JOURNAL OF CELLULAR PHYSIOLOGY 148:295-305 (19911

Regulation of Rat Proximal Tubule Epithelial Cell Growth by Fibroblast Growth Factors, Insulin-Like Growth Factor-I and Transforming Growth Factor+, and Analysis of Fibroblast Growth Factors in Rat Kidney GUOHONG ZHANG, TAKAHARU ICHIMURA, ALF WALLIN, M l K l O KAN, AND JAMES 1. STEVENS* The W. Alton ]ones Cell Science Center, Lake Placid, New York 12946

Growth factors may play an important role in regulating the growth of the proximal tubule epithelium. To determine which growth factors could be involved, we have investigated the mitogenicity of various purified factors in rat kidney proximal tubule epithelial (RPTE) cells cultured in defined medium. Fibroblast growth factors, aFGF (acidic FGF) and bFGF (basic FGF), stimulate DNA synthesis in a dose-dependent manner, with ED50 values of 4.5 and 3 . 2 ng/ml, respectively; their effects are not additive. With cholera toxin in the medium, both aFGF and bFGF can replace insulin or epidermal growth factor (EGF) to attain the maximum level of cell growth, but they cannot replace cholera toxin. Cholera toxin specifically potentiates the effects of FGFs on DNA synthesis. At high cell density, both insulin and insulin-like growth factor 1 (IGF-1) induce DNA synthesis more effectively than EGF, FGFs and cholera toxin. The high concentration (0.2-1 .O Fg/ml) of insulin required for cell growth can be replaced by a low concentration of IGF-1 (10-20 ng/ml), indicating that insulin probably acts through a low affinity interaction with the IGF-1 receptor. Transforming growth factor-pl (TGF-PI) inhibits DNA synthesis induced by individual factors and combinations of factors in a concentration-dependentmanner. Northern blot analysis shows that mRNA for TGF-Pl, IGF-1, and aFGF, but not bFGF are present in rat kidney. Western blot analysis and bioassay data confirmed that the majority of FGF-like protein in rat kidney is aFGF. The data suggest that in addition to EGF, IGFs, and TGF-P, FGFs may also be important kidney-derived regulators of proximal tubule epithelial cell growth in vivo and in vitro.

Growth of proximal tubule epithelial cells occurs during physiological processes, such as renal development (Ekblom 1989; Bacallao and Fine, 19891, and in pathophysiological processes, such as repair of damage to the tubular epithelium (Laurent e t al., 1988; Bacallao and Fine, 1989). Indeed, it has been suggested that there are clear parallels between growth during development and tubular regeneration after renal damage (Bacallao and Fine, 1989; Wallin et al., in press). However, the mechanisms which control proximal tubule epithelial cell growth in vivo are not clear. It has been suggested that polypeptide growth factors, some of which are derived from the kidney, may play a n important role in controlling cell proliferation and hypertrophy in the kidney under normal and pathological conditions (for reviews see Patt and Houck, 1983; Laurent et al., 1988; Mendley and Toback, 1988; Bacallao and Fine, 1989; Segal and Fine, 1989). Risau and Ekblom (1986) have suggested that fibroblast growth factors may regulate aspects of kidney organogenesis and both acidic and basic FGFs 0 1991 WILEY-LISS, INC.

(aFGF and bFGF; respectively; also called heparin binding growth factors 1 and 2; Burgess and Maciag, 1989) have been isolated from kidney tissue (GautschiSova et al., 1987; Baird et al., 1985). After unilateral nephrectomy there is a n increase in insulin-like growth factor-1 mRNA (IGF-1; Stiles et al., 1985; Fagin and Melmed, 1987) and immunoreactive IGF-1 (Andersson et al., 1988) in the hypertrophic remnant kidney. There is also a n increase in IGF-1 in the cortex during repair of ishemic damage to the S3 segment of the proximal tubule (Andersson and Jennische, 1988). Tsau et al. Received May 30, 1990; accepted April 3, 1991. *To whom reprint requestskorrespondence should be addressed. Abbreviations used basic fibroblast growth factor, bFGF; acidic fibroblast growth factor, aFGF; epidermal growth factor, EGF; transforming growth factor, TGF; insulin-like growth factor, IGF; rat kidney proximal tubule epithelial cells, RPTE; bovine serum albumin, BSA; human umbilical cord vein endothelial cells, HUV-EC; dithiothreitol, DTT.

296

ZHANG ET AL.

(1989) and Coimbra et al. (1990) have reported that infusion of EGF enhances the regeneration of proximal tubules after ischemic acute renal failure. TGF-p, which is predominantly a growth inhibitor for epithelial cells, has been isolated from bovine kidney (Roberts et al., 1985) and is produced by renal epithelial cells (Fine et al., 1985). Thus, physiologically important growth factors produced in the kidney may act as bipolar effectors to exert positive and negative effects on proximal tubule epithelial cell growth. Growth factors play a critical role in vitro in the growth of proximal tubule cells from several species (Chung et al., 1982; Taub, 1985; Hatzinger and Stevens, 1989). Both EGF and IGF-1 have been shown to be mitogens for proximal tubule epithelial cells in culture (Chung et al., 1982; Norman et al., 1987; Hatzinger and Stevens, 1989; Gander et al., 1990). In addition, Kanda et al. (1989) have shown that FGFs are mitogens for rabbit renal cortical epithelial cells. However, it is not clear whether or not these requirements in vitro reflect the physiological requirements for growth factors during growth in vivo. Indeed, it has even been suggested that given an appropriate extracellular matrix, RPTE cells will grow in the absence of growth factor stimulation (Rosenberg and Michalopoulos, 1987). Rat kidney proximal tubule epithelial (RPTE) cells cultured in defined medium are a valuable tool for evaluating biochemical mechanisms which control cell proliferation and differentiation because of the large data base accumulated in this species. The purpose of this investigation was to determine which growth factors might be potential regulators of RPTE growth in vivo by testing their activity as mitogens for RPTE cells in primary culture. We investigated the roles of aFGF and bFGF, insulin-like growth factor-1 (IGF-l), and transforming growth factor-pl (TGF-p1) as mitogens for primary cultured RPTE cells. The role of EGF as a proximal tubule mitogen has received considerable attention elsewhere (Taub, 1985; Mendley and Toback, 1989; Bacallao and Fine, 1989).TGF-P1 was found to be a potent growth inhibitor. IGF-1 replaced insulin as a mitogen for RPTE in agreement with studies in proximal tubule epithelial cells from other species (Gander et al., 1990). However, the activity of IGF-1 and insulin was markedly stronger than any other factors tested on high density cells. FGFs were found to be potent mitogens for RPTE in culture. Abundant message and protein for aFGF, but not bFGF, were both present in rat kidney tissue. The data suggest that FGFs could be important regulators of RPTE growth in vivo.

ng/ml), EGF (10 ng/ml), and insulin (10 pg/ml) unless specified otherwise. DMEMIF12 was supplied by Gibco (Grand Island, NY). Cholera toxin was added in accordance with our previous study, in which cholera toxin was determined to be a n effective growth promoter for RPTE as has been reported for some other epithelial cells (Hatzinger and Stevens, 1989). Insulin, cholera toxin, penicillin-G, and streptomycin sulfate were purchased from Sigma Chemical Co. (St.Louis, MO). EGF, aFGF, and bFGF were purchased from Upstate Biotechnology (Lake Placid, NY). IGF-1 was supplied by Imcera Bioproducts, Inc. (Terre Haute, IN). TGF-P1 was purchased from Collaborative Research, Inc. (Bedford, MA). Other chemicals were purchased from commercial sources.

Preparation of rat kidney proximal tubule epithelial cells Male Sprague-Dawley rats (Taconic Farms, Germantown, NY) weighing 150-250 g were used for the preparation of RPTE. The purified proximal tubule fragments were first isolated by a method previously described (Hatzinger and Stevens, 19891, with modification to the buffers as described (Zhang and Stevens, 1989). Tubular fragments were treated further by shaking with either trypsin (0.05% trypsin, Gibco, Grand Island, NY) for 10 min or collagenase (270 unitdm1 of Type-11; Worthington Biochemical Co., Freehold, NJ) for 30 min at 37°C under a n atmosphere of 5% co2:95% 02.The dissociated tubular fragments were then homogenized 10 strokes in a 7 ml Dounce homogenizer with a very loose pestle (greater clearance than a B type Potter-Elvejhem pestle) and collected by centrifugation a t 850 rpm for 3 min. The resulting pellet of small clusters and single cells was washed twice with DMEM/F12 by centrifugation. Based on protein assays (Bio-Rad, Richmond, CA) 15-30 mg of cells can be prepared. The viability of the cells was 85%, as determined by the trypan blue exclusion test.

Cell cultures Cells were seeded in the basal medium described above in 12-well dishes (Costar, Cambridge, MA) coated with 10 pg/ml collagen (type-I, Collaborative Research; Bedford, MA) at a density of 100-150 pg cellular protein per well and allowed to attach overnight. In some experiments, cells were inoculated into dishes which were coated with both collagen and fetal bovine serum (FBS, Gibco, Grand Island, NY) as described by Rosenberg and Michalopoulos (1987). Cultures were maintained at 37°C in a n atmosphere of 5% co2:95% air and were fed every two to three days. The cultures have been characterized for proximal MATERIALS AND METHODS tubular and epithelial character. The epithelial charCulture medium and supplements acter of the cultures has been confirmed previously by The basal culture medium (DMEMIF12) was a 1 :l positive keratin staining (Hatzinger and Stevens, mixture of Dulbecco’s Modified Eagle Medium (DMEM) 1988). In addition, the cultures have y-glutamyltransand Ham’s Nutrient Mixture F12 (F12) with penicillin- ferase activity, alkaline phosphatase activity, and G, streptomycin sulfate (PS: 50IU/ml each), and bovine sodium-dependent glucose transport characteristic of serum albumin (BSA; Sigma Chem. Co., St. Louis, MO) proximal tubule epithelial cells (data not shown; also as previously described (Hatzinger and Stevens, 19891, see Hatzinger and Stevens, 1989). except that BSA fraction V was used instead of a Growth and DNA synthesis determinations complex of fatty acid free BSA and oleic acid. For routine growth and maintenance of RPTE, the basal Cell growth was determined by [3H]thymidine incormedium was supplemented with cholera toxin (10 poration, DNA mass measurement or cell number. In

GROWTH FACTORS AND KIDNEY EPITHELIAL CELLS

general [3H]thymidine incorporation was used for experiments in which mitogenic potential was determined over a short time period (24 hr). DNA mass or cell number was used for long term growth experiments (6-10 days). Factors were added to cultures either in basal medium or basal medium supplemented with insulin, EGF, and cholera toxin as indicated in individual experiments. L3H]Thymidine (specific activity: 20 Ci/mmol, New England Nuclear, Boston, MA) incorporation into the trichloroacetic acid-insoluble materials was determined by adding L3H1thymidine (0.25 pCi/ml) on day 4 and measuring the incorporation 24 h r later as described (Freshney, 1987). When I3H1thymidine incorporation was performed on high density cells, the factors were first removed for 48 hr, before individual factors were added back to the cultures. [3H]Thymidine incorporation was determined over the next 24 hr. Increases in DNA mass were determined from cells first solubilized in buffer containing 20 mM Tris.HC1 (pH 8.0),1 mM EDTA, and 0.1% Triton X-100. DNA mass was then determined fluorometrically using Hoechst dye 33258 as reported (Downs and Wilfinger, 1983).Values for DNA mass and 13H]thymidine incorporation were compared in triplicate experiments with cell number increases determined by electronic particle counting (Coulter Electronics, Inc. Hialeah, FL). In the presence or absence of insulin, EGF, FGFs, or cholera toxin, equivalent growth stimulation was achieved by using any of the methods (data not shown).

Detection of growth factor expression in rat kidney Rat kidneys were decapsulated and immediately frozen in sterile 50 ml conical tubes placed in liquid nitrogen. The kidneys were pulverized to a fine powder with a mortar and pestle chilled in liquid nitrogen. The kidney powder was immediately mixed with RNAzol (Cinna/Biotex Laboratories, Friendswood, TX), and total RNA isolated according to the manufacturer’s instructions. Preparation of poly (A)’ RNA was carried out by one cycle of oligo-dT-Sepharose chromatography. After electrophoresis and blotting to nitrocellulose, Northern analysis was carried out using cDNA probes for rat aFGF (Goodrich et al., 1989), bovine bFGF (Abraham et al., 1986), human TGF-Pl (Derynck et al., 19851, and rat IGF-1 (Murphy et al., 1987). Analysis of fibroblast growth factor activity in rat kidneys Human umbilical cord vein endothelial cells (HUVEC) were used for bioassays of FGFs and FGF-like activity (Maciag et al., 1981; Gospodarowicz et al., 1985). HUV-EC were prepared and established in primary cultures in T25 flasks as described (Kan et al., 1985). Bioassays were done using cells between the third and seventh population doublings. Population doublings were counted after establishing the initial cultures in T25 flasks. Therefore, a doubling of HUVEC from the original T25 flask would be considered one population doubling. HUV-EC were seeded at a density of 1 X lo4 cells/well in 24-well dishes and maintained overnight in 1 ml of RITC 80-7 medium (Kan and Yamane, 1982) containing 2% fetal bovine serum and

297

100 pg/ml kanamycin. Samples were added to the individual wells in 1 ml of RITC 80-7 medium containing 10% fetal bovine serum and 1mg/ml kanamycin to prevent contamination from crude samples. The higher kanamycin concentration had no effect on the response of HUV-EC cells to bFGF. Cell number was determined by electronic particle counting 5 days later. FGF-like activity was quantitated by comparing the growth stimulation of HUV-EC produced by the unknown to standard curves prepared with known amounts of aFGF (0.1-10 ng/ml). A standard curve was included for each experiment. Bioassay of FGF-like activity in kidney tissue was determined after extraction and concentration on heparin-sepharose CL-6B (Pharmacia, Upsalla, Sweden). Kidneys were homogenized in 5 volumes of extraction buffer containing 20 mM Tris (pH 7.4), 10 mM EDTA, 0.2 mM D,L-dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride (added in a minimal volume of ethanol immediately before use), 0.5 kg/ml leupeptin, and 0.1% 3-[(3-~holamidopropyl)dimethylammoniol-l-propanesulfonate (CHAPS; Bio Rad, Richmond, CAI. Solid sodium chloride was added to 0.35 M and the sample was centrifuged 7 min a t 6000g. The 0.35 M salt in the buffer facilitates extraction of FGFs but does not prevent FGF binding to heparin-sepharose. The pellet was extracted with 2 ml of extraction buffer and the supernatants were combined and centrifuged for 1 h r at 28,OOOg. When Western blots were performed, the FGF-like activity in the extract was concentrated on heparinsepharose and the complex boiled directly in sample preparation buffer. Western blotting was carried out using standard techniques and rabbit antibodies specific for aFGF, a generous gift from Dr. T. Maciag (American Red Cross), and bFGF, a generous gift from Dr. E. Kardami (University of Manitoba). In some experiments, FGF-like activity bound to heparin-sepharose was recovered by stepwise elution with increasing amounts of salt. Heparin-sepharose (0.3 ml) was packed i n a disposable column (Bio Rad, Richmond, CA) and washed with 10 ml of Dulbecco’s phosphate buffer saline (PBS),10 ml of PBS containing 1 mg/ml bovine serum albumin (BSA), 10 ml of 2.5 M NaC1, and 10 ml of 0.5 M NaCl in water. The kidney extract was applied to the column and the first effluent was repeatedly cycled through the column over a 2 hr period. After washing with 15 ml of 0.5 M sodium chloride in water, FGF-like activity was eluted by stepwise addition of 0.5 ml aliquots of 1.0, 1.5, 2.0, and 2.5 M NaCl in water and 0.5 ml fractions were collected. FGF-like activity in the fractions was assayed directly after adding BSA (1 mg/ml) and DTT (1mM), final concentrations. Crude kidney extracts contained growth inhibitory substances which interfered with the bioassay and prevented direct quantitation of FGF-like activity, making it difficult to determine recovery from the heparin-sepharose column. Recovery of aFGF was determined using lZ5I-labeled aFGF, pregared as described (Kan et al., 1991). An aliquot of [l 511aFGF was added to a kidney extract (prepared as described above) and applied to the column. Of the [12511aFGFapplied to the column, 68%bound, and 86% of the bound material eluted at 1.5 M NaCl or less.

ZHANG ET AL.

298

IC-IEC 6000

T

C+IEC (C+S)-IEC (C+S)+IEC

4000

2000

0 day 6

day 8

day 10

Fig. 1. Culture of rat kidney proximal tubule epithelial cells requires growth factors. Rat kidney proximal tubule epithelial cells were grown on collagen type I (C) or collagen type I plus serum (C+S) coated dishes in basal medium as described in Methods. Growth was determined by measuring the total DNA mass in the presence (+IEC) or absence (-IEC) of insulin (I), EGF (E) and cholera toxin (C) after 6

days in culture (see Methods for concentrations). The data are the mean t standard deviation of three separate experiments on cultures from different animals (n = 3). Significant differences (P4 ng/ml), cells began to detach from the cultured dishes (data not shown). Therefore, TGF-P1 is a potent but nonspecific inhibitor of RPTE growth.

Factor

Heparin

Cells X 10-5

% of Control

none none aFGF aFGF bFGF bFGF

-

0.54 0.48 0.73 2.50 2.50 1.87

100 89 135 463 463 346

+

-

+ -

+

'Rat kidney proximal tubulecells were preparedasdescribed in Methodsand seeded a t a density of 150 pg of tubular protein in 12-well dishes. Cells were grown for six days in the presence or absence of fibroblast growth factors (10 ng/ml) before cell number was determined by electronic particle counting. Heparin was included in some cultures a t a concentration of 10 pg/ml. The data are from a representative experiment.

Effects of fibroblast growth factors on DNA synthesis and replacement of insulin and epidermal growth factor Because FGFs have not been reported previously to be mitogens for proximal tubule epithelial cells, we further characterized their activity in order to determine if FGFs act as discrete factors or in concert with other factors by similar mechanisms. To test these possibilities we performed subtraction experiments in which a single component was replaced by either aFGF or bFGF and the effects on total DNA mass were determined (Fig. 4). When aFGF or bFGF was added to the supplemented medium containing insulin, EGF, and cholera toxin, there was only a small increase in DNA synthesis. However, aFGF (Fig. 4A) and bFGF (Fig. 4B) replaced either insulin or EGF in the presence of cholera toxin, but could not replace cholera toxin in the presence of both insulin and EGF. Although FGFs appear to replace IGF-1 and insulin at low cell density, they do not have the same effect as IGF-1 and insulin on higher density cultures (Fig. 2). The data in Figure 4 suggested that cholera toxin might play a n important role in the ability of FGFs to replace insulin and EGF. To determine if cholera toxin and FGFs act synergistically, we tested the effects of various combinations of factors on [3H]thymidine incorporation. Each of the factors was added either individually or in combination. The [3H]thymidine incorporation data from the individual factors were added to yield the theoretical additive value (individual; Fig. 5) and were compared with the data when factors were added together (combination; Fig. 5). The data in Figure 5 clearly show that only the combination of cholera toxin and bFGF acted synergistically. Similar effects were observed by combining cholera toxin with aFGF (data not shown).

Analysis of fibroblast growth factors in rat kidney If the kidney produces factors which are involved in the growth and differentiation of proximal tubule epithelial cells in vivo, they must be expressed in renal tissue. Therefore, we probed poly (A)+RNA with cDNA probes for TGF-P1, IGF-1, aFGF, and bFGF. All the messages were easily detectable with the exception of bFGF (Fig. 6). The bovine bFGF probe used (Abraham

GROWTH FACTORS AND KIDNEY EPITHELIAL CELLS

301

Insulin

insulin

EGF

+

+

+

+

Cholera Toxin

EGF

Cholera Toxin

EGF

Fig. 3. Transforming growth factor pl inhibits growth stimulation by individual factors. Cells were grown in the presence or absence of TGF-pl (1 ngiml) with various combinations of factors (see Methods for concentrations). L3H1Thymidine was added on day 4 and incorpo-

et al., 1986) recognizes rat bFGF mRNA (Mansson et al., 1989). Western blot analysis and bioassays of FGFs in kidney extracts supported the RNA data. When the FGFs in kidney extracts were concentrated onto heparin-sepharose and the bound protein analyzed by Western blotting, there was an intense band for aFGF, but there was little immunoreactivity for bFGF (Fig. 7). Interestingly, the weak reactivity with bFGF was predominantly at 31.5 Kd, consistent with reports that bFGF can exist as a dimer under SDS and reducing conditions (Sato et al., 1989). We were able to confirm the Western blot data with bioassay data using a combination of stepwise elution from heparinsepharose and analysis of the heparin-dependence of the FGF-like activity. The data in Table 3A show that only aFGF required heparin for activity with HUV-EC in agreement with others (Burgess and Maciag, 1989). bFGF activity was inhibited slightly by addition of heparin. Bioassay results using HUV-EC showed that 86% of the HBGF-like activity eluted from heparinsepharose in 1.5M NaCl or less consistent with the elution of aFGF. This FGF-like material was dependent on the presence of heparin for activity with HUV-EC (Table 3B). Apparently, the majority of the FGF-like activity in rat kidney is due to aFGF in agreement with the mRNA data.

Insulin

+

Cholera Toxin

ration determined 24 hr later. The data are the mean i the standard deviation of three separate experiments (n = 3). An asterisk indicates a significant decrease (P < 0.05) when TGF-Pl was added. Significance was determined using Student’s t-test.

trolled by growth factors which act as autocrine or paracine regulators of cell growth (Laurent et al., 1988; Mendley and Toback, 1989; Bacallao and Fine, 1989). Previous results suggest that EGF may be an important stimulator of proximal tubule epithelial cell growth following nephrotoxic damage (Tsau et al., 1989, Coimbra et al., 1990).Others have suggested that IGF-1 synthesized in the distal tubule may stimulate the growth of the proximal tubule epithelial cells during regeneration (Andersson and Jennische, 1988; Hammerman, 1989). EGF is filtered and excreted in the urine and IGF-1 is present in plasma in a tight complex with IGF-1 binding proteins (for reviews see Underwood et al., 1986; Bacallao and Fine, 1989; Mendley and Toback, 1989; Hammerman, 1989). Therefore, it is possible that these factors are available to stimulate proximal tubule epithelial cell growth in vivo. To our knowledge, members of the FGF family have not previously been shown to be mitogens for rat proximal tubule epithelial cells. Our data are in agreement with the data of Kanda et al. (1989), who showed that FGFs are mitogens for rabbit cortical epithelial cells. Unlike EGF and IGF-1, aFGF and bFGF are not present as circulating factors, but are stored either inside cells or tightly bound to the extracellular matrix (for reviews see Burgess and Maciag, 1989; Goldfarb, 1990). Although there are other members of the family DISCUSSION which do have signal sequences, they are limited to The growth of proximal tubule epithelial cells during expression during embryogenesis or in selected tissues tubular regenerations or development may be con- in the adult. Thus, aFGF and bFGF are the most

302

ZHANG ET AL.

Fig. 4. Replacement of insulin and epidermal growth factor by fibroblast growth factors. Cells were grown in medium supplemented with various combinations of insulin (I), epidermal growth factor (El, and cholera toxin ( C ) (see Methods for concentrations). Each factor (I, E, or C) was replaced by aFGF (A) or bFGF (B) added a t a concen-

tration of 10 ngiml. Cells were harvested after 6 days and DNA mass was determined. The data are the mean 2 the standard deviation of three separate experiments (n = 3). Significant differences (P < 0.05) were determined using a one-way analysis of variance. Means with different letter designations are significantly different (see Methods).

abundant FGFs in adult tissue. Since neither aFGF or bFGF has an appropriate signal peptide for secretion, it is suggested that they are released from cells or the basement membrane matrix upon wounding and proteolysis, whereupon they stimulate the growth and migration of target cells (Folkman and Klagsbrun, 1987; Gajdusek and Carbon, 1989; Sato and Rifkin, 1988; McNeil et al., 1989). The majority of data on the role of FGFs in wounding pertains t o repair of the vascular endothelial cells, where they are believed to play a major role. However, the fact that FGFs are mitogens for a broad range of cells of both mesodermal and ectodermal origin suggests that they may play a role in the growth of other cells in vivo (Burgess and Maciag, 1989; Goldfarb, 1990). For example, aFGF has

been implicated in liver regeneration following partial hepatectomy (Kan et al., 1989). There is little information on the role of FGFs in renal function. However, FGFs have been isolated from embryonic and adult kidney tissue (Baird et al., 1985; Risau and Ekblom, 1986; Gautschi-Sova et al., 19871,as well as renal cell carcinomas (Nakamoto et al., 1988) and nephroblastomas (Witte et al., 1989).During organogenesis, FGFs may be important in inducing the differentiation of the metanephric mesenchyme (Risau and Ekblom, 1986). This is of particular interest given the parallels between tubular development and regeneration (Bacallao and Fine, 1989; Wallin et al., in press). Our Western and Northern blot results indicate that aFGF may be the most abundant FGF in rat

303

GROWTH FACTORS AND KIDNEY EPITHELIAL CELLS

e

F C

L

Insulin

Insulin

EGF

Insulin

EGF

Cholera Toxin

EGF

Cholera Toxin

bFGF

bFGF

+

+

Fig. 5. Synergistic interaction of cholera toxin and fibroblast growth factors on DNA synthesis. Cells were grown in basal medium or in the presence of individual factors or various combinations of factors; insulin (10 pg/ml), EGF (10 ng/ml), cholera toxin (10 ngiml), and bFGF (10 ngiml). r3H1Thymidine was added at day 4 and incorporation determined 24 hr later. The values designated Individual represent the theoretical additive value determined by adding the I3H1thymidine incorporation data from the separate cultures which contained the individual factors. The values designated Combination

Fig. 6. Northern analysis of kidney poly (A)' RNA. The data show Northern blot analysis of kidney mRNA (5 kg) using 32P-labeled cDNA probes for individua! factors (see Methods). After autoradiography, the bands were cut from photographs of the autoradiograms and aligned for comparison. Message sizes detected were as follows: aFGF, 4.4 kb; IGF-1, 7.0 kb; TGF-pl, 2.5 kb.

kidney, in agreement with the results from bovine kidney (Gautshi-Sova et al., 1987). The abundance of aFGF message in rat kidney agrees with the observations for human kidney (Wang et al., 1989). Further studies should elucidate the sites of synthesis of aFGF in the kidney. This is of particular interest since FGFs replace EGF in culture. EGF has received increasing attention as a possible regulator of nephrogenic repair (Tsau et al., 1989; Coimbra et al., 1990). The synergy between FGFs and cholera toxin suggests that increased cellular cAMP potentiates the effects of FGFs. Recently, Lee et al. (1989) sequenced a

+

+

+

Cholera Toxin

+

bFGF

represent the 13Hlthymidineincorporation when the indicated factors were added to cultures in combination. Both values have been corrected by subtracting the incorporation observed in the absence on any added factors and thus represent the specific incorporation due to the growth factors. The data are the mean of three separate experiments (n = 3).Significant differences (P< 0.05) between Combination and Individual means for each treatment were determined by Student's t-test. An asterisk indicates a significant difference.

Fig. 7. Western blot analysis of fibroblast growth factors in rat kidney extracts. FGFs were extracted from kidney homogenates and concentrated by adsorption on to heparin-sepharose as described in Methods. After boiling the heparin-sepharose in sample preparation buffer, the proteins were separated by SDS-gel electrophoresis and probed by Western blot with antibody specific for aFGF and bFGF. Standard FGFs were also included as controls. (A) Blot probed with aFGF antibody; Lane 1-30 ng aFGF, Lane 2-87 pg kidney extract. (B)Blot probed with bFGF antibody; Lane 1-20 ng bFGF, Lane 2-87 pg kidney extract. The aFGF standard contains two molecular weight forms of aFGF due to truncation of the higher molecular weight form (Burgess and Maciag, 1989).

cDNA clone for the chicken FGF receptor and found an intracellular tyrosine kinase domain. Mioh and Chen (1987) reported an increase in cAMP after treating smooth muscle cells with aFGF. It is possible that there is an interaction between the two second messenger

ZHANG ET AL.

304

TABLE 3. Stimulation of endothelial cell growth by rat kidney fibroblast mowth factor in the Dresence and absence of heparin] A. Heparin-dependence of fibroblast growth factor activity Heparin Cell number X 10-5 aFGF (10 ng/ml) bFGF (10 ng/ml)

-

++

0.12 f 0.05 1.0 0.10 1.5 f 0.28 0.73 0.07

*

*

B. Heparin-dependenceof fibroblast growth factor-activity in heparinsepharose eluates of rat kidney extracts Apparent FGF Heparin recovered (pg) Heparinsepharose eluate 0.48 3.55 of kidney extract

+

I The dependence of purified FGFs (A) or FGF-like activity purified from rat kidney extracts by Heparin-Sepharose chromatography (B) on the addition of heparin (10 fig/ml). Activity was determined using human umbilical cord vein endothelial cells (HUV-EC)and the bioassay conditions described inMethods.In part B, therecovery of FGF from kidney extracts represents the pooled recoveryfrom six 0.5 ml fractions derived from four 1 M NaCl and two 1.5 M NaCl washes(seeMeth0ds).The apparent FGF recovery in pg was quantitated by assaying each fraction in duplicate in the presence or absence of heparin and comparing the HUV-EC growth data with a standard curve constructed using aFGF (see Methods). Therefore, the data in the absenceofheparin representthe apparent decreaseinrecovery due tolossof activity. The sum of the results obtained for the individual fractions are shown. Of the activity recovered, 86%was eluted below 1.5 M NaCI.

systems, e.g., receptor-associated tyrosine kinase and CAMP, at some point in the signalling pathways for FGF-induced mitogenesis. The lack of synergy between cholera toxin and either EGF or IGF could be accounted for if the substrates for the receptor tyrosine kinases differ among the receptor (Kyriakis and Avruch, 1990). The effect of IGF-1 on RPTE is also worthy of further comment. It is possible that a decrease in receptor number could account for the decrease in responsiveness a t high cell density. The number of EGF receptors in a renal epithelial cell line, BSC-1, has been shown to be tenfold higher at low cell density than a t high density without a n apparent change in receptor affinity (Holley et al., 1977). Behrens et al. (1989) have shown that the number of EGF receptors increases following folate nephrotoxicity in the rat. Alternatively, higher density cells might make related factors in sufficient quantity to obviate the need for exogenous EGF or FGF. Further investigation is necessary to address this point. The low mitogenic index in normal adult renal tissues suggests that a negative feedback system could exist to limit cell growth. TGF-P1 is present in kidney (Roberts et al., 1983) and our data show that TGF-pl is a potent inhibitor of DNA synthesis in cultured RPTE cells. Therefore, TGF-P1 may act as a n important component of a feedback system to regulate renal cell proliferation similar to the situation in regenerating liver (Fausto and Mead, 1989). Following partial hepatectomy, TGF-P1 increases as regenerating liver cells begin to slow their growth, suggesting th at TGF-p1 may play a role in limiting the growth (Fausto and Mead, 1989). The role of TGF-P1 in hepatic fibrosis is already firmly established (Czaja et al., 1989). In conclusion, kidney-derived growth factors may play a n important role i n the repair of damage to the proximal tubule epithelium. However, the mechanisms by which these factors might regulate nephrogenic repair await further experimentation. The present

investigation defines a selected group of factors, including IGFs, FGFs, and TGF-p1, which may be important mediators of kidney epithelial cell growth. In vivo these factors could play a n important role in physiological and pathophysiological processes such as kidney development or tubular regeneration.

ACKNOWLEDGMENTS This work was supported by grants CA48197 and CA38925 from the National Cancer Institute and grant ES05670 from the National Institute of Environmental Health Sciences. Additional funding was received from the W. Alton Jones Foundation. LITERATURE CITED Abraham, J.A., Mergia, A., Whang, J.L., Tumolo, A., Friedman, J., Hjerrild, K.A., Gospodarowicz, D., and Fiddes, J.C. (1986) Nucleotide sequence of a bovine clone encoding the angiogenic protein, basic fibroblast growth factor. Science, 233.545-548. Andersson, G., and Jennische, E. 11988) IGF-1 immunoreactivity is expressed by regenerating renal tubular cells after ischaemic injury in the rat. Acta Physiol. Scand., 132t453-457. Andersson, G.L., Skottner, A,, and Jennische, E. (1988) Immunocytochemical and biochemical localization of insulin-like growth factor 1 in the kidney of rats before and after uninephrectomy. Acta Endocrinol., 119t555-560. Bacallao, R., and Fine, L.G. (1989) Molecular events in the organization of renal epithelium: From nephrogenesis to regeneration. Am. J. Physiol., 257tF913-F924. Baird, A., Esch, F., Bohlen, P., Ling, N., and Gospodarowicz, D. (1985) Isolation and partial characterization of a n endothelial cell growth factor from the bovine kidney: Homology with basic fibroblast growth factor. Regul. Pept., 12t201-213. Barnes, D., and Sato, G. (1980) Serum-free cell culture: A unifying approach. Cell, 22:649-655. Behrens, M.T., Corbin, A.L., and Hise, M.K. (1989) Epidermal growth factor receptor regulation in rat kidney: Two models of renal growth. Am. J . Physiol., 257tF1059-Fl064. Burgess, W.H., and Maciag, T. (1989)The heparin-binding (fibroblast) growth factor family of proteins. Annu. Rev. Biochem., 58:575-606. Centrella, M., McCarthy, T.L., and Canalis, E. (1987) Transforming gowth factor p is a bifunctional regulator of replication and collagen synthesis in osteoblast-enriched cell cultures from fetal rat bone. J. Biol. Chem., 26.2.2869-2874. Chung, S.D., Alavi, N., Livingston, D., Hiller, S.,andTaub, M. (1982) Characterization of primary rabbit kidney cultures that express proximal tubule functions in a hormonally defined medium. J. Cell. Biol., 95tl18-126. Coimbra, R., Cieslinski, D.A. and Humes, H.D. (1990) Epidermal growth factor accelerates renal repair in mercuric chloride nephrotoxicity. Am. J . Physiol., 259:F438-F443. Czaja, M.J., Weiner, F.R., Flanders, K.C., Giambrone, M.-A., Wind, R., Biempica, L., and Zern, M.A. (1989) In vitro and in vivo association of transforming growth factor-pl with hepatic fibrosis. J. Cell. Biol., 108:2477-2482. Derynck, R., Jarret, J.A., Chen, E.Y., Eaton, D.H., Bell, J.R., Assoian, R.K., Roberts, A.B., Sporn, A.B., and Goeddel, D.V. (1985) Human transforming growth factor-p complementary DNA sequence and expression in normal and transformed cells. Nature, 316.701-705. Downs, T.R., and Wilfinger, W.W. (1983) Fluorometric quantification of DNA in cells and tissue. Anal. Biochem., 131t538-547. Ekblom, P. (1989) Developmentally regulated conversion of mesenchyme to epithelium. FASEB J., 3.2141-2150 (1989). Fagin, J.A., and Melmed, S. (1987) Relative increase in insulin-like growth factor-1 messenger ribonucleic acid levels in compensatory renal hypertrophy. Endocrinology, 120:71&724. Fausto, N., and Mead, J.E. (1989) Regulation of liver growth: Protooncogenes and transforming growth factors. Lab. Invest., 60.4-13. Fine, L.G., Holley, R.W., Nasri, H., and Badie-Dezfooly, B. (1985) BSC-1 growth inhibitor transforms a mitogenic stimulus into a hypertrophic stimulus for renal proximal tubular cells. Relationship to Na+-H’ antiport activity. Roc. Natl. Acad. Sci. U.S.A., 82:61636166. Folkman, J., and Klagsbrun, M. (1987) Angiogenic factors. Science, 235t442-447. Freshney, R.I. (1987) “Culture of Animal Cells. A Manual of Basic Technique,” 2nd edition. Alan R. Liss, Inc., New York, pp. 236-237.

GROWTH FACTORS AND KIDNEY EPITHELIAL CELLS Gajdusek, C.M., and Carbon, S. (1989) Injury-induced release of basic fibroblast growth factor from bovine aortic endothelium. J . Cell. Physiol., 139:570579. Gander, T., Hsu W.-C., Gramling, S., Robinson, K.A., Buse, M.G., Blocker, N., Roy, L., Green, S., Garvin, A.J., and Sens, D.A. (1990) Growth factor binding and bioactivity in human kidney epithelial cell cultures. In Vitro Cell. Dev. Biol., 261285-290. Gautschi-Sova, P., Jiang, Z.-P., Frater-Schroder, M., and Bohlen, P. (1987)Acidic fibroblast growth factor is present in nonneural tissue: Isolation and chemical characterization from bovine kidney, Biochemisty, 26.58444847. Goldfarb, M. (1990) The fibroblast growth factor family. Cell Growth Diff., 1:439-445. Goodrich, S.P., Yan, G.-C., Bahrenburg, K., and Mansson, P.-E. (1989) The nucleotide sequence of rat fibroblast growth factor 1 (aFGF). Nucleic Acid Res., 173867. Gospodarowicz, D., Massoglia, S., Cheng, J., Ge-Ming, L., and Bohlen, P. (1985) Isolation of pituitary fibroblast growth factor by fast protein liquid chromatography (FPLC): Partial chemical and biological Characterization. J. Cell. Physiol., 122:323-332. Hammerman, M.R. (1989) The growth hormone-insulin-like growth factor axis in kidney. Am. J. Physiol., 257:F503-F514. Hatzinger, P.B., and Stevens, J.L. (1988) Alterations in intermediate filament proteins in rat kidney proximal tubule epithelial cells. Biochem. Biophys. Res. Commun. 157:1316-1322. Hatzinger, P.B., and Stevens, J.L. (1989) Rat kidney proximal tubule cells in defined medium: The roles of cholera toxin, extracellular calcium, and serum in cell growth and expression of y-glutamyltransferase. In Vitro Cell. Develop. Biol., 25:205-212. Holley, R.W., Armour, R., Baldwin, J.H., Brown, K.D., and Yeh Y.-C. (1977)Density-dependent regulation of growth of BSC-1 cells in cell culture: Control of growth by serum factors. Proc. Natl. Acad. Sci. U.S.A., 745046-5050. Jetten, A.M., Shirley, J.E., and Stoner, G. (1986) Regulation of proliferation and differentiation of respiratory tract epithelial cells by TGF-p. Exp. Cell Res., 167539-549. Kan, M., and Yamane, I. (1982) In vitro proliferation and lifespan of human diploid fibroblasts in serum-free BSA-containing medium. J . Cell. Physiol., 111:155-162. Kan, M., Huang, J., Mansson, P.E., Yasumitsu, H., Carr, B., and McKeehan, W.L. (1989) Heparin-binding growth factor type 1 (acidic fibroblast growth factor): A potential biphasic autocrine and paracrine regulator of hepatocyte regeneration. Proc. Natl. Acad. Sci. U.S.A., 86t7432-7436. Kan, M., Kato, M., and Yamane, I. (1985) Long-term cultivation and growth requirements for human umbilical vein endothelial cells. In Vitro Cell. Dev. Biol., 21:181-188. Kan, M, Shi, E., and McKeehan, W. (1991) Identification and assay of fibroblast growth factor receptor. Methods Enzymol., 198:(in press). Kanda, S., Nomata, K., Saha, P.K., Nishimura, N., Yamada, J., Kanatake, H., and Saito, Y.(1989) Growth factor regulation of the renal cortical tubular cells by epidermal growth factor, insulin-like growth factor-I, acidic and basic fibroblast growth factor, and transforming growth factor-p in serum free culture. Cell Biol. Int. Rep., 13:687-699. Kyriakis, J.M, and Avruch, J . (1990) Insulin, epidermal growth factor, and fibroblast growth factor elicit distinct patterns of protein tyrosine phosphorylation in BC3Hl cells. Biochim. Biophys. Acta, IO54:73-82. Laurent, G., Toubeau, G., Heuson-Stiennon, J.A., Tulkens, P., and Maldague, P. (1988) Kidney tissue repair after nephrotoxic injury: Biochemical and morphological characterization. CRC Crit. Rev. Toxicol., 19:147-183. Lee, P.L., Johnson, D.E., Cousens, L.S., Fried, V.A., and Williams, L.T. (1989) Purification and complementary DNA cloning of a receptor for basic fibroblast growth factor. Science, 245:57-60. Maciag, T., Hoover, G.A., Stemerman, M.B., and Weinstein, R. (1981) Serial propagation of human endothelial cells in vitro. J. Cell. Biol., 91t420-426. Mansson, P.-E., Adams, P., Kan, M. and McKeehan, W.L. (1989) Heparin-binding growth factor gene expression and receptor characteristics in normal rat prostate and two transplantable rat prostate tumors. Cancer Res., 49:2485-2494. Massague, J., Cheifetz, S., Endo, T., and Nadal-Ginard,B. (1986)Type p, transforming growth factor is a n inhibitor of myogenic differentiation. Proc. Natl. Acad. Sci. U.S.A., 83t8206-8210. McNeil, P.L., Muthukrishnan, L., Warder, E., and D’Amore, P.A. (1989) Growth factors are released by mechanically wounded endothelial cells. J. Cell. Biol., 109t811-822. Mendley, S.R., and Toback, F.G. (1989) Autocrine and paracrine

305

regulation of kidney epithelial cell growth. Annu. Rev. Physiol., 51:33-50. Mioh, H., and Chen, J.-K. (1987) Acidic fibroblast growth factor transiently activates aterial smooth muscle cells. Biochem. Biophys. Res. Commun., 146t771-776. Murphy, L.J., Bell, G.I., Duckworth, M.L., and Friesen, H.G. (1987) Identification, characterization, and regulation of a rat complementary deoxyribonucleic acid which encodes insulin-like growth factorI. Endocrinology, 121:684-691. Nakamoto, T., Usui, A,, Oshima, K., Ikemoto, H., Mitani, S.,and Usui, T. (1988) Analysis of growth factors in renal cell carcinoma. Biochem. Biophys. Res. Commun. 153:818-824. Norman, J., Badie-Dezfooly, B., Nord, E.P., Kurtz, 11, Schlosser, J., Chaudhari, A,, and Fine, L.G. (1987) EGF-induced mitogenesis in proximal tubular cells: Potentiation by angiotensin 11. Am. J . Physiol., 253:F299-F309. Patt, L.M., and Houck, J.C. (1983) Role of polypeptide growth factors in normal and abnormal growth. Kidney Int., 23:603-610. Risau, W., and Ekblom, P. (1986) Production of heparin-binding angiogenesis factor by the embryonic kidney. J . Cell Biol., 103:llOl-1107. Rizzino, A. (1988) Transforming growth factor-p: Multiple effect on cell differentiation and extracellular matrices. Dev. Biol., 130:411422. Rizzino, A,, Kazakoff, P., Ruff, E., Kuszynski, C., and Nebelsick, J . (1988) Regulatory effects of cell density on the binding of transforming growth factor p, epidermal growth factor, platelet-derived growth factor, and fibroblast growth factor. Cancer Res., 48t42664271. Roberts, A.B., Anzano, M.A., Wakefield, L.M., Roche, N.S., Stern, D.F.. and Suorn. M.B. (1985) Tvue B transformine mowth factor: A bifunctionai regulator of cellifar ’growth. Procy fiatl. Acad. Sci. U.S.A., 82r119-123. Roberts, A.B., Anzano, M.A.. Mevers, C.A., Wideman, J.,Blacher, R.. Pan, Y.X.E., Stein, S., Lehrman, R., Smith, J.M., Lamb, L.C., and Sporn, M.B. (1983) Purification and properties of a type beta transforming growth factor. Biochemistry, 22 5692-5698. Rosenberg, M.R., and Michalopoulos, G. (1987) Kidney proximal tubular cells isolated by collagenase perfusion grow in defined media in the absence of growth factors. J . Cell. Physiol., 131,107113. Sato, Y., and Rikin, D.B. (1988) Autocrine activities of basic fibroblast growth factor: Regulation of endothelial cell movement, plasminogen activator synthesis, and DNA synthesis. J. Cell Biol., 107: 1199-1205. Sato, Y., Murphy, P.R., Sato, R., and Friesen, H.G. (1989) Fibroblast growth factor release by bovine endothelial cells and human astrocytoma cells in culture is density dependent. Mol. Endocrinol., 3t744-748. Segal, R., and Fine, L.G. (1989) Polypeptide growth factors and the kidney. Kidney Int., 3 6 (suppl. 27): S2-S10. Sporn, M.B., Roberts, A.B., Wakefield, L.M., and de Crombrugghe, B. (1957) Some recent advances in the chemistry and biology of transforming growth factor-beta. J . Cell B i d , 105:1039-1045.. Stiles, A.D., Sosenko, I.R.S., DErcole, A.J., and Smith, B.T. (1985) Relation of kidney tissue somatomedin-C/insulin-like growth factor-1 to postnephrectomy renal growth in the rat. Endocrinology, 117:2397-2401. Taub, M. (1985) In: Methods for Serum-Free Culture of Epithelial and Fibroblastic Cells. M. Taub, ed. New York: Alan R. Liss, Inc., pp. 3-24. Tsau, Y.K., Norman, J.T., and Fine, L.G. (1989) Epidermal growth factor (EGF) enhances renal regeneration and accelerates recovery from ischemic acute renal failure. Kidney Int., 35:420 (abstract). Underwood, L.E., DErcole, A.J., Clemmons, D.R., and Van Wyk, J . J . (1986) Paracrine functions of somatomedins. Baillieres Clin. Endocrin. Metab., 1559-77. Wallin, A,, Zhang, G., Jaken, S., and Stevens, J.L. (in press) Nephroadministragenic repair following S-(1,2-dichlorovinyl)-L-cysteine tion. Lab. Invest. Wang, W.-P., Lehtoma, K., Varban, M.L., Krishnan, I., andChiu, I.-M. (1989) Cloning of the gene coding for human class I heparin-binding growth factor and its expression in fetal tissues. Mol. Cell Biol., 9:2387-2395. Witte, D.P., Nagasaki, T., Stambrook, P., and Lieberman, M.A. (1989) Identification of an acidic fibroblast growth factor-like activity in a mesoblastic nephroma. Lab. Invest., 60:353-359. Zhang, G.H., and Stevens, J.L. (1989) Transport and activation of S-(1,2-dichloroviny1)-L-cysteine and N-acetyl-S(1,2-dichlorovinyl)L-cysteine in rat kidney proximal tubules. Toxicol. Appl. Pharmacol., 100t51-61.