Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

In the mature organism, -adrenoceptor stimulation is circumscribed by receptor down-regulation and desensitization, which serve to limit or terminate responsiveness in the face of excessive or prolonged excitation. The uncoupling of receptors from their response elements proceeds by two distinct sets of events. Desensitization that is specific to -adrenoceptor input (homologous desensitization) requires phosphorylation of the receptor by G protein-coupled receptor kinase 2 (-adrenergic receptor kinase 1), which blocks the interaction of the receptor with G proteins and ultimately leads to the internalization of the receptor itself (Hausdorff et al., 1990). However, desensitization can also involve signaling elements downstream from the receptor; this type of heterologous effect produces loss of signals from multiple inputs that share the same effectors (Clark et al., 1989; Yamashita et al., 1989; Premont et al., 1992; Ping et al., 1995).

In contrast to the mature system, during fetal or early neonatal development there is no desensitization, neither after -adrenoceptor agonist administration nor during physiological overstimulation of adrenergic pathways (Lau et al., 1982; Boreus et al., 1986; Habib et al., 1991). Indeed, the physiological responses that are necessary to the perinatal transition, cardiac inotropy and chronotropy, lung liquid reabsorption, and surfactant synthesis, actually show enhanced -adrenoceptor effects in response to stimulatory signals that would ordinarily cause desensitization in the adult (Kudlacz et al., 1990; Kudlacz and Slotkin, 1990; Stein et al., 1992; Giannuzzi et al., 1995). At the cellular level, receptor down-regulation does not occur with repeated agonist administration, and furthermore, the -adrenoceptor signaling cascade undergoes heterologous sensitization at sites distal to the receptor (Giannuzzi et al., 1995; Zeiders et al., 1997, 1999). We recently demonstrated that, rather than uncoupling receptors from G proteins, agonist treatment in the neonate causes an enhancement of coupling (Zeiders et al., 1999). Nevertheless, it appears that the major site for heterologous sensitization of signaling is at the level of adenylyl cyclase (AC) itself. Agonist-induced sensitization in the neonate is accompanied by robust increases in total AC enzymatic activity that can account for the increased response to -adrenoceptor agonists and to other agonists, such as glucagon, that operate through different receptors but that share AC as the effector molecule (Giannuzzi et al., 1995; Zeiders et al., 1997). Indeed, it is now clear that the expression and catalytic activity of AC itself limits receptor-mediated responses, both in the neonate and in the adult (Gao et al., 1998, 1999).

Accordingly, the current study examines the mechanisms by which neonatal -adrenoceptor stimulation leads to heterologous sensitization at the levels of AC expression and catalytic function. We have used binding of the specific radioligand [³H]forskolin to characterize the number and properties of AC molecules during normal ontogeny of the rat heart and to evaluate the mechanism underlying agonist-induced neonatal sensitization. We also evaluated the expression of mRNAs encoding the major AC subtypes during development and in response to agonist stimulation and compared these results with changes in the catalytic properties of AC. Finally, in addition to studying the normal ontogenetic pattern and response to stimulation, we used neonatal lesioning with 6-hydroxydopamine (6-OHDA) to evaluate the role of the development of innervation in eliciting changes in AC expression or activity.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animal Treatments. All studies using animals were carried out in accordance with the declaration of Helsinki and with the Guide for the Care and Use of Laboratory Animals, as adopted and promulgated by the National Institutes of Health. Timed pregnant Sprague-Dawley rats were shipped by climate-controlled truck (Zivic-Miller Laboratories, Allison Park, PA; transit time, 12 h) and housed with free access to food and water. After birth, the pups were randomized and redistributed to the nursing dams with a litter size of 9 to 12 pups; additionally, dams were periodically reassigned randomly to different litters to distribute any dam-related differences equally among all litters. For studies of age-dependent differences in AC signaling, hearts were dissected at various ages from birth to postweaning (25 days). To examine the effects of neonatal isoproterenol administration, equal numbers of animals within each litter were assigned in a sex-matched design to control or isoproterenol treatment groups. Beginning at 2 days of age, animals were given a 4-day regimen of once-daily s.c. injections of either 1.25 mg/kg of l-isoproterenol HCl (Sigma Chemical Co., St. Louis, MO) or an equivalent volume (1 ml/kg) of vehicle, consisting of 0.9% saline plus 0.1% ascorbic acid; tissues were harvested 24 h after the last injection. This treatment has been shown to elicit maximal cardiac -adrenoceptor stimulation at all ages, and in neonatal rats, to cause agonist-induced sensitization of AC signaling, whereas the same treatment in adults produces receptor down-regulation and desensitization (Seidler and Slotkin, 1979; Lau et al., 1982; Giannuzzi et al., 1995; Zeiders et al., 1997, 1999).

Membrane Preparation. Tissues were homogenized (Polytron, Brinkmann Instruments, Westbury, NY) in 39 volumes of ice-cold buffer containing 145 mM NaCl, 2 mM MgCl₂, and 20 mM Tris (pH 7.5); where necessary, strained through several layers of cheesecloth to remove connective tissue; and then sedimented at 40,000g for 15 min. The pellets were washed twice by resuspension (Polytron) in homogenization buffer followed by resedimentation and were then dispersed with a homogenizer (smooth glass fitted with a Teflon pestle) in 40 volumes (based on original wet weight of tissue) of a buffer consisting of either 10 mM MgCl₂ and 50 mM Tris (pH 7.5) for forskolin binding assays, or 250 mM sucrose, 1 mM EGTA, and 10 mM Tris (pH 7.4) for assays of AC activity.

[³H]Forskolin Binding. Forskolin binding assays were carried out with standard radioligand binding techniques, using an adaptation of existing methods (Shu and Scarpace, 1994). Aliquots of membrane suspension (corresponding to 25 µg of protein) were incubated with 300 nCi of [³H]forskolin (New England Nuclear, Corp., Boston, MA; specific activity, 31 Ci/mmol), isotopically diluted with unlabeled forskolin (Sigma) to produce a concentration range of 40 to 1000 nM, for 60 min at room temperature in a total volume of 250 µl of buffer. Incubations were stopped by dilution with 3 ml of ice-cold buffer, and the labeled membranes were trapped by rapid vacuum filtration onto Whatman GF/C filters, which were then washed with additional buffer and counted by liquid scintillation spectrometry. Nonspecific binding was determined in identical samples containing 10 µM unlabeled forskolin. Because binding was assessed by isotopic dilution of the radioligand, the proportion of nonspecific binding increased substantially as the concentrations were raised. Accordingly, determinations were based on an upper limit of 1 to 3 µM forskolin. Under these conditions, nonspecific binding was typically 25 to 30% of the total binding; however, in the youngest animals, in which total binding was lower, and at 1 to 3 µM forskolin concentrations, in which isotopic dilution was the greatest, nonspecific binding ranged as high as 60%. In addition, preliminary experiments were carried out with 40 µM cytochalasin B (Sigma) to demonstrate that binding of forskolin to glucose transporter sites did not contribute significantly to the overall binding (data not shown).

AC Activity. Fifty-microliter aliquots of membrane suspension were incubated for 30 min at 30°C with final concentrations of 100 mM Tris-HCl (pH 7.4), 10 mM theophylline, 1 mM adenosine 5'-triphosphate, 10 mM MgCl₂, 1 mg BSA, and a creatine phosphokinase-ATP-regenerating system consisting of 10 mM sodium phosphocreatine and 8 I.U. phosphocreatine kinase, and 10 µM GTP in a total volume of 250 µl (all reagents were purchased from Sigma). The enzymatic reaction was stopped by placing the samples in a 90-100°C water bath for 5 min, followed by sedimentation at 3000g for 15 min, and the supernatant solution was assayed for cAMP using radioimmunoassay kits (Amersham Pharmacia Biotech, Chicago, IL). Preliminary experiments showed that the enzymatic reaction was linear well beyond the 30-min time period and was linear with membrane protein concentration; concentrations of cofactors were optimal and, in particular, the addition of higher concentrations of GTP produced no further augmentation of activity. The activity of the AC catalytic unit was evaluated with the response to either 100 µM forskolin or 10 mM MnCl₂ (Zeiders et al., 1997, 1999).

RNA Isolation and AC mRNA Quantitation. For analysis of AC mRNA, tissues were frozen rapidly on dry ice and stored at 80°C until assayed. Total cellular RNA was isolated using RNeasy Midi Kits (Qiagen, Valencia, CA) and the final concentration of RNA was determined by spectrophotometry at 260 nm. Purity was verified by the ratio of optical densities at 260 and 280 nm (which was always approximately 2), and RNA integrity was confirmed by examining the 28S and 18S ribosomal bands after gel electrophoresis and ethidium bromide staining.

Data Analysis. Data are presented as means and S.E.s, with intergroup differences established by ANOVA (data log-transformed whenever variance was heterogeneous), incorporating all relevant variables (age, in vivo treatment, in vitro stimulant, and AC mRNA subtype). Post hoc tests to establish differences between individual ages or treatments were carried out with Fisher's protected least significant difference test but only where the global test indicated an interaction between treatment and the other variables; in the absence of significant interactions, only main treatment or age effects are reported. Evaluations of AC activity involved multiple measurements carried out on the same membrane preparations (stimulation by forskolin or Mn²⁺, presence or absence of GTP) so that the in vitro stimulants were considered as a repeated measure. For all tests, significance was assumed at the level of p < .05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Measurement of [³H]forskolin binding to cardiac cell membranes indicated a single class of binding sites at either postnatal day 6 (PN6) or PN25 (Fig. 1). However, there were substantial differences both in the number of sites (B_max) and in the affinity of forskolin binding (K_d) between the two ages. On PN6, B_max was in the range of 300 to 400 fmol/mg protein but the value more than tripled by PN25. The ontogenetic increase in the number of binding sites was accompanied by a lowering of binding affinity, as characterized by a doubling of the K_d, from an initial value of 0.2 to 0.4 µM; the latter value is within the range of affinities reported previously for fully mature rat heart (Shu and Scarpace, 1994; Scarpace et al., 1996).

View larger version (35K): [in this window] [in a new window]   Fig. 1.   Scatchard analysis of [3H]forskolin binding to cardiac membrane preparations at different PN ages. Data represent means and bivariate S.E. values. Inset shows raw data as the displacement curve of radiolabeled forskolin by unlabeled forskolin. Statistical analyses of Kd and Bmax values appear at the bottom.

The increase in [³H]forskolin binding between PN6 and PN25 suggests that the actual number of AC molecules increases over that span. If agonist-induced sensitization of neonatal -adrenoceptor signaling reflects actions on the ontogeny of AC, then isoproterenol treatment should produce a similar shift in the binding parameters of [³H]forskolin. Accordingly, we treated neonatal rats on PN2-5 with isoproterenol daily and examined binding on PN6, 24 h after the last dose (Fig. 2). The isoproterenol-treated animals exhibited a significant increase in both B_max and K_d to values intermediate between those of normal (untreated) PN6 and PN25 animals.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Effects of four consecutive days of isoproterenol treatment from PN days 2 through 5 on [3H]forskolin binding to cardiac membrane preparations, evaluated on PN day 6. Data represent means and bivariate S.E. values. Inset shows raw data as the displacement curve of radiolabeled forskolin by unlabeled forskolin. Statistical analyses of Kd and Bmax values appear at the bottom.

The switch from agonist-induced sensitization to agonist-induced desensitization occurs coincidentally with the development of sympathetic innervation (Giannuzzi et al., 1995; Zeiders et al., 1997, 1999). Accordingly, we also determined whether destruction of sympathetic nerves immediately after birth would arrest the development of AC in its neonatal state. Animals were treated with 6-OHDA on PN1 and binding parameters examined in preparations from animals on PN23-25 (Fig. 3). Although B_max values were slightly lower in the lesioned group, the difference was not statistically significant; furthermore, the ontogenetic increase in K_d occurred in the 6-OHDA group just as it did in controls.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Effect of neonatal 6-OHDA treatment on [3H]forskolin binding to cardiac membrane preparations PN23 to PN25. Data represent means and bivariate S.E. values. Statistical analyses of Kd and Bmax values appear at the bottom.

The ontogenetic increase in the K_d of forskolin binding suggests that a change in the conformation of the AC molecule has occurred, effects that could influence catalytic activity of the enzyme. To determine whether this occurs, we examined differential stimulation of AC with forskolin or Mn²⁺ over the immediate period (PN6 to PN15) in which the transition from agonist-induced sensitization to desensitization occurs (Fig. 4). Although both stimulants work directly upon AC, they bind to different epitopes (Hurley, 1999), so that allosteric changes should display preferential effects; because forskolin-stimulated activity is enhanced or repressed by coincident binding of different types of G proteins (Seamon and Daly, 1986), we assessed activity with and without GTP. Forskolin-stimulated activity, measured either in the absence or presence of GTP, increased substantially between PN6 and PN15, and isoproterenol pretreatment increased activity significantly (p < .002 for comparison of PN6 control to PN6 isoproterenol treated). For forskolin, addition of GTP to the medium increased the net AC activity significantly across all three groups (main effect of GTP); the magnitude of the effect was small, as would be expected, because no  agonist was included to cause activation (inclusion of a  agonist would add the confounding problem of desensitization in the older animals). Using Mn²⁺ to stimulate AC, we also obtained a much larger response on PN15 than on PN6, as well as significant stimulation of the response by neonatal isoproterenol administration (p < .005 for PN6 control versus PN6 isoproterenol treated). However, unlike forskolin, addition of GTP to the medium decreased the effect of Mn²⁺, assessed across the three groups (main effect of GTP); again, the magnitude of effect was small, albeit statistically significant. Furthermore, when the effects of age or isoproterenol treatment on the forskolin response were compared with those for the Mn²⁺ response (Fig. 4, bottom), there were major differences between the two stimulants. On PN6, the response to Mn²⁺ was twice that of forskolin, whereas by PN15, forskolin was slightly more effective than Mn²⁺. Again, isoproterenol administration to neonatal rats initiated this transition prematurely, so that on PN6, the treated animals had values in between those of normal PN6 and PN15 animals. These effects were present regardless of the inclusion or exclusion of GTP in the medium, suggesting that they were not mediated by changes in the expression or function of G proteins.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Stimulation of cardiac AC activity by forskolin or Mn2+ in membrane preparations from control animals (con) and animals treated with isoproterenol (iso) on PN2 to PN5. Activities were evaluated in the absence or presence of GTP. Data represent means and S.E. values obtained from 8 to 10 preparations for each group. ANOVAs for main effects are shown with each panel, with rank orders established by post hoc analysis with Fisher's protected least significant difference test; tests of individual differences were not carried out because of the absence of interactions of treatment × GTP. Note the different ordinate scales for forskolin- and Mn2+-stimulated activity.

Because we found differential effects of GTP on the AC enzymatic response to forskolin and Mn²⁺, we verified that the enhancement of forskolin-stimulated activity seen with the addition of GTP was paralleled by comparable changes in the ability of the enzyme to interact with forskolin. We performed the [³H]forskolin binding assay on adult cardiac cell membranes, using a fixed, subsaturating concentration (40 nM) of forskolin to be able to measure affinity-related shifts in binding and compared values with and without addition of 20 µM GTP or 20 µM GppNHp (nonhydrolyzable GTP analog); the effects were also compared with addition of 10 mM NaF, which causes maximal activation of all G proteins. Both GTP and GppNHp caused significant stimulation of [³H]forskolin binding of approximately the same magnitude seen for the effect on AC catalytic activity: GTP, 19 ± 6% increase over control binding (p < .007, n = 8); GppNHp, 39 ± 10% increase (p < .002, n = 8). Fluoride caused a massive stimulation of binding (223 ± 36% increase, p < .0001, n = 3).

Changes in the binding affinity for forskolin and in the catalytic properties of AC could result from alterations in the subtype of AC being expressed. There are two main isoforms of AC in the rat heart, type V and type VI (Premont et al., 1996; Scarpace et al., 1996). We therefore examined whether development or isoproterenol treatment could affect the differential expression of the mRNAs for the two proteins (Fig. 5). Overall, we found higher expression for AC VI than AC V, with values for both subtypes peaking in the early neonatal period and then declining slightly by PN25. Thus, during a period in which the number of AC molecules is increasing, as indicated by [³H]forskolin binding, there is no corresponding increase in the expression of mRNAs encoding the proteins. Treatment of neonatal rats with isoproterenol did cause induction of both mRNAs to a small but statistically significant extent (Fig. 6), without preferential effects toward either subtype (no interaction of treatment × subtype). We also used neonatal treatment with 6-OHDA to determine whether the onset of sympathetic innervation played a role in the ontogenetic expression of either mRNA subtype. On PN14, we found slightly reduced overall expression of both subtypes in the lesioned group (type V, 92 ± 4% of control; type VI, 86 ± 3%, n = 8) but the effects did not achieve overall significance (p > .2 for main effect of treatment) and showed no subtype preference (no treatment × subtype interaction).

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Levels of mRNAs encoding AC types V and VI at different PN ages in control hearts. Data represent means and S.E.s obtained from 8 to 30 animals at each age. ANOVA across the two different subtypes indicates a significantly greater expression of type VI (p < .0001) and no distinction in the effects of age between the two subtypes (no interaction of subtype × age). For each subtype, ANOVA indicates a main effect of age (p < .0001); *, values that differ significantly from the PN2 group.

View larger version (43K): [in this window] [in a new window]   Fig. 6.   Effects of neonatal isoproterenol treatment from PN2 through PN5 on expression of AC V or AC VI mRNA measured on PN6. Data represent means and S.E.s obtained from 18 animals for each group. ANOVA indicates a greater expression for AC VI than for AC V (p < .0001) and a significant increase caused by isoproterenol (p < .0003); however, tests for individual subtypes were not carried out because of the absence of an interaction of treatment × subtype.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

During cardiac development, chronotropic responses to -adrenoceptor stimulation increase (Seidler and Slotkin, 1979; Lau et al., 1982), attendant upon a rise in reactivity at the cellular level, as exemplified by isoproterenol-stimulated AC (Navarro et al., 1991; Pracyk and Slotkin, 1991; Giannuzzi et al., 1995; Zeiders et al., 1997). In the current study, we saw a large increment in the number of AC molecules during this period, as assessed by [³H]forskolin binding. In keeping with the view that AC itself delimits -adrenergic reactivity (Gao et al., 1998, 1999), our findings account for the ontogenetic changes in physiological performance. It should be noted that these changes occur during the span in which sympathetic innervation becomes functional (Slotkin, 1986) and over which the -adrenoceptor/AC signaling cascade develops the ability to evoke agonist-induced desensitization (Giannuzzi et al., 1995; Zeiders et al., 1997). Ordinarily, then, one would expect to see a decreased responsiveness in weanling animals as opposed to neonates; because of the increase in AC, an increase is actually seen (Seidler and Slotkin, 1979; Lau et al., 1982). The expression of AC is thus a major determinant of the overall pattern of cardiac adrenergic reactivity.

In addition to increased numbers of AC molecules during development, we obtained evidence for shifts in catalytic properties of the enzyme. The binding affinity for forskolin was significantly lower (higher K_d) in PN25 animals, compared with PN6; in addition, newborn rats showed a much greater enzymatic response to Mn²⁺, compared with forskolin, a preference that was lost as early as PN15. The two stimulants act upon different epitopes of the AC molecule, with forskolin binding to the catalytic core (Hurley, 1999), whereas Mn²⁺ replaces Mg²⁺ at metal binding sites (Hurley, 1999). The two types of effect can be resolved by the relative impact of G protein association with the AC molecule: forskolin binding and activity are enhanced by G_s-AC association, whereas Mn²⁺ interferes with G protein association, leading to loss of receptor-mediated AC activity (Limbird and Macmillan, 1981). Indeed, we found that the addition of GTP led to increases in forskolin-stimulated AC activity but decreases in Mn²⁺-stimulated activity in membranes from developing heart. The promotional effect of GTP on forskolin-stimulated AC activity was paralleled by an increase in forskolin binding to AC; furthermore, massive activation of G protein association with AC, elicited with fluoride, led to a correspondingly large increase in forskolin binding.

Our finding of differential ontogenetic changes in forskolin- and Mn²⁺-stimulated AC activity, and of the age dependence of the binding affinity of AC for forskolin, raise the possibility that ontogenetic changes in the catalytic properties of AC are related to maturation of G proteins. Indeed, G protein coupling of -adrenoceptors to AC changes markedly over the developmental time frame studied here (Cros et al., 1988; Zeiders et al., 1999). However, the effects of GTP on forskolin-stimulated activity were roughly equivalent on PN6 and on PN15 (no interaction of GTP effect × age) and the age-dependent change in forskolin binding affinity of AC was a decrease (increased K_d), not an increase as would be expected from enhanced participation of G proteins. Accordingly, other mechanisms must contribute to the change in catalytic properties. One possibility is a change in AC subtype. The heart contains predominantly AC types V and VI (Tang and Gilman, 1992). Although both proteins have been cloned and sequenced (Premont et al., 1996), their structures do not differ sufficiently to enable analysis with selective antibodies, so instead, we examined the expression of their corresponding mRNAs. Despite the large increment in total AC enzymatic activity and in the number of AC molecules, we found only slight increases in mRNA between PN2 and PN6, followed by a decline so that levels were lower on PN25 than on PN6. We also did not see differential ontogenetic changes in the two subtypes. The lack of correlation of mRNA expression to AC protein suggests that the primary determinants of AC expression are post-transcriptional, conclusions supported by previous work with hearts from adult or aged animals (Shu and Scarpace, 1994; Premont et al., 1996; Scarpace et al., 1996). It is thus not possible at the present time to determine conclusively whether the ontogenetic differences in AC catalytic activity represent isoform shifts, as opposed to modifications such as phosphorylation (Premont et al., 1992) or allosteric changes secondary to changes in the membrane lipid milieu (Benediktsdottir et al., 1995).

Previous work on the ontogeny of AC mRNA subtypes has yielded conflicting information. In one study (Espinasse et al., 1995), expression was approximately equal for type V and type VI in the fetus, but then showed marked, selective increases in type V beginning immediately after birth, with an eventual 10-fold higher value for type V. In another study (Scarpace et al., 1996), the predominance of type V in young adulthood was only 2-fold and declined with increasing age. In our study, during the circumscribed neonatal period in which signaling switches from agonist-induced sensitization to desensitization, type VI predominated. The differences among the three studies are almost certainly methodological, because the one showing large preferential increases in type V used polymerase chain reaction and RNase protection analysis; both of the previous studies failed to carry out actual molar quantitation, which requires precise specific activity determinations and the inclusion of calibrated standards on each blot (Slotkin et al., 1995a), techniques that were used here. The denominator used to calculate the concentration of the relevant mRNAs is also inconsistent among the studies. When housekeeping genes are used to correct for RNA recovery or for blotting load, membrane transfer, and hybridization efficiency, the fact that these mRNAs are also changing differentially with development can introduce artifacts in the apparent expression of other mRNAs being tested. Accordingly, our study measured mRNA relative to total RNA, which is not subject to these problems, and we actually standardized each blot to known quantities of the test cDNA to permit actual molar concentrations of the corresponding mRNAs to be determined. Nevertheless, the lack of correlation of gene expression at the mRNA level, with the actual amount of AC protein present, indicates that mechanistic conclusions cannot be drawn from ontogenetic or treatment-induced changes at the mRNA level.

Our results provide a mechanistic explanation for the existence of agonist-induced sensitization of -adrenoceptor signaling in the neonatal heart. Isoproterenol treatment of neonates led to accelerated maturation of AC activity, exemplified by an augmented concentration of AC molecules in the cell membrane (increased [³H]forskolin B_max) and a premature shift in the catalytic properties of AC (increased K_d, loss of preference for Mn²⁺-induced stimulation). Measurements of mRNA levels after isoproterenol treatment indicated a small, but statistically significant induction of both subtypes V and VI, without specificity for either subtype. We did not determine whether this represents increased gene expression or reduced mRNA degradation, nor did we study a complete time course; with earlier measurements, induction might be greater. However, it should be kept in mind that, as shown above and in previous studies (Shu and Scarpace, 1994; Premont et al., 1996; Scarpace et al., 1996), mRNA levels do not provide a major predictor of AC protein expression. Accordingly, the actual intermediate steps connecting -adrenoceptor activation to increased AC expression and altered catalytic properties will need to be elucidated before we can understand why agonist-induced sensitization disappears with development.

The time frame over which the ontogenetic changes in the expression and properties of AC occur, and over which agonist-induced sensitization disappears, corresponds to the development of sympathetic innervation (Slotkin, 1986). To see whether these events are mechanistically related, we subjected neonatal rats to lesioning with 6-OHDA to interrupt this process (Slotkin et al., 1995b). The denervated animals still showed the normal transition of forskolin binding parameters, indicating that the shifts in numbers and affinity state of AC molecules took place despite the absence of neural input. Similarly, there was little or no difference in the mRNAs encoding the two AC subtypes. Previously, we have shown that denervation slows, but does not prevent, the disappearance of agonist-induced cardiac -adrenoceptor sensitization and its replacement by desensitization (Hou et al., 1989; Slotkin et al., 1996). The role of neural input in the transition is thus most likely elicited at the level of receptor-G protein interactions, rather than at the level of AC, a conclusion in keeping with earlier results that directly assessed receptor-G protein coupling (Zeiders et al., 1999).

The present study concentrates on developmental events in the immediate neonatal period. However, it is likely that changes in catalytic properties of AC, superimposed on changes in the absolute number of AC molecules, play a vital role in the subsequent efficiency of signal transduction. AC catalytic activity shows later ontogenetic increases and decreases that do not coincide with changes in the B_max for [³H]forskolin (Navarro et al., 1991; Pracyk and Slotkin, 1991). For example, AC catalytic responses to isoproterenol, glucagon, or forskolin-Mn²⁺ on PN25 are virtually the same as on PN6, despite the 3-fold difference in the number of AC molecules (Giannuzzi et al., 1995; Zeiders et al., 1997, 1999). This reinforces the idea that heterologous signaling changes occur, consequent to functional alterations at the levels of G protein expression, of phosphorylation state of G proteins or AC itself, and of membrane lipid milieu (Clark et al., 1989; Yamashita et al., 1989; Premont et al., 1992; Ping et al., 1995; Benediktsdottir et al., 1995). Thus, although the initial ontogenetic rise in AC activity and agonist-induced neonatal sensitization clearly include the important component of AC induction, stage-specific catalytic changes are also likely to determine the set point for cellular responses. The differences between responses to Mn²⁺ and forskolin seen here provide one example of changes in catalytic properties, but we have already shown that agonist-induced sensitization includes a G_s-specific component (Zeiders et al., 1999), and we also have preliminary evidence for neonatal isoproterenol-induced suppression of G_i expression (J.L.Z., F.J.S., and T.A.S., manuscript in preparation). One potential test of G protein components in agonist-induced neonatal sensitization, would be to examine the effects of clenbuterol, which, unlike isoproterenol, stimulates only G_s and not G_i.

Our study demonstrates that agonist-induced sensitization in response to neonatal -adrenoceptor stimulation occurs because of induction of AC itself, along with changes in the catalytic properties of the enzyme. This mechanism is critically important for the ability of the developing organism to make the cardiovascular, respiratory, and metabolic adjustments that occur with perinatal transition. Circulating catecholamine levels rise substantially at the end of gestation, culminating in an enormous surge during delivery, events that enable the necessary physiological adaptations to take place (Lagercrantz and Slotkin, 1986). The absence of agonist-induced desensitization in the face of intense and prolonged stimulation, and the presence instead of sensitization, allows for the maintenance of physiological responsiveness in this period. The fact that sensitization is heterologous, targeting a signaling component (AC) shared by many other inputs, allows for integration of disparate signals that have the same functional end points; thus, thyroid hormones, glucocorticoids, glucagon, and neurotransmitters can all serve cooperatively to foster the perinatal transition (Navarro et al., 1991; Pracyk and Slotkin, 1991; Bian et al., 1992; Birk et al., 1992; Kawai and Arinze, 1993; Slotkin et al., 1994; Wagner et al., 1994).