (5 to 7 jim thick) or methacrylate (1 jim) werethen sequentially incubated with various dilutions of the anti-receptor serum (1:800) for 18 hours at room temperature, with asecondary antibody sheep anti-rabbit IgG. (1:50) for 30 minutes,washed with buffer and incubated with the PAP complex (1:100)for 30 minutes. The peroxidase was detected with 3,3-diamino-benzidine and H202. Between incubations, the sections wererigorously washed with buffer (3 x 5 mm). A titration study of theantibody was carried out with different dilutions of the antiserumin order to use the highest possible dilution to avoid the nonspe-cific reaction of the antibody. Control studies included omissionor replacement of the primary antiserum by a pre-immune serum.In addition, some sections were stained with the vasopressinreceptor anti-serum that was previously absorbed with purified ratliver membranes, which are known to have only V1 receptor.

lodination procedure

Lysine vasopressin (LVP) or the specific V1 antagonist Phaa,D-Tyr(Me)2,Arg6,Tyr9(NH2)AVP (kindly provided by Prof. M Man-ning) were radiolabeled with 125J using lodogen as previouslydescribed [28] or by the chloramine T method. The monoradio-iodinated peptide was separated from the free iodine, the non-labeled and the diradioiodinated peptides by HPLC on a C18column with a linear gradient of acetonitrile.

AutoradiographyRats were killed by decapitation and the kidneys immediately

frozen in liquid nitrogen and then sections (16 jim thick) wereobtained using a cryostat at —20°C. Sections were freeze-dried,hydrated with 50 m phosphate buffer, pH 7.4, and incubated with50 or 100 M of the iodinated V1 antagonist Phaa,D-Tyr(Me)2,Arg6,Tyr(NH2)°AVP for 24 hours at 4°C. After incubation wascompleted the sections were washed with cold phosphate bufferand cold distilled water. In some experiments the labeled antag-onist was incubated in the presence of non-labeled antagonist,AVP or desmopressin (dDAVP; a V2 specific agonist). Thesections were dried for two hours at 37°C and exposed to KodaxXAR-5 film.

RESULTS

Antiserum characterization

To determine the specificity of the antibody, Western blotswere carried out using purified plasma membranes from liver andkidney and cell lines expressing either V1 or V2 receptors or cells,which have no detectable vasopressin binding sites. Nitrocelluloseimmunoblots containing membrane proteins separated by electro-phoresis showed one main band of Mr 55,000 for kidney mem-branes and two faint bands running slightly faster and slower thanthe main band (Fig. IC). Replacement or omission of the primaryantiserum showed no immunoreaction with membrane proteins

116—

66—

45—

29—

A B

Fig. 2. Characterization of the antibody against the vasopressin recep-tor. Western blot analyses of rat liver membranes (A) COS 7 cell, A-b celland LLC-PK1 cell membranes (B). A band of approximately 55,000daltons can be observed in rat liver membranes and in A-10 cells both ofwhich contain Via receptor whereas no detectable band can be observedin neither COS 7 nor LLC-PK1 cells, the former has no vasopressinbinding sites and the latter endow V2 vasopressin receptors.

A B C

Fig. 1. Molecular characterization of the V vasopressin receptor. Cross-linking experiments (A) of rat liver membranes which were incubated with'251-LVP and EGS lane 1 and in lane 2 an excess (1 j.M) of non-labeledLVP was added to the reaction mixture. The complete displacement of theradioactive LVP from the 60,000 daltons band can be observed. Autora-diograph of a polyacrylamide SDS gel containing an immunoprecipitateusing vasopressin receptor antiserum of 1251-labeled vasopressin receptorafter solubilization with CHAPS (B). Western blot analysis of rat kidneymembranes using an antiserum raised against the rat vasopressin receptor(C). A main band of approximately 55,000 daltons and minor ones of45,000 and 70,000 daltons can be seen.

(data not shown). Immunoblots of rat liver membranes (Fig. 2A)showed one band with a similar molecular weight to that of kidneymembranes (Fig. IC). The immunostaining of nitrocellulosecontaining membrane protein from COS 7 cells (Fig. 2B, lane 1),which lack detectable vasopressin binding sites [5], A-b cells (Fig.2B, lane 2), which express the Via receptor [29], a cell line derivedfrom rat aorta smooth muscle and LLC-PK1 cells (Fig. 2B, lane 3),which contain the V2 receptor [30], showed that only the A-bcells are reactive with the antibody as expected. The molecularweight of the immunoreactive band shown by the A-b cells wassimilar to that shown by rat liver membranes (Fig. 2A). Themolecular mass for the vasopressin receptor shown by the im-mublot technique was similar to that obtained by autoradiographyof SDS gels containing rat liver membranes that were incubatedwith '251-labeled lysine-vasopressin and then cross-linked withEGS [ethylene glycol bis(succinimidyl succinate); Fig. 1A]. More-

over, the antiserum raised against the affinity purified receptorwas able to immunoprecipitate the '251-LVP labeled receptorfrom CHAPS solubilized membranes that had been cross-linkedwith 1251-LVP and EGS (Fig. 1B).

AutoradiographyThe iodinated V1 specific linear antagonist was used to deter-

mine the V1 binding sites in the rat kidney. A heavy labeling wasobserved on the cortex with a radial pattern and in the inner stripeof the outer medulla and in the inner medulla. The pelvic wall wasalso labeled, as well as the wall of arteries and arterioles (Fig. 3).This labeling was readily displaced by the addition of an excess ofAVP (Fig. 3 E, F) or of non-radioactive antagonist (Fig. 3 B, C).However, dDAVP, a potent V2 agonist, was unable to displace theiodinated V1 antagonist (Fig. 3 H, I).

ImmunocytochemistryOmission of the primary antiserum (not shown) or incubation

with an antiserum that had been absorbed with rat liver mem-branes (Fig. 4C) showed no staining at all. Immunostaining ofkidney sections with the vasopressin receptor antiserum showedthe staining of tubules in the cortex, and a strong staining of thetubules in the inner stripe of the outer medulla and to a lesserextent the outer part of the inner medulla (Fig. 4 A, B). In thecortex, apparently two types of tubules were immunostained witha different pattern. The first seemed to be connecting tubules withnarrow lumen and consisted of large cube-shaped cells that gavea strong staining throughout the cytoplasm (Fig. 4A). The secondtype had the typical morphology of collecting ducts and had awider lumen (Fig. 4 A, B). The principal cells of the cortical and

1208 Gonzalez et al: Vasopressin receptors in the rat kidney

116—

66—

45—

29—

Fig. 3. Autoradiograms of coronal rat kidney sections using the V1 antagonist labeled with 125! (4,D, G). Autoradiograms of displacement experimentsusing: a 50 flM concentration of AVP (B), V1 antagonist (E) and dDAVP (H) and 500 n concentration of AVP (C), Vi antagonist (F), and dDAVP(I). Magnifications approximately X5 to X6.

m

-ii

C')

a S

t

w

1'

"I

T

S

Gonzalez ci al: Vasopressin receptors in the rat kidney 1209

It.

--V

-

. 7O

9t 0

c•- -

'r- .•

/ 1i

'gç.

a-

—--

- -

I.

fr-'.

-aA

C) D

00

I :i1

::.

A

C S-I

-ThL

2

'-4-

-I.-

--, iw

.-._

-'t

1210 Gonzalez et a!: Vasopressin receptors in the rat kidney

Fig. 4. Immunocytochemical staining of rat kidney sections with anti-vasopressin receptor (A and B) and with a vasopressin antiserum absorbed withpurified liver membranes enriched in V1 receptor (C). In A micrograph of the cortex (C) and the outer stripe (OS) of the outer medulla (OM)(magnification X50). The staining of connecting tubules (CT) and collecting ducts (CD) in the cortex (C) and also the staining of collecting ducts inthe outer medulla is seen. Because of the delicate staining of the basolateral membrane, this can only be seen at higher magnification (inset). In B amicrograph of the inner stripe (IS) of the outer medulla (OM) and the inner medulla is shown. The staining of the thick ascending loop of Henle (TAL)and the collecting ducts (CD) can be observed. In C, a section is similar to A but immunostained with the vasopressin receptor antiserum that has beenabsorbed with liver membranes enriched in V receptor.

the outer medullary collecting ducts showed a very strikingstaining pattern, that is, a very strong reaction was localized in thebasal portion of the cells and a faint staining in the adjacentcytoplasm (Fig. 5). In the inner medulla, only the collecting ductswere stained in addition to vasa recta, but a drastic change in thestaining was observed, that is, the cells of the collecting ducts withthe strong basal immunoreaction rapidly disappeared as thecollecting duct went into the inner medulla. Because the shapeand localization of some of the immunoreactive tubules of thecortex were similar to that of connecting tubules, we usedantibodies raised against tissue kallikrein as a marker for thisnephron segment. In the kidney, this enzyme has been shown tobe exclusively present in connecting tubules [26]. The immuno-staining of consecutive serial sections (1 jim) with antibodies totissue kallikrein and to the vasopressin receptor showed that thesetubules were, in fact, connecting tubules (Fig. 6). Moreover, theantibody against vasopressin receptor stained the same cells as didthe tissue kallikrein antibody, that is, the connecting tubule cellsbut not the intercalated cells (Fig. 6 B, D).

Immunostaining of human kidney sections using the antiserumagainst rat VI vasopressin receptor showed a similar stainingpattern to that of the rat kidney, that is, the staining of connectingtubules and the collecting ducts (Fig. 7 A, B). Immunoreactivetubular cells of the human kidney showed a homogenous staining

throughout the cytoplasm and only some of them showed a strongstaining in the apical or basal membranes (Fig. 7).

DISCUSSION

The eDNA for the Via and the V2 receptor have been recentlyshown to encode proteins of a Mr of 44,200 plus two potentialsites for N-glycosylation for the V1, and a Mr of 40,500 plus onepotential site for N-glycosylation for the V2 [5, 6]. In addition,both proteins contain several potential sites for phosphorylationthat could also produce some heterogeneity on these proteins.The molecular mass of the band shown by the antiserum is in goodagreement with the molecular mass of the Via receptor deter-mined by chemical cross-linking [241, the molecular mass of thereceptor from the rat liver, and also with the molecular mass ofthe protein deduced from the cDNA sequence of the V1 receptor.The difference in the molecular weight given by the immunoblotand cross-linking experiments is due to the fact that in the lattertechnique the receptor is covalently bound to the hormone, andthus runs as a higher molecular weight protein.

The immunostaining of the band on the nitrocellulose contain-ing membrane proteins from A-b cells and the lack of reaction onthose containing membrane proteins from COS 7 and LLC-PK1,which expresses the V2 receptor, and the immunoprecipitation of

'-'• 04I' ..tt.-S.' gs I.ea —st .

1*1

4,

-S

-I'4

-

ti- ,t

S

CD

- ..4 4.

.4—. S. -.

a•1

-I-. •y.

Gonzalez et al: Vasopressin receptors in the rat kidney 1211

Fig. 5. Immunocytochemical staining of a kidney section with the anti-receptor antiserum. A strong immunoreaction of the principal cells of the outermedullary collecting ducts (CD) can be observed. Note that the reaction is located in the basal side of the cells. An immunoreaction of thick ascendinglimb (TAL) cells can be also observed specially in the basal infolding (arrows). Magnifications x700, inset X 1,400.

the 1251-LVP labeled receptor strongly support the idea that theantibody is recognizing the Via receptor subtype, and it is notreacting with the V2 subtype. This contention is supported by thefact that the antiserum precleared against liver enriched in V1receptor did not show staining of kidney sections.

The presence of immunoreactive V1 receptors and the bindingof '251-antagonist to distal nephron segments prompted thequestion as to whether the V1 receptor was present on distalconvoluted or connecting tubules. The use of antibodies directedto rat urinary kallikrein, an enzyme exclusively present in con-necting tubules, showed the labeling of the same structure byantibodies against kallikrein and against the vasopressin receptor.Therefore, these results clearly indicate that one of the corticaltubules stained by the anti-vasopressin receptor antiserum corre-sponds to connecting tubules.

The labeling pattern given by autoradiography, using the V1antagonist, was similar to the immunostaining pattern given byimmunocytochemistry except in the inner medulla where theantibody showed a label that decreases towards the papilla,whereas the autoradiography with the antagonist showed a strongbinding of the inner medulla. This discrepancy might be due to thereaction of the antagonist with other types of receptors, forexample, the Vib or the oxytocin receptors, which have beenshown to be present in the kidney [20, 311. In fact, this antagonistis able to bind to rat uterus membrane with a relatively highaffinity [32]. The pattern of immunostaining of the antibody is verysimilar to the distribution of the V1 binding sites determined by

autoradiography using a tritiated non-peptide antagonist, the SR49059 [33].

According to their shape and trajectory some of the tubules ofthe cortex labeled by the antagonist seem to be connecting tubulesand the other collecting ducts. The binding of the iodinated-antagonist was displaced by AVP and the non-labeled Via antag-onist but not by dDAVP, which is a specific V2 agonist devoid ofvasopressor or anti-vasopressor activities. These results suggestthat the receptor shown by autoradiographic studies is a V1subtype. We have previously shown that this anti-vasopressinreceptor antiserum stains the liver vasopressin receptor andreveals a particular distribution of this protein in the hepaticacinus [341.

The presence of V1 receptor in the kidney has been demon-strated by autoradiographic and pharmacological studies usingselective ligands. V1 binding sites detected by autoradiographyhave been located in collecting ducts of the outer medulla andalso in the cortical collecting ducts [15, 33]. In addition, labeling ofvascular elements such as vasa recta, arcuate, interlobular and therenal arteries, have also been shown. V1 binding sites in thecortical collecting ducts have also been detected on microdis-sected kidney tubules by pharmacological studies using specificligands [13]. In the present study, we showed the immunolabelingmainly of the cortical and outer medullary collecting ducts, butalso, to a minor extent, of the inner medullary collecting ducts.Our results are in good agreement with previous autoradiographicstudies [15, 331 with respect to the localization and the proportion

-'at C --. ... A. ti. . II. a t..s.. -. •.?jjp- - —

.—- '[• —L.' C•a \

'1%

• b

PT "tCv

1212 Gonzalez et al: Vasopressin receptors in the rat kidney

Fig. 6. Immunostaining of paraffin (A and C) and methacrylate (B and D) serial sections with antiserum against tissue kallikrein (A and B) and withanti.vasopressin receptor antiserum (C and D). The connecting tubules numbered from I to 7 stained by the anti-kallikrein antiserum of picture A arealso stained by the anti-vasopressin receptor antiserum (C). Magnification of A and C X500. In B, a thin serial methacrylate section stained with theanti-kallikrein antiserum showing the localization of the enzyme in the connecting tubule cells (CNTc), whereas the intercalated cells (Ic) appearunstained. The consecutive serial thin section which was stained with the anti-vasopressin receptor antiserum (D) shows the reaction on the sameconnecting tubule cells. Immunoreaction in the proximal tubule (PT) can be also seen in picture B probably representing a reabsorption process offiltered kallikrein. Note that the same proximal tubule appears negative when the anti-receptor antiserum was used. Magnification of B and D ,350.

of the label in the different regions. These result are also in goodagreement with those given by reverse transcription polymerascchain reaction (RT-PCR) where V1 transcripts were found in thecortical and outer medullary collecting ducts [17, 181.

We showed the presence of the V1 vasopressin receptor in theconnecting cells and its absence in the intercalated cells of theconnecting tubules. In the cortical and outer medullary collectingducts some cells were stained by the antibody whereas other cellsappeared to he negative. It is likely, as it happens in theconnecting tubules, that the vasopressin receptors in the collectingducts are located in the principal cells. This assumption is in goodagreement with the fact that a fluorescent analogue of AVP bindsto the principal cells in the collecting ducts [191.

The staining of cells of cortical tubules of human kidney is notsurprising since the rat and the human Via receptors have a highhomology (77%), whereas rat Via has a low homology (40%) withthe rat or human V2 receptor.

The staining pattern of collecting duct and connecting tubulecells was different. While cells from the former showed thestaining in the apical and mainly in the basolateral membranes,the latter showed the staining in relation to the apical andbasolateral membranes as well as throughout the cytoplasm. Thisstaining pattern of connecting tubules might reflect a receptor-mediated endocytosis process, a mechanism that has been associ-ated with the internalization of the ligand-receptor complex bycoated pits and the dissociation of the complex in early endo-somes [35]. In fact, it has been shown that both subtypes ofvasopressin receptors can be internalized by endocytosis of thehormone-receptor complex and receptors recycled to the cellsurface [36—381.

In the present work we found that the thick ascending limb wasstained by the antibody, suggesting the presence of V receptorsubtype in this part of the nephron. The presence of V1, receptorsubtype in this segment is still a controversial issue. Firsov and

-S

> S

4.

02

• ,9

ttsP

s ,i

-..,

'-.

•- .

.. A

Gonzalez et a!: Vasopressin receptors in the rat kidney 1213

Fig. 7. Immunostaining of human kidney sections with the vasopressinreceptor antiserum. Micrograph showing the immunostaining of tubulesin the cortex of the human kidney (A) a higher magnification of positivetubular cells (B).

coworkers [17] have found no detectable Via transcript in thissegment by RT-PCR, whereas Terada and colleagues [18] foundsmall amount of Via mRNA in this segment by the sametechnique. Further work will be necessary to substantiate thisissue.

It is generally accepted that, in the kidney, vasopressin produceswater and Na reabsorption and K secretion. Most evidenceindicates that the V2 receptor mediates changes in the urea andwater permeability, thus regulating the water reabsorption. How-ever, the role of the V1 receptor in the kidney remains unknown.The high expression of the V1 mRNA during the ontogeny of thekidney and the well-known mitogenic capacity of vasopressinthrough the V1 receptor led Ostrosky and co-workers [161 topropose a role for the V1 receptor during kidney development.However, the presence of the V1 receptor in the adult kidneyindicates that this might not be the only function of the V1receptor in that organ.

Previous studies have described an association between AVPand renal tissue kallikrein. In fact, experiments in which AVP wasinfused into dogs and rats showed a dose-dependent increase inthe urinary excretion of renal tissue kallikrein [39]. Tn addition,experiments performed using isolated renal cortical slices haveshown that AVP stimulates the release of renal tissue kallikrein[40]. It is interesting that no such effect was observed when thecortical slices were prepared from kidneys of spontaneously

hypertensive animals, suggesting a secondary defect to the devel-opment of hypertension [41]. The presence of V1 receptors inconnecting tubule cells, the site where tissue kallikrein is pro-duced, supports the action of AVP on the release of this enzymefrom connecting tubule cells, and it suggests that AVP actions inthe kidney may be closely associated to those driven by thekallikrein-kinin system. Kinins, the peptides released from kini-nogens by tissue kallikrein [42], have been shown to inhibit thehydrosmotic effect of antidiuretic hormone in perfused corticalcollecting tubules, but only when the kinin was added to thebasolateral side [43]. Like AVP V1 receptor, kinin B2 receptorsare also extremely abundant in the basal portion of collecting ductcells [44]. Interaction between both peptides and their corre-sponding receptors at this level may result in a balanced excretion!reabsorption of water and electrolytes by the kidney.

It has been shown that vasopressin increases the osmotic waterflow, an action mediated by V2 receptor subtypes, in the corticaland medullary collecting ducts [45, 46]. These observations are inagreement with the fact that vasopressin-sensitive water channelsare located in the cortical and medullary collecting ducts, espe-cially in the inner portion [47]. In the present study, we found thatthe V1 receptor subtype is located in the cortical and outermedullary collecting ducts and to a minor extent in the innermedullary collecting tubules. It is probably that both subtypes ofreceptors coexist in the same cell. In this context, it is interestingthat a modulation of the vasopressin-stimulated water reabsorp-tion has been proposed that is elicited through a V2 type receptor,by vasopressin itself through a V1 subtype receptor by theproduction of PGE2 [48].

Although a strong staining of the basolateral membrane ofcortical and outer medullary collecting tubules was observed, aweak staining of the apical membranes was also seen in bothsegments. However, this apical staining was more evident in theconnecting tubules and in the inner medullary collecting ducts. Asimilar staining pattern, that is, the apical and basolateral mem-branes, has been found in collecting ducts using vasopressinanti-idiotypic monoclonal antibodies [23]. These observations arein agreement with the fact that some of the effects of AVP can beelicited only from the basolateral side, whereas others only can beelicited when AVP is applied on the luminal side [49—51].Moreover, AVP is able to increase cytosolic Ca2, an effectmediated by the V1 receptor, when it is applied at both theluminal and the basolateral sides of rabbit cortical collecting ducts[52]. In this context, it is interesting to note that the concentrationof AVP in the urine is higher than that found in plasma [53].

By using immunocytochemistry with anti-receptor antibodiesand autoradiography with selective a V antagonist we have beenable to show the location of the V1 subtype receptor in differentnephron segments of the rat kidney. By using immunocytochem-istry of serial sections immunostained for the vasopressin receptorand for tissue kallikrein, we have been able to show, for the firsttime, the presence of V vasopressin receptors in connectingtubule cells (tissue kallikrcin-producing cells). These results willallow further studies on the role of the V1 receptor in renalfunction.

ACKNOWLEDGMENTS

This work was supported by grants from FONDECYT 1930368,1961171 and DIUACH S-93-45. We thank F. OyarzOn and C. Lizama fortheir technical assistance.

1214 Gonzalez et al: Vasopressin receptors in the rat kidney

Reprint requests to Dr. Carlos B. Gonzalez, Department of Physiology, P.O.Box 567, Universidad Austral de Chile, Valdivia, Chile.

REFERENCES

1. GONZALEZ CB, CA0RsI CE, FIGUEROA CD: Structure of neurosecre-tory granules and the chemistry of exocytosis. Ann NY Acad Sci689:59—73, 1993

2. MICHELL RH, KIRK Ci, BILLAH MM: Hormonal stimulation ofphosphatidyl-inositol breakdown, with special reference to the hepaticeffects of vasopressin. Biochem Soc Trans 7:861—865, 1979

3. Ro'e C, BARTH T, JARD S: Vasopressin-sensitive kidney adenylatecyclase. J Biol Chem 250:3157—3168, 1975

4. ANTONI FA: Novel ligand specificity of pituitary vasopressin receptorsin the rat. IVeuroendocrinology 39:186—188, 1984

5. LOLAIT Si, O'CARROL AM, MCBRIDE WW, KONING M, MOREL A,BROWSTEIN MJ: Cloning and characterization of a vasopressin V2receptor; chromosomal localization of gene suggests link to hereditarynephrogenic diabetes insipidus. Nature 357:336—339, 1992

6. MOREL A, O'CARROLL AM, BROWNSTEIN MJ, LOLAIT SJ: Molecularcloning and expression of rat V, arginine vasopressin receptor.Nature 356:523—526, 1992

7. CANTAU B, KEPPENS 5, DE WULF H, JARD 5: [3H]vasopressin bindingto isolated rat hepatocytes and liver membranes: Regulation by GTPand relation to glycogen phosphorylase activation. J Recept Res1:137—168, 1980

8. PENIT J, FAURE M, JARD S: Vasopressin and angiotensin II receptorsin rat aorta smooth muscle cells in culture. Am J Physiol 244:E72—E82, 1983

9. PERLMUTFER AF, CONSTANTINI MG, LOESER B: Characterization of3h-Avp binding sites in particulate preparations of rat brain. Peptides4:335—341, 1983

10. SlEss W, STIFEL M, BINDER H, WEBER PC: Activation of Vi receptorsby vasopressin stimulates inositol phospholipid hydrolisis and archi-donate metabolism in human plateles. Biochem J 233:83—91, 1986

11. BOCKAERT J, ROY C, RAJERISON R, JARD S: Specific binding of[3H]lysine-vasopressin to pig kidney plasma membranes. J Biol Chem248:5922—5931, 1973

12. MAGGI M, DE Rossi M, AMENTA F, GENAZZANI AD, RODBARD D,SERb DM: Similarity of vasopressin receptors in seminal vesicles andrenal medulla of pigs. J Reprod Fertil 84:40 1—407, 1988

13. AMMAR A, ROSEAU 5, BUTLEN D: Pharmacological characterization ofVia vasopressin receptors in the rat cortical collecting duct. Am JPhysiol 262:F546—F553, 1992

14. GERSTBERGER R, FAHRENHOLZ F: Autoradiographic localization ofV1 vasopressin binding sites in the brain and kidney. EurJPharmacol167:105—116, 1989

15. PHILLIPS PA, ABRAHAMS JM, KELLY J, MOOSER V, TRINDER D,JOHNSTON CI: Localization of vasopressin binding sites in rat tissuesusing specific V1 and V2 selective ligands. Endocrinology 126:1478—1484, 1990

16. O5TR0wSKI NL, YOUNG WS, KNEPPER MA, LOLAIT Si: Expression ofvasopressin V1, and V2 receptor messenger ribonucleic acid in theliver and kidney of embryonic, developing and adult rats. Endocrinol-ogy 133:1849—1859, 1993

17. FiRsov D, MANDON B, MOREL A, MEROT A-CJ, LE MAOUT J,BELLANGER A-C, ROUFFIGNAC C, ELALOUF J-M, BUHLER J-M: Mo-lecular analysis of vasopressin receptor in the rat nephron: Evidencefor the alternative splicing of the V2 receptor. Pfiugers Arch 429:79—89, 1994

18. TERADA Y, TOMITA K, NONOGUCHI H, YANG T, MARUMO F: Differentlocalization and regulation of two types of vasopressin receptormessenger RNA in microdissected rat nephron segments using re-verse transcription polymerase chain reaction. J Clin Invest 92:2339—2345, 1993

19. KIRK KL, BUKU A, EGGENA P: Cell specificity of vasopressin bindingin renal collecting duct: Computer-enhanced imaging of fluorescenthormone analogue. Proc NatlAcad Sci USA 84:6000—6004, 1987

20. TRIBOLLET E, BARBERIS C, DREIFUSS J-J, JARD S: Autoradiographiclocalization of vasopressin and oxytocin binding sites in rat kidney.Kidney Mt 33:959—965, 1988

21. FADOOL JM, AGGARWAL SK: Immunocytochemical demonstration ofvasopressin binding in rat kidney. J Histochem Cytochem 38:7—12, 1990

22. RAVID R, SWAAB DF, POOL CW: Immunocytochemical localization ofvasopressin-binding sites in the rat kidney. J Endocrinol 105:133—140,1985

23. JURZAK M, JANS DA, HAASE W, PETERS R, FAHRENHOLZ F: Gener-ation of anti-idiotypic monoclonal antibodies recognizing vasopressinreceptors in cultured cells and kidney sections. Exp Cell Res 203:182—191, 1992

24. ESTRADA EF, BARRA V, CAORSI CE, TRONCOSO 5, RuIz-OPAzo N,GONZALEZ CB: Identification of the V1 vasopressin receptor bychemical cross-linking and ligand affinity blotting. Biochemistry 30:8611—8616, 1991

25. GONZALEZ CB, CA0R5I CE, REYES CE, TRONCOSO S, BARRA V:Identification and purification of the vasopressin receptor from ratliver membranes. Ann NYAcad Sci 689:526—529, 1993

26. FIGUEROA CD, MACIVER AG, MACKENZIE JC, BHOou. KD: Local-ization of immunoreactive kininogen and tissue kallikrein in thehuman nephron. Histochemistiy 89:437—442, 1988

27. GONZALEZ CB, RODRIGUEZ EM: Ultrastructure and immunocyto-chemistry of neurons in the supraoptic and paraventricular nuclei ofthe lizard Lioleamus cyanogaster. Evidence for the intracisternallocation of the precursor of neurophysin. Cell Tiss Res 207:463—477,1980

28. GONZALEZ CB, REYE5 CE, CA0RsI CE, TRoNcoso 5: Modulation ofthe affinity of the vasopressin receptor by magnesium ions. BiochemInt 26:759—766, 1992

29. STASSEN FL, HECKMAN G, SCHMIDT D, AIYAR N, NAMBI P, CR0OKEST: Identification and characterization of vascular (V1) vasopressinreceptor of a established cell line. (abstract) Mol Pharmacol 259:266,1987

30. RoY C, AUSIELLO DA: Characterization of (8-lysine) vasopressinbinding sites on a pig kidney cell line (LLC-PK1). J Biol Chem258:3415—3422, 1981

31. SAITO M, SUGIMOTO T, TAHARA A, KAWASHIMA H: Molecular cloningand characterization of rat Vib vasopressin receptor: Evidence for itsexpression in extra-pituitary tissues. Biochem Biophys Res Comm212:751—757, 1995

32. SCHMIDT A, AUDIGUER S, BARBERIS C, JARD S, MANNING M, KOLOD-ZIEJCZYK AS, SWAYER WH: A radioiodinated linear vasopressinantagonist: A ligand with high affinity and specificity for Via recep-tors. FEBS Lett 282:77—81, 1991

33. SERRADEIL-LE GAL C, RAUFASTE D, MARTY E, GARCIA C, MARF-FRAND JP, LE FUR G: Autoradiographic localization of vasopressinVia receptors in rat kidney using [3H]-SR 49059. Kidney mt 50:499—505, 1996

34. NATHANSON MH, BURGSTALHER AD, MENNONE A, FALLON MB,GONZALEZ CB, SAEZ JC: Ca2 waves are organized among hepato-cytes in the intact organ. Am J Physiol 269:Gi67—G171, 1995

35. LUTZ W, SALISBURY JL, KUMAR R: Vasopressin receptor-mediatedendocytosis: Current view. Am J Physiol 261:F1—F13, 1991

36. KIM JK, SUMMER SN, SCHRIER RW: Arginine vasopressin receptorinternalization and recycling in rat renal collecting tubules. / ReceptRes 14:139—152, 1994

37. KIRK KL: Binding and internalization of a fluorescent vasopressinanalogue by collecting duct cells. Am J Physiol 255:C622—C632, 1988

38. LUTZ W, SANDERS M, SALISBURY J, KUMAR R: Internalization ofvasopressin analogues in kidney and smooth muscle cells: Evidencefor receptor-mediated endocytosis in cells with V2 or Vi receptors.ProcNatlAcadSci USA 87:6507—6511, 1990

39. FEJES-TOTH G, ZAHAJSZKY T, FILEP i: Effect of vasopressin on renalkallikrein excretion. Am / Physiol 239:F388—F392, 1980

40. LAUAR N, SHACKLADY M, BHOOLA KD: Factors influencing the invitro release of renal kallikrein. Agents Actions 9(Suppl):545—552, 1982

41. CHAPMAN ID, LEACH JT, BFJOOLA KD: Studies on renal and urinarykallikrein in spontaneously hypertensive rats. Adv Exp Med Bioli98B:219—224, 1986

42. BHOOLA KD, FIGUEROA CD, WORTHY K: Bioregulation of kinins:Kallikreins, kininogens, and kininases. Pharmacol Rev 44:1—80, 1992

43. SCHUSTER VL, KOKKO JP, JACOBSON HR: Interaction of lysil-brady-kinin and antidiuretic hormone in the rabbit cortical collecting tubule./ Clin Invest 73:1659—1667, 1984

Gonzalez et al: Vasopressin receptors in the rat kidney 1215

44. FIGUEROA CD, GONZALEZ CB, GRIG0RIEv 5, ABD ALLA S, HAASE-MANN M, JARNAGIN K, MULLER-ESTERL W: Probing for the bradyki-nm B2 receptor in the rat kidney by anti-peptide and anti-ligandantibodies. J Histochem Cytochem 43:137—148, 1995

45. ROCHA AS, KOKKO JP: Permeability of medullaiy nephron segmentsto urea and water: Effect of vasopressin. Kidney mt 6:379—387, 1974

46. ROCHA AS, KUDO LH: Water, urea, sodium, chloride and potassiumtransport in the in vitro isolated perfused papillary collecting duct.Kidney mt 22:473—484 1982

47. FUSHIMI K, UCHIDA 5, HARA Y, HIRATA Y, MARUMO F, SASAKI S:Cloning and expression of apical membrane water channel of ratkidney collecting tubule. Nature 361:549—552, 1993

48. SONNENBURG WK, SMnH WL: Regulation of cyclic AMP metabolismin rabbit cortical collecting tubule cells by prostaglandins. J Biol Chem263:6155—6160, 1988

49. ANDO Y, ASANO Y: Functional evidence for the apical V1 receptorin the rabbit cortical collecting duct. Am J Physiol 264:F467—F471,1993

50. ANDO Y, TABEI K, ASANO Y: Luminal vasopressin modulates trans-port in the rabbit collecting duct. J Clin Invest 88:952—959, 1991

51. SCHWARTZ IL, SHLATZ LI, KINNE-SAFFRAN E, KINNE R: Target cellpolarity and membrane phosphorylation in relation to the mechanismof action of antidiuretic hormone. Proc Nati Acad Sci USA 7 1:2595—2599, 1974

52. IKEDA M, Y0sHIT0MI K, IMAI M, KURORAWA K: Cell Ca2 responseto luminal vasopressin in cortical collecting tubule principal cells.Kidney mt 45:811—816, 1994

53. KIMURA T, SHARE L: Characterization of the renal handling ofvasopressin in the dog by stop-flow analysis. Endocrinology 109:2089—2094, 1981