Histopathological analysis of renal cystic epithelia in the

Pkd2WS25/- mouse model of ADPKD

R. Brent Thomson,1 SueAnn Mentone, 1 Robert Kim, 1 Karen Earle, 1 Eric Delpire,2 Stefan

Somlo, 1 and Peter S. Aronson1

1 Section of Nephrology, Department of Internal Medicine, Yale University School of

Medicine, PO Box 208029, New Haven, Connecticut 06520-8029

2 Department of Anesthesiology and Center for Molecular Neuroscience, Vanderbilt

University Medical Center, Nashville, Tennessee 37232

Running head: Histopathological analysis of renal cystic epithelia

Corresponding author: Peter S. Aronson

Section of Nephrology

Department of Internal Medicine

Yale University School of Medicine

1 Gilbert Street, CAB S-255C

P.O. Box 208029

New Haven, Connecticut 06520-8029

peter.aronson@yale.edu

phone: 203-785-4902 fax: 203-785-4904

Copyright (c) 2003 by the American Physiological Society.

Articles in PresS. Am J Physiol Renal Physiol (July 8, 2003). 10.1152/ajprenal.00153.2003

2

ABSTRACT

It has been proposed that ADPKD-affected renal epithelial cells undergo a phenotypic

transition from a highly differentiated absorptive state to a much less differentiated

secretory state during cystogenesis and that this transition is accompanied by loss of

epithelial cell polarity and mistargeting of specific membrane proteins. We have

conducted a detailed evaluation of this hypothesis in the Pkd2WS25/- mouse model of

ADPKD. Ultrastructural analysis of Pkd2WS25/- cysts by electron microscopy confirmed

that cystic epithelial cells progressively dedifferentiate with cyst enlargement.

Immunocytochemical analysis of both early and late stage cysts with antibodies directed

against the Na+/K+-ATPase, Ksp-cadherin, and E-cadherin failed to detect evidence of

altered cyst cell polarity. The Na+/K+-ATPase and Ksp-cadherin were expressed

exclusively on the basolateral membranes (BLM) of epithelial cells in all early cysts.

Expression levels of both the Na+/K+-ATPase and Ksp-cadherin decreased progressively

with the degree of cyst cell dedifferentiation, but neither protein was ever mislocalized.

Highly dedifferentiated cysts did not express immunodetectable levels of either the

Na+/K+-ATPase or Ksp-cadherin. E-cadherin was expressed prominently on the

basolateral membranes of all cysts. Cysts were subsequently stained with an antibody

directed against the secretory isoform of the Na+-K+-Cl- cotransporter NKCC1. NKCC1

expression was detected on the BLM of advanced cysts only. Our data are consistent

with a model of progressive cystic epithelial cell dedifferentiation in which fluid

accumulation in late stage cysts is mediated by transepithelial secretion of chloride rather

than secretion of sodium by apical Na+/K+-ATPase.

Keywords: polarity, Na+/K+-ATPase, NKCC1, Ksp-cadherin, E-cadherin

3

INTRODUCTION:

The physiological mechanisms involved in renal cyst formation in patients

affected by autosomal dominant polycystic kidney disease (ADPKD) are still largely

unknown. Intensive genetic analyses of families affected by ADPKD have revealed that

the majority of reported cases are directly attributable to mutations in the genes that

encode either polycystin-1 or polycystin-2. The functions of polycystin-1 and polycystin-

2 are not yet completely understood, but it is generally believed that their functions are

related and that a disruption in the function of either protein initiates a common

cystogenic pathway (for review see 18). In the currently accepted model of cystogenesis,

single progenitor cells are believed to enter a dedifferentiating hyperplastic pathway that

ultimately gives rise to the expansive fluid-filled cysts that are the hallmark of ADPKD.

Implicit in this model is the notion that the cystic epithelium undergoes a phenotypic

transition from a highly differentiated absorptive state to a much less differentiated

secretory state.

It has been suggested that alterations in epithelial cell polarity due to mistargeting

of specific membrane proteins may play a significant role in this transition. Charron et al

(8) reported that cultured human ADPKD cyst cells have a disrupted cytoarchitecture and

coupled with mislocalization of E-cadherin may have a generalized basolateral sorting

defect. Wilson et al (34; 35) proposed that the mistargeting of a fully functional Na+/K+-

ATPase to the apical membrane of cystic epithelial cells could result in net basal-to-

apical Na+ flux which in turn could drive luminal fluid accumulation and cyst expansion.

Although apical mislocalization of the Na+/K+-ATPase has been reported in several

different models of polycystic kidney disease (for example see 1 and 26), it has not been

universally accepted as the principal mechanism underlying either fluid secretion or cyst

expansion in ADPKD.

Grantham et al (13) have proposed that cyst luminal fluid accumulation may be

driven by net basal-to-apical Cl- movements in a manner similar to that observed in

classical secretory epithelia such as sweat duct or airway (see 30 for review) rather than

by apical secretion of Na+. In support of this hypothesis, Brill et al (6) reported that they

could find no evidence of mislocalization of the Na+/K+-ATPase in either excised human

ADPKD cysts or human ADPKD cyst cells in culture. Moreover, studies by Hanaoka et

4

al (14) and Brill et al (6) indicated that apical Cl- exit could be facilitated by the CFTR

protein, and inhibitor studies by Mangoo-Karim et al (23) suggested that the basolateral

entry step may be mediated by the Na+-K+-2Cl- cotransporter NKCC1.

To address this controversy and to determine if altered cell polarity plays a role in

cyst development in ADPKD, we have conducted an extensive ultrastructural and

immunocytochemical characterization of renal cysts in the Pkd2WS25/- adult mouse model

of ADPKD. Cyst initiation in this model is due to a somatic inactivation of the second

Pkd2 allele and cystogenesis has been shown to proceed in a manner similar to that

reported for human ADPKD (36; 37). Importantly, the use of an animal model permitted

us to perform rapid in situ perfusion fixation of the kidneys, eliminating the potential for

ischemia-induced mistargeting of the Na+/K+-ATPase (see 7 and 24) and affording

optimal preservation of both cellular ultrastructure and protein localization.

Ultrastructural analysis by both transmission and scanning electron microscopy

confirmed that cyst epithelial cells undergo a dramatic dedifferentiation as cystogenesis

proceeds. Immunocytochemical analysis of both early- and late-stage cysts indicate that

neither altered cell polarity nor mislocalization of the Na+/K+-ATPase appears to play a

role in cyst expansion and fluid secretion in ADPKD. Consistent with the model of

progressive cystic epithelial cell dedifferentiation in which fluid secretion is mediated by

transepithelial Cl- movements rather than apical secretion of Na+, we detected expression

of the Na+-K+-2Cl- cotransporter, NKCC1, on the basolateral membranes of advanced

cyst epithelial cells.

5

METHODS:

Animals and reagents. Pkd2WS25/- animals were generated by crossing Pkd2+/- and

Pkd2WS25/+ mice as described by Wu et al (36). To facilitate tissue preparation for

microscopy, all animals used in this study were one to two months of age. Kidneys from

animals of this age contained large numbers of cysts in all stages of development, but

retained enough non-cystic tissue to permit adequate perfusion fixation of both the cysts

and the unaffected renal parenchyma.

Two different mouse monoclonal antibodies directed against the |1 subunit of the

Na+/K+-ATPase were used in this study: antibody C464.6 (Upstate Biotechnology, Lake

Placid, NY) was used at a dilution of 1:50 and antibody Alpha-5 (Developmental Studies

Hybridoma Bank, Iowa City, IA; 22) was used at a dilution of 1:100. The anti-E-

cadherin mouse monoclonal antibody Clone 36 (Transduction Laboratories, Lexington,

KY) was used at a dilution of 1:100.

The anti-Ksp-cadherin antibody is an affinity purified rabbit polyclonal antibody

that we generated against the human isoform of Ksp-cadherin. Rabbits were immunized

with a maltose-binding protein (MBP; New England Biolabs, Beverly, MA) fusion

protein containing the C-terminal 267 amino acids (KspFP563-829) of the human

isoform of Ksp-cadherin. Sera was negatively purified against bacterial lysates

containing only the MBP and then positively purified on a column containing the MBP-

KspFP563-829 fusion protein. Specific reactivity was confirmed by Western analysis of

Cos-7 cells transiently transfected with full-length rabbit Ksp-cadherin. Specificity was

further verified by Western analysis of human, mouse, rabbit, and dog kidney

microsomes and immunocytochemical analysis of human and mouse kidney cryosections.

The purified serum was used for immunocytochemistry in the present study at a dilution

of 1:100.

The anti-NKCC1 antibody (19) is an affinity purified rabbit polyclonal antibody

directed against the mouse isoform of NKCC1 and was used at a dilution of 1:200. A

rabbit anti-mouse Tamm-Horsfall antibody (THP; a generous gift of Dr. John Hoyer) was

used at a dilution of 1:200 for identification of cysts of cortical thick ascending limb

origin (16). A rabbit anti-thiazide-sensitive Na-Cl cotransporter (NCC, TSC) antibody (a

generous gift of Dr. Steven Hebert) was used at a dilution of 1:200 for identification of

6

cysts of distal convoluted tubule origin (28). The fluorescein-conjugated lectin Lotus

tetragonolobus (LTA; Vector Laboratories, Burlingame, CA) was used at a dilution of

1:10 for identification of cysts of proximal tubule origin (20). Rhodamine-conjugated

Dolichos biflorus agglutinin (DBA; Vector Laboratories, Burlingame, CA) was used at a

dilution of 1:100 for identification of cysts of collecting duct origin (20). Alexa Fluor

anti-mouse IgG (488 and 594) and anti-rabbit IgG (488) secondary antibodies were used

at a dilution of 1:200 (Molecular Probes, Eugene, OR).

Electron microscopy. Mice were anesthetized by intraperitoneal injection of

sodium pentobarbital. The kidneys were cleared (PBS) and fixed (3%

paraformaldehyde/3% glutaraldehyde) by cardiac perfusion. The kidneys were then

removed, cut into 2-4 mm thick sagittal sections and post-fixed in the same fixative for 1

h at 4°C. Tissue destined for transmission electron microscopy (TEM) was further cut

into 2-4 mm blocks for subsequent processing. Cortical tissue only was used for the

TEM study. The tissue sections were washed in 0.1 M cocodylate buffer containing

7.5% sucrose (pH 7.4; 3 changes; 1 h; 4°C) and then incubated in veronal acetate-

buffered 1% osmium tetroxide for 2 h at 4°C (1 h for TEM sections). Tissue used for

scanning electron microscopy (SEM) was again washed in cocodylate buffer (3 changes;

1 h; 4°C), dehydrated through a graded ethanol series, critical point dried, mounted on

stubs, sputter coated with PdAu, and then visualized on an International Scientific

Instruments SS-40 scanning electron microscope. Tissue used for TEM was washed and

incubated in a Kellenberger veronal-acetate buffer containing 0.5% uranyl acetate for 2 h

at room temperature, rinsed in H2O, dehydrated through a graded ethanol series, and

embedded in Epon 812. Ultrathin 80 nm sections were cut on a Reichert Ultracut E

ultramicrotome, stained with uranyl acetate and lead citrate, and examined with a Zeiss

EM910 electron microscope.

Immunocytochemistry. Kidneys from six non-sibling mice were perfused and

cleared in the same manner as above, except that PLP (2% paraformaldehyde, 750 mM

lysine, and 10 mM sodium periodate in a phosphate buffer, pH 7.4) was used as the

fixative. The cortices were separated from the kidneys, cut into 2-4 mm blocks and

postfixed in the same fixative for an additional 4 h. To maintain the structural integrity

of the fragile cyst walls, all tissue blocks were embedded in Epon 812 as above except

that the tissue blocks were not exposed to osmium tetroxide or uranyl acetate. One

7

micron sections were cut on the Reichert Ultracut E ultramicrotome and mounted on

Superfrost Plus glass slides (Electron Microscopy Sciences, Fort Washington, PA).

Sections were etched for 5 minutes in a solution containing 2 g KOH, 10 ml 100%

methanol, and 5 ml propylene oxide, washed for 5 min in 100% methanol (2 changes),

rinsed in TBS (50 mM Tris-HCl, 100 mM NaCl, pH 7.4), and then subjected to antigen

retrieval as outlined in Biemesderfer et al (5). Briefly, sections were immersed in a

boiling 10 mM sodium citrate buffer (pH 6) and then heated for an additional 20 minutes

at the 40% power setting of a 700 watt microwave oven, allowed to cool, washed in TBS,

quenched in 0.5 M ammonium chloride with 0.1% BSA for 15 min, rinsed in TBS,

incubated in 1% SDS in TBS for 5 min, and finally washed in TBS. Sections to be

stained with mouse primary antibodies were incubated with unconjugated anti-mouse IgG

diluted 1:5 in TBS/0.1% BSA/ 10% goat serum for 1 h at room temperature. All other

sections were blocked with 0.1% BSA and 10% goat serum in TBS for 1 h at room

temperature. All sections were washed with TBS and then incubated with the respective

primary antibodies diluted in TBS/ 0.1% BSA/10% goat serum overnight at 4°C. The

next morning sections were washed in high salt TBS (527 mM NaCl) with 0.1% BSA

(5X; 5 minutes each wash) and then incubated with the appropriate fluorochrome-labeled

secondary antibody diluted in TBS/0.1% BSA/10% goat serum for 1 h at room

temperature. Sections were washed for 5 minutes in high salt TBS, 20 minutes in

standard TBS (4 changes), mounted in VectaShield (Vector Laboratories, Burlingame,

CA), and then visualized on a Zeiss Axiophot phase-contrast microscope. Antibodies and

lectins used for double labeling experiments were tested in control pilot studies to verify

lack of cross reactivity between reagents. Sequential serial sections were labeled with

specific cortical tubule segment markers to determine cyst tubule segments of origin.

Cyst staging. The criteria used to developmentally stage cysts were based on

cyst size, preliminary TEM ultrastructural surveys of PKD2WS25/- ADPKD cysts, and

recommendations by Devuyst et al (10). Generally, cysts were considered early-stage if

on cross-section there were less than 50 cyst-lining epithelial cells, intermediate-stage if

there were 51-200 cyst-lining epithelial cells, and advanced-stage if there were greater

than 200 cyst-lining epithelial cells. The relative position of a section within a cyst was

determined by examination of adjacent serial sections spanning approximately 40 µm in

each direction.

8

RESULTS:

Ultrastructural analysis of PKD2WS25/- ADPKD cysts. In small cysts that are

presumably in the early stages of cyst expansion, the ultrastructure of the cyst cells is

highly reminiscent of unaffected tubular epithelial cells (Fig. 1a). The cells are cuboidal

to columnar, have approximately normal numbers of mitochondria, have well developed

lateral junctions, well organized basement membranes, and prominent cilia. None of the

early cyst cells that we examined retained sufficient “normal” ultrastructural details to

unequivocally assess the tubule segment of origin. As cyst expansion proceeds, the cells

take on a more squamoid appearance (Fig. 1b) and lose all resemblance to unaffected

tubular epithelial cells. The mitochondria become much less prominent, but the cells

continue to retain well-defined lateral junctions and a well-organized basement

membrane. Significant deposits of collagen begin to appear in the region just below the

basement membrane. In very large cysts that presumably represent the final stages of

cyst expansion, the cyst cells are highly attenuated and acquire a pronounced endothelial-

like appearance (Fig. 1c). Mitochondria are extremely rare, but again the cells continue

to retain well-defined lateral junctions (Fig. 1c inset). The basement membrane loses

much of its organization and becomes intimately associated with large deposits of striated

collagen. We did not observe thickening or lamination of the basement membrane per se

at any stage of cyst progression.

Scanning electron microscopy of the lumens of similarly staged ADPKD cysts

provide further support for the cellular dedifferentiation hypothesis. In the initial stages

of cyst growth the cyst epithelial cells appear to be highly proliferative and form

elaborate chords, ridges, and labyrinths within the cyst lumens (Fig. 2a). Early cyst cells

maintain regular well-defined lateral borders and have prominent cilia (Fig. 2b). The

degree of secondary cell-surface ultrastructure (eg. microvilli) varies from cyst to cyst

and presumably reflects the cyst tubule segment of origin. As cyst size increases the cell

borders become less regular and the cilia less prominent (Fig. 2c). In many late-

intermediate and early-advanced stage cysts we observed more than one cell phenotype

(see for example Fig. 2C arrow). These atypical cyst cells were not present in large

numbers (1-5% of the total cyst cell population) and appeared to be randomly

interspersed throughout the cyst walls. It is unclear if these cells are differentially staged

9

progeny of the original cyst progenitor cell or if they are derived from outlying cell

populations. Cellular heterogeneity was never observed in extremely advanced cysts (Fig.

2d). The cells in advanced cysts appeared to be much larger than those seen in early

cysts, and they have irregular cell borders and inconspicuous cilia. The cyst walls are

very uniform and show little evidence of the cellular proliferation that was evident in the

earlier stages of cyst growth.

Immunocytochemical analysis of PKD2WS25/- ADPKD cysts. To determine the

localization of the Na+/K+-ATPase in PKD2WS25/- cysts and to facilitate comparison with

previously published reports, both Alpha-5 and C464.6 anti-Na+/K+-ATPase antibodies

were used in this study. Using antibody Alpha-5, Wilson et al (35) reported that the |-1

subunit of the Na+/K+-ATPase was localized to the apical membrane of ADPKD cyst

cells in immersion-fixed human nephrectomy samples and human ADPKD cyst cells in

culture. Using antibody 6H (clone C464.6), Brill et al (6) reported that the |-1 subunit of

the Na+/K+-ATPase was localized to the basolateral membrane of immersion-fixed

excised human ADPKD cysts and human ADPKD cyst cells in culture. Under the

conditions used in our study, the staining patterns for the two antibodies were identical

(Fig. 3). They each labeled the basolateral membranes of both normal and cystic

epithelia. Sixty cysts from six mice were positively labeled with the Na+/K+-ATPase

antibodies (see Table 1) and in no instance was apical localization ever detected.

In addition to assessing cyst Na+/K+-ATPase distribution, we also performed

double-labeling experiments with an antibody directed against Ksp-cadherin. Ksp-

cadherin is a kidney-specific member of the cadherin superfamily of cell adhesion

molecules and is expressed exclusively on the basolateral membrane of all tubular

segments of the mammalian kidney (33). If cystic epithelial cells truly have a

generalized basolateral sorting defect one would predict that other basolateral proteins,

such as Ksp-cadherin, would also be missorted. This proved not to be the case. As with

the Na+/K+-ATPase, apical localization of Ksp-cadherin in ADPKD cysts was never

detected (see Fig. 4 and Table 1).

Given our inability to detect evidence of a generalized basolateral sorting defect

or of altered cyst cell polarity, we suspected that E-cadherin may actually not be

mislocalized in native ADPKD cyst cells. Using the anti-E-cadherin antibody Clone 36,

Charron et al (8) reported that E-cadherin was not expressed on the lateral membranes of

10

ADPKD cells in culture, but rather that it was sequestered in perinuclear vesicular

structures. Using the same antibody in the PKD2WS25/- ADPKD native cysts, we found no

evidence of cystic epithelial intracellular sequestration of E-cadherin. Rather, we

observed prominent E-cadherin staining on the lateral membranes of all cystic epithelial

cells (see Fig. 5 and Table 1).

While neither the Na+/K+-ATPase nor Ksp-cadherin was ever detected on the

apical membrane of PKD2WS25/- native cysts, we did observe a progressive decrease in the

level of expression of both proteins that appeared to be concomitant with the extent of

cyst expansion and the degree of cyst cell dedifferentiation from a cuboidal to a squamoid

phenotype. Na+/K+-ATPase immunostaining was never observed in cysts with either a

highly attenuated squamoid epithelium or extensive surrounding fibrotic tissue (Fig. 6a).

Ksp-cadherin immunostaining was also typically absent in advanced cysts, but was

occasionally detected in squamoid epithelial cysts that were not surrounded by extensive

fibrotic tissue. In Figure 6b, for example, both cysts have a highly attenuated squamoid

epithelium, but only the upper cyst was totally surrounded by fibrotic tissue. In the region

below the lower cyst (not included in the field shown in Fig. 6b), normal appearing

tubules were in direct apposition with the cyst wall and there was no evidence of

surrounding fibrotic tissue. We interpret this to indicate that the lower cyst is less

advanced than the upper cyst. The decrease in Na+/K+-ATPase and Ksp-cadherin

expression levels could not be correlated with the cyst tubule segment of origin because

advanced cysts also tended to lose expression of the tubule origin markers used in this

study (see Table 1). In the field shown in Figure 6a and 6b, the lower cyst was THP

positive whereas the upper cyst was negative for all tubule segment markers. This

observation is consistent with the notion that the upper cyst is more highly advanced and

dedifferentiated than the lower cyst. E-cadherin immunostaining, on the other hand, was

detected at high levels on the lateral membranes of all cysts regardless of the degree of

cyst expansion or dedifferentiation. In Figure 6c, for example, the anti-E-cadherin

antibody is shown labeling cyst epithelial cells that were negative for the Na+/K+-ATPase,

Ksp-cadherin, and all tubule segment markers.

If the Na+/K+-ATPase is not mislocalized to the apical membrane of ADPKD

cysts, primary active luminal Na+ extrusion is unlikely to account for cyst fluid secretion.

Sullivan et al (32) suggest that ADPKD cyst fluid accumulation is not driven by apical

11

sodium transport, but rather by net basal to apical chloride flux. They propose that the

cystic epithelium undergoes a transition to a secretory phenotype similar to that observed

in airway epithelia and that the transepithelial chloride flux is mediated by the basolateral

bumetanide-sensitive Na+-K+-2Cl- cotransporter NKCC1 and the apical chloride channel

CFTR. Hanaoka et al (14) and Brill et al (6) have provided compelling evidence for

CFTR involvement in the apical exit step, but to date there has not been a clear

demonstration of NKCC1 localization on the basolateral membranes of native ADPKD

cyst cells.

In normal kidneys, NKCC1 expression is confined to the basolateral membrane of

the inner medullary collecting duct (IMCD), the glomerular and extraglomerular

mesangium, the glomerular afferent arterioles, and the papillary epithelium (19).

Consistent with these findings, NKCC1 expression in the non-cystic regions of the

cortices of PKD2WS25/- ADPKD cystic kidneys was restricted to structures associated with

glomeruli (Figure 7a). The vast majority of the cortical cysts that we observed did not

express immunodetectable levels of NKCC1 (see Table 1). In particular, anti-NKCC1

antibody labeling was never observed in cyst cells that expressed immunodetectable

levels of the Na+/K+-ATPase. For example, the tissue section illustrated in Figure 7a

depicting NKCC1-negative cysts is a serial section subsequent to that labeled with the

anti- Na+/K+-ATPase antibody shown in Figure 3a. Moderate levels of NKCC1

expression were detected on the basolateral membrane of squamoid-cell cysts whose

epithelia had typically dedifferentiated to the point of no longer expressing detectable

levels of the tubule segment-specific markers (see Figure 7b-c and Table 1). NKCC1

expression was occasionally detected in squamoid-cell cysts that were positive for

segment-specific marker labeling, but in the two observed cases the segment-specific

marker labeling was weak and inconsistent. Co-expression of NKCC1 with Ksp-cadherin

was rare and occurred in less than 10% of the cysts that expressed NKCC1.

12

DISCUSSION:

The ultrastructural morphology of the mouse PKD2WS25/- ADPKD cysts is virtually

indistinguishable from that of human ADPKD cysts (see for example 9, 12, 11, and 7).

As reported for human ADPKD cystic kidneys, we observed two predominant cyst types:

those with highly differentiated cuboidal or columnar epithelial cells and those with

attenuated dedifferentiated squamoid epithelial cells. Our data suggest that the two cyst

cell populations are related and that they represent different stages of cyst development.

Small early cysts were predominantly lined by cuboidal or columnar epithelial cells, and

large advanced cysts were predominantly lined by dedifferentiated squamoid epithelial

cells. Intermediate cysts had less clearly demarcated cell types and frequently appeared

to have both cuboidal and squamoid cells in the same cyst. Carone et al (7) observed a

similar cellular phenotypic heterogeneity in a subpopulation of human ADPKD cysts.

The cyst cell phenotype appears to directly reflect the degree of cyst progression. In the

early stages of cystogenesis the cyst cells are well-differentiated and resemble normal

tubular epithelial cells. The tissue surrounding the cyst at this stage shows little or no

evidence of fibrosis or abnormal collagen deposition. As cyst expansion proceeds the

cyst cells gradually dedifferentiate to the attenuated squamoid phenotype seen in

advanced macro-cysts. The surrounding tissue becomes extremely fibrotic and large

deposits of collagen become intimately associated with the highly disorganized cyst

basement membrane.

Early studies of human ADPKD cystic kidneys also identified two cyst types, but

it was largely believed that the two cyst types were unrelated and that the phenotypic

differences reflected the respective cyst tubule segments of origin (for review see 32).

Cysts lined by dedifferentiated squamoid epithelial cells were reported to contain a fluid

with a high sodium content that was similar to that observed in serum and proximal

nephron segments. These cysts were therefore believed to be derived from proximal

tubules and were referred to as “proximal” or “non-gradient” cysts. Cysts lined with

cuboidal or columnar cells had sodium concentrations that were low relative to serum

levels and very similar to that observed in distal nephron segments. These cysts were

considered to be derived from the distal nephron and were referred to as “distal” or

“gradient” cysts. The cyst tubule segment of origin was determined solely on the basis of

13

luminal sodium concentrations and an ability to maintain a transepithelial sodium

gradient. The identification of the tubule segment of origin was rarely corroborated by

cyst cell ultrastructure since the vast majority of the sampled cysts had typically

dedifferentiated to the point of no longer expressing recognizable segment-specific

characteristics (see for example 12). The ultrastructure of the “distal” cyst cells was

reminiscent of cells observed in distal segments of unaffected nephrons, but the

“proximal” cyst cells had no resemblance to normal proximal tubule cells. Statistical

analyses of human ADPKD cyst fluid compositions suggested that roughly 67% of the

sampled cysts were of proximal origin and that the remaining 33% were of distal origin

(17; 11). The ratio of “proximal” cysts to “distal” cysts closely reflected the expected

ratio of proximal nephron mass to distal nephron mass and was therefore taken as

additional supporting evidence that the two cyst types reflected the nephron segment of

origin. This conclusion implies that cystogenesis is a totally random event without a

clear segment-specific predilection for the site of cyst formation.

Our data (see Table 1) and that of Wu et al (36) argue strongly against a simple

stochastic model for the site of cyst formation in ADPKD-affected kidneys. Seventy to

eighty percent of the cysts that were observed in the PKD2WS25/- mouse renal cortex could

be unequivocally identified as being of either distal tubule or collecting duct origin. It is

possible that this correlation is specific to the PKD2WS25/- model, but recent

immunocytochemical studies on human ADPKD cystic kidneys suggest that this is not

the case. In two studies examining immunolocalization of aquaporin-1 (10) and of gp330

(2) only 30-44% of over one thousand cysts were considered to be of proximal origin.

Similar studies of aquaporin-2 immunolocalization (10;15) suggested that 30-33% of the

cysts were of collecting duct origin. Presumably the bulk of the remaining 33% of the

cysts are derived from other distal segments arguing that 66% and not 33% of the cysts

are of distal origin, and 33% and not 67% of the cysts are of proximal origin. If this is

true, the cyst fluid sodium concentration and hence the cyst cell ultrastructure would be

unlikely to correlate with the cyst segment of origin in the studies of human ADPKD

kidneys described above.

Like Grantham et al (12), we observed cellular phenotypic heterogeneity in a

significant subpopulation of ADPKD cysts. The source of the secondary cyst cell types

is unclear. The data presented by Grantham et al (12) suggest that the cysts in question

14

in their study had a collecting tubule origin and that the secondary cells were most likely

intercalated cells. The same may be true to some extent in our study, but the

ultrastructural evidence for a collecting tubule cyst origin is less compelling. Regardless

of the cyst tubule segment of origin, it is difficult to reconcile the cyst cellular

heterogeneity with the hypothesis of Qian et al’ (29) for cyst cell monoclonality. Using a

PCR-based cell clonality assay with human ADPKD cysts, Qian et al demonstrated that

at least 82% of the cysts (and quite probably more) were derived from single cyst

progenitor cells. If only collecting tubule-derived cysts showed cellular heterogeneity

one could postulate that both the intercalated- and principal-like cells were derived from

a single cyst progenitor cell. The exact origin of intercalated cells in vivo is unclear, but

it appears that both they and principal cells may be derived from a common ureteric bud

cell lineage during nephrogenesis (31). Perhaps ADPKD-induced cellular

dedifferentiation places collecting tubule-derived cyst progenitor cells into a limited

pleuripotent state that allows the subsequent transition to either an intercalated- or

principal-like cyst cell. A greater difficulty comes with cysts that do not appear to be of

collecting duct origin. In these instances one could postulate a polyclonal cyst origin or

the recruitment of cyst cells from an outlying cell population. The strongest argument

against either of those possibilities in our study is that the secondary cells were always

isolated and randomly dispersed throughout the cyst wall. If there truly was an

independently-derived mitotically-active subpopulation of cells within a cyst one would

expect to see clustering of phenotypically-identical cells. Moreover, the cellular

phenotypic heterogeneity was not preserved in all cyst stages. Heterogeneity was only

observed in intermediate or early advanced cysts coincident with the general timing of

the transition from a cuboidal to a squamoid cyst cell phenotype. This suggests that the

apparent “atypical” cells may actually be normal cyst cells in a slightly delayed state of

dedifferentiation and that they would eventually be phenotypically indistinguishable from

their neighbors upon completion of the transition to the squamoid dedifferentiated

phenotype.

The role of the primary cilium in the etiology of polycystic kidney disease is

currently receiving considerable attention. Studies in mouse models of PKD suggest that

defects in primary cilia may directly induce formation of renal cysts (for review see 18).

In orpk mice, for example, a mutation in the polaris gene has been shown to lead to

15

severe stunting of the primary cilia in the collecting tubule and concomitant cyst

formation (27). In the PKD2WS25/- model of ADPKD all cells in both early and

intermediate stage cysts, regardless of the cyst tubule segment of origin, have well-

developed normal appearing cilia. Cells in advanced cysts also have primary cilia, but

they appear to be much less prominent. Grantham et al (12) reported similar findings in

human ADPKD renal cysts. These observations suggest that if a ciliary defect is

involved in cyst formation in ADPKD, it likely does not have an ultrastructural

component. This is not wholly unexpected given that mutations in the C. elegans

orthologs of the PKD1 and PKD2 genes induced functional disruptions of the sensory

cilia without producing grossly obvious ciliary ultrastructural defects (3; 4). Nauli et al

(25) have recently proposed that polycystin-1 and polycystin-2 form heterodimeric

complexes on the surface of the primary cilium and that these complexes act as

mechanotransductive elements to convey information on luminal fluid flow and lumen

diameter to the morphoregulatory centers of renal tubular epithelial cells. They suggest

that a functional disruption of this complex could lead to cellular dedifferentiation and

tissue remodeling, and that cyst formation due to mutations in PKD1 or PKD2 is likely

due to a dysfunction in this signaling pathway rather than an overt alteration of ciliary

ultrastructure.

The only ultrastructural detail that appears to be significantly different between

PKD2WS25/- ADPKD cysts and human ADPKD cysts is the degree of basement membrane

perturbation. Cuppage et al (9) reported that there were abnormalities in human

ADPKD cyst basement membranes, but that there was not a consistent pattern in the

defect. Carone et al (7), however, reported that all of the human ADPKD cysts observed

in their study had thickened reticulated basement membranes. We did not observe either

thickening or reticulation of cyst basement membranes at any stage of cyst growth. The

basement membranes of all early and intermediate cysts were not noticeably different

from those of unaffected tubule segments. It was not until the cyst epithelial cells were

highly dedifferentiated that disorganization of the cyst basement membrane became

apparent. Consistent with the human ADPKD studies, we observed a significant increase

in interstitial fibrosis and collagen deposition in regions in direct apposition with

advanced cysts. The factors responsible for the fibrosis and the collagen deposition in

both the PKD2WS25/- and human ADPKD cystic kidneys remain uncertain.

16

It is clear that cyst cells undergo a dramatic dedifferentiation as cyst expansion

proceeds. However, from an ultrastructural perspective we could detect no evidence to

suggest that the cystic epithelium ever dedifferentiates to the point of losing cellular

polarity. Well-defined lateral junctions were preserved in even the most advanced cysts.

Consistent with this observation, the Na+/K+-ATPase, Ksp-cadherin, and E-cadherin were

all correctly targeted in cysts derived from every segment of the tubular nephron and the

collecting system. Moreover, we detected significant levels of anti-E-cadherin

immunostaining on the lateral membranes of all cyst cells regardless of the stage of cyst

expansion or cyst cell dedifferentiation. It is possible that some proteins may truly be

mislocalized in ADPKD cystic epithelia, but our observations in the PKD2WS25/- mice

suggest that this would be an uncommon event and that it would likely be due to a

perturbation in a processing pathway that is specific for that particular protein and not

due to a generalized basolateral sorting defect.

Our immunolocalization data do not support the model of apical Na+/K+-ATPase -

mediated, sodium-driven cyst fluid secretion. In advanced cysts at least, the data are

consistent with the model of chloride-driven cyst fluid accumulation proposed by

Sullivan et al (32) in which chloride uptake is mediated by the basolateral Na+/K+/2Cl-

cotransporter NKCC1. In a preliminary study of NKCC1 expression in human ADPKD

cystic kidneys, Lebeau et al (21) report that approximately 36% of all cysts examined

expressed immunodetectable levels of NKCC1 on their basolateral membranes. This

number compares favorably with our observations in PKD2WS25/- cystic mouse kidneys

(see Table 1), but it is unclear if they observed a similar correlation of NKCC1

expression with cyst expansion and dedifferentiation. Because we specifically focused

on cortical cysts in our study, the NKCC1 expression that we observed must have been a

direct result of an ADPKD-induced phenotypic transformation rather than the

consequence of cystic expansion of the inner medullary nephron segments that normally

express NKCC1. If advanced ADPKD cyst cells are using basolateral NKCC1 to drive

luminal fluid secretion in a manner similar to that proposed for airway epithelial cells,

one would presume that Na+/K+-ATPase activity would also be required on the

basolateral membrane of the same fluid secreting cells. Grantham et al (13) clearly

demonstrated that basolateral ouabain inhibited chloride-dependent cyst fluid secretion in

excised human ADPKD cysts. In our study the Na+/K+-ATPase was never detected in the

17

advanced cyst cells that expressed NKCC1. We suspect that the Na+/K+-ATPase may

have been present on the basolateral membrane, but that it was expressed at such low

levels that it was below the level of detection with the antibodies that were used in our

study. Given the extraordinarily low rate of fluid secretion that must occur in cysts in

vivo, it is possible that very few copies of the Na+/K+-ATPase are actually required to

maintain the necessary ionic gradients. The disparity of NKCC1 immunostaining

between early- and late-stage cysts suggests that the mechanism of cyst fluid

accumulation evolves as cystogenesis proceeds. It is likely that the earliest cysts retain

connections with the nephron segment of origin and that the cyst fluid is initially

glomerular filtrate. As the cyst expands and the connection with the parent nephron is

lost, fluid accumulation likely occurs by NKCC1-mediated NaCl and fluid secretion.

In summary, we have confirmed that ADPKD cyst cells undergo a dramatic

dedifferentiation as cystogenesis proceeds. We could detect no evidence of a generalized

basolateral sorting defect or of mislocalization of either the Na+/K+-ATPase, E-cadherin,

or Ksp-cadherin, and we therefore conclude that altered cyst cell polarity does not play a

role in either cyst expansion or cyst fluid secretion in ADPKD. We have directly

demonstrated the expression of the Na+/K+2Cl- cotransporter isoform NKCC1 on the

basolateral membrane of late cysts, supporting the model of chloride and fluid secretion

proposed by Grantham and colleagues.

18

ACKNOWLEDGEMENTS:

We thank Mr. Barry Piekos and Mr. Thomas Ardito for expert technical

assistance with the scanning and transmission electron microscopy.

This work was supported by National Institute of Diabetes and Digestive and

Kidney Diseases Grant DK-57328 (Yale PKD Center).

19

REFERENCES:

1. Avner ED, Sweeney WE and Nelson JW. Abnormal sodium pump distribution

during renal tubulogenesis in congenital murine polycystic kidney disease. Proc. Nat.

Acad. Science USA 89: 7447-7451, 1992.

2. Bachinsky DR, Sabolic I, Emmanouel DS, Jefferson DM, Carone FA, Brown D

and Perrone RD. Water channel expression in human ADPKD kidneys. Am. J. Physiol.

268: F398-F403, 1995.

3. Barr MM and Sternberg PW. A polycystic kidney-disease gene homologue required

for male mating behavior in C. elegans. Nature 401: 386-389, 1999.

4. Barr MM, DeModena J, Braun D, Nguyen CQ, Hall DH and Sternberg PW. The

Caenorhabditis elegans autosomal dominant polycystic kidney disease gene homologs

lov-1 and pkd-2 act in the same pathway. Current Biol. 11: 1341-1346, 2001.

5. Biemesderfer D, DeGray B and Aronson PS. Active (9.6S) and inactive (21S)

oligomers of NHE3 in microdomains of the renal brush border. J. Biol. Chem. 276:

10161-10167, 2001.

6. Brill SR, Ross KE, Davidow CJ, Ye M, Grantham JJ and Caplan MJ.

Immunolocalization of ion transport proteins in human autosomal dominant polycystic

kidney epithelial cells. Proc. Nat. Acad. Science USA 93: 10206-10211, 1996.

7. Carone FA, Nakamura S, Caputo M, Bacallao R, Nelson JW and Kanwar YS.

Cell polarity in human renal cystic disease. Lab. Invest. 70: 648-655, 1994.

8. Charron AJ, Nakamura S, Bacallao R and Wandinger-Ness A. Compromised

Cytoarchitecture and Polarized Trafficking in Autosomal Dominant Polycystic Kidney

Disease Cells. J. Cell Biol. 149: 111-124, 2000.

9. Cuppage FE, Huseman RA, Chapman A and Grantham JJ. Ultrastructure and

function of cysts from human adult polycystic kidneys. Kidney Int. 17: 372-381, 1980.

10. Devuyst O, Burrow CR, Smith BL, Agre P, Knepper MA and Wilson PD.

Expression of aquaporins-1 and -2 during nephrogenesis and in autosomal dominant

polycystic kidney disease. Am. J. Physiol. 271: F169-F183, 1996.

11. Gardner KD, Jr., Burnside JS, Skipper BJ, Swan SK, Bennett WM, Connors BA

and Evan AP. On the probability that kidneys are different in autosomal dominant

polycystic disease. Kidney Int. 42: 1199-1206, 1992.

20

12. Grantham JJ, Geiser JL and Evan AP. Cyst formation and growth in autosomal

dominant polycystic kidney disease. Kidney Int. 31: 1145-1152, 1987.

13. Grantham JJ, Ye M, Gattone VH and Sullivan LP. In vitro fluid secretion by

epithelium from polycystic kidneys. J. Clin. Invest. 95: 195-202, 1995.

14. Hanaoka K, Devuyst O, Schwiebert EM, Wilson PD and Guggino WB. A role for

CFTR in human autosomal dominant polycystic kidney disease. Am. J. Physiol. 270:

C389-C399, 1996.

15. Hayashi M, Yamaji Y, Monkawa T, Yoshida T, Tsuganezawa H, Kitajima W,

Sasaki S, Ishibashi K, Maurmo F and Saruta T. Expression and localization of the

water channels in human autosomal dominant polycystic kidney disease. Nephron 75:

321-326, 1997.

16. Hoyer JR, Sisson SP and Vernier RL. Tamm-Horsfall glycoprotein: ultrastructural

immunoperoxidase localization in rat kidney. Lab. Invest. 41: 168-173, 1979.

17. Huseman R, Grady A, Welling D and Grantham JJ. Macropuncture study of

polycystic disease in adult human kidneys. Kidney Int. 18: 375-385, 1980.

18. Igarashi P and Somlo S. Genetics and pathogenesis of polycystic kidney disease.

JASN 13: 2384-2398, 2002.

19. Kaplan MR, Plotkin MD, Brown D, Hebert SC and Delpire E. Expression of the

mouse Na-K-2Cl cotransporter, mBSC2, in the terminal inner medullary collecting duct,

the glomerular and extraglomerular mesangium, and the glomerular afferent arteriole. J.

Clin. lnvest. 98: 723-730, 1996.

20. Laitinen L, Virtanen I and Saxén L. Changes in the glycosylation pattern during

embryonic development of mouse kidney as revealed with lectin conjugates. J.

Histochem. and Cytochem. 35: 55-65, 1987.

21. Lebeau C, Hanaoka K, Moore-Hoon ML, Guggino WB, Beauwens R and

Devuyst O. BSC2 is the basolateral chloride transporter in a subset of cysts in autosomal

dominant polycystic kidney disease. JASN 10: 36A, 1999.

22. Lebovitz RM, Takeyasu K and Fambrough D. Molecular characterization and

expression of the Na,K-ATPase |-subunit in Drosophila melanogaster. EMBO 8: 193-

202, 1989.

21

23. Mangoo-Karim R, Ye M, Wallace DP, Grantham JJ and Sullivan LP. Anion

secretion drives fluid secretion by monolayers of cultured human polycystic cells. Am. J.

Physiol. 269: F381-F388, 1995.

24. Molitoris BM. Putting the actin cytoskeleton into perspective: pathophysiology of

ischemic alterations. Am. J. Physiol. 272: F430-F433, 1997.

25. Nauli SM, Alenghat FJ, Luo Y, Williams E, Vassilev P, Li X, Elia AEH, Lu W,

Brown EM, Quinn SJ, Ingber DE and Zhou J. Polycystins 1 and 2 mediate

mechanosensation in the primary cilium of kidney cells. Nature Genetics 33: 129-137,

2003.

26. Ogborn MR, Sareen S, Tomobe K, Takahashi H and Crocker JF. Renal tubule

Na,K-ATPase polarity in different animal models of polycystic kidney disease. J.

Histochem. and Cytochem. 43: 785-790, 1995.

27. Pazour GJ, Dickert BL, Vucica Y, Seeley ES, Rosenbaum JL, Witman GB and

Cole DG. Chlamydomonas IFT88 and its mouse homologue, polycystic kidney disease

gene Tg 737 are required for assembly of cilia and flagella. J. Cell Biol. 151: 709-718,

2000.

28. Plotkin MD, Kaplan MR, Verlander JW, Lee WS, Brown D, Poch E, Gullans SR

and Hebert SC. Localization of the thiazide sensitive Na-Cl cotransporter rTSC1 in the

rat kidney. Kidney Int. 50: 174-183, 1996.

29. Qian F, Watnick TJ, Onuchic LF and Germino G. The molecular basis of focal

cyst formation in human autosomal dominant polycystic kidney disease Type 1. Cell 87:

979-987, 1996.

30. Riordan JR. The cystic fibrosis transmembrane conductance regulator. Ann. Rev.

Physiol. 55: 609-630, 1993.

31. Schumacher K, Klotz-Vangerow S, Tauc M and Minuth W. Embryonic renal

collecting duct cell differentiation is influenced in a concentration-dependent manner by

the electrolyte environment. Am. J. Nephrol. 21: 165-175, 2001.

32. Sullivan LP, Wallace DP and Grantham JJ. Chloride and fluid secretion in

polycystic kidney disease. JASN 9: 903-916, 1998.

33. Thomson RB and Aronson PS. Immunolocalization of Ksp-cadherin in the adult

and developing rabbit kidney. Am. J. Physiol. 277: F146-F156, 1999.

22

34. Wilson PD, Sherwood AC, Palla K, Du J, Watson R and Norman JT. Reversed

polarity of Na-K-ATPase: mislocalization to apical plasma membrane. Am. J. Physiol.

260: F420-F430, 1991.

35. Wilson PD, Devuyst O, Li X, Gatti L, Falkenstein D, Robinson S, Fambrough D

and Burrow CR. Apical Plasma Membrane Mispolarization of NaK-ATPase in

Polycystic Kidney Disease Epithelia is Associated with Aberrant Expression of the }2

Isoform. Am. J. Path. 156: 253-268, 2000.

36. Wu G, D'Agati V, Cai Y, Markowitz G, Park JH, Reynolds DM, Maeda Y, Le

TC, Hou H, Kucherlapati R, Edelmann W and Somlo S. Somatic inactivation of Pkd2

results in polycystic kidney disease. Cell 93: 177-188, 1998.

37. Wu G, Markowitz G, Li L, D'Agati VD, Factor SM, Geng L, Tibara S, Tuchman

J, Cai Y, Park JH, van Adelsberg J, Hou H, Kucherlapati R, Edelmann W and

Somlo S. Cardiac defects and renal failure in mice with targeted mutations in Pkd2.

Nature Genetics 24: 75-78, 2000.

23

FIGURE LEGENDS:

Figure 1. Representative transmission electron micrographs of early- through

late-stage PKD2WS25/- ADPKD cysts. Panel A. Early cyst (magnification

x8595). Panel B. Late intermediate cyst (magnification x6440). Panel C.

Extremely advanced cyst (magnification x14291). Panel C inset. Lateral

junction of cystic epithelial cells in extremely advanced cyst (magnification

x54351). A total of 48 cysts were analyzed by TEM; 29 were considered to

be early stage, 13 were considered to be intermediate stage, and 6 were

considered to be advanced. arrows indicate basement membranes; asterisks

indicate collagen deposits.

Figure 2. Representative scanning electron micrographs of PKD2WS25/- ADPKD

cysts. Panel A. Luminal surface of a comparatively small 123 µm diameter

advanced stage cyst illustrating the extent to which cyst cells are capable of

proliferating (magnification x544). Panel B. High magnification of an

early cyst cell. Note the regular cell borders and the prominent

cilia.(magnification x2367) Panel C. Luminal surface of a late intermediate

cyst. The arrow indicates an atypical cyst cell of the type frequently

observed in cysts of this size class. (magnification x1878) Panel D.

Luminal surface of an extremely advanced 3.2 mm diameter cyst.

(magnification x1801). Total number of cysts analyzed = 64.

Figure 3. Immunolocalization of the Na+/K+-ATPase in PKD2WS25/- ADPKD cystic

epithelia. Panel A. Overview of antibody C464.6 staining depicting

basolateral labeling in both normal and cystic epithelia. Cysts 1 and 2 were

DBA positive indicating collecting duct origin and Cyst 3 was NCC (TSC)

positive indicating a distal convoluted tubule origin. (magnification x98).

Panel B. Higher magnification of Cyst 3 labeled with antibody C464.6

(magnification x258). Panel C. Basolateral immunostaining of both normal

and cystic epithelia with anti-Na+/K+-ATPase antibody Alpha-5. The cyst

24

depicted in this panel was positively labeled with the anti-THP antibody

indicating a thick ascending limb origin. (magnification x707).

Figure 4. Immunolocalization of Ksp-cadherin in PKD2WS25/- ADPKD cystic

epithelia. Panel A. Overview of basolateral Ksp-cadherin staining in both

normal and cystic epithelia. The immunolabeled section depicted in this

panel is the anti-Ksp-cadherin portion of a double-labeling experiment done

with the anti-Na+/K+-ATPase antibody C464.6. The anti-Na+/K+-ATPase

portion of the experiment was presented in Panels A and B of Figure 3.

(magnification x98). Panel B. High magnification of the cyst labeled with

the asterisk in Panel A (magnification x258).

Figure 5. Immunolocalization of E-cadherin in PKD2WS25/- ADPKD cystic

epithelia. The anti-E-cadherin antibody Clone 36 clearly labels the

basolateral membranes of cystic epithelia. The cyst depicted in this panel

was DBA positive indicating a collecting tubule origin. (magnification

x354).

Figure 6. Immunolocalization of Na+/K+-ATPase, Ksp-cadherin and E-cadherin

in advanced PKD2WS25/- ADPKD cysts. Panel A. The tissue section

depicted in this panel was labeled with the anti-Na+/K+-ATPase antibody

C464.6. Positive immunostaining was not detected in either the highly

dedifferentiated cystic epithelium or the surrounding fibrotic tissue. Areas of

the section outside the field depicted in this panel showed strong basolateral

staining in both unaffected tubule segments and early cysts. The lower cyst

was THP positive indicating a thick ascending limb origin, but the upper

cyst was not positively labeled by any of the tubule segment markers used in

this study. (magnification x327). Panel B. Double-labeling of the same

tissue section depicted in Panel A with the anti-Ksp-cadherin antibody.

Note the prominent basal immunostaining in the lower THP positive cyst.

(magnification x327). Panel C. Prominent lateral E-cadherin

immunostaining of cystic epithelial cells that were not positively labeled by

25

anti-Na+/K+-ATPase or anti-Ksp-cadherin antibodies or tubule segment-

specific markers. (magnification x211).

Figure 7. Immunolocalization of NKCC1 in PKD2WS25/- ADPKD cysts. Panel A.

Immunolabeling of the cystic renal cortex with the anti-NKCC1 antibody.

Note the positive staining in the glomerular associated structures indicated

by the arrowheads and the lack of staining in the early cysts indicated by the

asterisks. See Figures 3a and 4a for sequential serial-section labeling with

anti-Na+/K+-ATPase and anti-Ksp-cadherin antibodies. (magnification x92)

Panel B. Anti-NKCC1 immunolabeling of an advanced cyst that does not

express detectable levels of the Na+/K+-ATPase, Ksp-cadherin, or tubule

segment-specific markers. The arrowhead indicates glomerular mesangial

staining. (magnification x278). Panel C. Heterogeneous anti-NKCC1 cyst

staining. Cyst 1 is a DBA /Ksp-cadherin positive, Na+/K+-ATPase negative

cyst (data not shown) that stains moderately with the anti-NKCC1 antibody.

Cyst 2 is a NKCC1-negative glomerular cyst that is also negative for all

tubule segment markers, the Na+/K+-ATPase and Ksp-cadherin. Note the

lack of anti-NKCC1 mesangial staining in the glomerular rudiment

indicated by the asterisk. Cyst 3 expressed high levels of basolateral

NKCC1, but was negative for all tubule segment markers, the Na+/K+-

ATPase and Ksp-cadherin. (magnification x740).

26

Table 1. Nephron segment of origin of Na+/K+-ATPase-, Ksp-cadherin-, E-cadherin-, and NKCC1-

positive cysts.

Cyst Segment of Origin

PT TAL DCT CT Indeterminate Total # of Total # of(+/tot) (+/tot) (+/tot) (+/tot) (+/tot) positive cysts cysts examined

Na+/K+-ATPase 3/3 21/21 2/3 34/36 0/19 60 82

Ksp-cadherin 3/3 21/21 3/3 36/36 3/19 66 82

E-cadherin 2/2 20/20 5/5 32/32 14/14 73 73

NKCC1 0/2 1/17 0/1 1/28 16/17 18 65

Cyst segment of origin was determined by lectin and antibody labeling as described in the Methods.

Cysts not labeled with the segment-specific markers used in this study were considered to be of an

indeterminate origin. PT, proximal tubule; TAL, thick ascending limb; DCT, distal convoluted tubule;

CT, collecting tubule; +/tot, number of positively labeled cysts/total number of cysts observed in each

cyst segment category.