INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

INSULIN-LIKE GROWTH FACTOR-I (IGF-I) is a polypeptide with similar structure and biological function to insulin. It is produced by virtually all tissues and can act in an endocrine as well as a paracrine/autocrine manner (27). The primary role of circulating IGF-I, which is derived mainly from the liver, is its inhibition of growth hormone (GH) secretion (53). This negative-feedback control of GH is well documented in humans, nonprimate mammals (4, 5, 51, 55), and more recently in fish species (6, 17, 42, 47). We discovered a novel role for IGF-I as a specific and direct stimulator of prolactin (PRL) release (17). IGF-I disparately alters the release of these two pituitary hormones through distinct cellular signaling pathways that likely involve activation of specific tyrosine kinase-linked receptors (16).

Physiological effects of IGF-I occur through high-affinity binding of IGF-I to the type-I IGF receptor. This receptor is a heterotetramer consisting of two - and two -subunits linked by disulfide bonds. It is a member of the receptor tyrosine kinase family which, when bound, causes autophosphorylation of tyrosine residues in the carboxyl end of the intracellular domain. This initial activation is followed by additional phosphorylations of intracellular protein kinases and eventually leads to a cellular response (32). In addition to IGF-I, other insulin-like peptides are known to bind the IGF-I receptor. In mammals, IGF-II binds with a 15-20-fold lower affinity to the IGF-I receptor, whereas in the few published reports on fish, it appears to bind with a similar affinity as IGF-I (6, 10, 11, 32, 44). In all vertebrates studied, insulin has a 100-1,000-fold lower affinity for this receptor (32, 44).

Pituitary IGF-I receptors have been identified in mammals (1, 9, 20, 34, 46), amphibians (8), and tumor cell lines (45, 54), as well as in a single species of fish, the rainbow trout (6). The sites of IGF-I action within the fish pituitary gland, however, have not been established. Although studies suggest IGF-I alters GH release by binding to receptors on somatotrophs (6, 45, 54, 55), evidence for direct mediation of IGF-I action on pituitary lactotrophs is lacking. Specifically, no studies to date have localized IGF-I receptors to pituitary PRL cells in vertebrates. The unique anatomy of the teleost pituitary, where PRL-secreting cells are segregated as a nearly homogeneous population in the rostral portion of the pars distalis, allows for easy assessment of IGF-I receptor affinity as well as localization to PRL cells within an intact tissue. In vitro autoradiography and immunohistochemistry were used to localize IGF-I receptors on tissue sections of hybrid striped bass (Morone saxatilis × M. chrysops) pituitaries. The affinity and specificity of the putative pituitary IGF-I receptor was characterized using quantitative microdensitometry.

Despite the established role of IGF-I in regulating PRL and GH synthesis and secretion in vertebrates as well as the multitude of studies demonstrating that IGF-I receptors are ubiquitously expressed (32), there have been few studies that have correlated IGF-I binding affinity to its potency in regulating both PRL and GH secretion. It is also unclear whether the disparate effects of IGF-I on PRL and GH secretion are modulated by IGF binding proteins (IGFBPs). Using des(1-3)IGF-I, an IGF-I analog that specifically binds with high affinity to IGF-I receptors but not IGFBPs, we investigated whether the actions of IGF-I on PRL and GH secretion may be modulated by IGFBPs.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals. Juvenile hybrid striped bass (Morone saxatilis × M. chrysops; body wt 30-60 g) were obtained from freshwater ponds at the Pamlico Aquaculture Field Laboratory of North Carolina State University or from the Vernon James Research and Extension Center (Plymouth, NC). All fish were maintained for at least 3 wk in recirculating tanks supplied with freshwater (22 ± 2°C) under simulated natural photoperiod conditions before experiments. Fish were fed ad libitum twice daily with a pelleted feed (Southern States, Richmond, VA). All experimental procedures were performed in accordance with the NIH Guiding Principles for Care and Use of Laboratory Animals and were approved by the North Carolina State University Animal Care and Use Committee.

Hormones and antibodies. Recombinant human (rh) IGF-I, rhIGF-II, and des(1-3)IGF-I were purchased from Gropep Pty (Adelaide, Australia) and porcine insulin was from Sigma Chemical (St. Louis, MO). 3-[¹²⁵I]iodotyrosyl-rhIGF-I was purchased from Amersham (Arlington Heights, IL). The hybrid striped bass (hsb) PRL antibody was kindly provided by Dr. Craig Sullivan (North Carolina State University). This homologous antibody was previously characterized and shown to specifically bind hsbPRL and not other pituitary hormones by immunohistochemistry, radioimmunoassay (25), and Western blotting procedures (17, 26). The IGF-I receptor antibody (C-20) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). This antibody was raised against a portion of the human IGF-I receptor (carboxy terminus), which is highly conserved among vertebrates. A blocking peptide corresponding to the carboxy terminus of the IGF-I receptor (Santa Cruz Biotechnology) was used to verify specificity of the antibody for pituitary bass IGF-I receptors.

Localization of IGF-I saturable binding. Radioligand binding and in vitro autoradiography studies were performed to localize and characterize IGF-I binding sites on hybrid striped bass pituitary glands using a modified protocol (8, 34). Pituitaries were collected, embedded in Tissue-Tek OCT compound (Miles, Elkhart, IN) and immediately frozen on dry ice. The glands were serially sectioned at 20 µm, thaw-mounted onto Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA), and stored desiccated at 80°C. The slide-mounted tissue sections were brought to room temperature and preincubated for 15 min in 25 mM Tris · HCl buffer (pH 7.4) that contained 10 mM MgCl₂, 0.1% BSA, and 1 mg/ml bacitracin. After preincubation, the slides were dried in a stream of air at 4°C for 30 min and placed under vacuum with desiccant for an additional 45 min at room temperature. The slides were then incubated in the preincubation buffer (containing 100 pM ¹²⁵I-rhIGF-I) for 2 h at 22 ± 2°C, the temperature at which the fish were maintained. Preliminary investigations established this concentration of IGF-I as saturable, similar to that shown in other autoradiographic binding studies (9, 34), and within the same order of magnitude as that used in fish membrane-binding studies (11) or semipurified receptor extracts (22, 33). Nonsaturable binding was defined as binding in the presence of 1 µM nonradioactive rhIGF-I. After incubation, slides were washed three times for 1 min in Tris · HCl buffer (pH 7.4) at 4°C and rinsed twice in distilled water at 4°C for 5 s each. The slides were dried under a stream of air at 4°C for 1 h and stored in desiccant at room temperature overnight. The next day, the slides were exposed to -max Hyperfilm (Amersham, Arlington Heights, IL) for 6 days. The film was then developed with Kodak D-19 developer and fixed. Slides were stained with hematoxylin and eosin and coverslipped.

Characterization of IGF-I specific binding. Quantitative microdensitometry of the silver grains in the autoradiograms was performed using the Bioquant Meg IV system (R and M Biometrics, Nashville, TN) with a charge-coupled device-7 solid-state video camera mounted on a Wild M5A stereomicroscope with a light-field base. Images were digitized by a Targa M8 image-capture board. Because of the nonlinear sensitivity of the film, ¹²⁵I-microscale standards (Amersham) were exposed to Hyperfilm together with the tissue sections and were used to correct the sample densities. Saturable binding was obtained by subtracting the density of nonsaturable binding (binding in the presence of 1 µM nonradioactive rhIGF-I) from the density measured in the total binding samples. Competitive inhibition curves were constructed by incubating serial tissue sections with ¹²⁵I-rhIGF-I (100 pM) with increasing concentrations of nonradioactive rhIGF-I, rhIGF-II, porcine insulin, and des(1-3)IGF-I (0.01-1,000 nM). To estimate the concentration of nonradioactive IGF-I, IGF-II, insulin, and des(1-3)IGF-I that inhibited 50% of saturable binding (IC₅₀), data were analyzed by nonlinear regression using the Prism software program (GraphPad Software, San Diego, CA).

Immunohistochemistry. Pituitary glands were collected, fixed in 4% paraformaldehyde and embedded in paraffin. Pituitary glands were cut into 4-µm sections. Sequential sections were used for IGF-I receptor and PRL staining. Slides were deparaffinized and blocked with 3.0% hydrogen peroxide followed by 10% normal goat serum, then washed in phosphate-buffered saline (PBS). Serial sections were later incubated overnight at 4°C with optimal dilutions of primary antibody (a 1:500 dilution for hsbPRL, see Ref. 25; and a 1:50 dilution for IGF-I receptor) that contained 1.0% normal goat serum and Triton X-100. Slides were washed in PBS with Triton X-100 and treated with a biotinylated goat anti-rabbit antibody (Vector Laboratories, Burlingame, CA) for 60 min. Binding of the primary antibodies was detected by the avidin-biotinylated peroxidase method (VectaStain kit, Vector Laboratories). Slides were counterstained with hematoxylin, dehydrated in ethanol, coverslipped, and analyzed using light microscopy. Negative controls were prepared using serial sections and replacing primary antibody with normal goat serum. In addition, the IGF-I receptor antibody was preabsorbed with excess IGF-I receptor peptide to show specificity of IGF-I receptor staining.

Static tissue cultures and gel electrophoresis. Pituitary glands were collected, dissected, and incubated according to our previously described methods (17). Briefly, pituitaries were dissected into rostral pars distalis (RPD, containing ~95% PRL cells) and proximal pars distalis/pars intermedia (PPD/PI, containing GH and other pituitary cell types) and placed in defined medium consisting of a Krebs bicarbonate Ringer solution containing glucose, L-glutamine, and MEM essential amino acids (without L-glutamine, GIBCO). The control media was adjusted to 325 mosmolal, which represents the blood osmotic pressure of freshwater bass (Morone spp.). Tissues were preincubated for 2 h with control medium. After the pretreatment, media was removed and replaced with experimental media. Tissues were incubated on a gyratory platform (60 rpm) for 18-20 h at 25 ± 1°C in a humidified atmosphere containing 95% O₂-5% CO₂.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

IGF-I binding was widely distributed throughout the pituitary gland of hybrid striped bass (Fig. 1). All three anatomically distinct regions of the pituitary gland: RPD (containing predominantly PRL cells), PPD (containing GH, thyrotropin-stimulating hormone, and gonadotropin cells), and PI (containing melanocyte stimulating hormone and somatolactin cells) contain high-affinity binding sites for IGF-I. Nonradioactive IGF-II (1 µM) was able to displace ¹²⁵I-IGF-I from its receptor, whereas a similar dose of insulin could not (Fig. 2).

View larger version (9K): [in this window] [in a new window]   Fig. 1.   Autoradiographic localization of 125I-insulin-like growth factor (IGF)-I binding sites in the hybrid striped bass pituitary gland. a: bright-field photomicrograph of a hematoxylin and eosin (H&E) stained section of a hybrid striped bass pituitary gland incubated with 0.1 nM 125I-IGF-I. b: dark-field autoradiogram of the tritium-sensitive film that overlaid a. Areas of 125I-IGF-I binding are illuminated white. c: autoradiogram of a section adjacent to and treated identically to b, except that excess nonradioactive IGF-I (1 µM) was added to the incubation medium to reveal nonsaturable binding (NSB). Saturable binding is the difference in radioactivity between b and c. RPD, rostral pars distalis; PPD, proximal pars distalis; PI, pars intermedia; TB, total binding.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Autoradiographic analysis of the effects of recombinant human (rh) IGF-II and porcine insulin on the displacement of 125I-IGF-I binding in the hybrid striped bass pituitary. a, d: bright-field photomicrographs of a hematoxylin and eosin-stained section of a hybrid striped bass pituitary gland incubated with 0.1 nM 125I-IGF-I. b, e: dark-field autoradiograms of the tritium-sensitive film that overlaid a and d, respectively. Areas of 125I-IGF-I binding are illuminated white. c, f: autoradiograms of sections adjacent to and treated identically to TB (b or e), except that excess nonradioactive IGF-II (1 µM) or insulin (1 µM) was added to the incubation medium. Note that IGF-II completely blocks 125I-IGF-I binding, whereas insulin does not.

To determine the specificity and affinity of the IGF-I binding site, competitive inhibition curves were generated using ¹²⁵I-labeled rhIGF-I and increasing concentrations of competing nonradiolabeled rhIGF-I, rhIGF-II, and porcine insulin. At concentrations ranging from 0.01 to 1,000 nM, IGF-I inhibited saturable binding of ¹²⁵I-IGF-I. The binding data were analyzed by nonlinear regression and results indicated the presence of a single-class, high-affinity binding site. The dose that half maximally displaced ¹²⁵I-IGF-I from its receptor (IC₅₀) was 2.9 nM (Fig. 3). Nonradioactive IGF-II was as equipotent as IGF-I in inhibiting saturable ¹²⁵I-IGF-I binding. Insulin was altogether ineffective at displacing IGF-I from its receptor within a concentration range as high as 1 µM (Fig. 3). The IC₅₀ values for IGF-II and insulin were 3.9 and >1,000 nM, respectively. Unlabeled IGF-I also displaced ¹²⁵I-IGF-I binding on the RPD (data not shown). The IC₅₀ for this portion of the pituitary containing 95-99% PRL cells was 4 nM, which is similar to that seen with the entire gland. Consistent with IGF-I binding on the RPD, immunohistochemistry colocalized IGF-I receptors and PRL in the same cells of the bass pituitary gland (Fig. 4, c and d). Binding of the IGF-I receptor antibody on PRL cells was completely displaced when preabsorbed with IGF-I receptor peptide, which shows that IGF-I receptor staining was specific (Fig. 4b). As previously demonstrated by Jackson (25), PRL immunoreactivity was restricted to the RPD, where PRL cells are naturally segregated.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Competitive inhibition of 125I-IGF-I saturable binding by nonradioactive rhIGF-I, rhIGF-II, and porcine insulin in the hybrid striped bass pituitary gland. Serial sections were incubated with 0.1 nM 125I-IGF-I and graded concentrations of nonradioactive peptides.

View larger version (130K): [in this window] [in a new window]   Fig. 4.   Immunohistochemical staining of serial sections of the RPD of the bass pituitary gland with IGF-I receptor (c) and PRL (d) antibodies. Positive staining for these proteins appears as a brown color. Staining was specific as demonstrated by the lack of staining with normal serum (a) and when the IGF-I receptor antibody was preabsorbed with a fivefold excess of IGF-I receptor peptide (b). Note the tremendous overlap in IGF-I receptor and PRL staining within the same cells throughout the RPD.

In a separate experiment, tissue sections were incubated with ¹²⁵I-labeled rhIGF-I and varying doses of nonradiolabeled des(1-3)IGF-I, a truncated analog of IGF-I that binds with high affinity to the IGF-I receptor but weakly to IGFBPs. Des(1-3)IGF-I was just as effective as the full-length IGF-I in displacing ¹²⁵I-IGF-I (Fig. 5). In this experiment, the IC₅₀ for IGF-I binding was similar to that shown for IGF-I in the previous experiment (see Fig. 3).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Competitive inhibition of 125I-IGF-I saturable binding by nonradioactive rhIGF-I and des(1-3)IGF-I in the hybrid striped bass pituitary gland. Serial sections were incubated with 0.1 nM 125I-IGF-I and graded concentrations of the nonradioactive peptides.

The effects of des(1-3)IGF-I versus IGF-I were evaluated to determine whether IGF-I might regulate PRL and GH release through activation of a specific IGF-I receptor, high-affinity IGFBPs, or both. Dissected RPD and PPD/PI were exposed separately to graded concentrations of both IGF-I and the analog. During 18-20-h incubations, des(1-3)IGF-I markedly inhibited GH release in a dose-dependent fashion at concentrations ranging from 10 to 1,000 ng/ml (P < 0.0001; Fig. 6, top). The lowest concentration of des(1-3)IGF-I that significantly inhibited GH release was 10 ng/ml, whereas that for IGF-I was 100 ng/ml. In contrast to GH, PRL release from the RPD increased in a dose-dependent manner in response to des(1-3)IGF-I and became significantly different from controls at 10 ng/ml (P < 0.001). Compared with des(1-3)IGF-I, a 10-fold higher dose of IGF-I was required to augment PRL release (Fig. 6, bottom).

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Effect of rhIGF-I and des(1-3)IGF-I on hybrid striped bass GH (top) and PRL (bottom) release during an 18-20-h static culture. Each bar represents the mean ± SE of 6-8 tissues/treatment from three separate experiments (i.e., total of 18-24 tissues/treatment). Values are normalized as percent of control. * P < 0.05, *** P < 0.0001, significant differences from control.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

This investigation is the first to localize and characterize IGF-I receptors on teleost pituitary glands using in vitro receptor autoradiography and quantitative microdensitometry while simultaneously correlating the affinity of the IGF-I receptor to an established physiological action, namely, the regulation of PRL and GH secretion. This study clearly demonstrates that saturable, specific, high-affinity IGF-I binding sites are located on hybrid striped bass pituitary glands, including those areas containing GH- and PRL-secreting cells. IGF-II and the truncated IGF-I analog des(1-3)IGF-I showed equal potency as IGF-I in displacing ¹²⁵I-IGF-I from its receptor, whereas insulin was unable to compete for IGF-I binding sites at concentrations as high as 1 µM. Des(1-3)IGF-I, like the full-length peptide, was effective in stimulating PRL and inhibiting GH release, which provides additional evidence that IGF-I elicits its disparate effects through activation of IGF-I receptors. The more potent responses seen with des(1-3)IGF-I compared to IGF-I point to the possibility that the regulation of PRL and GH secretion by IGF-I may be modulated by the presence of one or more IGFBPs.

Recently we showed that IGF-I disparately regulates PRL and GH synthesis and release from fish pituitary glands. These actions occur through distinct as well as overlapping cellular signaling pathways that involve regulation of phosphatidylinositol 3-kinase and mitogen-activated protein kinase (16, 17). Before these investigations, it was unknown whether IGF-I actions were mediated through high-affinity receptors located within distinct regions of the fish pituitary gland containing PRL and GH cells. IGF-I receptors are widely expressed in vertebrate tissues including the pituitary gland (1, 27, 32). In fish, IGF-I binding has been demonstrated in several tissues (7, 11, 22, 23, 31, 36, 41, 44) including one study that reported IGF-I binding to pituitary glands (6). However, localization of IGF-I receptors within these tissues has been limited mainly due to the method of tissue preparation. With the exception of one study that examined IGF-I receptors in goldfish retina (7), most investigations, including that shown in trout pituitaries, characterized IGF-I receptors using membrane preparations or semipurified receptor extracts. These techniques, although well established, do not allow for localization of receptors within the tissues examined nor do they characterize binding in tissues in their normal, in situ configurations. Using thinly sectioned, frozen tissue preparations, we clearly show that IGF-I specifically binds IGF-I receptors throughout the entire pituitary gland, including the RPD and PPD or PRL- and GH-containing cell regions, respectively. IGF-I binds a single high-affinity site that presumably reflects a receptor based on regression analysis of competitive displacement curves and results from des(1-3)IGF-I, which is further discussed below.

Owing to the heterogeneous distribution of cell types within most vertebrate pituitaries, it has been difficult to localize receptors to specific cell types while correlating IGF-I binding affinities to the regulation of pituitary hormones by IGF-I. Fish anterior pituitary glands, unlike those of most other vertebrates including mammals, are segregated into distinct regions containing discrete cell types. This unique morphological characteristic of the teleost pituitary allows for the in vitro study of aggregated populations of distinct cell types (24, 37). We exploited this model system and localized high-affinity IGF-I binding sites exclusively in the RPD where there is a nearly homogeneous population (>95%) of PRL cells (21, 24). The binding was widespread throughout the RPD and of similar affinity (~3 nM) as that shown in the whole pituitary (~4 nM), which suggests specific receptors are likely important for mediating the IGF-I induction of PRL release. Colocalization of IGF-I receptors to PRL immunoreactive cells further indicates that IGF-I is likely to alter PRL release by directly acting on lactotrophs.

Competitive inhibition of saturable ¹²⁵I-IGF-I binding by nonradioactive IGF-I, IGF-II, and insulin revealed a pattern of specificity for IGF-I. We found that IGF-I and IGF-II bind with high affinity to hybrid striped bass pituitary glands (IC₅₀ = 3 and 4 nM, respectively) whereas insulin resulted in a much reduced affinity and was unable to completely displace ¹²⁵I-IGF-I from its receptor even at the highest dose measured (1 µM). A previous study in rainbow trout showed high-affinity IGF-I binding in a pituitary membrane preparation (K_d = 0.23 ± 0.73; Ref. 6). The discrepancy between this study and that reported here might be associated with differences in tissue preparation, receptor binding protocols, and/or species. The binding affinity of IGF-I for its receptor seen in the present study is similar to those found in mammals and in other investigations where radioreceptor autoradiography was used rather than binding in membrane preparations or semipurified receptor extracts.

The binding affinity of IGF-I in hybrid striped bass RPD as well as whole pituitaries correlates well with the EC₅₀ values over which IGF-I was shown to effectively augment PRL and inhibit GH release (4.6 and 3.8 nM, respectively; Ref. 17). A displacement of 50% of IGF-I from its receptor and the concentration of IGF-I that half-maximally stimulates PRL and inhibits GH are also well within the range of plasma IGF-I levels (free and bound to IGFBPs) recently measured in hybrid striped bass (50-300 ng/ml or 6.5-39 nM; unpublished data). In addition to this circulating source, IGF-I is also produced locally in hybrid striped bass pituitary glands (unpublished data; Ref. 15). This local source in addition to circulating IGF-I would result in substantially higher IGF-I concentrations that may activate pituitary receptors and cause release of these hormones. Taken together, these results further support the physiological significance of IGF-I as a disparate regulator of PRL and GH cell function.

In mammals, two types of IGF-I receptors have been characterized: IGF-I and IGF-II/mannose 6-phosphate (IGF-II/M6P) receptors. Although the precise function of the IGF-II/MP6 receptor is still uncertain, it is thought to play a role in internalization and degradation of IGF-II (27, 39, 50). Despite many attempts to identify a type II IGF receptor in oviparous species, it has only recently been shown in trout embryos that IGF-II may bind an IGF-II/MP6 receptor (35). The physiological significance of this receptor is still unclear. Similar to mammalian IGF-II/MP6 receptors, it lacks tyrosine kinase activity and therefore may have a similar function of binding and degrading IGF-II to regulate the mitogenic effects of this growth factor. Although IGF-II/MP6 receptors may exist in all vertebrates, the physiological actions of IGFs (both IGF-I and IGF-II) in fish, as in mammals, are thought to be mediated by the type I IGF receptor (11, 32, 43). Insofar as the pituitary is concerned, results from this and another study in rainbow trout (6) show that IGF-II and IGF-I bind with similar affinities whereas insulin binds with a 100-1,000-fold lower affinity to the IGF-I receptor. In mammals, IGF-II binds with a 15-20-fold lower affinity to the IGF-I receptor, and this is typically followed by a less-potent cellular response compared with IGF-I. Despite the similar affinities for the IGF-I receptor, we found that IGF-II was 10-20-fold less potent than IGF-I in regulating the release of both PRL and GH (17). There is evidence suggesting that IGF-II binds a site distinct from IGF-I (18). Thus it is possible that IGF-II binds the type I IGF receptor with high affinity but is unable to activate the appropriate intracellular messengers critical to hormone release. Future studies should assess whether IGF-II may block IGF-I-evoked PRL and GH release and/or whether it is ineffective in activating intracellular protein kinases that lead to the stimulation of PRL and inhibition of GH by IGF-I.

Binding proteins (at least six in mammals and four in fish) are carriers of circulating IGF-I that regulate availability of the growth factor and alter its bioactivity at specific target tissues. Binding proteins also are known to associate with cell membranes either by interacting with components in the extracellular matrix or binding membrane receptor sites (28, 29, 40). Des(1-3)IGF-I is a naturally occurring, truncated analog of IGF-I. Removal of the first three amino acids allows for high-affinity binding to IGF-I receptors but weak binding to IGFBPs (14, 38). In this study, des(1-3)IGF-I inhibited saturable ¹²⁵I-rhIGF-I binding in a dose-dependent manner and was equally potent as IGF-I in doing so, whereas it was at least 10 times more effective than IGF-I in stimulating PRL and inhibiting GH release. These results demonstrate that IGF-I is likely binding with high affinity to IGF-I pituitary receptors to regulate PRL and GH release.

Although IGFBPs have been shown to modulate the mitogenic activity of IGF-I in various tissues, virtually nothing is known regarding the function of IGFBPs in regulating IGF-I-evoked secretory responses generally and pituitary hormone secretion specifically. Voss et al. (52) recently found that IGF-I inhibited GH mRNA expression over a narrow range of concentrations in MtT/S cells (a rat somatotroph cell line that produces GH but not PRL). This ultrasensitive effect of IGF-I on GH mRNA was reduced when an IGF-I analog that binds the receptor with high affinity but with low affinity for IGFBPs was added, which suggests that IGFBPs were responsible for modulating the actions of IGF-I on GH mRNA expression at least in this clonal cell line. Our findings in physiologically relevant "normal" cells indicate that IGFBPs may also modulate the ability of IGF-I to regulate the release of GH as well as PRL because des(1-3)IGF-I had a significantly greater potency than the full-length IGF-I in eliciting pituitary responses. Studies examining protein synthesis in L6 myoblasts show that the increased biological potency occurring with des(1-3)IGF-I is not likely due to glycosylation or other modifications to the growth factor nor to increased affinity for the type I receptor (2, 3, 13). The latter is supported by our studies demonstrating similar competitive inhibition curves for the truncated analog and full-length growth factor. We postulate that pituitary IGFBPs complex with IGF-I and prevent its access to the receptor, which would result in a less-potent response than that seen with des(1-3)IGF-I, which binds weakly to IGFBPs and therefore has more continuous access to the receptor. In mammals, all six IGFBPs have been localized to the anterior pituitary gland (1, 19). Based on Western blot analysis, pituitary production of at least one IGFBP (a 28-30 kDa, IGFBP-2-like protein) has been demonstrated in striped bass (48). In mammals as well as zebrafish, this IGFBP has been shown to inhibit IGF-I-mediated action (12, 29). Thus it is possible that an IGFBP-2-like molecule produced by bass pituitary glands reduces PRL and GH sensitivity to IGF-I by modulating the accessibility of IGF-I to its receptor.

In summary, this study demonstrates localization of a specific, high-affinity IGF-I receptor to PRL cells and throughout the entire teleost pituitary gland. Pituitary presence and a strong correlation between receptor affinity and IGF-I activity provide powerful evidence that IGF-I binds the PRL- and GH-containing regions of the pituitary to directly and differentially regulate PRL and GH secretion. It also appears that IGFBPs significantly modulate IGF-I-evoked PRL and GH cell responses in the pituitary gland. Future studies will examine whether the type I IGF receptor and IGFBPs might be regulated by conditions known to enhance or impair these two pituitary hormones.