Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Nicotine intake is associated with large increases in circulating epinephrine and norepinephrine in humans and animals (Cryer et al., 1976; Benowitz, 1986; Van Loon et al., 1987). These increases in circulating catecholamines are at least partially responsible for a number of the peripheral pharmacological effects of nicotine, such as increases in blood pressure and cardiac output, platelet activation, stimulation of the renin-angiotensin system, and the elevation of circulating glucose and free fatty acids (Cryer et al., 1976). Numerous studies have shown that nicotine stimulates the secretion of these catecholamines from adrenal chromaffin cells (Wakade and Wakade, 1983; Boksa and Livett, 1984). In situ and cell culture studies have shown that this secretion is blocked by nicotinic acetylcholine receptor (nAChR) antagonists and requires the presence of extracellular calcium. Hence, circulating nicotine is thought to stimulate catecholamine secretion from adrenal chromaffin cells by direct interaction with nAChRs present on these cells, leading to the influx of calcium through voltage-gated calcium channels and consequent exocytosis.

Nicotine also activates and induces adrenal tyrosine hydroxylase (TH), the enzyme that catalyzes the rate-limiting step in catecholamine biosynthesis (Fossom et al., 1991a,b; Haycock and Wakade, 1992; Hiremagalur and Sabban, 1995; Jahng et al., 1997). These increases in TH are thought to represent adaptive responses, which lead to enhanced epinephrine and norepinephrine synthesis, so as to replenish the neurohormone stores lost during enhanced secretion. Even though it is clear that nicotine activates TH and TH gene transcription rate by interaction with nAChRs on adrenal chromaffin cells in cell culture or in situ model systems, the mechanisms responsible for the response to systemically administered nicotine under in vivo conditions have not been fully defined.

In previous studies we have shown that s.c. injection of nicotine leads to rapid activation of adrenal TH (Fossom et al., 1991b; Tank et al., 1998). The conventional explanation for this response is that circulating nicotine interacts with nAChRs present on adrenal chromaffin cells. In agreement with this hypothesis, nicotine administration activates TH and TH gene transcription rate in adrenal glands in which the splanchnic nerve has been transected, abolishing input to the adrenal medulla from the central nervous system (Fossom et al., 1991b; Tank et al., 1996). The nicotine-mediated activation of TH in denervated glands is totally blocked by the ganglionic nAChR antagonist, hexamethonium. In contrast, hexamethonium does not block this response in innervated adrenal glands (Fossom et al., 1991b; Tank et al., 1998). This latter unexpected result has led us to postulate that nicotine regulates adrenal TH activity via at least two mechanisms: 1) by direct interaction of circulating nicotine with nAChRs on adrenal chromaffin cells; and 2) by stimulating the splanchnic nerve, causing increased release of neurotransmitters from splanchnic nerve terminals and the consequent activation of cognate receptors on adrenal chromaffin cells.

Splanchnic nerves release multiple types of neurotransmitters, including acetylcholine (ACh), ATP, and secretin-like neuropeptides such as vasoactive intestinal polypeptide and pituitary adenylyl cyclase-activating peptide. Hence, the release of these multiple neurotransmitters would be expected to activate multiple cognate receptors linked to different intracellular signaling pathways in adrenal chromaffin cells. Because TH is activated by multiple signaling pathways [see Kumer and Vrana (1996) for review], we have postulated that the inability of hexamethonium to block the enzyme activation by nicotine in innervated adrenal glands is due to the stimulation of these multiple receptors via this trans-synaptic mechanism. In support of this hypothesis, agonists of muscarinic acetylcholine receptors activate adrenal TH in vivo (Tank et al., 1998), and neuropeptides of the secretin family activate TH in cultured adrenal medullary and PC12 cells (Roskoski et al., 1989; Waymire et al., 1991; Haycock and Wakade, 1992). However, it remains unclear how nicotine elicits this trans-synaptic regulation of adrenal TH. Does it work centrally to activate brain pathways leading to splanchnic nerve activation? Or does it work locally, possibly presynaptically on splanchnic nerve terminals, to produce increased release of neurotransmitters from splanchnic nerves?

In this report we present evidence supporting the hypothesis that nicotine can act centrally to activate TH and TH gene transcription rate in the rat adrenal medulla. We also demonstrate that this centrally mediated response to nicotine does not desensitize after repeated nicotine injections and that the induction of adrenal TH mRNA elicited by repeated nicotine treatment is dependent on this centrally mediated mechanism.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Intraventricular (i.v.t.) Injections of the Rats. Male Sprague-Dawley rats (200-250 g) purchased from Charles-River (St. Louis, MO) were used in this study. The rats were anesthetized by i.p. injection of chloral hydrate and sodium pentobarbital before surgery. Surgical implantation of catheters into the lateral ventricle was performed using standard stereotaxic surgical procedures. Appropriate coordinates (1.5 mm lateral and 0.8 mm caudal to the bregma and 7.0 mm below the surface of the skull) for implantation of the catheter into the ventricle were determined using the atlas of Paxinos and Watson (1986). The catheters were made from PE-20 polyethylene tubing. Approximately 4-cm lengths of tubing were cut, and the stylet from a 27-gauge needle was inserted into the tubing. The middle of the catheter was heated over a soldering iron until the plastic began to melt, forming a swelling. The plastic was allowed to cool, and the stylet was removed. Catheters were then cut with a razor blade to a specific size using a template. The i.v.t. portion of the catheter was cut at a 45° angle 7.0 mm from the swelling. The portion of the catheter above the skull was cut squarely 2.5 cm from the swelling. Just before insertion into the brain, each catheter was flushed with sterile saline and the outside end of the catheter was sealed with a soldering iron. Rats were allowed to recover for at least 7 days after the surgical implantation of the catheter. For 4 to 5 days before the i.v.t. injections, the animals were daily handled and placed for 30 min in the cages in which they would be administered drug.

Denervation of Left Adrenal Gland. In some experiments the rats were hemisplanchnicotomized before the i.v.t. injections of nicotine. The splanchnic nerve to the left adrenal gland was surgically transected as described in our previous studies (Fossom et al., 1991b; Tank et al., 1998). The right adrenal glands were left intact. Immediately after this surgical transection, the i.v.t. catheters were implanted as described above, while the animals were still under anesthesia. The animals were allowed to recover from these surgeries for at least 7 days before the i.v.t. injections. To monitor whether the left adrenal gland was denervated, choline acetyltransferase activity was measured in the denervated and innervated glands. The results presented in this study were obtained from animals in which 70% or greater of the choline acetyltransferase activity was lost on the denervated side. This percentage of loss of choline acetyltransferase activity was shown previously to be associated with 80% or greater loss of the trans-synaptic activation of adrenal TH elicited by either nicotine administration or decapitation (Fossom et al., 1991b).

Enzyme Assays. Adrenal glands were removed under anesthesia, rapidly frozen on dry ice, and stored at 80°C. All subsequent procedures were performed at 4°C. Frozen adrenal glands were homogenized in 250 µl of 30 mM potassium phosphate (pH 6.8), 50 mM NaF, and 10 mM EDTA, and the homogenate was centrifuged at 20,000g for 15 min. When appropriate, a 25-µl aliquot of the supernatant was removed for assaying choline acetyltransferase activity. The remainder of the supernatant was used for measuring TH activity.

Nuclear Run-on Assays. Relative TH gene transcription rate was measured in rat adrenal medulla using nuclear run-on assays, essentially as described previously (Fossom et al., 1991a) with a number of minor modifications. Briefly, adrenal glands were removed, and medullae were dissected away from the cortex under a dissecting microscope. Nuclei were isolated from the adrenal medullae from a single animal and incubated for 30 min with [³²P]UTP and appropriate buffers to promote the elongation of nascent RNA strands. Radiolabeled RNA was isolated and hybridized to a nitrocellulose filter on which the following plasmids were applied using a slot blotter: pTHg6.3, p28S1.5, and pGem7Zf. pTHg6.3 is a genomic clone encoding 6.3 kilobases of the rat TH gene (Fossom et al., 1991a). p28S1.5 encodes 1.5 kilobases of the human 28S rRNA gene and was purchased from ATCC (Manassas, VA; catalog number 77235). The 28S rRNA cDNA was used to provide signals for normalization of the TH signals, so as to control for loss of radiolabeled RNA during the assay and for hybridization efficiency. pGem7Zf was purchased from Promega Corp. (Madison, WI) and was used to provide background hybridization signals. The amount of [³²P]UTP incorporated into nascent RNA and the cpm of radiolabeled RNA put into the hybridization reactions were measured using DE81 filter assays as described by Sambrook et al. (1989). After hybridization the filters were washed and the hybridized radioactivity was visualized using autoradiography. Autoradiographic signals were quantitated by scanning the autoradiograms with a Hewlett Packard ScanJet 4C scanner, with a transparency adaptor and computer-assisted imaging analysis using IMAGE software (National Institutes of Health) to calculate density units. Care was taken to use density values that were within the linear range of the autoradiographic film. Density units were converted to cpm by comparison to a standard curve, which was constructed by spotting known amounts (cpm) of [³²P]UTP onto the nitrocellulose filter just before autoradiography. The signals for pGem7Zf hybrids were subtracted from the pTHg6.3 or p28S1.5 hybrid signals to calculate signals that represented radiolabeled RNA specifically hybridized to either TH or 28S gene sequences, respectively. The specifically hybridized TH signal was then divided by the specifically hybridized 28S signal for each sample to obtain the relative TH gene transcription rate.

Measurement of TH mRNA. Adrenal glands were removed, rapidly frozen on dry ice, and stored at 80°C. Total cellular RNA was isolated using the guanidinium hydrochloride/phenol/chloroform extraction procedure described previously (Fossom et al., 1991a). When TH mRNA was measured in adrenals from hemisplanchnicotomized animals, each adrenal gland was homogenized in a buffer containing 140 mM NaCl, 1.5 mM MgCl₂, 10 mM Tris (pH 8.6), 1 mM dithiothreitol, and 1000 U/ml RNAsin. An aliquot was removed for measurement of choline acetyltransferase activity, and the remaining suspension was rapidly added to the denaturing guanidinium hydrochloride solution used for RNA preparation. TH mRNA was measured using an RNase protection assay, as described previously (Piech-Dumas et al., 1999). Radiolabeled antisense RNA probes used for measuring TH mRNA and 28S rRNA were synthesized from linearized TH.3 and TRI RNA 28S plasmids. pTH.3 contains a 280-base pair insert encoding sequences 1241 to 1520 of the rat TH cDNA (Fossom et al., 1991a). pTRI RNA 28S contains a 115-base pair insert encoding rat 28S rRNA sequences and was purchased from Ambion, Inc. (Austin, TX). The autoradiographic density units obtained for TH mRNA duplex bands were normalized to the density units obtained from a standard curve using known amounts of TH sense riboprobe to calculate the picograms of TH mRNA present in the hybridization reactions as described by Piech-Dumas et al. (1999). These values were converted to attomoles of TH mRNA and then normalized to picomoles of 28S rRNA, which was calculated from the density units obtained from the 28S rRNA duplex bands in the same samples.

Statistical Analyses. The results were analyzed by one-way ANOVA, using the computer program INSTAT. Comparisons between groups were made using the Student-Newman-Keul or Dunnett multiple comparisons test. A level of p < 0.05 was considered statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of i.v.t. Injections of Nicotine on Adrenal TH Activity. To test directly whether nicotine activates adrenal TH via its actions in the central nervous system, we examined a number of different protocols for delivering nicotine i.v.t. In our initial experiments we injected nicotine i.v.t. while the rats were under anesthesia, however, these injections produced no effect on adrenal TH activity, possibly due to the actions of the anesthetics (sodium pentobarbital and chloral hydrate). Next, we surgically implanted a catheter into the lateral ventricle and injected nicotine via this catheter while the animal was conscious, but restrained. We did observe activation of adrenal TH in some animals using this procedure, but the response was very variable. Basal TH activity in saline-treated rats was also variable and relatively high. We postulated that this variability was due to the stress of the handling and restraint during the injection procedure. Hence, we developed a protocol to inject the animals i.v.t. while the rats were conscious, freely moving, and not restrained (see Materials and Methods). We also handled the animals for 30 to 60 s daily for 4 to 5 days before the i.v.t. injections, to accustom them to the handling necessary for insertion of the polyethylene tubing onto the permanently implanted catheter. Finally, the animals were placed in the Plexiglas cages in which they were housed during the experiment for 30 min daily for 4 to 5 days before the experiment, to minimize the stress of a new environment. With these modifications, we were able to obtain reproducibly low basal adrenal TH activity in saline-treated control animals (ranging from 0.06 to 0.13 nmol/min/mg of protein in different experiments) and also relatively reproducible responses to nicotine.

Effect of Splanchnic Nerve Transection on Adrenal TH Activation Elicited by i.v.t. Administration of Nicotine. In this set of experiments the splanchnic nerve of the left adrenal gland was surgically transected. The right adrenal gland was left intact and served as a control. Six to seven days after the surgery, the animals were administered a single i.v.t. injection of either saline (2 µl) or nicotine (360 nmol), and adrenal glands were removed under sodium pentobarbital anesthesia 20 min after the injection. Adrenal denervation did not significantly alter adrenal TH activity in response to i.v.t. saline injection (Table 2). When nicotine was injected, TH activity increased ~2-fold in the innervated adrenal gland. In contrast, TH activity did not increase in the denervated adrenal gland after a single i.v.t. injection of nicotine.

Effect of Hexamethonium on Adrenal TH Activation Elicited by i.v.t. Administration of Nicotine. Rats were injected with either saline or nicotine i.v.t. For some animals hexamethonium was also administered i.v.t. 10 min before the nicotine (or saline) injection. In other animals hexamethonium was administered i.p. 10 min before the nicotine (or saline) injection. Intraventricularly administered nicotine (360 nmol) produced a slightly less than 2-fold activation of adrenal TH (Table 3). When administered i.v.t., hexamethonium did not affect adrenal TH activity in rats injected i.v.t. with saline. However, when hexamethonium was administered i.v.t. before the injection of nicotine, the nicotine-mediated activation of adrenal TH was completely abolished. In contrast, when hexamethonium was administered i.p. before the nicotine injections, adrenal TH activity still increased ~2-fold after the i.v.t. injections of nicotine. In previous experiments we showed that i.p. administration of hexamethonium by itself does not significantly affect adrenal TH activity (Fossom et al., 1991b).

Effect of Repeated i.v.t. Injections of Nicotine on Adrenal TH Activity. We next tested whether the adrenal response to nicotine desensitized when the drug was administered i.v.t. repeatedly. Nicotine (360 nmol) was injected i.v.t. seven times (injections spaced 30 min apart) over a 3-h period. Saline was similarly injected repeatedly i.v.t. in control animals. During this entire injection period, the rats were maintained conscious and unrestrained in cylindrical Plexiglas cages, as described under Materials and Methods. Adrenal glands were removed under sodium pentobarbital anesthesia 20 min after the final i.v.t. injection. Adrenal TH activity in rats injected i.v.t. repeatedly with saline was similar to that observed in rats injected once with saline (Table 4). Repeated i.v.t. nicotine injections over 3 h were associated with 3-fold increases in adrenal TH activity.

Effect of i.v.t. Injections of Nicotine on TH Gene Transcription Rate and TH mRNA Levels in Adrenal Medulla. In the first set of experiments we tested whether adrenal TH gene transcription rate was stimulated after a single i.v.t. injection of nicotine. Rats were injected i.v.t. once with either saline (2 µl) or nicotine (360 nmol), and adrenal glands were isolated under sodium pentobarbital anesthesia 20 min after the injection. Adrenal medullae were dissected and nuclear run-on assays were performed to measure relative TH gene transcription rate. Results from these experiments are presented in Table 5. Intraventricularly injected nicotine produced an ~2-fold stimulation of adrenal TH gene transcription rate.

Is the Centrally Mediated Trans-Synaptic Stimulation of the Adrenal Medulla Essential for the Induction of TH mRNA by Systemically Administered Nicotine? The left adrenal gland was surgically denervated, and 7 days after the surgery the hemisplanchnicotomized rats were repeatedly administered 1.6 mg/kg nicotine s.c. over 3 h (injections spaced 30 min apart). In previous studies we showed that this acute repeated nicotine treatment stimulated TH gene transcription rate for at least 3 h and induced both TH mRNA and TH protein in innervated rat adrenal medulla (Fossom et al., 1991a). We tested whether this induction of TH mRNA is dependent upon splanchnic nerve innervation. As expected, adrenal TH mRNA was induced in the innervated adrenal gland (Table 6). In contrast, TH mRNA was not induced in the denervated adrenal gland.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Cigarette smoking stimulates the release of catecholamines from adrenal medulla and sympathetic neurons, leading to increased blood pressure and pulse rate (Cryer et al., 1976). These effects are thought to be partially mediated by direct interaction of nicotine with nAChRs present on adrenal chromaffin cells and sympathetic ganglia. However, a number of studies using rats, mice, and dogs have shown that centrally administered nicotine also activates the sympathetic nervous system, increasing blood pressure and other sympathetic responses (Lang and Rush, 1973; Kubo and Misu, 1981; Yokotani et al., 1987; Kiritsy-Roy et al., 1990; Siren and Feuerstein, 1990; Buccafusco and Yang, 1993; Song et al., 1999). These centrally evoked effects of nicotine are apparently mediated to a large degree by the release of epinephrine from the adrenal medulla (Kubo and Misu, 1981; Yokotani et al., 1987; Kiritsy-Roy et al., 1990). Our results support these findings and extend them to show that the increased release of epinephrine from the adrenal medulla initiated by the injection of nicotine i.v.t. is associated with activation of adrenal TH and stimulation of adrenal TH gene transcription rate. Furthermore, when nicotine is injected i.v.t. repeatedly over a number of hours, TH mRNA is also induced.

In previous studies we have shown that the systemic administration of nicotine is associated with activation of TH enzyme and induction of TH gene expression in rat adrenal medulla (Fossom et al., 1991a,b; Tank et al., 1996, 1998). We have postulated that these adrenal responses to nicotine are due to at least two mechanisms: 1) direct interaction of circulating nicotine with nAChRs present on adrenal chromaffin cells; and 2) stimulation of multiple adrenal chromaffin cell receptors by neurotransmitters released from splanchnic nerves presynaptic to adrenal chromaffin cells. The evidence supporting this hypothesis is as follows: 1) Systemically administered nicotine stimulates both TH enzyme activity and TH gene transcription rate in denervated adrenal glands (Fossom et al., 1991b; Tank et al., 1998). This observation demonstrates that mechanisms independent of splanchnic nerve innervation are capable of mediating at least part of these responses. Furthermore, hexamethonium blocks the activation of adrenal TH in denervated glands, suggesting that agonist occupation of adrenal chromaffin cell nAChRs is sufficient to mediate this response. 2) Systemically administered hexamethonium does not block the nicotine-mediated activation of TH enzyme or TH gene transcription rate in innervated adrenal glands (Fossom et al., 1991a,b; Tank et al., 1998). This result suggests that agonist occupation of chromaffin cell receptors other than nAChRs by neurotransmitters released from splanchnic nerves may compensate for the blockade of the nAChRs by hexamethonium. We have postulated that this trans-synaptic mechanism is due to the effect of nicotine in the brain, leading to increased stimulation of the splanchnic nerve. The present studies strongly support this hypothesis. Our results do not rule out the possibility that systemically administered nicotine may also be acting peripherally to produce this trans-synaptic stimulation of the adrenal medulla, but they do indicate that the central actions of nicotine are sufficient to elicit activation of TH enzyme, stimulation of the TH gene, and induction of TH mRNA.

Because nicotine is a very lipid-soluble compound, it is likely to diffuse rapidly across the blood-brain barrier into the periphery after the i.v.t. injection. Hence, one confounding issue in these experiments is whether the centrally administered nicotine produces its effect on the adrenal medulla by passing into the blood and acting directly on adrenal chromaffin cell nAChRs. This possibility seems remote, because the dose of nicotine administered i.v.t. in our studies (360 nmol/rat) is equivalent to 0.2 to 0.3 mg/kg nicotine administered s.c.; this dose does not activate TH or induce TH mRNA in rat adrenal medulla (Fossom et al., 1991a,b). However, to test directly for this possibility, two experiments were performed. First, we have shown that denervation of the adrenal gland completely abolishes the response to centrally administered nicotine (Table 2). Hence, circulating nicotine derived from the i.v.t. injections does not accumulate to a high enough level to activate adrenal TH by direct interaction with adrenal chromaffin cell nAChRs. Secondly, we have shown that the i.p. administration of hexamethonium does not block the activation of adrenal TH elicited by i.v.t.-administered nicotine. Because hexamethonium is highly charged, it does not readily cross the blood-brain barrier; hence, its effect is to block peripheral nAChRs, such as those in the adrenal medulla, when administered via this systemic route. Consequently, these results are consistent with the hypothesis that, in the presence of a nAChR antagonist, the nicotine-mediated activation of adrenal TH is mediated by neurotransmitters released from the splanchnic nerve, which interact with non-nAChRs on adrenal chromaffin cells. These results also agree with those reported in our previous studies, in which systemically administered hexamethonium does not block the activation of TH enzyme or TH gene transcription rate elicited by the s.c. administration of nicotine. Taken together, our results support the argument that, under the conditions of our experiments, i.v.t.-administered nicotine acts centrally, not peripherally, to regulate adrenal TH.

Very little is known about the central mechanisms by which nicotine stimulates the outflow of the sympathetic nervous system. Our results show that i.v.t.-administered hexamethonium completely blocks the effect of i.v.t.-administered nicotine on adrenal TH activity. The dose of hexamethonium used in our studies also blocks the increase in blood epinephrine levels elicited by i.v.t.-administered nicotine (Kiritsy-Roy et al., 1990). These results suggest that central neuronal-type nAChRs are essential for these effects on adrenal medullary function. However, these results do not shed light on which subtype of neuronal nAChR participates in the response, nor does it provide information concerning the brain region involved in mediating nicotine's actions. Previous studies have shown that nAChRs present in hypothalamus and medulla oblongata produce marked effects on cardiovascular function (Bhargava et al., 1978; Dev and Loeschcke, 1979). More recent studies have shown that selective administration of nicotine into different brainstem nuclei stimulates adrenocorticotropin hormone secretion to different degrees. More work is needed to determine which brain regions and nAChR subtype(s) participate in the response of adrenal TH to nicotine.

As mentioned under Results, a single injection of nicotine either s.c. or i.v.t. does not elicit a significant induction of adrenal TH mRNA (Fossom et al., 1991a). However, nicotine does induce adrenal TH mRNA, when it is administered repeatedly either chronically or acutely. Single injections of nicotine administered chronically once or twice daily for 7 to 14 days elicit relatively large increases in adrenal TH mRNA and TH activity (Seidler and Slotkin, 1976; Hiremagalur and Sabban, 1995). In studies using an acute repeated injection protocol, we have shown that repeated injections of nicotine over a 3-h period (injections spaced 30 min apart) produces sustained increases in adrenal TH activity and TH gene transcription rate, leading eventually to induction of TH mRNA and TH protein (Fossom et al., 1991a,b). These sustained increases in TH activation state and TH gene transcription rate in these acute studies are surprising, because nAChRs desensitize rapidly after prolonged or repeated exposure to agonists. Indeed, adrenal TH is not activated in denervated adrenal glands after repeated administration of nicotine over a 3-h period (Fossom et al., 1991b). This result suggests that adrenal chromaffin cell nAChRs desensitize during this acute repeated nicotine treatment, but in innervated glands TH activity and TH gene transcription are maintained at elevated rates due to sustained stimulation of the splanchnic nerve and consequently sustained activation of non-nAChRs by neurotransmitters released from the splanchnic nerve. One problem with this interpretation is it doesn't explain why the nAChRs that mediate the sustained stimulation of the splanchnic nerve don't become desensitized. The results of the present study do not address this question directly, but they do explain the previous observations. Repeated i.v.t. administration of nicotine for 3 h elicits a sustained activation of adrenal TH. This sustained activation is dependent on an intact splanchnic nerve, because denervation of the adrenal gland completely blocks the sustained response. These results indicate that the central nAChRs mediating these adrenal responses do not desensitize during this 3-h repeated nicotine administration paradigm. The identities of these nAChRs and the reasons why they do not desensitize remain obscure. Desensitization of nAChRs depends upon a number of parameters, including nAChR subunit composition, post-translational modifications of the receptor, and the environmental milieu of the neurons expressing the receptor (Dani et al., 2000). Further work is needed to clarify this issue, but our results suggest that the central nAChRs that mediate the trans-synaptic regulation of adrenal TH differ functionally from those present on adrenal chromaffin cells, which regulate the enzyme in response to circulating nicotine.

Finally, we have shown that repeated i.v.t. injections of nicotine for 3 h induce TH mRNA in rat adrenal gland. Furthermore, we have shown that the induction of adrenal TH mRNA by repeated s.c. injections of nicotine is totally blocked by denervation of the adrenal gland. Taken together, these results support the hypothesis that the induction of adrenal TH mRNA elicited by systemically administered nicotine is dependent on trans-synaptic mechanisms initiated by central nAChRs that do not readily desensitize during this acute repeated nicotine treatment. In addition, taken together with our previous finding that the TH enzyme response to nicotine desensitizes in denervated adrenal glands after acute repeated systemic administration of the drug (Fossom et al., 1991b), these results indicate that sustained agonist occupation of adrenal chromaffin cell nAChRs by circulating nicotine is not sufficient to induce TH mRNA. Hence, we conclude that the central response to acute repeated nicotine administration is essential for the long-term induction of TH mRNA in rat adrenal medulla.