Influence of estradiol on insulin-like growth factor-1-induced luteinizing hormone secretion

Brain Research 1013 (2004) 91 – 97 www.elsevier.com/locate/brainres

Research report

Influence of estradiol on insulin-like growth factor-1-induced luteinizing hormone secretion Jill K. Hiney, Vinod Srivastava, Robert K. Dearth, W. Les Dees * Department of Veterinary Anatomy and Public Health, College of Veterinary Medicine, Texas A & M University, College Station, TX, 77843-4458, USA Accepted 31 March 2004 Available online 6 May 2004

Abstract Several studies suggest an interrelationship between estradiol (E2) and insulin-like growth factor-1 (IGF-1) at the hypothalamic level. The present study was designed to discern if the capability of IGF-1 to release LH and influence the timing of female puberty is influenced by E2. Twenty-eight-day-old female rats were ovariectomized (OVEX), then implanted with a third ventricular (3V) cannula. Two weeks later, these animals received subcutaneous (s.c.) injection of oil, or either one or two injections of E2 in the form of estradiol benzoate (1 Ag). Forty-eight hours later, four basal blood samples were drawn then the animals received IGF-1 (200 ng) or saline via the 3V and four more blood samples were taken. Results indicated that E2 replacement lowered basal LH levels and IGF-1 induced a significant LH release in only animals that had E2 levels above 20 pg/ml. These levels of E2 were also associated with increases ( p < 0.05) in the expression of both IGF-1 receptor (IGF-1R) mRNA and protein. In order to further support the hypothesis that the action of IGF-1 at the time of puberty is influenced by E2, 24day-old intact female rats received s.c. injection of sesame oil or 0.1 Ag of E2. The next day, the E2-treated animals also received twice daily s.c. injections of either IGF-1 (500 ng) or saline until vaginal opening (VO) occurred. The animals that received E2 plus IGF-1 showed VO at 31.1 days, which was 2.5 days earlier ( p < 0.01) than E2-treated animals and 4 days earlier ( p < 0.001) than IGF-1-treated and saline control animals. Taken together, these results indicate that the hypothalamic action of IGF-1 to stimulate LH release and advance female pubertal development is dependent upon the influence of E2. D 2004 Elsevier B.V. All rights reserved. Theme: Development and regeneration Topic: Hormones and Development Keywords: Insulin-like growth factor-1; Estradiol; Puberty

1. Introduction Insulin-like growth factor-1 (IGF-1) has been identified as an important factor involved in the progression of the pubertal process [6,15,17,22]. IGF-1 levels in circulation increase prior to puberty [5,13,15], along with a concomitant increase in estradiol (E2) [15]. Immunoreactive IGF-1 increases in the tanycytes of the arcuate nucleus (ARC) and median eminence (ME) area at the time of puberty, whereas ovariectomy causes a decrease in the IGF-1 peptide in this area [8]. The gene expression of the type 1 IGF-1 receptor (IGF-1R) also rises at first proestrus [15] and the binding * Corresponding author. Tel.: +1-979-845-1430; fax: +1-979-8479038. E-mail address: [email protected] (W.L. Dees). 0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2004.03.054

activity is sensitive to changes in E2 [3,8,19]. Several findings further support an interaction between IGF-1 and E2 signaling in cells of the hypothalamus [9,11,12,18]. In vitro studies using hypothalamic neurons and glia have shown that E2 stimulates and controls IGF-1R activity and binding proteins [9,18]. Both E2 and IGF-1 and their respective receptors are necessary to stimulate differentiation of hypothalamic neurons and glia [9,11,12]. Furthermore, estrogen-induced changes in astroglial and neuronal contacts in the hypothalamus that occur during proestrus and estrus have been shown to be dependent on both E2 and IGF-1 receptors [11,12]. Estrogen receptors and IGF-1Rs have also been found to be colocalized in neurons and astroglia in several areas of the brain including the hypothalamus [4], indicating that the interrelationship between E2 and IGF-1 in the brain may include activity at the intracellular level.

92

J.K. Hiney et al. / Brain Research 1013 (2004) 91–97

Less is known about IGF-1 and E2 interactions with regard to gonadotrophin release during pubertal development, whereas several studies in adult animals have been done. In adult ovariectomized rats, Quesada and Etgen [19] demonstrated that E2 replacement increased IGF-1 receptor binding. Also, they showed that central administration of the IGF-1R antagonist, JB-1, blocks the E2/progesterone-induced LH surge [20]. Studies using adult castrated sheep [1] also demonstrated a relationship between E2, IGF-1 and LH secretion in fed and fasted animals. In immature animals, central administration of IGF-1 stimulated LH release, with the effect being observed throughout the peripubertal period [15]. However, we noted that IGF-1induced LH release varied between animals depending on their individual stage of pubertal development; hence, indirectly suggesting the different responses to IGF-1 were due to varying levels of gonadal steroid in serum. Also, in juvenile monkeys, chronic IGF-1 administration lowered the negative feedback effect of E2 on LH release thereby advancing the progression of puberty [21]. Due to the significance of IGF-1 at the time of puberty, the present study was conducted to discern directly whether IGF-1induced LH release is influenced by E2 and if this is due to changes in the IGF-1R.

2. Materials and methods 2.1. Animals All rats used in the present study were of the Sprague – Dawley line and were purchased from Charles River (Boston, MA). All rats were maintained under controlled conditions of light (lights on: 0600 h, lights off: 1800 h) and temperature (23 jC) with ad libitum access to food and water. All of the following procedures were approved by the University Laboratory Animal Committee of Texas A&M University. Tribromoethanol (2.5%) was used as the anesthesia for the surgical procedures. 2.2. Experiment 1: IGF-1-induced LH release after ovariectomy and E2 replacement Twenty-eight-day-old female rats were bilaterally ovariectomized and allowed 10 days for recovery. The animals were stereotaxically implanted with a stainless steel cannula (23 ga) in the third ventricle (3V) of the brain as previously described [2]. Two days after 3V surgery, animals were divided into three groups: group 1 received a subcutaneous (s.c.) injection of sesame oil, 48 h prior to the experiment; group 2 received a single s.c. injection of estradiol benzoate (EB;1 Ag, Sigma, St. Louis, MO) 48 h prior to the experiment; group 3 received two s.c. injections of EB, one at 72 h and the other at 48 h prior to the experiment. Twenty-four hours before the experiment, a silastic cannula was inserted into the right external

jugular vein of each rat according to a previously described method [14]. On the day of the experiment, an extension of tubing was attached to each cannula then flushed with heparinized saline (100 IU/ml). Four basal blood samples were drawn from each freely moving animal at 10-min intervals, immediately followed by a 3V injection of IGF-1 (200 ng/3 Al, rat IGF-1, Gropep, Australia), or an equal volume of saline. This dose was used based on results of our previous dose response studies [15]. Following the respective injection, four more samples were drawn for a total of eight samples. Blood samples were taken between 1000 and 1200 h. After the experiment, the brains were inspected to verify proper placement of the cannulae. Samples were centrifuged at 4 jC and the serum stored at 70 jC until assayed for LH. A single blood sample was taken prior to experiment for serum E2 measurement. 2.3. Experiment 2: IGF-1R mRNA and protein levels after ovariectomy and E2 replacement Animals were OVEX at 28 days and after 2 –3 weeks were divided into the same three groups as described above. Forty-eight hours after the last EB injection, the animals were killed by decapitation and the brain removed. The medial basal hypothalamus (MBH) was dissected out by making a cut caudal to the optic chiasm, another cut at the rostral border of the mammillary bodies, and then using the borders of the tuberinfundibular sulcus as the lateral limits. A final cut was made dorsal to the ventral limits about 3 mm. The MBH was frozen at 70 jC until Western blot and PCR analysis were performed. Trunk blood was taken and serum removed for E2 analysis. 2.4. Experiment 3: effect of E2 and IGF-1 administration on vaginal opening Twenty-four-day-old female rats were divided into four groups: Group 1 received a s.c. injection of sesame oil. Group 2 received a s.c. injection of 0.1 Ag of EB in sesame oil. Group 3 received sesame oil, then the following day two s.c. injections of IGF-1 (500 ng/100 Al, Gropep) were given at 1400 and 1600 h. Group 4 received 0.1 Ag of EB and the following day two s.c. injections of IGF-1 were given at 1400 and 1600 h. These injections of IGF-1 were given until the animal showed vaginal opening (VO). When VO occurred, animals were killed and classified as to the phase of pubertal development. 2.5. Western blot analysis for IGF-1R IGF-1R immunodetection was performed on plasma membrane preparations isolated from MBH. The frozen tissues were homogenized at 4 jC in ice-cold homogenization buffer containing 50 mM Tris Cl (pH 7.4), 50 mM NaCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 10 Ag/ml of aprotinin, 10 Ag/ml of leupeptin, 1 Ag/ml of

J.K. Hiney et al. / Brain Research 1013 (2004) 91–97

pepstatin and 1 mM sodium orthovanadate (Na3VO4). The homogenates were centrifuged at 2900  g for 20 min at 4 jC and the supernatants were centrifuged again at 29,000  g for 45 min at 4 jC. Pellets were resuspended in homogenization buffer containing 0.25% nadeoxycholate and 1% NP40, and centrifuged at 15,000  g for 20 min at 4 jC. The total protein concentration in the supernatant was quantitated by the Bradford protein assay (Bio-Rad Laboratories, Hercules, CA) using bovine serum albumin as the standard. Aliquots of membrane preparation of protein (40 Ag) were separated on 8% SDS-polyacrylamide gel electrophoresis (PAGE) under reducing conditions and electrophoretically blotted onto polyvinylidene-difluoride (PVDF) membranes. Additionally, a prestained molecular weight marker (Bio-Rad Laboratory, Hercules, CA) was used with each run to verify the complete and efficient transfer of the protein onto the membrane. Following transfer, the gel was stained with Coomassie Brilliant Blue R-250 followed by destaining to check the transfer efficiency of the protein. In this regard, the destained gel did not show any Coomassiestained protein bands. Membranes were blocked with 5% nonfat dry milk in phosphate-buffered saline (PBS, pH 7.4) overnight at 4 jC and subsequently incubated at room temperature for 3 h with the chicken polyclonal anti-human IGF-1 receptor a-subunit (4 Ag/ml; Upstate Biotechnology, Lake Placid, NY). Following incubation, membranes were washed in PBS and incubated for 2 h with goat antichicken IgG conjugated with horseradish peroxide at a dilution of 1:15,000 (Jackson ImmunoResearch Lab., West Grove, PA). After another wash, the IGF-1 receptor protein of 130 kDa was detected by the enhanced chemiluminescence method (PerkinElmer Life Sciences, Boston, MA), and quantified using Quantity one software (Bio-Rad Laboratories, Hercules, CA). 2.6. Reverse transcriptase – polymerase chain reaction (RT-PCR) Total RNA was isolated from MBH using RNA Zol B followed by precipitation with isopropanol and ethanol washes according to the manufacturer’s instructions (TeleTest, Friendswood, TX).The integrity of the RNA was determined by the visualization of the ethidium bromidestained 28S and 18S ribosomal RNA bands, and quantitation was done by measuring its absorbance at 260 nm. Total RNA (2 Ag) from each sample was reverse transcribed by using Promega RT system (Promega Life Science, Madison, WI) to first strand complementary DNA (cDNA) with 14 U of AMVRT reverse transcriptase in the presence of 5 mM MgCl2, 20 U RNase inhibitor, 1 mM dNTPs and 25 Ag/ml random primer in a total volume of 20 Al. The reaction mixture of first strand cDNA synthesis was incubated at room temperature for 15 min followed by the incubation at 42 jC for 30 min. Afterward, the tubes were heated at 95 jC for 5 min to stop the reaction, then was kept at 4 jC until

93

polymerase chain reaction (PCR) was performed. The primers employed in the PCR were designed from the coding region of the rat IGF-1 receptor cDNA sequence. The forward primer 5V-TTACCTCTCCACCATAGACT-3V was from bases 495 –510 and the reverse primer 5V-GGGATACAGTACATGCTCT-3Vspan the region 995 –1014 bp of the rat IGF-1 receptor cDNA. The expected size of the amplified fragment was 519 bp. PCR was performed in a total volume of 100 Al with 20 Al reverse transcribed product, 1.5 mM MgCl2, 0.2 mM dNTP, 50 pmol forward primer, 50 pmol reverse primer. Tubes were placed in the DNA icycler and incubated at 95 jC (initial denaturation) for 2 min, then 2.5 units of Taq DNA polymerase was added to each tube. This was followed by 36 cycles of denaturation at 95 jC for 2 min, annealing at 45 jC for 1 min, elongation at 72 jC for 1 min and the final extension was done for 7 min at 72 jC. Each sample was amplified along with a 315-bp fragment of the ribosomal 18S RNA which serves as an internal control to which data are normalized. The PCR product was analyzed on 2% agarose gel containing 0.5 Ag/ml ethidium bromide using TBE (89 mM Tris base, 89 mM boric acid, 2 mM EDTA) buffer. The molecular sizes of the amplified products (IGF-1R and18S) were determined by comparison with the mol wt markers run in parallel with the RT-PCR products. Bands were visualized by Gel Doc 2000 Imager and quantified using Quantity one software (Bio-Rad Laboratories, Hercules, CA). 2.7. Assays and statistics 2.7.1. E2 and LH radioimmunoassays Serum E2 levels were measured as previously described [15] using a kit purchased from Diagnostic Products (Los Angeles, CA). Serum LH was measured by a modification of previously described methods [15]. The antisera, rLH-S11, as well as the reagents were obtained from Dr. A.F. Parlow and the National Hormone and Pituitary Program. Results are expressed in terms of LH-RP-3 standards, respectively. The sensitivity of the E2 and LH assays was 8 pg/ml and 0.07 ng/ml respectively. The inter- and intraassay variations for both assays were less than 10%. 2.7.2. Statistical analysis Differences between groups were initially analyzed using analysis of variance (ANOVA), with post hoc testing using the Student – Newman – Keuls multiple range test (SNK), and when appropriate, the unpaired Student’s t-test. The arithmetic mean area under the LH curve is determined by the trapezoid rule that calculates the area under the curve that occurred within the pre and post injection samples, respectively. LH pulse assessment and mean area under the curve were determined by Prism 3.0 software. All statistical tests were conducted using INSTAT and Prism software for the IBM PC (GraphPad, San Diego, CA). Probability values < 0.05 were considered to be statistically significant.

94

J.K. Hiney et al. / Brain Research 1013 (2004) 91–97

3. Results 3.1. Influence of circulating E2 on IGF-1-induced LH release following ovariectomy Giving either one or two injections of EB (1 Ag/injection) allowed us to vary the levels of E2 exposure prior to IGF-1 stimulation. Fig. 1, panel A shows that, 2 weeks following

ovariectomy, when serum E2 levels were below assay sensitivity, IGF-1 was unable to increase LH over the elevated basal levels. E2 replacement lowered basal LH levels as seen in panels B and C. A single dose of EB resulted in steroid levels between 9.2 and 20.4 pg/ml (mean E2 level = 16.14 pg/ ml). This level of E2 was not high enough in the serum to influence an IGF-1-induced release of LH (panel B, p>0.05). However, animals that received two injections of EB showed elevated levels ranging between 29.2 and 94.8 pg/ml (mean E2 level = 46.7 F 6.2 pg/ml) resulting in a marked ( p < 0.01) IGF-1 stimulation of LH release (panel C). 3.2. Influence of serum E2 on IGF-1R mRNA expression and protein levels in the MBH We evaluated whether varying levels of serum E2 could alter IGF-1R mRNA or protein levels in the MBH. The representative autoradiogram shown in Fig. 2A demonstrates the RT-PCR generated band of IGF-1R (519 bp) in the MBH of OVEX and EB-treated animals. Fig. 2B shows the composite graph after normalizing the values of the IGF1R RNA to the internal control, 18S ribosomal RNA. In those animals with low E2 levels (mean = 13.3 F 7.4) the expression for IGF-1R did not change, but in animals with higher E2 levels (mean = 36.8 F 5.7) the expression for IGF1R mRNA was elevated ( p < 0.05) in the MBH. Translational changes in IGF-1R protein mirror the effect that occurred at the transcriptional level. Fig. 3A shows a representative autoradiogram that illustrates an immunoreactive band of 130 kDa IGF-1R protein detected in the MBH of both ovariectomized and EB-treated rats. Fig. 3B illustrates a composite graph of the densitometric quantitation of IGF-1R protein in OVEX and EB-treated rats. The IGF-1R protein levels did not increase significantly in those animals with low E2 levels ( p>0.05). However, induction of higher levels of E2 augmented the increase ( p < 0.05) in the protein content of IGF-1R over those animals with E2 levels below 20 pg/ml and even higher ( p < 0.01) over the OVEX group. 3.3. Effect of E2 priming on IGF-1-induced puberty

Fig. 1. Effect of ovariectomy and E2 replacement on IGF-1-induced LH release. A – C: Composite IGF-1-induced LH secretory profiles of 2-week OVEX (panel A), OVEX with E2 levels < 20 pg/ml (panel B) or OVEX with E2 levels >20 pg/ml (panel C). Note that IGF-1 was unable to induce an increase in LH release over the already elevated basal levels in the OVEX rats (panel A). E2 replacement lowered basal LH levels, but IGF-1 was unable to stimulate a significant response in animals with E2 levels below 20 pg/ml (panel B). However, IGF-1 stimulated a significant LH response in those animals that had E2 levels higher than 20 pg/ml (panel C). Open bar indicates the mean ( F S.E.M.) area under the LH curve of the first four basal samples prior to IGF-1 administration. Solid bar depicts the mean ( F S.E.M.) area under the LH curve of the four samples following IGF-1 injection. N = 11 for OVEX only, 14 for OVEX with E2 < 20 pg/ml, 11 for OVEX with E2>20 pg/ml. **p < 0.01.

Priming 24-day-old intact female rats with E2 and IGF-1 raised the steroid and peptide levels in these animals to levels that are normally encountered prior to pubertal onset. Fig. 4 demonstrates that a single low dose (0.1 Ag) of EB and then subsequent injections of IGF-1 (500 ng/injection) caused VO to occur 2.5 days earlier ( p < 0.01 by SNK; 31.1 F 0.64 days of age) than in the animals that received EB only (33.6 F 0.75 days of age). The advancement in VO was even more significant (4 days; p < 0.001 by SNK) when compared to the animals that received IGF-1 alone (35.4 F 0.3), or the control animals that received saline (35.3 F 0.56 days of age). The EB only group showed a slight advance in VO ( p>0.05, not significantly different by SNK) over the IGF-1-treated and control groups by 1.8 and 1.7 days, respectively. All of the vaginal smears showed

J.K. Hiney et al. / Brain Research 1013 (2004) 91–97

Fig. 2. Effect of E2 replacement on IGF-1R gene expression in the MBH of ovariectomized rats. (A) A representative gel of PCR amplification showing the effects of increasing levels of E2 on IGF-1 receptor mRNA expression in the MBH of ovariectomized rats. Lane 1 represents a 100 bp DNA ladder. Total RNA (2 Ag) isolated from MBH fragments of OVEX late juvenile female rats (lanes 2 – 4), OVEX with E2 levels < 20 pg/ml (lanes 5 – 7) and OVEX with E2 levels >20 pg/ml (lanes 8 – 10) was reverse transcribed and the cDNA was amplified for 36 cycles using IGF-1 receptor and 18S ribosomal primers. PCR products were electrophoresed on 2% agarose gel and photographed. Ethidium bromide stained gel showed the expected RT-PCR products for IGF-1 receptor at 519 bp and 18S at 315 bp. (B) Composite graph that shows the densitometric quantitation of the bands from three gels that correspond to the IGF-1R mRNA. The data were normalized to the internal control, 18S mRNA, and represent the mean F S.E.M. The IGF-1R gene expression showed a slight nonsignificant elevation over the OVEX animals when E2 levels were below 20 pg/ml, but when E2 levels were raised over 20 pg/ml, the level of gene expression significantly increased over the OVEX animals and animals with E2 < 20 pg/ml. *p < 0.05 vs. OVEX and E2 < 20 pg/ml. N = 5 for OVEX, 7 for OVEX with E2 < 20 pg/ml and 7 for OVEX with E2>20 pg/ml.

95

Fig. 3. Effect of E2 replacement on IGF-1R protein expression in the MBH of ovariectomized rats. (A) Representative immunoblot of the 130-kDa IGF-1R protein isolated from MBH fragments of OVEX late juvenile female rats (lanes 1 – 3), OVEX with E2 < 20 pg/ml (lanes 4 – 6) and OVEX with E2>20 pg/ml (lanes 7 – 9). Total membrane proteins (40 Ag) were separated on 8% SDS-PAGE and electroblotted onto a PVDF membrane. The blots were incubated with a chicken polycolonal anti-human IGF-1 receptor a-subunit antibody. The specific protein was detected by ECL method. (B) Composite graph that shows the densitometric quantitation of all of the bands from three blots that correspond to IGF-1R protein. Note that animals with E2 levels less than 20 pg/ml (solid bar) did not show an increase in IGF-1R protein levels over OVEX animals (open bar). Raising the levels of E2 over 20 pg/ml (hatched bar) caused a marked increase in IGF-1R protein over OVEX (open bar) and low E2 (solid bar). **p < 0.01 vs. the OVEX and +p < 0.05 vs. E2 < 20 pg/ml. N = 7 for OVEX; 7 for OVEX with E2 < 20 pg/ml and 8 for OVEX with E2>20 pg/ml.

have demonstrated that prepubertal OVEX does not alter the increase in circulating IGF-1 that occurs at puberty [13,23], indicating that the prepubertal rise in peripherally derived IGF-1, that acts centrally to induce LH release [15], is a gonadal independent event. Importantly, Duenas et al. [8] showed that the increase in immunoreactive IGF1 in the medial basal hypothalamus that occurs prior to the first LH surge is dependent on ovarian E2 levels. In this

cornified epithelium indicating first estrus. Investigation of the ovaries showed corpus hemoragicum and luteum indicating ovulation.

4. Discussion In our earlier study [15], we found that in all stages of pubertal development, IGF-1 significantly induced LH release, but that within late juvenile and proestrus stages, the level of LH stimulation varied among animals. Although the serum levels of E2 were not measured in those animals, the fact that E2 levels are known to vary during those stages of development suggested that E2 could play a role in this action. Therefore, it was critical to discern if different circulating levels of E2 were associated with the varied capabilities of IGF-1 to release LH. Previous reports

Fig. 4. The effect of E2 priming on IGF-1-induced precocious puberty. Open and diagonal hatched bars represent the mean age in days at VO for the control and IGF-1-treated rats, respectively. Vertical hatched and solid bars represent the E2- and E2/IGF-1-treated rats, respectively. Note that E2 priming followed by IGF-1 treatments advanced the age of VO by 2.5 days vs. E2-treated and by 4 days vs. control and IGF-1-treated rats. **p < 0.01 vs. E2 only; +p < 0.001 vs. control and IGF-1-treated rats. Significance determined by ANOVA followed by SNK. N = 10 for all groups.

96

J.K. Hiney et al. / Brain Research 1013 (2004) 91–97

regard, they showed that OVEX decreased the IGF-1 immunoreactivity in the glial cells of the medial basal hypothalamus and then increased dose dependently after E2 replacement [8]. In the present study, OVEX blocked the ability of IGF-1 to stimulate LH release, whereas this IGF-1 effect was restored following short-term E2 replacement. The level of E2 in serum had to be over 20 pg/ml in order for IGF-1-induced LH release to occur, since animals below this level did not show a LH response to IGF-1. While we have shown this LH response to IGF-1 is clearly influenced by E2, additional research is needed to determine more clearly the serum level above 20 pg/ml that may represent the threshold, and to assess the importance of the increase in E2 vs. length of E2 exposure to the elevated levels. This effect of E2 replacement on IGF-1induced LH release was also demonstrated in adult castrated sheep. Administration of low doses of IGF-1 into the circulation stimulated LH release in both the castrated and EB-treated animals, but steroid replacement enhanced the IGF-1 stimulation of LH secretion [1]. This enhancing effect of E2 on hypothalamic-pituitary responsiveness to IGF-1 was also demonstrated by E2-induced upregulation of pituitary IGF-1 binding in OVEX rats [16]. Variation in IGF-1-induced LH release may also be due to E2 effects on synthesis of the IGF-1R. In this study, elevating the level of E2 in serum over 20 pg/ml stimulated gene expression and increased the protein levels of IGF1R, whereas lower steroid levels had no effect. The influence of E2 on IGF-1R has also been demonstrated recently by utilizing the IGF-1R antagonist, JB-1. In adult OVEX rats, central administration of JB-1 does not affect the elevated basal levels of LH, but blocks the E2/progesterone-induced LH surge. This blockade of the IGF-1 receptor suggests that the positive feedback of E2 on LH release requires IGF-1R activity [20]. Additionally, in the ME, IGF-1R mRNA levels are higher during the afternoon of first proestrus than during juvenile development [15]. This suggests that E2 is acting on the hypothalamus to induce IGF-1R synthesis at a time when steroid levels are increasing. In adult and developing rats, accumulation of IGF-1 immunoreactivity in the tanycytes that line the ME has been described to be dependent on E2 [8,10]. During the late stage of proestrus, IGF-1 receptors located on tanycytes [10] accumulate IGF-1 arriving from peripheral circulation. This increase in IGF-1 immunoreactivity coincides with E2-induced changes in synaptic plasticity [10]. These authors have further demonstrated in adult rats, that estrogen-induced changes in tanycytes regulate the availability of IGF-1 to hypothalamic neurons [11,12]. Additionally, the synaptic and glial plasticity that occurs in the ARC/ME region prior to ovulation requires E2 and the activation of the IGF-1 receptor [4]. Several points are worth discussing which are directly related to the E2/IGF-1 interrelationship within the hypothalamus and LHRH/LH secretion. Present data suggest that E2 facilitates the important action of IGF-1 to stimulate

LHRH/LH release throughout pubertal development and further supports the fact that endogenous IGF-1 can stimulate LH release in the presence of high levels of E2. Our earlier report [15] showed that IGF-1-induced LH release during first proestrus and estrus, when E2 levels are rising, indicating that changes in gonadal steroids influence the actions of IGF-1 during pubertal development. Specifically, the present study further demonstrates the importance of E2 positive feedback in that priming the hypothalamus of 24day-old rats with E2 augments the ability of IGF-1 administered peripherally to advance the timing of female puberty, while IGF-1 or E2 alone were ineffective, or less effective, respectively, in that regard. Previously, we showed that central administration of IGF-1 advanced female puberty, but in those animals the IGF-1 administration was initiated on day 28, an age when E2 levels are beginning to rise [15]. We suggest that the increase in E2 is necessary to prime the Type 1 IGF-1Rs in the hypothalamus and ME prior to the pubertal rise in circulating IGF-1, an event which we showed previously to be accompanied by increased serum LH in both rats [15] and rhesus monkeys [7]. In conclusion, our results demonstrate a clear and significant interaction between the serum levels of E2, changes in IGF-1R mRNA and protein, and the ability of IGF-1 peptide to induce LH release and advance female puberty. While these collective results are important and extend our knowledge of E2 and IGF-1 interactions, more research is needed to determine precisely the mechanism of action that critically governs these events as puberty approaches.

Acknowledgements This work was supported by NIH grants AA-07216 (to W.L.D.) and ES-09106 (to the TAMU Center for Environmental and Rural Health).

References [1] C.L. Adam, P.A. Findlay, A.H. Moore, Effects of insulin-like growth factor-l on luteinizing hormone secretion in sheep, Anim. Reprod. Sci. 50 (1997) 45 – 56. [2] J. Antunes-Rodigues, S.M. McCann, Water, sodium chloride, and food intake induced injections of cholinergic and adrenergic drugs into the third ventricle of the rat brain, Proc. Soc. Exp. Biol. Med. 133 (1970) 1464 – 1469. [3] G.P. Cardona-Gomez, J.L. Trejo, A.M. Fernandez, L.M. GarciaSegura, Estrogen receptors and insulin-like growth factor-1 receptors mediate estrogen-dependent synaptic plasticity, NeuroReport 11 (2000) 1735 – 1738. [4] G.P. Cardona-Gomez, L. DonCarlos, L.M. Garcia-Segura, Insulin-like growth factor I receptors and estrogen receptors colocalize in female rat brain, Neuroscience 99 (2000) 751 – 760. [5] K.C. Copeland, T.J. Kuehl, V.D. Castracane, Pubertal endocrinology of the baboon: elevated somatomedin-C/insulin-like growth factor-1 at puberty, J. Clin. Endocrinol. Metab. 55 (1982) 1198 – 1201. [6] N. Danilovich, D. Wernsing, J.J. Coschigano, J.J. Kopchick,

J.K. Hiney et al. / Brain Research 1013 (2004) 91–97

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14] [15]

A. Bartke, Deficits in female reproductive function in GH-R-KO mice; role of IGF-1, Endocrinology 140 (1999) 2637 – 2640. W.L. Dees, G.A. Dissen, J.K. Hiney, F. Lara, S.R. Ojeda, Alcohol ingestion increased secretion of puberty-related hormones in the developing female Rhesus monkey, Endocrinology 141 (2000) 1325 – 1331. M. Duenas, S. Luquin, J.A. Chowen, I. Torres-Aleman, F. Naftolin, L.M. Garcia-Segura, Gonadal hormone regulation of insulin-like growth factor-1-like immunoreactivity in hypothalamic astroglia of developing and adult rats, Neuroendocrinology 59 (1994) 528 – 538. M. Duenas, I. Torres-Aleman, F. Naftolin, L.M. Garcia-Segura, Interaction of insulin-like growth factor-1 and estradiol signaling pathways on hypothalamic neuronal differentiation, Neuroscience 74 (1996) 531 – 539. M.C. Fernandez-Galaz, I. Torres-Aleman, L.M. Garcia-Segura, Endocrine-dependent accumulation of IGF-1 by hypothalamic glia, NeuroReport 8 (1996) 373 – 377. M.C. Fernandez-Galaz, E. Morschl, J.A. Chowen, I. Torres-Aleman, F. Naftolin, L.M. Garcia-Segura, Role of astroglia and insulin-like growth factor-1 in gonadal hormone-dependent synaptic plasticity, Brain Res. Bull. 44 (1997) 525 – 531. M.C. Fernandez-Galaz, F. Naftolin, L.M. Garcia-Segura, , Phasic synaptic remodeling of the rat arcuate nucleus during the estrous cycle depends on insulin-like growth factor-1 receptor activation, J. Neurosci. Res. 55 (1999) 286 – 292. D.J. Handelsman, J.A. Spaliviero, C.D. Scott, R.C. Baxter, Hormonal regulation of the peripubertal surge of insulin-like growth factor-1 in the rat, Endocrinology 120 (1987) 491 – 496. P.G. Harms, S.R. Ojeda, A rapid simple procedure for chronic cannulation of the rat jugular vein, J. Appl. Phys. 36 (1974) 391 – 392. J.K. Hiney, V. Srivastava, C.L. Nyberg, S.R. Ojeda, W.L. Dees, Insulin-like growth factor-1 of peripheral origin acts centrally to accel-

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

97

erate the initiation of female puberty, Endocrinology 137 (1996) 3717 – 3727. K.M. Michels, W.H. Lee, A. Seltzer, J.M. Saavedra, C.A. Bondy, Up-regulation of pituitary [125]insulin-like growth factor-1 (IGF-1) binding and IGF binding protein-2 and IGF-1 gene expression by estrogen, Endocrinology 132 (1993) 23 – 29. F. Pazos, F. Sanches-Franco, J. Balsa, J. Lopez-Fernandez, J. Escalada, L. Cacicedo, Regulation of Gonadal and somatotropic axis by chronic intraventricular infusion of insulin-like growth factor-1 antibody at the initiation of puberty in male rats, Neuroendocrinology 69 (1999) 408 – 416. S. Pons, I. Torres-Aleman, Estradiol modulates insulin-like growth factor 1 receptors and binding proteins in neurons from the hypothalamus, J. Neuroendocrinol. 5 (1993) 267 – 271. A. Quesada, A. Etgen, Insulin-like growth factor-1 regulation of a1adrenergic receptor signaling is estradiol dependent in the preoptic area and hypothalamus of female rats, Endocrinology 142 (2001) 599 – 607. A. Quesada, A. Etgen, Functional interactions between estrogen and insulin-like growth factor-1 in the regulation of a1B-adrenoceptors and female reproductive function, J. Neurosci. 22 (2002) 2401 – 2406. M.E. Wilson, IGF-1 administration advances the decrease in hypersensitivity to oestradiol negative feedback inhibition of serum LH in adolescent female rhesus monkeys, J. Endocrinol. 145 (1995) 121 – 130. M.E. Wilson, Premature elevation in serum insulin-like growth factor1 advances first ovulation in rhesus monkeys, J. Endocrinol. 158 (1998) 247 – 257. M.E. Wilson, Regulation of the growth hormone-insulin-like growth factor I axis in developing and adult monkeys is affected by estradiol replacement and supplementation with insulin-like growth factor I, J. Clin. Endocrinol. Metab. 83 (1998) 2018 – 2028.