Insulin-Like Growth Factor-I (IGF-I) Attenuation of Growth Hormone Is Enhanced by Overexpression of Pituitary IGF-I Receptors
Hironori Yamasaki, Diane Prager, Saba Gebremedhin, and Shlomo Melmed Department of Medicine Cedars-Sinai Medical Center-University of California Los Angeles School of Medicine Los Angeles, California 90048
directly regulates the pituitary secretion of GH by participating in a negative feedback regulatory loop both at the level of the hypothalamus as well as directly at the pituitary (3, 4). This feedback regulation may interact directly with the effects of hypothalamic GH-releasing hormone and somatostatin, respectively, on the somatotroph (5, 6). IGF-I attenuates transcription of the rat and human GH genes (7), suppresses basal and stimulated GH mRNA levels (8, 9), and suppresses the secretion of GH (10, 11). Specific binding of IGF-I has been characterized in normal and clonal rat pituitary cells (12). Abundant IGF-l-binding sites (Kd,~ 0.5 nM) have been demonstrated in rat anterior pituitary cells (13, 14), and their number appears to be regulated in vivo (15). IGF-I suppression of GH secretion is also dependent upon the presence of IGF-I receptors in pituitary tumor cells (16). These studies suggest that the IGF-I receptor plays an important role in the feedback regulation of the somatotroph cell. To further evaluate mechanisms for the signal transduction of IGF-I to the GH gene, a human IGF-I receptor cDNA was overexpressed in pituitary cells, and their responsiveness to IGF-I was assessed.
Insulin-like growth factor-l (IGF-I) attenuates GH gene expression by a receptor-mediated mechanism in pituitary cells. We, therefore, isolated neomycin-resistant stable GC cell transfectants overexpressing human IGF-I receptor cDNA (IGFIRcDNA) cloned in an Rous sarcoma virus-directed expression vector. A transfection control contained the IGFIR-cDNA cloned in the reverse orientation. Southern analysis confirmed incorporation of human IGFIR-cDNA sequences into rat genomic DNA. Immunoprecipitation of metabolically labeled [35S]methionine stably transfected cells revealed a 200-kDa human IGF-I receptor precusor protein. Growth rate and basal GH secretion were not altered in transfected cells. Although transfected and control cells had a similar Kd for IGF-I binding (0.43 and 0.40 nwi, respectively), IGF-l-binding sites were induced 17fold (384,000 vs. 22,000 sites/cell). Treatment of cells with IGF-I (6.5 nivi) maximally attenuated GH secretion by 80% compared to 40% attenuation in control cells (P < 0.0001). Maximal suppression of GH in transfectants occurred within 15 h of treatment, and GH secretion by control cells was only maximally suppressed after 42 h. The ED50 of IGF-I suppression of GH secretion in transfectants after 15 h was 0.5 nM. These results demonstrate that transfectants overexpressing human IGF-I receptor are hyperresponsive to exogenous IGF-I. These data indicate that IGF-I receptor number plays an important role in mediating the signal transduction of IGFI to the GH gene. (Molecular Endocrinology 5: 890896,1991)
RESULTS Selection of Stable Transfectants GC cells were transfected with an IGF-I receptor cDNA (pRSV-IGFIR; Fig. 1). To assess the integration of the exogenous IGF receptor cDNA plasmid into rat GC cell genomic DNA, a Southern blot analysis of transfectant cells was performed. A construct in the reverse orientation, (pRSV-rlGFIR) serving as a transfection control was also analyzed. The results of DNA hybridizations obtained from nine separate colonies isolated after transfection is shown in (Fig. 2). SamHI- or EcoRIdigested genomic DNA isolated from the control cells (wild type and clones R1 through R6) and transfectants (clones S1, S20 and S21) was hybridized with a 32P-
INTRODUCTION Peripheral insulin-like growth factor-l (IGF-I) mediates many of the growth effects of GH (1, 2). IGF-I also 0888-8809/91 /0890-0896$03.00/0 Molecular Endocrinology Copyright © 1991 by The Endocrine Society
890
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 06:21 For personal use only. No other uses without permission. . All rights reserved.
891
IGF-I Attenuation of GH
EcoRI
labeled human IGF-I receptor cDNA probe. The receptor probe hybridized to all transfectant DNA samples tested, while the probe did not hybridize to DNA extracted from wild-type untransfected GC cells. The different sized bands confirmed the integration of human IGF-I receptor DNA sequences into the genomic DNA of these cells. Ndel
BamH I
IGF-1 Rec cDNA (4.2 Kb)
Synthesis of Human IGF-I Receptor by Transfected Cells To confirm the presence of human IGF-I receptor protein synthesized by the forward transfectants (S20 and S21), lysates of [35S]methionine-labeled transfected and wild-type cells were immunoprecipitated by «IR3 antibody, which specifically recognizes human IGF-I receptor and not the rat IGF-I receptor. A 200-kDa band was observed in the transfectants, representing the unprocessed IGF-I receptor precursor, and smaller degradation products were also observed. alR3 cross-reacted minimally with wild-type GC cell proteins (Fig. 3).
Xhol
31
IGF-I Binding to Transfected Cells
Fig. 1. Schematic Diagram of the Plasmid Used for Transfection A 4.2-kb Hinc\\Sst\ fragment of the human IGF-I receptor cDNA was cloned into the polylinker site of mammalian expression vector Rexp.
Binding of [125I]IGF-I was tested in wild-type GC cells, reverse transfected control cells, R2, and forward transfectants, S20 and S21. Figure 4A shows that increasing
Mrx10" 3 — 200 -•-97 S1 R1 R2 R3 R4 R5 R6
Kb
S20S21
Kb
•5.0 •5.0
•3.0
•3.0
A
B
Fig. 2. Southern Blot Analysis of Genomic DNA from Stable Transfectants Containing pRSV-IGFIR (S1, S20, and S21) and pRSV-rlGFIR(R1-R6) Wild-type cells were untransfected. DNA isolation and blot hybridization were performed as described in Materials and Methods. Digestion of DNA with BamH\ (A) and EcoRI (B) was followed by hybridization to the separated DNA with the human IGF-I receptor cDNA. Size markers (kilobases) are indicated on the right.
Fig. 3. Immunoprecipitation of [35S]Methionine-Labeled IGF-I Receptor Protein Metabolically labeled GC and transfectant cells were immunoprecipitated with a monoclonal antibody to the human IGFI receptor, alR3. The immunoprecipitates were analyzed on denaturing SDS-polyacrylamide gels, and an autoradiograph of the gel is depicted. 14C-Labeled mol wt (Mr) markers are indicated.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 06:21 For personal use only. No other uses without permission. . All rights reserved.
MOL ENDO-1991 892
Vol 5 No. 7
D O
wild type R2
derived from the intercept of the abscissa was increased 17-fold in S20 (384,000 sites/cell) and 10-fold in S21 (234,000 sites/cell), respectively, compared to that in wild-type cells. These observations further confirm that S20 and S21 overexpress an IGF-I receptor. Effects of IGF-I on GH Secretion
(3
2-
10
B
1 00
(nM)
Fig. 4. Binding of [125I]IGF-I to Wild-Type GC Cells, R2 Control Cells (A), and S20 and S21 Transfected Cells (B) Cells were incubated with increasing doses of IGF-I, and binding of [125I]IGF-I was performed in suspension at 15 C for I80 min. Each point represents mean of three samples representative of two separate experiments.
amounts of unlabeled IGF-I displaced [125I]IGF-I binding from wild-type GC and R2 control cells. The specific binding of IGF-I was similar in these two control cell types, 7.6% and 6.9%, respectively. Fifty percent displacement of maximal binding was achieved with 0.50.6 nM IGF-I. This similarity suggests that the transfection procedure itself did not alter the binding activity of IGF-I. Figure 4B shows that increasing amounts of unlabeled IGF-I displaced [125I]IGF-I binding from forward transfectants S20 and S21. In contrast to wildtype GC and R2 control cells, however, the specific binding in S20 and S21 cells was markedly increased to 55% and 47%, respectively. Fifty percent displacement of the maximal binding was achieved with concentrations of IGF-I necessary for binding to wild-type GC and R2 control cells, ranging from 0.5-1 nM. When these binding displacement data were subjected to Scatchard analysis, a linear plot was obtained for all cell types, consistent with the presence of a single class of high affinity receptors in these pituitary cells. The distribution constant (Kd) for [125I]IGF-I binding derived from the slope of the plot was similar in all four cell types (Fig. 5). The number of [125I]IGF-I receptors
To test the time course of IGF-I on GH secretion, wildtype, R2, and S20 cells growing in serum-free defined medium were treated with 6.5 nM IGF-I for up to 42 h. Basal GH secretion by untreated S20 cells was not lower than that in wild-type cells. Wild-type and R2 control cells showed similar suppression of GH to 60% of that in control untreated cells by the end of the incubation period. In contrast, GH secretion by the forward transfectant S20 overexpressing the human IGF-I receptor was suppressed to 20% of control secretion by the end of the incubation period. IGF-I suppressed GH secretion at earlier time points in S20 compared to wild-type and R2 control cells, with maximal GH suppression evident after only 15 h of incubation (Fig. 6). The dose response of the effects of IGF-I on GH secretion was, therefore, tested after 15 h of treatment (Fig. 7). IGF-I suppressed GH secretion in S20 cells in a dose-dependent manner, but GH secretion was not appreciably altered in wild-type and R2 control cells at this time. Fifty percent of the maximal GH suppression was achieved with 0.5 nM IGF-I in S20 cells, a concentration similar to the Ko for IGF-I binding in these cells.
DISCUSSION
The human IGF-I receptor was stably transfected into GC rat pituitary tumor cells. The fidelity of the expression of human IGF-I receptors was confirmed by demonstrating integration of IGF receptor sequences into genomic DNA and by immunoprecipitation of IGF-I receptor proteins. Those transfectants overexpressing human IGF-I receptor also exhibited enhanced responsiveness to IGF-I. IGF-I is an extracellular growth factor regulating cellular proliferation (17). Alterations in growth factor receptor activity can result in uncontrolled growth and neoplastic transformation. For example, overexpressed epidermal growth factor receptors (18) and IGF-I receptors (19) cause ligand-dependent cellular transformation in NIH 3T3 cells. Modification of epidermal growth factor receptor structure can also lead to mitogenic activation (20, 21). In contrast to these studies, overexpression of IGF-I receptors did not promote cellular growth of GC pituitary cells, even at relatively high concentrations of IGF-I (data not shown). Possibly, further overexpression of IGF-I receptors or other unknown cell-specific factors is required to stimulate proliferation of these cells. However, these cells are already transformed and terminally differentiated; therefore, a proliferative effect
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 06:21 For personal use only. No other uses without permission. . All rights reserved.
893
IGF-I Attenuation of GH
150-
100-
Kd
Hironori Yamasaki, Diane Prager, Saba Gebremedhin, and Shlomo Melmed Department of Medicine Cedars-Sinai Medical Center-University of California Los Angeles School of Medicine Los Angeles, California 90048
directly regulates the pituitary secretion of GH by participating in a negative feedback regulatory loop both at the level of the hypothalamus as well as directly at the pituitary (3, 4). This feedback regulation may interact directly with the effects of hypothalamic GH-releasing hormone and somatostatin, respectively, on the somatotroph (5, 6). IGF-I attenuates transcription of the rat and human GH genes (7), suppresses basal and stimulated GH mRNA levels (8, 9), and suppresses the secretion of GH (10, 11). Specific binding of IGF-I has been characterized in normal and clonal rat pituitary cells (12). Abundant IGF-l-binding sites (Kd,~ 0.5 nM) have been demonstrated in rat anterior pituitary cells (13, 14), and their number appears to be regulated in vivo (15). IGF-I suppression of GH secretion is also dependent upon the presence of IGF-I receptors in pituitary tumor cells (16). These studies suggest that the IGF-I receptor plays an important role in the feedback regulation of the somatotroph cell. To further evaluate mechanisms for the signal transduction of IGF-I to the GH gene, a human IGF-I receptor cDNA was overexpressed in pituitary cells, and their responsiveness to IGF-I was assessed.
Insulin-like growth factor-l (IGF-I) attenuates GH gene expression by a receptor-mediated mechanism in pituitary cells. We, therefore, isolated neomycin-resistant stable GC cell transfectants overexpressing human IGF-I receptor cDNA (IGFIRcDNA) cloned in an Rous sarcoma virus-directed expression vector. A transfection control contained the IGFIR-cDNA cloned in the reverse orientation. Southern analysis confirmed incorporation of human IGFIR-cDNA sequences into rat genomic DNA. Immunoprecipitation of metabolically labeled [35S]methionine stably transfected cells revealed a 200-kDa human IGF-I receptor precusor protein. Growth rate and basal GH secretion were not altered in transfected cells. Although transfected and control cells had a similar Kd for IGF-I binding (0.43 and 0.40 nwi, respectively), IGF-l-binding sites were induced 17fold (384,000 vs. 22,000 sites/cell). Treatment of cells with IGF-I (6.5 nivi) maximally attenuated GH secretion by 80% compared to 40% attenuation in control cells (P < 0.0001). Maximal suppression of GH in transfectants occurred within 15 h of treatment, and GH secretion by control cells was only maximally suppressed after 42 h. The ED50 of IGF-I suppression of GH secretion in transfectants after 15 h was 0.5 nM. These results demonstrate that transfectants overexpressing human IGF-I receptor are hyperresponsive to exogenous IGF-I. These data indicate that IGF-I receptor number plays an important role in mediating the signal transduction of IGFI to the GH gene. (Molecular Endocrinology 5: 890896,1991)
RESULTS Selection of Stable Transfectants GC cells were transfected with an IGF-I receptor cDNA (pRSV-IGFIR; Fig. 1). To assess the integration of the exogenous IGF receptor cDNA plasmid into rat GC cell genomic DNA, a Southern blot analysis of transfectant cells was performed. A construct in the reverse orientation, (pRSV-rlGFIR) serving as a transfection control was also analyzed. The results of DNA hybridizations obtained from nine separate colonies isolated after transfection is shown in (Fig. 2). SamHI- or EcoRIdigested genomic DNA isolated from the control cells (wild type and clones R1 through R6) and transfectants (clones S1, S20 and S21) was hybridized with a 32P-
INTRODUCTION Peripheral insulin-like growth factor-l (IGF-I) mediates many of the growth effects of GH (1, 2). IGF-I also 0888-8809/91 /0890-0896$03.00/0 Molecular Endocrinology Copyright © 1991 by The Endocrine Society
890
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 06:21 For personal use only. No other uses without permission. . All rights reserved.
891
IGF-I Attenuation of GH
EcoRI
labeled human IGF-I receptor cDNA probe. The receptor probe hybridized to all transfectant DNA samples tested, while the probe did not hybridize to DNA extracted from wild-type untransfected GC cells. The different sized bands confirmed the integration of human IGF-I receptor DNA sequences into the genomic DNA of these cells. Ndel
BamH I
IGF-1 Rec cDNA (4.2 Kb)
Synthesis of Human IGF-I Receptor by Transfected Cells To confirm the presence of human IGF-I receptor protein synthesized by the forward transfectants (S20 and S21), lysates of [35S]methionine-labeled transfected and wild-type cells were immunoprecipitated by «IR3 antibody, which specifically recognizes human IGF-I receptor and not the rat IGF-I receptor. A 200-kDa band was observed in the transfectants, representing the unprocessed IGF-I receptor precursor, and smaller degradation products were also observed. alR3 cross-reacted minimally with wild-type GC cell proteins (Fig. 3).
Xhol
31
IGF-I Binding to Transfected Cells
Fig. 1. Schematic Diagram of the Plasmid Used for Transfection A 4.2-kb Hinc\\Sst\ fragment of the human IGF-I receptor cDNA was cloned into the polylinker site of mammalian expression vector Rexp.
Binding of [125I]IGF-I was tested in wild-type GC cells, reverse transfected control cells, R2, and forward transfectants, S20 and S21. Figure 4A shows that increasing
Mrx10" 3 — 200 -•-97 S1 R1 R2 R3 R4 R5 R6
Kb
S20S21
Kb
•5.0 •5.0
•3.0
•3.0
A
B
Fig. 2. Southern Blot Analysis of Genomic DNA from Stable Transfectants Containing pRSV-IGFIR (S1, S20, and S21) and pRSV-rlGFIR(R1-R6) Wild-type cells were untransfected. DNA isolation and blot hybridization were performed as described in Materials and Methods. Digestion of DNA with BamH\ (A) and EcoRI (B) was followed by hybridization to the separated DNA with the human IGF-I receptor cDNA. Size markers (kilobases) are indicated on the right.
Fig. 3. Immunoprecipitation of [35S]Methionine-Labeled IGF-I Receptor Protein Metabolically labeled GC and transfectant cells were immunoprecipitated with a monoclonal antibody to the human IGFI receptor, alR3. The immunoprecipitates were analyzed on denaturing SDS-polyacrylamide gels, and an autoradiograph of the gel is depicted. 14C-Labeled mol wt (Mr) markers are indicated.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 06:21 For personal use only. No other uses without permission. . All rights reserved.
MOL ENDO-1991 892
Vol 5 No. 7
D O
wild type R2
derived from the intercept of the abscissa was increased 17-fold in S20 (384,000 sites/cell) and 10-fold in S21 (234,000 sites/cell), respectively, compared to that in wild-type cells. These observations further confirm that S20 and S21 overexpress an IGF-I receptor. Effects of IGF-I on GH Secretion
(3
2-
10
B
1 00
(nM)
Fig. 4. Binding of [125I]IGF-I to Wild-Type GC Cells, R2 Control Cells (A), and S20 and S21 Transfected Cells (B) Cells were incubated with increasing doses of IGF-I, and binding of [125I]IGF-I was performed in suspension at 15 C for I80 min. Each point represents mean of three samples representative of two separate experiments.
amounts of unlabeled IGF-I displaced [125I]IGF-I binding from wild-type GC and R2 control cells. The specific binding of IGF-I was similar in these two control cell types, 7.6% and 6.9%, respectively. Fifty percent displacement of maximal binding was achieved with 0.50.6 nM IGF-I. This similarity suggests that the transfection procedure itself did not alter the binding activity of IGF-I. Figure 4B shows that increasing amounts of unlabeled IGF-I displaced [125I]IGF-I binding from forward transfectants S20 and S21. In contrast to wildtype GC and R2 control cells, however, the specific binding in S20 and S21 cells was markedly increased to 55% and 47%, respectively. Fifty percent displacement of the maximal binding was achieved with concentrations of IGF-I necessary for binding to wild-type GC and R2 control cells, ranging from 0.5-1 nM. When these binding displacement data were subjected to Scatchard analysis, a linear plot was obtained for all cell types, consistent with the presence of a single class of high affinity receptors in these pituitary cells. The distribution constant (Kd) for [125I]IGF-I binding derived from the slope of the plot was similar in all four cell types (Fig. 5). The number of [125I]IGF-I receptors
To test the time course of IGF-I on GH secretion, wildtype, R2, and S20 cells growing in serum-free defined medium were treated with 6.5 nM IGF-I for up to 42 h. Basal GH secretion by untreated S20 cells was not lower than that in wild-type cells. Wild-type and R2 control cells showed similar suppression of GH to 60% of that in control untreated cells by the end of the incubation period. In contrast, GH secretion by the forward transfectant S20 overexpressing the human IGF-I receptor was suppressed to 20% of control secretion by the end of the incubation period. IGF-I suppressed GH secretion at earlier time points in S20 compared to wild-type and R2 control cells, with maximal GH suppression evident after only 15 h of incubation (Fig. 6). The dose response of the effects of IGF-I on GH secretion was, therefore, tested after 15 h of treatment (Fig. 7). IGF-I suppressed GH secretion in S20 cells in a dose-dependent manner, but GH secretion was not appreciably altered in wild-type and R2 control cells at this time. Fifty percent of the maximal GH suppression was achieved with 0.5 nM IGF-I in S20 cells, a concentration similar to the Ko for IGF-I binding in these cells.
DISCUSSION
The human IGF-I receptor was stably transfected into GC rat pituitary tumor cells. The fidelity of the expression of human IGF-I receptors was confirmed by demonstrating integration of IGF receptor sequences into genomic DNA and by immunoprecipitation of IGF-I receptor proteins. Those transfectants overexpressing human IGF-I receptor also exhibited enhanced responsiveness to IGF-I. IGF-I is an extracellular growth factor regulating cellular proliferation (17). Alterations in growth factor receptor activity can result in uncontrolled growth and neoplastic transformation. For example, overexpressed epidermal growth factor receptors (18) and IGF-I receptors (19) cause ligand-dependent cellular transformation in NIH 3T3 cells. Modification of epidermal growth factor receptor structure can also lead to mitogenic activation (20, 21). In contrast to these studies, overexpression of IGF-I receptors did not promote cellular growth of GC pituitary cells, even at relatively high concentrations of IGF-I (data not shown). Possibly, further overexpression of IGF-I receptors or other unknown cell-specific factors is required to stimulate proliferation of these cells. However, these cells are already transformed and terminally differentiated; therefore, a proliferative effect
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 06:21 For personal use only. No other uses without permission. . All rights reserved.
893
IGF-I Attenuation of GH
150-
100-
Kd
