Insulin regulates insulin-like in rat hepatocytes
growth factor I mRNA
M. BijNI-SCHNETZLER, C. SCHMID, P. J. MEIER, AND E. R. FROESCH Metabolic Unit and Division of Clinical Pharmacology, Department of Medicine, University Hospital, 8091 Ziirich, Switzerland
B~NI-SCHNETZLER, M.,C. SCHMID, P.J. MEIER,AND E.R. FROESCH. Insulin regulates insulin-like growth factor I mRNA in rat hepatocytes. Am. J. Physiol. 260 (Endocrinol. Metab. 23): E846-E851, 1991.-To evaluate the regulatory role of growth hormone (GH) and insulin on insulin-like growth factor I (IGFI) mRNA levels, we employed primary rat hepatocytes. Cells were incubated for 16 h with 10 nM insulin, 10 nM GH, or a combination thereof, and IGF-I mRNA levels were analyzed by Northern blotting. Insulin results in 2.5-fold and GH in 3.8fold higher IGF-I mRNA levels than hormone-free controls, and a combination of insulin and GH had an additive effect (6.7-fold). The effect of 10 nM insulin was constant at variable GH concentrations. Therefore, GH and insulin affect IGF-I mRNA levels independently of each other. The half-maximal effective dose of insulin was 4.7 x 10sl’ M, and, in kinetic experiments, insulin was effective within 2 h. These findings demonstrate that insulin modulates hepatic IGF-I production by a direct regulation of the transcript levels of IGF-I. growth growth
hormone; regulation
insulin-like
growth
factor binding
proteins;
INSULIN-LIKE GROWTH FACTOR I (IGF-I) is a growthpromoting hormone with structural similarities to proinsulin (25). Growth indices are positively correlated with serum IGF-I concentrations (30). These appear to be regulated at the IGF-I mRNA levels in the liver (4), which is the predominant site of synthesis of circulating IGF-I (31). The primary regulators of IGF-I are growth hormone (GH) and the nutritional status (13, 23, 26). In addition, insulin therapy restores IGF-I mRNA levels toward normal in insulin-deficient growth-arrested rats (4). It has been assumed that insulin affects IGF-I levels in these diabetic rats indirectly via restoration of the impaired GH secretion (7, 27). However, there are three different observations in vivo that suggest that insulin might also have a direct effect on serum IGF-I levels. 1) Insulin treatment in combination with a high-carbohydrate diet significantly augmented skeletal growth in GH-deficient rats (28). 2) Poorly controlled diabetic children with normal or increased GH levels exhibit reduced growth that is associated with low levels of IGFI, whereas normal growth has been reported in hyperinsulinemic children, despite GH deficiency (6, 18, 34). 3) The delivery of insulin via the portal vein is more effective than systemic insulin in restoring growth and IGFI levels in diabetic and growth-arrested rats, indicating that insulin may, indeed, directly act on the liver to augment serum IGF-I concentrations (16). E846
0193-1849/91
$1.50 Copyright
In blood, IGF-I does not circulate as free hormone. It is tightly associated with high-affinity IGF binding proteins (IGFBPs). It was reported that these IGFBPs influence the half-life of IGF-I in vivo (35) and that, in different cell culture systems, they either inhibit or enhance IGF-I effects (12, 21). Insulin may affect IGF-I serum levels not only by stimulating hepatic IGF-I synthesis but also by regulating the levels of the various IGFBPs. Thus we have recently shown that insulin suppresses an IGFBP mRNA, the levels of which are high in the liver of insulin-deficient growth-arrested rats (4) and in primary hepatocytes of adult rats incubated without insulin (5). This IGFBP, recently termed IGFBP-2 (l), is abundant in fetal serum. It is virtually absent from serum of healthy adult animals where IGF-I is predominantly bound to IGFBP-3 (1). In the intact animal, there is a positive correlation between the GH and nutritional status (including insulin) on one hand and the levels of IGF-I and IGFBP-3 on the other hand. Because insulin influences GH secretion in vivo (7), it is rather difficult to discern in intact animals what the relative contributions of GH and of insulin with regard to IGF-I synthesis might be. We therefore resorted to an in vitro system consisting of primary hepatocytes from adult rats and evaluated the effect of insulin and of GH on IGF-I mRNA levels separately from each other. Because GH is a well-known stimulator of hepatic IGF-I production (23, 26), we focused this study on the characterization of the role of insulin. MATERIALS
AND
METHODS
Preparation of hepatocytes. Hepatocytes were isolated from livers of 180- to 220-g male Sprague-Dawley rats by Ca2+-free collagenase perfusion (3). The cells were then plated at a density of 7 x lo4 cells/cm2 on type I collagen-coated (Serva, Heidelberg, FRG) 60-mm dishes. The viability was assessedby trypan blue exclusion and was 285%. Th e culture media was phenol red-free Williams E (Animed, Basel, Switzerland) supplemented with penicillin (100 U/ml), streptomycin (0.1 mg/ml), insulin (10D7 M), dexamethasone (10B7 M), and 10% fetal calf serum. After a 3-h attachment period, the media were replaced by the above-described media devoid of fetal calf serum. Experiments were started 16-24 h after plating the cells. Addition of hormones to hepatocyte cultures. Before the start of experiments, hepatocytes were washed for a period of 2 h at 37°C by three changes of serum- and
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
INSULIN
STIMULATES
IGF-I
MRNA
hormone-free Williams E medium. Experiments were initiated by replacing the medium with 2.9 ml of fresh Williams E medium and by adding 0.1 ml of human serum albumin (HSA) without or with hormones to yield final concentrations to 10 nM insulin (gift of Novo Biolabs), 10 nM GH (Nordisk), or a combination thereof and 0.1 g/l of HSA. HSA was obtained from the Swiss Red Cross and was charcoal-treated before use for cell cultures (8). Hepatocytes were then incubated for 16 h or the times indicated in the respective experiments. RNA isolation and Northern blotting. Incubations with hormones were stopped by washing the cells three times with ice-cold phosphate-buffered saline. Subsequently, cells were removed from culture dishes using a rubber policeman and collected by centrifugation at 4°C. The pellets containing 5 x lo6 cells were then lysed by the addition of 2.5 ml of homogenization buffer [6 M guanidinium isothiocyanate (BRL, Gaithersburg, MD), 5 mM sodium citrate, pH 7.0, 0.1 M P-mercaptoethanol, and 0.5% sarkosyl (Sigma, St. Louis, MO)] and by vigorous vortexing. Total RNA was purified by a standard CsCl centrifugation method followed by extraction with a 4:l mixture of chloroform and l-butanol (22). RNA samples were finally dissolved in diethyl pyrocarbonate-treated water, and concentrations were determined spectrophotometrically. Denatured RNA (20 pg/lane) was electrophoresed in 1% formaldehyde containing agarose gels (22). RNA was transferred (22) to nylon membranes (Hybond-N, Amersham, UK) and were crosslinked with ultraviolet radiation (9). There were no indications for RNA degradation, as assessed by agarose gel electrophoresis in the presence of ethidium bromide and subsequent visual inspection. Hybridization and evaluation of blots. The following probes were used for hybridization: rat IGF-I cDNA corresponding to the genomic sequences between nucleotide 2,054 of exon 1 and nucleotide 868 of exon 5 (33) and a P-actin cDNA (10). They were labeled by random primer extension (kit from Boehringer, Mannheim, FRG) using [a-32P]deoxycytidine 5’triphosphate (Amersham, UK) to a specific activity of l-4 X 10’ counts per min (cpm)/pg. Prehybridization and hybridization were carried out in a hybridization incubator (GFL model 7601, Burgwedel, GFR). Prehybridization solutions consist of ~5 standard sodium citrate (SSC), ~5 Denhardt’s solution, 100 mM Na2HP04 (pH 6.5), 50% formamide, 0.5% sodium dodecyl sulfate (SDS), and 100 pg/ml salmon sperm DNA. All stock solutions used to mix prehybridization and hybridization solutions were prepared according to Maniatis et al. (22). For hybridization, 1 rig/ml of labeled probe was added to a solution consisting of ~5 sodium-EDTA-tris(hydroxymethyl)aminomethane, ~5 Denhardt’s solution, 100 mM Na2HP04 (pH 6.5), 50% formamide, 100 pg/ml salmon sperm DNA, and 0.2% SDS. After hybridization for 48 h, membranes were washed three times for 20 min at 50°C with x0.2 SSC and 0.1% SDS and were exposed to Kodak X-Omat AR film in the presence of a Cronex lightening plus enhancer screen. Quantitative estimates of the relative mRNA levels were obtained either by excising the 32Plabeled bands from the membranes and by counting their Cerenkov radiation or by the determination of the optical densities of bands from autoradiograms using a video
LEVELS
IN
E847
HEPATOCYTES
densitometer (Bio-Rad, model 620). Specific hybridization was directly proportional to the amount of RNA applied if there was no more than 1.25 pg RNA/10 mm2 filter area applied as determined by slot blotting. Furthermore, the optical density (OD) of a band on the autoradiogram was proportional to the radioactivity contained in this band up to an OD of 12/10 mm2 filter area. Film exposure times of 6-16 h typically yielded bands with an intensity 80% of the total amount of IGF-I mRNA, whereas the other two smaller species were much less abundant. Within each blot, the ratio of the large 8.2-kb message to the smaller messages was found to be constant and independent of the treatment of cells before RNA isolation. We therefore
28s~
18s~
8 7 6 5 & 3 2 1 GH INS IGF
Jh 16h-
-Oh I mRNA (GH).
1. Insulin-like growth factor I (IGF-I) of hepatocytes with insulin, IGF-I, or growth hormone Primary hepacultures established as described under MATERIALS AND METHODS were treated for 16 h with 10 nM insulin (INS), 10 nM GH. a combination thereof, with 10 nM human recombinant IGF-I (gift of W. R. Rutter, Chiron, Emeryville, CA, and of J. Niiesch, Ciba-Geigy, Basel. Switzerland), or a combination of GH and IGF-I. RNA was then extracted and subjected to Northern blotting using a rat cDNA probe for IGF-I. For comparison, a sample was also taken at start of incubations with hormones (0 h). Top: autoradiograph. Bottom: quantitative estimate of amount of large 8.2-kb message, as determined by scanning densitometry of autoradiogram. Relative levels of 8.2-kb IGF-I mRNA species are expressed as optical densities (OD). FIG.
treated tocyte
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E848
INSULIN
TABLE
on IGF-I
STIMULATES
1. Effect of 10 nM insulin and/or mRNA levels in hepatocytes Expt
Insulin
GH
1
3.53 1.90 2.85 1.81 2.32 2.48k0.71
4.56 4.13 4.33 2.26 3.78 3.81kO.91
2 3 4 5 Means
k SD
IGF-I
MRNA
+ GH
HEPATOCYTES
Dose dependence of the effect of insulin on IGF-I mRNA levels. In all previous experiments, we used 10 nM insu-
7.65 7.57 7.71 3.65 6.84 6.68t1.73
Hepatocyte cultures were treated for 16 h, and RNA was extracted and analyzed by Northern blotting for insulin-like growth factor I (IGF-I) mRNA levels as described in legend to Fig. 1. Relative IGF-I mRNA levels are expressed as ratio of optical densities of hormonetreated to hormone-free cultures. GH, growth hormone.
used the signal of the 8.2-kb message to determine relative levels of IGF-I mRNA.
IN
more potent (381 t 91%). In the presence of insulin and GH, IGF-I mRNA levels were 668 t 173% of the control levels, i.e., the effects of the two hormones were additive, and a synergism was not discernible.
10 nM GH Insulin
LEVELS
the
Effect of insulin, GH, and IGF-I on IGF-I mRNA levels in hepatocytes. To examine the contributions of GH and
of insulin to the regulation of hepatocyte IGF-I mRNA levels, we incubated cells with the respective hormones at a concentration of 10 nM. After 16 h, RNA was isolated and subjected to Northern blotting. Figure 1 shows that the level of IGF-I mRNA is very low in the absence of hormones and‘is markedly elevated by GH. In the presence of insulin alone, IGF-I mRNA levels are also increased, and a combination of insulin and GH resulted in an even more pronounced rise. No significant effect on IGF-I mRNA levels was observed upon treatment with 10 nM IGF-I, for which the lack of IGF-I receptors on liver cells of adult rats may be responsible. RNA isolated at the beginning of the hormonal treatment (time 0) contains higher IGF-I mRNA levels than RNA isolated from cells that were incubated for 16 h in the absence of hormones (see also Fig. 4). Table 1 shows a quantitative estimate of the relative contributions of insulin and GH, each alone and in combination, to the elevation of IGF-I mRNA levels. The results stem from five independent preparations of primary hepatocytes exposed to hormones for 16 h. In the presence of insulin, IGF-I mRNA levels were consistently higher, i.e., 248 t 71% (SD), than in hormone-free controls. GH alone was
lin, i.e., a dose that saturates insulin receptors and one that is expected to produce maximal effects. To find out whether insulin is effective at physiological concentrations, we determined the dose dependency of the insulin stimulation of IGF-I mRNA levels. Figure 2 shows a representative experiment. In three different hepatocyte preparations the insulin concentration eliciting a halfmaximal response averaged 4.7 X 10ml*t 1.6 X lo-‘* M. This dose corresponds to the dissociation constant of insulin binding to the rat liver insulin receptor (24). We conclude that insulin regulates IGF-I mRNA levels well within physiological concentrations in a dose-dependent manner. Contributions of various insulin and GH concentrations to the regulation of IGF-I mRNA levels. The finding that
insulin alone can elevate IGF-I mRNA levels relative to hormone-free controls and that the increase adds up to that obtained with GH indicates that the effect of insulin is independent of GH. To test this prediction, we incubated hepatocytes for 16 h with various concentrations of GH in the presence or absence of 10 nM insulin and subjected the RNA of these cells to Northern blotting. Figure 3A, top, shows the Northern blot of IGF-I mRNA, and Fig. 3A, bottom, shows that of P-actin mRNA. The levels of actin mRNA were constant in all samples, indicating that equal amounts of RNA were loaded per lane and that there is little, if any, RNA degradation. This was a consistent finding in all experiments with hepatocytes where hybridizations with actin cDNA were performed. When the same blot was hybridized with an IGF-I cDNA probe, the increase of IGF-I mRNA levels depended on the GH concentration. A quantitation of this experiment is shown in Fig. 3B, left. The stimulatory effect of GH alone reached a plateau at 10ml*M. When insulin was added together with GH the net effect of insulin was smaller at low GH concentrations than at
I
0
10-l’
,;-I0
10‘-9 insulin
I-8 10
10’ -7
FIG. 2. Dose dependence of regulation of IGF-I mRNA levels by insulin. Dose dependence of insulin (n-axis) is expressed as a function of percent of maximal stimulation (y-axis) of IGF-I mRNA level. Hepatocytes were incubated for 16 h with insulin at indicated concentrations, and RNA was extracted and subsequently subjected to Northern blotting. OD of 8.2-kb bands were determined by scanning densitometry. To calculate dose eliciting half-maximal response, value obtained in absence of insulin was defined as 0% and value obtained at 3 x lOA M insulin as 100%. OD ratio of value obtained with insulin at 3 x lo-" M and that of control was 1.9:1. Values obtained at different insulin concentrations are expressed as percent of maximal effect of insulin.
(M)
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
INSULIN
GH (M)
0
16”
IO -10 ,69
,dS
,o-7
I
STIMULATES
o
,611
--INSULIN
,$O
,69
IGF-I
,&I
MRNA
LEVELS
IN
HEPATOCYTES
,67
FIG. 3. Effect of 10 nM insulin at variable GH concentrations and effect of 10 nM GH at variable insulin concentrations. Hepatocytes were incubated with 10 nM insulin at various dilutions of GH. After 16 h, relative IGF-I mRNA levels were determined by Northern blotting. A, top: IGF-I mRNA. A, bottom: actin message. Signals for actin did not significantly differ from each other. B, left: quantitation of IGF-I mRNA content obtained by excising 8.2-kb bands from blot and by counting their Cerenkov radiation. Solid lines: relative IGF-I mRNA in absence of insulin (-ins). Dashed line: levels in presence of insulin (+ins). B, right: IGF-I mRNA levels at variable insulin concentrations in absence (-GH) or presence of 10 nM GH (+GH). Results of both halves of B are derived from separate experiments.
+INSULIN
ACTIN*i
15oa
300
,A
‘\
/’ : /
‘a--
-0
E849
*ins
:250 Y 2 a200
E
z
150
;-GH
100
0
1 o-1’
10-10
GH
10-g
10-S
10-7
A
&I
(M)
high GH concentrations. However, at every GH dose the presence of insulin constantly resulted in 156 + 13% higher IGF-I mRNA levels. The results of the opposite experiment in which the effects of 10 nM GH at various insulin concentrations were examined are shown in Fig. 3B, right. Note that both halves of Fig. 3B contain unprocessed data (cpm of bound radioactive cDNA probe) and therefore the numbers are not directly comparable. However, the same conclusion, namely that insulin and GH independently regulate IGF-I mRNA, is obtained. Because the effect of insulin on IGF-I mRNA levels is smaller than that of GH, the dose-response curve with insulin alone is less steep than that with GH alone.
,‘o.lO
,;-9
,;-a
,;.7
insulin(M)
Consequently, there is also slightly more variation of the stimulation by GH, which averaged 340 + 61%. We conclude that GH and insulin affect IGF-I mRNA levels independently of each other and that both hormones together have strictly additive effects. Time course of insulin-induced elevation of IGF-I mRNA levels. In all previous experiments, hepatocytes
were incubated with or without insulin for a period of 16 h. In the absence of insulin, IGF-I mRNA levels were lower after 16 h than at the start of the incubation (see also Fig. 1). A time-dependent decline was also observed when IGF-I was analyzed at the protein level (20). The extent of the decline at the mRNA level was variable
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E850
INSULIN
STIMULATES
IGF-I
MRNA
-3 8 2 a2 E
HOURS
FIG. 4. Time course of IGF-I mRNA levels in presence or absence of insulin. One-day hepatocyte cultures were washed with 3 changes of hormone-free media over a period of 2 h. Subsequently, fresh media containing 10 nM insulin (solid line) or no insulin (dashed line) was at various times after start of added (time 0) ,a nd RNA was extracted by Northern blotting, and relative kinetics. IGF-I m RNA was analyzed levels of 8.2-kb message were determined by scanning densitometry and are expressed as OD as described in Fig. 1.
Growth
Hormone
0
0
B IGF I FIG. 5 . Schematic representation mRNA and IGF binding of IGF-I in rat liver.
Insulin
0
IGF BP2 of proposed hormonal regulation protein 2 (IGFBP-2) mRNA levels
among different hepatocyte preparations and most likely was due to dedifferentiation of these cells, as indicated by a decrease of the expression of several hepatocytespecific functions (14,17). Therefore, the question arises whether the elevated IGF-I transcript level in the presence of insulin reflects a stimulation or a diminished decline of IGF-I mRNA levels. The former is the case if IGF-I mRNA levels rise above the starting level during the course of the incubation with insulin, and the latter is the case if the starting level is never exceeded. To distinguish between these two possibilities, we examined the time course of the effect of insulin on IGF-I message levels. Figure 4 shows that, in the presence of insulin, IGF-I transcript levels increased above the O-point value and reached a maximum after 2 h. Thereafter, they gradually declined below the O-point level but always remained above the levels of the hormone-free controls (1.55-fold higher). These data show that insulin treatment not only diminishes the decline of IGF-I mRNA levels but also results in a stimulation above the level at the start of the experiment. DISCUSSION
Using a primary hepatocyte culture system, we examined the question of whether insulin is capable of regulating IGF-I mRNA levels and what the relative contri-
LEVELS
IN
HEPATOCYTES
butions of GH, a well known regulator of IGF-I synthesis, and of insulin might be. From the results obtained with this system, we can draw the following conclusions. Insulin rapidly stimulates IGF-I mRNA levels (2.5fold stimulation), with a half-maximal effect at 4.7 x 10ml’ M, and this insulin effect is independent of GH. GH alone is more potent than insulin (3.8-fold stimulation), and, when both hormones are added together, their effects are additive (6.7-fold stimulation). Johnson et al. (19) published results that differ from ours. They reported that an increased level of IGF-I transcripts occurred after exposure of hepatocytes to GH but that the presence of insulin did not further enhance the GHmediated accumulation of IGF-I transcripts. All other previous reports examined the effect of insulin on IGF-I production at the protein level and in different systems. In liver perfusion systems, insulin alone had no effect, only in combination with GH (11). In organ cultures and in monolayer cultures, 1.3- to 1.45-fold higher IGF-I levels were observed after a culture period of 3 days in the presence of insulin (2, 11, 20). Although the effects of insulin in these systems and at the IGF-I protein level are not entirely consistent and not as pronounced as at the mRNA level, they are in agreement with our finding that insulin treatment results in elevated IGF-I mRNA levels when compared with controls. Most likely the less marked stimulations observed at the protein level are due to the fact that, to have sufficient material to measure IGF-I concentrations, IGF-I released over a period of several days was analyzed in most studies. At the mRNA level, momentary levels can be determined at any time point, and there are marked differences under the influence of insulin, GH, or both. These advantages enabled us to examine time courses and concentration dependencies and to obtain better quantitative estimates of the relative contributions of GH and of insulin. The finding that insulin augments IGF-I mRNA levels relative to controls in primary hepatocytes strongly supports in vivo observations, indicating that insulin not only affects IGF-I levels by correcting metabolic perturbations but, in addition, by directly regulating IGF-I synthesis in the liver. Interestingly, the regulatory effect of insulin is not restricted to the stimulation of hepatic IGF-I mRNA levels but, in addition, insulin also appears to regulate the hepatic expression of IGF binding proteins. We have previously shown that insulin therapy of diabetic rats suppresses the elevated levels of hepatic IGFBP-2 mRNA (4). Similarly, insulin in vitro suppressed IGFBP-2 mRNA accumulation in cultured hepatocytes (5). The regulatory role of insulin for IGF-I production in the liver is summarized in Fig. 5. Insulin, together with GH, positively regulates IGF-I mRNA expression. GH is more potent than insulin in augmenting IGF-I mRNA levels, but it has no effect on IGFBP2. Insulin strongly suppresses IGFBP-2 mRNA levels and consequently IGFBP-2 synthesis (5). Both effects of insulin in vitro were observed within 2-4 h at physiological concentrations of the hormone (5). Because hepatic IGF-I mRNA levels are correlated with circulating IGF-I concentrations (4), which in turn are strongly correlated with growth parameters (30), the physiological significance of an upregulation of hepatic IGF-I mRNA levels is evident. However, the role of the
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
INSULIN
STIMULATES
IGF-I MRNA
suppression of IGFBP-2 mRNA levels by insulin remains to be elucidated. Taken together, the finding that insulin directly regulates hepatic IGF-I mRNA levels as well as hepatic IGFBP-2 mRNA levels emphasizes the importance of insulin as a regulator of hepatic IGF-I synthesis. We thank J. Schwander and P. Rotwein for providing cDNA probes for rat IGF-I. We also acknowledge the excellent technical assistance of Ani Ohannessian, as well as the competent secretarial help of Martha Salman. This work was supported by grants 3.914-0.88 and 3.046-0.87 from the Swiss National Science Foundation. Address reprint requests to M. Boni-Schnetzler. Received 7 June 1990; accepted in final form 5 February 1991. REFERENCES 1. BALLARD, J., R. C. BAXTER, M. BINOUX, D. CLEMMONS, S. DROP, K. HALL, R. L. HINTZ, M. RECHLER, E. RUTANEN, AND J. SCHWANDER. On the nomenclature of the IGF binding proteins. Acta Endocrinol. 121: 751-752, 1989. 2. BINOUX, M., C. LASSARE, AND N. HARDOUIN. Somatomedin production by rat liver in organ culture. Acta Endocrinol. 99: 422-430, 1982. 3. BOELSTERLI, U. A., P. BONIS, J.-F., BROUILLARD, AND P. DONATSCH. In vitro toxicity assessment of cyclosporin A and its analogs in a primary rat hepatocyte culture model. Z’oxicol. Appl. Pharmacol. 96: 212-221,1988. 4. B~NI-SCHNETZLER, M., K. BINZ, J.-L. MARY, C. SCHMID, J. SCHWANDER, AND E. R. FROESCH. Regulation of hepatic expression of IGF I and fetal IGF binding protein mRNA in streptozotocin-diabetic rats. FEBS Lett. 251: 253-256, 1989. 5. B~NI-SCHNETZLER, M., C. SCHMID, J.-L. MARY, B. ZIMMERLI, P. J. MEIER, J. ZAPF, J. SCHWANDER, AND E. R. FROESCH. Insulin regulates the expression of the insulin-like growth factor binding protein 2 (IGFBP-2) mRNA in rat hepatocytes. Mol. Endocrinol. 4:1320-1326,199O. 6. BUCHER, H., J. ZAPF, T. TORRESANI, A. PRADER, E. R. FROESCH, AND R. ILLIG. Insulin-like growth factor I and II, prolactin and insulin in 19 growth hormone-deficient children with excessive, normal, or decreased longitudinal growth after operation for craniopharyngioma. lV. Engl. J. Med. 309: 1142-1146, 1983. 7. CARLSSON, L. M. S., R. G. CLARK, A. SKOTTNER, AND I. C. A. F. ROBINSON. Growth hormone and growth in diabetic rats: effects of insulin and insulin-like growth factor I infusions. J. Endocrinol. 122:661-670,1989. 8. CHEN, R. F. Removal of fatty acids from serum albumin by charcoal treatment. J. Biol. Chem. 242: 173-181, 1967. 9. CHURCH, G. M., AND W. GILBERT. Genomic sequencing. Proc. Natl. Acad. Sci. USA 81: 1991-1995, 1984. 10. CLEVELAND, D. W., M. A. LOPATA, R. J. MAC DONALD, N. J. COWAN, W. J. RUTTER, AND M. W. KIRSCHNER. Number and evolutionary conservation of cy- and P-tubulin and cytoplasmic pand gamma-actin genes using specific cloned cDNA probes. CeZl 20:95-105,198O. 11. DAUGHADAY, W. H., L. S. PHILLIPS, AND M. C. MUELLER. The effects of insulin and growth hormone on the release of somatomedin by isolated rat liver. Endocrinology 98: 1214-1219, 1976. 12. ELGIN, R., W. BUSBY, AND D. CLEMMONS. An insulin-like growth factor (IGF) binding protein enhances the biological response to IGF I. Proc. Natl. Acad. Sci. USA 84: 3254-3258, 1987. 13. EMLER, C. A., AND D. S. SCHALCH. Nutritionally-induced changes in hepatic insulin-like growth factor I (IGF I) gene expression in rats. Endocrinology 120: 832-834, 1987. 14. FOLIOT, A., D. GLAISE, S. ERLINGER, AND C. GUGUEN-GUILLOUZO. Long-term maintenance of taurocholate uptake by adult rat hepatocytes co-cultured with a liver epithelial cell line. Hepatology5: 215-219,1985. 15. FROESCH, E. R., H.-P. GULER, C. SCHMID, M. ERNST, P. ZENOBI, AND J. ZAPF. Growth promotion by insulin-like growth factor I: endocrine and autocrine regulation. In: Perspectives in the Science of Growth and Development, edited by J. M. Tanner. London:
LEVELS
l6
17.
18
.
IN HEPATOCYTES
E851
Smith-Gordon, 1989, p. 251-263. GRIFFEN, S. C., S. M. RUSSELL, L. S. KATZ, AND C. S. NICOLL. Insulin exerts metabolic and growth-promoting effects by a direct action on the liver in vivo: clarification of the functional significance of the portal vascular link between the beta cells of the pancreatic islets and the liver. Proc. Natl. Acad. Sci. USA 84: 73007304,1987. GUGUEN-GUILLOUZO, C., B. CLEMENT, G. BAFFET, C. BEAUMONT, E. MOREL-CHANY, D. GLAISE, AND A. GUILLOUZO. Maintenance and reversibility of active albumin secretion by adult rat hepatocytes co-cultured with another liver epithelial cell type. Exp. CelZ Res. 143:47-54, 1983. HOSKINS, P. J., R. D. G. LESLIE, AND D. A. PYKE. Height at diagnosis of diabetes in children: a study in identical twins. Br. Med.
J. 290: 278-280,1985.
19. JOHNSON, T. R., B. K. BLOSSEY, C. W. DENKO, AND J. ILAN. Expression of insulin-like growth factor I in cultured rat hepatocytes: effects of insulin and growth hormone. MoZ. EndocrinoZ. 3: 580-587,1989. M., K. TAKUNO, K. ASAHAWA, N. HIZUHA, T. TSUSH20. KOGAWA, IMA, AND K. SHIZUME. Insulin stimulation of somatomedin A production in monolayer cultures of rat hepatocytes. Acta Endocrinol. 103: 385-390, 1983. 21. KNAUER, D. J., AND G. L. SMITH. Inhibition of biologic activity of multiplication-stimulating activity by binding to its carrier protein. Proc. Natl. Acad. Sci. USA 77: 7252-7256, 1980. T., E. F. FRITSCH, AND J. SAMBROOK. Molecular Clon22. MANIATIS, ing. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1982, p. 196-205. 23. MATHEWS, L. S., G. NORSTEDT, AND R. D. PALMITER. Regulation of insulin-like growth factor I gene expression by growth hormone. Proc. Natl. Acad. Sci. USA 83: 9343-9347, 1986. 24. O’HARE, T., AND P. F. PILCH. Intrinsic kinase activity of the insulin receptor. J. Biol. Chem. 264: 602-610, 1989. E., AND R. E. HUMBEL. The amino acid sequence 25. RINDERKNECHT, of human insulin-like growth factor I and its structural homology with proinsulin. J. Biol. Chem. 253: 2769-2776, 1978. C. T., A. L. BROWN, D. E. GRAHAM, S. SEELIG, S. 26. ROBERTS, BERRY, K. H. GABBAY, AND M. M. RECHLER. Growth hormone regulates the abundance of insulin-like growth factor I RNA in adult rat liver. J. Biol. Chem. 261: 10025-10028, 1986. 27. SALAMON, E. A., J. Luo, AND L. J. MURPHY. The effect of acute and chronic insulin administration on insulin-like growth factor-I expression in the pituitary-intact and hypophysectomized rat. Diabetologia 32: 348-353, 1989. 28. SALTER, J., AND C. H. BEST. Insulin as a growth hormone. Br. Med.
J. 2: 353-356,1953.
29. SCHALCH, D. S., U. E. HEINRICH, B. DRAZUIN, C. J. JOHNSON, AND L. L. MILLER. Role of liver in regulating somatomedin activity: hormonal effects on the synthesis and release of insulin-like growth factor and its carrier protein by the isolated perfused rat liver. Endocrinology 104:1143-1151,1979. E., H.-P. GULER, J. MERRYWEATHER, C. SCAN30. SCHEIWILLER, DELLA, W. MAERKI, J. ZAPF, AND E. R. FROESCH. Growth restoration of insulin-deficient diabetic rats by recombinant human insulin-like growth factor I. Nature Land. 323: 169-171, 1986. 31. SCHWANDER, J. C., C. HAURI, J. ZAPF, AND E. R. FROESCH. Synthesis and secretion of insulin-like growth factor and its binding protein by perfused rat liver: dependence on growth hormone status. Endocrinology 113: 297-305, 1983. 32. SCOTT, C. D., J. MARTIN, AND R. C. BAXTER. Rat hepatocyte insulin-like growth factor I and binding protein: effect of growth hormone in vitro and in vivo. Endocrinology 116: 1102-1107,1985. 33. SHIMATSU, A., AND P. ROTWEIN. Mosaic evolution of the insulinlike growth factors. Organization, sequence, and expression of the rat insulin-like growth factor I gene. J. BioZ. Chem. 262: 7894-7900, 1987. 34. WINTER, R. J., L. S. PHILLIPS, 0. C. GREEN, AND H. S. TRAISMAN. Somatomedin activity in the Mauriac syndrome. J. Pediatr. 97: 598-600,198O. AND E. R. FROESCH. Acute 35. ZAPF, J., C. HAURI, M. WALDVOGEL, metabolic effects and half-lives of intravenously administered insulin-like growth factors I and II in normal and hypophysectomized rats. J. Clin. Invest. 77: 1768-1775, 1986.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
growth factor I mRNA
M. BijNI-SCHNETZLER, C. SCHMID, P. J. MEIER, AND E. R. FROESCH Metabolic Unit and Division of Clinical Pharmacology, Department of Medicine, University Hospital, 8091 Ziirich, Switzerland
B~NI-SCHNETZLER, M.,C. SCHMID, P.J. MEIER,AND E.R. FROESCH. Insulin regulates insulin-like growth factor I mRNA in rat hepatocytes. Am. J. Physiol. 260 (Endocrinol. Metab. 23): E846-E851, 1991.-To evaluate the regulatory role of growth hormone (GH) and insulin on insulin-like growth factor I (IGFI) mRNA levels, we employed primary rat hepatocytes. Cells were incubated for 16 h with 10 nM insulin, 10 nM GH, or a combination thereof, and IGF-I mRNA levels were analyzed by Northern blotting. Insulin results in 2.5-fold and GH in 3.8fold higher IGF-I mRNA levels than hormone-free controls, and a combination of insulin and GH had an additive effect (6.7-fold). The effect of 10 nM insulin was constant at variable GH concentrations. Therefore, GH and insulin affect IGF-I mRNA levels independently of each other. The half-maximal effective dose of insulin was 4.7 x 10sl’ M, and, in kinetic experiments, insulin was effective within 2 h. These findings demonstrate that insulin modulates hepatic IGF-I production by a direct regulation of the transcript levels of IGF-I. growth growth
hormone; regulation
insulin-like
growth
factor binding
proteins;
INSULIN-LIKE GROWTH FACTOR I (IGF-I) is a growthpromoting hormone with structural similarities to proinsulin (25). Growth indices are positively correlated with serum IGF-I concentrations (30). These appear to be regulated at the IGF-I mRNA levels in the liver (4), which is the predominant site of synthesis of circulating IGF-I (31). The primary regulators of IGF-I are growth hormone (GH) and the nutritional status (13, 23, 26). In addition, insulin therapy restores IGF-I mRNA levels toward normal in insulin-deficient growth-arrested rats (4). It has been assumed that insulin affects IGF-I levels in these diabetic rats indirectly via restoration of the impaired GH secretion (7, 27). However, there are three different observations in vivo that suggest that insulin might also have a direct effect on serum IGF-I levels. 1) Insulin treatment in combination with a high-carbohydrate diet significantly augmented skeletal growth in GH-deficient rats (28). 2) Poorly controlled diabetic children with normal or increased GH levels exhibit reduced growth that is associated with low levels of IGFI, whereas normal growth has been reported in hyperinsulinemic children, despite GH deficiency (6, 18, 34). 3) The delivery of insulin via the portal vein is more effective than systemic insulin in restoring growth and IGFI levels in diabetic and growth-arrested rats, indicating that insulin may, indeed, directly act on the liver to augment serum IGF-I concentrations (16). E846
0193-1849/91
$1.50 Copyright
In blood, IGF-I does not circulate as free hormone. It is tightly associated with high-affinity IGF binding proteins (IGFBPs). It was reported that these IGFBPs influence the half-life of IGF-I in vivo (35) and that, in different cell culture systems, they either inhibit or enhance IGF-I effects (12, 21). Insulin may affect IGF-I serum levels not only by stimulating hepatic IGF-I synthesis but also by regulating the levels of the various IGFBPs. Thus we have recently shown that insulin suppresses an IGFBP mRNA, the levels of which are high in the liver of insulin-deficient growth-arrested rats (4) and in primary hepatocytes of adult rats incubated without insulin (5). This IGFBP, recently termed IGFBP-2 (l), is abundant in fetal serum. It is virtually absent from serum of healthy adult animals where IGF-I is predominantly bound to IGFBP-3 (1). In the intact animal, there is a positive correlation between the GH and nutritional status (including insulin) on one hand and the levels of IGF-I and IGFBP-3 on the other hand. Because insulin influences GH secretion in vivo (7), it is rather difficult to discern in intact animals what the relative contributions of GH and of insulin with regard to IGF-I synthesis might be. We therefore resorted to an in vitro system consisting of primary hepatocytes from adult rats and evaluated the effect of insulin and of GH on IGF-I mRNA levels separately from each other. Because GH is a well-known stimulator of hepatic IGF-I production (23, 26), we focused this study on the characterization of the role of insulin. MATERIALS
AND
METHODS
Preparation of hepatocytes. Hepatocytes were isolated from livers of 180- to 220-g male Sprague-Dawley rats by Ca2+-free collagenase perfusion (3). The cells were then plated at a density of 7 x lo4 cells/cm2 on type I collagen-coated (Serva, Heidelberg, FRG) 60-mm dishes. The viability was assessedby trypan blue exclusion and was 285%. Th e culture media was phenol red-free Williams E (Animed, Basel, Switzerland) supplemented with penicillin (100 U/ml), streptomycin (0.1 mg/ml), insulin (10D7 M), dexamethasone (10B7 M), and 10% fetal calf serum. After a 3-h attachment period, the media were replaced by the above-described media devoid of fetal calf serum. Experiments were started 16-24 h after plating the cells. Addition of hormones to hepatocyte cultures. Before the start of experiments, hepatocytes were washed for a period of 2 h at 37°C by three changes of serum- and
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
INSULIN
STIMULATES
IGF-I
MRNA
hormone-free Williams E medium. Experiments were initiated by replacing the medium with 2.9 ml of fresh Williams E medium and by adding 0.1 ml of human serum albumin (HSA) without or with hormones to yield final concentrations to 10 nM insulin (gift of Novo Biolabs), 10 nM GH (Nordisk), or a combination thereof and 0.1 g/l of HSA. HSA was obtained from the Swiss Red Cross and was charcoal-treated before use for cell cultures (8). Hepatocytes were then incubated for 16 h or the times indicated in the respective experiments. RNA isolation and Northern blotting. Incubations with hormones were stopped by washing the cells three times with ice-cold phosphate-buffered saline. Subsequently, cells were removed from culture dishes using a rubber policeman and collected by centrifugation at 4°C. The pellets containing 5 x lo6 cells were then lysed by the addition of 2.5 ml of homogenization buffer [6 M guanidinium isothiocyanate (BRL, Gaithersburg, MD), 5 mM sodium citrate, pH 7.0, 0.1 M P-mercaptoethanol, and 0.5% sarkosyl (Sigma, St. Louis, MO)] and by vigorous vortexing. Total RNA was purified by a standard CsCl centrifugation method followed by extraction with a 4:l mixture of chloroform and l-butanol (22). RNA samples were finally dissolved in diethyl pyrocarbonate-treated water, and concentrations were determined spectrophotometrically. Denatured RNA (20 pg/lane) was electrophoresed in 1% formaldehyde containing agarose gels (22). RNA was transferred (22) to nylon membranes (Hybond-N, Amersham, UK) and were crosslinked with ultraviolet radiation (9). There were no indications for RNA degradation, as assessed by agarose gel electrophoresis in the presence of ethidium bromide and subsequent visual inspection. Hybridization and evaluation of blots. The following probes were used for hybridization: rat IGF-I cDNA corresponding to the genomic sequences between nucleotide 2,054 of exon 1 and nucleotide 868 of exon 5 (33) and a P-actin cDNA (10). They were labeled by random primer extension (kit from Boehringer, Mannheim, FRG) using [a-32P]deoxycytidine 5’triphosphate (Amersham, UK) to a specific activity of l-4 X 10’ counts per min (cpm)/pg. Prehybridization and hybridization were carried out in a hybridization incubator (GFL model 7601, Burgwedel, GFR). Prehybridization solutions consist of ~5 standard sodium citrate (SSC), ~5 Denhardt’s solution, 100 mM Na2HP04 (pH 6.5), 50% formamide, 0.5% sodium dodecyl sulfate (SDS), and 100 pg/ml salmon sperm DNA. All stock solutions used to mix prehybridization and hybridization solutions were prepared according to Maniatis et al. (22). For hybridization, 1 rig/ml of labeled probe was added to a solution consisting of ~5 sodium-EDTA-tris(hydroxymethyl)aminomethane, ~5 Denhardt’s solution, 100 mM Na2HP04 (pH 6.5), 50% formamide, 100 pg/ml salmon sperm DNA, and 0.2% SDS. After hybridization for 48 h, membranes were washed three times for 20 min at 50°C with x0.2 SSC and 0.1% SDS and were exposed to Kodak X-Omat AR film in the presence of a Cronex lightening plus enhancer screen. Quantitative estimates of the relative mRNA levels were obtained either by excising the 32Plabeled bands from the membranes and by counting their Cerenkov radiation or by the determination of the optical densities of bands from autoradiograms using a video
LEVELS
IN
E847
HEPATOCYTES
densitometer (Bio-Rad, model 620). Specific hybridization was directly proportional to the amount of RNA applied if there was no more than 1.25 pg RNA/10 mm2 filter area applied as determined by slot blotting. Furthermore, the optical density (OD) of a band on the autoradiogram was proportional to the radioactivity contained in this band up to an OD of 12/10 mm2 filter area. Film exposure times of 6-16 h typically yielded bands with an intensity 80% of the total amount of IGF-I mRNA, whereas the other two smaller species were much less abundant. Within each blot, the ratio of the large 8.2-kb message to the smaller messages was found to be constant and independent of the treatment of cells before RNA isolation. We therefore
28s~
18s~
8 7 6 5 & 3 2 1 GH INS IGF
Jh 16h-
-Oh I mRNA (GH).
1. Insulin-like growth factor I (IGF-I) of hepatocytes with insulin, IGF-I, or growth hormone Primary hepacultures established as described under MATERIALS AND METHODS were treated for 16 h with 10 nM insulin (INS), 10 nM GH. a combination thereof, with 10 nM human recombinant IGF-I (gift of W. R. Rutter, Chiron, Emeryville, CA, and of J. Niiesch, Ciba-Geigy, Basel. Switzerland), or a combination of GH and IGF-I. RNA was then extracted and subjected to Northern blotting using a rat cDNA probe for IGF-I. For comparison, a sample was also taken at start of incubations with hormones (0 h). Top: autoradiograph. Bottom: quantitative estimate of amount of large 8.2-kb message, as determined by scanning densitometry of autoradiogram. Relative levels of 8.2-kb IGF-I mRNA species are expressed as optical densities (OD). FIG.
treated tocyte
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E848
INSULIN
TABLE
on IGF-I
STIMULATES
1. Effect of 10 nM insulin and/or mRNA levels in hepatocytes Expt
Insulin
GH
1
3.53 1.90 2.85 1.81 2.32 2.48k0.71
4.56 4.13 4.33 2.26 3.78 3.81kO.91
2 3 4 5 Means
k SD
IGF-I
MRNA
+ GH
HEPATOCYTES
Dose dependence of the effect of insulin on IGF-I mRNA levels. In all previous experiments, we used 10 nM insu-
7.65 7.57 7.71 3.65 6.84 6.68t1.73
Hepatocyte cultures were treated for 16 h, and RNA was extracted and analyzed by Northern blotting for insulin-like growth factor I (IGF-I) mRNA levels as described in legend to Fig. 1. Relative IGF-I mRNA levels are expressed as ratio of optical densities of hormonetreated to hormone-free cultures. GH, growth hormone.
used the signal of the 8.2-kb message to determine relative levels of IGF-I mRNA.
IN
more potent (381 t 91%). In the presence of insulin and GH, IGF-I mRNA levels were 668 t 173% of the control levels, i.e., the effects of the two hormones were additive, and a synergism was not discernible.
10 nM GH Insulin
LEVELS
the
Effect of insulin, GH, and IGF-I on IGF-I mRNA levels in hepatocytes. To examine the contributions of GH and
of insulin to the regulation of hepatocyte IGF-I mRNA levels, we incubated cells with the respective hormones at a concentration of 10 nM. After 16 h, RNA was isolated and subjected to Northern blotting. Figure 1 shows that the level of IGF-I mRNA is very low in the absence of hormones and‘is markedly elevated by GH. In the presence of insulin alone, IGF-I mRNA levels are also increased, and a combination of insulin and GH resulted in an even more pronounced rise. No significant effect on IGF-I mRNA levels was observed upon treatment with 10 nM IGF-I, for which the lack of IGF-I receptors on liver cells of adult rats may be responsible. RNA isolated at the beginning of the hormonal treatment (time 0) contains higher IGF-I mRNA levels than RNA isolated from cells that were incubated for 16 h in the absence of hormones (see also Fig. 4). Table 1 shows a quantitative estimate of the relative contributions of insulin and GH, each alone and in combination, to the elevation of IGF-I mRNA levels. The results stem from five independent preparations of primary hepatocytes exposed to hormones for 16 h. In the presence of insulin, IGF-I mRNA levels were consistently higher, i.e., 248 t 71% (SD), than in hormone-free controls. GH alone was
lin, i.e., a dose that saturates insulin receptors and one that is expected to produce maximal effects. To find out whether insulin is effective at physiological concentrations, we determined the dose dependency of the insulin stimulation of IGF-I mRNA levels. Figure 2 shows a representative experiment. In three different hepatocyte preparations the insulin concentration eliciting a halfmaximal response averaged 4.7 X 10ml*t 1.6 X lo-‘* M. This dose corresponds to the dissociation constant of insulin binding to the rat liver insulin receptor (24). We conclude that insulin regulates IGF-I mRNA levels well within physiological concentrations in a dose-dependent manner. Contributions of various insulin and GH concentrations to the regulation of IGF-I mRNA levels. The finding that
insulin alone can elevate IGF-I mRNA levels relative to hormone-free controls and that the increase adds up to that obtained with GH indicates that the effect of insulin is independent of GH. To test this prediction, we incubated hepatocytes for 16 h with various concentrations of GH in the presence or absence of 10 nM insulin and subjected the RNA of these cells to Northern blotting. Figure 3A, top, shows the Northern blot of IGF-I mRNA, and Fig. 3A, bottom, shows that of P-actin mRNA. The levels of actin mRNA were constant in all samples, indicating that equal amounts of RNA were loaded per lane and that there is little, if any, RNA degradation. This was a consistent finding in all experiments with hepatocytes where hybridizations with actin cDNA were performed. When the same blot was hybridized with an IGF-I cDNA probe, the increase of IGF-I mRNA levels depended on the GH concentration. A quantitation of this experiment is shown in Fig. 3B, left. The stimulatory effect of GH alone reached a plateau at 10ml*M. When insulin was added together with GH the net effect of insulin was smaller at low GH concentrations than at
I
0
10-l’
,;-I0
10‘-9 insulin
I-8 10
10’ -7
FIG. 2. Dose dependence of regulation of IGF-I mRNA levels by insulin. Dose dependence of insulin (n-axis) is expressed as a function of percent of maximal stimulation (y-axis) of IGF-I mRNA level. Hepatocytes were incubated for 16 h with insulin at indicated concentrations, and RNA was extracted and subsequently subjected to Northern blotting. OD of 8.2-kb bands were determined by scanning densitometry. To calculate dose eliciting half-maximal response, value obtained in absence of insulin was defined as 0% and value obtained at 3 x lOA M insulin as 100%. OD ratio of value obtained with insulin at 3 x lo-" M and that of control was 1.9:1. Values obtained at different insulin concentrations are expressed as percent of maximal effect of insulin.
(M)
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
INSULIN
GH (M)
0
16”
IO -10 ,69
,dS
,o-7
I
STIMULATES
o
,611
--INSULIN
,$O
,69
IGF-I
,&I
MRNA
LEVELS
IN
HEPATOCYTES
,67
FIG. 3. Effect of 10 nM insulin at variable GH concentrations and effect of 10 nM GH at variable insulin concentrations. Hepatocytes were incubated with 10 nM insulin at various dilutions of GH. After 16 h, relative IGF-I mRNA levels were determined by Northern blotting. A, top: IGF-I mRNA. A, bottom: actin message. Signals for actin did not significantly differ from each other. B, left: quantitation of IGF-I mRNA content obtained by excising 8.2-kb bands from blot and by counting their Cerenkov radiation. Solid lines: relative IGF-I mRNA in absence of insulin (-ins). Dashed line: levels in presence of insulin (+ins). B, right: IGF-I mRNA levels at variable insulin concentrations in absence (-GH) or presence of 10 nM GH (+GH). Results of both halves of B are derived from separate experiments.
+INSULIN
ACTIN*i
15oa
300
,A
‘\
/’ : /
‘a--
-0
E849
*ins
:250 Y 2 a200
E
z
150
;-GH
100
0
1 o-1’
10-10
GH
10-g
10-S
10-7
A
&I
(M)
high GH concentrations. However, at every GH dose the presence of insulin constantly resulted in 156 + 13% higher IGF-I mRNA levels. The results of the opposite experiment in which the effects of 10 nM GH at various insulin concentrations were examined are shown in Fig. 3B, right. Note that both halves of Fig. 3B contain unprocessed data (cpm of bound radioactive cDNA probe) and therefore the numbers are not directly comparable. However, the same conclusion, namely that insulin and GH independently regulate IGF-I mRNA, is obtained. Because the effect of insulin on IGF-I mRNA levels is smaller than that of GH, the dose-response curve with insulin alone is less steep than that with GH alone.
,‘o.lO
,;-9
,;-a
,;.7
insulin(M)
Consequently, there is also slightly more variation of the stimulation by GH, which averaged 340 + 61%. We conclude that GH and insulin affect IGF-I mRNA levels independently of each other and that both hormones together have strictly additive effects. Time course of insulin-induced elevation of IGF-I mRNA levels. In all previous experiments, hepatocytes
were incubated with or without insulin for a period of 16 h. In the absence of insulin, IGF-I mRNA levels were lower after 16 h than at the start of the incubation (see also Fig. 1). A time-dependent decline was also observed when IGF-I was analyzed at the protein level (20). The extent of the decline at the mRNA level was variable
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E850
INSULIN
STIMULATES
IGF-I
MRNA
-3 8 2 a2 E
HOURS
FIG. 4. Time course of IGF-I mRNA levels in presence or absence of insulin. One-day hepatocyte cultures were washed with 3 changes of hormone-free media over a period of 2 h. Subsequently, fresh media containing 10 nM insulin (solid line) or no insulin (dashed line) was at various times after start of added (time 0) ,a nd RNA was extracted by Northern blotting, and relative kinetics. IGF-I m RNA was analyzed levels of 8.2-kb message were determined by scanning densitometry and are expressed as OD as described in Fig. 1.
Growth
Hormone
0
0
B IGF I FIG. 5 . Schematic representation mRNA and IGF binding of IGF-I in rat liver.
Insulin
0
IGF BP2 of proposed hormonal regulation protein 2 (IGFBP-2) mRNA levels
among different hepatocyte preparations and most likely was due to dedifferentiation of these cells, as indicated by a decrease of the expression of several hepatocytespecific functions (14,17). Therefore, the question arises whether the elevated IGF-I transcript level in the presence of insulin reflects a stimulation or a diminished decline of IGF-I mRNA levels. The former is the case if IGF-I mRNA levels rise above the starting level during the course of the incubation with insulin, and the latter is the case if the starting level is never exceeded. To distinguish between these two possibilities, we examined the time course of the effect of insulin on IGF-I message levels. Figure 4 shows that, in the presence of insulin, IGF-I transcript levels increased above the O-point value and reached a maximum after 2 h. Thereafter, they gradually declined below the O-point level but always remained above the levels of the hormone-free controls (1.55-fold higher). These data show that insulin treatment not only diminishes the decline of IGF-I mRNA levels but also results in a stimulation above the level at the start of the experiment. DISCUSSION
Using a primary hepatocyte culture system, we examined the question of whether insulin is capable of regulating IGF-I mRNA levels and what the relative contri-
LEVELS
IN
HEPATOCYTES
butions of GH, a well known regulator of IGF-I synthesis, and of insulin might be. From the results obtained with this system, we can draw the following conclusions. Insulin rapidly stimulates IGF-I mRNA levels (2.5fold stimulation), with a half-maximal effect at 4.7 x 10ml’ M, and this insulin effect is independent of GH. GH alone is more potent than insulin (3.8-fold stimulation), and, when both hormones are added together, their effects are additive (6.7-fold stimulation). Johnson et al. (19) published results that differ from ours. They reported that an increased level of IGF-I transcripts occurred after exposure of hepatocytes to GH but that the presence of insulin did not further enhance the GHmediated accumulation of IGF-I transcripts. All other previous reports examined the effect of insulin on IGF-I production at the protein level and in different systems. In liver perfusion systems, insulin alone had no effect, only in combination with GH (11). In organ cultures and in monolayer cultures, 1.3- to 1.45-fold higher IGF-I levels were observed after a culture period of 3 days in the presence of insulin (2, 11, 20). Although the effects of insulin in these systems and at the IGF-I protein level are not entirely consistent and not as pronounced as at the mRNA level, they are in agreement with our finding that insulin treatment results in elevated IGF-I mRNA levels when compared with controls. Most likely the less marked stimulations observed at the protein level are due to the fact that, to have sufficient material to measure IGF-I concentrations, IGF-I released over a period of several days was analyzed in most studies. At the mRNA level, momentary levels can be determined at any time point, and there are marked differences under the influence of insulin, GH, or both. These advantages enabled us to examine time courses and concentration dependencies and to obtain better quantitative estimates of the relative contributions of GH and of insulin. The finding that insulin augments IGF-I mRNA levels relative to controls in primary hepatocytes strongly supports in vivo observations, indicating that insulin not only affects IGF-I levels by correcting metabolic perturbations but, in addition, by directly regulating IGF-I synthesis in the liver. Interestingly, the regulatory effect of insulin is not restricted to the stimulation of hepatic IGF-I mRNA levels but, in addition, insulin also appears to regulate the hepatic expression of IGF binding proteins. We have previously shown that insulin therapy of diabetic rats suppresses the elevated levels of hepatic IGFBP-2 mRNA (4). Similarly, insulin in vitro suppressed IGFBP-2 mRNA accumulation in cultured hepatocytes (5). The regulatory role of insulin for IGF-I production in the liver is summarized in Fig. 5. Insulin, together with GH, positively regulates IGF-I mRNA expression. GH is more potent than insulin in augmenting IGF-I mRNA levels, but it has no effect on IGFBP2. Insulin strongly suppresses IGFBP-2 mRNA levels and consequently IGFBP-2 synthesis (5). Both effects of insulin in vitro were observed within 2-4 h at physiological concentrations of the hormone (5). Because hepatic IGF-I mRNA levels are correlated with circulating IGF-I concentrations (4), which in turn are strongly correlated with growth parameters (30), the physiological significance of an upregulation of hepatic IGF-I mRNA levels is evident. However, the role of the
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
INSULIN
STIMULATES
IGF-I MRNA
suppression of IGFBP-2 mRNA levels by insulin remains to be elucidated. Taken together, the finding that insulin directly regulates hepatic IGF-I mRNA levels as well as hepatic IGFBP-2 mRNA levels emphasizes the importance of insulin as a regulator of hepatic IGF-I synthesis. We thank J. Schwander and P. Rotwein for providing cDNA probes for rat IGF-I. We also acknowledge the excellent technical assistance of Ani Ohannessian, as well as the competent secretarial help of Martha Salman. This work was supported by grants 3.914-0.88 and 3.046-0.87 from the Swiss National Science Foundation. Address reprint requests to M. Boni-Schnetzler. Received 7 June 1990; accepted in final form 5 February 1991. REFERENCES 1. BALLARD, J., R. C. BAXTER, M. BINOUX, D. CLEMMONS, S. DROP, K. HALL, R. L. HINTZ, M. RECHLER, E. RUTANEN, AND J. SCHWANDER. On the nomenclature of the IGF binding proteins. Acta Endocrinol. 121: 751-752, 1989. 2. BINOUX, M., C. LASSARE, AND N. HARDOUIN. Somatomedin production by rat liver in organ culture. Acta Endocrinol. 99: 422-430, 1982. 3. BOELSTERLI, U. A., P. BONIS, J.-F., BROUILLARD, AND P. DONATSCH. In vitro toxicity assessment of cyclosporin A and its analogs in a primary rat hepatocyte culture model. Z’oxicol. Appl. Pharmacol. 96: 212-221,1988. 4. B~NI-SCHNETZLER, M., K. BINZ, J.-L. MARY, C. SCHMID, J. SCHWANDER, AND E. R. FROESCH. Regulation of hepatic expression of IGF I and fetal IGF binding protein mRNA in streptozotocin-diabetic rats. FEBS Lett. 251: 253-256, 1989. 5. B~NI-SCHNETZLER, M., C. SCHMID, J.-L. MARY, B. ZIMMERLI, P. J. MEIER, J. ZAPF, J. SCHWANDER, AND E. R. FROESCH. Insulin regulates the expression of the insulin-like growth factor binding protein 2 (IGFBP-2) mRNA in rat hepatocytes. Mol. Endocrinol. 4:1320-1326,199O. 6. BUCHER, H., J. ZAPF, T. TORRESANI, A. PRADER, E. R. FROESCH, AND R. ILLIG. Insulin-like growth factor I and II, prolactin and insulin in 19 growth hormone-deficient children with excessive, normal, or decreased longitudinal growth after operation for craniopharyngioma. lV. Engl. J. Med. 309: 1142-1146, 1983. 7. CARLSSON, L. M. S., R. G. CLARK, A. SKOTTNER, AND I. C. A. F. ROBINSON. Growth hormone and growth in diabetic rats: effects of insulin and insulin-like growth factor I infusions. J. Endocrinol. 122:661-670,1989. 8. CHEN, R. F. Removal of fatty acids from serum albumin by charcoal treatment. J. Biol. Chem. 242: 173-181, 1967. 9. CHURCH, G. M., AND W. GILBERT. Genomic sequencing. Proc. Natl. Acad. Sci. USA 81: 1991-1995, 1984. 10. CLEVELAND, D. W., M. A. LOPATA, R. J. MAC DONALD, N. J. COWAN, W. J. RUTTER, AND M. W. KIRSCHNER. Number and evolutionary conservation of cy- and P-tubulin and cytoplasmic pand gamma-actin genes using specific cloned cDNA probes. CeZl 20:95-105,198O. 11. DAUGHADAY, W. H., L. S. PHILLIPS, AND M. C. MUELLER. The effects of insulin and growth hormone on the release of somatomedin by isolated rat liver. Endocrinology 98: 1214-1219, 1976. 12. ELGIN, R., W. BUSBY, AND D. CLEMMONS. An insulin-like growth factor (IGF) binding protein enhances the biological response to IGF I. Proc. Natl. Acad. Sci. USA 84: 3254-3258, 1987. 13. EMLER, C. A., AND D. S. SCHALCH. Nutritionally-induced changes in hepatic insulin-like growth factor I (IGF I) gene expression in rats. Endocrinology 120: 832-834, 1987. 14. FOLIOT, A., D. GLAISE, S. ERLINGER, AND C. GUGUEN-GUILLOUZO. Long-term maintenance of taurocholate uptake by adult rat hepatocytes co-cultured with a liver epithelial cell line. Hepatology5: 215-219,1985. 15. FROESCH, E. R., H.-P. GULER, C. SCHMID, M. ERNST, P. ZENOBI, AND J. ZAPF. Growth promotion by insulin-like growth factor I: endocrine and autocrine regulation. In: Perspectives in the Science of Growth and Development, edited by J. M. Tanner. London:
LEVELS
l6
17.
18
.
IN HEPATOCYTES
E851
Smith-Gordon, 1989, p. 251-263. GRIFFEN, S. C., S. M. RUSSELL, L. S. KATZ, AND C. S. NICOLL. Insulin exerts metabolic and growth-promoting effects by a direct action on the liver in vivo: clarification of the functional significance of the portal vascular link between the beta cells of the pancreatic islets and the liver. Proc. Natl. Acad. Sci. USA 84: 73007304,1987. GUGUEN-GUILLOUZO, C., B. CLEMENT, G. BAFFET, C. BEAUMONT, E. MOREL-CHANY, D. GLAISE, AND A. GUILLOUZO. Maintenance and reversibility of active albumin secretion by adult rat hepatocytes co-cultured with another liver epithelial cell type. Exp. CelZ Res. 143:47-54, 1983. HOSKINS, P. J., R. D. G. LESLIE, AND D. A. PYKE. Height at diagnosis of diabetes in children: a study in identical twins. Br. Med.
J. 290: 278-280,1985.
19. JOHNSON, T. R., B. K. BLOSSEY, C. W. DENKO, AND J. ILAN. Expression of insulin-like growth factor I in cultured rat hepatocytes: effects of insulin and growth hormone. MoZ. EndocrinoZ. 3: 580-587,1989. M., K. TAKUNO, K. ASAHAWA, N. HIZUHA, T. TSUSH20. KOGAWA, IMA, AND K. SHIZUME. Insulin stimulation of somatomedin A production in monolayer cultures of rat hepatocytes. Acta Endocrinol. 103: 385-390, 1983. 21. KNAUER, D. J., AND G. L. SMITH. Inhibition of biologic activity of multiplication-stimulating activity by binding to its carrier protein. Proc. Natl. Acad. Sci. USA 77: 7252-7256, 1980. T., E. F. FRITSCH, AND J. SAMBROOK. Molecular Clon22. MANIATIS, ing. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1982, p. 196-205. 23. MATHEWS, L. S., G. NORSTEDT, AND R. D. PALMITER. Regulation of insulin-like growth factor I gene expression by growth hormone. Proc. Natl. Acad. Sci. USA 83: 9343-9347, 1986. 24. O’HARE, T., AND P. F. PILCH. Intrinsic kinase activity of the insulin receptor. J. Biol. Chem. 264: 602-610, 1989. E., AND R. E. HUMBEL. The amino acid sequence 25. RINDERKNECHT, of human insulin-like growth factor I and its structural homology with proinsulin. J. Biol. Chem. 253: 2769-2776, 1978. C. T., A. L. BROWN, D. E. GRAHAM, S. SEELIG, S. 26. ROBERTS, BERRY, K. H. GABBAY, AND M. M. RECHLER. Growth hormone regulates the abundance of insulin-like growth factor I RNA in adult rat liver. J. Biol. Chem. 261: 10025-10028, 1986. 27. SALAMON, E. A., J. Luo, AND L. J. MURPHY. The effect of acute and chronic insulin administration on insulin-like growth factor-I expression in the pituitary-intact and hypophysectomized rat. Diabetologia 32: 348-353, 1989. 28. SALTER, J., AND C. H. BEST. Insulin as a growth hormone. Br. Med.
J. 2: 353-356,1953.
29. SCHALCH, D. S., U. E. HEINRICH, B. DRAZUIN, C. J. JOHNSON, AND L. L. MILLER. Role of liver in regulating somatomedin activity: hormonal effects on the synthesis and release of insulin-like growth factor and its carrier protein by the isolated perfused rat liver. Endocrinology 104:1143-1151,1979. E., H.-P. GULER, J. MERRYWEATHER, C. SCAN30. SCHEIWILLER, DELLA, W. MAERKI, J. ZAPF, AND E. R. FROESCH. Growth restoration of insulin-deficient diabetic rats by recombinant human insulin-like growth factor I. Nature Land. 323: 169-171, 1986. 31. SCHWANDER, J. C., C. HAURI, J. ZAPF, AND E. R. FROESCH. Synthesis and secretion of insulin-like growth factor and its binding protein by perfused rat liver: dependence on growth hormone status. Endocrinology 113: 297-305, 1983. 32. SCOTT, C. D., J. MARTIN, AND R. C. BAXTER. Rat hepatocyte insulin-like growth factor I and binding protein: effect of growth hormone in vitro and in vivo. Endocrinology 116: 1102-1107,1985. 33. SHIMATSU, A., AND P. ROTWEIN. Mosaic evolution of the insulinlike growth factors. Organization, sequence, and expression of the rat insulin-like growth factor I gene. J. BioZ. Chem. 262: 7894-7900, 1987. 34. WINTER, R. J., L. S. PHILLIPS, 0. C. GREEN, AND H. S. TRAISMAN. Somatomedin activity in the Mauriac syndrome. J. Pediatr. 97: 598-600,198O. AND E. R. FROESCH. Acute 35. ZAPF, J., C. HAURI, M. WALDVOGEL, metabolic effects and half-lives of intravenously administered insulin-like growth factors I and II in normal and hypophysectomized rats. J. Clin. Invest. 77: 1768-1775, 1986.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 8, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
