Effects of insulin on total RNA, poly(A)+ in primary cultures of rat hepatocytes CHING-JU
HSU, SCOT R. KIMBALL,
DAVID
RNA, and mRNA
A. ANTONETTI,
AND LEONARD
S. JEFFERSON
Department of Cellular and Molecular Physiology, College of Medicine, The Pennsylvania State University, Hershey, Pennsylvania 17033 Hsu, Ching-Ju, Scot R. Kimball, David A. Antonetti, and Leonard S. Jefferson. Effects of insulin on total RNA, poly(A)+ RNA, and mRNA in primary cultures of rat hepatocytes. Am. J. Physiol. 263 (Endocrinol. Metab. 26): Ell06Elll2,1992.-The purpose of this study was to examine mechanisms involved in the regulation of protein synthesis in primary cultures of rat hepatocytes. Hepatocytes were maintained in a chemically defined serum-free medium in the presence or absence of insulin. The rate of protein synthesis in hepatocytes deprived of insulin between days 2 and 5 of culture was reduced to 67% of the rate observed in insulin-maintained controls. The decrease in protein synthetic rate was accompanied by a proportional fall in the content of both total RNA and poly(A)+ RNA, suggesting that the capacity for protein synthesis was reduced in the absence of insulin. Both total RNA and poly(A)+ RNA contents and the protein synthetic rate were returned to control values after 3 days of insulin resupplementation. In addition, the effect of insulin on the expression of specific mRNAs was assessed by in vitro translation of total RNA followed by two-dimensional gel analysis of radiolabeled translation products. Only 13 of the greater than 150 spots discernible on the two-dimensional gels were altered in response to insulin. The mRNAs that were altered include examples of repression and stimulation of expression in response to insulin deprivation. Thus, in isolated rat hepatocytes, insulin regulates the capacity of both overall protein synthesis as well as the capacity for the synthesis of specific proteins. ribonucleic acid; deoxyribonucleic acid; serum-free medium; two-dimensional gel electrophoresis THE IMPORTANCE OF insulin in regulating the rate of protein synthesis in liver has been demonstrated by numerous studies describing the inhibition of protein synthesis that occurs in liver during diabetes and the corresponding increase in hepatic protein synthesis observed after insulin treatment of diabetic animals (reviewed in Ref. 11). Furthermore, the rate of protein synthesis is reduced in isolated perfused livers from diabetic rats (22). Addition of insulin to the perfusion medium stimulates protein synthesis in livers of mildly diabetic rats but has no effect on the rate of synthesis in livers from more severely diabetic animals (22). Also, treatment of diabetic rats with insulin restores protein synthesis to control levels (12). The inhibition of hepatic protein synthesis associated with diabetes involves the synthesis of proteins both retained within the liver as well as those secreted into the circulation (12). In addition to its effects on overall protein synthesis, insulin also regulates the synthesis of specific proteins in liver (reviewed in Ref. 20). Both positive and negative regulation of the synthesis of particular proteins has been described (20). Examples of proteins that are positively regulated in liver by insulin include glucokinase, pyruvate kinase, and glyceraldehyde-3-phosphate dehydrogenase. In contrast, insulin inhibits the expression of phosphoenolpyruvate carboxykinase. In each of these El106
0193-1849/92
$2.00
Copyright
cases, regulation of the expression of the protein by insulin involves changes in the amount of mRNA coding for that particular protein (20). In the present study, we have examined the regulation by insulin of protein synthesis in primary cultures of rat hepatocytes. In particular, the effect of insulin on total RNA and poly(A)+ RNA content was examined. Furthermore, changes in total RNA content were correlated with modulation of the rate of protein synthesis in response to insulin deprivation and resupplementation. Finally, in vitro translation of isolated total RNA followed by analysis of the translation products by twodimensional gel electrophoresis and fluorography was used to assess changes in the expression of specific mRNAs in response to insulin. EXPERIMENTALPROCEDURES Materials. Radioactive compounds ( [3H]uridylic acid and [ 35S]methionine) were purchased from Amersham. Ampholytes were from LKB. Collagenase (type 1A) was from either Sigma or Worthington. Hoechst dye 33258 was from CalbiochemBehring. Oligo dT-cellulose (type 7) was from Pharmacia, and insulin (protamine-zinc) and glucagon (crystalline) were from Eli Lilly. Polyadenylic acid and the chemicals used to prepare the chemically defined medium used for cell culture were purchased from Sigma. Preparation of hepatocytes. Hepatocytes were obtained from livers of rats perfused aseptically by the method described previously (5). After isolation, the cells were washed as described (5) with an additional centrifugation through Percoll, as described by Kreamer et al. (17); to separate viable from nonviable hepatocytes. The cells were placed and maintained in culture in serum-free chemically defined medium as previously described (5, 19) with the following modification. Before use, culture dishes (60 mm) were coated with collagen by applying to each dish 1 ml of a solution of 0.25% (wt/vol) rat tail collagen in 0.1% (vol/vol) acetic acid. The dishes were placed in a laminar flow hood and left overnight to dry. The dishes were then exposed to ultraviolet light for 30 min, washed two times with cold phosphate-buffered saline, 4 ml of medium were added, and the dishes were placed in an incubator at 37°C. After l-2 h of equilibration at 37”C, 2.7 x lo6 cells were plated on each dish, and the cultures were maintained exactly as described (19). Isolation
and quantitation
of
total RNA and poly(A)+
RNA.
Total RNA was extracted from cells harvested from 5-10 culture dishes using a modification (5) of the method described by Chirgwin et al. (3). The final sample was dissolved in sterile water at a concentration of l-2 mg/ml and stored at -70°C. The ratio of absorbance at 260 nm to that at 280 nm of the final solution was 2.0-2.2. The amount of poly(A)+ RNA present in the samples of total RNA was determined by a modification (14) of the method described by Zahringer et al. (31). Briefly, [3H]poly(U) was hybridized to each sample of total RNA in parallel with known amounts of a loo-base poly(A) standard. Single-stranded RNA
0 1992 The
American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
INSULIN-INDUCED
CHANGES
El107
IN RNA IN RAT HEPATOCYTES
was then digested with ribonuclease, and the resulting doublestranded RNA was collected on a nitrocellulose filter. The filter was dissolved in 10 ml Filtron X scintillation cocktail (National Diagnostics), and the radioactivity in the sample was measured in a Beckman LS-3801 scintillation counter. The concentration of poly(A) was then converted to total RNA using the previously established value of 4.3% as the amount of mRNA that is (3% Determination
POlY(A)
of poly(A) length. The average length of poly(A) in poly(A)+ RNA was determined as described by Rosen and Monahan (24). Briefly, poly(A)+ RNA was isolated from total RNA by chromatography on oligo dT-cellulose (26). Poly(A)+ RNA (40 pg) was then digested at 37°C for 1 h with 2 U ribonuclease T, and 6 U ribonuclease A in a total volume of 50 ~1of solution consisting of (in mM) 20 tris(hydroxymethyl)aminomethane. HCl, pH 7.6, 2 EDTA, and 300 NaCl. The reaction was stopped by placing the tubes on ice, and the degradation products were resolved by polyacrylamide gel electrophoresis. After electrophoresis, the gel was sliced into 3-mm pieces, and RNA was extracted from the slices by grinding the gel slices in 100 ~1 of 0.3 M NaCl followed by overnight incubation at 4°C. Gel slices containing poly(A) were determined by hybridization of the extracted RNA to [3H]poly(U), as described above. The length of poly(A) was estimated by comparison to poly(A) standards of known length electrophoresed in parallel with the samples. of
I
23 K
p
Measurement of rates of protein synthesis in primary cultures rat hepatocytes. Before the start of an experiment, the culture
medium was changed to medium containing 2.5 mM leucine, and the cells were incubated for 1 h at 37°C. At that time, 50 ~1 [3H]leucine (50 &i) were added to each plate of cells, and incubation at 37°C was continued for 20 min. These conditions permit the measurement of synthesis of total protein, i.e., newly synthesized secretory protein as well as retained protein (4). The cells were then washed three times with ice-cold phosphate-buffered saline, and the cells were harvested by scraping in 0.5 ml phosphate-buffered saline containing 0.5% Triton X-100. Each plate was then washed three times with 0.25 ml of the same buffer, and the washes were combined with the original sample. The extracts from two plates of cells were combined, and the samples were stored at -70°C. The incorporation of [3H]leucine into protein was determined by a slight modifi-
1
,
I
,
I
,
I
,
I
I
2
I
4
I
6
I
I
8
I
I 10
Day of Culture Fig. 1. Effect of insulin on RNA content of hepatocytes in primary culture. Hepatocytes were maintained in primary culture as described under EXPERIMENTAL PROCEDURES. Group A cells (0) were maintained in medium containing insulin throughout experiment, group B cells (w) were deprived of insulin from day 2 onward, and group C cells (0) were deprived of insulin between days 2 and 5 but were returned to insulincontaining medium on day 5. RNA and DNA content were measured as describedunder EXPERIMENTAL PROCEDURES.
c
30
t
0 1
B
.
0
I
I
2
I
I
4
I
6
I
I
I
8
Day of Culture Fig. 2. Effect of insulin on poly(A)+ RNA content of hepatocytes in primary culture. Conditions for these experiments were as described in legend to Fig. 1. Poly(A)+ RNA content was determined using equal amounts of total RNA from various samples, as described under EXPERIMENTAL PROCEDURES using established value that, on average, poly(A) tail represents 4.3% of length of average mRNA in rat liver (31).
cation (16) of the method previously described (4). Briefly, cells were lysed by the addition of Triton X-100 and sodium deoxycholate (final concn of 1% each) followed by vigorous vortexing. Aliquots (25 ~1) of the homogenate were spotted onto 24-mm filter disks (Whatman 540), and the filters were immediately placed in ice-cold 10% trichloroacetic acid. The filters were washed as described previously (16). Finally, protein was solubilized from the filters using Protosol (Du Pont-New England Nuclear) and Econofluor (National Diagnostics) before counting in a Beckman LS-3801 scintillation counter. It was previously established that the specific radioactivity of leucine bound to leucyl-tRNA was the same as the specific radioactivity of leucine in the medium under conditions similar to those used in the present experiment (6). Isolation
1
40
I
of
ribosomal
subunits
on sucrose density gradients.
Ribosomal subunits were isolated by sucrose density gradient centrifugation essentially as described by Flaim et al. (6). Briefly, hepatocytes were harvested by the addition of collagenase to the culture medium to a final concentration of 1 mg/ml and subsequent incubation at 37°C for 10 min. The cells were then washed three times with ice-cold phosphate-buffered saline, and the final cell pellet was resuspended in 500 ~1 of a solution consisting of (in mM) 25 N-2-hydroxyethylpiperazineN’-2-ethanesulfonic acid (HEPES), pH 7.5, 250 KCI, 1 magnesium acetate, 1 dithiothreitol, 250 sucrose, and 0.1 EDTA. The hepatocytes were then homogenized using a motor-driven glasson-glass homogenizer, and the homogenate was centrifuged at 10,000 g for 10 min at 4°C. The supernatant was made 1% in Triton X-100 and deoxycholate, and 400 ~1 of sample were layered onto 25-68% exponential sucrose gradients made up in buffer containing (in mM) 25 HEPES, pH 7.5, 250 KCl, 1
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
El108
INSULIN-INDUCED
CHANGES
IN RNA IN RAT HEPATOCYTES
2 I I
0.0
I I
I I
I I
I I
4
Day of Culture I
Fig. 4. Effect of insulin on number of free ribosomal subunits isolated on sucrose density gradients. Hepatocytes were maintained in primary culture in medium containing insulin (open bars) for either 2 or 4 days. Alternatively, cells were deprived of insulin for 2 h on day 2 (hatched bar) or between days 2 and 4 (filled bar). Cells were homogenized, and ribosomal subunits were isolated by sucrose density gradient centrifugation, as described under EXPERIMENTAL PROCEDURES. RNA content of fractions containing 40 S and 60 S ribosomal subunits was measured as described under EXPERIMENTAL PROCEDURES and are expressed as percentage of total RNA present in 10,000 g supernatant that was applied to sucrose density gradients. Results represent means t SE of 4-6 determinations in each group. 0.0
1
I
I
I
I
I
I
0
1
2
3
4
5
6
Day of Culture Fig. 3. Effect of insulin on rate of protein synthesis in hepatocytes in primary culture. Conditions for these experiments were as described in legend to Fig. 1. Rate of protein synthesis was determined by incorporation of [3H]leucine into total protein. Media was changed 1 h before start of experiment to media containing 2.5 mM leucine. [3H]leucine (5 Ci/mol; 50 &i/plate) was then added to media on plates, and plates were returned to incubator for 20 min. Under these conditions, synthesis of both secreted and retained proteins is measured because newly synthesized secretory proteins are not exported during 20-min labeling period (4). After labeling, plates were washed 3 times with ice-cold phosphate-buffered saline, and cells were harvested by scraping. Samples were stored frozen at -70°C until assayed for incorporation of [3H]leucine into protein (16). RNA and DNA analyses were performed on same samples as were used to measure protein synthesis and as described under EXPERIMENTAL PROCEDURES.
dithiothreitol, 0.1 EDTA, and 2 magnesium acetate. The gradients were then centrifuged at 35,000 revolutions/min in an SW41 rotor (Beckman Instruments) for 20 h at 4°C. Absorbance at 254 nm of the gradients was monitored, and fractions were collected using a density gradient fractionator (Instrument Specialties, Lincoln, NE). RNA translation, two-dimensional gel electrophoresis, and radiolabeled protein quantitation. RNA was translated in a RNA-
dependent reticulocyte lysate (Bethesda Research Laboratory), which had been optimized for potassium, spermidine, and RNA, using [ 35S]methionine (~1,000 Ci/mol) to radiolabel translation products. Before translation, RNA was heated to 65°C for 10 min and then rapidly cooled on ice. Translation reactions were performed in a total assay volume of 30 ~1 for 60 min at 30°C. An aliquot (5 ~1)of the reaction mixture was used to determine the amount of [35S]methionine incorporated into protein, as described previously (14). Two-dimensional gel electrophoresis was performed on equivalent amounts (200,000 counts/min) of sample by a slight modification of the procedure described previously (14). The two modifications include using a 6.2% rather than a 5.2% polyacrylamide gel in the first dimension and electrofocusing for 18,000 rather than 11,000 V h. The final gels were stained with l
Coomassie Blue R-250, destained, and subjected to fluorography, as described by Kent et al. (14). Fluorograms were scanned with a BioMolecular Dynamics laser scanner connected to a Sun 386i/150 workstation, and the results were analyzed using PDQuest I and II software (Protein Database) as previously described (14). Other analytical procedures. RNA was determined by alkaline hydrolysis (7) and DNA by the procedure of Labarca and Paigen (18) Statistics. The data are expressed as means t SE and compared with the appropriate controls using Student’s t test. P values ~0.05 were considered significant. RESULTS
The effect of insulin on the content of total RNA in primary cultures of rat hepatocytes is illustrated in Fig. 1. Cells were maintained in culture for various periods of time in the presence or absence of insulin, and RNA and DNA measurements were made after combining two dishes of cells per sample at each time point (samples were in triplicate). The RNA-to-DNA ratio did not change in the cells maintained with insulin over the period from day 2 to day 8 in culture (these cells will be referred to as group A) and was similar to the value (1.8 pg RNA/hg DNA) previously reported for freshly isolated hepatocytes (4). In contrast, depriving the cells of insulin for 3 days resulted in a reduction in RNA content to 63% of the control value (insulin-deprived cells will be referred to as group B). Extending the period of insulin deprivation to 6 days did not further influence the RNA-to-DNA ratio. Insulin addition to cells deprived of insulin for 3 days resulted in a return of the RNA-to-DNA ratio to the control value within 3 days (these cells will be referred to as group C). The amount of poly(A)+ RNA in isolated hepatocytes, calculated as a function of total RNA, was measured using the technique of Zahringer et al. (31) in which known quantities of total RNA and a 100-base poly(A) standard
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
INSULIN-INDUCEDCHANGESINRNAINRATHEPATOCYTES
are hybridized to [3H]poly(U). The accuracy of this method is dependent upon the samples being compared having, on average, the same length poly(A) tails. Therefore, the average length of poly(A) in the RNA samples was determined as described under EXPERIMENTAL PROCEDURES. The average poly(A) length in poly(A)+ RNA isolated from group A cells [66 + 5 nucleotides (nt)] was not significantly different than that in cells from group B (70 + 5 nt). The data shown in Fig. 2A demonstrate that insulin had no effect on the proportion of total RNA present as poly(A)+ RNA in isolated hepatocytes. However, since the content of total RNA decreased during insulin deprivation, the content of poly(A)+ RNA, calculated on the basis of DNA content, decreased in insulindeprived hepatocytes (Fig. 2B). Also, insulin restored poly(A)+ RNA content to control values within 3 days of insulin addition to insulin-deprived cells. Because ribosomal RNA represents 85% of total RNA (30), a reduction in total RNA to 63% of the control value in insulin-deprived hepatocytes (Fig. 1) should result in a decrease in the number of ribosomes and thus the capacity for protein synthesis. To ascertain whether or not this was the case, the rate of protein synthesis was measured in hepatocytes maintained in insulin-containing medium for 2 or 5 days and in hepatocytes deprived of insulin between days 2 and 5. The rate of protein synthesis in insulin-deprived cells was reduced to 67% (P < 0.05) of the rate observed in hepatocytes maintained in insulincontaining medium (Fig. 3A). However, if the rate of protein synthesis was expressed relative to the RNA content of the cells, no significant difference was observed
El109
between the two conditions (Fig. 3B). These results indicate that the capacity for protein synthesis was reduced in insulin-deprived hepatocytes compared with cells maintained in insulin-containing medium. A change in the efficiency of protein synthesis typically leads to an alteration in the cellular content of free ribosomal subunits. For example, if the rate of initiation of protein synthesis is slowed relative to elongation, then polysomes become disaggregated, and the content of free ribosomal subunits increases. In contrast, an inhibition of elongation relative to initiation would result in a decrease in free ribosomal subunits. However, insulin had no effect on the content of free ribosomal subunits in primary cultures of rat hepatocytes (Fig. 4). Therefore, the inhibition of protein synthesis that was observed in hepatocytes deprived of insulin was probably entirely due to a decrease in the capacity for protein synthesis with no change in efficiency. To assess changes in specific mRNAs occurring in isolated hepatocytes, total RNA was isolated from cells identified as groups A, B, and C. The RNA was translated in a messenger-dependent in vitro translation system, and [35S]methionine-labeled translation products were separated by two-dimensional gel electrophoresis, visualized by fluorography, and individual spots quantitated by densitometry. Of the ~150 spots discernable on each fluorogram, only 13 appeared to be differentially altered in response to insulin. The pattern of change in intensity of the 13 spots could be classified into one of two categories. Six spots, representing the two patterns of change, were circled and numbered in a separate key (Fig. 5). The first
Fig. 5. Effect of insulin on expression of mRNA isolated from primary cultures of rat hepatocytes. Total RNA was extracted from primary hepatocytes and translated in messenger-dependent in vitro translation system. Translation products were analyzed by two-dimensional gel electrophoresis. Gels were then treated for fluorography, dried, and exposed to X-ray film for 3 wk, as describedunder EXPERIMENTAL PROCEDURES. Photographs of fluorograms are shown. Translation products demonstrating major changes are circled in each panel and are identified by numbers in key. Group A cells were harvested on day 5 (A), group B cells were harvested on either day 5 (B) or day 8 (D), and group C cells were harvested on day 8 (C).
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
El110
INSULIN-INDUCED
CHANGES
pattern of change, represented by spots 3601, 4502, 5205, and 7101, showed a decrease after insulin deprivation of isolated hepatocytes (Fig. 6). A majority of the spots (11 of the 13 spots) exhibiting changes in response to insulin followed this pattern. In contrast, the second pattern, an increase in intensity in response to insulin deprivation, was only observed for two spots, 5301 and 5302. It should be noted that the 150 spots discernable on the fluorograms (Fig. 5) represent only a fraction of the total number of proteins present in hepatocytes. Proteins with isoelectric points outside of the range of the pH gradient or molecular weights larger or smaller than resolved by the gels used in these experiments would not have been detected. In addition, proteins with low-abundance mRNAs, with mRNAs that are inefficiently translated in reticulocyte lysate, or whose expression in response to insulin treatment or deprivation was only transient would not have been detected in these experiments. Therefore, the 13 spots that were observed to change in response to insulin in isolated hepatocytes in these experiments probably represent only a portion of the total number of proteins that can respond to insulin in these cells. DISCUSSION
Insulin has been shown to stimulate protein synthesis in a wide variety of insulin-responsive tissues, including liver, skeletal muscle, heart, and fat (reviewed in Ref. 11). The effect of insulin on protein synthesis in most tissues appears to be twofold. The initial effect is a stimulation of the efficiency of translation, i.e., an increase in the rate of protein synthesis without an accompanying change in
IN RNA IN RAT HEPATOCYTES
RNA content. In perfused skeletal muscle preparations, insulin causes an increase in the binding of [ 35S]methionine to 43s preinitiation complexes (13). Furthermore, insulin treatment of diabetic rats causes an increase in eukaryotic-initiation factor (eIF) -2p activity measured in muscle extracts (15). Together, these results suggest that the primary effect of insulin on the efficiency of protein synthesis in skeletal muscle occurs through a mechanism involving the initiation factor(s) eIF-2 and/or eIF-2B. Whether or not a similar mechanism is involved in the inhibition of the efficiency of protein synthesis in other tissues is unknown. However, it has been shown that insulin treatment of isolated adipocytes results in an increase in the aggregation of polysomes (29), and we have observed an increase in the aggregation of polysomes in livers from insulin-treated diabetic rats (unpublished observations). An increase in the state of aggregation of polysomes suggests that the initiation of protein synthesis is stimulated relative to elongation (25). The second effect of insulin on protein synthesis involves a stimulation of the capacity for protein synthesis (reviewed in Ref. 11). An increase in the capacity for protein synthesis could, in theory, involve an increase in the amount of any component of the translational apparatus, including initiation and elongation factors, ribosomes, individual mRNAs, aminoacyl-tRNA synthetases, ATP, and GTP. Tissue RNA content decreases significantly during diabetes in liver (1,12), skeletal muscle (l), and heart (1). Because the majority of RNA in liver is ribosomal, these results suggest that the content of ribosomes is decreased by insulin deprivation. Indeed, the spot4502
100
.
80 60
I
40 L
Fig. 6. Densitometric determination of changes in specific mRNAs in response to insulin in primary cultures of rat hepatocytes. Fluorograms illustrated in Fig. 5 were scanned with BioMolecular Dynamics scanner and analyzed using PDQuest software, as described under EXPERIMENTAL PROCEDURES. Changes in intensity of spots are presented as percentage of maximum observed for that spot. Group A cells were harvested on day 5 (condition A), group B cells were harvested on either day 5 (condition B) or day 8 (condition D), andgroup C cells were harvested on day 8 (condition
80
-
60
-
40
-
C).
Condition Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
INSULIN-INDUCED
CHANGES
accumulation of ribosomes is severely impaired in both gastrocnemius and heart from diabetic animals (1). In part, the fall in content of ribosomes in muscle during diabetes is due to an increase in the degradation of ribosomal proteins, and presumably ribosomes, relative to total tissue protein (1). In addition, in mouse myoblasts, insulin stimulates the translation of ribosomal protein mRNAs and the transcription of ribosomal DNA (9). Furthermore, insulin stimulates the expression of other mRNAs in liver (20) and skeletal muscle (14). Therefore, insulin regulates the capacity for protein synthesis through changes in both the amount of particular mRNAs as well as the number of ribosomes. In the present study, the capacity for protein synthesis in primary cultures of isolated hepatocytes was decreased by insulin deprivation. The decrease involved a reduction in both total RNA and poly(A)+ RNA contents. Insulin treatment of insulin-deprived cells resulted in a concomitant increase in both total and poly(A)+ RNA contents. The similarities in response of total RNA, which presumably reflects ribosomal RNA, and poly(A)+ RNA to insulin stimulation and deprivation suggest that similar mechanisms are involved in bringing about the alterations in these two parameters. Similar conclusions were reached in studies examining the effect of insulin on protein synthesis in gastrocnemius (14). The basis for the lack of effect of insulin on the efficiency of protein synthesis in primary cultures of hepatocytes is unknown. However, glucocorticoids have been shown to regulate efficiency in skeletal muscle and heart (23). Therefore, the presence of 0.1 PM dexamethasone in the culture medium in the present study may have masked any effect of insulin on the efficiency of protein synthesis in these cells. The changes that have been observed in the present study in the translational activities of a few mRNAs complement recent studies on insulin-induced transcriptional changes in skeletal (14) and cardiac muscles (2). As in those studies, we have found that insulin both increases and decreases the expression of specific mRNAs. However, in the other studies, insulin treatment of diabetic animals was used as an experimental model. During diabetes, the plasma concentrations of a number of hormones are altered. Thus a number of protein products were altered similarly by both insulin and other hormones, such as 3,5,3’-triiodothyronine, in heart (2). In the present study, isolated hepatocytes were maintained in serum-free medium containing defined concentrations of hormones. Therefore, the specific effect of insulin on mRNA expression could be examined. It should be noted that the use of in vitro translational assays to indirectly quantitate specific changes in gene expression must be interpreted with caution since the expression of some mRNAs is known to be translationally regulated in vivo. However, the expression of a number of proteins, as assessed by translational assay of isolated RNA (10,21,28), has been verified using cDNA probes for the same mRNAs (8, 27). In summary, we have shown that protein synthesis is inhibited in primary cultures of rat hepatocytes deprived of insulin and that insulin treatment of deprived cells
El111
IN RNA IN RAT HEPATOCYTES
completely restores the rate of protein synthesis to control levels. The basis for the inhibition of overall protein synthesis involves a decrease in capacity, with both total and poly(A)+ RNA content reduced by insulin deprivation. Finally, hepatocytes exhibit insulin-induced alterations in the transcriptional activities of specific genes that demonstrate patterns of either repression or increased expression in response to insulin administration. We are grateful to Bonnie Kleinhans and Jill Wolfgang for technical assistance and to Sharon Rannels for assistance in preparing the figures. This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-13499. Current address of C.-J. Hsu: Stranger-Cornell Cancer Research Laboratory, 510 East 73rd St., New York, NY 10021. Address for reprint requests: L. S. Jefferson, Dept. of Cellular and Molecular Physiology, College of Medicine, The Pennsylvania State University, PO Box 850, Hershey, PA 17033. Received 13 April 1992; accepted in final form 13 July 1992. REFERENCES 1. Ashford, A. J., and V. M. Pain. Effect of diabetes on the rates of synthesis and degradation of ribosomes in rat muscle and liver in vivo. J. Biol. Chem. 261: 4059-4065, 1986. 2. Barrieux, A., W. E. Neeley, and W. H. Dillman. Diabetesinduced alterations in the translational activity of specific messenger ribonucleic acids isolated from rat hearts. Circ. Res. 57: 296-303, 1985. 3. Chirgwin, J. Rutter.
J. M.,
A. E. Przybyla,
R. J. MacDonald,
and
W.
Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18: 5294-5299, 1979.
4. Everson, W. V., K. and L. S. Jefferson.
E. Flaim,
D.
M.
Susco,
S. R. Kimball,
Effect of amino acid deprivation on initiation of protein synthesis in rat hepatocytes. Am. J. Physiol. 256 (Cell Physiol. 25): C18-C27, 1989.
5. Flaim, Shiman,
K.
6. Flaim, Jefferson.
K.
E., S. M. Hutson, and L. S. Jefferson.
C. E. Lloyd,
J. M.
Taylor,
R.
Direct effect of insulin on albumin gene expression in primary cultures of rat hepatocytes. Am. J. Physiol. 249 (Endocrinol. Metab. 12): E447-E453, 1985. E.,
D.
E.
Peavy,
W.
E.
Everson,
and
L.
S.
The role synthesis in perfused resulting from amino through perfusion. J.
7.
8.
9.
10. 11.
of amino acids in the regulation of protein rat liver. I. Reduction in rates of synthesis acid deprivation and recovery during flowBiol. Chem. 257: 2932-2938, 1982. Fleck, A., and H. N. Munro. The precision of ultraviolet absorption measurements in the Schmidt-Tannhauser procedure for nucleic acid estimation. Biochim. Biophys. Acta 55: 571-583, 1962. Granner, D., T. Andreone, K. Sasaki, and E. Beale. Inhibition of transcription of the phosphoenolpyruvate carboxykinase gene by insulin. Nature Land. 305: 549-551, 1983. Hammond, M. L., and L. H. Bowman. Insulin stimulates the translation of ribosomal proteins and the transcription of rDNA in mouse myoblasts. J. Biol. Chem. 263: 17785-17791, 1988. Hill, R. E., K. L. Lee, an.d F. T. Kenney. Effects of insulin on mRNA activities in rat liver. J. Biol. Chem. 256: 1510-1513, 1981. Jefferson, L. S., K. E. Flaim, and D. E. Peavy. Protein metabolism. In: Diabetes Mellitus. Theory and Practice, edited by M. Ellenberg and H. Rifkin. New Hyde Park: Medical Examination, 1983, p. 47-59.
12. Jefferson,
L. S., W. S. L. Liao, and J. M. Taylor.
D. E. Peavy,
T. B. Miller,
Diabetes-induced alterations in liver protein synthesis. Changes in the relative abundance of mRNAs for albumin and other plasma proteins. J. Biol. Chem. 258: 1369-1375, 1983. 13. Kelly, F. J., and L. S. Jefferson. Control of peptide-chain initiation in rat skeletal muscle. Development of methods for preparation of native ribosomal subunits and analysis of the effect of insulin on formation of 40 S initiation complexes. J. Biol. Chem. 260: 6677-6683, 1985. 14. Kent, J. D., S. R. Kimball, and L. S. Jefferson. Effect of diabetes and insulin treatment of diabetic rats on total RNA, poly(A)+ RNA, and mRNA in skeletal muscle. Am: J. Physiol. 260 M. C. Appel,
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
El 112 (Cell
INSULIN-INDUCED Physiol.
29): C409-C416,
CHANGES
1991.
Kimball, S. R., and L. S. Jefferson. Effect of diabetes on guanine nucleotide exchange factor activity in skeletal muscle and heart. Biochem. Biophys. Res. Commun. 156: 706-711, 1988. 16. Kimball, S. R., and L. S. Jefferson. Mechanism of inhibition of protein synthesis by vasopressin in rat liver. J. Biol. Chem. 265:
IN
RNA
1990.
Kreamer, B., J. L. Staecker, V. Sawada, G. L. Sattler, M. T. S. Hsia, and H. C. Pitot. Use of a low-speed, iso-density Percoll centrifugation method to increase the viability of isolated rat hepatocyte preparations. J. Cell. Dev. Biol. 22: 201-211, 1986. 18. Labarca, C., and K. Paigen. A simple, rapid, and sensitive DNA assay procedure. Anal. Biochem. 102: 344-352, 1980. 19. Lloyd, C. E., J. E. Kalinyak, S. M. Hutson, and L. S. Jefferson. Stimulation of albumin gene transcription by insulin in primary cultures of rat hepatocytes. Am. J. Physiol. 252 (Cell
17.
Physiol.
21):
C205C214,
1987.
Meisler, M. H., and G. Howard. Effects of insulin on gene transcription. Annu. Rev. Physiol. 51: 701-714, 1989. 21. Palmiter, R. D. Differential rates of initiation of conalbumin and ovalbumin messenger ribonucleic acid in reticulocyte lysate. 20.
J. BioZ. Chem.
249:
6779-6787,
1974.
22.
Penhos, J. C., and M. E. Krahl. Stimulus of leucine incorporation into perfused liver protein by insulin. Am. J. Physiol. 204:
23.
Rannels, S. R., D. E. Rannels, A. Pegg, and L. S. Jefferson. Glucocorticoid effects on peptide-chain initiation in skeletal muscle and heart. Am. J. Physiol. 235 (Endocrinol. Metab. Gas-
140-142,
trointest.
1963.
Physiol.
4): E134-E139,
1978.
HEPATOCYTES
Rosen, J. M., and J. Monahan. Messenger RNA isolation, characterization, and hybridization analysis. In: Laboratory Meth-
24.
15.
16794-16798,
IN RAT
ods Manual for 1987, p. 4-25.
Hormone
Action
and
Molecular
Endocrinology.
25 Safer, B., R. Jagus, and W. M. Kemper. Analysis of initiation factor function in highly fractionated and unfractionated reticulocyte lysate systems. Methods Enzymol. 60: 61-87, 1979. Sambrook, J., E. F. Fritsch, and T. Maniatis. Molecular 26. Cloning: A Laboratory Manual, edited by C. Nolan. New York: Cold Spring Harbor Laboratory, 1989, vol. 2. 27. Sibrowski, W., and H. J. Seitz. Rapid action of insulin and cyclic AMP in the regulation of functional mRNA coding for glucokinase in rat liver. J. BioZ. Chem. 259: 343-346, 1984. Spence, J. T. Levels of translatable mRNA coding for rat liver 28* glucokinase. J. BioZ. Chem. 258: 9143-9146, 1983. 29. Vydelingum, N., R. L. Drake, J. Etienne, and A. H. Kissebah. Insulin regulation of fat cell ribosomes, protein synthesis, and lipoprotein lipase. Am. J. Physiol. 245 (Endocrinol. l
Metab.
8): E121-E131,
1983.
Young, V. R. The role of skeletal muscle and cardiac muscle in the regulation of protein metabolism. In: Mammalian Protein Metabolism, edited by M. N. Munro. New York: Academic, 1970, vol. 4, p. 585-674. J., N. Pritzl, and G. Stab. Quantitation of cardiac 31. Zahringer, polysomal mRNA by hybridization to [3H]poly(U). J. Cell. Dev.
30.
BioZ.
14: 539-550,
1982.
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
HSU, SCOT R. KIMBALL,
DAVID
RNA, and mRNA
A. ANTONETTI,
AND LEONARD
S. JEFFERSON
Department of Cellular and Molecular Physiology, College of Medicine, The Pennsylvania State University, Hershey, Pennsylvania 17033 Hsu, Ching-Ju, Scot R. Kimball, David A. Antonetti, and Leonard S. Jefferson. Effects of insulin on total RNA, poly(A)+ RNA, and mRNA in primary cultures of rat hepatocytes. Am. J. Physiol. 263 (Endocrinol. Metab. 26): Ell06Elll2,1992.-The purpose of this study was to examine mechanisms involved in the regulation of protein synthesis in primary cultures of rat hepatocytes. Hepatocytes were maintained in a chemically defined serum-free medium in the presence or absence of insulin. The rate of protein synthesis in hepatocytes deprived of insulin between days 2 and 5 of culture was reduced to 67% of the rate observed in insulin-maintained controls. The decrease in protein synthetic rate was accompanied by a proportional fall in the content of both total RNA and poly(A)+ RNA, suggesting that the capacity for protein synthesis was reduced in the absence of insulin. Both total RNA and poly(A)+ RNA contents and the protein synthetic rate were returned to control values after 3 days of insulin resupplementation. In addition, the effect of insulin on the expression of specific mRNAs was assessed by in vitro translation of total RNA followed by two-dimensional gel analysis of radiolabeled translation products. Only 13 of the greater than 150 spots discernible on the two-dimensional gels were altered in response to insulin. The mRNAs that were altered include examples of repression and stimulation of expression in response to insulin deprivation. Thus, in isolated rat hepatocytes, insulin regulates the capacity of both overall protein synthesis as well as the capacity for the synthesis of specific proteins. ribonucleic acid; deoxyribonucleic acid; serum-free medium; two-dimensional gel electrophoresis THE IMPORTANCE OF insulin in regulating the rate of protein synthesis in liver has been demonstrated by numerous studies describing the inhibition of protein synthesis that occurs in liver during diabetes and the corresponding increase in hepatic protein synthesis observed after insulin treatment of diabetic animals (reviewed in Ref. 11). Furthermore, the rate of protein synthesis is reduced in isolated perfused livers from diabetic rats (22). Addition of insulin to the perfusion medium stimulates protein synthesis in livers of mildly diabetic rats but has no effect on the rate of synthesis in livers from more severely diabetic animals (22). Also, treatment of diabetic rats with insulin restores protein synthesis to control levels (12). The inhibition of hepatic protein synthesis associated with diabetes involves the synthesis of proteins both retained within the liver as well as those secreted into the circulation (12). In addition to its effects on overall protein synthesis, insulin also regulates the synthesis of specific proteins in liver (reviewed in Ref. 20). Both positive and negative regulation of the synthesis of particular proteins has been described (20). Examples of proteins that are positively regulated in liver by insulin include glucokinase, pyruvate kinase, and glyceraldehyde-3-phosphate dehydrogenase. In contrast, insulin inhibits the expression of phosphoenolpyruvate carboxykinase. In each of these El106
0193-1849/92
$2.00
Copyright
cases, regulation of the expression of the protein by insulin involves changes in the amount of mRNA coding for that particular protein (20). In the present study, we have examined the regulation by insulin of protein synthesis in primary cultures of rat hepatocytes. In particular, the effect of insulin on total RNA and poly(A)+ RNA content was examined. Furthermore, changes in total RNA content were correlated with modulation of the rate of protein synthesis in response to insulin deprivation and resupplementation. Finally, in vitro translation of isolated total RNA followed by analysis of the translation products by twodimensional gel electrophoresis and fluorography was used to assess changes in the expression of specific mRNAs in response to insulin. EXPERIMENTALPROCEDURES Materials. Radioactive compounds ( [3H]uridylic acid and [ 35S]methionine) were purchased from Amersham. Ampholytes were from LKB. Collagenase (type 1A) was from either Sigma or Worthington. Hoechst dye 33258 was from CalbiochemBehring. Oligo dT-cellulose (type 7) was from Pharmacia, and insulin (protamine-zinc) and glucagon (crystalline) were from Eli Lilly. Polyadenylic acid and the chemicals used to prepare the chemically defined medium used for cell culture were purchased from Sigma. Preparation of hepatocytes. Hepatocytes were obtained from livers of rats perfused aseptically by the method described previously (5). After isolation, the cells were washed as described (5) with an additional centrifugation through Percoll, as described by Kreamer et al. (17); to separate viable from nonviable hepatocytes. The cells were placed and maintained in culture in serum-free chemically defined medium as previously described (5, 19) with the following modification. Before use, culture dishes (60 mm) were coated with collagen by applying to each dish 1 ml of a solution of 0.25% (wt/vol) rat tail collagen in 0.1% (vol/vol) acetic acid. The dishes were placed in a laminar flow hood and left overnight to dry. The dishes were then exposed to ultraviolet light for 30 min, washed two times with cold phosphate-buffered saline, 4 ml of medium were added, and the dishes were placed in an incubator at 37°C. After l-2 h of equilibration at 37”C, 2.7 x lo6 cells were plated on each dish, and the cultures were maintained exactly as described (19). Isolation
and quantitation
of
total RNA and poly(A)+
RNA.
Total RNA was extracted from cells harvested from 5-10 culture dishes using a modification (5) of the method described by Chirgwin et al. (3). The final sample was dissolved in sterile water at a concentration of l-2 mg/ml and stored at -70°C. The ratio of absorbance at 260 nm to that at 280 nm of the final solution was 2.0-2.2. The amount of poly(A)+ RNA present in the samples of total RNA was determined by a modification (14) of the method described by Zahringer et al. (31). Briefly, [3H]poly(U) was hybridized to each sample of total RNA in parallel with known amounts of a loo-base poly(A) standard. Single-stranded RNA
0 1992 The
American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
INSULIN-INDUCED
CHANGES
El107
IN RNA IN RAT HEPATOCYTES
was then digested with ribonuclease, and the resulting doublestranded RNA was collected on a nitrocellulose filter. The filter was dissolved in 10 ml Filtron X scintillation cocktail (National Diagnostics), and the radioactivity in the sample was measured in a Beckman LS-3801 scintillation counter. The concentration of poly(A) was then converted to total RNA using the previously established value of 4.3% as the amount of mRNA that is (3% Determination
POlY(A)
of poly(A) length. The average length of poly(A) in poly(A)+ RNA was determined as described by Rosen and Monahan (24). Briefly, poly(A)+ RNA was isolated from total RNA by chromatography on oligo dT-cellulose (26). Poly(A)+ RNA (40 pg) was then digested at 37°C for 1 h with 2 U ribonuclease T, and 6 U ribonuclease A in a total volume of 50 ~1of solution consisting of (in mM) 20 tris(hydroxymethyl)aminomethane. HCl, pH 7.6, 2 EDTA, and 300 NaCl. The reaction was stopped by placing the tubes on ice, and the degradation products were resolved by polyacrylamide gel electrophoresis. After electrophoresis, the gel was sliced into 3-mm pieces, and RNA was extracted from the slices by grinding the gel slices in 100 ~1 of 0.3 M NaCl followed by overnight incubation at 4°C. Gel slices containing poly(A) were determined by hybridization of the extracted RNA to [3H]poly(U), as described above. The length of poly(A) was estimated by comparison to poly(A) standards of known length electrophoresed in parallel with the samples. of
I
23 K
p
Measurement of rates of protein synthesis in primary cultures rat hepatocytes. Before the start of an experiment, the culture
medium was changed to medium containing 2.5 mM leucine, and the cells were incubated for 1 h at 37°C. At that time, 50 ~1 [3H]leucine (50 &i) were added to each plate of cells, and incubation at 37°C was continued for 20 min. These conditions permit the measurement of synthesis of total protein, i.e., newly synthesized secretory protein as well as retained protein (4). The cells were then washed three times with ice-cold phosphate-buffered saline, and the cells were harvested by scraping in 0.5 ml phosphate-buffered saline containing 0.5% Triton X-100. Each plate was then washed three times with 0.25 ml of the same buffer, and the washes were combined with the original sample. The extracts from two plates of cells were combined, and the samples were stored at -70°C. The incorporation of [3H]leucine into protein was determined by a slight modifi-
1
,
I
,
I
,
I
,
I
I
2
I
4
I
6
I
I
8
I
I 10
Day of Culture Fig. 1. Effect of insulin on RNA content of hepatocytes in primary culture. Hepatocytes were maintained in primary culture as described under EXPERIMENTAL PROCEDURES. Group A cells (0) were maintained in medium containing insulin throughout experiment, group B cells (w) were deprived of insulin from day 2 onward, and group C cells (0) were deprived of insulin between days 2 and 5 but were returned to insulincontaining medium on day 5. RNA and DNA content were measured as describedunder EXPERIMENTAL PROCEDURES.
c
30
t
0 1
B
.
0
I
I
2
I
I
4
I
6
I
I
I
8
Day of Culture Fig. 2. Effect of insulin on poly(A)+ RNA content of hepatocytes in primary culture. Conditions for these experiments were as described in legend to Fig. 1. Poly(A)+ RNA content was determined using equal amounts of total RNA from various samples, as described under EXPERIMENTAL PROCEDURES using established value that, on average, poly(A) tail represents 4.3% of length of average mRNA in rat liver (31).
cation (16) of the method previously described (4). Briefly, cells were lysed by the addition of Triton X-100 and sodium deoxycholate (final concn of 1% each) followed by vigorous vortexing. Aliquots (25 ~1) of the homogenate were spotted onto 24-mm filter disks (Whatman 540), and the filters were immediately placed in ice-cold 10% trichloroacetic acid. The filters were washed as described previously (16). Finally, protein was solubilized from the filters using Protosol (Du Pont-New England Nuclear) and Econofluor (National Diagnostics) before counting in a Beckman LS-3801 scintillation counter. It was previously established that the specific radioactivity of leucine bound to leucyl-tRNA was the same as the specific radioactivity of leucine in the medium under conditions similar to those used in the present experiment (6). Isolation
1
40
I
of
ribosomal
subunits
on sucrose density gradients.
Ribosomal subunits were isolated by sucrose density gradient centrifugation essentially as described by Flaim et al. (6). Briefly, hepatocytes were harvested by the addition of collagenase to the culture medium to a final concentration of 1 mg/ml and subsequent incubation at 37°C for 10 min. The cells were then washed three times with ice-cold phosphate-buffered saline, and the final cell pellet was resuspended in 500 ~1 of a solution consisting of (in mM) 25 N-2-hydroxyethylpiperazineN’-2-ethanesulfonic acid (HEPES), pH 7.5, 250 KCI, 1 magnesium acetate, 1 dithiothreitol, 250 sucrose, and 0.1 EDTA. The hepatocytes were then homogenized using a motor-driven glasson-glass homogenizer, and the homogenate was centrifuged at 10,000 g for 10 min at 4°C. The supernatant was made 1% in Triton X-100 and deoxycholate, and 400 ~1 of sample were layered onto 25-68% exponential sucrose gradients made up in buffer containing (in mM) 25 HEPES, pH 7.5, 250 KCl, 1
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
El108
INSULIN-INDUCED
CHANGES
IN RNA IN RAT HEPATOCYTES
2 I I
0.0
I I
I I
I I
I I
4
Day of Culture I
Fig. 4. Effect of insulin on number of free ribosomal subunits isolated on sucrose density gradients. Hepatocytes were maintained in primary culture in medium containing insulin (open bars) for either 2 or 4 days. Alternatively, cells were deprived of insulin for 2 h on day 2 (hatched bar) or between days 2 and 4 (filled bar). Cells were homogenized, and ribosomal subunits were isolated by sucrose density gradient centrifugation, as described under EXPERIMENTAL PROCEDURES. RNA content of fractions containing 40 S and 60 S ribosomal subunits was measured as described under EXPERIMENTAL PROCEDURES and are expressed as percentage of total RNA present in 10,000 g supernatant that was applied to sucrose density gradients. Results represent means t SE of 4-6 determinations in each group. 0.0
1
I
I
I
I
I
I
0
1
2
3
4
5
6
Day of Culture Fig. 3. Effect of insulin on rate of protein synthesis in hepatocytes in primary culture. Conditions for these experiments were as described in legend to Fig. 1. Rate of protein synthesis was determined by incorporation of [3H]leucine into total protein. Media was changed 1 h before start of experiment to media containing 2.5 mM leucine. [3H]leucine (5 Ci/mol; 50 &i/plate) was then added to media on plates, and plates were returned to incubator for 20 min. Under these conditions, synthesis of both secreted and retained proteins is measured because newly synthesized secretory proteins are not exported during 20-min labeling period (4). After labeling, plates were washed 3 times with ice-cold phosphate-buffered saline, and cells were harvested by scraping. Samples were stored frozen at -70°C until assayed for incorporation of [3H]leucine into protein (16). RNA and DNA analyses were performed on same samples as were used to measure protein synthesis and as described under EXPERIMENTAL PROCEDURES.
dithiothreitol, 0.1 EDTA, and 2 magnesium acetate. The gradients were then centrifuged at 35,000 revolutions/min in an SW41 rotor (Beckman Instruments) for 20 h at 4°C. Absorbance at 254 nm of the gradients was monitored, and fractions were collected using a density gradient fractionator (Instrument Specialties, Lincoln, NE). RNA translation, two-dimensional gel electrophoresis, and radiolabeled protein quantitation. RNA was translated in a RNA-
dependent reticulocyte lysate (Bethesda Research Laboratory), which had been optimized for potassium, spermidine, and RNA, using [ 35S]methionine (~1,000 Ci/mol) to radiolabel translation products. Before translation, RNA was heated to 65°C for 10 min and then rapidly cooled on ice. Translation reactions were performed in a total assay volume of 30 ~1 for 60 min at 30°C. An aliquot (5 ~1)of the reaction mixture was used to determine the amount of [35S]methionine incorporated into protein, as described previously (14). Two-dimensional gel electrophoresis was performed on equivalent amounts (200,000 counts/min) of sample by a slight modification of the procedure described previously (14). The two modifications include using a 6.2% rather than a 5.2% polyacrylamide gel in the first dimension and electrofocusing for 18,000 rather than 11,000 V h. The final gels were stained with l
Coomassie Blue R-250, destained, and subjected to fluorography, as described by Kent et al. (14). Fluorograms were scanned with a BioMolecular Dynamics laser scanner connected to a Sun 386i/150 workstation, and the results were analyzed using PDQuest I and II software (Protein Database) as previously described (14). Other analytical procedures. RNA was determined by alkaline hydrolysis (7) and DNA by the procedure of Labarca and Paigen (18) Statistics. The data are expressed as means t SE and compared with the appropriate controls using Student’s t test. P values ~0.05 were considered significant. RESULTS
The effect of insulin on the content of total RNA in primary cultures of rat hepatocytes is illustrated in Fig. 1. Cells were maintained in culture for various periods of time in the presence or absence of insulin, and RNA and DNA measurements were made after combining two dishes of cells per sample at each time point (samples were in triplicate). The RNA-to-DNA ratio did not change in the cells maintained with insulin over the period from day 2 to day 8 in culture (these cells will be referred to as group A) and was similar to the value (1.8 pg RNA/hg DNA) previously reported for freshly isolated hepatocytes (4). In contrast, depriving the cells of insulin for 3 days resulted in a reduction in RNA content to 63% of the control value (insulin-deprived cells will be referred to as group B). Extending the period of insulin deprivation to 6 days did not further influence the RNA-to-DNA ratio. Insulin addition to cells deprived of insulin for 3 days resulted in a return of the RNA-to-DNA ratio to the control value within 3 days (these cells will be referred to as group C). The amount of poly(A)+ RNA in isolated hepatocytes, calculated as a function of total RNA, was measured using the technique of Zahringer et al. (31) in which known quantities of total RNA and a 100-base poly(A) standard
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
INSULIN-INDUCEDCHANGESINRNAINRATHEPATOCYTES
are hybridized to [3H]poly(U). The accuracy of this method is dependent upon the samples being compared having, on average, the same length poly(A) tails. Therefore, the average length of poly(A) in the RNA samples was determined as described under EXPERIMENTAL PROCEDURES. The average poly(A) length in poly(A)+ RNA isolated from group A cells [66 + 5 nucleotides (nt)] was not significantly different than that in cells from group B (70 + 5 nt). The data shown in Fig. 2A demonstrate that insulin had no effect on the proportion of total RNA present as poly(A)+ RNA in isolated hepatocytes. However, since the content of total RNA decreased during insulin deprivation, the content of poly(A)+ RNA, calculated on the basis of DNA content, decreased in insulindeprived hepatocytes (Fig. 2B). Also, insulin restored poly(A)+ RNA content to control values within 3 days of insulin addition to insulin-deprived cells. Because ribosomal RNA represents 85% of total RNA (30), a reduction in total RNA to 63% of the control value in insulin-deprived hepatocytes (Fig. 1) should result in a decrease in the number of ribosomes and thus the capacity for protein synthesis. To ascertain whether or not this was the case, the rate of protein synthesis was measured in hepatocytes maintained in insulin-containing medium for 2 or 5 days and in hepatocytes deprived of insulin between days 2 and 5. The rate of protein synthesis in insulin-deprived cells was reduced to 67% (P < 0.05) of the rate observed in hepatocytes maintained in insulincontaining medium (Fig. 3A). However, if the rate of protein synthesis was expressed relative to the RNA content of the cells, no significant difference was observed
El109
between the two conditions (Fig. 3B). These results indicate that the capacity for protein synthesis was reduced in insulin-deprived hepatocytes compared with cells maintained in insulin-containing medium. A change in the efficiency of protein synthesis typically leads to an alteration in the cellular content of free ribosomal subunits. For example, if the rate of initiation of protein synthesis is slowed relative to elongation, then polysomes become disaggregated, and the content of free ribosomal subunits increases. In contrast, an inhibition of elongation relative to initiation would result in a decrease in free ribosomal subunits. However, insulin had no effect on the content of free ribosomal subunits in primary cultures of rat hepatocytes (Fig. 4). Therefore, the inhibition of protein synthesis that was observed in hepatocytes deprived of insulin was probably entirely due to a decrease in the capacity for protein synthesis with no change in efficiency. To assess changes in specific mRNAs occurring in isolated hepatocytes, total RNA was isolated from cells identified as groups A, B, and C. The RNA was translated in a messenger-dependent in vitro translation system, and [35S]methionine-labeled translation products were separated by two-dimensional gel electrophoresis, visualized by fluorography, and individual spots quantitated by densitometry. Of the ~150 spots discernable on each fluorogram, only 13 appeared to be differentially altered in response to insulin. The pattern of change in intensity of the 13 spots could be classified into one of two categories. Six spots, representing the two patterns of change, were circled and numbered in a separate key (Fig. 5). The first
Fig. 5. Effect of insulin on expression of mRNA isolated from primary cultures of rat hepatocytes. Total RNA was extracted from primary hepatocytes and translated in messenger-dependent in vitro translation system. Translation products were analyzed by two-dimensional gel electrophoresis. Gels were then treated for fluorography, dried, and exposed to X-ray film for 3 wk, as describedunder EXPERIMENTAL PROCEDURES. Photographs of fluorograms are shown. Translation products demonstrating major changes are circled in each panel and are identified by numbers in key. Group A cells were harvested on day 5 (A), group B cells were harvested on either day 5 (B) or day 8 (D), and group C cells were harvested on day 8 (C).
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
El110
INSULIN-INDUCED
CHANGES
pattern of change, represented by spots 3601, 4502, 5205, and 7101, showed a decrease after insulin deprivation of isolated hepatocytes (Fig. 6). A majority of the spots (11 of the 13 spots) exhibiting changes in response to insulin followed this pattern. In contrast, the second pattern, an increase in intensity in response to insulin deprivation, was only observed for two spots, 5301 and 5302. It should be noted that the 150 spots discernable on the fluorograms (Fig. 5) represent only a fraction of the total number of proteins present in hepatocytes. Proteins with isoelectric points outside of the range of the pH gradient or molecular weights larger or smaller than resolved by the gels used in these experiments would not have been detected. In addition, proteins with low-abundance mRNAs, with mRNAs that are inefficiently translated in reticulocyte lysate, or whose expression in response to insulin treatment or deprivation was only transient would not have been detected in these experiments. Therefore, the 13 spots that were observed to change in response to insulin in isolated hepatocytes in these experiments probably represent only a portion of the total number of proteins that can respond to insulin in these cells. DISCUSSION
Insulin has been shown to stimulate protein synthesis in a wide variety of insulin-responsive tissues, including liver, skeletal muscle, heart, and fat (reviewed in Ref. 11). The effect of insulin on protein synthesis in most tissues appears to be twofold. The initial effect is a stimulation of the efficiency of translation, i.e., an increase in the rate of protein synthesis without an accompanying change in
IN RNA IN RAT HEPATOCYTES
RNA content. In perfused skeletal muscle preparations, insulin causes an increase in the binding of [ 35S]methionine to 43s preinitiation complexes (13). Furthermore, insulin treatment of diabetic rats causes an increase in eukaryotic-initiation factor (eIF) -2p activity measured in muscle extracts (15). Together, these results suggest that the primary effect of insulin on the efficiency of protein synthesis in skeletal muscle occurs through a mechanism involving the initiation factor(s) eIF-2 and/or eIF-2B. Whether or not a similar mechanism is involved in the inhibition of the efficiency of protein synthesis in other tissues is unknown. However, it has been shown that insulin treatment of isolated adipocytes results in an increase in the aggregation of polysomes (29), and we have observed an increase in the aggregation of polysomes in livers from insulin-treated diabetic rats (unpublished observations). An increase in the state of aggregation of polysomes suggests that the initiation of protein synthesis is stimulated relative to elongation (25). The second effect of insulin on protein synthesis involves a stimulation of the capacity for protein synthesis (reviewed in Ref. 11). An increase in the capacity for protein synthesis could, in theory, involve an increase in the amount of any component of the translational apparatus, including initiation and elongation factors, ribosomes, individual mRNAs, aminoacyl-tRNA synthetases, ATP, and GTP. Tissue RNA content decreases significantly during diabetes in liver (1,12), skeletal muscle (l), and heart (1). Because the majority of RNA in liver is ribosomal, these results suggest that the content of ribosomes is decreased by insulin deprivation. Indeed, the spot4502
100
.
80 60
I
40 L
Fig. 6. Densitometric determination of changes in specific mRNAs in response to insulin in primary cultures of rat hepatocytes. Fluorograms illustrated in Fig. 5 were scanned with BioMolecular Dynamics scanner and analyzed using PDQuest software, as described under EXPERIMENTAL PROCEDURES. Changes in intensity of spots are presented as percentage of maximum observed for that spot. Group A cells were harvested on day 5 (condition A), group B cells were harvested on either day 5 (condition B) or day 8 (condition D), andgroup C cells were harvested on day 8 (condition
80
-
60
-
40
-
C).
Condition Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
INSULIN-INDUCED
CHANGES
accumulation of ribosomes is severely impaired in both gastrocnemius and heart from diabetic animals (1). In part, the fall in content of ribosomes in muscle during diabetes is due to an increase in the degradation of ribosomal proteins, and presumably ribosomes, relative to total tissue protein (1). In addition, in mouse myoblasts, insulin stimulates the translation of ribosomal protein mRNAs and the transcription of ribosomal DNA (9). Furthermore, insulin stimulates the expression of other mRNAs in liver (20) and skeletal muscle (14). Therefore, insulin regulates the capacity for protein synthesis through changes in both the amount of particular mRNAs as well as the number of ribosomes. In the present study, the capacity for protein synthesis in primary cultures of isolated hepatocytes was decreased by insulin deprivation. The decrease involved a reduction in both total RNA and poly(A)+ RNA contents. Insulin treatment of insulin-deprived cells resulted in a concomitant increase in both total and poly(A)+ RNA contents. The similarities in response of total RNA, which presumably reflects ribosomal RNA, and poly(A)+ RNA to insulin stimulation and deprivation suggest that similar mechanisms are involved in bringing about the alterations in these two parameters. Similar conclusions were reached in studies examining the effect of insulin on protein synthesis in gastrocnemius (14). The basis for the lack of effect of insulin on the efficiency of protein synthesis in primary cultures of hepatocytes is unknown. However, glucocorticoids have been shown to regulate efficiency in skeletal muscle and heart (23). Therefore, the presence of 0.1 PM dexamethasone in the culture medium in the present study may have masked any effect of insulin on the efficiency of protein synthesis in these cells. The changes that have been observed in the present study in the translational activities of a few mRNAs complement recent studies on insulin-induced transcriptional changes in skeletal (14) and cardiac muscles (2). As in those studies, we have found that insulin both increases and decreases the expression of specific mRNAs. However, in the other studies, insulin treatment of diabetic animals was used as an experimental model. During diabetes, the plasma concentrations of a number of hormones are altered. Thus a number of protein products were altered similarly by both insulin and other hormones, such as 3,5,3’-triiodothyronine, in heart (2). In the present study, isolated hepatocytes were maintained in serum-free medium containing defined concentrations of hormones. Therefore, the specific effect of insulin on mRNA expression could be examined. It should be noted that the use of in vitro translational assays to indirectly quantitate specific changes in gene expression must be interpreted with caution since the expression of some mRNAs is known to be translationally regulated in vivo. However, the expression of a number of proteins, as assessed by translational assay of isolated RNA (10,21,28), has been verified using cDNA probes for the same mRNAs (8, 27). In summary, we have shown that protein synthesis is inhibited in primary cultures of rat hepatocytes deprived of insulin and that insulin treatment of deprived cells
El111
IN RNA IN RAT HEPATOCYTES
completely restores the rate of protein synthesis to control levels. The basis for the inhibition of overall protein synthesis involves a decrease in capacity, with both total and poly(A)+ RNA content reduced by insulin deprivation. Finally, hepatocytes exhibit insulin-induced alterations in the transcriptional activities of specific genes that demonstrate patterns of either repression or increased expression in response to insulin administration. We are grateful to Bonnie Kleinhans and Jill Wolfgang for technical assistance and to Sharon Rannels for assistance in preparing the figures. This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-13499. Current address of C.-J. Hsu: Stranger-Cornell Cancer Research Laboratory, 510 East 73rd St., New York, NY 10021. Address for reprint requests: L. S. Jefferson, Dept. of Cellular and Molecular Physiology, College of Medicine, The Pennsylvania State University, PO Box 850, Hershey, PA 17033. Received 13 April 1992; accepted in final form 13 July 1992. REFERENCES 1. Ashford, A. J., and V. M. Pain. Effect of diabetes on the rates of synthesis and degradation of ribosomes in rat muscle and liver in vivo. J. Biol. Chem. 261: 4059-4065, 1986. 2. Barrieux, A., W. E. Neeley, and W. H. Dillman. Diabetesinduced alterations in the translational activity of specific messenger ribonucleic acids isolated from rat hearts. Circ. Res. 57: 296-303, 1985. 3. Chirgwin, J. Rutter.
J. M.,
A. E. Przybyla,
R. J. MacDonald,
and
W.
Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18: 5294-5299, 1979.
4. Everson, W. V., K. and L. S. Jefferson.
E. Flaim,
D.
M.
Susco,
S. R. Kimball,
Effect of amino acid deprivation on initiation of protein synthesis in rat hepatocytes. Am. J. Physiol. 256 (Cell Physiol. 25): C18-C27, 1989.
5. Flaim, Shiman,
K.
6. Flaim, Jefferson.
K.
E., S. M. Hutson, and L. S. Jefferson.
C. E. Lloyd,
J. M.
Taylor,
R.
Direct effect of insulin on albumin gene expression in primary cultures of rat hepatocytes. Am. J. Physiol. 249 (Endocrinol. Metab. 12): E447-E453, 1985. E.,
D.
E.
Peavy,
W.
E.
Everson,
and
L.
S.
The role synthesis in perfused resulting from amino through perfusion. J.
7.
8.
9.
10. 11.
of amino acids in the regulation of protein rat liver. I. Reduction in rates of synthesis acid deprivation and recovery during flowBiol. Chem. 257: 2932-2938, 1982. Fleck, A., and H. N. Munro. The precision of ultraviolet absorption measurements in the Schmidt-Tannhauser procedure for nucleic acid estimation. Biochim. Biophys. Acta 55: 571-583, 1962. Granner, D., T. Andreone, K. Sasaki, and E. Beale. Inhibition of transcription of the phosphoenolpyruvate carboxykinase gene by insulin. Nature Land. 305: 549-551, 1983. Hammond, M. L., and L. H. Bowman. Insulin stimulates the translation of ribosomal proteins and the transcription of rDNA in mouse myoblasts. J. Biol. Chem. 263: 17785-17791, 1988. Hill, R. E., K. L. Lee, an.d F. T. Kenney. Effects of insulin on mRNA activities in rat liver. J. Biol. Chem. 256: 1510-1513, 1981. Jefferson, L. S., K. E. Flaim, and D. E. Peavy. Protein metabolism. In: Diabetes Mellitus. Theory and Practice, edited by M. Ellenberg and H. Rifkin. New Hyde Park: Medical Examination, 1983, p. 47-59.
12. Jefferson,
L. S., W. S. L. Liao, and J. M. Taylor.
D. E. Peavy,
T. B. Miller,
Diabetes-induced alterations in liver protein synthesis. Changes in the relative abundance of mRNAs for albumin and other plasma proteins. J. Biol. Chem. 258: 1369-1375, 1983. 13. Kelly, F. J., and L. S. Jefferson. Control of peptide-chain initiation in rat skeletal muscle. Development of methods for preparation of native ribosomal subunits and analysis of the effect of insulin on formation of 40 S initiation complexes. J. Biol. Chem. 260: 6677-6683, 1985. 14. Kent, J. D., S. R. Kimball, and L. S. Jefferson. Effect of diabetes and insulin treatment of diabetic rats on total RNA, poly(A)+ RNA, and mRNA in skeletal muscle. Am: J. Physiol. 260 M. C. Appel,
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
El 112 (Cell
INSULIN-INDUCED Physiol.
29): C409-C416,
CHANGES
1991.
Kimball, S. R., and L. S. Jefferson. Effect of diabetes on guanine nucleotide exchange factor activity in skeletal muscle and heart. Biochem. Biophys. Res. Commun. 156: 706-711, 1988. 16. Kimball, S. R., and L. S. Jefferson. Mechanism of inhibition of protein synthesis by vasopressin in rat liver. J. Biol. Chem. 265:
IN
RNA
1990.
Kreamer, B., J. L. Staecker, V. Sawada, G. L. Sattler, M. T. S. Hsia, and H. C. Pitot. Use of a low-speed, iso-density Percoll centrifugation method to increase the viability of isolated rat hepatocyte preparations. J. Cell. Dev. Biol. 22: 201-211, 1986. 18. Labarca, C., and K. Paigen. A simple, rapid, and sensitive DNA assay procedure. Anal. Biochem. 102: 344-352, 1980. 19. Lloyd, C. E., J. E. Kalinyak, S. M. Hutson, and L. S. Jefferson. Stimulation of albumin gene transcription by insulin in primary cultures of rat hepatocytes. Am. J. Physiol. 252 (Cell
17.
Physiol.
21):
C205C214,
1987.
Meisler, M. H., and G. Howard. Effects of insulin on gene transcription. Annu. Rev. Physiol. 51: 701-714, 1989. 21. Palmiter, R. D. Differential rates of initiation of conalbumin and ovalbumin messenger ribonucleic acid in reticulocyte lysate. 20.
J. BioZ. Chem.
249:
6779-6787,
1974.
22.
Penhos, J. C., and M. E. Krahl. Stimulus of leucine incorporation into perfused liver protein by insulin. Am. J. Physiol. 204:
23.
Rannels, S. R., D. E. Rannels, A. Pegg, and L. S. Jefferson. Glucocorticoid effects on peptide-chain initiation in skeletal muscle and heart. Am. J. Physiol. 235 (Endocrinol. Metab. Gas-
140-142,
trointest.
1963.
Physiol.
4): E134-E139,
1978.
HEPATOCYTES
Rosen, J. M., and J. Monahan. Messenger RNA isolation, characterization, and hybridization analysis. In: Laboratory Meth-
24.
15.
16794-16798,
IN RAT
ods Manual for 1987, p. 4-25.
Hormone
Action
and
Molecular
Endocrinology.
25 Safer, B., R. Jagus, and W. M. Kemper. Analysis of initiation factor function in highly fractionated and unfractionated reticulocyte lysate systems. Methods Enzymol. 60: 61-87, 1979. Sambrook, J., E. F. Fritsch, and T. Maniatis. Molecular 26. Cloning: A Laboratory Manual, edited by C. Nolan. New York: Cold Spring Harbor Laboratory, 1989, vol. 2. 27. Sibrowski, W., and H. J. Seitz. Rapid action of insulin and cyclic AMP in the regulation of functional mRNA coding for glucokinase in rat liver. J. BioZ. Chem. 259: 343-346, 1984. Spence, J. T. Levels of translatable mRNA coding for rat liver 28* glucokinase. J. BioZ. Chem. 258: 9143-9146, 1983. 29. Vydelingum, N., R. L. Drake, J. Etienne, and A. H. Kissebah. Insulin regulation of fat cell ribosomes, protein synthesis, and lipoprotein lipase. Am. J. Physiol. 245 (Endocrinol. l
Metab.
8): E121-E131,
1983.
Young, V. R. The role of skeletal muscle and cardiac muscle in the regulation of protein metabolism. In: Mammalian Protein Metabolism, edited by M. N. Munro. New York: Academic, 1970, vol. 4, p. 585-674. J., N. Pritzl, and G. Stab. Quantitation of cardiac 31. Zahringer, polysomal mRNA by hybridization to [3H]poly(U). J. Cell. Dev.
30.
BioZ.
14: 539-550,
1982.
Downloaded from www.physiology.org/journal/ajpendo (116.110.023.093) on August 24, 2019.
