Early Cyst Growth Is Associated with the Increased Nuclear Expression of Cyclin D1/Rb Protein in an...

Fax +41 61 306 12 34E-Mail [email protected]

Original Paper

Nephron Exp Nephrol 2011;117:e93–e103 DOI: 10.1159/000320149

Early Cyst Growth Is Associated with theIncreased Nuclear Expression of Cyclin D1/Rb Protein in an Autosomal-Recessive Polycystic Kidney Disease Rat Model

Kristina G. Schwensen a Jane S. Burgess a Nicole S. Graf b Stephen I. Alexander b David C. Harris a Jacqueline K. Phillips c, d Gopala K. Rangan a

a Centre for Transplant and Renal Research, Westmead Millennium Institute, Sydney Medical School – Western, The University of Sydney, and b Children’s Hospital at Westmead, and c Australian School of Advanced Medicine, Macquarie University, Sydney, N.S.W. , and d Murdoch University, Perth, W.A. , Australia

cyst growth in rats and humans, suggesting that there might be a therapeutic window in which cyclin-dependent kinase inhibitors are most effective in preventing kidney enlarge-ment in ARPKD. Copyright © 2010 S. Karger AG, Basel

Introduction

Cystic renal diseases are a family of disorders that are characterized by the formation of renal cysts, renal fibro-sis, and the development of end-stage kidney failure [1] . The two most common forms are autosomal-dominant (AD) and autosomal-recessive polycystic kidney disease (ARPKD) [1] . ADPKD is typically diagnosed in adults, with kidney failure occurring beyond the fourth decade [1] . The cysts in ADPKD form sporadically as focal diver-ticular protuberances from the distal nephron which, with further growth ( 1 200 � m in diameter), detach and continue to enlarge as isolated cysts embedded in the re-nal parenchyma [2] . In contrast, ARPKD is diagnosed in early infancy and the renal disease is characterized by synchronized and diffuse cystic dilatation of collecting

Key Words

Polycystic kidney disease � Proliferation � Cyclin

Abstract

Background: In this study we hypothesised that prolifera-tion, and the increased expression of G 1 -phase cyclins (D1, E) and phosphorylated retinoblastoma protein (p-Rb) is re-stricted to the early period of synchronized cyst growth in autosomal-recessive polycystic kidney disease (ARPKD). Methods: Lewis polycystic kidney disease ( lpk ) rats (model of ARPKD; postnatal weeks 1, 3, 6, 12 and 24; n = 6 each) as well as human juvenile cystic renal disease tissue (n = 2) were examined. Results: Between weeks 1 and 3, the percentage cyst area increased 6-fold in lpk rats, followed by a more pro-gressive rise (1.5-fold increase) until week 24. The number of Ki-67-, cyclin D1- and p-Rb-positive cells increased in lpk rats and peaked at week 3, declining thereafter. By serial sec-tions, cysts co-expressed Ki-67, cyclin D1 and p-Rb. The ex-pression of cyclin E was variable, and peaked at week 24. In human tissue, small cysts had a higher expression of p-Rb. Conclusion: Proliferation and the increased nuclear expres-sion of cyclin D1 and p-Rb coincide with the early phase of

Received: August 22, 2009 Accepted: May 11, 2010 Published online: October 2, 2010

Dr. G. Rangan Centre for Transplant and Renal Research Level 2, Clinical Sciences Block Westmead Hospital, Westmead, Sydney, N.S.W. 2145 (Australia) Tel. +61 2 9845 6962, Fax +61 2 9633 9351, E-Mail g.rangan @ wmi.usyd.edu.au

© 2010 S. Karger AG, Basel1660–2129/11/1174–0093$38.00/0

Accessible online at:www.karger.com/nee

http://dx.doi.org/10.1159%2F000320149

Schwensen /Burgess /Graf /Alexander /Harris /Phillips /Rangan

Nephron Exp Nephrol 2011;117:e93–e103e94

ducts, which do not detach and become isolated cysts. Currently, there are no specific therapies to prevent cyst formation and enlargement in any type of cystic renal disease [3] .

An increased rate in the proliferation of the epithelial cells lining cysts is one of the key factors in the initiation and enlargement of cysts in both ADPKD and ARPKD [4–8] . Cell proliferation is mediated by proximal mito-genic signal transduction pathways which converge on the molecular mechanisms governing cell cycle activa-tion [9] . The cell cycle is regulated by the sequential acti-vation and inactivation of holoenzymes, consisting of cy-clins partnered with an obligate cyclin-dependent kinase (Cdk) [10, 11] . The classical model of the cell cycle indi-cates that D-type cyclins partnered with Cdk4 are re-quired for G 1 - to S-phase transition, whereas cyclin E partnered with Cdk2 is required for progression of both the G 1 - and S-phases [10] . The intracellular level of cy-clins are the critical determinant of Cdk activity [12] , and the overexpression of cyclins D1 and E are frequently im-plicated in mediating other proliferative diseases, such as cancer [13, 14] . In addition, an important outcome of cy-clin D1-Cdk4 and cyclin E-Cdk2 activation is phosphor-ylation of the retinoblastoma protein (p-Rb), a pocket protein transcriptional regulator, that activates substrates that promote proliferation [15] .

Although several studies have shown that G 1 -phase cyclins and p-Rb are increased in small animal models of cystic renal disease [16–18] , the time-dependent changes in their expression, however, are not known. Understanding the temporal pattern of cyclin and p-Rb expression in cystic renal disease has critical importance because it is likely to determine the therapeutic window during which Cdk inhibitors will be effective in prevent-ing kidney enlargement [19] . This is of particular rele-vance to ARPKD where cyst growth tends to be synchro-nized to the early period of the disease [3] . In this study, we hypothesised that renal cell proliferation, the expres-sion of G 1 -specific cyclins (D1, E) and a marker of Cdk activation (p-Rb) are restricted to the early phase ofARPKD. To test this hypothesis, we used the Lewis poly-cystic kidney disease (lpk ) rat model [20] and also inves-tigated human tissue. The lpk rat model arose from a spontaneous mutation and phenotypically recapitulates the renal pathology of human ARPKD (with the absence of hepatobiliary disease) [20] . It is characterized by early postnatal severe cystic renal disease, arising from dilated collecting duct segments, and kidney failure over a 24-week period, making it an ideal model to test the hypoth-esis of this study [20] .

Materials and Methods

Animals and Experimental Design Archival paraffin sections from a previous study [20] were ex-

amined at weeks 1, 3, 6, 12 and 24 (n = 6 per group) and compared to Lewis (control) animals (n = 3 per group).

Human Tissue Human juvenile cystic kidney and normal kidney tissue were

also examined to determine the expression of cyclin D1, cyclin E and p-Rb protein. Archival paraffin samples were obtained from the Department of Histopathology, Children’s Hospital at West-mead. Two normal kidney and two juvenile cystic kidney disease tissue samples were obtained. One of the control samples was from the normal part of a kidney affected by a Wilms’ tumour (sample 1) and the other was normal kidney from a partial resec-tion from a patient with reflux nephropathy with duplex kidney (sample 2). The two juvenile cystic renal disease samples were from a 22-weeks’ gestation male foetus (post-mortem) with cystic renal disease (sample 3) and the other was from a 14-year-old fe-male subject with a history of cystic renal disease associated with congenital hepatic fibrosis (sample 4). Studies were approved by the Human Research Ethics Committee at the Children’s Hospital at Westmead.

Histology and Immunohistochemistry Rat kidneys were fixed in either 4% formalin or methyl Car-

noy’s solution for 24 h before being embedded in paraffin. Human tissue was fixed in 10% formalin. Immunostaining was per-formed on 4- � m-thick sections, as previously described [20] . The following primary antibodies were used: (1) rabbit monoclonal IgG antibody reactive against rat and human Ki-67 (1: 100 over-night at 4 ° C; clone SP6, Lab Vision Laboratories, Calif., USA); (2) rabbit monoclonal IgG antibody reactive against rat and human cyclin D1 (1: 25 dilution for 30 min at room temperature; Clone SP4, Neomarkers, Calif., USA); (3) rabbit polyclonal IgG antibody reactive against human cyclin E (1: 200 dilution for 1 h at room temperature; Neomarkers, Calif., USA); (4) rabbit polyclonal IgG antibody reactive against rat and human phosphorylated (serine 780) Rb protein (1: 200 dilution for 30 min at room temperature; Cell Signaling Technologies, USA). All formalin-fixed sections underwent antigen retrieval in citrate buffer (pH 6.0) followed by microwave oven heating for 10 min. Immunoreactivity was visu-alised with Vectastain Elite ABC reagent (Vector Laboratories) and the chromogen diaminobenzidine. Slides were counter-stained either with methyl-green (cyclin D1 and p-Rb) or haema-toxylin (cyclin E and Ki-67).

Analysis of Histology Two methods of computer-based quantitative image analysis

(Optimas, Media Cybernetics and Aperio Technologies) were used to quantify the histological changes. Using the Optimas method, slides were viewed with a light microscope (BX51, Olym-pus, Olympus Australia), and JPEG images were taken with a dig-ital camera (Olympus DP11) and viewed on a desktop computer. To assess the total cyst area, cystic spaces were traced manually using the computer mouse in three non-overlapping digital im-ages (at a magnification of ! 40) of trichrome-stained sections. The percentage area of each image occupied by cystic spaces was calculated using the percent analysis tool in the Optimas pro-

Cyclin D1/Rb and ARPKD Nephron Exp Nephrol 2011;117:e93–e103 e95

gram, and a mean was calculated for each animal. Similarly, to quantify Ki-67, cyclin D1, cyclin E and p-Rb expression, the area of DAB-stained nuclei in a digital image was determined using the Optimas program, and expressed as a percentage. For Ki-67, cyclin D1, cyclin E and p-Rb, digital images were randomly cap-tured from 3 fields from the outer medulla plus 3 fields from the cortex for each kidney section (at a magnification of ! 200). A uniformly random cluster method was used to determine micro-scopic fields for evaluation [21] .

To assess the expression of Ki-67, cyclin D1, cyclin E and p-Rb in cystic and non-cystic tissue, further analysis was undertaken in whole-slide digital images using the Aperio system. Whole slide digital images were created from slides stained with either Ki-67, cyclin D1, cyclin E or p-Rb, scanned at a magnification of ! 200–400 using a scanner (Aperio Scanscope CS, Aperio Tech-nologies) [22] . Digitalized images were then viewed on a desktop computer using the Imagescope program (version 9, Aperio Tech-nologies). The nuclear algorithm of the Aperio software was used to determine the percentage of DAB-stained Ki-67, cyclin D1, cy-clin E and p-Rb-positive nuclei as a proportion of total nuclei present. Using the software, each of the Ki-67, cyclin D1, cyclin E and p-Rb stains were individually calibrated and a specific macro was developed for each of them. The quantitation was performed in the entire section (total positive nuclei) (data not shown) or in specific areas of the section (cystic and non-cystic tissue). For the latter, the analysis was performed in five randomly selected cysts or non-cystic tissue areas per section, and these areas were traced manually using the computer mouse. For analysis non-cystic tis-sue at week 1, medullary rays (which contain a high constitutive expression of Ki-67, cyclin D1) were excluded. The mean percent-age of DAB-positive cells as a proportion of total nuclei (DAB-positive and DAB-negative) was calculated for each section/ani-mal. Only strongly positive DAB nuclei (designated 3+ in the Ape-rio software) were included in the calculations.

Statistics Results were presented as mean 8 SEM. The data were ana-

lysed with the JMP statistical software package (version 4.04,SAS Institute, Carey, N.C., USA). Comparisons between the ex-perimental groups were performed by ANOVA, followed by a post-hoc analysis with the Tukey-Kramer HSD test. p ! 0.05 in-dicated statistical significance.

Results

Cyst Growth Is Synchronized and Peaks between Weeks 1 and 3 in lpk Rats At week 1 in lpk rats, cystic renal disease was only

mild, characterized by focal areas of dilated tubules in the outer medulla ( fig. 1 ), whereas by week 3, there was dif-fuse fusiform dilatation of tubular segments ( fig. 1 ). By quantitation, the greatest increase in the percentage cyst area occurred from weeks 1 to 3, rising 6-fold (week 1: 8.9 8 0.3 vs. week 3: 56.2 8 1.6%; mean 8 SEM, p ! 0.001) ( fig. 2 ). In contrast, cyst growth between weeks 3 and 24 was less marked, with the percentage cyst area increasing

only a further 1.5-fold and almost reaching the maxi-mum at week 12 (week 6: 62.8 8 3.1; week 12: 79.5 8 3.5; week 24: 86.8 8 3.3%; p ! 0.01 compared to week 3) ( fig. 2 ).

Cyst Proliferation Peaks between Weeks 1 and 3 in lpk Rats Numerous cells were positive for Ki-67 in the non-di-

lated tubules of medullary rays at week 1, consistent with postnatal nephrogenesis ( fig. 3, 4 ). Although there were no differences in Ki-67 expression between the two groups at this time-point ( fig. 4 a, b), epithelial cells lining focally dilated cysts were positive for Ki-67 in lpk rats. In contrast, between weeks 3 and 24, Ki-67 declined in con-trol animals, whereas it remained elevated in lpk rats ( fig. 4 a, b). Positive staining was predominantly found in cells lining dilated cystic tubules in lpk rats. The increase in Ki-67 in lpk rats peaked at week 3, and gradually de-creased at later time-points, such that there was no sig-nificant difference in proliferation with the Lewis group at week 24 ( fig. 4 a, b). The temporal increase in the num-ber of Ki-67 cells in lpk rats was similar in both the cortex and outer medulla ( fig. 4 a, b).

The percentage of Ki-67-positive nuclei was also ana-lysed in cystic and non-cystic tissue during weeks 1–6 using the Aperio software in whole-slide digital images ( table 1 ). In non-cystic tissue, Ki-67 peaked at week 3 compared to Lewis controls. In cysts, the proportion of epithelial cells positive for Ki-67 peaked at week 1 ( ta-ble 1 ).

Cyclin D1-Positive Cysts Peak between Weeks 1 and 3 in lpk Rats At week 1, there was strong constitutive nuclear ex-

pression of cyclin D1 in both Lewis and lpk kidneys ( fig. 4 c), which had a similar cellular distribution (non-dilated medullary rays) as Ki-67, at this time-point. At week 3, the constitutive expression of cyclin D1 declined in control animals, and at this and all other time-points, proximal cortical and outer medulla tubular nuclei were weakly positive for cyclin D1 in normal Lewis animals ( fig. 3 b). In the absence of positivity for Ki-67, this would suggest that proximal tubules are primed for (but not un-dergoing) proliferation, as hypothesized previously [23] . In lpk animals, the number of epithelial cells positive for cyclin D1 was significantly increased ( fig. 3 e). At weeks 1 and 3, cystic epithelial cells lining dilated tubular seg-ments in the outer medulla were strongly positive for cy-clin D1. In general, smaller cysts contained a greater number of cyclin D1-positive cells. By quantitative image



Week 1

Week 1

Week 3

Week 3

Week 6

Week 6

Week 12

Week 12

Week 24

Week 24

Lewis

Ipk

Fig. 1. a Whole-slide low-power digital images of trichrome-stained sections at various time-points (week 1, 3, 6, 12 and 24) in Lewis (control, upper panels) and lpk rats (lower panels). Cystic renal disease increased markedly between postnatal weeks 1 to 3. At week 24, interstitial collagen deposition was relatively diffuse. b Photomicrographs of trichrome-stained sections of lpk rats at weeks 1, 3, 6, 12 and 24. ! 40. At week 1, there were focal areas of

dilated tubules in the outer medulla. By week 3, there was a sig-nificant increase in the size of the cystic dilatations, which were rectangular and perpendicular to the cortical outline of the kid-ney, indicative of a collecting duct origin. The severity and fre-quency of cystic dilatation of the tubules progressively worsened with time. At later time-points, this was accompanied by intersti-tial inflammation and fibrosis.

0

25

75

50

100

Week 1 Week 3 Week 6 Week 12 Week 24

**

**

Perc

enta

ge

Fig. 2. Time-course of percentage cyst area in lpk rats as assessed by computer-assisted image analysis. Data expressed as mean 8 SEM; p ! 0.05 compared to week 1.

Colo

r ver

sion

ava

ilabl

e on

line

a

b


analysis, the number of cyclin D1-positive cells in the out-er medulla peaked at week 3 in lpk rats, and declined at later time-points ( fig. 4 c).

Further subanalysis of cyclin D1 in cystic and non-cystic tissue in LPK rats was undertaken. In non-cystic tissue, cyclin D1 peaked at week 3 compared to control, whereas for cystic tissue this was highest between weeks 1 and 3 ( table 1 ).

pRb-Positive Cysts Peak between Weeks 1 and 3 in lpk Rats Similar to Ki-67 and cyclin D1, there was also strong

nuclear expression of p-Rb in both Lewis and lpk ratsat week 1 ( fig. 4 d). A marked decline in the expressionof p-Rb was found in Lewis control animals at week 3, whereas increased levels were maintained in the lpk rats ( fig. 3 f, 4 d). After week 3, p-Rb expression was reduced in both Lewis control and lpk rats and was not significantly different between the two groups ( fig. 4 d).

Subanalysis of the tissue in whole-slide digital images showed that p-Rb peaked between weeks 1 and 3 in both cystic and non-cystic areas in LPK rats ( table 1 ).

Proliferating Cysts Are Positive for Cyclin D1 and p-Rb Protein at Week 3 in lpk Rats To determine whether cysts positive for Ki-67 co-ex-

pressed cyclin D1 and p-Rb, serial sections from lpk rats at week 3 were immunostained for these proteins. As shown in figure 5 , cysts expressing Ki-67 ( fig. 5 b, e) were also positive for cyclin D1 ( fig. 5 a, d) and p-Rb ( fig. 5 c, f). To determine if there was a correlation between Ki-67, cyclin D1 and p-Rb, the percentage of positive nuclei was

quantified in five cysts from each animal using whole-slide digital images. By quantitation the mean percent-ages of cells positive in cysts for Ki-67, D1 and p-Rb were 30.2 8 18.4, 36.2 8 16.4 and 34.5 8 7.8%, respectively. There was a significant correlation between Ki-67 and cyclin D1 (r 2 = 0.74, p = 0.03). However, the association was less strong between Ki-67 and p-Rb (r 2 = 0.65, p = 0.05), as well as between cyclin D1 and p-Rb (r 2 = 0.50,p = 0.12).

Cyclin E-Positive Cysts Peak at Week 24 in lpk Rats At week 1, cyclin E-positive cells were present in the

outer medulla and medullary rays in both lpk and control rats (data not shown). In addition, some dilated cysts were positive for cyclin E at week 1 in lpk rats. The number of positive cells in lpk rats compared to Lewis control for cyclin E was variable between weeks 3 and 24 ( fig. 6 ). However, with increasing age and maturity of the cysts, a greater number of cyclin E-positive cystic epithelial cells were detected. By week 24, the number of cyclin E-posi-

Table 1. Percentage of nuclei positive for Ki-67, cyclin D1 and p-Rb in Lewis and LPK rats

Lewis L PK

week 1 week 3 week 6 week 1 week 3 week 6

Ki-67, %Non-cystic 6.082.1 1.181.7 0.580 5.785.3 4.181.5* 2.781.4*Cystic – – – 3.281.4 1.981.0 1.980.9

Cyclin D1, %Non-cystic 5.681.2 8.685.6 5.184.4 8.285.0 19.881.0* 7.184.2Cystic – – – 24.185.6 32.885.0 12.285.5

p-Rb, %Non-cystic 2.281.8 080 080 13.8814.2* 8.289.5* 0.0280.03*Cystic – – – 34.4817.2 17.2812.9 0.1980.18

All data expressed as mean 8 SD. * p < 0.05 compared to time-respective Lewis group.

Table 2. Percentage of nuclei positive for Ki-67, cyclin D1 and p-Rb in small and large cysts in juvenile human cystic renal dis-ease (sample 3)

n Meandiameter�m

Ki-67%

CyclinD1 %

p-Rb%

Small cysts 5 68825 1.681.4 1.182.7 7.9811.6Large cysts 6 362883 0.580.6 0.180.2 0.280.4



tive cysts had increased in lpk rats. Furthermore, the in-tensity of the immunostaining as well as the number of cystic epithelial cells positive for cyclin E was increased at week 24. By quantitation, the percentage of cells posi-tive in cysts was highest at week 24 (week 1: 2.1 8 1.8; week 3: 17.1 8 13.4; week 6: 38.2 8 11.4; week 12: 12.0 8 11.1; week 24: 42.5 8 6.0).

Small Cysts Strongly Express p-Rb in Early Juvenile Human Cystic Renal Disease The expression of Ki-67, cyclin D1 and p-Rb was next

assessed in juvenile human cystic renal disease. The con-

trol samples (1 and 2) had normal kidney morphology on PAS-stained sections ( fig. 7 ). Sample 3, which was from a 22-weeks’ gestation male foetus with cystic renal disease, was characterized by diffuse areas of dilated cystic tu-bules (extending from the cortex to the medulla) in con-junction with epithelial hyperplasia and interstitial space expansion. Cystic dilatations were predominantly be-tween 60 and 100 � m in diameter (defined as small cysts), and to a lesser extent between 300 and 400 � m in diam-eter. The ratio of small to large cysts in sample 3 was ap-proximately 10 to 1. In the cortex of sample 3, the neph-rogenic zone, containing immature glomeruli, appeared

Ki-67 Cyclin D1 p-Rb

Negativecontrol

lpk

Lewis

a b c

d e f

g h i

Fig. 3. Photomicrographs of sections of Lewis ( a – c ) and lpk rats ( d – f ) showing immunostaining for Ki-67 ( a , d ), cyclin D1 ( b , e ) and p-Rb ( c , f ) at week 3. Magnification ! 400. Epithelial cells lin-ing cysts (lumens denoted by asterisks) had increased nuclear ex-

pression of Ki-67, cyclin D1 and p-Rb, as shown by positive nucle-ar staining. Negative control (primary antibody replaced with di-luent) are shown in panels g–i for Ki-67, cyclin D1 and p-Rb, respectively.

Colo

r ver

sion

ava

ilabl

e on

line


normal. In contrast, sample 4, which was from a 14-year-old female subject with cystic renal disease associated with congenital hepatic disease, showed much more ad-vanced cystic renal disease. This was characterized pre-dominantly by diffuse areas of large cysts (ranging in size from 2 to 6 mm in diameter), lined by flattened and de-differentiated cystic epithelial cells, in addition to inter-stitial fibrosis. Sample 4 also contained smaller cysts (ap-proximately 100 �m in diameter) to a much lesser degree, and these were located in the interstitial spaces between the larger cysts.

Immunostaining for Ki-67, cyclin D1 and p-Rb pro-tein was performed in serial sections on samples 1–4 ( fig. 7 ). In both of the normal human kidney samples (1 and 2), cyclin D1 immunoreactivity was weakly expressed in the nuclei of cortical proximal tubules and some me-

sangial cells ( fig. 7 ). Ki-67 was present only sporadically in the nuclei of some cortical tubular epithelial cells, whereas p-Rb immunoreactivity was weakly present in the nuclei and cytoplasm of most distal tubules and podo-cytes. In juvenile cystic renal disease, the epithelial cells lining the cysts in sample 3 were sporadically positive for cyclin D1 and Ki-67 whereas strong immunoreactivity for p-Rb was detected in clusters of dilated cystic tubules. In sample 4, the expression of Ki-67, cyclin D1 and p-Rb was much weaker than in sample 3. Epithelial cells lining small cysts were weakly and sporadically positive for cy-clin D1, Ki-67 and p-Rb.

Quantitative analysis of Ki-67 and cyclin D1/p-Rb was undertaken in small (60–100 � m in diameter) and large (greater than 300 � m in diameter) cysts from sample 3. As shown in table 2 , the expression of Ki-67, cyclin D1

0

5

10

15

Week 1c Week 3 Week 6

Cyclin D1

Week 12 Week 24

*

Perc

enta

ge

0

5

10

15

20

25

Week 1d Week 3 Week 6

p-Rb

Week 12 Week 24

*

Perc

enta

ge

0

1

2

3

4 Controllpk rats

Week 1a Week 3 Week 6

Ki-67 (cortex)

Week 12 Week 24

**Pe

rcen

tage

0

5

3

2

1

4

7

6

Week 1b Week 3 Week 6

Ki-67 (outer medulla)

Week 12 Week 24

*

**

Perc

enta

ge

Fig. 4. Graphs showing the percentage immunoreactivity of Ki-67 (cortex and outer medulla; a , b ), cyclin D1 (outer medulla, c ) and p-Rb (outer medulla, d ) in control (Lewis) and lpk rats at the var-ious time-points (weeks 1–24), as determined by quantitative im-age analysis. Ki-67 peaked at week 3 but was still increased be-

tween weeks 6 and 12, compared to control. Cyclin D1 and p-Rb increased in lpk rats and peaked at week 3 (n = 6 per time-point, data are expressed as mean 8 SEM; * p ! 0.05 compared to the control groups at each time-point).



Cyclin D1 Ki-67 p-Rb

a b c

d e f

and p-Rb was higher in small cysts. Furthermore, small cysts also exhibited a higher expression of pRb relative to Ki-67 and cyclin D1. Nevertheless, no correlation be-tween Ki-67 and p-Rb/cyclin D1 was present in either small or large cysts (all p 1 0.05).

Discussion

Human and experimental studies support the hypoth-esis that excessive proliferation is one of the major mech-anisms leading to cyst formation in ADPKD and ARPKD [4, 24, 25] . To date, in vivo studies in PKD have examined the expression of cell cycle proteins in a cross-sectional design or over short-term periods, and none (to our knowledge) have investigated in a comprehensive time-course manner. In this study, we have for the first time analysed the time-dependent relationship between cyst growth, proliferation and the expression of cyclin D1, cy-clin E and p-Rb over a 6-month period in the lpk rat mod-el of ARPKD, as well as in human juvenile cystic renal disease tissue. The retention of cyclin D1, cyclin E and phosphorylated Rb protein in the cell nucleus is an im-portant mechanism promoting the G 1 - to S-phase transi-

tion and cell proliferation [11] . Our data provide two im-portant and new insights: (1) that the upregulation of cy-clin D1 and p-Rb primarily occurs during the early, but not the later, phase of cystic growth in rats and humans, and (2) that, in contrast, the expression of cyclin E in cys-tic tissue did not correspond to the temporal pattern of cystic growth, Ki-67, cyclin D1 and p-Rb. Collectively, these data therefore suggest that proliferation, increased cyclin D1 expression and the phosphorylation of Rb are likely to mediate cyst growth in lpk rats.

In this study, cyst growth was assessed by computer-assisted area morphometry and the results showed that the rate of cyst enlargement was most rapid during the first three postnatal weeks in lpk rats. Consistent with our main hypothesis, this early period of synchronized cystic growth also coincided with the peak postnatal expression of Ki-67-, cyclin D1- and p-Rb-positive cells in lpk rats. Although total Ki-67 remained elevated at weeks 6 and 12 compared to control (at a time when total cyclin D and p-Rb had decreased), this was lower than at earlier weeks. Furthermore, subanalysis of cysts showed that the ex-pression of all three proteins at week 6 was still substan-tially increased compared to non-cystic tissue in Lewis control animals. To extend these findings we also exam-

Fig. 5. Photomicrographs showing serial sections of cyclin D1 ( a , d ), Ki-67 ( b , e ) and p-Rb ( c , f ) in lpk rats at week 3. Two examples are shown at low power ( ! 100 magnification, a – c ) and high power ( ! 400 magnification, d – f ). a – c are three consecutive sections, whereas d – f are another three consecutive sections. The nuclei of epi-thelial cells lining cysts were simultaneously positive for cyclin D1, Ki-67 and p-Rb.

Colo

r ver

sion

ava

ilabl

e on

line


ined serial sections of individual cysts in lpk rats at week 3 and found that cystic epithelial cells were simultane-ously positive for Ki-67 as well as cyclin D1 and p-Rb. Together, these data support the hypothesis that the in-creased synthesis of cyclin D1 and p-Rb have a patho-genic role in mediating cyst growth in lpk rats.

Contrary to the classical model of the cyclin regula-tion in the cell cycle and unlike cyclin D1 [10, 11] , the ex-pression of cyclin E in cysts peaked at week 24, at a time when total proliferation had largely subsided. The mo-lecular mechanisms underlying the discrepancy between cyclin D1 and cyclin E in the lpk rat are presently not clear. However, such discrepancies have been reported in other diseases, in particular in breast cancer. Here, pro-liferation in estrogen receptor (ER)-positive cancers is driven by increased levels of cyclin D1 and requires acti-vation of p-Rb [26, 27] . In contrast, in ER-negative breast cancers, proliferation is dependent on cyclin E and does not require cyclin D1 or Rb [26, 27] . Cyclin E overexpres-sion in breast cancer is also associated with a more ag-

gressive type of tumour and was associated with chromo-somal instability leading to inactivation of the tumour suppressor p53 [28] . Interestingly, our data for cyclin D1 and E contrast with in vitro studies which show that poly-cystin-1 negatively regulates cyclin E-Cdk2, rather than cyclin D1-Cdk4 [29, 30] . Whether this discrepancy is due to the type of ciliary protein that is mutated in the lpk rat or other factors in vivo, is presently unclear and requires further investigation. Nevertheless, the increased expres-sion of cyclin E in the lpk rat at week 24 suggests that this protein may have a pathological role in this model, but one that is independent of mediating proliferation.

It is noteworthy that the expression of Ki-67, cyclin D1 and p-Rb were elevated at week 1, in both the control and the lpk rat. This constitutive expression was largely due to their increased nuclear expression in cells within med-ullary rays. These findings are consistent with the fact that nephrogenesis (primarily lengthening of tubules) is only completed during the early postnatal period in rats [31] . Although there was a reduction in these parameters

Lewis

lpk

Cyclin E

a

0

5

10

15 Controllpk rats

Week 1b Week 3 Week 6 Week 12 Week 24

% a

rea

stai

ned

posi

tive

for c

yclin

E

Fig. 6. a Photomicrographs of sections of Lewis (upper panel) and lpk rats (lower panel) showing immunostaining for cyclin E. Mag-nification ! 400. Epithelial cells lining cysts had increased nucle-ar expression of cyclin E as shown by positive nuclear staining. b Graph showing the percentage immunoreactivity of cyclin E (outer medulla) in control (Lewis) and lpk rats at the various time-points (weeks 1–24), as determined by quantitative image analy-sis. Cyclin E expression was variable in lpk rats and peaked at week 24 compared to control (n = 3–6 per time-point, data are ex-pressed as mean 8 SEM; * p ! 0.05 compared to the control groups at each time-point).

Colo

r ver

sion

ava

ilabl

e on

line



Nor

mal

kid

ney

(No.

1)

Juve

nile

cys

ticre

nal d

isea

se (N

o. 3

)Ju

veni

le c

ystic

rena

l dis

ease

(No.

4)

PAS Ki-67 Cyclin D1 p-Rb

a b c d

e f g h

i j k l

at week 3 in Lewis animals, they remained increased in the lpk rat, suggesting that cystic renal disease in the lat-ter is due to dysregulation of renal maturation [32] .

The present study also examined for the first time the expression of Ki-67, cyclin D1 and p-Rb in juvenile cystic renal disease in humans. Two contrasting samples were available for review (samples 3 and 4). Sample 3, from the 22-weeks’ gestation male foetus, represented an early phase of disease, and the most obvious feature identified was that clusters of cysts were strongly immunoreactive against p-Rb with weaker expression of Ki-67 and p-Rb. In contrast, sample 4, which was from the 14-year-old fe-male subject with cystic renal disease, contained much larger cysts which were lined by flattened dedifferenti-ated cystic epithelial cells, which were sporadically posi-tive for Ki-67, cyclin D1 and p-Rb. Although the number of human samples that were examined in this study was small, these results complement our observations in the rat model and, importantly, suggest that the age (and size)

of the cyst may be an approximate indicator of its cell cycle activation and whether it is likely to be responsive to treatment with cell cycle inhibitors. In this case, larger cysts presumably represent a later phase of disease and, due to their absence of positive staining for p-Rb, are un-likely to respond to treatment with cell cycle inhibitors. However, in contrast to the rat data, no correlation be-tween Ki-67, p-Rb and cyclin D1 was present. There may be a number of reasons for this discrepancy: (1) the sam-ple size of the human samples was limited, and (2) there may have been variability in the fixation of the human tissue and the ability to unmask Ki-67 and cyclin D1 in comparison to pRb.

In conclusion, the results of this paper show that syn-chronized cyst growth in the lpk rat coincides with in-creased cystic epithelial cell proliferation and the nuclear expression of cyclin D1 and p-Rb, but not that of cyclin E. These data suggest a specific role for cyclin D1 and p-Rb in mediating cyst growth, and raise the idea that

Fig. 7. Photomicrographs of serial sections showing PAS staining ( a , e , i ) and immunostaining for Ki-67 ( b , f , j ), cyclin D1 ( c , g , k ) and p-Rb ( d , h , l ) in normal human kidney (sample 1, a – d ) and hu-man juvenile cystic renal disease (samples 3 and 4, e – l ). In normal human kidney (sample 1), there was occasional nuclear staining for Ki-67 (arrow, b ). In contrast, cyclin D1-positive nuclei were weakly present in the majority of proximal tubules (arrows, c ) and mesan-gial cells, whereas p-Rb immunoreactivity was detected in podo-

cytes (arrow, d ) and distal tubules (not shown). In juvenile cystic kidney disease, sample 3, which was from a 22-weeks’ gestation male foetus, showed clusters of small cysts that were strongly posi-tive for p-Rb ( h ) and these also co-expressed Ki-67 (arrows, f ) and cyclin D1 (arrows, g ). In contrast, sample 4, which was from a 14-year-old female subject, cystic renal disease was more advanced, containing much larger cysts and weaker staining (arrows) forKi-67 ( j ), cyclin D1 ( k ) and p-Rb ( l ). Magnification ! 400.

Colo

r ver

sion

ava

ilabl

e on

line


there might be a therapeutic window in which Cdk in-hibitors are most effective in preventing kidney enlarge-ment in cystic renal diseases. In this regard, it would be interesting to verify whether the degree of p-Rb immu-noreactivity in cysts could serve as a surrogate biomarker of treatment response to Cdk inhibitors.

Acknowledgements

This study was supported by funding from the National Health and Medical Research Council of Australia (457575G.K.R./D.C.H.; 38708 J.K.P.).

References

1 Wilson PD: Polycystic kidney disease. N Engl J Med 2004; 350: 151–164.

2 Grantham JJ: Clinical practice: autosomal dominant polycystic kidney disease. N Engl J Med 2008; 359: 1477–1485.

3 Sweeney WE Jr, Avner ED: Molecular and cellular pathophysiology of autosomal reces-sive polycystic kidney disease (ARPKD). Cell Tissue Res 2006; 326: 671–685.

4 Nadasdy T, Laszik Z, Lajoie G, Blick KE, Wheeler DE, Silva FG: Proliferative activity of cyst epithelium in human renal cystic dis-eases. J Am Soc Nephrol 1995; 5: 1462–1468.

5 Evan AP, Gardner KD Jr, Bernstein J: Polypoid and papillary epithelial hyperplasia: a poten-tial cause of ductal obstruction in adult poly-cystic disease. Kidney Int 1979; 16: 743–750.

6 Grantham JJ, Geiser JL, Evan AP: Cyst for-mation and growth in autosomal dominant polycystic kidney disease. Kidney Int 1987; 31: 1145–1152.

7 Wang X, Gattone V, 2nd, Harris PC, Torres VE: Effectiveness of vasopressin V2 receptor antagonists OPC-31260 and OPC-41061 on polycystic kidney disease development in the PCK rat. J Am Soc Nephrol 2005; 16: 846–851.

8 Nagao S, Nishii K, Yoshihara D, Kurahashi H, Nagaoka K, Yamashita T, Takahashi H, Yamaguchi T, Calvet JP, Wallace DP: Calci-um channel inhibition accelerates polycystic kidney disease progression in the cy/+ rat. Kidney International 2008; 73: 269–277.

9 Kumar V, Abbas AK, Fausto N: Tissue re-newal and repair: regeneration, healing and fibrosis; in Kumar V, Abbas AK, Fausto N (eds): Robbins and Cotran Pathologic Basis of Disease. Philadelphia, Elsevier Saunders, 2005, vol 1.

10 Obaya AJ, Sedivy JM: Regulation of cyclin-CDK activity in mammalian cells. Cell Mol Life Sci 2002; 59: 126–142.

11 Vermeulen K, Van Bockstaele DR, Berne-man ZN: The cell cycle: a review of regula-tion, deregulation and therapeutic targets in cancer. Cell Proliferation 2003; 36: 131–149.

12 Malumbres M, Barbacid M: Cell cycle, CDKS and cancer: a changing paradigm. Nat Rev Cancer 2009; 9: 153–166.

13 Kim JK, Diehl JA: Nuclear cyclin d1: an on-cogenic driver in human cancer. J Cell Physi-ol 2009; 220: 292–296.

14 Hwang HC, Clurman BE: Cyclin E in normal and neoplastic cell cycles. Oncogene 2005; 24: 2776–2786.

15 Genovese C, Trani D, Caputi M, Claudio PP: Cell cycle control and beyond: Emerging roles for the retinoblastoma gene family. On-cogene 2006; 25: 5201–5209.

16 Bukanov NO, Smith LA, Klinger KW,Ledbetter SR, Ibraghimov-Beskrovnaya O: Long-lasting arrest of murine polycystic kid-ney disease with CDK inhibitor roscovitine. Nature 2006; 444: 949–952.

17 Park JY, Schutzer WE, Lindsley JN, Bagby SP, Oyama TT, Anderson S, Weiss RH: P21 is decreased in polycystic kidney disease and leads to increased epithelial cell cycle pro-gression: roscovitine augments p21 levels. BMC Nephrol 2007; 8: 12.

18 Alcalay NI, Sharma M, Vassmer D, Chap-man B, Paul B, Zhou J, Brantley JG, Wallace DP, Maser RL, van den Heuvel GB: Accelera-tion of polycystic kidney disease progression in CPK mice carrying a deletion in the home-odomain protein cux1. Am J Physiol Renal Physiol 2008; 295:F1725–F1734.

19 Sugiyama K, Akiyama T, Shimizu M, Tama-oki T, Courage C, Gescher A, Akinaga S: De-crease in susceptibility toward induction of apoptosis and alteration in g1 checkpoint function as determinants of resistance of hu-man lung cancer cells against the antisignal-ing drug UCN-01 (7-hydroxystaurosporine). Cancer Res 1999; 59: 4406–4412.

20 Phillips JK, Hopwood D, Loxley RA, Ghatora K, Coombes JD, Tan YS, Harrison JL, Mc-Kitrick DJ, Holobotvskyy V, Arnolda LF, Rangan GK: Temporal relationship between renal cyst development, hypertension and cardiac hypertrophy in a new rat model of au-tosomal recessive polycystic kidney disease. Kidney Blood Pressure Res 2007; 30: 129–144.

21 Rangan GK, Tesch GH: Quantification of re-nal pathology by image analysis. Nephrology 2007; 12: 553–558.

22 Krajewska M, Smith LH, Rong J, Huang X, Hyer ML, Zeps N, Iacopetta B, Linke SP, Ol-son AH, Reed JC, Krajewski S: Image analy-sis algorithms for immunohistochemical as-sessment of cell death events and fibrosis in tissue sections. J Histochem Cytochem 2009; 57: 649–663.

23 Vogetseder A, Picard N, Gaspert A, Walch M, Kaissling B, Le Hir M: Proliferation ca-pacity of the renal proximal tubule involves the bulk of differentiated epithelial cells. Am J Physiol Cell Physiol 2008; 294:C22–C28.

24 Cowley BD Jr, Chadwick LJ, Grantham JJ, Calvet JP: Elevated proto-oncogene expres-sion in polycystic kidneys of the c57bl/6j (cpk) mouse. J Am Soc Nephrol 1991; 1: 1048–1053.

25 Cowley BD Jr, Smardo FL Jr, Grantham JJ, Calvet JP: Elevated c-myc protooncogene ex-pression in autosomal recessive polycystic kidney disease. Proc Natl Acad Sci USA 1987; 84: 8394–8398.

26 Aaltonen K, Amini RM, Landberg G, Eerola H, Aittomaki K, Heikkila P, Nevanlinna H, Blomqvist C: Cyclin d1 expression is associ-ated with poor prognostic features in estro-gen receptor positive breast cancer. Breast Cancer Res Treat 2009; 113: 75–82.

27 Loden M, Stighall M, Nielsen NH, Roos G, Emdin SO, Ostlund H, Landberg G: The cy-clin d1 high and cyclin e high subgroups of breast cancer: Separate pathways in tumoro-genesis based on pattern of genetic aberra-tions and inactivation of the prb node. On-cogene 2002; 21: 4680–4690.

28 Lindahl T, Landberg G, Ahlgren J, Nordgren H, Norberg T, Klaar S, Holmberg L, Bergh J: Overexpression of cyclin E protein is associ-ated with specific mutation types in the p53 gene and poor survival in human breast can-cer. Carcinogenesis 2004; 25: 375–380.

29 Bhunia AK, Piontek K, Boletta A, Liu L, Qian F, Xu PN, Germino FJ, Germino GG: Pkd1 induces p21(waf1) and regulation of the cell cycle via direct activation of the jak-stat sig-naling pathway in a process requiring pkd2. Cell 2002; 109: 157–168.

30 Li X, Luo Y, Starremans PG, McNamara CA, Pei Y, Zhou J: Polycystin-1 and polycystin-2 regulate the cell cycle through the helix-loop-helix inhibitor ID2. Nat Cell Biol 2005; 7: 1202–1212.

31 McCausland JE, Ryan GB, Alcorn D: Angio-tensin converting enzyme inhibition in the postnatal rat results in decreased cell prolif-eration in the renal outer medulla. Clin Exp Pharmacol Physiol 1996; 23: 552–554.

32 Henry A, Reyes-Mugica M: Autosomal re-cessive polycystic kidney disease associated with nephrogenic rests: a possible conse-quence of arrested renal maturation. Pediatr Dev Pathol 2009: 12:493–494.

Early Cyst Growth Is Associated with the Increased Nuclear Expression of Cyclin D1/Rb Protein in an...

Documents

Transcript of Early Cyst Growth Is Associated with the Increased Nuclear Expression of Cyclin D1/Rb Protein in an...