Glucocorticoids regulate the ontogenetic transition of adrenergic receptor subtypes in rat liver

Life Sciences, Vol. 48, pp. 1059-1065 Printed in the U.S.A.

Pergamon Press

G L U C O C O R T I C O I D S R E G U L A T E THE O N T O G E N E T I C T R A N S I T I O N OF A D R E N E R G I C R E C E P T O R SUBTYPES IN RAT L I V E R R.A. Huff, F.J. Seidler and T.A. Slotkin Department of Pharmacology Duke University Medical Center Durham, North Carolina 27710 USA (Received in final form January I0, 1991)

Summarv

During neonatal development, adrenergic control of hepatic glucose metabolism undergoes a transition from ~-receptor to a~-receptor-mediated dominance, coincident with the onset of function of the hypothalamus-pituitary-adrenocortical axis at the conclusion of the third to fourth week postpartum. To determine whether glucocorticoids contribute to this switch, neonatal rats were given 1 mg/kg of dexamethasone on postnatal days 13, 14 and 15 and the adrenergic receptor population examined by radioligand binding techniques. Dexamethasone accelerated the maturational replacement of fl-receptors with the ai-subtype; the loss of [3receptors was not reversible upon discontinuing treatment. When the glucocorticoid was given earlier, on days 7, 8 and 9, similar effects were obtained, but the suppression of the fi-subtype was only temporary; treatment before parturition (gestational days 17, 18 and 19) failed to suppress 13-receptor binding. These results suggest that, during a critical period, adrenocorticosteroids provide an important signal for the transition of adrenergic control of hepatic function. Hepatic regulation of glucose metabolism is mediated, in part, through adrenergic mechanisms. In the adult, catecholamines act primarily through ~tl-receptors, whereas I~-receptor actions predominate in the fetus and neonate (1-5). This transition is a logical adaptation to the development of sympathetic innervation, as there is an associated switch from circulating epinephrine from the adrenal medulla to norepinephrine derived from sympathetic nerves (6-8). Secondly, the change in dietary composition at the approach of weaning favors al-receptor mechanisms: milk provides a carbohydrate-poor diet, requiting gluconeogenesis which is closely linked to fl-adrenergic function, whereas the mature diet is carbohydrate-rich and thus emphasizes the need for glycogenolysis which is cq-receptor-dependent (9). The ontogenetic transition in adrenergic control of hepatic function coincides with maturation of hypothalamus-pituitary-adrenocortical function at the end of the third to fourth postnatal week in the rat (8,10,11). In light of the close association of glucocorticoids and adrenergic receptor mechanisms in the mature organism (12), glucocorticoids might thus provide an initiating factor in the shift in adrenergic reactivity. Accordingly, the current study examines the effects of dexamethasone on adrenergic receptor subtypes in neonatal rat liver during the transition from 13to txl-receptor dependence, as well as in earlier phases.

Sprague-Dawley rat dams and pups (Zivic-Miller Laboratories, Allison Park, PA) were housed with free access to food and water. Pups received 1 mg/kg s.c. of dexamethasone phosphate (Merck Sharp and Dohme, Rahway, NJ) either on postnatal days 7, 8 and 9 or on days 13, 14 and 15, whereas controls received an equivalent volume (1 ml/kg) of saline vehicle on the same schedule. Just before each injection, pups were randomized within their respective treatment 0024-3205/91 $3.00 +.00 Copyright (c) 1991 Pergamon Press plc

1060

Adrenergic Receptor Development

Vol. 48, No. 11, 1991

groups and redistributed to the nursing dams with litter size maintained at 8 to 12 pups. Randomization was repeated before each experiment and, for each, pups were chosen from several different cages. Animals were weighed and decapitated, and biochemical analyses were carded out on fresh tissue. Weaning occurred at 23 days. Liver membrane fractions were prepared by the method of Witkin and Harden (13). Tissues were weighed and homogenized (Polygon) in 39 volumes of ice-cold Tris-HCl (pH 7.5), filtered through two layers of gauze, and sedimented at 40,000 x g for 15 min. The pellets were washed twice by resuspension (Polytron) in homogenization buffer and resedimentation. The final pellets were dispersed with a smooth glass homogenizer fitted with a Teflon pestle in 3 volumes (based on original wet weight of tissue) of 250 mM sucrose, 2 mM MgC12 and 50 mM Tris-HC1 (pH 7.5) and used for radioligand binding and protein analysis (14). Both cq- and fl-adrenergic receptor binding capabilities were assessed by methods described in earlier publications (8,15,16). The overall strategy was to examine binding at a single ligand concentration in preparations from every animal. At several points of maximal alterations caused by dexamethasone, this procedure was followed in a separate cohort by full Scatchard analyses on a subsequent preparation pooled from 5 animals selected from 1 or 2 litters, in order to identify whether binding alterations resulted from changes in Kd or Bmax. For cq-receptors in individual animals, 2.2 nM [3H]prazosin (DuPont Medical Products, Wilmington, DE) was incubated for 50 min at 4 ° C with tissue membrane preparations in a medium consisting of 50 mM Tris-HC1 (pH 7.5) and 10 mM MgC12; the incubations were stopped by dilution with 5 ml of ice-cold buffer followed by rapid vacuum filtration onto Whatman GF/C filters, which were then washed twice with 3 ml of buffer and counted. Nonspecific binding was defined as binding in the presence of 10 ~tM phentolamine (Sigma Chemical Co., St. Louis, MO). The subsequent Scatchard analyses covered the range from 1 - 10 nM [3H]prazosin. For measurement of B-receptor binding capabilities in individual animals, 67 pM [~Zq]pindolol (DuPont) was incubated with membrane preparations in a medium consisting of 20 mM TrisHC1 (pH 7.5), 145 mM NaC1, 2 mM MgCI2 and 1 mM sodium ascorbate; incubations lasted 20 min at room temperature and were stopped by dilution with 3 ml of ice-cold buffer, followed by filtration, washing and counting. Nonspecific binding was determined in the presence of 100 IxM dl-isoproterenol (Sigma). Scatchard analyses covered the concentration range from 30 1000 pM [lZq]pindolol. It should be noted that, although individual values within a given cohort remain within 5-10% of the mean, across different cohorts hepatic adrenergic receptors in developing rats can show substantial variations in both Kd and Bmax. Other investigators have found differences as large as 50% in sequential experiments on different cohorts (8,17). Accordingly, absolute values between different control cohorts in the current study do not necessarily correspond one-to-one, although the inter-cohort differences seen here are generally smaller (=30%) than those reported by others; the inclusion of a matched control cohort in each study is therefore a necessity in identifying effects attributable to dexamethasone. Data are presented as means and standard errors. Statistical comparisons utilized ANOVA (data log transformed whenever variance was heterogeneous): longitudinal differences for each dosage regimen were evaluated by two-way ANOVA (factors of age and drug treatment group), whereas comparisons between regimens were conducted by three-way ANOVA (factors of regimen, days post-treatment and drug treatment group). Duncan's Multiple Range Test was used post-hoc to assess differences attributable to dexamethasone at individual age points, only where the ANOVA indicated an interaction of age x drug treatment; where a main treatment effect was not interactive with age, the post-hoc testing was not undertaken and only the overall treatment effect was reported. Scatchard plots were fitted to a single-site model, using linear least squares analysis. Significant differences for all tests were evaluated at both the levels of p < 0.05 and p < 0.01. Results Administration of dexamethasone affected both body and liver weights (Fig. 1). With either treatment regimen, the body weights in the dexamethasone group were 10-20% lower than in

Vol. 48, No. ii, 1991

Adrenergle Receptor Development

1061

controls. Liver weights were also affected, but in a biphasic manner: initial elevations were succeeded by deficits comparable to those seen for body weight. The initial increases did not necessarily represent enhanced liver growth: protein concentrations were reduced by the same proportion as the increase in tissue weight (data not shown), suggesting that changes in tissue water content contributed to the enhanced weight. Accordingly, the results of receptor binding studies were evaluated as binding per mg protein so as not to include contributions from altered tissue weight. Consistent with previous work (8), liver a~-receptors in control animals rose with the approach of weaning (end of third postnatal week) and in the immediate postweaning period (Fig. 2). With either treatment regimen, dexamethasone precipitated an initial, marked increase in binding which subsequently returned to normal values. Scatchard analysis performed at the age of peak effect (postnatal day 16 with the later treatment group) indicated effects attributable solely to changes in receptor concentration (Bmax) and not affinity (Kd). During the period immediately preceding and following weaning, fi-receptor binding in control animals showed the 50% maturational decrease as identified earlier (8) (Fig. 3). Administration of dexamethasone caused significant decreases in the time period directly following treatment. However, in this case, the patterns of effects differed between the two treatment regimens. With the postnatal day 7, 8, 9 regimen, the initial effect was relatively small, intensified by the fourth day after the last injection, and then reestablished the normal developmental pattern. In contrast, dexamethasone given on days 13, 14 and 15 produced an accelerated decline in 13-receptor binding that was not subsequently reversed: binding was already maximally reduced by the day after the last drug treatment and these animals reached and maintained the mature pattern of low B-receptor levels much earlier than in controls. Scatchard analyses performed on postnatal day 13 for the early treatment cohort and on day 16 for the later treatment, indicated that changes in binding were again attributable only to decreases in receptor concentrations, not affinity (Fig. 3). Because these results suggested the existence of a late postnatal critical period in which the maturational pattern of fl-receptor decline could be "programmed" by glucocorticoids (as distinct from reversible pharmacological effects), we performed an additional experiment in which dexamethasone was given even earlier. Pregnant rats were treated on gestational days 17, 18 and 19 and hepatic B-receptor binding evaluated in fetuses on gestational day 20. In this case, the treatment failed to reduce [125I]pindolol binding; in fact, binding was slightly increased: control, 18.2 + 0.7 fmol/mg protein (fetuses from 6 dams); dexamethasone, 21.6 + 1.3 fmol/mg protein (fetuses from 6 dams, p < 0.05). Discussion.. Results obtained in this study indicate that gIucocorticoids participate in the developmental transition of hepatic adrenergic receptors. The effects of dexamethasone can be divided into two distinct categories. First, the ontogenetic increase of txrreceptors is fostered by administration of the glucocorticoid. Nevertheless, this does not appear to be a true shift in the maturational pattern of tx~-receptors, but rather a short-term pharmacological manipulation of receptor expression. Accordingly, a~-receptors always returned to the normal developmental pattern immediately upon discontinuing dexamethasone. In contrast, the effects exerted on fl-adrenergic receptors did appear to represent a shift in the programming of receptor maturation. When treatment occurred in the week preceding the normal switchover from B-to tx~-receptors, dexamethasone evoked a premature loss of the B-subtype without subsequent resumption of the normal pattern. Glucocorticoid-induced loss of the ~-subtype displayed a critical period of sensitivity in that irreversible effects were not elicited by administration on days 7, 8 and 9; indeed, when given before birth, dexamethasone failed to evoke any decrease in I~-receptors whatsoever. Taken together, these results indicate that the ability of glucocorticoids to program the ontogenetic disappearance of hepatic B-receptors reaches a peak of sensitivity just before the period in which natural adrenocortical function matures (10,11). The hypothesis that the development of adrenal steroid secretion provides a normal maturational cue for hepatic

1062

Adrenergic Receptor Development

Vol. 48, No. Ii, 1991

Treatment on Days 13, 14, 15

Treatment on Days 7, 8, 9

,,,

-# E ~-, J::

6O I

50

--" Control -- z~-- Dex

E

40-

~

0-

~'

2O-

~

4~

.Z&'"

70 60

.~ 4o

.......

I

m

~~. ,,-'~'"

.

A"

30 "

Dex < Con p < 0.01 i

Dex < Con p < 0.01

....... I ,

I

1

2'0 2'2 24 &

lo 12 14 1'6 1'8

Postnatal Age (days)

Postnatal Age (days)

Treatment on Days 13, 14, 15

Treatment on Days 7, 8, 9 ........

~"

E

2

............

,,,

~-Controi -z~-- Dex

/ ~

so

ZX"

10-

--" Control "

8O

I

, , ,, ,,,

I

/~

~'

E

-

,o

4

,

,

or"ro'I

.

-- -Dex

I

"°

3

1.5

2

.4 0.5

"'" ,,4

'

I

I

I

10

12

14

16

I

¢

1

Dex x Age p < 0.01 I

s" "" ""

t

Q)

Dex x Age p < 0.01 i

18 20

!

i

I

I

I

1

16

18

20

22

24

26

Postnatal Age (days)

Postnatal Age (days) Fig. 1

Effects of dexamethasone (Dex) on the development of body and liver weights. Data represent means and standard errors from 8-9 rats in each treatment group at each age. For treatment on either days 7, 8 and 9 or days 13, 14 and 15, ANOVA indicates a significant overall deficit in body weight in the dexamethasone group compared to controls (Con), as well as age-dependent alterations in liver weight. Individual ages at which the dexamethasone group differs from controls were tested only where a treatment x age interaction was found, and are indicated by asterisks (p < 0.05) and daggers (p < 0.01).

post-hoc

adrenoceptor expression is supported by preliminary evidence that neonatal adrenalectomy delays the switch from I~- to eta-receptors (18). The failure of glucocorticoids to evoke a similar, irreversible shift in the programming of eta-receptor ontogeny suggests that other factors, possibly the development of sympathetic innervation (7), provides the major input for that subtype.

Vol. 48, No. 11, 1991

Adrenergic Receptor Development

Treatment ,

70

60

on D a y s 7, 8, 9 ,,

,,,, ,,

~ ,

I --*-c°ntr°'

"

:

T

I

I - - ' - - Dex

t k,

/.

Dex>Con p<0.05

,,

1063

z~

50

40 II

|

10

12

, ,=,,= ,,

•

I

i

•

14

16

18

20

Postnatal Age (days) T r e a t m e n t on Days 13, 14, 15 ,,,

"',, ',

~ Ioo

~. ~)

90

(~

80

',..

~

,,,

~Control

120

£

, , , , ,

Dex > Con p < 0.01 Dex x A g e p < 0.01

.-

70 60 •

•

|

I

I

BI

16

18

20

22

24

26

Postnatal Age (days) Effects of dexamethasone on hepatic al-receptors. Data represent means and standard errors from 8-9 rats in each treatment group at each age. With either regimen, ANOVA indicates a significant overall increase in binding attributable to dexamethasone treatment as well as age-dependent effects; asterisks (p < 0.05) and daggers (p < 0.01) indicate individual ages at which the dexamethasone groups differ from corresponding controls. The pattern of effect of dexamethasone is not statistically different between the two regimens (equivalent proportional changes, three-way ANOVA, no significant interaction of regimen x drug treatment or regimen x post-treatment day x drug treatment). Inset shows Scatchard analysis of membrane preparations pooled from 5 animals onpostnatal day 16 after treatment on days 13, 14 and 15. Calculated values: control Kd = 1.4 + 0.2 nM, Bmax = 33 + 1 pmol/mg protein; dexamethasone Kd = 1.1 + 0.1 nM (N.S.), Bmax = 81 + 1 fmol/mg protein (p < 0.01).

1064

Vol. 48, No. ii, 1991

Adrenerglc Receptor Develoment

Treatment

o n D a y s 7, 8, 9

....

10

I

,u

'1~"' Day 13

I

(3 Ea.

Dox
Control I --zx•

_

10

•

,

Treatment

"1

I

l

I

_1

14 16 18 Postnatal A g e (days) 12

i

10

p<0.01

Dex x Age p < 0.01

Dex

on

Days

13, 14, 15 .

|ll

--" (~°ntr°l i

~.~.D,,,,

.

4

20

B

D e x < C o n p<0.01 Dex x Age p < 0.05 22 24 16 18 20 Postnatal Age (days) I

•

|

._

|

....

I

•

26

Fi~. 3 Effects of dexamethasone on hepatic fl-receptors. Data represent means and standard errors from 8-9 rats in each treatment group at each age. With either regimen, ANOVA indicates a significant overall decrease in binding attributable to dexamethasone treatment as well as age-dependent effects; asterisks (p < 0.05) and daggers (p < 0.01) indicate individual ages at which the dexamethasone groups differ from corresponding controls. The pattern of effect of dexamethasone also differs significantly between the two treatment regimens (three-way ANOVA, interaction of regimen x post-treatment day x drug treatment, p < 0.05). Inset shows Scatchard analyses of membrane preparations pooled from 5 animals on postnatal day 13 (dexamethasone treatment on days 7, 8 and 9) or day 16 (treatment on days 13, 14 and 15). Calculated values" treatment on days 7, 8, 9, control Kd = 84 + 3 pM, Bmax = 25.0 __.0.3 fmol/mg protein; dexamethasone, Kd = 75 + 6 pM (N.S.), Brnax = 11.8 + 0.3 fmol/mg protein ( p < 0.01); treatment on days 13, 14, 15, control Kd = 194 + 27 pM, Bmax = 9.7 + 0.4 fmol/mg protein; dexamethasone, Kd = 164 + 6 pM (N.S.), Bmax = 6.8 -+ 0.1 fmol/mg protein (p < 0.01).

Vol. 48, No. ii, 1991

AdrenergSe Receptor Development

1065

These results are also of potential significance for clinical conditions involving fetal/neonatal glucocorticoid exposure, notably in the therapy of Respiratory Distress Syndrome or after enhanced steroid secretion associated with maternal stress (19-21). The premature loss of 13receptors and their replacement by oq-receptors would interfere with hepatic glucose metabolism early in development, when sympathetic nerves have not developed and adrenomedullary epinephrine represents the primary source of catecholamines (7,22). In light of the transient nature of steroid promotion of etl-receptor development and the potentially lasting depletion of 13receptors, the liver may be rendered hyporesponsive to situations requiting rapid glucose mobilization during metabolic stress, such as that experienced during delivery (22). Glucocorticoids may thus contribute to increased perinatal risk. Acknowledgements v

Supported by USPHS HD-09713. The authors thank Dr. C.M. Kuhn for helpful suggestions and comments and S.E. Lappi and M.I. Tayyeb for technical assistance. References 1. B.W. PALMISANO, P.S. CLIFFORD, R.L. COON, J.L. SEAGARD, R.G. HOFFMANN and J.P. KAMPINE, Pediat. Res. 27 148-152 (1990). 2. B.B. HOFFMAN, T. MICHEL, D.M. KILPATRICK, R.J. LEFKOWITZ, M.E.M. TOLBERT, H. GILMAN and J.N. FAIN, Proc. Natl. Acad. Sci. USA 77 4569-4573 (1980). 3. M. AGGERBECK, G. GUELLARN and J. HANOUNE, Biochem. Pharmacol. 29 643645 (1980). 4. P. SHERLINE, H. EISEN and W. GLINSMANN, Endocrinology 94 935-939 (1974). 5. M.L.J. MONCANY and C. PLAS, Endocrinology 107 1667-1675 (1980). 6. W.W. LAUTT, Med. Hypoth. 5 1287-1296 (1979). 7. T.A. SLOTKIN, In Developmental Neurobiology of the Autonomic Nervous System, ed. P.M. Gootman ed., pp. 97-133, Humana Press, Clifton, NJ (1986). 8. M.K. McMILLIAN, S.M. SCHANBERG and C.M. KUHN, J. Pharmacol. Exp. Ther. 227 181-186 (1983). 9. N.G. MORGAN, P.F. BLACKMORE and J.H. EXTON, J. Biol. Chem. 258 51035109 (1983). 10. S. MIYABO, K.I. YANAGISAWA, E. OOYA, T. HISADA and S. KISHIDA, Endocrinology 106 636-642 (1980). 11. W.H. LAMERS and P.G. MOOREN, Mech. Aging Dev. 15 93-118 (1981). 12. A.O. DAVIES and R.J. LEFKOWITZ, Ann. Rev. Physiol. 46 119-130 (1984). 13. R.M. WITKIN and T.K. HARDEN, J. Cyclic Nucleotide Res. 7 235-246 (1981). 14. O.H. LOWRY, N.J. ROSEBROUGH, A.L. FARR and R. RANDALL, J. Biol. Chem. 193 265-270 (1951). 15. T.A. SLOTKIN, L. ORBAND, T. COWERY, R.J. KAVLOCK and J. BARTOLOME, Toxicol. Lett. 35 285-295 (1987). 16. T.A. SLOTKIN, W.L. WHITMORE, L. ORBAND-MILLER, K.L. QUEEN and K. HAIM, J. Pharmacol. Exp. Ther. 243 101-109 (1987). 17. C.M. KUHN, M.K. McMILLIAN, G.E. EVONIUK and S.M. SCHANBERG, J. Pharmacol. Exp. Ther. 235 361-367 (1985). 18. M.K. McMILLIAN, S.M. SCHANBERG and C.M. KUHN, Fed. Proc. 42 383 (1983). 19. M.E. AVERY, Brit. Med. Bull. 31 13-17 (1975). 20. P.L. BALLARD, Hormones and Lung Maturation, Springer-Verlag, Berlin (1986). 21. E.M. KUDLACZ, H.A. NAVARRO, J.P. EYLERS, S.S. DOBBINS, S.E. LAPPI and T.A. SLOTKIN, J. Dev. Physiol. 12 129-134 (1989). 22. H. LAGERCRANTZ and T.A. SLOTKIN, Sci. Amer. 254 100-107 (1986).