Introduction

Summary Introduction Procedures Results Discussion References

Numerous studies have demonstrated that neuronal nicotinic receptors exhibit an extensive diversity. One nicotinic receptor subtype that is very widely distributed in both the central and peripheral nervous system is the -BGT-sensitive nicotinic receptor (1, 2). This subtype contains the 7 -BGT binding subunit in mammalian species (3-6) and an 7 and/or 8 subunit in chick (7-9). These subunits form functional homo-oligomeric channels after injection or transfection of 7 cDNA (cRNA) into oocytes or into an appropriate cell line, respectively (3, 5-7, 10). Although homo-oligomeric channels are formed under experimental conditions, the nature of the -BGT receptor in vivo is still uncertain. Current evidence suggests that the receptor in chick nervous tissue also contains other as yet unidentified subunits in addition to 7 (11), whereas experiments with rat 7/-BGT receptors suggest that the properties of native rat brain receptors are similar to those for expressed 7 receptors (12).

-BGT receptors are ligand-gated ion channels, which desensitize very rapidly, a property that had made their initial detection difficult (7, 13, 14). Studies indicate that 7 receptor activation results in an altered calcium flux across the membrane, which may be due to an increased calcium flux through the nicotinic receptor channel and/or to influx through voltage-dependent calcium channels. 7-Nicotinic receptor-mediated alterations in calcium flux have been demonstrated both for homo-oligomeric  receptors expressed in oocytes (3, 7, 15) and for endogenously expressed receptors in rat and chick ganglionic (16-18) and CNS neurons (19-21).

The functional role of the nicotinic 7/-BGT receptor in neuronal tissue is still under investigation. The very ancient lineage of the 7 receptor and its conservation throughout the course of evolution may suggest that it plays a significant role in neuronal function (22). Experimental evidence has implicated this receptor in a number of different functions, including CNS development (23-26), the regulation of neurite outgrowth (27, 28), sensory gating (29), the modulation of synaptic transmission (21), as well as others (23). Recent evidence suggests that at least some of these activities may be mediated through nicotinic receptor-mediated alterations in intracellular calcium levels (3, 7, 15-21). The purpose of the present work is to further study the functional properties of the nicotinic receptor-mediated changes in calcium flux and the factors that regulate this response.

Because responses from endogenously expressed 7/-BGT nicotinic receptors have proved difficult to study, possibly due to the rapidly desensitizing nature of the response and/or the relatively small magnitude of the response and the limited amount of tissue/cells available, we are using rat 7 receptors (7/GH₄C₁) expressed in the GH₄C₁ pituitary cell line (12). The present results show that the functional characteristics of 7/GH₄C₁ receptors are similar to oocyte-expressed and endogenous 7 receptors, suggesting that these receptors represent an appropriate model to study receptor function. The data then show that 7/-BGT receptors are regulated in a biphasic manner by nicotinic agonists, with an enhancement of receptor-mediated calcium influx after preexposure to subthreshold concentrations of agonists and desensitization after preexposure to high agonist concentrations.

	Experimental Procedures

Summary Introduction Procedures Results Discussion References

Materials. ¹²⁵I--BGT (10-20 µCi/µg) was purchased from DuPont-New England Nuclear (Boston, MA). Fura-2 AM was obtained from Molecular Probes, Inc. (Eugene, OR), MLA was obtained from Research Biochemicals, Inc. (Natick, MA), and nicotine, -BGT, cytisine, DMPP, TEA and TTX, 8-bromo-cAMP, dibutyryl cAMP, and d-tubocurarine from Sigma Chemical (St. Louis, MO). Supplies and chemicals for culture were purchased from GIBCO (Grand Island, NY). All other chemicals were purchased from standard commercial sources.

Cell culture. GH₄C₁ cells transfected with the expression vector pCEP4 containing the 7 nicotinic receptor cDNA insert have been described previously (12). A subclone of this cell line, expressing 30 fmol 7/-BGT receptor per 10⁶ cells, was used in the present study. Cells were grown in F10 medium supplemented with 8% fetal bovine serum, 50 units/ml penicillin and 50 µg/ml streptomycin and maintained in a humidified, 95% air/5% carbon dioxide atmosphere at 37°.

[Ca²⁺]_i measurements. Cells were grown to confluency on a 100-mm dish (final cell density of approximately 10-15 × 10⁶ cells per dish). The cells were removed from the dish following a 2- to 3-min exposure to 2 ml assay buffer (118 mM NaCl, 4.6 mM KCl, 20 mM HEPES, 10 mM D-glucose, and 1 mM CaCl₂, pH 7.2) supplemented with 5 mM EDTA. The cells were centrifuged for 3 min at 800 rpm and the resulting pellet resuspended in assay buffer at a concentration of 5 × 10⁶ cells/ml. Intracellular calcium was measured using the calcium chelating dye Fura-2 AM; cytosolic esterases cleave off the ester moiety, leaving the membrane-impermeant species of the dye behind to chelate free intracellular calcium. Fura-2 AM (dissolved in dimethyl sulfoxide) was added to the cell suspension to achieve a final concentration of 4 µM and the cells were incubated for 40 min at 30° in a shaker to ensure proper mixing of the dye with the cells. To remove residual dye not taken up into the cells, the samples were centrifuged at 800 rpm for 3 min and resuspended in 10 ml of assay buffer. This procedure was repeated. Assays were performed in a cuvette equipped with a magnetic stirrer in a Perkin Elmer Luminescence Spectrofluorometer LS50 (Perkin-Elmer Cetus, Norwalk, CT) with excitation of 340/380 nm and emission of 540 nm. Drugs dissolved in assay buffer were added directly to the cuvette in aliquots of 20 µl. Intracellular free calcium was calculated according to the formula of Grynkiewicz et al. (30): [Ca²⁺]_i = K_d(R  R_min/R_max  R)B, where K_d is the Fura-2 AM binding constant (224), R is the ratio of fluorescence of the cells at 340 and 380 nm, R_max and R_min are the ratios for Fura-2 free acid at 340 and 380 nm in the presence of saturating calcium (with 0.002% Triton X-100) and 10 mM EGTA, respectively, and B is the ratio of fluorescence at 380 nm in the presence of EGTA to the fluorescence at 380 nm in the presence of Triton X-100. Base-line [Ca²⁺]_i of 7/GH₄C₁ values ranged from 50 to 150 nM, agonist-stimulated values ranged from 40 to 100 nM, and potassium-evoked responses ranged from 250 to 600 nM (n = 50 experiments).

¹²⁵I--BGT receptor binding assays. Cells (0.4 × 10⁶) were plated in a 24-multiwell dish and grown to confluency (10⁶ cells/well). Before assay, the cells in culture were washed twice with 1 ml DMEM containing 3.7 mM NaHCO₃ and 0.1% bovine serum albumin (DMEM buffer). Cells were then preincubated for 60 min at 37° in the absence or presence of d-tubocurarine, followed by incubation in the presence of ¹²⁵I--BGT (1.0 nM) for 90 min at 37°. Binding was terminated by removal of the incubation medium followed by four 1-ml washes with DMEM buffer. The cells were resuspended in 500 µl of 0.5 N NaOH, with shaking, and the radioactivity was counted using a  counter. Nonspecific binding was defined as binding in the presence of 10⁴ M d-tubocurarine and represented approximately 5 to 10% of total binding at 1 nM of the radioligand.

Cell counts. 7/GH₄C₁ cells were washed twice with Ca²⁺- and Mg²⁺-free Hanks' balanced salt solution and subsequently incubated for 3 min in 0.05% trypsin to allow cell detachment. An aliquot of cells was removed from the well, trypan blue was added, and the cells were counted using a hemacytometer.

Data analysis. EC₅₀ and IC₅₀ values were determined by fitting the data to a sigmoidal logistic equation using the program Prism (GraphPAD Software, San Diego, CA). All values are expressed as the mean ± standard error. Results were analyzed using Student's t test or a one-way analysis of variance followed by the Newman-Keuls post hoc test, as indicated.

	Results

Summary Introduction Procedures Results Discussion References

Nicotinic receptor agonists increase [Ca²⁺]_i in 7/GH₄C₁ cells. Exposure of 7/GH₄C₁ cells to nicotinic agonists resulted in a dose-dependent increase in peak [Ca²⁺]_i. No increase was observed in control GH₄C₁ cells or hygromycin-resistant cells transfected with 7/pCEP4 cDNA but not expressing -BGT receptors. The rank order of potency (EC₅₀ values) of the different agonists to evoke an increase in [Ca²⁺]_i correlated well (R² = 0.86) with the ability of these ligands to interact at the -BGT receptor in binding assays (Table 1). Maximal responses for the different agonists were similar to those for DMPP (100 ± 9%; n = 11). On the other hand, the maximal response for cytisine was 69 ± 10% (n = 7), which was statistically significantly different from that for DMPP (p < 0.05 using Student's t test). As well, cytisine was the only agonist for which there was appreciable desensitization (data not shown) at higher concentrations of the agonist.

View this table: [in this window] [in a new window]   TABLE 1 Comparison of EC50 values of agonist-induced [Ca2+]i and IC50 values of the drugs to inhibit 125I--BGT binding to intact cells in culture Each value represents the mean ± standard error of three to five experiments.

Inhibition of nicotinic receptor mediated increases in [Ca²⁺]_i by antagonists. Agonist-induced responses were blocked by the 7-selective nicotinic antagonists MLA and -BGT as well as by d-tubocurarine. The data were fitted to one- and two-site competition curves. -BGT and d-tubocurarine both fit best to a one-site model with R² values of 0.85 and 0.94, respectively. The IC₅₀ value for -BGT was 2.7 ± 1.1 × 10¹⁰ M (n = 3) and d-tubocurarine 3.3 ± 0.6 × 10⁴ M (n = 4). The MLA inhibition data were best fit by a two-site model (data not shown) with R² = 0.86, whereas that for a one-site model was 0.78. The EC₅₀ values for MLA were 7.3 ± 1.5 × 10¹³ M (n = 3) and 2.0 ± 1.0 × 10¹⁰ M (n = 4). These data may indicate that MLA interacts with two distinct receptor sites; however, it is also possible that the biphasic curve occurred because of the extremely high affinity of MLA for the receptor such that the concentration of ligand did not saturate the receptor at the lower antagonist concentrations.

Nicotinic agonists increase [Ca²⁺]_i in 7/GH₄C₁ cells via voltage-gated calcium channels. The nicotinic agonist evoked increase in [Ca²⁺]_i was dependent on the presence of extracellular calcium. No significant increase in [Ca²⁺]_i was observed in the presence of CdCl₂, which blocks voltage-gated calcium channels but not nicotinic currents (data not shown). These results suggest that nicotinic agonists increase [Ca²⁺]_i through an enhanced flux through voltage-gated calcium channels. The voltage-gated calcium channel blockers nifedipine and methoxyverapamil, both at 10 µM, also blocked receptor-mediated responses (data not shown), although it is not clear that these agents are specific for only voltage-gated calcium channels.

Exposure to subthreshold and activating concentrations of agonist enhance and inhibit, respectively, subsequent agonist-induced calcium influx. Previous work has shown that nicotinic receptor responses mediated by varying  receptor subunits are desensitized after exposure to agonists at concentrations as low as 10⁹ M. The present results (Fig. 1) show that preexposure to subthreshold concentrations of the agonists epibatidine, nicotine, and DMPP resulted in an enhanced response to a subsequent application of a maximal concentration of the same agonist. When the concentration of agonist is increased to one that induces a maximal effect, the response to a second activating concentration of agonist is reduced; this is most likely due to receptor desensitization. Preexposure of the cells to subthreshold concentrations of agonist had no effect on potassium-evoked calcium influx.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Exposure to subthreshold concentrations of agonist enhance subsequent agonist-induced calcium influx, whereas activating agonist concentrations inhibit responsiveness. Dispersed 7/GH4C1 cells (1-2 × 106 cells/ml) were preexposed to the indicated concentrations of agonists; a maximal concentration of the agonist (3 × 107 M epibatidine, 3 × 105 M nicotine or 3 × 105 M DMPP) was then added to the cuvette 40 sec later. The values represent the mean ± standard error of 4-25 determinations from two to eight experiments. Dotted line, control agonist-induced calcium responses. Significance of difference from control (no agonist preexposure) using one-way analysis of variance followed by Newman-Keuls post hoc test: *, p < 0.05; **, p < 0.01; ***, p < 0.001.

View this table: [in this window] [in a new window]   TABLE 2 Modulation of nicotinic receptor sensitivity by varying agonist concentrations Dispersed cells (1-2 × 106) were exposed to subthreshold concentrations (S1) of DMPP (107 M or 3 × 107M) followed 40 sec later by exposure (S2) to a half-maximal (3 × 106 M) or maximal concentration of agonist (105 M). The values represent the mean ± standard error of 8 to 20 determinations from two to five experiments. Significance of difference from control: *p < 0.01; **p < 0.001.

View this table: [in this window] [in a new window]   TABLE 3 Effect of length of exposure time with subthreshold agonist on the agonist-induced response Dispersed cells (1-2 × 106) were exposed to subthreshold concentrations of DMPP or buffer followed 20-160 sec later by exposure to a maximal concentration of DMPP (3 × 105 M). The values represent the mean ± standard error of four to six determinations from two experiments. Significance of difference from control: *p < 0.01.

Potassium pretreatment enhances 7 receptor binding and function. The results in Fig. 2A show that potassium treatment of 7/GH₄C₁ cells results in a dose-dependent increase in -BGT receptor binding, which is due to an increase in B_max (12). Fig. 2B shows that these up-regulated receptors are functional and that the increase in the agonist-induced responses for all three agonists correspond well with the increase in receptor number. Although there were significant increases in nicotinic receptor-mediated responses after potassium pretreatment, both base-line [Ca²⁺]_i and potassium-evoked responses were similar to those observed in control cells. Values of the base-line [Ca²⁺]_i for control and potassium-treated cells were 63 ± 4 nM (n = 20) and 67 ± 6 nM (n = 19), respectively, whereas peak potassium-evoked [Ca²⁺]_i was 255 ± 22 nM (n = 20) and 234 ± 11 (n = 19) for control and potassium-treated cells, respectively. Potassium treatment did not increase cell number compared with control cells.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Potassium pretreatment enhances 7 receptor binding and function. 7/GH4C1 cells were exposed to increasing potassium concentrations for 2-3 days in culture. The potassium-containing medium was removed 3-5 hr before assay. 125I--BGT binding (A) was determined as described in Experimental Procedures. Control binding was 27.6 ± 0.8 fmol/106 cells; bars, mean ± standard error of three to seven experiments. (B) The effects of 3 × 105 M DMPP, 3 × 105 M nicotine, or 3 × 108 M epibatidine on [Ca2+]i in control and potassium-pretreated cells. Bars, mean ± standard error of six to seven determinations from two experiments. Significance of difference from its own control using one-way analysis of variance followed by Newman-Keuls post hoc test: *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Preexposure to subthreshold concentrations of agonist enhances nicotinic agonist sensitivity after potassium pretreatment. The results of Fig. 1 show that exposure to subthreshold concentrations of agonists resulted in an enhanced responsiveness to a subsequent exposure of a maximal concentration of agonist. To determine whether receptors up-regulated in the presence of potassium exhibited similar characteristics, experiments were performed to determine the effect of subthreshold concentrations of agonist on a subsequent agonist-evoked response in cells pretreated with 50 mM potassium for 2-3 days in culture. The data in Fig. 3 show that a similarly enhanced response to agonist was observed after potassium treatment (Fig. 3; Table 5). These results indicate that a greatly enhanced receptor response can occur under conditions associated with chronic depolarization, if receptors are exposed initially to low agonist. Preexposure of up-regulated receptors to high agonist resulted in receptor desensitization similar to control cells (Table 5).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Preexposure to subthreshold concentrations of agonist enhances nicotinic agonist sensitivity after potassium pretreatment. Lower traces, control agonist-induced increases in [Ca2+]i; upper traces, response that occurs after preexposure to subthreshold doses of agonist (108 M epibatidine, 3 × 107 M nicotine, or 3 × 107 M DMPP) followed 40 sec later by exposure to a concentration of the same agonist, which elicits a maximal response (3 × 10-7 M epibatidine, 3 × 10-5 M nicotine or 3 × 10-5 M DMPP). Horizontal bar, 10 sec; vertical bar, 50 nM [Ca2+]i. The response to nicotinic agonists is enhanced after potassium treatment (compare lower traces in each set). Preexposure to agonist results in a larger response compared with control in both untreated and potassium-treated cells.

View this table: [in this window] [in a new window]   TABLE 5 Agonist preexposure modulates nicotinic receptor sensitivity following long-term potassium treatment 7/GH4C1 cells were exposed to 50 mM potassium for 2 to 3 days in culture. The potassium-containing medium was removed 3 to 5 hr before measurement of [Ca2+]i. For assay, dispersed cells (1-2 × 106) were exposed to a low or high concentrations of agonist (S1) followed 40 sec later by exposure (S2) to a maximal concentration of agonist, as indicted. The values represent the mean ± standard error of 4 to 10 determinations from two to four experiments. Significance of difference from control: *p < 0.05; **p < 0.01; ***p < 0.001.

Effect of the neurotransmitters substance P and 5-HT on nicotinic receptor mediated [Ca²⁺]_i. Experiments were performed to determine whether neurotransmitter compounds that had been shown previously to alter nicotinic receptor-mediated activity affected 7 responses. One such compound is substance P (31), which modulates nicotinic receptor responsiveness in both neuronal and muscle cells in a noncompetitive manner. Table 6 shows that substance P decreased nicotinic receptor-mediated calcium influx. The decrease was maximal with 1 × 10⁶ M; higher concentrations did not further decrease calcium influx. On the other hand, 5-HT, which also has been shown to result in noncompetitive blocking effects or to modulate desensitization at some nicotinic receptors (32, 33), had no appreciable effect (Table 6).

View this table: [in this window] [in a new window]   TABLE 6 Effect of various drugs on nicotinic receptor-mediated calcium mobilization in 7/GH4C1 cells Dispersed 7/GH4C1 cells (1-2 × 106) were exposed to the indicated concentration of drug at least 20 min before exposure to 3 × 105 M DMPP. Each value represents the mean ± standard error determinations from two to four experiments. Control [Ca2+]i was 68 ± 3 (n = 43). Significance of difference from control: *p < 0.05; **p < 0.001.

Modulation of nicotinic receptor-mediated [Ca²⁺]_i by second messenger systems. An extensive body of evidence has shown that alterations in the activity of protein kinase A and/or protein kinase C activity may modulate both muscle and neuronal nicotinic receptor-mediated responses under specific experimental conditions (2). Therefore, experiments were performed to determine whether these second messenger systems might be involved in 7-mediated calcium influx (Table 6). Pretreatment of the cells with 10³ M 8-bromo-cAMP for 5-40 min resulted in a significant increase in agonist-induced [Ca²⁺]_i, with a trend for an increase at the lower concentrations tested. A tendency for an enhanced responsiveness also was observed with 10³ M dibutyryl cAMP; however, this was not statistically significantly different from control. Neither 8-bromo-cAMP nor dibutyryl cAMP altered basal or potassium-stimulated [Ca²⁺]_i at any concentrations tested. As an alternate approach to elevate cyclic nucleotide levels, the effect of forskolin, which stimulates adenylate cyclase, was also tested. It resulted in a complete block of 7 receptor-mediated responses at concentrations (1 µM) much lower than those required to inhibit enzymic activity (data not shown). This is most likely due to forskolin's ability to block nicotinic receptor-mediated activity (34), particularly because addition of 8-bromo-cAMP and dibutyryl cAMP enhanced or did not affect 7-induced responses.

8

7

8

8

7

7

9

8

7

7

6

5

	Discussion

Summary Introduction Procedures Results Discussion References

The present results show that nicotinic receptor-mediated increases in [Ca²⁺]_i are readily detectable in 7/GH₄C₁ cells, with no responses occurring in untransfected cells or 7 transfected cells not expressing -BGT receptors. The rank order of potency of the different nicotinic receptor agonists correlated well (R² = 0.86) with the interaction of these ligands at the -BGT receptor in binding assays and is similar to that observed for the homomeric rat, chick, human, and bovine 7 receptors expressed in oocytes and/or human embryonic kidney 293 cells (3-6). The 7 mediated responses were blocked by the selective -BGT sensitive nicotinic receptor antagonists -BGT and MLA (35) and also by d-tubocurarine, which inhibits most nicotinic receptor-mediated responses; these results would suggest that the agonist-induced increase in intracellular calcium concentrations is mediated by an -BGT-sensitive nicotinic receptor.

The responses to nicotinic receptor agonists were dependent on the presence of calcium in the external medium, as were potassium-evoked responses. These results suggest that the increase in [Ca²⁺]_i requires the influx of extracellular calcium and is not the result of a mobilization of intracellular stores. The observation that the nicotinic receptor-mediated responses were blocked by CdCl₂, which blocks voltage-gated calcium channels but does not inhibit nicotinic -BGT receptor-mediated currents (17), suggests that the increase in [Ca²⁺]_i elicited by nicotinic agonists occurs through voltage-gated calcium channels. The voltage-gated sodium channel blocker TTX had no effect on the agonist-induced response at maximally effective concentrations (17, 19). The present data are thus in agreement with previous work which showed that, in cultured ciliary ganglion (16, 17, 28) and in sympathetic (18) and hippocampal neurons (19), the 7/-BGT-mediated rise in [Ca²⁺]_i in response to nicotinic agonist also occurs through voltage-gated calcium channels.

Desensitization of nicotinic -BGT-sensitive nicotinic responses is a well-documented phenomenon (3, 5-7, 10, 13, 14). The present results show that agonist-induced [Ca²⁺]_i similarly desensitizes. The novel aspect of the present work is the observation that exposure of the receptors to subthreshold concentrations of agonist (i.e., concentrations that do not themselves elicit a response) results in an enhanced response to a subsequent application of a higher concentration of agonist. Similar types of studies in which the effect of varying agonist concentrations were determined on brain and peripheral nervous tissue nicotinic responses, possibly mediated by 3- or 4-containing receptors (36, 37), indicate that desensitization is the general response. Therefore, 7 receptors seem to respond to subthreshold in a manner distinct from other nicotinic receptor subtypes.

Time-course studies to determine the effect of varying the time period between the addition of subthreshold agonist and a subsequent maximal concentration of agonist showed that the enhanced response was maximal 40-60 sec after the exposure to subthreshold agonist; the response declined back to control after longer exposure periods. These results suggest that there is a desensitization to subthreshold agonist with continued exposure time. Therefore, desensitization may be dependent on both length of time of exposure to agonist, as suggested from the present results, and on agonist concentration, as shown previously by others. Although the molecular mechanisms responsible for this effect are not clear, they may involve receptor phosphorylation or dephosphorylation by various protein kinases or phosphatases, especially as the present data and that of others has shown that protein kinase A modulates nicotinic receptor responsiveness. These phenomena may represent receptor regulatory mechanisms that allow for optimal neuronal responsiveness in vivo to varying agonist exposure.

Electrical activity plays an important role in the regulation of the functional properties of neurons. Depolarization-induced increases have been demonstrated both for neuronal and muscle -BGT receptors. Exposure of cells in culture to elevated potassium for several days results in an increase in receptor number and receptor-mediated activity (38). In the present study, treatment of cultured cells to potassium for 2-3 days resulted in a significant increase in -BGT receptor binding, which is due to an increase in B_max (12). The increase in receptor number is associated with a corresponding increase in 7 receptor-mediated function. The characteristics of these up-regulated receptors were similar to control receptors in terms of agonist potencies. Interestingly, these up-regulated receptors also exhibited an enhanced response after preexposure to subthreshold concentrations of agonist. Previous studies have shown that neuronal nicotinic acetylcholine responses on PC12 cells and sympathetic neurons in culture are not regulated by depolarization, indicating that this regulatory mechanism exhibits specificity (18, 39).

The above results indicate that low agonist exposure and depolarizing inputs may act alone or in combination to result in an amplified neuronal responsiveness. It is possible that in vivo, 7 nicotinic receptor-bearing cells may be exposed to a diffuse release of agonist, which may result in a concentration of agonist insufficient to elicit a receptor-mediated response. On the other hand, this subthreshold agonist exposure may serve as a priming influence for a more enhanced response to subsequent pulses of the transmitter. The results with 7/GH₄C₁ cells exposed to long-term potassium suggest that even when the cells are chronically stimulated to result in receptor up-regulation, they are still more susceptible to nicotinic receptor-mediated depolarization after low agonist preexposure when compared with the normal resting condition. Numerous studies have shown that -BGT receptor are located extrasynaptically (40), a finding that suggests that these receptors may be involved in nonsynaptic signaling. If one of these nonsynaptically mediated functions for the -BGT receptors is neuritic retraction/extension and remodeling, as has been suggested (27, 28), the above described regulatory mechanisms may provide the means for optimal growth and development.

The neuropeptide substance P, at µM concentrations, acts as a noncompetitive inhibitor of nicotinic receptor function in both neuronal and muscle. Furthermore, -BGT has been shown to interact with rat brain tachykinin receptors and substance P to inhibit binding of ¹²⁵I--BGT to muscle and brain membranes with a maximal inhibition of 60% (31). The present results show that substance P also decreases 7 receptor-mediated [Ca²⁺]_i, with a maximal inhibition of approximately 40% at 10⁶ M. Concentrations up to 10⁵ M did not result in a further decrease in function. These results correlate well with the receptor-binding studies of Weiland et al. (31), which suggest that substance P may have a modulatory action on -BGT-sensitive nicotinic receptors.

The 7 nicotinic receptor and 5-HT₃ receptor are both ligand-gated ion channels with a similar molecular organization and some similar activation and desensitization properties. Furthermore, serotonergic ligands have been shown to affect nicotinic receptors. Garcia-Colunga and Miledi (32) had shown that serotonergic agents block neuronal nicotinic acetylcholine receptors in a noncompetitive manner, whereas Cross et al. (33) found that 5-HT enhanced the rate of desensitization of the acetylcholine current response in muscle and neuronal receptor subtypes. Despite these effects on muscle and other neuronal nicotinic receptors, the present experiments suggest that 5-HT exposure does not affect 7-mediated responses.

Second messenger have been implicated in the regulation of nicotinic receptor-mediated function, particularly cAMP (1, 2). The present data show that exposure of the 7/GH₄C₁ cells to a nonhydrolyzable analog of cAMP resulted in a significant enhancement of the response at the higher concentrations of the nucleotide, with a trend for an increase at the lower concentrations. These data suggest that phosphorylation by protein kinase A may be a mechanism involved in mediating 7 nicotinic receptor-mediated changes in [Ca²⁺]_i.

In summary, nicotinic agonist-mediated increases in calcium influx in 7/GH₄C₁ cells have the same characteristics as those observed in cultured peripheral and CNS neurons with respect to agonist/antagonist potencies and up-regulation by potassium, suggesting that they represent a good model system to study 7 receptor-mediated function. The present data further show that 7 receptors exhibit an enhanced response to agonist after preexposure of the cells to subthreshold concentrations and that this increased responsiveness also occurs after up-regulation of receptor activity in response to membrane depolarization. Therefore, multiple mechanisms may be present to increase 7 receptor-mediated neuronal responsiveness.