Insulin Antagonism of Glucocorticoid Induction ... - Wiley Online Library

Eur J Biochem 118. 137-142 (1981) c‘ FFBS 1981

Insulin Antagonism of Glucocorticoid Induction of Tyrosine Aminotransferase in Cultured Foetal Hepatocytes Kevin K . W. HO, Max H. CAKE, Georgc c‘. T. YEOH, and Ivan T. OLIVER School or Environmental and Life Sciences, Murdoch University, and Departments o f Physiology and Biochemistry, University of Western Australia, Nedlands (Received April 1. 1981)

Whereas dexamethasone is unable to induce tbe premature formation of hepatic tyrosine aminotransferase whcn administered to foetal rats in utero, the stcroid can induce the enzyme in foetal rat liver if the liver is first removed from the environment in utero and grown in culture. Dexamethasone produced a significant induction of the enzyme at a concentration of 0.1 n M in cultured foetal hepatocytes, but for optimal induction the cells werc exposed to 10 n M for 15 h. Growing the hepatocytes in the presence of physiological concentrations of insulin had no effect on the enzyme activity in control cells. However, the induction of the enzyme by dexamethasone was markedly diminished in the presence of insulin. This effect of insulin is both time-dependent and dose-dependent with significant inhibition being obtained with 1 n M insulin. Growing foetal hepatocytes in the presence of insulin has no effect on either the cellular level of glucocorticoid receptor or on the ability of dcxamethasonereccptor complexes to undergo nuclear translocation suggesting that insulin inhibits some event subsequent to translocation. The results are discussed in relation to the postnatal appearance of tyrosine aminotransferase and suggest that the marked decline in the plasma concentration of insulin, that is known to occur at birth, is a major contributor to the postnatal induction of the enzyme. It is well documented that tyrosine aminotransferase is absent from foetal rat liver. Enzyme activity develops within a few hours of birth [I], even after premature delivery [2], and glucocorticoids have been implicated in this process. However, the administration of glucocorticoids to foetal rats in uzero fails to induce tyrosine aminotransferase even though these agents are capable of markedly enhancing the activity of the enzyme in postnatal rats [1,3-51. Singer and Litwack [6] demonstrated that the foetus is unresponsive despite an adequate hepatic accumulation of the administered steroid. A recent report [7] indicates that foetal hepatocytes acquire a functional glucocorticoid receptor system on or about the 18th day of gestation. This correlates well with the stage of development when exogenously administered glucocortocoids are able to precociously induce glycogen deposition in the liver [S]. These observations suggest that the hormonal environment of the foetus is not conducive to tyrosine aminotransferase synthesis. That is, that there is some factor or factors prescnt irz utero which repress the action of glucocorticoids on tyrosine aminotransferase synthesis, but which do not block glycogen synthesis. Indeed, it has been shown that glucocorticoids can induce tyrosine aminotransferase in foetal rat liver if the liver is first removed from the environment in utero and cultured either as a n explant [9] o r as a monolayer of cells [10,11]. In a recent report, Raiha and Edkins [I21 have shown that insulin abolished the dexamethasone-induced increase of argininosuccinate synthetase (EC 6.3.4.5) and

argininosuccinate lyase (EC 4.3.2.1) activities in cultured foetal rat liver. Circulating insulin levels are known to bc high in the foctus and to fall soon after birth [13, 141, and insulin administration to prematurely delivered rats markedly inhibits the postnatal appearance of tyrosine aminotransferase that would otherwise occur [15]. In this paper we report on studies of thc effect of insulin on the glucocorticoid induction of tyrosine aminotransferase activity in foetal hcpatocytes in culture. Whereas insulin has previously been shown to be essential for induction of tyrosine aminotransferasc by glucocorticoids in cultured hepatocytes derived from adult liver [16], we have found that in cultured foetal hepatocytes, insulin has an inhibitory effect on enzyme induction by glucocorticoids. This difference in the response of foetal hepatocytes to insulin is discussed in relation to the inability of exogenously administered steroids to induce tyrosine aminotransferase in uiero.

~.

Chemiculs

EKJPW.~. Tyrosine aminotransferase or L-tyrosine : 2-oxoglutarate aminotransferase (EC 2.6.1.5); argininosuccinatc synthetasc (EC 6.3.4.5); argininosuccinate lyase (EC 4.3.2.1). Triviul Numes. Corticoatcrone, 1l/1,21-dihydroxy-pregn-4-cne-3,20dione; dexamcthasone, 9-fluoro- 11/1,17,21-trihydroxy-16x-methyl-pregn1,4-diene-3,20-dione.

MATERIALS AND METHODS

A n imu Is Rats of the Wistar albino strain of Rurtus norwegicus were used. Gestational age was determined from the time of a positive vaginal smear for spermatozoa and is accurate to & 8 h. These animals have a gestation period of 22 days.

1,-[3,5-~H]Tyrosineand [I ,2-’HH]dexamethasone were purchased from the Radiochemical Centre (Amersham, Bucks, UK). L-Tyrosine and sodium diethyldithiocarbamatc were from Merck (Darmstadt, FRG). Pyridoxal 5’-phosphate, 2-oxoglutarate, dexamethasone, corticostcrone, calf thymus

138

DNA, and activated charcoal were obtained from Sigma Chemical Co. (St Louis, MO, USA). Collagenase (grade 11) was purchased from Boehringer Mannheim (Victoria, Australia). 2,4-Dinitrophenylhydrazine was supplied by Ajax Chemical Co. (Sydney, Australia) and diaminobenzoic acid and Coomassie brilliant blue G250 by the Eastman Kodak Co. (Rochester, NY, USA). Dextran T500 and CM-Sephadex C-50 were obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). Bovine insulin B.P. was supplied by the Commonwealth Serum Laboratories (Parkville, Victoria, Australia). Eagle's Minimal Essential Medium and foetal calf serum were from Flow Laboratories (Annandale, N.S.W., Australia). Fungizone, penicillin/streptomycin, and glutamine were obtained from Grand Island Biological Co. (Grand Island, NY, USA). Hepatocyte Isolation and Culture Livers from 19-day foetal rats were chopped and incubated with collagenase in balanced salts solution [17], as described by Yeoh et al. [Ill. Cells were harvested and washed twice in balanced salts solution by centrifugation at 50 x ,g for 2 min, then suspended in culture medium and dispensed into 90-mmdiameter plastic culture dishes previously coated with collagen. The basal culture medium was modified Eagle's minimal essential medium supplemented with 10 7; charcoal-treated foetal calf serum, glutamine (2.4 mM final concentration) fungizone (2.8 lg/ml) and penicillin/streptomycin (57 units/ml and 570 Fg/ml respectively). Further additions to the medium are indicated in the text. In order to remove endogenous steroids, the foetal calf serum was charcoal-treated prior to medium supplementation. The serum was incubated for 30 min at 37 "C in the presence of activated charcoal (50 mg/ml serum), cooled and then centrifuged (20000 x g , 20 min, 4 "C). Using a tracer dose of ['H]corticosterone it was shown that this process removed more than 99 of the endogenous corticosteroids. The cultures were maintained at 37 "C in a water-saturated atmosphere of 5 7;co2 / 9 5 air and were given fresh medium at 24-h intervals after plating unless otherwise stated. The hepatocytes formed a monolayer atcached to the collagen substratum but most of the haemopoietic cells remained in suspension and were discarded at the first change of medium. The residual haemopoietic cells were completely removed by subsequent media replacement. Preparation of Hepatocyte Extracts The cultured cells were washed vigorously twice with balanced salts solution, removed from the dishes with a Teflon scraper and suspended in balanced salts solution. The cellular suspension was centrifuged at 1600 x g for 2 min, the supernatant removed and the weight of the pellet determined. The cellular pellet was resuspended in either 2 vol. of ice-cold 0.05 M potassium phosphate buffer pH 6.5 for enzyme assay or 4 vol. of 0.05 M Tris/HCl pH 7.6, containing 3 mM MgC12 and 10 mM 2-mercaptoethanol for glucocorticoid receptor assay. The cell suspensions were sonicated in a Branson Sonifier (model B12, Branson Instruments, Conn., USA) at 2 A for 15 s using a microprobe. A sample of the sonicated preparation was taken for DNA estimation. The remainder was centrifuged in a Beckman Airfuge at approximately 210 kPa (165000 x g max) for 12 min in an A-I00 rotor.

Assay of Tyrosine Aminotransferase Before assay the cytosol samples were subjected to ionexchange chromatography on CM-Sephadex C50 in order to separate tyrosine aminotransferase from contaminating aspartate aminotransferase [I 11. Fractions eluted with 0.33 M KCI in 0.05 M potassium phosphate buffer pH 6.5 containing the enzyme were assayed for tyrosine aminotransferase activity using the radiochemical assay described previously [I I]. Specific activity was expressed as nmol p-hydroxyphenylpyruvate produced x h-' x mg protein-'. Assay of Glucocorticoid Receptor The hepatocyte cytosol was incubated with 50 nM [3H]dexamethasone in the presence and absence of a 667-fold excess of non-radioactive steroid. After a 3-h incubation at 0 "C, when steroid binding to receptor was essentially complete, the specific macromolecular-bound fraction was determined using the dextran-coated charcoal technique [18]. Results are expressed as pmol bound/mg DNA.

Measurement of Nuclear Uptake of Glucocorticoids

To determine the specific nuclear binding of glucocorticoid in the intact cell, cell cultures were incubated with 5 0 n M [3H]dexamethasonein the presence and absence of 5 pM nonradioactive steroid. After a I-h incubation at 37°C the cultures were subjected to the following fractionation procedures, which were carried out at 0-4°C. The cells were washed with ice-cold buffered salts solution, harvested (as previously described) and homogenized in 4 vol. of 0.05 M Tris/HCl pH 7.6, containing 3 mM MgClz and 10 mM 2-mercaptoethanol in a Potter-Elvehjem homogenizer. The homogenate was adjusted to 0.25 M sucrose, an aliquot taken for DNA estimation and the remainder centrifuged at 1500 x g for 10 min. The supernatant was used to obtain a cytosol fraction by further centrifugation in a Beckman Airfuge. Specifically bound steroid was determined using the dextrancoated charcoal technique [18]. Nuclei were purified from the pellet of the low-speed centrifugation of the cell homogenate. The pellet was resuspended in 2 vol. of 0.05 M Tris/HCl pH 7.6 containing 0.25 M sucrose and 3 mM MgC12, and then diluted with 2.5 vol. of 0.05 M Tris/HCl pH 7.6 containing 2.2 M sucrose and 3 mM CaC12. The resulting suspension was centrifuged through a cushion of 2.15 M sucrose-calcium buffer at 73000 x g for 30 min. After washing the pellet with 0.05 M Tris/HCl pH 7.6 0.25 M sucrose, 3 mM MgC12 the nuclear fraction was resuspended in this buffer and duplicate samples assayed for DNA and radioactivity. Specific binding represents the amount of radioactivity bound in the absence of unlabelled competitor less the amount bound in the presence of competitor [19]. Results for both cytosol and nuclear binding are expressed as pmol steroid bound/mg DNA.

Assays of Protein and D N A Protein concentration was determined using the dyebinding method of Bradford [20] incorporating the modification of Beardon [21] and using bovine serum albumin as a standard. DNA was assayed by a fluorometric procedure [22] using highly polymerized calf thymus DNA as a standard.

139 >-

c ._ > .c 0 ; m .- 1200

$ 2

,m g1000 c

;F L

800

.

c

.E 1, ._ =

E p -

600

400

a , "

c"

200 /

A

n

d/ V

0

0

0.1 1 10 100 1000 Dexamethasone concentration (nM)

6

12 18 Time of exposure ( h )

24

Fig. 1. Induction of tyrosine aminotransferase by dexamethasone in cultured jbetal rat hepatocytes. Hepatocytes from 19-day foetal rats were isolated and cultured as described in Materials and Methods. (A) After 32 h of culture the media was supplemented with the indicated concentration of dexamethasone and culture continued for 16 h. (B) Cultures were supplemented with 10 nM dexamethasone at various times prior to 48 h of culture, such that the exposure time to the steroid was varied. At the completion of both experiments cells were harvested and assayed for tyrosine aminotransferase. The results represent the mean + S.E.M. of three independent cell preparations for each experiment

Table 1. Effect of insulin on induction of tyrosine aminotransferase hy dexamethasone in cultured foetal rat hepatocytes Hepatocytes from 19-day foetal rats were cultured for 32 h in the absence or presence of 100 nM insulin (as indicated) with the media being replaced at least once during this period. After 32 h half of the cultures were treated with 10 nM dexamethasone, the other half receiving an equivalent volume of 1,2-propanediol (vehicle) and culture continued for a further 16 h. Cells were harvested and assayed for tyrosine aminotransferase. The results represent the mean S.E.M. of eight separate experiments

+

Addition to culture medium -

~~

insulin

Tyrosine aminotransferase activity ~

~

dexamethasone

mean f S E M nmol x h ~~~

-

+

+

-

+ + -

45 0 700 3

increase

' x mg protein '

+ 107 + 104 8

~ _ _ _

655 3 599, 133 2902 377 230 3 (inhibition of increase = 65 %)

+

RESULTS It has previously been shown that dexamethasone is able to induce tyrosine aminotransferase in cultured foetal hepatocytes provided the cells are isolated from foetal rats close to term [7,10,11]. As is evident from Fig. 1A the effect of dexamethasone is dose-dependent with a significant induction being obtained with 0.1 nM dexamethasone ( P < 0.05). The response is relatively slow with an exposure time in excess of 12 h being required for maximal response (Fig. I B). As a consequence all further experiments reported here employed a 15-h exposure to 10 n M dexamethasone as the optimal induction conditions. Growing the hepatocytes in the presence of 100 nM insulin failed to affect the enzyme activity in control cells. However, the presence of insulin markedly diminished the response of those cells to dexamethasone (Table 1). The presence of 100 nM insulin resulted in a 65 % inhibition (P< 0.01) of the induction by the steroid. A similar effect of insulin was observed when 100 nM of the natural glucocorticoid corticosterone, was used as inducer (data not shown). In this case, the induction was inhibited 67%. This effect of insulin on tyrosine aminotransferase induction by glucocorticoids is

dose-dependent (Fig. 2A). In fact, within the range 1 - 100 n M insulin there is a significant linear regression between the logarithm of the insulin concentration and the degree of inhibition (r = 0.993, P < 0.01). A significant inhibition was obtained with 1 nM insulin ( P < 0.01) and the insulin concentration required to inhibit glucocorticoid induction by 50 was calculated to be 4 nM. It can be seen from Fig. 2 B that the effect of insulin is also time-dependent. A significant inhibition was obtained if the insulin was added only 1 h before the addition of dexamethasone (48% inhibition, P < 0.05). Note, however, that the actual exposure period to insulin was much longer than this ; cells exposed to insulin for 1 h prior to the addition of the steroid were also in the presence of insulin throughout the 15-h induction period. Maximal inhibition by insulin required that the cells be pre-exposed to insulin for at least 20 h prior to the addition of the inducer. Previous studies have shown that the degree of tyrosine aminotransferase induction by glucocorticoids in cultured hepatocytes increases progressively during the first few days in culture [7,10]. This can be clearly seen in Fig. 3. Moreover, it can be seen that the degree of inhibition by insulin of the glucocorticoid induction of tyrosine aminotransferase is independent of the actual cellular response to the steroid. After 24 h of culture, when the hepatocyte response to glucocorticoids is low insulin inhibits the induction by 55%. After a further 48 h of culture, when the induction by dexamethasone is approximately sevenfold greater, the inductive effect of the steroid is inhibited to the same extent by insulin (62 "/,). Hepatocytes derived from 19-day foetal rats and cultured for 48 h have been shown previously to contain a specific cytoplasmic receptor for glucocorticoids [7]. The cell-free binding data presented in Table 2 indicate that growing the cells in the presence of insulin had no effect on the concentration of the glucocorticoid receptor in these cells. This finding is confirmed in studies with intact cells (Table 3), although the total amount of receptor detected in these studies was much higher. This may be a reflection of a greater stability of the receptor when complexed with ligand a5 compared to the free receptor as has been seen in studies with HTC cells [23]. The data in Table 3 also indicates that the activity of the receptor, with respect to nuclear translocation, is unaffected by the presence of insulin in the culture media. Whether the hepatocytes were grown in the presence or absence of insulin about 50 % of the specifically bound steroid is nuclear located after a I-h incubation with the steroid.

1 40 r

I

B

1 0 1

Insulin concentration ( n M )

10 20 Time of exposure (h)

30

Fig. 7 . E / f i ~ rof insulin on induction of tyrosine uminorrun.$wise by dexamethasone in culturedJbeia1 rat hepatocytes. Hepatocytes from 19-day foetal rats were cultured as described in Materials and Methods. (A) The culture media was supplemented with the indicated concentrations of insulin. The medium was replaced and fresh insulin added after 24 h and 31 h in culture. One hour later either 10 nM dexamethasone (stippled bars) or an equivalent volume of the vehicle, 1,2-propanediol (open bars), was added and culture continued for 16 h. (B) Cells were exposed to 10 nM dexamethasone after 32 h in culture but only after the media was supplemented with 100 nM insulin for the times indicated prior to the addition of dexamethasone. In those cases where the pre-exposure to insulin exceeded 10 h the medium and insulin supplementation were replaced every 10 h. In both experiments the cells were harvested after 48 h in culture and the cytoplasm assayed for tyrosine aminotransferase activity. For both experiments the data represents the mean f S.E.M. for three separate cell preparations

Table 3. Effect qf insulin on the nuclear uptake of' dexameihasonereceptor complex in cultured foetal hepatocytes Hepatocytes from 19-day foetal rats were cultured for 48 h in the absence or presence of 100 n M insulin. The cell cultures were then incubated a further hour at 37 "C with 50 nM [3H]dexamethasone in the presence and absence of a 100-fold excess of unlabelled steroid. The cells were harvested, cytosol and nuclear fractions prepared and specific binding of dexamethasone estimated as described. The data represent the mean k S.E.M. of three separate cell preparations Fraction

Receptor concentration ~

~~

~~

~~

control media pmol x mg DNA-'

Time in culture (days)

Fig. 3. EfSect of culture rime oil insulin unrugonrsm qf'r,vrosrne umitioirans/erase induction by dexamethasone. Hepatocytes from 19-day foetal rats were cultured in the absence (0, A) or presence ( 0 ,A) of 100 nM insulin. Cells were harvested after 24, 48, and 72 h in culture but after having been exposed to 10 n M dexamethasone (A, A) or an equivalent volume of the vehicle, 1,2-propanediol (0, 0) for the previous 16 h . Cell extracts were analysed for tyrosine aminotransferase activity and the results represent the mean f S.E.M. of three separate cell preparations

insulin-supplemented media

-~

~

(x) ~

~~

Cytoplasmic Nuclear

1 36 f 0 16 1 28 +_ 0 35 (48 5)

1 53 f 0 37 1 44 +_ 0 35 (48 5)

Total

2 64 +_ 0 37

2 97 i 0 65

DISCUSSION

Table 2. Effect of insulin on the cell-free binding of dexamethasone by .specific cytoplasmic receptors Hepatocytes from 19-day foetal rats were cultured for 48 h in the absence or presence of the indicated concentration of insulin with the media being replaced every 24 h. Cells were harvested and a cytosol fraction prepared and assayed for glucocorticoid receptor concentration determined. The results represent the mean i S.E.M. of duplicatc determinations on each of three independent cell preparations Addition to culture medium

Receptor concentration pmol x mg DNA-'

Nil 10 nM insulin 100 nM insulin

0.62 i 0.08 0.69 0.06 0.63 f 0.03

Glucocorticoids are unable to induce the premature formation of tyrosine aminotransferase when administered to foetal rats in utero even though these agents are capable of markedly enhancing the activity of the enzyme in postnatal rats [l, 3 - 51. However glucocorticoids can induce tyrosine aminotransferase in foetal rat liver if the liver is cultured, either as an explant [9] or as a monolayer of cells [10,11]. Dexamethasone produced a significant induction of the enzyme at a concentration of 0.1 nM in foetal hepatocytes grown in culture, but for optimal induction a concentration of 10 n M was required (Fig. 1). This is somewhat lower than the concentration (1 00 nM) required to optimally induce tyrosine aminotransferase in cultured adult hepatocytes [24] and hepatoma cells [25,26].Recent findings indicate that the degree of response of cultured foetal hepatocytes to dexamethasone is directly correlated with the cellular concentration of the glucocorticoid receptor [7]. Hepatocytes derived

141

from foetal rats at an early stage of gestation initially lack the receptor and are unresponsive to dexamethasone added to the culture medium. However, with continued culture both receptor activity and steroid responsiveness are acquired. On the other hand, hepatocytes derived from foetal animals close to term contain the receptor and are responsive to exogenous steroid from the commencement of culture. The absence of a specific glucocorticoid receptor from hepatocytes may well be the cause of the unresponsiveness of the foetal liver in rats prior to the 18th day of gestation. However, the fact that the tyrosine aminotransferase system does not respond to exogenous steroid until after delivery indicates that other factors must be contributing to the unresponsiveness of foetal rats close to term. Moreover, since hepatocytes from these animals are responsive if first removed from the environment in utero and grown in culture it is possible that there is some Factor or factors present in u ~ e r owhich repress the action of glucocorticoids on the tyrosine aminotransferase system. This study suggests that the high levels of insulin known to be present in the plasma of foetal rats could antagonize the action of glucocorticoids on the tyrosine aminotransferase system. During the final stage of gestation in the rat the insulin concentration in the foetal plasma can reach levels as high as 1.8 n M but declines to low levels (0.2 nM) within 2 h of delivery [14,27]. The induction of tyrosine aminotransferase by dexamethasone in cultured foetal hepatocytes was markedly inhibited by insulin in a dose-dependent manner ; significant inhibition occurred with an insulin concentration as low as 1 n M (Fig.2A). It was calculated that an insulin concentration of 4 nM, somewhat higher than physiological concentrations, is required to inhibit by 50 % the glucocorticoid induction of tyrosine aminotransferase. However, one must bear in mind the likely possibility that the insulin, when added to the culture media, is partially inactivated or degraded by the hepatocytes during the course of the 15-h incubation. If this is the case then the calculated insulin concentration required to inhibit the glucocorticoid response is high due to an overestimation. Alternatively, it may be that insulin is only one of several factors present in the environment in utero that act in cohort to totally repress the glucocorticoid response of this particular enzyme system. The suggestion that insulin is at least one of the factors responsible for foetal unresponsiveness to glucocorticoids is supported by the finding that insulin administration to prematurely delivered rats markedly inhibits the postnatal appearance of tyrosine aminotransferase [15]. Insulin has been shown to antagonize the response of a number of other systems to glucocorticoids. Smith et al. [28] reported that insulin abolishes the stimulatory effect of cortisol on lecithin synthesis in cultured foetal lung cells. In a recent report, Raiha and Edkins [I21 have shown that insulin also markedly inhibits the dexamethasone induction of argininosuccinate synthetase and argininosuccinate lyase in foetal rat liver explants maintained in organ culture. The inhibitory effect of insulin is not confined to foetal tissues as it has been shown that insulin inhibits the glucocorticoid induction of phosphoenolpyruvate carboxykinase in Reuber H35 cells [25, 291 and of tryptophan oxygenase in cultured adult rat hepatocytes [30]. Nevertheless the inhibitory effect of insulin on the tyrosine aminotransferase system may well be a characteristic that is particular to the foetal stage of liver development as insulin has been shown to induce the enzyme in perfused adult rat liver [31] and to be essential for glucocorticoid

induction of tyrosine aminotransferase in cultured adult hepatocytes [16]. The level of glucocorticoid receptor in a tissue almost certainly determines, at least in part, the degree of responsiveness of that tissue to glucocorticoids [7]. Indeed, it has been shown that in cases of reduced sensitivity of a tissue to glucocorticoids the concentration of cytoplasmic receptor is low [32- 341. For this reason we examined the possibility that culturing the cells in the presence of insulin resulted in diminished levels of the receptor. From the data presented in Tables 2 and 3 this is clearly not the case; hepatocytes grown in insulin-supplemented media have a normal concentration of the glucocorticoid receptor. Moreover, the cellular uptake of dexamethasone, binding of the steroid to the receptor and translocation of the steroid-receptor complex to the nucleus are unaffected by the presence of insulin in the culture medium (Table 3). Clearly, insulin must be affecting some step subsequent to nuclear uptake of the receptor complex. Current investigations are aimed at elucidating not only the site of insulin action but also the mechanism by which it effects this inhibition. Glucocorticoids [I, 151, glucagon [4,15], adrenaline and cyclic AMP [4,15,35] have all been implicated in the postnatal appearance oftyrosine aminotransferase. Insulin has also been implicated on the basis that the decline in the plasma concentration of insulin that occurs at birth would result in a postnatal increase in the hepatic concentration of cyclic AMP. However, from the data presented in this paper it can be assumed that, in addition, the responsiveness or the hepatic tyrosine aminotransferase system to circulating glucocorticoids will increase dramatically as a result of the diminished concentration of insulin. Thus the postnatal decline in the plasma insulin concentration may well be one of the major contributors to the postnatal induction of the enzyme. We thank Dr J. G. Steele Tor his critical assessment of the manuscript. This work was supported by grant from the Australian Research Grants Committee and the National Health and Medical Research Council of Australia. K . K . W. Ho was the recipient of a Murdoch University Research Studentship.

REFERENCES 1 . Sereni, F., Kenney, F. T. & Kretchmer, N. (1959) J . BkJl. C’herti. 234, 609 - 612. 3. Holt, P. G. & Oliver, 1. T. (1968) Biockrm. J . 108, 333-338. 3. Franz, J. M. & Knox, W. E. (1967) Bioc/zcmislr.y, 6, 3464-3471. 4. Greengard, 0. (1969) Bioclzem. J . 115, 19 -24. 5 . Cake, M. H., Ghisalberti, A. V. & Oliver, I. T. (1973) Biochenz. Bioplzys. Res. C‘ommun. 54, 983 - 990. 6. Singer, S. & Litwack, G. (1971) Lndocriiiology, 88,1448-1455. 7. Cake, M. H., Ycoh, G., Oliver, 1. T. & Litwack, G. (1979) Adv. Biosci. 25, 263 - 272. 8. Grecngard, 0. & Dewey, H. K. (1970). Dev. Biol. 21, 452-461. 9. Wicks, W. D. (1968). J . BioL Ch(w. 243, 900-906. 10. Croizat, B., Granclli-Piperno. A. & Gros, F. (1974) Bioc/zimic iPari,s) 56, 305-313. 1 1 . Yeoh, G. C. T., Bennett, F. A. & Oliver, I . T. (1979) Bioclzevn. J . 180, 153 - 160. 12. Raiha, N . C . R. & Edkins, E. (1977) B i d . Neoncr/e. 31, 266-270. 13. Cohen, N. M. & Turner, R. C. (1972) Bid. Neonule, 21, 107 - 11 1 . 14. Girard, J. R., Pcrre, P., Kervran, A., Pegorier, J . P. & Assan, R. ’ IC.i?lC (1977) in Glucugo,i: Its Role in Phy,sioloxy arid Clirricul .hfcd’ (Foa, P. P., Bajaj, J . S. & Foa, N . L., eds) pp. 563-581, SpringerVerlag, New York. IS. Holt, P. G. & Oliver, 1. T. (1969) Biochemistrj., 8, 1429-1437.

142 16. Michalopoulos, G. & Pitot, H. C. (1975) Exp. Cell Res. 94, 70-78. 17. Hanks, J. H. & Wallace, R. E. (1949) Proc. Soc. Exp. Biol. Med. 71, 196- 200. 18. Beato, M. & Feigelson, P. (1972) J . Biol. Chem. 247, 7890-7896. 19. Rousseau, G. G., Baxter, J. D., Higgins, S. J. & Tomkins, G. M. (1973) J . Mol. Biol. 79, 539-554. 20. Bradford, M. M. (1976) Anal. Biochem. 72,248-254. 21. Bearden, J. C. (1978) Biochim. Biophys. Acta, 533, 525-529. 22. Hinegardner, R. T. (1971) Anal. Biochem. 39, 197-201. 23. Baxter, J. D. & Tomkins, G. M. (1971) Proc. Null Acad. Sci. USA, 68, 932-937. 24. Ernest, M. J., Chen, C.-L. & Feigelson, P. (1977) J . Biol. Chenz. 252, 6783-6791. 25. Barnett, C. A. & Wicks, W. D. (1971) J . Biol. Chenz. 246, 72017206. 26. Granner, D. K., Lee, A. & Thompson, E. B. (1977) J . Biol. Chem. 252, 3891 -3897.

27. Di Marco, P. M., Ghisalberti, A. V., Martin, C. E. & Oliver, I. T. (1978) Eur. J . Biochem. 87, 243-247. 28. Smith, B. T., Giroud, C. J. P., Robert, M. & Avery, M. E. (1975) J . Pediatr. 87, 953-955. 29. Wicks, W. D., Barnett, C. A. & McKibbin, J. B. (1974) Fed. Proc. 33, 1105-1111. 30. Nakamura, T., Shinno, H. & Ichihara, A. (1980) J . Biol. Chern. 255, 7533-7535. 31. Hager, C. B. & Kenney, F. T. (1965) J. Biol. Chem. 243.3296-3300. 32. Hackney, J. F., Gross, S. R., Aronow, L. & Pratt, W. B. (1970) Mol. Pharmacol. 6 , 500-512. 33. Baxter, J. D., Harris, A. W., Tomkins, G. M. & Cohn, M. (1971) Science (Wash. DC) 171, 189-191. 34. Rosenau, W., Baxter, J. D., Rousseau, G. G. & Tomkins, G. M. (1972) Nut. New Biol. 237, 20-24. 35. Ghisalberti, A. V., Steele, J. G., Cake, M. H., McGrath, M. C. & Oliver, I. T. (1980) Biochenz. J . 190, 685-690.

K. I(.W. Ho and M. H. Cake, School of Environmental and Life Sciences, Murdoch University, Murdoch, Western Australia, Australia 61 50 G. C. T. Yeoh, Department of Physiology, University of Western Australia, Nedlands, Western Australia, Australia 6009 I. T. Oliver, Department of Biochemistry, University of Western Australia, Nedlands, Western Australia, Australia 6009