Effects of Partial Hepatectomy on Hepatic Insulinlike Growth Factor Binding Protein- 1 Expression A ~ I GHAHARY, Z GERALD Y. MINUK,JIANGMING Luo, TONYGAUTHIER AND LIAMJ . MURPHY Departments of Internal Medicine and Physiology, Faculty of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada R3E 1R9
Insulinlike growth factor binding proteins modulate the action of the insulinlike growth factors in various bioassays and may regulate the bioavailability of the insulinlike growth factors in uiuo. Because the insulinlike growth factors may influence hepatic regeneration, we have examined the effect of partial hepatectomy on serum insulinlike growth factor binding proteins and on the abundance of insulinlike growth factor binding protein-1 messenger RNA in the liver. All rats were fasted before and after partial hepatectomy or sham operation to avoid the confounding effects of difference in food intake. Using a conventional protocol, 70% of the liver was removed, and groups of four or five rats were killed at different intervals after partial hepatectomy. Sham-operated rats served as controls. Pooled sera from each group of rats were analyzed by ligand blotting with lZsIinsulinlike growth factor-I. Liver RNA from individual rats was analyzed by slot-blot and Northern-blot hybridization. A small decrease in the 39- to 42-kD insulinlike growth factor binding protein was apparent in sera from both the sham-operatedand partial hepatectomized rats. In contrast, a dramatic increase (fivefold)in the 29-kD serum insulinlike growth factor binding protein (insulinlike growth factor binding protein-1) was apparent only in the partial hepatectomized rats. Hepatic insulinlike growth factor binding protein- 1 messenger RNA abundance was significantly increased (1.99 f 0.18-fold;p < 0.05) at 1hr, reached a peak of 2.32 f 0.22-fold (p < 0.01) at 3 hr after partial hepatectomy and returned to basal levels over the subsequent 6 to 12 hr. Interestingly, the abundance of insulinlike growth factor binding protein-1 messenger RNA was also significantly increased in the kidneys of partially hepatectomized rats at 3 hr (1.47 f 0.23-fold; p < 0.05), and peaked at 6 hr (2.50 0.39-fold; p < 0.05) after partial hepatectomy. No significant
*
Received May 7, 1991; accepted January 17, 1992. This research was supported by grants from the Medical Research Council of Canada and the Manitoba Health Research Council. A. Ghahary is a recipient of Manitoba Health Research Council postdoctoral fellowship.G. Y. Minuk is a Manitoba Medical Services Foundation Professor. L. Murphy is a Medical Research Council Scholar. Address reprint requests to: Dr. Gerald Y. Minuk, Director, Liver Diseases Unit, GF407, Health Sciences Centre, 820 Sherbrook Street, Winnipeg, Manitoba, Canada R3A 1R9. 31/1/36767
change in hepatic or renal insulinlike growth factor binding protein-1 messenger RNA abundance was a p parent in sham-operated rats. For comparison, the abundance of the growth hormone-dependentbinding protein insulinlike growth factor binding protein-3 and insulinlike growth factor-I messenger RNAs was also examined. A decrease in both hepatic insulinlike growth factor binding protein-3 and insulinlike growth factor-1 messenger RNA was observed in shamoperated and hepatectomized rats. These data demonstrate that partial hepatectomy is associated with increased expression of insulinlike growth factor binding protein-1 in both the liver and kidney. The increase in tissue and circulating insulinlike growth factor binding protein-1 may modulate the mitogenic effects of insulinlike growth factor-I during liver re1992;15:1125-1131.) generation. (HEPATOLOGY
Insulinlike growth factors I and I1 (IGF-I and 11) are mitogenic peptides that play an important role in growth and development. IGF-I is expressed in most if not all tissues (1-3). In the rodent, IGF-I1 appears t o be important in fetal growth but has little growthstimulating activity in the postnatal animal (4).In contrast, IGF-I can stimulate growth in the growth hormondeficient rodent (4,5). Although IGF-1 messenger RNA (mRNA)is expressed in many other tissues, the liver is the main site of IGF-1 biosynthesis and circulating IGF-1 (6, 7). The changes in serum IGF-I levels after partial hepatectomy (PHx) are variable (8-10). Unterman and Phillips (9) found that somatomedin activity measured in a cartilage bioassay fell to approximately 50% of controls after PHx. However, when appropriate account was taken for the reduced food intake, serum from PHx rats had higher somatomedin activity than did pair-fed, food-restricted, sham-operated controls (9). They interpreted these data to indicate that regenerating liver may produce more somatomedins or fewer somatomedin inhibitors than normal liver tissue. In contrast, Russell, D’Ercole and Underwood (10) found that serum IGF-I, as measured by RIA in unextracted rat serum, was significantly higher in the first few hours after hepatectomy and fell over the subsequent 12 to 24 hr. In the latter study, serum IGF-I levels were indistinguishable in PHx and pair-fed, sham-operated rats 22 hr postop-
1125
1126
GHAHARY ET AL.
HEPATOLOGY
TABLE 1. The effect of PHx and sham operation on serum glucose concentrations Timehours Group
0.5
Control ratsa Sham-operated PHx
4.2 t 0.6 6.0 ? 0.9 5.8 ? 0.7
1
1.1 +. 0.2 2.6 t 0.4'
6.6 4.6
? ?
12
6
3
3.0 2.9
0.8 1.4'
___________
? ?
1.1 0.5
2.9 2.8
? ?
24
1.1 2.2
5.8 z 0.8 3.6 ? 0.6'
~~~
Data represent the mean ? S.E.M. Serum glucose is expressed as mmol/L for 5 rats/group. "Control group of age-matched rats fasted for 16 hr. 'The significant difference between the sham and Phx rats, p < 0.01. 'The significant difference between the sham and PHx rats, p < 0.05.
eration. However, the IGF-I concentrations in hepatic extracts from PHx rats were significantly lower than those seen in sham-operated rats (10). Serum IGF-I1 concentrations do not appear t o be reduced after PHx, although an increase in mannose-6-phosphate/IGF-II receptors in liver membranes after P H x has been reported (11). The IGFs are present in serum and other biological fluids in association with binding proteins (12-14). In both Unterman and Phillip's (9) study and that of Russell, D'Ercole and Underwood (lo), changes in IGF binding proteins after PHx may have influenced the results obtained. The IGF binding proteins have been reported t o inhibit IGF-I action in a variety of bioassays, including the somatomedin cartilage bioassay (15-17) and in most laboratories, serum binding proteins interfere with IGF-I RIA determinations i n unextracted rat serum. At least four distinct insulinlike growth factor binding proteins (IGFBPs)have been isolated, characterized a n d cloned (18-21). Several lines of evidence suggest that there may be additional members of this gene family (19-22). T h e binding proteins may regulate the bioavailability of t h e IGFs and therefore modulate their growthpromoting and metabolic actions of the IGFs. Of the four IGFBPs cloned so far, all appear t o be highly expressed in the liver (18-21). In the rodent, however, the abundance of IGFBP-2 mRNA declines rapidly in the liver after birth (20). In this study, we examined the effects of PHx on serum IGFBPs and, using a rat IGFBP-1 complementary DNA (cDNA) quantified the changes i n hepatic and renal IGFBP-1 expression after PHx in the rat.
MATERIALS AND METHODS Surgical Procedure and Tissue Preparation. Adult male SpragueDawley rats weighing 250 to 300 gm were obtained from the University of Manitoba breeding facility (Winnipeg, Manitoba, Canada). All animals were housed under conditions of controlledtemperature and light with free access to food and water. Both PHx and sham-operated rats were fasted from 3 PM on the day before surgery. The operative procedures were performed in the morning between 8 and 10 AM with rats under light ether anesthesia. For PHx, the median and left lateral lobes (approximately 70% of the liver mass) were excised according to the method of Higgins and Anderson (24). This tissue was used as the zero time control liver tissue. In the sham operations, the appropriate portions of the liver were
exteriorized for the same length of time as rats undergoing PHx. Groups of four to five rats were then killed at different intervals after PHx. Because no kidney tissue was collected from the time 0 rats, the changes in renal IGFBP-1 mRNA abundance are expressed in terms of the 0.5 hr sham-operated controls. Both sham and PHx rats were allowed free access to water containing 20% sucrose immediately on recovery from anesthesia but were deprived of other sources of food for the entire duration of the experiment. Trunk blood from each rat was collected and kept at 4" C for 2 to 3 hr before serum separation. Sera were then stored at - 20" C. Because no blood was collected from the rats at the time of operation, blood was collected from a comparable group of age-matched, fooddeprived rats. This blood was used to prepare time 0 control sera for ligand binding and serum glucose determinations. The livers and kidneys of individual rats were rapidly removed and stored at - 70" C. Determination of Serum Glucose and ZGFBPs. Serum glucose concentrations were measured with a glucose oxidase assay system (Sigma Chemical Co., St. Louis, MO). Serum IGFBPs were detected by the ligand binding technique according to the method of Hossenlopp et al. (25). Three microliters of pooled sera from four to five rats at each time point were subjected to electrophoresis on a 12% SDSpolyacrylamide gel under nonreducing conditions. Separated proteins were then transferred to nitrocellulose membranes with Towbin buffer (15 mmol/L Tris, 120 mmol/L glycine, [pH 8.31 and 20% methanol). Filters were dried at 37" C for 5 min, and nonspecific binding was blocked by successive incubation of nitrocellulose membranes in 3%NP40 for 30 min, 1%BSA for 2 hr and 0.1% Tween 20 in ice-cold Tris-saline solution (0.15 moVL NaC1, 0.01 m o m Tris-HC1, 0.5 mg/ml sodium azide [pH 7.41) for 10 min. Nitrocellulose filters were then incubated overnight at 4" C with approximately lo6 cpm of 1251-labeledIGF-1 (Amersham Canada, Ltd., Oakville, Ontario, Canada) and then washed twice with ice-cold Tris-saline solution containing 0.1% Tween 20 for 15 min each and once in Tris-saline solution alone for 15 min. Filters were then air dried, and radioactivity was visualized by autoradiography. RNA Extraction and Hybridization. Total RNA from each individual rat liver or kidney tissue was extracted by the guanidinium isothiocyanate/cesium chloride technique (26). RNA was analyzed by electrophoresis in a 1%agarose and 2.2 mol/L formaldehydegel and transferred to nitrocellulose filter. In addition, serial dilutions of total RNA from each individual rat were also analyzed by slot-blot hybridization. Filters were then prehybridized for 4 to 6 hr at 42" C in a solution containing 50% formamide, 20 mmoVL NaH,PO, (pH 7), 4 x standard saline citrate ( l x = 0.15 mol/L NaC1-0.015 mol/L sodium citrate), 2 mmol/L EDTA, 4 x Denhardt's solution (1x = 0.02% BSA, Ficoll and polyvinylpyrolidine), 1% SDS and 100 Fg/ml sonicated denatured salmon sperm
Vol. 15, No. 6, 1992
1127
IGFBP-1 EXPRESSION AFTER PARTIAL HEPATECTOMY
DNA. Hybridization was performed at 42" C in the same solution for 16 to 20 hr with a rat IGFBP-1 cDNA (271, a rat IGF-I cDNA (28) or an oligonucleotide with a sequence corresponding to the complement of nucleotides 54 to 114 of the published sequences for rat IGFBP-3 (29).The cDNA probes were labeled by nick translation, whereas the oligonucleotide probe was radiolabeled with T4 polynucleotide kinase. Filters were initially washed at room temperature with 2 x standard saline citrate and 0.1% x SDS, and finally washed for 20 min at 65" C in 0.1 x standard saline citrate and 0.1% SDS, except in the case of the IGFBP-3 oligonucleotide probe where the final wash was in 1 x standard saline citrate and 0.1% SDS at 45" C for 15 min. Statistical Analysis. Dunnett's t test was used to determine the statistical significance of difference between control and treatment groups.
RESULTS In both the sham-operated rats and the PHx rats, there was biphasic change in serum glucose. A marked drop in serum glucose was observed 1h r after surgery. This decline in serum glucose was more marked in the sham-operated rats than in the PHx rats (Table 1). At least five IGFBPs were detected in rat sera from both sham-operated and hepatectomized rats (Fig. 1). The larger binding proteins with molecular mass of 39 to 42 kD most probably represent IGFBP-3. A small decline in the abundance of this binding protein from time 0 was observed in both sham-operated and hepatectomized rats. In contrast, the 29-kD IGFBP-1 was increased in sera from PHx rats compared with sham-operated rats at 3 hr and remained elevated up to 6 hr after the operation. The expression of hepatic IGFBP-1 mRNA was markedly increased within 1 hr, peaked at 3 hr and returned to basal values at 6 to 12 hr after PHx (Fig. 2). Ethidium bromidestained 18s and 28s ribosomal RNA shown in the lower panels of Fig. 2 indicate that the apparent increase in hepatic IGFBP-1 mRNA expression after PHx was not due to gel overloading. Only a small increase in the IGFBP-1 mRNA level was observed after the sham operation. The abundance of hepatic IGFBP-1 mRNA expression was also determined by slot-blot hybridization with serial dilutions of hepatic total RNA from sham-operated or hepatectomized rats. The intensity of the hybridization signal was quantitated by densitometry and expressed in terms of the time 0 control. As shown in Figure 3, hepatic IGFBP-1 mRNA was significantly increased (1.99 k 0.18-fold; p < 0.05) at 1 hr and reached a peak (2.32 f 0.22-fold; p < 0.01) 3 hr after PHx. The abundance of IGFBP-1 mRNA did not increase significantly in sham-operated rats (Fig. 3, open bars). Of interest, the abundance of kidney IGFBP-1 mRNA was also significantly increased (1.47 k 0.23-fold; n = 5 ; p < 0.05) after 3 hr and peaked (2.5 k 0.39; n = 5; p < 0.01) 6 h r after PHx (Fig. 4).No significant change in renal IGFBP-1 mRNA levels was seen in the sham-operated rats. In contrast to the increase in hepatic IGFBP-1 expression, a slight decrease in hepatic IGFBP-3 mRNA abundance was apparent in both the sham-operated and
A
29KDa-
29KDag
0
1
3
6
12
24
TIME(hr)
500
-
400 -
300
-
200-
fa
100-
w
v)
TIME ( hr ) FIG.1.Effect of PHx on serum IGFBPs. Pooled serum samples were analyzed by the ligand-blotting technique using lZ5I-IGF-Ias a tracer. (A) Serum from sham-operated control rats was analyzed, whereas the data obtained for serum from hepatectomized rats are shown in (B). Changes in abundance of the 29-kD binding protein, presumed to be IGFBP-1, were quantified by densitometry and are plotted in (C). Solid symbols show the results for serum from PHx rats, whereas open symbols represent the sham-operated controls.
the PHx rats (Fig. 5 ) . Similarly, a decrease in hepatic IGF-I mRNA abundance was observed in both groups of rats (Fig. 6 ) . When hepatic IGF-I mRNA was quantitated by dot-blot hybridization, it was apparent that the decrease in IGF-I mRNA was more marked in the sham-operated rats than in the PHx rats (Fig. 7). DISCUSSION
Liver cell mass is carefully regulated, and partial resection is accompanied by a cycle of hepatocyte proliferation such that the liver mass in the rat is rapidly restored to near normal within 48 to 72 hr. Although the
1128
GHAHARY ET AL.
HEPATOLOGY
6
A
1.6kb
I G FB P-1
28s
18s 0
0.5
3
1
6
12
0 0.5
24
1
3
6
12 24
T I M E(hr) FIG. 2. Effect of PHx on hepatic IGFBP-1 mRNA abundance. Total RNA (60 pg/lane) pooled from four to five rats killed at various times after either PHx (A) or sham operation (B) was electrophoresed and transferred to nitrocellulose filters. The upper panel shows the pattern of hybridization obtained with a rat IGFBP-1 cDNA, whereas the lower panel depicts the ethidium bromide-stained gel. The size of IGFBP-1 transcript was determined by comparison with the position of the 18s and 28s ribosomal RNAs.
** *
z
0 m m W
2-
a a X
11I T
W
W
zF 4W a
1
0l / ’ I 0
3
6
0.5
12
24
TIME ( hr )
FIG.3. Relative changes in hepatic IGFBP-1mRNA abundance after sham operation and PHx. Serial dilutions of total RNA from individual rat liver samples a t various time points after either sham operation (open histograms) or PHx (solid histograms) were analyzed by slot-blot hybridization. Data were quantitated by densitometery, and slope of the line of best fit for each RNA sample was expressed in term of that of the time 0 control, which was an arbitrary attributed value of 1. Data represent the means i- S.E.M. for n = 4 or 5igroup. The significant difference between the control, time 0 group and other groups are depicted as * and ** for p < 0.05 and p < 0.01, respectively.
liver is the major site of IGF biosynthesis, and the IGFs are mitogenic for a variety of cell types, the role of the IGFs in hepatic regeneration is not clear. In this study, we have shown that there is an increase in hepatic IGFBP-1 mRNA abundance after PHx. A parallel change in the serum 29-kD IGFBP was also observed. This IGFBP presumably represents IGFBP-1. Although the effects of PHx on serum IGF-I concentrations have
been reported previously, to our knowledge this study is the first report where effects of PHx on IGFBP-1 expression have been examined. Hepatic IGFBP-1mRNA expression was significantly increased within 1hr after operation and peaked by 3 hr. In comparison, the increase in kidney IGFBP-1 mRNA expression was slightly delayed, with peak levels achieved 6 hr after PHx. In several species, including the rat (27, 30, 31), food deprivation results in enhanced IGFBP-1 expression. This cannot be the only explanation for the effect of PHx on IGFBP-1 expression, because both sham and PHx rats were food-deprived prior to and following the operative procedures. The only caloric source was sucrose-supplemented water, and water consumption was similar in both groups of animals. Furthermore, hypoglycemia was even more marked in the sham-operated rats than in the PHx rats. Whereas hypoglycemia does not appear to be the direct cause of the enhanced IGFBP-1 expression in the PHx rats, it is probable that the stress response to PHx is considerably greater than sham operation. We have previously demonstrated that glucocorticoids, growth hormone and glucagon, in addition to insulin, can regulate expression of IGFBP-1 (32-34). Thus disturbances in these and other hormones after PHx may account for the enhanced IGFBP-1 expression. The increase in renal IGFBP-1 expression after PHx was an unexpected finding, but supports the notion that perhaps local and/or humoral factors are involved in the enhanced IGFBP-1 expression during liver regeneration. The role of IGFBP-1 remains controversial. It appears to be able to both inhibit and enhance the action of IGF-I in a variety of bioassays (15,351.The different effects of IGFBP-1 on IGF-I action may result from different postranslational modifications such as phosphorylation
Vol. 15, No. 6, 1992
1129
IGFBP-1 EXPRESSION AFTER PARTIAL HEPATECTOMY
B
A
I G FB P-1
1*6kb
28s
18s 0.5
1
3
6
12
24
0.5
3
1
6
12
24
TIME( h r ) FIG.4. Effect of PHx on renal IGFBP-1 mRNA abundance. Total RNA (60 p,g/lane) pooled from four to five rats killed a t various times after either PHx (A) or sham operation (B) was electrophoresed and transferred to nitrocellulose filters. The upper panel shows the pattern of hybridization obtained with a rat IGFBP-1 cDNA, whereas the lower panel depicts the ethidium bromide-stained gel. The size of IGFBP-1 transcript was determined by comparison with the position of the 18s and 28s ribosomal RNAs.
IGFBP-3
IGFBP-1 0 0.5 1 3 6 SHAM
12 24 0.5
1
3
6 12 24 PH
FIG.5. Comparison of the effect of PHx on hepatic IGFBP-1 and IGFBP-3 mRNA abundance. Total RNA (50 @/lane) pooled from four to five rats killed a t various times after either PHx or sham operation was electrophoresed and transferred to nitrocellulose filters. The upperpanel shows the pattern of hybridization obtained with a rat IGFBP-3 oligonucleotide probe, whereas the lower panel depicts the hybridization signal obtained with the IGFBP-1 cDNA.
that have been documented (36). The state of the IGFBP-1 present in sera from PHx rats and its functional effect on IGF-1 action are not known. Although serum somatomedin bioactivity declines after PHx (9), this may simply reflect the decline in circulating IGF-I levels rather than increased levels of IGFBP-1 or other somatomedin inhibitors. Several studies have demonstrated that circulating IGF-I levels fall after PHx (8-11); however, it is unclear whether this decline can be explained by loss of liver mass, changes in growth hormone or other hormones, hepatocellular injury or reduced food consumption. In this regard, the experiments reported here provide some
additional information. Our data demonstrate that there is a decrease in IGF-I mRNA levels in the remnant hepatic tissue after PHx. Interestingly, the decline in IGF-I mRNA levels was more marked in the shamoperated rats than the PHx rats. In a study reported by Russell, D’Ercole and Underwood (101, where serum IGF-I was measured by RIA, serum IGF-I levels were higher in PHx rats than in fasted control rats. Based on our findings, the increase in serum IGF-I levels after PHx demonstrated by Russell, D’Ercole and Underwood (lo) are unlikely to have resulted from enhanced IGF-I synthesis, but may have been caused by the effects of increased circulating IGFBP-1 concentrations on serum
1130
HEPATOLOGY
GHAHARY ET AL.
* 28%
* 18~+
0
0.5
1
3
6
12
24
-
0-5
1
3
SHAM
6
12
24
PH
-
T I M E AFTER OPERATION hours FIG. 6. Effect of PHx on hepatic IGF-1 mRNA abundance. Total RNA (60 kg/lane) pooled from four to five rats killed a t various times after either sham operation or PHx were electrophoresed and transferred to nitrocellulose filters. The upper panel shows the pattern of hybridization obtained with a rat IGF-1 cDNA, whereas the lower panel depicts the ethidium bromide-stained gel.
2
120 100
40
20
0
I
0.5
IGF-I levels or IGFBP-1 interference in the unextracted IGF-I RIA used by these authors. The decline in hepatic IGF-I and in IGFBP-3 mRNA expression reported here is likely to be caused at least in part by caloric restriction and the reduced pituitary growth hormone secretion that occurs in fasted rats (37). In conclusion, the results of this study clearly demonstrate that IGFBP-1 expression is up-regulated after PHx. Furthermore, this response to PHx appears to involve factors other than food deprivation. The physiological significance of the increased circulating levels of IGFBP-1 following hepatic resection remains to be determined. ADDENDUM
1
3
6
12
24
-
TIME hours
FIG.7. The effect of PHx on hepatic IGF-1 mRNA abundance. Serial dilutions of total RNA from each individual rat liver samples at various time points after either sham operation (open histograms) or PHx (solid histograms) were analyzed by slot-blot hybridization. The data were quantitated by densitometery, and slope of the line of best fit for each RNA sample was expressed in terms of that of the time 0 control, which was an arbitrary attributed value of 100. Data represent the means +- S.E.M. for n = 4 or 5/group. The significant difference between the control, time 0 control group and other groups are depicted as * and ** for p < 0.05 and p < 0.01, respectively.
After submission of this manuscript, a similar observation was reported by Mohn et al. (381, who noted a more prompt and pronounced increase in IGFBP abundance in the regenerating liver after hepatectomy. REFERENCES 1. Murphy U, Bell GI, Friesen HG. Tissue distribution of insulinlike growth factor I and I1 messenger ribonucleic acid in the adult rat. Endocrinology 1987;120:1279-1282. 2. Lund PK, Moats-Staats BM, Hynes MA, Simmons JD, Jansen M, D’Ercole AJ, Van Wyk J J . Somatomedidinsulinlike growth factor-I and insulinlike growth factor-I1 mRNAs in rat fetal and adult tissues. J Biol Chem 1986;261:14539-14544.
Vol. 15, No. 6, 1992
IGFBP-1 EXPRESSION AFTER PARTIAL HEPATECTOMY
3. Lowe WL, Roberts CT, Lasky SR, LeRoith D. Differential expression of alternative 5’ untranslated regions in mRNAs encoding rat insulinlike growth factor I. Proc Natl Acad Sci USA 1987;84:8946-8950. 4. Schoenle E, Zapf J , Hauri C, Steiner T, Froesch ER. Comparison of in vivo effects of insulinlike growth factor-I and I1 and growth hormone in hypophysectomized rats. Acta Endocrinol 1985;108: 167-174. 5. Schoenle E, Zapf J , Humbel RE, Froesch ER. Insulinlike growth factor I stimulates growth in hypophysectomized rats. Nature 1985;296:252-253. 6. McConaghey P, Sledge CB. Production of sulfation factor by the perfused rat liver. Nature 1970;225:1249-1250. 7. D’Ercole AJ, Stiles AD, Underwood LE. Tissue concentration of somatomedin C: further evidence for multiple sites of synthesis and paracrine or autocrine mechanism of action. Proc Natl Acad Sci USA 1984;81:935-939. 8. Uthne K, Uthne T. Influence of liver resection and regeneration on somatomedin (sulphation factor) activity in sera from normal and hypophysectomized rats. Acta Endocrinol 1972;71:255-264. 9. Unterman TG, Phillips LS. Circulating somatomedin activity during hepatic regeneration. Endocrinology 1986;119:185-192. 10. Russell WE, D’Ercole AJ, Underwood LE. Somatomedin C/insulinlike growth factor I during liver regeneration in the rat. Am J Physiol 1985;248:E618-E623. 11. Scott CD, Baxter RC. Insulinlike growth factor/mannose-6phosphate receptors are increased in hepatocytes from regenerating rat liver. Endocrinology 1990;126:2543-2549. 12. Zapf J , Waldvogel M, Froesch ER. Binding of non-suppressible insulinlike activity to human serum. Arch Biochem Biophys 1975;168:638-645. 13. Romanus JA, Terrell JE, Yang YWH, Nessley SP, Rechler MM. Insulinlike growth factor carrier proteins in neonatal and adult rat serum are immunologically different: demonstration using a new radioimmunoassay for the carrier protein from BRL-3A rat liver cells. Endocrinology 1986;118:1743-1758. 14. Martin JL, Baxter RC. Insulinlike growth factor binding proteins from human plasma. J Biol Chem 1986;261:8754-8760. 15. Burch WM,Correa J , Shively JE, Powell DR. The 25 kilodalton insulinlike growth factor (1GF)-bindingprotein inhibits both basal and IGF-I mediated growth of chick embryo pelvic cartilage in vitro. J Clin Endocrinol Metab 1990;70:173-180. 16. Meuli C, Zapf J, Froesch ER. NSILA-carrier protein abolishes the action of nonsuppressible insulinlike activity (NSILA-S) on perfused rat heart. Diabetologia 1978;14:255-259. 17. Knauer DJ, Smith GL. Inhibition of biological activity of multiplication-stimulating activity by binding to its carrier protein. Proc Natl Acad Sci USA 1980;77:7252-7256. 18. Brinkman A, Groffen C, Kortleve DJ, Geurts van Kessel A, Drop SLS. Isolation and characterization of a cDNA encoding the low molecular weight insulinlike growth factor binding protein (IBP-1) EMBO J 1988;7:2417-2423. 19. Wood WI, Cachianes G, Henzel WJ, Winslow GA, Spencer SA, Hellmiss R, Martin JL, Baxter RC. Cloning and expression of the growth hormone-dependent insulinlike growth factor-binding protein. Mol Endocrinol 1988;2:1176-1185. 20. Brown AL, Chiariotti L, Orlowski CC, Mehlem T, Burgess WH, Ackerman EJ, Bruni CB, Rechler MM. Nucleotide sequence and expression of a cDNA clone encoding a fetal rat binding protein for insulinlike growth factors. J Biol Chem 1989;264:5148-5154. 21. Shimasaki S, Uchiyama F, Shimonaka M, Ling N. Molecular cloning of the cDNAs encoding a novel insulinlike growth factor-binding protein from rat and human. Mol Endocrinol 1990;4:1451-1458. 22. Roghani M, Hossenlopp P, Lepage P, Balland A, Binoux M. Isolation from human cerbrospinal fluid of a new insulinlike
1131
growth factor-binding protein with a selective affinity for IGF-11. FEBS Lett 1989;255:253-258. 23. Zapf J , Kiefer M, Merryweather J, Masiarz F, Bauer D, Born W, Fischer JA, Froesch ER. Isolation from adult human serum of four insulinlike growth factor (IGF) binding proteins and molecular cloning of one of them that is increased by IGF-I administration and in extrapancreatic tumor hypoglycemia. J Biol Chem 1990; 265: 14892-14898. 24. Higgins GM, Anderson RM. Experimental pathology of liver: restoration of liver of the white rat following partial surgical removal. Arch Pathol 1931;12:186-202. 25. Hossenlopp P, Seurin D, Segovia-Quinson B, Hardouin S, Binoux M. Analysis of serum insulinlike growth factor binding proteins using western blotting: use of the method for titration of the binding proteins and competitive binding studies. Anal Biochem 1986;154:138-143. 26. Chirgwin J, Pryzbyla A, MacDonald R, Rutter W. Isolation of biologically active ribonucleic acid from sources enriched for ribonuclease. Biochemistry 1979;18:5294-5599. 27. Murphy W, Senevirantne C, Ballejo G, Croze F, Kennedy TG. Identification and characterization of a rat decidual insulinlike growth factor-binding protein complementary DNA. Mol Endocrinol 1990;4:329-336. 28. Murphy LJ, Bell GI, Duckworth ML, Friesen HG. Identification, characterization and regulation of a rat cDNA which encodes insulinlike growth factor-I. Endocrinology 1987;121:684-691. 29. Shimasaki S, Koba A, Mercado M, Shimonaka M, Ling N. Complementary DNA structure of the high molecular weight rat insulinlike growth factor binding protein (IGFBP 3) and tissue distribution of its mRNA. Biochem Biophys Res Commun 1989; 165:907-912. 30. Busby WH, Snyder DK, Clemmons DR. Radioimmunoassay of a 26,000-dalton plasma insulinlike growth factor-binding protein: control by nutritional variables. J Clin Endocrinol Metab 1988; 67:1225-1230. 31. McCusker RH, Campion DR, Jones WK, Clemmons DR. The insulinlike growth factor-binding proteins of porcine serum: endocrine and nutritional regulation. Endocrinology 1989;125: 501-509. 32. Luo J , Reid RE, Murphy W. Dexamethasone increases hepatic insulinlike growth factor binding protein-1 (IGFBP-1) mRNA and serum IGFBP-1 concentration in the rat. Endocrinology 1990; 127:1456-1462. 33. Seneviratne C, Luo JM,Murphy W. Transcriptional regulation of rat insulinlike growth factor binding protein-1 by growth hormone. Molec Endocrinol 1990;4:1199-1204. 34. Kacchra Z, Yannopoulos C, Barash I, Guyda HJ, Murphy U, Posner BI. The differential regulation by glucagon and growth hormone of IGF-I and IGF binding proteins (IGF-BPs) in cultured rat hepatocytes [Abstract 11331. 72th Annual Meeting of the Endocrine Society, Atlanta, GA 1990. 35. Elgin RG, Busby WH, Clemmons DR. An insulinlike growth factor (IGF) binding protein enhances the biological response to IGF-I. Proc Natl Acad Sci USA 1987;84:3254-3258. 36. Jones JI, Clemmons DR. The bioactivity of an insulinlike growth factor binding protein is modulated by serine phosphylation [Abstract]. Clin Res 1990;38:978A. 37. Tannenbaum GS, Epelbaum J , Colle E, Brazeau P, Martin JB. Antiserum to somatostatin reverses starvation-induced inhibition of growth hormone but not insulin secretion. Endocrinology 1978;102:1909-1914. 38. Mohn KL, Melby AE, Tewari DS, Laz TM, Taub R. The gene encoding rat insulinlike growth factor-binding protein 1is rapidly and highly induced on regenerating liver. Mol Cell Biol 1 9 9 l ; l l : 1393-1401.
Insulinlike growth factor binding proteins modulate the action of the insulinlike growth factors in various bioassays and may regulate the bioavailability of the insulinlike growth factors in uiuo. Because the insulinlike growth factors may influence hepatic regeneration, we have examined the effect of partial hepatectomy on serum insulinlike growth factor binding proteins and on the abundance of insulinlike growth factor binding protein-1 messenger RNA in the liver. All rats were fasted before and after partial hepatectomy or sham operation to avoid the confounding effects of difference in food intake. Using a conventional protocol, 70% of the liver was removed, and groups of four or five rats were killed at different intervals after partial hepatectomy. Sham-operated rats served as controls. Pooled sera from each group of rats were analyzed by ligand blotting with lZsIinsulinlike growth factor-I. Liver RNA from individual rats was analyzed by slot-blot and Northern-blot hybridization. A small decrease in the 39- to 42-kD insulinlike growth factor binding protein was apparent in sera from both the sham-operatedand partial hepatectomized rats. In contrast, a dramatic increase (fivefold)in the 29-kD serum insulinlike growth factor binding protein (insulinlike growth factor binding protein-1) was apparent only in the partial hepatectomized rats. Hepatic insulinlike growth factor binding protein- 1 messenger RNA abundance was significantly increased (1.99 f 0.18-fold;p < 0.05) at 1hr, reached a peak of 2.32 f 0.22-fold (p < 0.01) at 3 hr after partial hepatectomy and returned to basal levels over the subsequent 6 to 12 hr. Interestingly, the abundance of insulinlike growth factor binding protein-1 messenger RNA was also significantly increased in the kidneys of partially hepatectomized rats at 3 hr (1.47 f 0.23-fold; p < 0.05), and peaked at 6 hr (2.50 0.39-fold; p < 0.05) after partial hepatectomy. No significant
*
Received May 7, 1991; accepted January 17, 1992. This research was supported by grants from the Medical Research Council of Canada and the Manitoba Health Research Council. A. Ghahary is a recipient of Manitoba Health Research Council postdoctoral fellowship.G. Y. Minuk is a Manitoba Medical Services Foundation Professor. L. Murphy is a Medical Research Council Scholar. Address reprint requests to: Dr. Gerald Y. Minuk, Director, Liver Diseases Unit, GF407, Health Sciences Centre, 820 Sherbrook Street, Winnipeg, Manitoba, Canada R3A 1R9. 31/1/36767
change in hepatic or renal insulinlike growth factor binding protein-1 messenger RNA abundance was a p parent in sham-operated rats. For comparison, the abundance of the growth hormone-dependentbinding protein insulinlike growth factor binding protein-3 and insulinlike growth factor-I messenger RNAs was also examined. A decrease in both hepatic insulinlike growth factor binding protein-3 and insulinlike growth factor-1 messenger RNA was observed in shamoperated and hepatectomized rats. These data demonstrate that partial hepatectomy is associated with increased expression of insulinlike growth factor binding protein-1 in both the liver and kidney. The increase in tissue and circulating insulinlike growth factor binding protein-1 may modulate the mitogenic effects of insulinlike growth factor-I during liver re1992;15:1125-1131.) generation. (HEPATOLOGY
Insulinlike growth factors I and I1 (IGF-I and 11) are mitogenic peptides that play an important role in growth and development. IGF-I is expressed in most if not all tissues (1-3). In the rodent, IGF-I1 appears t o be important in fetal growth but has little growthstimulating activity in the postnatal animal (4).In contrast, IGF-I can stimulate growth in the growth hormondeficient rodent (4,5). Although IGF-1 messenger RNA (mRNA)is expressed in many other tissues, the liver is the main site of IGF-1 biosynthesis and circulating IGF-1 (6, 7). The changes in serum IGF-I levels after partial hepatectomy (PHx) are variable (8-10). Unterman and Phillips (9) found that somatomedin activity measured in a cartilage bioassay fell to approximately 50% of controls after PHx. However, when appropriate account was taken for the reduced food intake, serum from PHx rats had higher somatomedin activity than did pair-fed, food-restricted, sham-operated controls (9). They interpreted these data to indicate that regenerating liver may produce more somatomedins or fewer somatomedin inhibitors than normal liver tissue. In contrast, Russell, D’Ercole and Underwood (10) found that serum IGF-I, as measured by RIA in unextracted rat serum, was significantly higher in the first few hours after hepatectomy and fell over the subsequent 12 to 24 hr. In the latter study, serum IGF-I levels were indistinguishable in PHx and pair-fed, sham-operated rats 22 hr postop-
1125
1126
GHAHARY ET AL.
HEPATOLOGY
TABLE 1. The effect of PHx and sham operation on serum glucose concentrations Timehours Group
0.5
Control ratsa Sham-operated PHx
4.2 t 0.6 6.0 ? 0.9 5.8 ? 0.7
1
1.1 +. 0.2 2.6 t 0.4'
6.6 4.6
? ?
12
6
3
3.0 2.9
0.8 1.4'
___________
? ?
1.1 0.5
2.9 2.8
? ?
24
1.1 2.2
5.8 z 0.8 3.6 ? 0.6'
~~~
Data represent the mean ? S.E.M. Serum glucose is expressed as mmol/L for 5 rats/group. "Control group of age-matched rats fasted for 16 hr. 'The significant difference between the sham and Phx rats, p < 0.01. 'The significant difference between the sham and PHx rats, p < 0.05.
eration. However, the IGF-I concentrations in hepatic extracts from PHx rats were significantly lower than those seen in sham-operated rats (10). Serum IGF-I1 concentrations do not appear t o be reduced after PHx, although an increase in mannose-6-phosphate/IGF-II receptors in liver membranes after P H x has been reported (11). The IGFs are present in serum and other biological fluids in association with binding proteins (12-14). In both Unterman and Phillip's (9) study and that of Russell, D'Ercole and Underwood (lo), changes in IGF binding proteins after PHx may have influenced the results obtained. The IGF binding proteins have been reported t o inhibit IGF-I action in a variety of bioassays, including the somatomedin cartilage bioassay (15-17) and in most laboratories, serum binding proteins interfere with IGF-I RIA determinations i n unextracted rat serum. At least four distinct insulinlike growth factor binding proteins (IGFBPs)have been isolated, characterized a n d cloned (18-21). Several lines of evidence suggest that there may be additional members of this gene family (19-22). T h e binding proteins may regulate the bioavailability of t h e IGFs and therefore modulate their growthpromoting and metabolic actions of the IGFs. Of the four IGFBPs cloned so far, all appear t o be highly expressed in the liver (18-21). In the rodent, however, the abundance of IGFBP-2 mRNA declines rapidly in the liver after birth (20). In this study, we examined the effects of PHx on serum IGFBPs and, using a rat IGFBP-1 complementary DNA (cDNA) quantified the changes i n hepatic and renal IGFBP-1 expression after PHx in the rat.
MATERIALS AND METHODS Surgical Procedure and Tissue Preparation. Adult male SpragueDawley rats weighing 250 to 300 gm were obtained from the University of Manitoba breeding facility (Winnipeg, Manitoba, Canada). All animals were housed under conditions of controlledtemperature and light with free access to food and water. Both PHx and sham-operated rats were fasted from 3 PM on the day before surgery. The operative procedures were performed in the morning between 8 and 10 AM with rats under light ether anesthesia. For PHx, the median and left lateral lobes (approximately 70% of the liver mass) were excised according to the method of Higgins and Anderson (24). This tissue was used as the zero time control liver tissue. In the sham operations, the appropriate portions of the liver were
exteriorized for the same length of time as rats undergoing PHx. Groups of four to five rats were then killed at different intervals after PHx. Because no kidney tissue was collected from the time 0 rats, the changes in renal IGFBP-1 mRNA abundance are expressed in terms of the 0.5 hr sham-operated controls. Both sham and PHx rats were allowed free access to water containing 20% sucrose immediately on recovery from anesthesia but were deprived of other sources of food for the entire duration of the experiment. Trunk blood from each rat was collected and kept at 4" C for 2 to 3 hr before serum separation. Sera were then stored at - 20" C. Because no blood was collected from the rats at the time of operation, blood was collected from a comparable group of age-matched, fooddeprived rats. This blood was used to prepare time 0 control sera for ligand binding and serum glucose determinations. The livers and kidneys of individual rats were rapidly removed and stored at - 70" C. Determination of Serum Glucose and ZGFBPs. Serum glucose concentrations were measured with a glucose oxidase assay system (Sigma Chemical Co., St. Louis, MO). Serum IGFBPs were detected by the ligand binding technique according to the method of Hossenlopp et al. (25). Three microliters of pooled sera from four to five rats at each time point were subjected to electrophoresis on a 12% SDSpolyacrylamide gel under nonreducing conditions. Separated proteins were then transferred to nitrocellulose membranes with Towbin buffer (15 mmol/L Tris, 120 mmol/L glycine, [pH 8.31 and 20% methanol). Filters were dried at 37" C for 5 min, and nonspecific binding was blocked by successive incubation of nitrocellulose membranes in 3%NP40 for 30 min, 1%BSA for 2 hr and 0.1% Tween 20 in ice-cold Tris-saline solution (0.15 moVL NaC1, 0.01 m o m Tris-HC1, 0.5 mg/ml sodium azide [pH 7.41) for 10 min. Nitrocellulose filters were then incubated overnight at 4" C with approximately lo6 cpm of 1251-labeledIGF-1 (Amersham Canada, Ltd., Oakville, Ontario, Canada) and then washed twice with ice-cold Tris-saline solution containing 0.1% Tween 20 for 15 min each and once in Tris-saline solution alone for 15 min. Filters were then air dried, and radioactivity was visualized by autoradiography. RNA Extraction and Hybridization. Total RNA from each individual rat liver or kidney tissue was extracted by the guanidinium isothiocyanate/cesium chloride technique (26). RNA was analyzed by electrophoresis in a 1%agarose and 2.2 mol/L formaldehydegel and transferred to nitrocellulose filter. In addition, serial dilutions of total RNA from each individual rat were also analyzed by slot-blot hybridization. Filters were then prehybridized for 4 to 6 hr at 42" C in a solution containing 50% formamide, 20 mmoVL NaH,PO, (pH 7), 4 x standard saline citrate ( l x = 0.15 mol/L NaC1-0.015 mol/L sodium citrate), 2 mmol/L EDTA, 4 x Denhardt's solution (1x = 0.02% BSA, Ficoll and polyvinylpyrolidine), 1% SDS and 100 Fg/ml sonicated denatured salmon sperm
Vol. 15, No. 6, 1992
1127
IGFBP-1 EXPRESSION AFTER PARTIAL HEPATECTOMY
DNA. Hybridization was performed at 42" C in the same solution for 16 to 20 hr with a rat IGFBP-1 cDNA (271, a rat IGF-I cDNA (28) or an oligonucleotide with a sequence corresponding to the complement of nucleotides 54 to 114 of the published sequences for rat IGFBP-3 (29).The cDNA probes were labeled by nick translation, whereas the oligonucleotide probe was radiolabeled with T4 polynucleotide kinase. Filters were initially washed at room temperature with 2 x standard saline citrate and 0.1% x SDS, and finally washed for 20 min at 65" C in 0.1 x standard saline citrate and 0.1% SDS, except in the case of the IGFBP-3 oligonucleotide probe where the final wash was in 1 x standard saline citrate and 0.1% SDS at 45" C for 15 min. Statistical Analysis. Dunnett's t test was used to determine the statistical significance of difference between control and treatment groups.
RESULTS In both the sham-operated rats and the PHx rats, there was biphasic change in serum glucose. A marked drop in serum glucose was observed 1h r after surgery. This decline in serum glucose was more marked in the sham-operated rats than in the PHx rats (Table 1). At least five IGFBPs were detected in rat sera from both sham-operated and hepatectomized rats (Fig. 1). The larger binding proteins with molecular mass of 39 to 42 kD most probably represent IGFBP-3. A small decline in the abundance of this binding protein from time 0 was observed in both sham-operated and hepatectomized rats. In contrast, the 29-kD IGFBP-1 was increased in sera from PHx rats compared with sham-operated rats at 3 hr and remained elevated up to 6 hr after the operation. The expression of hepatic IGFBP-1 mRNA was markedly increased within 1 hr, peaked at 3 hr and returned to basal values at 6 to 12 hr after PHx (Fig. 2). Ethidium bromidestained 18s and 28s ribosomal RNA shown in the lower panels of Fig. 2 indicate that the apparent increase in hepatic IGFBP-1 mRNA expression after PHx was not due to gel overloading. Only a small increase in the IGFBP-1 mRNA level was observed after the sham operation. The abundance of hepatic IGFBP-1 mRNA expression was also determined by slot-blot hybridization with serial dilutions of hepatic total RNA from sham-operated or hepatectomized rats. The intensity of the hybridization signal was quantitated by densitometry and expressed in terms of the time 0 control. As shown in Figure 3, hepatic IGFBP-1 mRNA was significantly increased (1.99 k 0.18-fold; p < 0.05) at 1 hr and reached a peak (2.32 f 0.22-fold; p < 0.01) 3 hr after PHx. The abundance of IGFBP-1 mRNA did not increase significantly in sham-operated rats (Fig. 3, open bars). Of interest, the abundance of kidney IGFBP-1 mRNA was also significantly increased (1.47 k 0.23-fold; n = 5 ; p < 0.05) after 3 hr and peaked (2.5 k 0.39; n = 5; p < 0.01) 6 h r after PHx (Fig. 4).No significant change in renal IGFBP-1 mRNA levels was seen in the sham-operated rats. In contrast to the increase in hepatic IGFBP-1 expression, a slight decrease in hepatic IGFBP-3 mRNA abundance was apparent in both the sham-operated and
A
29KDa-
29KDag
0
1
3
6
12
24
TIME(hr)
500
-
400 -
300
-
200-
fa
100-
w
v)
TIME ( hr ) FIG.1.Effect of PHx on serum IGFBPs. Pooled serum samples were analyzed by the ligand-blotting technique using lZ5I-IGF-Ias a tracer. (A) Serum from sham-operated control rats was analyzed, whereas the data obtained for serum from hepatectomized rats are shown in (B). Changes in abundance of the 29-kD binding protein, presumed to be IGFBP-1, were quantified by densitometry and are plotted in (C). Solid symbols show the results for serum from PHx rats, whereas open symbols represent the sham-operated controls.
the PHx rats (Fig. 5 ) . Similarly, a decrease in hepatic IGF-I mRNA abundance was observed in both groups of rats (Fig. 6 ) . When hepatic IGF-I mRNA was quantitated by dot-blot hybridization, it was apparent that the decrease in IGF-I mRNA was more marked in the sham-operated rats than in the PHx rats (Fig. 7). DISCUSSION
Liver cell mass is carefully regulated, and partial resection is accompanied by a cycle of hepatocyte proliferation such that the liver mass in the rat is rapidly restored to near normal within 48 to 72 hr. Although the
1128
GHAHARY ET AL.
HEPATOLOGY
6
A
1.6kb
I G FB P-1
28s
18s 0
0.5
3
1
6
12
0 0.5
24
1
3
6
12 24
T I M E(hr) FIG. 2. Effect of PHx on hepatic IGFBP-1 mRNA abundance. Total RNA (60 pg/lane) pooled from four to five rats killed at various times after either PHx (A) or sham operation (B) was electrophoresed and transferred to nitrocellulose filters. The upper panel shows the pattern of hybridization obtained with a rat IGFBP-1 cDNA, whereas the lower panel depicts the ethidium bromide-stained gel. The size of IGFBP-1 transcript was determined by comparison with the position of the 18s and 28s ribosomal RNAs.
** *
z
0 m m W
2-
a a X
11I T
W
W
zF 4W a
1
0l / ’ I 0
3
6
0.5
12
24
TIME ( hr )
FIG.3. Relative changes in hepatic IGFBP-1mRNA abundance after sham operation and PHx. Serial dilutions of total RNA from individual rat liver samples a t various time points after either sham operation (open histograms) or PHx (solid histograms) were analyzed by slot-blot hybridization. Data were quantitated by densitometery, and slope of the line of best fit for each RNA sample was expressed in term of that of the time 0 control, which was an arbitrary attributed value of 1. Data represent the means i- S.E.M. for n = 4 or 5igroup. The significant difference between the control, time 0 group and other groups are depicted as * and ** for p < 0.05 and p < 0.01, respectively.
liver is the major site of IGF biosynthesis, and the IGFs are mitogenic for a variety of cell types, the role of the IGFs in hepatic regeneration is not clear. In this study, we have shown that there is an increase in hepatic IGFBP-1 mRNA abundance after PHx. A parallel change in the serum 29-kD IGFBP was also observed. This IGFBP presumably represents IGFBP-1. Although the effects of PHx on serum IGF-I concentrations have
been reported previously, to our knowledge this study is the first report where effects of PHx on IGFBP-1 expression have been examined. Hepatic IGFBP-1mRNA expression was significantly increased within 1hr after operation and peaked by 3 hr. In comparison, the increase in kidney IGFBP-1 mRNA expression was slightly delayed, with peak levels achieved 6 hr after PHx. In several species, including the rat (27, 30, 31), food deprivation results in enhanced IGFBP-1 expression. This cannot be the only explanation for the effect of PHx on IGFBP-1 expression, because both sham and PHx rats were food-deprived prior to and following the operative procedures. The only caloric source was sucrose-supplemented water, and water consumption was similar in both groups of animals. Furthermore, hypoglycemia was even more marked in the sham-operated rats than in the PHx rats. Whereas hypoglycemia does not appear to be the direct cause of the enhanced IGFBP-1 expression in the PHx rats, it is probable that the stress response to PHx is considerably greater than sham operation. We have previously demonstrated that glucocorticoids, growth hormone and glucagon, in addition to insulin, can regulate expression of IGFBP-1 (32-34). Thus disturbances in these and other hormones after PHx may account for the enhanced IGFBP-1 expression. The increase in renal IGFBP-1 expression after PHx was an unexpected finding, but supports the notion that perhaps local and/or humoral factors are involved in the enhanced IGFBP-1 expression during liver regeneration. The role of IGFBP-1 remains controversial. It appears to be able to both inhibit and enhance the action of IGF-I in a variety of bioassays (15,351.The different effects of IGFBP-1 on IGF-I action may result from different postranslational modifications such as phosphorylation
Vol. 15, No. 6, 1992
1129
IGFBP-1 EXPRESSION AFTER PARTIAL HEPATECTOMY
B
A
I G FB P-1
1*6kb
28s
18s 0.5
1
3
6
12
24
0.5
3
1
6
12
24
TIME( h r ) FIG.4. Effect of PHx on renal IGFBP-1 mRNA abundance. Total RNA (60 p,g/lane) pooled from four to five rats killed a t various times after either PHx (A) or sham operation (B) was electrophoresed and transferred to nitrocellulose filters. The upper panel shows the pattern of hybridization obtained with a rat IGFBP-1 cDNA, whereas the lower panel depicts the ethidium bromide-stained gel. The size of IGFBP-1 transcript was determined by comparison with the position of the 18s and 28s ribosomal RNAs.
IGFBP-3
IGFBP-1 0 0.5 1 3 6 SHAM
12 24 0.5
1
3
6 12 24 PH
FIG.5. Comparison of the effect of PHx on hepatic IGFBP-1 and IGFBP-3 mRNA abundance. Total RNA (50 @/lane) pooled from four to five rats killed a t various times after either PHx or sham operation was electrophoresed and transferred to nitrocellulose filters. The upperpanel shows the pattern of hybridization obtained with a rat IGFBP-3 oligonucleotide probe, whereas the lower panel depicts the hybridization signal obtained with the IGFBP-1 cDNA.
that have been documented (36). The state of the IGFBP-1 present in sera from PHx rats and its functional effect on IGF-1 action are not known. Although serum somatomedin bioactivity declines after PHx (9), this may simply reflect the decline in circulating IGF-I levels rather than increased levels of IGFBP-1 or other somatomedin inhibitors. Several studies have demonstrated that circulating IGF-I levels fall after PHx (8-11); however, it is unclear whether this decline can be explained by loss of liver mass, changes in growth hormone or other hormones, hepatocellular injury or reduced food consumption. In this regard, the experiments reported here provide some
additional information. Our data demonstrate that there is a decrease in IGF-I mRNA levels in the remnant hepatic tissue after PHx. Interestingly, the decline in IGF-I mRNA levels was more marked in the shamoperated rats than the PHx rats. In a study reported by Russell, D’Ercole and Underwood (101, where serum IGF-I was measured by RIA, serum IGF-I levels were higher in PHx rats than in fasted control rats. Based on our findings, the increase in serum IGF-I levels after PHx demonstrated by Russell, D’Ercole and Underwood (lo) are unlikely to have resulted from enhanced IGF-I synthesis, but may have been caused by the effects of increased circulating IGFBP-1 concentrations on serum
1130
HEPATOLOGY
GHAHARY ET AL.
* 28%
* 18~+
0
0.5
1
3
6
12
24
-
0-5
1
3
SHAM
6
12
24
PH
-
T I M E AFTER OPERATION hours FIG. 6. Effect of PHx on hepatic IGF-1 mRNA abundance. Total RNA (60 kg/lane) pooled from four to five rats killed a t various times after either sham operation or PHx were electrophoresed and transferred to nitrocellulose filters. The upper panel shows the pattern of hybridization obtained with a rat IGF-1 cDNA, whereas the lower panel depicts the ethidium bromide-stained gel.
2
120 100
40
20
0
I
0.5
IGF-I levels or IGFBP-1 interference in the unextracted IGF-I RIA used by these authors. The decline in hepatic IGF-I and in IGFBP-3 mRNA expression reported here is likely to be caused at least in part by caloric restriction and the reduced pituitary growth hormone secretion that occurs in fasted rats (37). In conclusion, the results of this study clearly demonstrate that IGFBP-1 expression is up-regulated after PHx. Furthermore, this response to PHx appears to involve factors other than food deprivation. The physiological significance of the increased circulating levels of IGFBP-1 following hepatic resection remains to be determined. ADDENDUM
1
3
6
12
24
-
TIME hours
FIG.7. The effect of PHx on hepatic IGF-1 mRNA abundance. Serial dilutions of total RNA from each individual rat liver samples at various time points after either sham operation (open histograms) or PHx (solid histograms) were analyzed by slot-blot hybridization. The data were quantitated by densitometery, and slope of the line of best fit for each RNA sample was expressed in terms of that of the time 0 control, which was an arbitrary attributed value of 100. Data represent the means +- S.E.M. for n = 4 or 5/group. The significant difference between the control, time 0 control group and other groups are depicted as * and ** for p < 0.05 and p < 0.01, respectively.
After submission of this manuscript, a similar observation was reported by Mohn et al. (381, who noted a more prompt and pronounced increase in IGFBP abundance in the regenerating liver after hepatectomy. REFERENCES 1. Murphy U, Bell GI, Friesen HG. Tissue distribution of insulinlike growth factor I and I1 messenger ribonucleic acid in the adult rat. Endocrinology 1987;120:1279-1282. 2. Lund PK, Moats-Staats BM, Hynes MA, Simmons JD, Jansen M, D’Ercole AJ, Van Wyk J J . Somatomedidinsulinlike growth factor-I and insulinlike growth factor-I1 mRNAs in rat fetal and adult tissues. J Biol Chem 1986;261:14539-14544.
Vol. 15, No. 6, 1992
IGFBP-1 EXPRESSION AFTER PARTIAL HEPATECTOMY
3. Lowe WL, Roberts CT, Lasky SR, LeRoith D. Differential expression of alternative 5’ untranslated regions in mRNAs encoding rat insulinlike growth factor I. Proc Natl Acad Sci USA 1987;84:8946-8950. 4. Schoenle E, Zapf J , Hauri C, Steiner T, Froesch ER. Comparison of in vivo effects of insulinlike growth factor-I and I1 and growth hormone in hypophysectomized rats. Acta Endocrinol 1985;108: 167-174. 5. Schoenle E, Zapf J , Humbel RE, Froesch ER. Insulinlike growth factor I stimulates growth in hypophysectomized rats. Nature 1985;296:252-253. 6. McConaghey P, Sledge CB. Production of sulfation factor by the perfused rat liver. Nature 1970;225:1249-1250. 7. D’Ercole AJ, Stiles AD, Underwood LE. Tissue concentration of somatomedin C: further evidence for multiple sites of synthesis and paracrine or autocrine mechanism of action. Proc Natl Acad Sci USA 1984;81:935-939. 8. Uthne K, Uthne T. Influence of liver resection and regeneration on somatomedin (sulphation factor) activity in sera from normal and hypophysectomized rats. Acta Endocrinol 1972;71:255-264. 9. Unterman TG, Phillips LS. Circulating somatomedin activity during hepatic regeneration. Endocrinology 1986;119:185-192. 10. Russell WE, D’Ercole AJ, Underwood LE. Somatomedin C/insulinlike growth factor I during liver regeneration in the rat. Am J Physiol 1985;248:E618-E623. 11. Scott CD, Baxter RC. Insulinlike growth factor/mannose-6phosphate receptors are increased in hepatocytes from regenerating rat liver. Endocrinology 1990;126:2543-2549. 12. Zapf J , Waldvogel M, Froesch ER. Binding of non-suppressible insulinlike activity to human serum. Arch Biochem Biophys 1975;168:638-645. 13. Romanus JA, Terrell JE, Yang YWH, Nessley SP, Rechler MM. Insulinlike growth factor carrier proteins in neonatal and adult rat serum are immunologically different: demonstration using a new radioimmunoassay for the carrier protein from BRL-3A rat liver cells. Endocrinology 1986;118:1743-1758. 14. Martin JL, Baxter RC. Insulinlike growth factor binding proteins from human plasma. J Biol Chem 1986;261:8754-8760. 15. Burch WM,Correa J , Shively JE, Powell DR. The 25 kilodalton insulinlike growth factor (1GF)-bindingprotein inhibits both basal and IGF-I mediated growth of chick embryo pelvic cartilage in vitro. J Clin Endocrinol Metab 1990;70:173-180. 16. Meuli C, Zapf J, Froesch ER. NSILA-carrier protein abolishes the action of nonsuppressible insulinlike activity (NSILA-S) on perfused rat heart. Diabetologia 1978;14:255-259. 17. Knauer DJ, Smith GL. Inhibition of biological activity of multiplication-stimulating activity by binding to its carrier protein. Proc Natl Acad Sci USA 1980;77:7252-7256. 18. Brinkman A, Groffen C, Kortleve DJ, Geurts van Kessel A, Drop SLS. Isolation and characterization of a cDNA encoding the low molecular weight insulinlike growth factor binding protein (IBP-1) EMBO J 1988;7:2417-2423. 19. Wood WI, Cachianes G, Henzel WJ, Winslow GA, Spencer SA, Hellmiss R, Martin JL, Baxter RC. Cloning and expression of the growth hormone-dependent insulinlike growth factor-binding protein. Mol Endocrinol 1988;2:1176-1185. 20. Brown AL, Chiariotti L, Orlowski CC, Mehlem T, Burgess WH, Ackerman EJ, Bruni CB, Rechler MM. Nucleotide sequence and expression of a cDNA clone encoding a fetal rat binding protein for insulinlike growth factors. J Biol Chem 1989;264:5148-5154. 21. Shimasaki S, Uchiyama F, Shimonaka M, Ling N. Molecular cloning of the cDNAs encoding a novel insulinlike growth factor-binding protein from rat and human. Mol Endocrinol 1990;4:1451-1458. 22. Roghani M, Hossenlopp P, Lepage P, Balland A, Binoux M. Isolation from human cerbrospinal fluid of a new insulinlike
1131
growth factor-binding protein with a selective affinity for IGF-11. FEBS Lett 1989;255:253-258. 23. Zapf J , Kiefer M, Merryweather J, Masiarz F, Bauer D, Born W, Fischer JA, Froesch ER. Isolation from adult human serum of four insulinlike growth factor (IGF) binding proteins and molecular cloning of one of them that is increased by IGF-I administration and in extrapancreatic tumor hypoglycemia. J Biol Chem 1990; 265: 14892-14898. 24. Higgins GM, Anderson RM. Experimental pathology of liver: restoration of liver of the white rat following partial surgical removal. Arch Pathol 1931;12:186-202. 25. Hossenlopp P, Seurin D, Segovia-Quinson B, Hardouin S, Binoux M. Analysis of serum insulinlike growth factor binding proteins using western blotting: use of the method for titration of the binding proteins and competitive binding studies. Anal Biochem 1986;154:138-143. 26. Chirgwin J, Pryzbyla A, MacDonald R, Rutter W. Isolation of biologically active ribonucleic acid from sources enriched for ribonuclease. Biochemistry 1979;18:5294-5599. 27. Murphy W, Senevirantne C, Ballejo G, Croze F, Kennedy TG. Identification and characterization of a rat decidual insulinlike growth factor-binding protein complementary DNA. Mol Endocrinol 1990;4:329-336. 28. Murphy LJ, Bell GI, Duckworth ML, Friesen HG. Identification, characterization and regulation of a rat cDNA which encodes insulinlike growth factor-I. Endocrinology 1987;121:684-691. 29. Shimasaki S, Koba A, Mercado M, Shimonaka M, Ling N. Complementary DNA structure of the high molecular weight rat insulinlike growth factor binding protein (IGFBP 3) and tissue distribution of its mRNA. Biochem Biophys Res Commun 1989; 165:907-912. 30. Busby WH, Snyder DK, Clemmons DR. Radioimmunoassay of a 26,000-dalton plasma insulinlike growth factor-binding protein: control by nutritional variables. J Clin Endocrinol Metab 1988; 67:1225-1230. 31. McCusker RH, Campion DR, Jones WK, Clemmons DR. The insulinlike growth factor-binding proteins of porcine serum: endocrine and nutritional regulation. Endocrinology 1989;125: 501-509. 32. Luo J , Reid RE, Murphy W. Dexamethasone increases hepatic insulinlike growth factor binding protein-1 (IGFBP-1) mRNA and serum IGFBP-1 concentration in the rat. Endocrinology 1990; 127:1456-1462. 33. Seneviratne C, Luo JM,Murphy W. Transcriptional regulation of rat insulinlike growth factor binding protein-1 by growth hormone. Molec Endocrinol 1990;4:1199-1204. 34. Kacchra Z, Yannopoulos C, Barash I, Guyda HJ, Murphy U, Posner BI. The differential regulation by glucagon and growth hormone of IGF-I and IGF binding proteins (IGF-BPs) in cultured rat hepatocytes [Abstract 11331. 72th Annual Meeting of the Endocrine Society, Atlanta, GA 1990. 35. Elgin RG, Busby WH, Clemmons DR. An insulinlike growth factor (IGF) binding protein enhances the biological response to IGF-I. Proc Natl Acad Sci USA 1987;84:3254-3258. 36. Jones JI, Clemmons DR. The bioactivity of an insulinlike growth factor binding protein is modulated by serine phosphylation [Abstract]. Clin Res 1990;38:978A. 37. Tannenbaum GS, Epelbaum J , Colle E, Brazeau P, Martin JB. Antiserum to somatostatin reverses starvation-induced inhibition of growth hormone but not insulin secretion. Endocrinology 1978;102:1909-1914. 38. Mohn KL, Melby AE, Tewari DS, Laz TM, Taub R. The gene encoding rat insulinlike growth factor-binding protein 1is rapidly and highly induced on regenerating liver. Mol Cell Biol 1 9 9 l ; l l : 1393-1401.
