0013-7227/91/1296-3240$03.00/0 Endocrinology Copyright © 1991 by The Endocrine Society
Vol. 129, No. 6 Printed in U.S.A.
Gonadal and Adrenal Effects on Hepatic Epidermal Growth Factor Receptor Expression in a Murine Model* BERT SCOCCIA, PETER KOVAR, AND ROBERT BENVENISTE Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Humana Hospital-Michael Reese; University of Illinois College of Health Sciences (B.S., P.K.); and Illinois Institute of Technology (R.B.), Chicago, Illinois 60616
ABSTRACT. We hypothesize that the actions of epidermal growth factor (EGF) may be modulated by changes in cell surface EGF receptor (EGF-R) expression under endocrine influences. Mouse liver cell membrane preparations were used in a RRA. During ontogenesis, both sexes showed a significant increase {P < 0.005) in hepatic EGF-R numbers at puberty; however, males demonstrated significantly higher levels than females (P < 0.005). Gonadectomy of adult males and females resulted in a significant (P < 0.05) decrease and increase, respectively, in hepatic EGF-R expression. Prepubertal gonadectomy in both sexes resulted in EGF-R levels similar to those observed in adult females. Adrenalectomy of adult animals of both sexes had no effect on hepatic EGF-R numbers, but gonadectomy plus adrenalectomy virtually obliterated EGF-R expression. Short term treatment with testosterone of adult females or gonadectomized
E
PIDERMAL growth factor (EGF) is a polypeptide with a mol wt of 6045 that induces a variety of cellular effects, including cell proliferation and differentiation (1). EGF shares a significant sequence homology with transforming growth factor-a (TGFa), and both bind to the EGF receptor (EGF-R) (2). The EGF-R has a mol wt of 170,000 and is a transmembrane glycoprotein that includes a cytoplasmic domain with tyrosine kinase activity (3). EGF-R has a significant sequence homology with the v-erb-B oncogene protein (4), and EGF-R is often overexpressed in transformed cells (5), lending support to the belief that EGF and the EGF-R play an important regulatory function in cells. Despite a large volume of information on the cellular effects of EGF and the acute biological responses elicited Received July 22,1991. Address requests for reprints to: Bert Scoccia, M.D., Department of Obstetrics and Gynecology, K-100, Humana Hospital-Michael Reese, Lake Shore Drive at 31st Street, Chicago, Illinois 60616. * Presented in part at the 36th Annual Meeting of the Society for Gynecologic Investigation, San Diego, CA, 1989 (Abstract 16), and at the 71st Annual Meeting of The Endocrine Society, Seattle WA, 1989 (Abstract 1034). This work was supported in part by the Michael Reese Medical Research Institute Council and NIH Biomedical Research Support Grant SO7-RR-05476 (to B.S.).
female and male mice significantly increased EGF-R numbers (P < 0.05) to adult male levels. 17/?-Estradiol given short term to adult males or gonadectomized male and female mice did not significantly alter EGF-R levels. EGF-R expression in androgeninsensitive male mice was significantly reduced (P < 0.005) to female levels. We conclude that 1) hepatic EGF-R numbers increase postpubertally in both sexes; 2) hepatic EGF-R expression is significantly stimulated by testosterone, and this effect depends on a functional androgen receptor; 3) the ovary has an inhibitory effect on adult hepatic EGF-R numbers; however, this effect does not appear to be mediated by estrogens; and 4) the adrenal gland has a stimulatory effect on adult hepatic EGF-R expression. (Endocrinology 129: 3240-3246,1991)
by EGF administration in vivo (6), the physiological role of EGF is not clearly understood. This gap is due in part to the difficulty in documenting that a change in circulating EGF levels precedes a normal physiological event (7, 8). One possibility is that EGF acts locally in a paracrine or autocrine fashion (9, 10). Alternatively, we hypothesize that the physiological role of EGF may depend on target tissue changes in the expression of EGFR, which may be modulated by endocrine influences. To study this hypothesis further, we have used a rodent liver model to examine, under physiological conditions, changes in EGF-R expression. EGF-R are highly expressed in rodent livers, and both EGF and TGFa play an important role in liver growth and maintenance (11). We have previously shown that there is a sexual dimorphism in the expression of hepatic EGF-R in cell membrane preparations (12) and at the cell surface (13). Most dramatic is the approximately 50-fold difference in EGF-R numbers observed between freshly plated male and female hepatocytes (13). The objectives of the present study are to determine 1) at which developmental stage the expression of EGFR becomes sexually dimorphic; 2) what effect prepubertal and adult gonadectomy, alone or in conjunction with adrenalectomy, has on EGF-R expression; 3) what is the
3240
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:56 For personal use only. No other uses without permission. . All rights reserved.
HEPATIC EGF-R EXPRESSION effect oftestosterone and 17/?-estradiol supplementation on intact and gonadectomized male and female mice; 4) what is the expression of EGF-R in livers from androgeninsensitive male mice.
Materials and Methods Animals and treatments Intact, adrenalectomized, or sham-operated BALB-c male and female mice (5-10 weeks old) were purchased from Harlan Laboratory (Indianapolis, IN). Newborn pups were obtained from in-house breeding and used at 2-4 weeks of age. Gonadectomies and sham operations were performed under inhalational anesthesia (methoxyflurane, Pittman-Moore, Washington Crossing, NJ) either at 2 (prepubertal) or 10 (adult) weeks of age. Adult 10-week-old androgen-insensitive mice (Tfm/Y, C57BL/6J-Aw-j-Ta) and normal littermate controls were purchased from Jackson Laboratory (Bar Harbor, ME). Adult 12week-old intact or gonadectomized mice of both sexes were used in the short term (4 days) steroid supplementation experiments. Testosterone (1 mg/mouse • day) or 17/3-estradiol (1 mg/ mouse-day; Sigma Chemical Co., St. Louis, MO) were suspended in 0.02 ml vehicle (sesame oil and 10% ethyl alcohol) and injected sc. Mice were housed under controlled lighting and temperature conditions and fed purina chow ad libitum. In addition, adrenalectomized animals were fed a solution of normal saline. All animal experiments were approved by the Committee on Animal Care and Use at Humana Hospital-Michael Reese. Animals were maintained until 13 weeks of age, unless otherwise indicated, and were killed by cervical dislocation. The livers were quickly excised, frozen in liquid nitrogen, and stored at —80 C until processing. Livers from three animals were pooled for each EGF-R measurement, and each data point consisted of three separate EGF-R measurements (nine livers).
3241
The x-axis intercept of the Scatchard plot curves was used to calculate the total number of EGF-binding sites, corresponding to both high and low affinity states. EGF-R numbers, expressed in femtomoles, were normalized per mg membrane preparation protein. The intra- and interassay coefficients of variation for the RRA were 10% and 14%, respectively. Statistics Data were analyzed by analysis of variance (17) or Student's t test, using the ABSTAT (Anderson-Bell Corp., Parker, CO) computer program. Data were expressed as the mean ± SD, and P < 0.05 was considered significantly different.
Results Figure 1 shows Scatchard plots of EGF binding data in prepubertal (2-week-old; top panel) and adult (13week-old; bottom panel) liver cell membrane preparations from male and female mice. Levels of EGF-R, expressed in femtomoles per mg protein, in prepubertal mice were not significantly different between males (5 ± 3) and females (5 ± 1; top panel). There was also no significant difference in percent specific binding in prepubertal animals (males, 7.9 ± 1.0%; females, 5.6 ± 1.5%). The expression of EGF-R in adult mice (bottom panel) is sexually dimorphic, with males (451 ± 28) having significantly higher EGF-R numbers than females (65 ± 26; P
o Male • Female
PRE•PUBERTAL
Liver membrane preparations and binding assay Cell membranes from either male or female murine livers were used in a RRA to calculate the affinity and number of EGF-R (14). Briefly, 2 g minced mouse liver were homogenized at 4 C in Tris-sucrose buffer (TSB; 20 mM Tris-hydrochloride, 2 mM magnesium chloride, 2 mM potassium chloride, and 0.15 M sucrose, pH 7.4). The homogenate was centrifuged for 30 min at 48,000 X g, and the pellet was resuspended in TSB (~20 mg protein/ml) and stored at -80 C for use as binding preparation. The RRA consisted of a series of duplicate tubes (1.2 ml total volume/tube) containing 0.1 ml binding preparation (diluted to 1000 Mg protein), 0.1 ml radioiodinated (14) mouse EGF ([12BI]EGF)f 0.1 ml standard mouse EGF at concentrations ranging from 0.2-165 nM, and 0.9 ml TSB. The nonspecific binding (mean ± SD; 4.1 ± 0.8%) was determined using an excess of EGF (825 nM). The protein content of each binding preparation was measured by the method of Lowry et al. (15). Receptor-bound and -free EGF were separated by centrifugation at 20,000 x g after a 24-h incubation at 4 C. A best-fit analysis was used to compute the number of EGF-R and affinity constants in each binding preparation by Scatchard plot analysis (16).
50
100
150
200
250 300
350 400
450
500
BOUND EGF (moles x 10 l5 /mg)
FIG. 1. Scatchard analysis of EGF binding to hepatic receptor preparations from prepubertal (2-week-old; top panel) and adult (13-weekold; bottom panel) male and female mice. Each plot presents data obtained from three groups of three pooled livers (nine mice). The ordinate indicates the ratio of bound to free 126I-labeled EGF, and the abscissa indicates bound EGF in femtomoles per mg membrane preparation protein.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:56 For personal use only. No other uses without permission. . All rights reserved.
HEPATIC EGF-R EXPRESSION
3242
< 0.005). Male mice also demonstrate significantly higher percent specific binding (35.8 ± 7.5%) than females (12.9 ± 5.8%; P < 0.05). The upward concavities of the Scatchard plots in adult preparations (n = 18) provide the low (Ke) and high (Kf) affinity constants, expressed in liters per mol X 109. There was no significant difference in either Ke (male, 4.0 ± 0.4; female, 4.8 ± 1.1) or Kf (male, 0.8 ± 0.2; female, 0.7 ± 0.3) between the adult sexes. The ontogenesis of EGF-R in male and female mice between 2-13 weeks of age is shown in Fig. 2, left panel. Hepatic EGF-R expression did not differ significantly between males and females up to 3 weeks of age (males, 37 ± 32; females, 6 ± 2; P = NS). There was a significant increase in EGF-R expression in males at 4 weeks compared to females (247 ± 14 and 12 ± 9, respectively; P < 0.005). This dimorphism of EGF-R expression became progressively more marked until 13 weeks of age (males, 451 ± 28; females, 65 ± 26; P < 0.005). Figure 2, right panel, shows that the postpubertal increase in hepatic EGF-R expression is not limited to male mice. Indeed, EGF-R expression in adult females was significantly higher than that in prepubertal animals (65 ± 26 and 5 ± 1, respectively; P < 0.05). The effects of castration and adrenalectomy of male mice on the expression of EGF-R are shown in Fig. 3. No significant difference in the expression of EGF-R was observed between adult sham-operated (413 ± 20) and adrenalectomized (400 ± 178) male mice. Castration of adult males was followed by a significant decrease in the expression of EGF-R (90 ± 6) compared to that in sham-operated controls (P < 0.005). Castration plus adrenalectomy of adult mice resulted in a further significant decrease in the expression of EGF-R (10 ± 8) compared to that in adult castrated (P < 0.005), shamoperated (P < 0.005), or adrenalectomized (P < 0.05) males. Also shown in Fig. 3 is the expression of EGF-R
Endo'1991 Vol 129 • No 6
in adult males that had undergone previous prepubertal (2 weeks old) castration. EGF-R expression (34 ± 8) was significantly higher than that in either intact prepubertal (2-week-old) mice (5 ± 3; P < 0.005) or adult mice that had undergone both castration and adrenalectomy (P < 0.05). In contrast, adult mice castrated prepubertally had lower EGF-R expression than that in animals castrated as adults (P < 0.005). Figure 4 shows the effects of ovariectomy and adrenalectomy on EGF-R numbers in female mice. No significant difference in the expression of EGF-R was observed between adult sham-operated (53 ± 16) and adrenalectomized (41 ± 12) females. Ovariectomy of adult mice resulted in a significant increase (158 ± 44; P < 0.05) in the expression of EGF-R compared to levels in sham-operated females. In contrast, ovariectomy plus adrenalectomy resulted in markedly decreased EGF-R levels (11 ± 6) compared to those in ovariectomized (P < 0.005), adrenalectomized (P < 0.05), or sham-operated adult females (P < 0.05). Ovariectomy plus adrenalectomy of adult mice led to a decrease in the expression of EGF-R to levels not significantly different from those in prepubertal (2-week-old) intact animals (5 ± 1). Also shown in Fig. 4 is the expression of EGF-R in adult females after prepubertal (2 weeks old) ovariectomy. In these animals EGF-R expression (41 ± 6) was significantly higher (P < 0.005) than that in prepubertal (2week-old) intact and adult ovariectomized plus adrenalectomized females, but not significantly different from levels in sham-operated adult females. The EGF-R expression in adult mice that received prepubertal ovariectomy was significantly lower than that in mice receiving adult ovariectomy (P < 0.05). Table 1 documents the effect of steroid supplementation for 4 days on adult, intact, and gonadectomized male and female mice. Testosterone (1 mg/day • mouse) administered sc to intact females and gonadectomized female
FlG. 2. Left panel, Ontogenesis of hepatic EGF-R expression in male and female mice between 2-13 weeks of age. Right panel, Prepubertal (2-week-old) vs. adult (13-week-old) male and female murine liver EGF-R numbers. Each bar graph represents the mean ± SD for three groups of three pooled livers (nine mice), a, P < 0.05; b, P < 0.005.
2 3 4
8 AGE (Weeks)
13
Pre-Pubertal
Post'Pubertal
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:56 For personal use only. No other uses without permission. . All rights reserved.
HEPATIC EGF-R EXPRESSION
icantly different from that in control females (21 ± 19; P = NS).
(b) 450-1 (a)
3243
|
400-
Discussion
|»350-
We have hypothesized that hormonally induced changes in hepatic EGF-R expression may help modulate cn the autocrine, paracrine, and possible endocrine effects 0) S of endogenous EGF and TGFa on rodent livers. Indeed, E 200I 150EGF and TGFa are both produced by the rodent liver (c) a ioo(2, 11) and have a significant effect on hepatic regenerUJ ation (11). Additionally, EGF is secreted in relatively (e) 50large quantities by rodent salivary glands and kidneys JSL JO. Adult Adult PrePreAdult Adult (1). Sham- AOX Pubertal Pubenal CTX AOX Operated CTX Intact The present study on mouse livers confirms our preCTX vious finding of a sexually dimorphic EGF-R expression FIG. 3. Effects of castration and adrenalectomy on EGF-R number in in adult rats, with males having significantly higher adult (13-week-old) and prepubertal (2-week-old) male mice. Each bar EGF-R numbers than females (12). We have previously graph represents the mean ± SD for three groups of three pooled livers shown that the lower level of EGF-R expression in (nine mice). CTX, Castration; ADX, adrenalectomy; CTX+ADX, castration plus adrenalectomy. P = NS for a vs. b, and d vs. f; P < 0.05 females is not due to an excess of free EGF in female for b us. c, b vs. d, and d us. e; P < 0.005 for a vs. c, a vs. d, a us. e, a vs. liver preparations when measured after acid extraction f, c vs. d, c vs. e, and e vs. f. (12). In this study we have found that the affinity con250 stants at high and low levels of receptor occupancy are not significantly different between adult male and female W mice. These data indicate that the sex difference in EGF 95 200 binding to hepatic membranes is due to a difference in m receptor number, rather than an effect on the affinity of S 150the receptor. The nonlinear Scatchard plots of the binding data, with an upward concavity, have been previously § 100described (3, 12, 13, 18,19). These curvilinear plots point to a double class of binding sites, with high and low O en. affinities, assuming no negative cooperativity (3,18, 19). (j) The study of murine liver ontogenesis shows that EGFR levels do not differ significantly until they become Adult Adult Adult Adult PrePreSham- ADX ADX Pubertal Pubertal sexually differentiated at the onset of puberty (30-40 • Intact CTX Operated CTX days). We have also found that hepatic EGF-R numbers in females increase significantly compared to neonatal FIG. 4. Effects of ovariectomy and adrenalectomy on EGF-R expreslevels. Previous studies have shown that during intrasion in adult (13-week-old) and prepubertal (2-week-old) female mice. Each bar graph represents the mean ± SD for three groups of three uterine development, hepatic EGF-R are minimally expooled livers (nine mice). CTX, Ovariectomy; ADX, adrenalectomy; pressed and are most noticeable at the end of rodent CTX+ADX, ovariectomy plus adrenalectomy. P = NS for g vs. h, g vs. k, and j vs. 1; P < 0.05 for g vs. i, g vs. j , g vs. 1, h vs. i, h us. j , and i us. gestation (20, 21). In addition, mouse hepatic EGF production itself has been shown to increase and become k; P < 0.005 for i vs. j , j vs. k, and k vs. 1. sexually dimorphic during rodent development, with and male mice increased EGF-R numbers significantly males having significantly higher EGF levels than fe(P < 0.05) to levels seen in adult males. On the other males (22). Our results are in agreement with the findings hand, 17/3-estradiol (1 mg/day • mouse) given sc to intact of Kashimata et al. (23), who observed an apparent sex males and gonadectomized male and female mice did not difference in specific binding starting in the pubertal significantly affect EGF-R expression. period. Figure 5 compares levels of EGF-R in livers obtained The present study points to a complex interaction from adult androgen-insensitive male mice (Tfm/Y) to between hepatic EGF-R expression and gonadal influthose in male and female littermates that do not express ences. Gonadectomy of adult male and female mice rethis genetic deficiency. The expression of EGF-R in Tfm/ sults in a decrease and an increase, respectively, in EGFY mice (43 ± 3) was significantly lower than that in R numbers toward an intermediate level. We have precontrol males (151 ± 25; P < 0.005), but was not signifviously shown in a mouse hepatocyte model that there is ^
300-
x 250-
n
9
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:56 For personal use only. No other uses without permission. . All rights reserved.
3244
HEPATIC EGF-R EXPRESSION
Endo • 1991 Vol 129 • No 6
TABLE 1. Hepatic EGF-R expression (mean ± SD) in intact and gonadectomized male and female mice after steroid supplementation Sex
CT
Male Female
500 ± 58 35 ± 10
CT + T
CT + E
GDX
GDX + T
GDX + E
490 ± 75
200 ± 75° 193 ± 70d
510 ± 89* 542 ± 101'
216 ± 85 175 ± 53
NS
NS
NS
550 ± 100c
Vol. 129, No. 6 Printed in U.S.A.
Gonadal and Adrenal Effects on Hepatic Epidermal Growth Factor Receptor Expression in a Murine Model* BERT SCOCCIA, PETER KOVAR, AND ROBERT BENVENISTE Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Humana Hospital-Michael Reese; University of Illinois College of Health Sciences (B.S., P.K.); and Illinois Institute of Technology (R.B.), Chicago, Illinois 60616
ABSTRACT. We hypothesize that the actions of epidermal growth factor (EGF) may be modulated by changes in cell surface EGF receptor (EGF-R) expression under endocrine influences. Mouse liver cell membrane preparations were used in a RRA. During ontogenesis, both sexes showed a significant increase {P < 0.005) in hepatic EGF-R numbers at puberty; however, males demonstrated significantly higher levels than females (P < 0.005). Gonadectomy of adult males and females resulted in a significant (P < 0.05) decrease and increase, respectively, in hepatic EGF-R expression. Prepubertal gonadectomy in both sexes resulted in EGF-R levels similar to those observed in adult females. Adrenalectomy of adult animals of both sexes had no effect on hepatic EGF-R numbers, but gonadectomy plus adrenalectomy virtually obliterated EGF-R expression. Short term treatment with testosterone of adult females or gonadectomized
E
PIDERMAL growth factor (EGF) is a polypeptide with a mol wt of 6045 that induces a variety of cellular effects, including cell proliferation and differentiation (1). EGF shares a significant sequence homology with transforming growth factor-a (TGFa), and both bind to the EGF receptor (EGF-R) (2). The EGF-R has a mol wt of 170,000 and is a transmembrane glycoprotein that includes a cytoplasmic domain with tyrosine kinase activity (3). EGF-R has a significant sequence homology with the v-erb-B oncogene protein (4), and EGF-R is often overexpressed in transformed cells (5), lending support to the belief that EGF and the EGF-R play an important regulatory function in cells. Despite a large volume of information on the cellular effects of EGF and the acute biological responses elicited Received July 22,1991. Address requests for reprints to: Bert Scoccia, M.D., Department of Obstetrics and Gynecology, K-100, Humana Hospital-Michael Reese, Lake Shore Drive at 31st Street, Chicago, Illinois 60616. * Presented in part at the 36th Annual Meeting of the Society for Gynecologic Investigation, San Diego, CA, 1989 (Abstract 16), and at the 71st Annual Meeting of The Endocrine Society, Seattle WA, 1989 (Abstract 1034). This work was supported in part by the Michael Reese Medical Research Institute Council and NIH Biomedical Research Support Grant SO7-RR-05476 (to B.S.).
female and male mice significantly increased EGF-R numbers (P < 0.05) to adult male levels. 17/?-Estradiol given short term to adult males or gonadectomized male and female mice did not significantly alter EGF-R levels. EGF-R expression in androgeninsensitive male mice was significantly reduced (P < 0.005) to female levels. We conclude that 1) hepatic EGF-R numbers increase postpubertally in both sexes; 2) hepatic EGF-R expression is significantly stimulated by testosterone, and this effect depends on a functional androgen receptor; 3) the ovary has an inhibitory effect on adult hepatic EGF-R numbers; however, this effect does not appear to be mediated by estrogens; and 4) the adrenal gland has a stimulatory effect on adult hepatic EGF-R expression. (Endocrinology 129: 3240-3246,1991)
by EGF administration in vivo (6), the physiological role of EGF is not clearly understood. This gap is due in part to the difficulty in documenting that a change in circulating EGF levels precedes a normal physiological event (7, 8). One possibility is that EGF acts locally in a paracrine or autocrine fashion (9, 10). Alternatively, we hypothesize that the physiological role of EGF may depend on target tissue changes in the expression of EGFR, which may be modulated by endocrine influences. To study this hypothesis further, we have used a rodent liver model to examine, under physiological conditions, changes in EGF-R expression. EGF-R are highly expressed in rodent livers, and both EGF and TGFa play an important role in liver growth and maintenance (11). We have previously shown that there is a sexual dimorphism in the expression of hepatic EGF-R in cell membrane preparations (12) and at the cell surface (13). Most dramatic is the approximately 50-fold difference in EGF-R numbers observed between freshly plated male and female hepatocytes (13). The objectives of the present study are to determine 1) at which developmental stage the expression of EGFR becomes sexually dimorphic; 2) what effect prepubertal and adult gonadectomy, alone or in conjunction with adrenalectomy, has on EGF-R expression; 3) what is the
3240
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:56 For personal use only. No other uses without permission. . All rights reserved.
HEPATIC EGF-R EXPRESSION effect oftestosterone and 17/?-estradiol supplementation on intact and gonadectomized male and female mice; 4) what is the expression of EGF-R in livers from androgeninsensitive male mice.
Materials and Methods Animals and treatments Intact, adrenalectomized, or sham-operated BALB-c male and female mice (5-10 weeks old) were purchased from Harlan Laboratory (Indianapolis, IN). Newborn pups were obtained from in-house breeding and used at 2-4 weeks of age. Gonadectomies and sham operations were performed under inhalational anesthesia (methoxyflurane, Pittman-Moore, Washington Crossing, NJ) either at 2 (prepubertal) or 10 (adult) weeks of age. Adult 10-week-old androgen-insensitive mice (Tfm/Y, C57BL/6J-Aw-j-Ta) and normal littermate controls were purchased from Jackson Laboratory (Bar Harbor, ME). Adult 12week-old intact or gonadectomized mice of both sexes were used in the short term (4 days) steroid supplementation experiments. Testosterone (1 mg/mouse • day) or 17/3-estradiol (1 mg/ mouse-day; Sigma Chemical Co., St. Louis, MO) were suspended in 0.02 ml vehicle (sesame oil and 10% ethyl alcohol) and injected sc. Mice were housed under controlled lighting and temperature conditions and fed purina chow ad libitum. In addition, adrenalectomized animals were fed a solution of normal saline. All animal experiments were approved by the Committee on Animal Care and Use at Humana Hospital-Michael Reese. Animals were maintained until 13 weeks of age, unless otherwise indicated, and were killed by cervical dislocation. The livers were quickly excised, frozen in liquid nitrogen, and stored at —80 C until processing. Livers from three animals were pooled for each EGF-R measurement, and each data point consisted of three separate EGF-R measurements (nine livers).
3241
The x-axis intercept of the Scatchard plot curves was used to calculate the total number of EGF-binding sites, corresponding to both high and low affinity states. EGF-R numbers, expressed in femtomoles, were normalized per mg membrane preparation protein. The intra- and interassay coefficients of variation for the RRA were 10% and 14%, respectively. Statistics Data were analyzed by analysis of variance (17) or Student's t test, using the ABSTAT (Anderson-Bell Corp., Parker, CO) computer program. Data were expressed as the mean ± SD, and P < 0.05 was considered significantly different.
Results Figure 1 shows Scatchard plots of EGF binding data in prepubertal (2-week-old; top panel) and adult (13week-old; bottom panel) liver cell membrane preparations from male and female mice. Levels of EGF-R, expressed in femtomoles per mg protein, in prepubertal mice were not significantly different between males (5 ± 3) and females (5 ± 1; top panel). There was also no significant difference in percent specific binding in prepubertal animals (males, 7.9 ± 1.0%; females, 5.6 ± 1.5%). The expression of EGF-R in adult mice (bottom panel) is sexually dimorphic, with males (451 ± 28) having significantly higher EGF-R numbers than females (65 ± 26; P
o Male • Female
PRE•PUBERTAL
Liver membrane preparations and binding assay Cell membranes from either male or female murine livers were used in a RRA to calculate the affinity and number of EGF-R (14). Briefly, 2 g minced mouse liver were homogenized at 4 C in Tris-sucrose buffer (TSB; 20 mM Tris-hydrochloride, 2 mM magnesium chloride, 2 mM potassium chloride, and 0.15 M sucrose, pH 7.4). The homogenate was centrifuged for 30 min at 48,000 X g, and the pellet was resuspended in TSB (~20 mg protein/ml) and stored at -80 C for use as binding preparation. The RRA consisted of a series of duplicate tubes (1.2 ml total volume/tube) containing 0.1 ml binding preparation (diluted to 1000 Mg protein), 0.1 ml radioiodinated (14) mouse EGF ([12BI]EGF)f 0.1 ml standard mouse EGF at concentrations ranging from 0.2-165 nM, and 0.9 ml TSB. The nonspecific binding (mean ± SD; 4.1 ± 0.8%) was determined using an excess of EGF (825 nM). The protein content of each binding preparation was measured by the method of Lowry et al. (15). Receptor-bound and -free EGF were separated by centrifugation at 20,000 x g after a 24-h incubation at 4 C. A best-fit analysis was used to compute the number of EGF-R and affinity constants in each binding preparation by Scatchard plot analysis (16).
50
100
150
200
250 300
350 400
450
500
BOUND EGF (moles x 10 l5 /mg)
FIG. 1. Scatchard analysis of EGF binding to hepatic receptor preparations from prepubertal (2-week-old; top panel) and adult (13-weekold; bottom panel) male and female mice. Each plot presents data obtained from three groups of three pooled livers (nine mice). The ordinate indicates the ratio of bound to free 126I-labeled EGF, and the abscissa indicates bound EGF in femtomoles per mg membrane preparation protein.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:56 For personal use only. No other uses without permission. . All rights reserved.
HEPATIC EGF-R EXPRESSION
3242
< 0.005). Male mice also demonstrate significantly higher percent specific binding (35.8 ± 7.5%) than females (12.9 ± 5.8%; P < 0.05). The upward concavities of the Scatchard plots in adult preparations (n = 18) provide the low (Ke) and high (Kf) affinity constants, expressed in liters per mol X 109. There was no significant difference in either Ke (male, 4.0 ± 0.4; female, 4.8 ± 1.1) or Kf (male, 0.8 ± 0.2; female, 0.7 ± 0.3) between the adult sexes. The ontogenesis of EGF-R in male and female mice between 2-13 weeks of age is shown in Fig. 2, left panel. Hepatic EGF-R expression did not differ significantly between males and females up to 3 weeks of age (males, 37 ± 32; females, 6 ± 2; P = NS). There was a significant increase in EGF-R expression in males at 4 weeks compared to females (247 ± 14 and 12 ± 9, respectively; P < 0.005). This dimorphism of EGF-R expression became progressively more marked until 13 weeks of age (males, 451 ± 28; females, 65 ± 26; P < 0.005). Figure 2, right panel, shows that the postpubertal increase in hepatic EGF-R expression is not limited to male mice. Indeed, EGF-R expression in adult females was significantly higher than that in prepubertal animals (65 ± 26 and 5 ± 1, respectively; P < 0.05). The effects of castration and adrenalectomy of male mice on the expression of EGF-R are shown in Fig. 3. No significant difference in the expression of EGF-R was observed between adult sham-operated (413 ± 20) and adrenalectomized (400 ± 178) male mice. Castration of adult males was followed by a significant decrease in the expression of EGF-R (90 ± 6) compared to that in sham-operated controls (P < 0.005). Castration plus adrenalectomy of adult mice resulted in a further significant decrease in the expression of EGF-R (10 ± 8) compared to that in adult castrated (P < 0.005), shamoperated (P < 0.005), or adrenalectomized (P < 0.05) males. Also shown in Fig. 3 is the expression of EGF-R
Endo'1991 Vol 129 • No 6
in adult males that had undergone previous prepubertal (2 weeks old) castration. EGF-R expression (34 ± 8) was significantly higher than that in either intact prepubertal (2-week-old) mice (5 ± 3; P < 0.005) or adult mice that had undergone both castration and adrenalectomy (P < 0.05). In contrast, adult mice castrated prepubertally had lower EGF-R expression than that in animals castrated as adults (P < 0.005). Figure 4 shows the effects of ovariectomy and adrenalectomy on EGF-R numbers in female mice. No significant difference in the expression of EGF-R was observed between adult sham-operated (53 ± 16) and adrenalectomized (41 ± 12) females. Ovariectomy of adult mice resulted in a significant increase (158 ± 44; P < 0.05) in the expression of EGF-R compared to levels in sham-operated females. In contrast, ovariectomy plus adrenalectomy resulted in markedly decreased EGF-R levels (11 ± 6) compared to those in ovariectomized (P < 0.005), adrenalectomized (P < 0.05), or sham-operated adult females (P < 0.05). Ovariectomy plus adrenalectomy of adult mice led to a decrease in the expression of EGF-R to levels not significantly different from those in prepubertal (2-week-old) intact animals (5 ± 1). Also shown in Fig. 4 is the expression of EGF-R in adult females after prepubertal (2 weeks old) ovariectomy. In these animals EGF-R expression (41 ± 6) was significantly higher (P < 0.005) than that in prepubertal (2week-old) intact and adult ovariectomized plus adrenalectomized females, but not significantly different from levels in sham-operated adult females. The EGF-R expression in adult mice that received prepubertal ovariectomy was significantly lower than that in mice receiving adult ovariectomy (P < 0.05). Table 1 documents the effect of steroid supplementation for 4 days on adult, intact, and gonadectomized male and female mice. Testosterone (1 mg/day • mouse) administered sc to intact females and gonadectomized female
FlG. 2. Left panel, Ontogenesis of hepatic EGF-R expression in male and female mice between 2-13 weeks of age. Right panel, Prepubertal (2-week-old) vs. adult (13-week-old) male and female murine liver EGF-R numbers. Each bar graph represents the mean ± SD for three groups of three pooled livers (nine mice), a, P < 0.05; b, P < 0.005.
2 3 4
8 AGE (Weeks)
13
Pre-Pubertal
Post'Pubertal
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:56 For personal use only. No other uses without permission. . All rights reserved.
HEPATIC EGF-R EXPRESSION
icantly different from that in control females (21 ± 19; P = NS).
(b) 450-1 (a)
3243
|
400-
Discussion
|»350-
We have hypothesized that hormonally induced changes in hepatic EGF-R expression may help modulate cn the autocrine, paracrine, and possible endocrine effects 0) S of endogenous EGF and TGFa on rodent livers. Indeed, E 200I 150EGF and TGFa are both produced by the rodent liver (c) a ioo(2, 11) and have a significant effect on hepatic regenerUJ ation (11). Additionally, EGF is secreted in relatively (e) 50large quantities by rodent salivary glands and kidneys JSL JO. Adult Adult PrePreAdult Adult (1). Sham- AOX Pubertal Pubenal CTX AOX Operated CTX Intact The present study on mouse livers confirms our preCTX vious finding of a sexually dimorphic EGF-R expression FIG. 3. Effects of castration and adrenalectomy on EGF-R number in in adult rats, with males having significantly higher adult (13-week-old) and prepubertal (2-week-old) male mice. Each bar EGF-R numbers than females (12). We have previously graph represents the mean ± SD for three groups of three pooled livers shown that the lower level of EGF-R expression in (nine mice). CTX, Castration; ADX, adrenalectomy; CTX+ADX, castration plus adrenalectomy. P = NS for a vs. b, and d vs. f; P < 0.05 females is not due to an excess of free EGF in female for b us. c, b vs. d, and d us. e; P < 0.005 for a vs. c, a vs. d, a us. e, a vs. liver preparations when measured after acid extraction f, c vs. d, c vs. e, and e vs. f. (12). In this study we have found that the affinity con250 stants at high and low levels of receptor occupancy are not significantly different between adult male and female W mice. These data indicate that the sex difference in EGF 95 200 binding to hepatic membranes is due to a difference in m receptor number, rather than an effect on the affinity of S 150the receptor. The nonlinear Scatchard plots of the binding data, with an upward concavity, have been previously § 100described (3, 12, 13, 18,19). These curvilinear plots point to a double class of binding sites, with high and low O en. affinities, assuming no negative cooperativity (3,18, 19). (j) The study of murine liver ontogenesis shows that EGFR levels do not differ significantly until they become Adult Adult Adult Adult PrePreSham- ADX ADX Pubertal Pubertal sexually differentiated at the onset of puberty (30-40 • Intact CTX Operated CTX days). We have also found that hepatic EGF-R numbers in females increase significantly compared to neonatal FIG. 4. Effects of ovariectomy and adrenalectomy on EGF-R expreslevels. Previous studies have shown that during intrasion in adult (13-week-old) and prepubertal (2-week-old) female mice. Each bar graph represents the mean ± SD for three groups of three uterine development, hepatic EGF-R are minimally expooled livers (nine mice). CTX, Ovariectomy; ADX, adrenalectomy; pressed and are most noticeable at the end of rodent CTX+ADX, ovariectomy plus adrenalectomy. P = NS for g vs. h, g vs. k, and j vs. 1; P < 0.05 for g vs. i, g vs. j , g vs. 1, h vs. i, h us. j , and i us. gestation (20, 21). In addition, mouse hepatic EGF production itself has been shown to increase and become k; P < 0.005 for i vs. j , j vs. k, and k vs. 1. sexually dimorphic during rodent development, with and male mice increased EGF-R numbers significantly males having significantly higher EGF levels than fe(P < 0.05) to levels seen in adult males. On the other males (22). Our results are in agreement with the findings hand, 17/3-estradiol (1 mg/day • mouse) given sc to intact of Kashimata et al. (23), who observed an apparent sex males and gonadectomized male and female mice did not difference in specific binding starting in the pubertal significantly affect EGF-R expression. period. Figure 5 compares levels of EGF-R in livers obtained The present study points to a complex interaction from adult androgen-insensitive male mice (Tfm/Y) to between hepatic EGF-R expression and gonadal influthose in male and female littermates that do not express ences. Gonadectomy of adult male and female mice rethis genetic deficiency. The expression of EGF-R in Tfm/ sults in a decrease and an increase, respectively, in EGFY mice (43 ± 3) was significantly lower than that in R numbers toward an intermediate level. We have precontrol males (151 ± 25; P < 0.005), but was not signifviously shown in a mouse hepatocyte model that there is ^
300-
x 250-
n
9
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:56 For personal use only. No other uses without permission. . All rights reserved.
3244
HEPATIC EGF-R EXPRESSION
Endo • 1991 Vol 129 • No 6
TABLE 1. Hepatic EGF-R expression (mean ± SD) in intact and gonadectomized male and female mice after steroid supplementation Sex
CT
Male Female
500 ± 58 35 ± 10
CT + T
CT + E
GDX
GDX + T
GDX + E
490 ± 75
200 ± 75° 193 ± 70d
510 ± 89* 542 ± 101'
216 ± 85 175 ± 53
NS
NS
NS
550 ± 100c
