Epidermal growth factor (EGF) expression in the congenital polycystic mouse kidney

Kidney International, Vol. 39 (1991), pp. 57—62

Epidermal growth factor (EGF) expression in the congenital polycystic mouse kidney SATOSHI H0RIK0sHI, SHuNIcHIR0 KUBOTA, GEORGE R. MARTIN, YOsHIHIK0 YAMADA, and PAUL E. KLOTMAN Laboratory of Developmental Biology and Anomalies, National Institute of Dental Research, NIH, Bethesda, Maryland, USA

Epidermal growth factor (EGF) expression in the congenital polycystic mouse kidney. The mechanisms responsible for renal cyst formation in congenital polycystic kidney disease remain unknown, although abnormalities of cellular metabolism, basement membrane components, and growth factors have been suggested. In the present study, we examined a potential role for epidermal growth factor (EGF) in cyst formation in a mouse model. We measured growth factor activity and concentration in renal cyst fluid, urine, and serum obtained from mice with congenital polycystic kidney disease (cpk). In affected mice, both growth factor activity of urine and the urinary EGF concentration were much lower than unaffected littermate controls even when corrected for creatinine concentration. Although the growth factor activity was much lower in affected mice, there were significant differences in the regional distribution of EGF in animals with cysts. Both growth factor activity and EGF concentration were greater in cyst fluid when compared to urine. Growth factor activity in cysts was completely inhibited by anti-EGF

model, the autosomal recessive congenital polycystic kidney (cpk) mouse [17], to explore the role of growth factors in cyst formation. In this model, affected mice progressively develop large cysts (Fig. 1) which replace renal parenchyma leading to renal failure and death at approximately three weeks of age. In the present study, we measured the ability of serum, urine and

expression of EGF mRNA in kidneys from affected mice was markedly decreased when compared to littermate controls. Theses results suggest that decreased EGF production and local differences in EGF concentration may contribute to cyst formation.

expression by Northern hybridization.

cyst fluid from affected mice and littermate controls to stimulate

parameters of growth using BALB/MK keratinocytes and tubular epithelial cells from mouse kidney. To characterize the growth factor(s) in cyst fluid we performed heparin-Sepharose column chromatography to fractionate growth factor activity and we inhibited growth factor activity using anti-EGF anti-

body. We also measured EGF concentrations in the same

antibody using BALB/MK epithelial keratinocytes as targets. The samples by radioimmunoassay and examined EGF mRNA

Methods

Materials

Cpk mice which arose spontaneously in a branch of the epithelial cell proliferation and fluid accumulation in large cysts [1—3]. The mechanisms responsible for this aberrant growth and cyst formation are not known. Recent reports suggest that both

C57BL/6J inbred line (Jackson Lab.) were used in these experiments. Cyst fluid, serum and bladder urine were obtained from three-week-old affected cpk mice and their littermates. Prior to sacrifice, animals were anesthetized with methoxyflurane (Pit-

the extracellular environment and intracellular factors play important roles. For example, alterations in basement membrane protein production may result in a weakening of the support structure for tubular cells, which could lead to cyst

aspiration. Animals were then sacrificed by cervical dislocation and bladder urine and renal cyst fluid were removed. Cyst fluid obtained from the same animal was pooled as a single sample.

Congenital polycystic kidney disease is characterized by

formation [4—9]. Furthermore, changes in transtubular transport manifested by increases in Na-K ATPase activity may contribute to the accumulation of fluid in cysts. Thus, both an increase in growth as well as an increase in Na-K ATPase are associated with cyst formation [6, 10—16].

man-Moore Inc.) and blood was collected by retro-orbital

All samples were stored individually at —20°C. Cyst fluid samples from several affected animals were pooled for column

chromatography. In addition, kidneys were harvested from two- and three-week-old mice, snap frozen in liquid nitrogen, and stored at —70°C for future RNA extraction.

We hypothesized that growth factors found in cysts are Assay of DNA synthesis in quiescent culture important in producing the disordered epithelial cell growth characteristic of the congenital polycystic kidney. FurtherBALB/MK keratinocytes, provided by Dr. Aaronson (NIH, more, we attempted to identify the specific growth factor(s) NCI Bethesda, Maryland, USA), were grown in minimal essenwhich contributes to cyst formation. We utilized an animal tial medium (MEM) with 0.05 mrvt CaC12 and 10 mg/liter Ca Received for publication January 24, 1990 and in revised form July 3, 1990 Accepted for publication August 13, 1990

pantothenate supplemented with 10% dialyzed fetal calf serum (FCS; Gibco Laboratories, Grand Island, New York, USA) and 4 nglml of EGF (Collaborative Research) [18, 19]. Incorporation of [3H]thymidine was utilized as an index of cell proliferation. BALB/MK cells (2 x 104/well) were grown in 96 well plates

58

Horikoshi et a!: EGF expression in cpk mouse kidney

A

B

Antibody blocking studies To assess the specific role of EGF in stimulating growth, a second set of studies was performed using a polyclonal sheep

A

anti-human EGF (Onco Inc.) as an antagonist. The cross reactivity of anti-EGF antibody to mouse EGF was confirmed •4

I i: r

_3, _4_

••

*

-

by Western blot and immunostaining using normal mouse kidney frozen sections. An identical concentration of non-

,

immunized sheep IgG (Cappel Inc., Cochranville, Pennsylvania, USA) was used as a control. Heparin-Sepharose column chromatography To characterize the growth factor(s) in cyst fluid, heparinSepharose (Pharmacia Diagnostics, Uppsala, Sweden) column

a

4-

--

—

I

'C

Fig. 1. Coronal section of kidneys obtained from 3-week-old unaf-

chromatography (column size 1 ml) was performed. Three milliliters of pooled cyst fluid was applied to the column and washed with 20 mt Tris-HC1 (pH 7.4). The column was then eluted with a linear gradient of 0.2 to 2.0 M NaCl in Tns-HCI (pH 7.4). Each fraction was concentrated five fold by Centrifree MPS-l (Amicon, Lexington, Massachusetts, USA) with a molecular weight cut off of 5 kDa. Concentrate (20 d) was assayed

for biological activity by [3H]thymidine incorporation using BALB/MK keratinocytes as described above.

fected (A) and affected (B) mice. (7.5 X Magnification)

EGF and creatinine concentration Research). When cells became confluent, they were washed The concentration of EGF in serum, urine and cyst fluid of twice with PBS and the medium was changed to 200 d of serum mice was measured using a radioimmunoassay kit (Amersham). free MEM supplemented with 5 ,ttg/ml transferrin and 1 IxM Briefly, antisera to EGF and [12511 EGF were added to diluted selenium. After an additional 48 hours, the medium was samples and incubated for 16 hours at 4°C. A second antibody changed and serum, urine, or cyst fluid were added in a range of (anti-rabbit IgG) was then added and incubated for 10 minutes concentrations. After 15 hours, 4 Ci/ml of [methyl-3H]thymi- at room temperature. After centrifugation for 10 minutes at 1500 dine (Amersham Corp., Arlington Heights, Illinois, USA) was xg, supernatants were removed and the radioactivity incorpoadded to each well and incubated for five hours at 37°C. Wells rated into pellets was measured with a gamma spectrometer were then washed once with ice-cold PBS and twice with (Packard Instruments, Downers Grove, Illinois, USA). Creatiice-cold 5% trichloroacetic acid (TCA). The TCA-insoluble nine concentration was measured in urine and cyst fluid samfraction was solubilized with 0.25 M NaOH, incubated for 30 ples with a modified colorimetric assay using the Jaffé reaction minutes at 37°C, and incorporated radioactivity determined (Sigma Diagnostics, Sigma Chemical Co., St. Louis, Missouri, with a liquid scintillation spectrometer (Beckman Instruments, USA). The concentration of EGF in cyst fluid was normalized Fullerton, California, USA). All assays of {3H]thymidine incor- by creatinine concentration and expressed as ng/mg creatinine. poration were performed in duplicate and assays were repeated Comparisons of cyst fluid and urine from affected and unafat least three times. To account for differences in baseline fected mice were made using the Student's t-test. [3H]thymidine incorporation between assays, counts (cpm/well) were normalized for the positive control values (cells exposed to 10% FCS containing media). Comparisons of growth factor

activities between affected and unaffected mouse urine, and between cyst fluid and urine from affected mice were made using the Student's t-test. Tubular epithelial cells were isolated from two-week-old affected and unaffected mouse kidneys by a modification of methods previously described [20]. The cells were grown in a

RNA extraction and Northern hybridization

Total kidney RNA was prepared from mice two and three weeks old using guanidine-thiocyanate according to Chirgwin et al [211. Total RNA (10 g) was electrophoresed through formaldehyde-agarose gels and transferred to nitrocellulose filters.

The filters were hybridized with the 754 bp SmaI/PvuII fragment (bases 2886-3639) of preproEGF cDNA clone (pmegflO) 1:1 mixture of Dulbecco's Modified Eagle's Medium and Ham's [22, 231 and /3-actin probe labeled with [32PIdCTP by the F12 Medium supplemented with 10 mM Hepes buffer, 10 nM random prime method [24]. After washing, the filter was exposed to X-ray film either for three days (f3-actin probe) or two selenium, 5 ig/ml bovine insulin, 5 g/ml human transferrin, 5 x 10— 12 M triiodothyronine, 5 x i0 M hydrocortisone, and 25 weeks (EGF probe). ng/ml prostaglandin E1. Fifteen thousand renal tubule cells Results were plated in each of 96 wells coated with I ig/cm2 of human fibronectin. After the addition of cyst fluid or mouse EGF Kidneys of polycystic mice were markedly enlarged by three (Collaborative Research), [3H]thymidine incorporation assays weeks of age (Fig. 1). Renal parenchyma was largely replaced were performed by the same methods described above for by expanding cysts in affected mice. In contrast, unaffected littermates had normal kidney size and function. BALB/MK keratinocytes.

59

Horikoshi et a!: EGF expression in cpk mouse kidney 0.14 C

0

0.12

6

2.0

5

1.0

4

0.8

Co

0 C

0

CO

0.08 >.

I

I

0.06

0.6

3

C

a)0

E > .CU

0C

C)

0.10

a)

00 00

z a) CD

a

0.4

2

0

0.04 0

30

20

10

Fraction no. 0.02

0

Fig. 3. Growth factor activity of fractions eluted from a heparin0.5

1.0

2.0

Sepha rose column. The peak biological activity did not bind to hepann. The gradient of elution is indicated on the right ordinate. Details of the chromatography are described in methods.

Concentration of mitogens % Volume of rn/to gen/media 110

Fig. 2. Effect ofurine and cyst fluid on [3H]thymidine incorporation by

BALB/MK keratinocytes. The percent of the total volume of media constituted by urine and cyst fluid is indicated on the abscissa and the

100

counts/mm of [3H] incorporated in cells (expressed as the fraction of the

90

counts obtained with 10% FCS) are plotted on the ordinate. The maximum number of cpm/well for cells exposed to 10% FCS was 155,570 49,532 cpm. Bars represent the mean standard error of the mean for urine from affected mice (; N = 7), cyst fluid (D; N = 11), and urine from unaffected littermates (; N = 4). The asterisks indicate < 0.005. < 0.01, the level of significance where * < 0.05,

80 70 C

0

60

Co

E

The bioassay of growth factor activity in urine and cyst fluid as measured by {3H]thymidine incorporation of BALB/MK keratinocyte target cells is illustrated in Figure 2. Urine from cpk mice contained significantly less (30 to 50%) growth factor activity than that of unaffected littermates. Despite this marked reduction in EGIF activity in the urine of animals with cysts, significant differences in the distribution of EGF activity were found in cyst fluid and urine of affected animals. Growth factor activity of cyst fluid was significantly greater than that of urine from affected animals. As shown in Figure 3, the active fraction of the cyst fluid was contained in a single peak which did not bind to heparin-Sepharose. This result suggested that most of the growth factor activity in cyst fluid did not bind to heparin as would be expected for transforming growth factor beta (TGF/3), fibroblast growth factor (FGF), or platelet-derived growth factor (PDGF). In order to explore the possibility that the majority of growth factor activity in cyst fluid was due to EGF, we used anti-EGF

antibody as an inhibitor in vitro. As indicated in Figure 4, anti-EGF antibody inhibited 98% of the growth factor activity in cyst fluid. In contrast, anti-EGF antibody inhibited only 30% of

the growth factor activity found in normal mouse serum (data not shown). Renal function of cystic animals is markedly impaired by two to three weeks. Therefore, we measured creatinine concentra-

50

C/)

40

30 20 10

0 —10

0

1:25000

1:5000

1:1000

1:200

Antibody dilution

Fig. 4. [3H]thymidine incorporation of BALB/MK cells in response to cyst fluid in the presence of sheep anti-EGF IgG antibody (0). In each well, media contained 2% by volume cyst fluid. Antibody dilution is indicated on the abscissa and [3Hlthymidine incorporation is indicated on the ordinate expressed as a % of control without antibody (100%). The absolute value of [3H]thymidine incorporation for the cells exposed to 2% cyst fluid without antibody was 5782 949 cpm. The mitogenic activity of 2% cyst fluid was determined in the presence of the same concentration and dilution of normal sheep IgG (•) and the mitogenic activity of anti-EGF antibody was determined in the absence of cyst fluid (A). Points represent the mean SEM of three experiments.

of normal littermates (Fig. 5). However, similar to the results obtained for growth factor activity, the concentration of EGF/ tion in the same cyst fluid and urine samples in which the creatinine in cyst fluid was significantly greater than that of concentration of immunoreactive EGF was determined. In urine. In order to determine whether the growth factor in cyst fluid urine of affected mice, EGF/Cr was significantly less than that

60

Horikoshi el al: EGF expression in cpk mouse kidney

provides a good model for the evaluation of potentially important modifying factors. In the cpk mouse, changes in extracellular matrix components, cell proliferation, and transepithelial transport of solute and water likely contribute to the development of cysts. For example, an increase in the expression of mRNA of basement

*

70 60 -

0

membrane components has been reported in cpk mice [9]. Various chemical agents, such as the antioxidants diphenylamine (DPA) [25], nordihydroguaiaretic acid (NDGA) [1] and 2-amino-4,5-diphenylthiazole (DPT) [4, 5, 26], and hormones, such as triiodothyronine [15] and hydrocortisone [27] have been reported to cause tubule cell hyperplasia and cyst formation in rats. In addition to these effects, DPT decreases tubule proteoglycan synthesis, one of the major basement membrane components [6, 71. Studies using organ or cell culture systems

50 —

x .2

40 — E

suggest that increased Na-K ATPase activity and increased intracellular cyclic AMP concentration induce cyst formation [6, 12, 15, 28, 29] and Na concentration influences kidney epithelial cell growth in culture [11, 14]. Furthermore, EGF activates Na transport in fibroblasts and has been implicated

H

*

10

8

in breast cyst formation as well [30—32]. Therefore, we hypoth-

6 4 2 0

Affected urine

N=8

Affected cyst fluid

N=8

Unaffected

urine

N=8

Fig. 5. EGF concentration of urine and cyst fluid measured by radio-

esized that alterations in growth factor activity, particularly EGF, might be important in cyst formation since EGF stimulates both cell growth and Na/K transport. In the present study, we found that urine from affected mice had profoundly reduced growth factor activity when compared to unaffected littermates. However, cyst fluid had much higher

growth factor activity than urine from affected mice. The

growth factor activity which we measured using BALB/MK mouse epithelial keratinocytes as targets stimulated renal epiAsterisk indicates significance where *13 < 0.001. thelial cell growth as well. In order to better define this factor, we characterized the major growth factor activity in cyst fluid using heparin-Sepharose column chromatography. The majorwas biologically active for renal cells, we evaluated the effect of ity of growth factor activity for BALB/MK epithelial keraticyst fluid on DNA synthesis of tubular epithelial cells (Fig. 6). nocytes was found in a single peak which did not bind to Kidney epithelial cells isolated from normal and cystic mice heparin. Furthermore, growth factor activity of this peak was were capable of responding to the growth factors in cyst fluid as almost completely inhibited by anti-EGF antibody, suggesting well as to authentic EGF in a manner similar to that of that EGF was the predominant growth factor for epithelial cells BALB/MK keratinocytes. The maximal stimulation of [3H]thy- in cyst fluid. In order to confirm this possibility, we measured midine incorporation occurred when 1% of the medium by EGF concentration in cyst fluid, urine and serum in both volume was composed of cyst fluid. Although there were affected and unaffected mice. These values were normalized for marked and significant differences in urinary growth factor creatinine concentration since cystic animals have markedly activity and EGF/creatinine concentration between affected impaired renal function and, as a result, likely do not concenand unaffected littermates, the differences in serum growth trate urine normally. In affected animals, the urinary EGFI factor activity and serum EGF concentration were not signifi- creatinine of affected mice was much less than normal littermates. The ratio of EGF/creatinine concentration in cyst fluid cant (Fig. 7). The expression of mRNA for preproEGF in affected and was much greater than that of urine from affected animals. unaffected littermate mouse kidneys was analyzed by Northern These results paralleled the findings observed for growth factor hybridization (Fig. 8). The preproEGF mRNA from affected activities as measured by bioassay. The reason for the increase mouse kidneys was markedly less than that from unaffected in cyst fluid EGF/creatinine concentration in affected animals littermates at both two and three weeks. In contrast, beta actin when compared to urine is unclear, but probably reflects local mRNA was increased in affected mice when compared to sites of production and accumulation. unaffected controls at three weeks, probably reflecting the To evaluate whether the decrease in EGF was due to a reduction in gene expression, whole kidney RNA was probed marked proliferative state of cyst formation [9, 34]. immunoassay. Data were normalized by creatinine concentration of each sample. Bars represent mean standard error of the mean.

Discussion

for the presence of EGF mRNA. Affected animals had a marked

reduction in EGF message when compared to normal littermates. Normally, preproEGF mRNA expression increases in formation, particularly in the autosomal dominant forms of the kidney for the two weeks following birth [23, 33]. While polycystic kidney disease. While this may be less true in the there was a slight increase in preproEGF mRNA in cpk mice autosomal recessive forms of the disease, the cpk mouse between two and three weeks, the level was markedly reduced Both genetic and environmental factors are important for cyst

61

Horikoshi et a!: EGF expression in cpk mouse kidney

2 6.0

7.0 B 6.0

0o

5.0

5.0

4.0

4.0

3.0

3.0

7.0 C

U)

0

.E

0

2.0

LI

1.0

1.0

0

2.0

4.0

6.0

Concentration of EGF, ng/mI

1.0

0.0

2.0

3.0

4.0

Concentration of cyst fluid, % B

A

0.20

1.4 1.2

C

0

0.15

00 .4-

cc E° >-

.cLL

I

Fig. 6. Mitogenic effects of EGF and cyst fluid using tubular epithelial target cells obtained from normal (0) and cystic (•) mouse kidneys. Various amounts of EGF and cyst fluid were added to the media and TCA-insoluble incorporation of [3Hlthymidine was measured as described in methods. The counts/mm of [3H]thymidine incorporated into cells are expressed as a relative value compared to control (without mitogens). The absolute values for {3H]thymidine incorporation for affected and unaffected mouse cells in the absence of mitogen were 361 cpm and 1,313 cpm, respectively.

0.10

c0 1.0 c 0.8

0.05

C 00 0.6 U0w 0.4

21

Fig. 7. A. [3H]thymidine incorporation assay of affected () and unaffected (•) mouse serum using BALB/MK keratinocyte target cells. The volume percent of serum in media is plotted on the abscissa and counts of incorporated [3H] normalized for the control (relative value in cpm/ well for [3H]thymidine incorporation exposed to 10% FCS containing media) are plotted on the ordinate. B. EGF concentration in serum of affected and unaffected mice measured by

0.2

C0

0

0 1 2 0.5 % Volume of serum/media

Affected

serum N=5

when compared to unaffected littermates. These results also paralleled the findings observed for EGF concentration as measured by radioimmunoassay. Gattone et al reported defective preproEGF expression in cpk mice by immunohistochemical study and Northern hybridization [34] and our data also strongly suggest that EGF regulation is impaired in the cpk model. While EGF/creatinine concentration in urine is reduced

Unaffected serum

N=

radioimmunoassay.

11

2Ws

I U

II A

2Ws U

A

A

— EGF (4.9 Kb)

in cpk mice, EGF accumulating locally in renal cysts may act as

a mitogen for renal tubular epithelial cells and may stimulate Na-K ATPase. It appears likely that the presence of EGF is required for normal renal development and that decreased EGF production

may contribute to renal cyst formation. As a result, local differences in EGF concentration may determine the specific location and distribution of cysts which form in the kidney. However, it is conceivable that the response of epithelial cells from cystic animals is impaired as well. Carone et al reported that cyst-derived epithelial cells obtained from human autosoma! dominant polycystic kidneys do not respond to EGF to the same extent as normal cells [35]. However, in the present study, tubule epithelial cells from affected and unaffected mice responded to EGF and cyst fluid similarly. The differences between these findings and those of Carone et al may be due to species differences as well as differences in the genetics of the disease. In addition, the cell types responsible for cyst formation are not clear. Taub et al reported that renal tubule epithelial cells prepared from the cpk mouse model contained abundant

B

C

3-actin (2.0 Kb)

'—EtBr(28S)

Fig. 8. Northern hybridization of EGF and f3-actin. A. EGF mRNA

expression of 2- and 3-week-old affected mice and normal littermates. After hybridization with a cDNA probe for EGF, washed filters were exposed for 14 days. B. The same filter was hybridized with a p-actin cDNA probe. C. Ethidium bromide staining of the gel.

alkaline phosphatase activity, suggesting a proximal tubule origin [36]. However, Preminger et al [17] and Gattone et al [37]

have reported that renal cyst formation in the mouse model occurs first in proximal tubules and then involves the collecting

62

Horikoshi ci a!: EGF expression in cpk mouse kidney

duct secondarily. As a result, different cell types may be involved at different stages of the disease. In summary, cyst formation is associated with marked cellular proliferation. Although EGF is markedly reduced in affected

mice, the local accumulation of EGF in cysts may play an important role in cyst formation. Furthermore, the marked changes in EGF regulation may be critical for the disorganized growth pattern typical of the cpk mouse model. Acknowledgments BALB/MK keratinocytes used in this study was a gift from Dr. Stuart Aaronson. The authors thank Mr. Stuart Cohn for technical assistance and Drs. Hynda Kleinman, Jeffrey Kopp, and Patricio Ray for careful review of the manuscript.

ical modification of cell proliferation and fluid secretion in renal cysts. Kidney mt 35:1379—1389, 1989 17. PREMINGER GM, KOCHWE, FRIED FA, MCFARLAND E, MURPHY ED, MANDELL J: Murine congenital polycystic kidney disease: A model for studying development of cystic disease. J Urol 127:556— 560,

1982

BE, AARONSON SA: BALB and Kirsten murine sarcoma viruses alter growth and differentiation of EGF-dependent BALB/c mouse epidermal keratinocyte lines. Cell 32:599—606, 1983

18. WEISSMAN

19.

FIORE PPD, FALCO J, BORRELLO I, WEISSMAN B, AARONSON SA:

The calcium signal for BALB/MK keratinocyte terminal differentiation counteracts epidermal growth factor (EGF) very early in the EGF-induced proliferative pathway. Mo! Cell Biol 8:557—563, 1988 20. TAUB M, SATO G: Growth of functional primary cultures of kidney epithelial cells in defined medium: J Cell Physiol 105:369—378, 1980 21. CHIRGWIN JM, PRZYBYLA AE, MACDONALD Ri, RUTTER WJ:

Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18:5294—5299, 1979

Reprint requests to Paul E. Kiotman, M.D., Chief, Molecular Medicine Section, Laboratory of Developmental Biology and Anomalies, National Institute of Dental Research/NIH, 9000 Rockville Pike, Bethesda, Maryland 20892, USA.

References

22. RALL LB, SCOTT J, BELL GI: Mouse prepro-epidermal growth factor synthesis by the kidney and other tissues: Nature 313:228— 231, 1985 23. SALIDO EC, YEN PH, SHAPIRO U, FISHER DA, BARAJAS U: In situ

hybridization of prepro-epidermal growth factor mRNA in the mouse kidney. Am J Physiol 256:F632—F638, 1989 24. FEINBERG AP, VOGELSTEIN B: A technique for radiolabelling DNA

restriction endonuclease fragments to high specific activity. Anal 1. EVAN AP, GARDNER KD: Nephron obstruction in nordihydroguaiaretic acid-induced renal cystic disease. Kidney Int 15:7—19, 1979 2. EVAN AP, GARDNER KD, BERNSTEIN J: Polypoid and papillary

Biochem 132:6—13, 1983 25. GARDNER KD, SOLOMON 5, FITZGERREL WW, EVAN AP: Function

and structure in the diphenylamine-exposed kidney. J Clin Invest

epithelial hyperplasia: A potential cause of ductal obstruction in adult polycystic disease. Kidney mt 16:743—750, 1979 3. GRANTHAM JJ, GEISER JL, EVAN AP: Cyst formation and growth in autosomal dominant polycystic kidney disease. Kidney Int 31:1145—

57:796—806, 1976

26. GARDNER KD, EVAN AP: Renal cystic disease induced by diphenylthiazole. Kidney mt 24:43—52, 1983 27. AVNER ED, PlEsco NP, SWEENEY WE, STUDNICKI FM, FETTER-

MAN GH, ELLIS D: Hydrocortisone-induced cyst metanephric maldevelopment in serum-free organ culture. Lab Invest 50(2):208—

1152, 1987

4. KANWAR YS, CAR0NE FA: Reversible changes of tubular cell and basement membrane in drug-induced renal cystic disease. Kidney 5.

mt 26:35—43, 1984 BUTKOWSKI RJ, CARONE FA, GRANTHAM JJ, HUDSON BG: Tubu-

lar basement membrane changes in 2-amino-4,5-diphenylthiazolinduced polycystic disease. Kidney mt 28:744—751, 1985 6. WILSON PD, SCHRIER RW, BRECKON RD, GABOW PA: A new

method for studying human polycystic kidney disease epithelia in culture. Kidney liii 30:371—378, 1986 7. LELONGT B, CARONE FA, KANwAR YS: Decreased de novo synthesis of proteoglycans in drug-induced renal cystic disease. Proc Nail Acad Sci USA 85:9047—9051, 1988 8. CARONE FA, MAKINO H, KANWAR YS: Basement membrane antigens in renal polycystic disease. Am J Pathol 130:466—471, 1988

2l8, 1984 28. AVNER ED, SWEENEY WE, ELLIS D: cyst

In vitro modulation of tubular

regression in murine polycystic kidney disease. Kidney mt

36:960—968, 1989

29. MANGOO-KARIM R, UcHic ME, GRANT M, SHUMATE WA, CALVET JP, PARK CH, GRANTHAM JJ: Renal epithelial fluid secretion

and cyst growth: The role of cyclic AMP. FASEB J 3:2629—2632, 1989 30. MOOLENAAR WH, YARDEN Y, DE LAAT SW, SCHLESSINGER J:

Epidermal growth factor induces electrically silent Na influx in human fibroblasts. J Biol Chem 257:8502—8506, 1982 31. MOOLENAAR WH, TSIEN RY, VAN DER SAAG PT, DE LAAT SW:

Na/K exchange and cytoplasmic pH in the action of growth

9. EBIHARA I, KILLEN PD, LAURIE GW, HUANG T, YAMADA Y, MARTIN GR, BROWN KS: Altered mRNA expression of basement

factors in human fibroblasts. Nature 304:645—648, 1983 32. SMITH K, MILLER WR, FENNELLY JA, MATTHEWS JNS, SCOTT

membrane components in a murine model of polycystic kidney disease. Lab Invest 58:262—269, 1988 10. CARONE FA, HOLLENBERG PF, NAKAMURA S, PUNYARIT P, GL0-

WN, HARRIS AL: Quantification of epidermal growth factor in human breast cyst fluids: Correlation with dehydroepiandrosterone-sulphate and electrolyte concentrations. liii J Cancer 44:229—

GOWSKI W, FLOURET G: Tubular basement membrane change occurs pan passu with the development of cyst formation. Kidney

232, 1989 33. GUBITS RM, SHAW PA, GRESIK EW, ONETTI-MUDA A, BARKA T:

Epidermal growth factor gene expression is regulated differently in mouse kidney and submandibular gland. Endocrinology 119:1382—

Int 35:1034—1040, 1989 11.

TOBACK FG: Induction of growth in kidney epithelial cells in culture by Nat Proc Nat! Acad Sci USA 77:6654—6656, 1980

12. AVNER ED, SWEENEY WE, FINEGOLD DN, PlEsco NP, ELLIS D:

Sodium-potassium ATPase activity mediates cyst formation in metanephric organ culture. Kidney Int 28:447—455, 1985 13. PERR0NE RD: In vitro function of cyst epithelium from human polycystic kidney. J Clin Invest 76:1688—1691, 1985

1387, 1986

34. GATTONE II VH, ANDREWS GK, FU-WEN N. CHADWICK Li, KLEIN RM, CALVET JP: Defective epidermal growth factor gene

expression in mice with polycystic kidney disease. Devop Biol 138:225—230, 1990

14. WALSH-REITZ MM, GLUCK SL, WAACK S, TOBACK FG: Lowering

35. CARONE FA, NAKAMURA 5, SCHUMACHER BS, PUNYARIT P, BAUER KD: Cyst-derived cells do not exhibit accelerated growth or

extracellular Na concentration releases autocrine growth factors

features of transformed cells in vitro. Kidney Int 35:1351—1357,

from renal epithelial cells, Proc Nat! Acad Sci USA 83:4764—4768, 1986

15. AVNER ED, SWEENEY WE, PlEsco NP, ELLIS D: Triiodothyronine-induced cyst formation in metanephric organ culture: The role of increased Na-K-adenosine triphosphatase activity. J Lab Clin Med 109:441—453, 1987 16. GRANTHAM JJ, Uc-iIc M, CRAGOE EJ, KORNHAUS J, GRANTHAM JA, DoNoso V, MANGOO-KARIM R, EVANA, MCATEER J: Chem-

1989

36. TAUB M, LAURIE GW, MARTIN GR, KLEINMAN HK: Altered basement membrane protein biosynthesis by primary cultures of cpklcpk mouse kidney. Kidney Int 37:1090—1097, 1990 37.

GATTONE 11 VH, CALVET JP, COWLEY BD, EVAN AP, SHAVER TS,

HELMSTADTER K, GRANTHAM JJ: Autosomal recessive polycystic

kidney disease in murine model. A gross and microscopic description. Lab Invest 59:23 1—238, 1988