0013-7227/90/1275-2217$02.00/0 Endocrinology Copyright © 1990 by The Endocrine Society
Vol. 127, No. 5 Printed in U.S.A.
Developmental and Tissue-Specific Regulation of Proglucagon Gene Expression* Y. C. LEEf, P. L. BRUBAKERt AND D. J. DRUCKER§ Departments of Medicine (D.J.D., Y.C.L.), Clinical Biochemistry (D.J.D.), Genetics (D.J.D.), and Physiology (P.L.B.), Toronto General Hospital, University of Toronto, Toronto M5G 2C4, Ontario, Canada
ABSTRACT. The pattern of glucagon gene expression and the posttranslational processing of proglucagon was studied in the fetal and neonatal rat. Pancreatic immunoreactive glucagon (IRG) and glucagon-like immunoreactivity (GLI) were low in both fetal pancreas and intestine, respectively. Immediately after birth, pancreatic IRG rose markedly and reached a peak concentration at postnatal day 7, followed by a gradual return to its adult level. Intestinal GLI was low until postnatal day 7 and rose steadily thereafter to adult levels. The levels of GLI in the hypothalamus were much lower than in intestine, yet the developmental accumulation of hypothalamic GLI resembled the pattern observed in intestine. In contrast, the levels of GLI and IRG in the brain stem were higher in the fetus and neonate, and
decreased to adult levels. Proglucagon mRNA transcripts, uniform in size, were detected in RNA isolated from fetal or adult brainstem, pancreas, and intestine. However, fetal proglucagon mRNA transcripts were larger than adult proglucagon mRNA transcripts in pancreas and intestine, but not brainstem. The results of RNAse mapping studies, including analysis of both the 5'-and 3'-ends of the mRNA transcripts, demonstrated that the larger fetal mRNA transcripts could be accounted for by an increase in the length of the polyadenylate tail in the fetal tissues. These observations demonstrate that the developing rat exhibits tissue-specific differences in both proglucagon gene expression and the pattern of posttranslational processing of the prohormone. (Endocrinology 127: 2217-2222, 1990)
T
HE GENE encoding proglucagon is expressed in the A cells of the endocrine pancreas, the L cells of the intestine, and in brainstem and hypothalamic neurons (1-4). Tissue-specific posttranslational processing of proglucagon gives rise to a different pattern of glucagonrelated peptides in pancreas, intestine, and brain. Whereas 29 amino acid glucagon predominates in the pancreas, glicentin and the glucagon-like peptides are the major forms liberated in the intestine. In contrast, the processing of proglucagon in the brain produces glucagon, glicentin, and the glucagon-like peptides. The isolation and analysis of the cDNAs and genes which encode proglucagon has led to a resurgence of interest in the biology of the glucagon-related peptides. A truncated form of glucagon-like peptide 1, GLP-1, has been shown to possess potent insulinotropic properties both in vitro and in vivo (5, 6). However, the actions of additional peptides derived from the proglucagon precursor remain poorly understood. The development of antibodies and nucleic acid probes has facilitated the study Received June 13,1990. Address all correspondence and reprint requests to: D. J. Drucker, M.D., Toronto General Hospital, 200 Elizabeth Street, CCRW3-838, Toronto M5G 2C4,Ontario, Canada. * This work was supported by grants from the MRC, CDA, and NCI Canada. t CDA Fellow. X Diabetes Canada Scholar. § Career scientist of the Ontario Ministry of Health.
of the sites of proglucagon biosynthesis. The results of these studies have demonstrated that the glucagon gene is expressed in both fetal and adult tissues. Immunocytochemical analyses have shown that peptides with glucagon-like immunoreactivity can be detected in the developing fetal pancreas, intestine, and brain (3, 7-11). The results of these experiments have led to the suggestion that the glucagon gene may be the first peptide hormone gene expressed in the endocrine pancreas, and a role for glucagon in the development of pancreatic endocrine cells has been proposed (11). Additional evidence for the expression of the glucagon gene in fetal tissues derives from studies of transgenic animals. The brains of mice transgenic for a rat glucagon gene-SV 40 large t antigen transgene contained fusion gene mRNA transcripts at embryonic day 12, and the relative number of neurons which expressed the transgene increased with postnatal development (12). These experiments demonstrated that glucagon gene regulatory sequences modulate the activation of glucagon gene transcription in a tissue- and developmentally-specific manner. The observation that the glucagon gene is expressed in fetal tissues raised the possibility that one or more of the proglucagon-derived peptides may play a role in fetal or neonatal development. Surprisingly little information is available, however, about proglucagon gene expression and the specific peptides liberated from proglucagon in fetal tissues. As a first step toward understanding the
2217
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 21:39 For personal use only. No other uses without permission. . All rights reserved.
2218
REGULATION OF PROGLUCAGON GENE EXPRESSION
Endo • 1990 Vol 127 • No 5
importance of the proglucagon-derived peptides in the maturing fetus and neonate, we have studied proglucagon gene expression and the pattern of posttranslational processing in fetal, neonatal, and adult rat tissues. P-GLI P-IRG I-GLI I-IRG
Materials and Methods Animals Multiple litters of fetal (20 and 21-day timed gestation) or neonatal (1, 2, 7, 11, and 21-day) Wistar rats (Charles River Canada) were pooled to make n = 1, with up to 13 rats pooled from each litter. Reagents
F20/21
All chemical reagents were from Sigma (St. Louis, MO), Fisher (Pittsburgh, PA), or BDH (Toronto, Ontario, Canada). [32PJATP and [32P]UTP were from Amersham International. Restriction enzymes, DNA modifying enzymes, and DNA and RNA polymerases were from Pharmacia Canada.
d1/2
d7
d11/21
Adult
Age
c "55
Tissue preparation
s
Peptide extraction was performed as previously described (10,13,14). Peptides were extracted from whole intestine (fetal and day-1 to -11 neonatal rats) or 10-cm terminal ileum (day21 and adult rats), whole hypothalamus, whole medulla oblongata, or a section of pancreatic tail, by homogenization at 4 C in 1 N HCL containing 5% (vol/vol) formic acid, 1% (vol/ vol) trifluoroacetic acid (TFA), and 1% (vol/wt) NaCl, followed by reversed-phase adsorption to C18 silica (C18 SepPak, Waters Associates, Milford MA).
O) ,3 ^) O
12H-GLI H-IRG MO-GLI MO-IRG
Q.
4-
0 b
F20/21
d1/2
d7
d11/21
Adult
Age
Assays 601
Tissue extracts were assayed for immunoreactive glucagon (IRG) and glucagon-like immunoreactive peptides (GLI) using antisera O4A (Dr. R. H. Unger, Dallas, TX) and K4023 (Novo Alle, Bagsvaerd, Denmark), respectively, as described previously (13, 14). The antiserum O4A recognizes the free Cterminal end of glucagon and recognizes primarily glucagon, whereas antiserum K4023 recognizes the midsequence of glucagon and cross-reacts equally well with glucagon and N- or Cterminally extended forms of glucagon (i.e. the intestinal peptides glicentin and oxyntomodulin). Protein assays were carried out using the method of Lowry et al. (15).
45-
— P 30-
H
15-
RNA extraction F20/21
Tissues were harvested from killed rats and homogenized immediately in 4 M guanidinium isothiocyanate. Total RNA was isolated as previously described (16) and stored at -80 C until further analysis was performed. Northern blot analysis, cDNA probe labeling, and RNAse protection studies were carried out as previously described (3,16). The RNAse H digestion of poly(A) tract was carried out by a modification of the method of Robinson et al. (17). Briefly, 50-^g total cellular RNA was mixed with 10-^g oligo (dT)12-18 (Collaborative Research, Bedford, MA) and dissolved in 70-/ul (100 mM KCL/0.1 mM EDTA). The mixture was heated to 65 C for 2 min and annealed at room temperature for 30 min. Ribonuclease H, 5U, (Boehringer Mannheim, FRG) and 25-jtl incubation buffer (50 mM MgCl2, 0.4 M KC1, 0.25 M Tris-HCl, pH 7.5 and 5 mM DTT)
d1/2
d7
di 1/21
Adult
Age
FIG. 1. Levels of glucagon-like immunoreactive peptides (GLI), and immunoreactive glucagon (IRG) in pancreatic (P) and intestinal (I) extracts (Fig. la), and in hypothalamic (H), and medulla oblongata (MO) extracts (Fig. lb). The ratio of GLI to IRG during development in the rat is shown in lc. Pancreas (P), intestine (I), hypothalamus (H), and medulla oblongata (MO) were examined in fetal rats at 20-21 days gestation (F20/21; n = 4-17); in newborn rats at day 1/2 (n = 4), day 7 (n = 4-7), day 11/21 (n = 3-9); and in adult rats (n = 4-16). The data for H was taken from (10); H was studied on day 2 and day 11, not day 1 and day 21. The data for adult rat P, I, and MO were from (31). * P < 0.05, o P < 0.01, # P < 0.001.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 21:39 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF PROGLUCAGON GENE EXPRESSION
2219
were added and incubated at 37 C for 30 min. RNA was extracted with phenol/chloroform/isoamyl alcohol, ethanol precipitated, and analyzed by Northern blotting.
Results Pancreatic levels of IRG peptides were low in the fetus, but rose after birth to reach a peak at day 7-21, and then declined to adult levels (Fig. 1A). GLI levels were similar to those of the IRG peptides at all points during development, consistent with a predominence of glucagon in the pancreas. The ratio of GLI to IRG thus remained constant at 1.7 ± 0.2 in all pancreatic preparations studied (Fig. 1C). In contrast to the results obtained for the pancreatic proglucagon-derived peptides, intestinal GLI levels remained low from day 20-21 of gestation until day 7 of neonatal life, and rose dramatically thereafter, reaching highest levels in the adult (Fig. 1A). Heal concentrations of GLI peptides were identical to those of the pancreas in the adult rat (P > 0.2). Consistent with the absence of pancreatic glucagon in the intestine, the levels of the IRG peptides were very low in all age groups studied. Thus, the ratio of GLI to IRG peptides in the rat intestine remained constant at 47.6 ± 1.8 throughout development (Fig. 1C). Proglucagon-derived peptides were also detected in the developing brain, with the highest levels in the fetal and neonatal medulla oblongata (Fig. IB). Brainstem GLI and IRG peptide levels declined during development, whereas GLI levels in the hypothalamus rose to reach a peak in the adult (Fig. IB). The ratio of GLI to IRG in the hypothalamus increased during development from 2.6 ± 0.5 in the fetus to 46.2 ± 11.4 in the adult. A rise in the GLI/IRG ratio was also noted in the brainstem, (from 0.7 ± 0.3 to 10.3 ± 1.4 in the adult). Northern blot analysis of RNA prepared from fetal pancreas and intestine detected a single proglucagon mRNA transcript which was just slightly larger than the proglucagon mRNA transcript detected in the corresponding adult tissues (Fig. 2, a and b). Analysis of the identical samples after annealing with oligo d(T) and treatment with RNAse H revealed that the fetal and adult proglucagon mRNA transcripts became indistinguishable in size (Fig. 2, a and b), suggesting that the fetal transcript was larger due to an increase in the length of the poly A tail. A decrease (to adult levels) in the size of the fetal proglucagon mRNA transcripts was noted as early as postnatal day 1 (data not shown). To determine if the larger fetal mRNA transcripts was a specific finding confined to the glucagon gene, the Northern blot shown in Fig. lb was rehybridized with a rat somatostatin cDNA probe, shown in Fig. lc. The results of these studies demonstrated that the fetal intestinal somatostatin mRNA transcript was also larger than the adult transcript, and this difference in size also appeared to be due to a longer poly A tract. In contrast, the fetal and
I
£
2r
O-
M-
Vol. 127, No. 5 Printed in U.S.A.
Developmental and Tissue-Specific Regulation of Proglucagon Gene Expression* Y. C. LEEf, P. L. BRUBAKERt AND D. J. DRUCKER§ Departments of Medicine (D.J.D., Y.C.L.), Clinical Biochemistry (D.J.D.), Genetics (D.J.D.), and Physiology (P.L.B.), Toronto General Hospital, University of Toronto, Toronto M5G 2C4, Ontario, Canada
ABSTRACT. The pattern of glucagon gene expression and the posttranslational processing of proglucagon was studied in the fetal and neonatal rat. Pancreatic immunoreactive glucagon (IRG) and glucagon-like immunoreactivity (GLI) were low in both fetal pancreas and intestine, respectively. Immediately after birth, pancreatic IRG rose markedly and reached a peak concentration at postnatal day 7, followed by a gradual return to its adult level. Intestinal GLI was low until postnatal day 7 and rose steadily thereafter to adult levels. The levels of GLI in the hypothalamus were much lower than in intestine, yet the developmental accumulation of hypothalamic GLI resembled the pattern observed in intestine. In contrast, the levels of GLI and IRG in the brain stem were higher in the fetus and neonate, and
decreased to adult levels. Proglucagon mRNA transcripts, uniform in size, were detected in RNA isolated from fetal or adult brainstem, pancreas, and intestine. However, fetal proglucagon mRNA transcripts were larger than adult proglucagon mRNA transcripts in pancreas and intestine, but not brainstem. The results of RNAse mapping studies, including analysis of both the 5'-and 3'-ends of the mRNA transcripts, demonstrated that the larger fetal mRNA transcripts could be accounted for by an increase in the length of the polyadenylate tail in the fetal tissues. These observations demonstrate that the developing rat exhibits tissue-specific differences in both proglucagon gene expression and the pattern of posttranslational processing of the prohormone. (Endocrinology 127: 2217-2222, 1990)
T
HE GENE encoding proglucagon is expressed in the A cells of the endocrine pancreas, the L cells of the intestine, and in brainstem and hypothalamic neurons (1-4). Tissue-specific posttranslational processing of proglucagon gives rise to a different pattern of glucagonrelated peptides in pancreas, intestine, and brain. Whereas 29 amino acid glucagon predominates in the pancreas, glicentin and the glucagon-like peptides are the major forms liberated in the intestine. In contrast, the processing of proglucagon in the brain produces glucagon, glicentin, and the glucagon-like peptides. The isolation and analysis of the cDNAs and genes which encode proglucagon has led to a resurgence of interest in the biology of the glucagon-related peptides. A truncated form of glucagon-like peptide 1, GLP-1, has been shown to possess potent insulinotropic properties both in vitro and in vivo (5, 6). However, the actions of additional peptides derived from the proglucagon precursor remain poorly understood. The development of antibodies and nucleic acid probes has facilitated the study Received June 13,1990. Address all correspondence and reprint requests to: D. J. Drucker, M.D., Toronto General Hospital, 200 Elizabeth Street, CCRW3-838, Toronto M5G 2C4,Ontario, Canada. * This work was supported by grants from the MRC, CDA, and NCI Canada. t CDA Fellow. X Diabetes Canada Scholar. § Career scientist of the Ontario Ministry of Health.
of the sites of proglucagon biosynthesis. The results of these studies have demonstrated that the glucagon gene is expressed in both fetal and adult tissues. Immunocytochemical analyses have shown that peptides with glucagon-like immunoreactivity can be detected in the developing fetal pancreas, intestine, and brain (3, 7-11). The results of these experiments have led to the suggestion that the glucagon gene may be the first peptide hormone gene expressed in the endocrine pancreas, and a role for glucagon in the development of pancreatic endocrine cells has been proposed (11). Additional evidence for the expression of the glucagon gene in fetal tissues derives from studies of transgenic animals. The brains of mice transgenic for a rat glucagon gene-SV 40 large t antigen transgene contained fusion gene mRNA transcripts at embryonic day 12, and the relative number of neurons which expressed the transgene increased with postnatal development (12). These experiments demonstrated that glucagon gene regulatory sequences modulate the activation of glucagon gene transcription in a tissue- and developmentally-specific manner. The observation that the glucagon gene is expressed in fetal tissues raised the possibility that one or more of the proglucagon-derived peptides may play a role in fetal or neonatal development. Surprisingly little information is available, however, about proglucagon gene expression and the specific peptides liberated from proglucagon in fetal tissues. As a first step toward understanding the
2217
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 21:39 For personal use only. No other uses without permission. . All rights reserved.
2218
REGULATION OF PROGLUCAGON GENE EXPRESSION
Endo • 1990 Vol 127 • No 5
importance of the proglucagon-derived peptides in the maturing fetus and neonate, we have studied proglucagon gene expression and the pattern of posttranslational processing in fetal, neonatal, and adult rat tissues. P-GLI P-IRG I-GLI I-IRG
Materials and Methods Animals Multiple litters of fetal (20 and 21-day timed gestation) or neonatal (1, 2, 7, 11, and 21-day) Wistar rats (Charles River Canada) were pooled to make n = 1, with up to 13 rats pooled from each litter. Reagents
F20/21
All chemical reagents were from Sigma (St. Louis, MO), Fisher (Pittsburgh, PA), or BDH (Toronto, Ontario, Canada). [32PJATP and [32P]UTP were from Amersham International. Restriction enzymes, DNA modifying enzymes, and DNA and RNA polymerases were from Pharmacia Canada.
d1/2
d7
d11/21
Adult
Age
c "55
Tissue preparation
s
Peptide extraction was performed as previously described (10,13,14). Peptides were extracted from whole intestine (fetal and day-1 to -11 neonatal rats) or 10-cm terminal ileum (day21 and adult rats), whole hypothalamus, whole medulla oblongata, or a section of pancreatic tail, by homogenization at 4 C in 1 N HCL containing 5% (vol/vol) formic acid, 1% (vol/ vol) trifluoroacetic acid (TFA), and 1% (vol/wt) NaCl, followed by reversed-phase adsorption to C18 silica (C18 SepPak, Waters Associates, Milford MA).
O) ,3 ^) O
12H-GLI H-IRG MO-GLI MO-IRG
Q.
4-
0 b
F20/21
d1/2
d7
d11/21
Adult
Age
Assays 601
Tissue extracts were assayed for immunoreactive glucagon (IRG) and glucagon-like immunoreactive peptides (GLI) using antisera O4A (Dr. R. H. Unger, Dallas, TX) and K4023 (Novo Alle, Bagsvaerd, Denmark), respectively, as described previously (13, 14). The antiserum O4A recognizes the free Cterminal end of glucagon and recognizes primarily glucagon, whereas antiserum K4023 recognizes the midsequence of glucagon and cross-reacts equally well with glucagon and N- or Cterminally extended forms of glucagon (i.e. the intestinal peptides glicentin and oxyntomodulin). Protein assays were carried out using the method of Lowry et al. (15).
45-
— P 30-
H
15-
RNA extraction F20/21
Tissues were harvested from killed rats and homogenized immediately in 4 M guanidinium isothiocyanate. Total RNA was isolated as previously described (16) and stored at -80 C until further analysis was performed. Northern blot analysis, cDNA probe labeling, and RNAse protection studies were carried out as previously described (3,16). The RNAse H digestion of poly(A) tract was carried out by a modification of the method of Robinson et al. (17). Briefly, 50-^g total cellular RNA was mixed with 10-^g oligo (dT)12-18 (Collaborative Research, Bedford, MA) and dissolved in 70-/ul (100 mM KCL/0.1 mM EDTA). The mixture was heated to 65 C for 2 min and annealed at room temperature for 30 min. Ribonuclease H, 5U, (Boehringer Mannheim, FRG) and 25-jtl incubation buffer (50 mM MgCl2, 0.4 M KC1, 0.25 M Tris-HCl, pH 7.5 and 5 mM DTT)
d1/2
d7
di 1/21
Adult
Age
FIG. 1. Levels of glucagon-like immunoreactive peptides (GLI), and immunoreactive glucagon (IRG) in pancreatic (P) and intestinal (I) extracts (Fig. la), and in hypothalamic (H), and medulla oblongata (MO) extracts (Fig. lb). The ratio of GLI to IRG during development in the rat is shown in lc. Pancreas (P), intestine (I), hypothalamus (H), and medulla oblongata (MO) were examined in fetal rats at 20-21 days gestation (F20/21; n = 4-17); in newborn rats at day 1/2 (n = 4), day 7 (n = 4-7), day 11/21 (n = 3-9); and in adult rats (n = 4-16). The data for H was taken from (10); H was studied on day 2 and day 11, not day 1 and day 21. The data for adult rat P, I, and MO were from (31). * P < 0.05, o P < 0.01, # P < 0.001.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 21:39 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF PROGLUCAGON GENE EXPRESSION
2219
were added and incubated at 37 C for 30 min. RNA was extracted with phenol/chloroform/isoamyl alcohol, ethanol precipitated, and analyzed by Northern blotting.
Results Pancreatic levels of IRG peptides were low in the fetus, but rose after birth to reach a peak at day 7-21, and then declined to adult levels (Fig. 1A). GLI levels were similar to those of the IRG peptides at all points during development, consistent with a predominence of glucagon in the pancreas. The ratio of GLI to IRG thus remained constant at 1.7 ± 0.2 in all pancreatic preparations studied (Fig. 1C). In contrast to the results obtained for the pancreatic proglucagon-derived peptides, intestinal GLI levels remained low from day 20-21 of gestation until day 7 of neonatal life, and rose dramatically thereafter, reaching highest levels in the adult (Fig. 1A). Heal concentrations of GLI peptides were identical to those of the pancreas in the adult rat (P > 0.2). Consistent with the absence of pancreatic glucagon in the intestine, the levels of the IRG peptides were very low in all age groups studied. Thus, the ratio of GLI to IRG peptides in the rat intestine remained constant at 47.6 ± 1.8 throughout development (Fig. 1C). Proglucagon-derived peptides were also detected in the developing brain, with the highest levels in the fetal and neonatal medulla oblongata (Fig. IB). Brainstem GLI and IRG peptide levels declined during development, whereas GLI levels in the hypothalamus rose to reach a peak in the adult (Fig. IB). The ratio of GLI to IRG in the hypothalamus increased during development from 2.6 ± 0.5 in the fetus to 46.2 ± 11.4 in the adult. A rise in the GLI/IRG ratio was also noted in the brainstem, (from 0.7 ± 0.3 to 10.3 ± 1.4 in the adult). Northern blot analysis of RNA prepared from fetal pancreas and intestine detected a single proglucagon mRNA transcript which was just slightly larger than the proglucagon mRNA transcript detected in the corresponding adult tissues (Fig. 2, a and b). Analysis of the identical samples after annealing with oligo d(T) and treatment with RNAse H revealed that the fetal and adult proglucagon mRNA transcripts became indistinguishable in size (Fig. 2, a and b), suggesting that the fetal transcript was larger due to an increase in the length of the poly A tail. A decrease (to adult levels) in the size of the fetal proglucagon mRNA transcripts was noted as early as postnatal day 1 (data not shown). To determine if the larger fetal mRNA transcripts was a specific finding confined to the glucagon gene, the Northern blot shown in Fig. lb was rehybridized with a rat somatostatin cDNA probe, shown in Fig. lc. The results of these studies demonstrated that the fetal intestinal somatostatin mRNA transcript was also larger than the adult transcript, and this difference in size also appeared to be due to a longer poly A tract. In contrast, the fetal and
I
£
2r
O-
M-
