- Email: [email protected]
Fetal terbutaline exposure causes selective postnatal increases in cerebellar α-adrenergic receptor binding
Life Sciences, Vol. 47, pp. 2051-2057 Printed in the U.S.A.
Pergamon Press
FETAL TERBUTALINE EXPOSURE CAUSES SELECTIVE POSTNATAL INCREASES IN CEREBELLAR ¢z-ADRENERGIC R E C E P T O R BINDING T.A. S1otldn, E.M. Kudlacz, S.E. Lappi, M.I. Tayyeb and F.J. Seidler Department of Pharmacology Duke University Medical Center Durham, North Carolina 27710 USA (Received in final form September 24, 1990)
Summary 13-Adrenergic agonists used in therapy of premature labor and asthma cross the placenta and can affect development of the fetal nervous system. In the current study, pregnant rats were given 10 mg/kg of terbutaline on gestational days 17, 18 and 19 and adrenergic receptor binding capabilities examined in brain regions of the offspring. Despite the absence of body or brain growth impairment, selective increases were seen postnatally in cerebellar el- and oa-receptor subtypes, whereas the same receptor populations were decreased by small amounts in cerebral cortex and midbrain + bralnstem. I~-Adrenergic receptors showed little or no change in any region. The regional and subtype selectivity are compatible with primary deficits in the development of noradrenergic projections to the cerebellum identified in previous studies and provide further evidence that therapeutic use of [~-adrenergic agonists may produce neurobehavioral teratology. The therapeutic management of premature labor and maternal asthma has resulted in prenatal exposure of a large number of infants to 13-adrenergic agonists. The ability of these agents to cross the placenta has a beneficial effect on the fetus because 132-receptorsin the lung promote synthesis and secretion of surfactant, thus lessening the probability of neonatal respiratory distress syndrome (1-3). However, recent studies have shown that neural signals mediated through the same receptors also control the differentiation of developing adrenergic target ceils; hence, ~-agonists disrupt subsequent cellular development in the lung, resulting in deficits in the total number of ceils (4,5). A similar trophic role of adrenergic receptors in the control of cell replication and differentiation has been demonstrated in the central nervous system (6-8): during critical perinatal periods, B-receptors become closely coupled to regulation of omithine decarboxylase, an enzyme that has a major role in controlling cell differentiation (9,10); B-receptors can also directly mediate the termination of DNA synthesis (4,7). Thus, the therapeutic use of 13-agonists may be accompanied by alterations of neural development and function in the offspring. We have recently shown that late gestational exposure of fetal rats to terbutaline selectively compromises the development of cerebellar [3H]norepinephrine synaptosomal uptake, a biochemical marker for presynaptic noradrenergic terminals (11). The current study extends these findings to include evaluations of ontogeny of noradrenergic receptors in three brain regions that differ in their patterns of cellular maturation and their sensitivity to [3-receptor regulation of omithine decarboxylase and cell replication (7-10,12). Me~h0ds Animal treatments. Timed pregnant Sprague-Dawley rats (Zivic-Miller Laboratories, Allison Park, PA) were shipped by climate-controlled truck (total transit time, 12 hr). After arrival, animals were housed individually in breeding cages and allowed food and water ad libitum. On gestational days 0024-3205[90 $3.00 +.00 Copyright (e) 1990 Pergamon Press plc
2052
Fetal Terbutaline Exposure
Vol. 47, No. 22, 1990
17, 18 and 19, dams were given 10 mg/kg s.c. of terbutaline sulfate (Geigy Pharmaceuticals, Ardsley, NY); controls received an equivalent volume (1 ml/kg) of saline vehicle. This maternal dose regimen has been shown previously to stimulate fetal 13-adrenergic receptors (3,10,13). After birth, pups were randomized within their respective treatment groups and redistributed to the nursing dams with litter size maintained at 8-12 pups. Randomization was carded out again just prior to each experiment and in addition, pups were selected from different cages and sex-matched; each litter contributed only a single sample for any individual experiment. Pups were killed by decapitation and analyses carded out on tissues from individual animals except during the first postnatal week, where tissues had to be pooled from 2-3 animals for each determination. Brains were dissected into three regions: blunt cuts were made through the cerebellar peduncles, whereupon the cerebellum (including fiocculi) was lifted from the underlying tissue; a cut was then made rostral to the thalamus to separate "cerebral cortex" from "midbrain + brainstem." The tissues were weighed, frozen on dry ice and stored at -45°; preliminary studies indicated no degradation of receptor binding characteristics during storage. Membrane preparation and assavs. Tissues were thawed and homogenized (Polytron) in 39 volumes of ice-cold buffer containing 145 mM NaC1, 2 mM MgC12, 20 mM "Iris (pH 7.5) and sedimented at 40,000 x g for 15 rain. The pellets were washed twice by resuspension (Polytron) in homogenization buffer followed by recentrifugation. The final pellet was dispersed with a Teflonto-smooth-glass homogenizer in 4 volumes (based on original wet weight of tissue) of 250 mM sucrose, 2 mM MgC12, 50 mM Tris (pH 7.5) and the suspension was then used for ligand binding (see below) and for protein analysis (14). The properties of this membrane preparation in developing rat tissues have been described in detail previously (15,16). Radioligands were incubated with the tissue membrane preparations in a total volume of 0.25 ml. Incubations were stopped by dilution with 3 ml of ice-cold buffer, followed by rapid vacuum filtration onto Whatman GF/C filters, which were then washed with additional buffer. Non-specific binding was defined as binding of radioligand in the presence of an excess concentration of a specific displacing agent: 5 IxM phentolamine (Ciba Pharmaceuticals, Summit, NJ) for cq- and ct2receptors, and 100 ~tM dl-isoproterenol sulfate (Sigma Chemical Co., St. Louis, MO) for I~receptors. Studies were conducted at a single ligand concentration approximating the Kd for each receptor subtype (16,17), a strategy that was adopted because of the need to examine all three receptor subtypes in as many as 48 separate tissue preparations at each age point, and because of the impracticality of running full Scatchard analyses on the small amounts of tissue available; indeed, because of limitations in the amount of tissue, eq-receptor binding was not evaluated in cerebellum prior to postnatal day 11. A subsaturating concentration was chosen for each ligand so as to permit detection of changes in binding characteristics attributable either to Kd or Bmax, but the single concentration technique does not permit distinction between the two possible mechanisms. cq-Receptor binding was determined with [3I-I]prazosin (DuPont Medical Products, Wilmington, DE; specific activity 19 Ci/mmol). Aliquots of membrane preparation (50 to 500 Ixg of protein, depending upon age and region) were incubated with 2.2 nM radioligand in 10 mM MgC12, 50 mM Tris-HC1 (pH 7.5), at 4 ° for 50 min. Non-specific binding of [3H]prazosin generally constituted 1530% of total binding, depending on age and brain region, et2-Receptor binding was evaluated similarly, using 2.5 nM [3H]rauwolscine (DuPont; specific activity 79 Ci/mmol), and a 20 min incubation at room temperature; non-specific binding of [3I-I]rauwolscine was generally 25-40%. 13Adrenergic receptor binding was evaluated with 67 pM [l~I]pindolol (DuPont; specific activity 2200 Ci/mmol), aliquots of the membrane preparation corresponding to 20-50 ~tg of protein and an incubation medium consisting of 145 mM NaC1, 2 mM MgC12, 1 mM sodium ascorbate, 20 mM Tris (pH 7.5); non-specific binding was typically 5-20% of the total. Statistics. Data are presented as means and standard errors. Initial statistical comparisons were conducted by three-way ANOVA (data log-transformed whenever variance was heterogeneous) using factors of age, brain region and fetal treatment. Wherever a significant interaction of treatment with brain region was identified, a two-way ANOVA was then conducted to identify the region(s) affected; individual ages points at which the terbutaline group differed from control were then identified post-hoc with Duncan's Multiple Range Test. It should be noted that, when only a main treatment effect could be identified without selectivity for region or age, the lower-order tests of
Vol. 47, No. 22, 1990
Fetal Terbutaline Exposure
2053
individual regions or ages were not conducted. Significance for ANOVA was evaluated at the levels o f p < 0.05 andp < 0.01, and atp < 0.05 forpost-hoc tests. Results and Discussion In keeping with earlier reports (3,5,11), maternal terbutaline treatment had little or no deleterious effect on maternal weight gain, proportions of dams delivering pups, litter size, neonatal viability, or on postnatal body and brain region weights (data not shown). Nevertheless, there were significant alterations in the developmental profiles of both a~- and et2-adrenergic receptor subtypes. For cqreceptors, there was a net increase in binding in the cerebellum, whereas binding tended to be decreased by small or variable amounts in the midbrain + brainstem and cerebral cortex (Fig. 1). A similar pattern was seen with q-receptor binding, namely an overall increase in the cerebellum with little or no change elsewhere (Fig. 2). Although binding of [~I]pindolol to 13-receptors showed a statistically significant overall decrease elicited by fetal terbutaline exposure, the magnitude of the effect was quite small and did not exhibit any selectivity for region or age (Fig. 3). Recovery of membrane protein was unaffected by fetal terbutaline exposure and accordingly the same results were obtained with receptor binding calculated as fmol/mg protein (data not shown). Thus, even when body or brain region growth is unaffected, fetal terbutaline exposure elicits a regionally-selective increase in binding capabilities of the ctl- and o~-adrenergic receptor subtypes. As a fwst possibility, terbutaline could directly disrupt development of adrenergic target cells within the cerebellum (10). I~-Receptors are involved in the control of cell replication and differentiation in the central nervous system (6,7), but the majority of cerebellar I~-receptors are located on glia, not neurons (18). Cerebellar gliogenesis is primarily a late postnatal event, and thus the sensitivity of cerebellar DNA synthesis to prenatal or early postnatal exposure to a l~-agonist is relatively low (7,12). In fact, the midbrain + brainstem region, not the cerebellum, appears to be the region most susceptible to disruption of cell replication by 13-agonists (7). Furthermore, global interference with differentiation would be expected to show similar effects on all adrenergic receptor subtypes located on the affected cells, not specific actions directed toward a-receptors, as found here. Indeed, the specific archetypical patterns of each receptor subtype can be used as a specific test of the cell replication/differentiation hypothesis. As found previously (16), ct2-receptors are transiently overexpressed in both midbrain + brainstem and cerebellum and decay as maturation proceeds. Shifting the time course of cellular differentiation would thus be expected to shift the time course of receptor development; instead, fetal terbutaline exposure altered only the magnitude of receptor binding, not the timing of receptor expression and disappearance. Similarly, the overall timing of receptor acquisition of the subtypes and regions showing monotonic maturational increases was not perturbed by terbutaline exposure. It is therefore unlikely that the increases in cerebellar cq- and Ctz-receptor binding can be attributed solely to disruption of cellular replication/differentiation patterns by terbutaline. Instead, our data suggest that the effects of terbutaline on receptor sites are secondary to interference with development of presynaptic noradrenergic function. The regimen of fetal terbutaline exposure used here has been shown to impair the ingrowth of noradrenergic projections specifically to the cerebellum (11), the same region showing the receptor changes; thus, receptor up-regulation is evidence that presynaptic input to the cerebellum is chronically subnormal. The selectivity toward areceptor subtypes is also compatible with this view. The majority 13-receptor subtype in the cerebellum, the 1~2-receptor (19), is primarily (although not exclusively) glial (18), and can be regulated separately from neuronal a-receptor sites that are primarily juxtaposed to noradrenergic projections. Accordingly, even total destruction of noradrenergic inputs leads to no more than a 50% elevation in cerebellar 132-receptor binding sites (19), and it would therefore be surprising to find a robust effect with the smaller noradrenergic deficits elicited by prenatal terbutaline. A final point is that receptor expression during development, as found here and elsewhere, does not necessarily follow monotonically the development of noradrenergic projections, but rather exhibits transient peaks of activity (16). Two factors contribute to these differences between ontogeny of projections and of receptors: first, regions may express receptor subtypes that do not exist in the adult (20) but that nevertheless play an important physiological role during a specific critical period (8). Second, the anatomical development of projections does not take into account the fact that nerve activity can change drastically during development, typically involving a transient "surge" and
2054
Fetal Terbutaline Exposure
Vol. 47, No. 22, 1990
~I-RECEPTORS Cerebellum 1600 1200 800 - o-TER 400
I
CON < TER, p < 0.05 I
I
10
I
I
20 30 Postnatal Age (days)
40
Cerebral Cortex 6000 5000 4000 3000 2000
j
I -°-TER I
1000
CON > TER, p < 0.01 I
I
10
I
20 30 Postnatal Age (days)
I
40
Midbrain + Brainstem 6000 5000 4000 3000
I-° TE" I
2000 1000
treatment x age interaction, p < 0.05 i
10
I
l
20 30 Postnatal Age (days)
I
40
Fi~. 1 Effects of fetal terbutaline exposure on binding-of [3H]prazosin to al-receptors in brain regions. Data represent mean and SE of 6-8 determinations in each group at each age; where SE bars are not visible, they are smaller than the data symbol. ANOVA indicates a significant, regionally-dependent effect of terbutaline on binding (treatment x region interaction, p < 0.01); separate tests of each region are indicated above. For regions showing a significant treatment x age interaction, asterisks indicate individual points at which the terbutaline group differs significantly from control.
Volo 47, No. 22, 1990
Fetal Terbutaline Exposure
2055
~2-RECEPTORS Cerebellum 2000 1500
~
]--=--CON I
z
-
~
0-TER
1000 500
CON
I
10
I
20 30 Postnatal Age (days)
I
40
Cerebral Cortex 2500 2000 1500 - ' I - CO N
1000 500 t
I
10
I
l
20 30 Postnatal Age (days)
40
Midbrain + Brainstem
1200
800 l'-I-CON
- o-TER
400 treatment l
10
I
x age interactk:~,p < 0.01 I
I
20 30 Postnatal Age (days)
l
40
Fi~. 2 Effects of fetal terbutaline exposure on binding of [3H]rauwolscine to o~-receptors in brain regions. Data represent means and standard errors of 6-8 determinations in each group at each age; where SE bars are not visible, they are smaller than the data symbol. ANOVA indicates a significant, age- and regionally-dependent effect of terbutaline on binding (treatment x region interaction, p < 0.01; treatment x region x age interaction, p < 0.01); separate tests of each region are indicated above. For regions showing a significant treatment x age interaction, asterisks indicate the individual points at which the terbutaline group differs significantly from control.
2056
Fetal Terbutallne Exposure
Vol. 47, No. 22, 1990
[}-RECEPTORS Cerebellum 600 500 400 300 200 100 |
I
I
I
10
20 Postnatal
30
40
Age (days)
Cerebral Cortex 1000 800
• 600
?
i_o_co:I
400 200
10
20 30 Postnatal Age (days)
40
Midbrain + Brainstem 250 200 (3
150 100 50 I
10
I
I
20 30 Postnatal Age (days;)
I
40
Fi~. 3 Effects o f fetal terbutaline exposure on binding of [125I]pindolol to ~-receptors in brain regions. Data represent means and standard errors of 6-8 determinations in each group at each age; where SE bars are not visible, they are smaller than the data symbol. A N O V A indicates a significant overall reduction caused by terbutaline (main treatment effect, p < 0.05), without selectivity for region or age (no interactions of treatment x region, treatment x age or treatment x region x age); thus noposthoc testing of individual regions or ages was undertaken.
Vol. 47, No. 22, 1990
Fetal Terbutaline
Exposure
2057
decline into adulthood (21,22). The identification of specific mechanisms by which terbutaline alters receptor development will thus require examination of patterns of gene expression of receptor proteins, as well as elucidation of potential alterations in neural tone. In the latter case, we have already shown that peripheral noradrenergic neurons exhibit lowered activity after prenatal terbutaline exposure (23), and it is likely that similar changes occur in the central nervous system. Thus, clarifying the neurotransmitter pathways and receptor types affected by prenatal exposure to terbutaline can serve as a focus for characterization of potential neurobehavioral teratogenesis associated with maternal I]-adrenergic agonist therapy. Acknowledgement Supported by USPHS HD-09713. References 1. M.E. AVERY, Brit. Med. Bull. 31 13-16 (1975). 2. D. WARBURTON, L. PARTON, S. BUCKLEY, L. COSICO, G. ENNS and T. SALUNA, Pediat. Res. 24 166-170 (1988). 3. E.M. KUDLACZ, H.A. NAVARRO and T.A. SLOTKIN, J. Pharmacol. Exp. Ther. 250 236240 (1989). 4. T.A. SLOTKIN, W.L. WHITMORE, L. ORBAND-MILLER, K.L. QUEEN AND K. HAIM, J. Pharmacol. Exp. Ther. 243 101-109 (1987). 5. E.M. KUDLACZ, H.A. NAVARRO, J.P. EYLERS, S.E. LAPPI, S.S. DOBBINS and T.A. SLOTKIN, Pediat. Res. 25 617-622 (1989). 6. A. VERNADAKIS and D.A. GIBSON, Perinatal Pharmacolo~v: Problems and Priorities. J. Dancis and H.C. Hwang (eds), 65-76, Raven Press, New York (1974). 7. T.A. SLOTKIN, R. WINDH, W.L. WHITMORE and F.J. SEIDLER, Brain Res. Bull. 21 737-740 (1988). 8. C.P. DUNCAN, F.J. SEIDLER, S.E. LAPPI and T.A. SLOTKIN, Dev. Brain Res. 55 29-33 (1990). 9. T.A. SLOTKIN and J. BARTOLOME, Brain Res. Bull. 17 307-320 (1986). 10. G. MORRIS and T.A. SLOTKIN, J. Pharmacol. Exp. Ther. 233 141-147 (1985). 11. T.A. SLOTKIN, F.E. BAKER, S.S. DOBBINS, J.P. EYLERS, S.E. LAPPI and F.J. SEIDLER, Brain Res. Bull. 23 263-265 (1989). 12. S. REINIS and J.M. GOLDMAN, The Develonment of the Brain: Biological and Functional Persnectives. Charles C. Thomas, Springfield, IL (1980). 13. E.M~ KUDLACZ, H.A. NAVARRO, J.P. EYLERS and T.A. SLOTKIN, J. Pharmacol. Exp. Ther. 252 42-50 (1990). 14. O.H. LOWRY, N.J. ROSEBROUGH, A.L. FARR and R. RANDALL, J. Biol. Chem. 193 265-270 (1951). 15. R.M. WlTKIN and T.K. HARDEN, J. Cyclic Nucleotide Res. 7 235-246 (1981). 16. J.V. BARTOLOME, R.J. KAVLOCK, T. COWDERY, L. ORBAND-MILLER and T.A. SLOTKIN, Neurotoxicology 8 1-14 (1987). 17. F.J. SEIDLER, J.M. BELL and T.A. SLOTKIN, Pediat. Res. 27 191-197 (1990). 18. E. HOSLI and L. HOSLI, Neuroscience 7 2873-2881 (1982). 19. D. LORTON, J. BARTOLOME, T.A. SLOTKIN and J.N. DAVIS, Brain Res. Bull. 21 591600 (1988). 20. L.S. JONES, L.L. GAUGER, J.N. DAVIS, T.A. SLOTKIN and J . BARTOLOME, Neuroscience 15 1195-1202 (1985). 21. F.J. SEIDLER and T.A. SLOTKIN, Neuroscience 6 2081-2086 (1981). 22. H.A. NAVARRO, F.J. SEIDLER, J.P. EYLERS, F.E. BAKER, S.S. DOBBINS, S.E. LAPPI and T.A. SLOTKIN, J. Pharmacol. Exp. Ther. 251 894-900 (1989). 23. Q.-C. HOU and T.A. SLOTKIN, Pediat. Res. 26 554-557 (1989).
Pergamon Press
FETAL TERBUTALINE EXPOSURE CAUSES SELECTIVE POSTNATAL INCREASES IN CEREBELLAR ¢z-ADRENERGIC R E C E P T O R BINDING T.A. S1otldn, E.M. Kudlacz, S.E. Lappi, M.I. Tayyeb and F.J. Seidler Department of Pharmacology Duke University Medical Center Durham, North Carolina 27710 USA (Received in final form September 24, 1990)
Summary 13-Adrenergic agonists used in therapy of premature labor and asthma cross the placenta and can affect development of the fetal nervous system. In the current study, pregnant rats were given 10 mg/kg of terbutaline on gestational days 17, 18 and 19 and adrenergic receptor binding capabilities examined in brain regions of the offspring. Despite the absence of body or brain growth impairment, selective increases were seen postnatally in cerebellar el- and oa-receptor subtypes, whereas the same receptor populations were decreased by small amounts in cerebral cortex and midbrain + bralnstem. I~-Adrenergic receptors showed little or no change in any region. The regional and subtype selectivity are compatible with primary deficits in the development of noradrenergic projections to the cerebellum identified in previous studies and provide further evidence that therapeutic use of [~-adrenergic agonists may produce neurobehavioral teratology. The therapeutic management of premature labor and maternal asthma has resulted in prenatal exposure of a large number of infants to 13-adrenergic agonists. The ability of these agents to cross the placenta has a beneficial effect on the fetus because 132-receptorsin the lung promote synthesis and secretion of surfactant, thus lessening the probability of neonatal respiratory distress syndrome (1-3). However, recent studies have shown that neural signals mediated through the same receptors also control the differentiation of developing adrenergic target ceils; hence, ~-agonists disrupt subsequent cellular development in the lung, resulting in deficits in the total number of ceils (4,5). A similar trophic role of adrenergic receptors in the control of cell replication and differentiation has been demonstrated in the central nervous system (6-8): during critical perinatal periods, B-receptors become closely coupled to regulation of omithine decarboxylase, an enzyme that has a major role in controlling cell differentiation (9,10); B-receptors can also directly mediate the termination of DNA synthesis (4,7). Thus, the therapeutic use of 13-agonists may be accompanied by alterations of neural development and function in the offspring. We have recently shown that late gestational exposure of fetal rats to terbutaline selectively compromises the development of cerebellar [3H]norepinephrine synaptosomal uptake, a biochemical marker for presynaptic noradrenergic terminals (11). The current study extends these findings to include evaluations of ontogeny of noradrenergic receptors in three brain regions that differ in their patterns of cellular maturation and their sensitivity to [3-receptor regulation of omithine decarboxylase and cell replication (7-10,12). Me~h0ds Animal treatments. Timed pregnant Sprague-Dawley rats (Zivic-Miller Laboratories, Allison Park, PA) were shipped by climate-controlled truck (total transit time, 12 hr). After arrival, animals were housed individually in breeding cages and allowed food and water ad libitum. On gestational days 0024-3205[90 $3.00 +.00 Copyright (e) 1990 Pergamon Press plc
2052
Fetal Terbutaline Exposure
Vol. 47, No. 22, 1990
17, 18 and 19, dams were given 10 mg/kg s.c. of terbutaline sulfate (Geigy Pharmaceuticals, Ardsley, NY); controls received an equivalent volume (1 ml/kg) of saline vehicle. This maternal dose regimen has been shown previously to stimulate fetal 13-adrenergic receptors (3,10,13). After birth, pups were randomized within their respective treatment groups and redistributed to the nursing dams with litter size maintained at 8-12 pups. Randomization was carded out again just prior to each experiment and in addition, pups were selected from different cages and sex-matched; each litter contributed only a single sample for any individual experiment. Pups were killed by decapitation and analyses carded out on tissues from individual animals except during the first postnatal week, where tissues had to be pooled from 2-3 animals for each determination. Brains were dissected into three regions: blunt cuts were made through the cerebellar peduncles, whereupon the cerebellum (including fiocculi) was lifted from the underlying tissue; a cut was then made rostral to the thalamus to separate "cerebral cortex" from "midbrain + brainstem." The tissues were weighed, frozen on dry ice and stored at -45°; preliminary studies indicated no degradation of receptor binding characteristics during storage. Membrane preparation and assavs. Tissues were thawed and homogenized (Polytron) in 39 volumes of ice-cold buffer containing 145 mM NaC1, 2 mM MgC12, 20 mM "Iris (pH 7.5) and sedimented at 40,000 x g for 15 rain. The pellets were washed twice by resuspension (Polytron) in homogenization buffer followed by recentrifugation. The final pellet was dispersed with a Teflonto-smooth-glass homogenizer in 4 volumes (based on original wet weight of tissue) of 250 mM sucrose, 2 mM MgC12, 50 mM Tris (pH 7.5) and the suspension was then used for ligand binding (see below) and for protein analysis (14). The properties of this membrane preparation in developing rat tissues have been described in detail previously (15,16). Radioligands were incubated with the tissue membrane preparations in a total volume of 0.25 ml. Incubations were stopped by dilution with 3 ml of ice-cold buffer, followed by rapid vacuum filtration onto Whatman GF/C filters, which were then washed with additional buffer. Non-specific binding was defined as binding of radioligand in the presence of an excess concentration of a specific displacing agent: 5 IxM phentolamine (Ciba Pharmaceuticals, Summit, NJ) for cq- and ct2receptors, and 100 ~tM dl-isoproterenol sulfate (Sigma Chemical Co., St. Louis, MO) for I~receptors. Studies were conducted at a single ligand concentration approximating the Kd for each receptor subtype (16,17), a strategy that was adopted because of the need to examine all three receptor subtypes in as many as 48 separate tissue preparations at each age point, and because of the impracticality of running full Scatchard analyses on the small amounts of tissue available; indeed, because of limitations in the amount of tissue, eq-receptor binding was not evaluated in cerebellum prior to postnatal day 11. A subsaturating concentration was chosen for each ligand so as to permit detection of changes in binding characteristics attributable either to Kd or Bmax, but the single concentration technique does not permit distinction between the two possible mechanisms. cq-Receptor binding was determined with [3I-I]prazosin (DuPont Medical Products, Wilmington, DE; specific activity 19 Ci/mmol). Aliquots of membrane preparation (50 to 500 Ixg of protein, depending upon age and region) were incubated with 2.2 nM radioligand in 10 mM MgC12, 50 mM Tris-HC1 (pH 7.5), at 4 ° for 50 min. Non-specific binding of [3H]prazosin generally constituted 1530% of total binding, depending on age and brain region, et2-Receptor binding was evaluated similarly, using 2.5 nM [3H]rauwolscine (DuPont; specific activity 79 Ci/mmol), and a 20 min incubation at room temperature; non-specific binding of [3I-I]rauwolscine was generally 25-40%. 13Adrenergic receptor binding was evaluated with 67 pM [l~I]pindolol (DuPont; specific activity 2200 Ci/mmol), aliquots of the membrane preparation corresponding to 20-50 ~tg of protein and an incubation medium consisting of 145 mM NaC1, 2 mM MgC12, 1 mM sodium ascorbate, 20 mM Tris (pH 7.5); non-specific binding was typically 5-20% of the total. Statistics. Data are presented as means and standard errors. Initial statistical comparisons were conducted by three-way ANOVA (data log-transformed whenever variance was heterogeneous) using factors of age, brain region and fetal treatment. Wherever a significant interaction of treatment with brain region was identified, a two-way ANOVA was then conducted to identify the region(s) affected; individual ages points at which the terbutaline group differed from control were then identified post-hoc with Duncan's Multiple Range Test. It should be noted that, when only a main treatment effect could be identified without selectivity for region or age, the lower-order tests of
Vol. 47, No. 22, 1990
Fetal Terbutaline Exposure
2053
individual regions or ages were not conducted. Significance for ANOVA was evaluated at the levels o f p < 0.05 andp < 0.01, and atp < 0.05 forpost-hoc tests. Results and Discussion In keeping with earlier reports (3,5,11), maternal terbutaline treatment had little or no deleterious effect on maternal weight gain, proportions of dams delivering pups, litter size, neonatal viability, or on postnatal body and brain region weights (data not shown). Nevertheless, there were significant alterations in the developmental profiles of both a~- and et2-adrenergic receptor subtypes. For cqreceptors, there was a net increase in binding in the cerebellum, whereas binding tended to be decreased by small or variable amounts in the midbrain + brainstem and cerebral cortex (Fig. 1). A similar pattern was seen with q-receptor binding, namely an overall increase in the cerebellum with little or no change elsewhere (Fig. 2). Although binding of [~I]pindolol to 13-receptors showed a statistically significant overall decrease elicited by fetal terbutaline exposure, the magnitude of the effect was quite small and did not exhibit any selectivity for region or age (Fig. 3). Recovery of membrane protein was unaffected by fetal terbutaline exposure and accordingly the same results were obtained with receptor binding calculated as fmol/mg protein (data not shown). Thus, even when body or brain region growth is unaffected, fetal terbutaline exposure elicits a regionally-selective increase in binding capabilities of the ctl- and o~-adrenergic receptor subtypes. As a fwst possibility, terbutaline could directly disrupt development of adrenergic target cells within the cerebellum (10). I~-Receptors are involved in the control of cell replication and differentiation in the central nervous system (6,7), but the majority of cerebellar I~-receptors are located on glia, not neurons (18). Cerebellar gliogenesis is primarily a late postnatal event, and thus the sensitivity of cerebellar DNA synthesis to prenatal or early postnatal exposure to a l~-agonist is relatively low (7,12). In fact, the midbrain + brainstem region, not the cerebellum, appears to be the region most susceptible to disruption of cell replication by 13-agonists (7). Furthermore, global interference with differentiation would be expected to show similar effects on all adrenergic receptor subtypes located on the affected cells, not specific actions directed toward a-receptors, as found here. Indeed, the specific archetypical patterns of each receptor subtype can be used as a specific test of the cell replication/differentiation hypothesis. As found previously (16), ct2-receptors are transiently overexpressed in both midbrain + brainstem and cerebellum and decay as maturation proceeds. Shifting the time course of cellular differentiation would thus be expected to shift the time course of receptor development; instead, fetal terbutaline exposure altered only the magnitude of receptor binding, not the timing of receptor expression and disappearance. Similarly, the overall timing of receptor acquisition of the subtypes and regions showing monotonic maturational increases was not perturbed by terbutaline exposure. It is therefore unlikely that the increases in cerebellar cq- and Ctz-receptor binding can be attributed solely to disruption of cellular replication/differentiation patterns by terbutaline. Instead, our data suggest that the effects of terbutaline on receptor sites are secondary to interference with development of presynaptic noradrenergic function. The regimen of fetal terbutaline exposure used here has been shown to impair the ingrowth of noradrenergic projections specifically to the cerebellum (11), the same region showing the receptor changes; thus, receptor up-regulation is evidence that presynaptic input to the cerebellum is chronically subnormal. The selectivity toward areceptor subtypes is also compatible with this view. The majority 13-receptor subtype in the cerebellum, the 1~2-receptor (19), is primarily (although not exclusively) glial (18), and can be regulated separately from neuronal a-receptor sites that are primarily juxtaposed to noradrenergic projections. Accordingly, even total destruction of noradrenergic inputs leads to no more than a 50% elevation in cerebellar 132-receptor binding sites (19), and it would therefore be surprising to find a robust effect with the smaller noradrenergic deficits elicited by prenatal terbutaline. A final point is that receptor expression during development, as found here and elsewhere, does not necessarily follow monotonically the development of noradrenergic projections, but rather exhibits transient peaks of activity (16). Two factors contribute to these differences between ontogeny of projections and of receptors: first, regions may express receptor subtypes that do not exist in the adult (20) but that nevertheless play an important physiological role during a specific critical period (8). Second, the anatomical development of projections does not take into account the fact that nerve activity can change drastically during development, typically involving a transient "surge" and
2054
Fetal Terbutaline Exposure
Vol. 47, No. 22, 1990
~I-RECEPTORS Cerebellum 1600 1200 800 - o-TER 400
I
CON < TER, p < 0.05 I
I
10
I
I
20 30 Postnatal Age (days)
40
Cerebral Cortex 6000 5000 4000 3000 2000
j
I -°-TER I
1000
CON > TER, p < 0.01 I
I
10
I
20 30 Postnatal Age (days)
I
40
Midbrain + Brainstem 6000 5000 4000 3000
I-° TE" I
2000 1000
treatment x age interaction, p < 0.05 i
10
I
l
20 30 Postnatal Age (days)
I
40
Fi~. 1 Effects of fetal terbutaline exposure on binding-of [3H]prazosin to al-receptors in brain regions. Data represent mean and SE of 6-8 determinations in each group at each age; where SE bars are not visible, they are smaller than the data symbol. ANOVA indicates a significant, regionally-dependent effect of terbutaline on binding (treatment x region interaction, p < 0.01); separate tests of each region are indicated above. For regions showing a significant treatment x age interaction, asterisks indicate individual points at which the terbutaline group differs significantly from control.
Volo 47, No. 22, 1990
Fetal Terbutaline Exposure
2055
~2-RECEPTORS Cerebellum 2000 1500
~
]--=--CON I
z
-
~
0-TER
1000 500
CON
I
10
I
20 30 Postnatal Age (days)
I
40
Cerebral Cortex 2500 2000 1500 - ' I - CO N
1000 500 t
I
10
I
l
20 30 Postnatal Age (days)
40
Midbrain + Brainstem
1200
800 l'-I-CON
- o-TER
400 treatment l
10
I
x age interactk:~,p < 0.01 I
I
20 30 Postnatal Age (days)
l
40
Fi~. 2 Effects of fetal terbutaline exposure on binding of [3H]rauwolscine to o~-receptors in brain regions. Data represent means and standard errors of 6-8 determinations in each group at each age; where SE bars are not visible, they are smaller than the data symbol. ANOVA indicates a significant, age- and regionally-dependent effect of terbutaline on binding (treatment x region interaction, p < 0.01; treatment x region x age interaction, p < 0.01); separate tests of each region are indicated above. For regions showing a significant treatment x age interaction, asterisks indicate the individual points at which the terbutaline group differs significantly from control.
2056
Fetal Terbutallne Exposure
Vol. 47, No. 22, 1990
[}-RECEPTORS Cerebellum 600 500 400 300 200 100 |
I
I
I
10
20 Postnatal
30
40
Age (days)
Cerebral Cortex 1000 800
• 600
?
i_o_co:I
400 200
10
20 30 Postnatal Age (days)
40
Midbrain + Brainstem 250 200 (3
150 100 50 I
10
I
I
20 30 Postnatal Age (days;)
I
40
Fi~. 3 Effects o f fetal terbutaline exposure on binding of [125I]pindolol to ~-receptors in brain regions. Data represent means and standard errors of 6-8 determinations in each group at each age; where SE bars are not visible, they are smaller than the data symbol. A N O V A indicates a significant overall reduction caused by terbutaline (main treatment effect, p < 0.05), without selectivity for region or age (no interactions of treatment x region, treatment x age or treatment x region x age); thus noposthoc testing of individual regions or ages was undertaken.
Vol. 47, No. 22, 1990
Fetal Terbutaline
Exposure
2057
decline into adulthood (21,22). The identification of specific mechanisms by which terbutaline alters receptor development will thus require examination of patterns of gene expression of receptor proteins, as well as elucidation of potential alterations in neural tone. In the latter case, we have already shown that peripheral noradrenergic neurons exhibit lowered activity after prenatal terbutaline exposure (23), and it is likely that similar changes occur in the central nervous system. Thus, clarifying the neurotransmitter pathways and receptor types affected by prenatal exposure to terbutaline can serve as a focus for characterization of potential neurobehavioral teratogenesis associated with maternal I]-adrenergic agonist therapy. Acknowledgement Supported by USPHS HD-09713. References 1. M.E. AVERY, Brit. Med. Bull. 31 13-16 (1975). 2. D. WARBURTON, L. PARTON, S. BUCKLEY, L. COSICO, G. ENNS and T. SALUNA, Pediat. Res. 24 166-170 (1988). 3. E.M. KUDLACZ, H.A. NAVARRO and T.A. SLOTKIN, J. Pharmacol. Exp. Ther. 250 236240 (1989). 4. T.A. SLOTKIN, W.L. WHITMORE, L. ORBAND-MILLER, K.L. QUEEN AND K. HAIM, J. Pharmacol. Exp. Ther. 243 101-109 (1987). 5. E.M. KUDLACZ, H.A. NAVARRO, J.P. EYLERS, S.E. LAPPI, S.S. DOBBINS and T.A. SLOTKIN, Pediat. Res. 25 617-622 (1989). 6. A. VERNADAKIS and D.A. GIBSON, Perinatal Pharmacolo~v: Problems and Priorities. J. Dancis and H.C. Hwang (eds), 65-76, Raven Press, New York (1974). 7. T.A. SLOTKIN, R. WINDH, W.L. WHITMORE and F.J. SEIDLER, Brain Res. Bull. 21 737-740 (1988). 8. C.P. DUNCAN, F.J. SEIDLER, S.E. LAPPI and T.A. SLOTKIN, Dev. Brain Res. 55 29-33 (1990). 9. T.A. SLOTKIN and J. BARTOLOME, Brain Res. Bull. 17 307-320 (1986). 10. G. MORRIS and T.A. SLOTKIN, J. Pharmacol. Exp. Ther. 233 141-147 (1985). 11. T.A. SLOTKIN, F.E. BAKER, S.S. DOBBINS, J.P. EYLERS, S.E. LAPPI and F.J. SEIDLER, Brain Res. Bull. 23 263-265 (1989). 12. S. REINIS and J.M. GOLDMAN, The Develonment of the Brain: Biological and Functional Persnectives. Charles C. Thomas, Springfield, IL (1980). 13. E.M~ KUDLACZ, H.A. NAVARRO, J.P. EYLERS and T.A. SLOTKIN, J. Pharmacol. Exp. Ther. 252 42-50 (1990). 14. O.H. LOWRY, N.J. ROSEBROUGH, A.L. FARR and R. RANDALL, J. Biol. Chem. 193 265-270 (1951). 15. R.M. WlTKIN and T.K. HARDEN, J. Cyclic Nucleotide Res. 7 235-246 (1981). 16. J.V. BARTOLOME, R.J. KAVLOCK, T. COWDERY, L. ORBAND-MILLER and T.A. SLOTKIN, Neurotoxicology 8 1-14 (1987). 17. F.J. SEIDLER, J.M. BELL and T.A. SLOTKIN, Pediat. Res. 27 191-197 (1990). 18. E. HOSLI and L. HOSLI, Neuroscience 7 2873-2881 (1982). 19. D. LORTON, J. BARTOLOME, T.A. SLOTKIN and J.N. DAVIS, Brain Res. Bull. 21 591600 (1988). 20. L.S. JONES, L.L. GAUGER, J.N. DAVIS, T.A. SLOTKIN and J . BARTOLOME, Neuroscience 15 1195-1202 (1985). 21. F.J. SEIDLER and T.A. SLOTKIN, Neuroscience 6 2081-2086 (1981). 22. H.A. NAVARRO, F.J. SEIDLER, J.P. EYLERS, F.E. BAKER, S.S. DOBBINS, S.E. LAPPI and T.A. SLOTKIN, J. Pharmacol. Exp. Ther. 251 894-900 (1989). 23. Q.-C. HOU and T.A. SLOTKIN, Pediat. Res. 26 554-557 (1989).