Abstract
The recent increase in the use of nicotine products by teenagers has revealed an urgent need to better understand the impact of nicotine on the adolescent brain. Here, we sought to examine the actions of extracellular ATP as a neurotransmitter and to investigate whether ATP and nicotinic signaling interact during adolescence. With the GRABATP (G-protein-coupled receptor activation-based ATP sensor), we first demonstrated that nicotine induces extracellular ATP release in the medial habenula, a brain region involved in nicotine aversion and withdrawal. Using patch-clamp electrophysiology, we then demonstrated that activation of the ATP receptors P2X or P2Y1 increases the neuronal firing of cholinergic neurons. Surprisingly, contrasting interactive effects were observed with nicotine exposure. For the P2X receptor, activation had no observable effect on acute nicotine-mediated activity, but during abstinence after 10 d of nicotine exposure, coexposure to nicotine and the P2X agonist potentiated neuronal activity in female, but not male, neurons. For P2Y1 signaling, a potentiated effect of the agonist and nicotine was observed with acute exposure, but not following extended nicotine exposure. These data reveal a complex interactive effect between nicotinic and ATP signaling in the adolescent brain and provide mechanistic insights into extracellular ATP signaling with sex-specific alterations of neuronal responses based on prior drug exposure.
SIGNIFICANCE STATEMENT In these studies, it was discovered that nicotine induces extracellular ATP release in the medial habenula and subsequent activation of the ATP purinergic receptors increases habenular cholinergic neuronal firing in the adolescent brain. Interestingly, following extended nicotine exposure, nicotine was found to alter the interplay between purinergic and nicotinic signaling in a sex-specific manner. Together, these studies provide a novel understanding for the role of extracellular ATP in mediating habenular activity and reveal how nicotine exposure during adolescence alters these signaling mechanisms, which has important implications given the high incidence of e-cigarette/vape use by youth.
Introduction
Nicotine is the primary reinforcer underlying tobacco and e-cigarette dependence (Le Foll et al., 2022). While tobacco smoking has declined, this has been unfortunately paralleled by a dramatic increase in e-cigarette use, especially among adolescents (National Youth Tobacco Survey, Centers for Disease Control and Prevention; Available at: https://www.cdc.gov/tobacco/data_statistics/surveys/nyts/index.htm). Since nicotine can alter neurodevelopmental processes leading to persistent neurobiological changes (Luna et al., 2010; Siqueira, 2017; Pushkin et al., 2019; Ren and Lotfipour, 2019; Mahajan et al., 2021), it is critical to better understand how nicotine use during adolescence may induce changes in neuronal signaling. The medial habenula (MHb) projection to the interpeduncular nucleus has been shown to modulate nicotine intake, aversion, and withdrawal (Salas et al., 2004a, b, 2009; Fowler et al., 2011; Görlich et al., 2013; Dao et al., 2014; Fowler and Kenny, 2014; Shih et al., 2014; Antolin-Fontes et al., 2015). Cholinergic MHb neurons exhibit characteristic baseline spontaneous firing with a frequency of 2–10 Hz (Görlich et al., 2013; Dao et al., 2014). Nicotine acts on cholinergic MHb neurons via the nicotinic acetylcholine receptors (nAChRs), which include α3-, α5-, and β4-containing nAChRs (Fowler et al., 2011; Görlich et al., 2013; Dao et al., 2014; Shih et al., 2014; Elayouby et al., 2021). Acute nicotine increases MHb neuronal firing frequency (Görlich et al., 2013). Interestingly, following chronic nicotine exposure, there is no difference in baseline firing frequency, but nicotine re-exposure significantly potentiates neuronal activity (Görlich et al., 2013). Thus, MHb activity appears to become altered with nicotine use, which thereby underlies the withdrawal state. Indeed, injection of a nAChR antagonist into the MHb elicits withdrawal symptoms (Salas et al., 2009), providing further evidence for the role of the MHb nAChRs in withdrawal symptomology. The MHb cholinergic neurons also express neurokinin and hyperpolarization-activated cyclic nucleotide-gated (HCN) channels, which appear to interact with nicotinic signaling and can alter withdrawal symptoms (Görlich et al., 2013; Dao et al., 2014). Together, these findings demonstrate that cholinergic MHb neurons mediate nicotine withdrawal, and multiple cellular mechanisms may converge to regulate the withdrawal state.
ATP has canonically been recognized for its critical role in providing energy to support intracellular processes. However, ATP is also localized in synaptic regions and acts on purinergic P2X and P2Y receptors; moreover, interactions between purinergic receptors and nAChRs have been demonstrated at the cell membrane in vitro, and both receptor types are localized in MHb neurons (Edwards et al., 1992; Kanjhan et al., 1999; Morán-Jiménez and Matute, 2000; Khakh et al., 2005; Florenzano et al., 2008; Guo et al., 2008; Song et al., 2011). ATP activation leads to either influx of Na+ and Ca2+ via P2X (North, 2002; Burnstock, 2018) or increased intracellular Ca2+ through G-protein signaling via P2Y1 (Abbracchio et al., 2006; Guzman and Gerevich, 2016). Interestingly, in vitro, mutual inhibition has been shown between P2X receptors and α3β4-containing nAChRs (Khakh et al., 2000; Decker and Galligan, 2009), both of which are expressed in cholinergic MHb neurons. Thus, these prior findings support the contention that purinergic signaling may be an important modulator of MHb neuronal activity.
In these studies, we sought to determine whether nicotine acts on circuit function to alter ATP signaling and whether ATP neurotransmission regulates habenular firing during the naive state and after chronic nicotine exposure. To monitor ATP release activity in the MHb, we used the G-protein-coupled receptor activation-based ATP sensor (GRABATP), a modified human P2Y1 receptor that elicits a florescence signal with ATP binding (Wu et al., 2022). We then used patch-clamp electrophysiology to examine the effects of activation of P2X and P2Y1 receptors on MHb neuronal activity and to determine the further interaction of the receptors with nicotine, under both acute and nicotine withdrawal states, in male and female mice. Together, our studies reveal the functional relevance of extracellular ATP in the habenula and further demonstrate that purinergic signaling plays an important sex-specific role in habenular neuronal activity during nicotine abstinence in adolescence.
Materials and Methods
Animals.
C57BL/6J wild-type male (n = 39) and female (n = 44) mice (RRID:IMSR_JAX:000664) were derived from six breeders in our laboratory animal facilities. Subjects were investigated during adolescence [postnatal day 21 (P21) to P50; Yuan et al., 2015]. Mice were maintained in the vivarium with a 12 h reversed light/dark schedule. Food and water were provided ad libitum until killed for patch-clamp recording or GRABATP imaging. All experiments were performed in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at the University of California, Irvine.
Drugs.
The P2X receptor general agonist αβ-methylene ATP salt (catalog #3209, Tocris Bioscience), the P2Y1-specific agonist [(1R,2R,3S,4R,5S)−4-[6-amino-2-(methylthio)−9H-purin-9-yl]-2,3-dihydroxybicyclo[3.1.0]hex-1-yl]methyl] diphosphoric acid monoester trisodium salt (MRS2365; catalog #2157, Tocris Bioscience), the general P2 antagonist suramin (catalog #S2671, Sigma-Aldrich), and (–)-nicotine hydrogen tartrate salt (catalog #0215355491, MP Biomedicals) were dissolved in 0.9% sterile saline and adjusted to pH 7.4. Of note, we focused on the P2X general agonist since subtype-selective agonists are not available. Based on prior studies, the following concentrations were used: a 100 μm αβ-methylene adenosine 5'-triphosphate trisodium salt solution and a 30 μm suramin solution (Edwards et al., 1992; Pankratov et al., 2002). For the P2Y1-specific agonist, a moderate dose of 50 μm was selected (Liston et al., 2020), and this receptor subtype was focused on, given the availability of the selective agonist and the documented expression levels in the MHb. Based on previous electrophysiology studies in the MHb (Dao et al., 2014; Shih et al., 2014; Morton et al., 2018), 1 μm nicotine was used to validate that the recordings are retrieved from cholinergic neurons. To avoid a possible ceiling effect of neuronal firing, 0.5 μm nicotine was chosen to study the interaction between cholinergic and purinergic signaling mechanisms. Of relevance, concentrations in this range are consistent with that found in human smokers (Benowitz et al., 2009).
Nicotine exposure paradigm.
To investigate the role of purinergic signaling in the MHb during nicotine abstinence, adolescent mice were subject to subcutaneous nicotine injection at a concentration of 1.5 mg/kg for 10 d. The injection route and duration of this treatment was selected to provide controlled dosing over a moderate length of exposure period (Chellian et al., 2021). The 1.5 mg/kg dose was selected since mice have been previously shown to self-administer this amount in daily sessions (Fowler and Kenny, 2011). The GRABATP imaging and electrophysiology recordings were conducted 24 h after the last nicotine injection. Of note, the 24 h abstinence time point has been well established in the field as a significant point of withdrawal symptomology following nicotine exposure in animal models (Damaj et al., 2003; Matta et al., 2007). Further, humans report adverse symptoms of withdrawal after 4–24 h of cessation (Merritt et al., 2010; McLaughlin et al., 2015; Conti et al., 2020), including adolescent smokers at 24 h (Rubinstein et al., 2009), supporting the translational value of this time point.
Acute brain slicing.
Mice were fully anesthetized with isoflurane, followed by transcardiac perfusion of 10 ml of 4°C N-methyl-d-glucamine (NMDG)-HEPES artificial CSF (aCSF) containing the following (in mm): 92 NMDG, 2.5 KCl, 1.25 NaH2PO4, 30 NaHCO3, 20 HEPES, 25 glucose, 2 thiourea, 5 Na-ascorbate, 3 Na-pyruvate, 0.5 CaCl2 · 2 H2O, and 10 MgSO4 · 7 H2O, saturated with carbogen (95% O2/5% CO2) and a pH value adjusted to ∼7.4. The transcardiac perfusion of the prechilled NMDG-HEPES aCSF solution is used to rapidly slow the metabolism of the brain to decrease neuronal death rate during slicing (Ting et al., 2018). The brain was quickly extracted, and 250 μm coronal slices of the MHb were sectioned using a MicroSlicer Zero 1N (Ted Pella). After sectioning, slices were transferred into the beaker filled with 150 ml of 32°C NMDG-HEPES aCSF, and the Na+ spike-in procedure was performed by adding Na+ spike-in solution (2 m NaCl in NMDG-HEPES aCSF) every 5 min. The Na+ spike-in procedure allows the brain slices to gradually accommodate to a higher extracellular Na+ concentration (Ting et al., 2018). This procedure can both increase neuronal survival rate and reliability of the gigaseal formation during patch clamp (Ting et al., 2018). After adding the Na+ spike-in solution five times, the brain slices were transferred to a HEPES holding solution containing the following (in mm): 92 NaCl, 2.5 KCl, 1.25 NaH2PO4, 30 NaHCO3, 20 HEPES, 25 glucose, 2 thiourea, 5 Na-ascorbate, 3 Na-pyruvate, 2 CaCl2 · 2 H2O, and 2 MgSO4 · 7 H2O, saturated with carbogen (95% O2/5% CO2) and pH value adjusted to ∼7.4. The brain slices were incubated at room temperature at least 1 h before recording with the intracellular solution osmolarity of 290 mOsmol/kg and extracellular osmolarity of 310 mOsmol/kg.
Electrophysiology patch-clamp and drug application.
Brain slices were transferred to the recording chamber (SliceScope, Scientifica; RRID:SCR_018405) and submerged under 32°C recording solution containing the following (in mm): 124 NaCl, 2.5 KCl, 1.25 NaH2PO4, 24 NaHCO3, 5 HEPES, 12.5 glucose, 2 CaCl2 · 2 H2O, and 2 MgSO4 · 7 H2O, saturated with carbogen (95% O2/5% CO2) and with a pH value adjusted to ∼7.4. The rate of recording solution perfusing into the recording chamber was 5 ml/min. Ventral MHb neurons were visualized by infrared differential interference contrast microscopy (Olympus). Whole-cell recordings were obtained using borosilicate glass pipette electrodes (catalog #P-1000, Sutter Instrument; RRID:SCR_021042) with resistance of 3–5 MΩ. The electrodes were filled with a potassium methanesulfonate-based internal solution containing the following (in mm): 135 KMeSO3, 10 HEPES, 4 MgCl2, 4 MgCl2, 4 Na2ATP, 0.4 NaGTP, and 10 sodium creatine phosphate with a pH value adjusted to 7.3 using KOH.
Data were acquired by using an amplifier (Multiclamp 700B Amplifier, Molecular Devices; RRID:SCR_018455) and a digitizer (Digidata 1550B, Molecular Devices), and were recorded with pCLAMP 10 (Molecular Devices; RRID:SCR_011323). All signals were filtered at 5 kHz and sampled at 20 kHz. The series resistance was monitored by applying a hyperpolarizing 10 mV voltage step in voltage-clamp mode to calculate the resistance using Ohm's law. If the series resistance varied >20% across the experiment, the data were excluded because of concerns with the gigaseal state.
Ventral MHb neurons were maintained at −50 mV under current clamp to provide standardized conditions (Hsu et al., 2013; Vickstrom et al., 2020). The baseline was recorded for 5 min before drug application. The αβ-methylene ATP or MRS2365 was applied in 0.5 s pulses using a picospritzer (Parker Hannifin; RRID:SCR_018152). Drug action was quantified during the 5 s immediately after picospritzer application. The borosilicate glass pipettes (3–5 MΩ) containing αβ-methylene ATP or MRS2365 were placed ∼20 µm away from recorded neurons, and the pressure meter on the picospritzer was set to 10–12 psi. To assess the interaction between cholinergic and purinergic signaling, the baseline was recorded for 2 min followed by bath application of 0.5 μm nicotine for 10 min. With the picospritzer, αβ-methylene ATP or MRS2365 was applied onto the recorded neuron for 0.5 s while nicotine bath application was maintained for an additional 10 min. To examine cell activity following chronic nicotine treatment, brain slices were obtained from mice injected with nicotine daily for 10 prior days, with the same tissue processing and patch-clamp procedures described above. Given our focus on nAChR-containing neurons, cells were excluded if no response was elicited with application of nicotine at the end of the study.
To examine endogenous purinergic signaling in the presence of an antagonist, neurons were held at −50 mV in current clamp. Sections were bathed in aCSF for 5 min, followed by bath application of the following: (1) 5 min of suramin (30 μm), (2) 5 min of suramin (30 μm) and nicotine (0.5 μm), and (3) 10 min of nicotine (0.5 μm), in sequential order. Neuronal activity was recorded, and firing frequency was calculated during each drug bath application. To ensure a return to baseline levels following suramin application, nicotine was bath applied for 10 min, and the neuronal activity during the last 5 min was analyzed.
Stereotaxic surgery.
The adeno-associated virus (AAV)-GRABATP sensor was microinjected directly into the MHb using a stereotaxic device (Kopf). Mice were anesthetized with 1–3% of an isoflurane/oxygen mixture. The head was fixed and adjusted to flat-skull position on the Kopf stereotaxic frame. Microinjection of 0.5 μl of AAV9-hSyn-ATP1.0 (titer, 5 * 1013 vector genomes/ml; catalog #YL006001-AV9, WZ Biosciences; Wu et al., 2022) was administered bilaterally into the MHb (AP, −1.2 mm; midline, ±1.06; DV, −2.47 mm; injectors at a 20° angle toward midline) at a rate of 0.25 μl across 2 min. The injectors remained in place for an additional 3 min before moving 0.2 mm up the dorsal axis, at which time another 0.5 μl of AAV9-hSyn-ATP1.0 was administered at the same infusion rate across 2 min. The injectors remained in the injection site for 3 min before removal.
Fluorescence imaging of GRABATP in brain slices.
After a period of 2.5 weeks following the AAV9-hSyn-ATP1.0 injection, acute brain slices at a 250 µm thickness were obtained from the mice, as described above. Images were acquired on the electrophysiology recording microscope with a SciCam Pro Camera (Scientifica). The frame rate of the image was set to 2 frames/s, and the binning mode was set at 2 × 2 on the Ocular software (Scientifica; RRID:SCR_021035). MHb neurons expressing GRABATP were visualized using 488 nm LED light. The baseline activity of ATP release in MHb was recorded for 2.5 min before bath applying 1 μm nicotine. After 5 min, nicotine exposure (to allow the recording chamber to be fully submerged with nicotine solution), fluorescence activity indicative of ATP release was recorded for at least 8 min.
Statistical analysis.
The number of subjects for each experiment was based on the results of power analysis performed in G*Power (a priori estimated effect size: 0.6, p < 0.05; power, 0.95), and published guidelines for electrophysiological studies (Journal of Neuroscience, 2018). t tests or ANOVAs was performed to fit the experimental design. Repeated one-way ANOVAs were conducted to examine the effects of two or more treatments with a within-subject design. Following significant main and interaction effects, a Tukey's post hoc test was used, with correction for multiple comparisons. Data analysis was performed in GraphPad Prism 7.0 software (RRID:SCR_002798). The criterion of significance was α = 0.05.
Results
Nicotine induces ATP release in the MHb
The expression of purinergic receptors in the MHb have been documented; however, ATP release activity in this brain region remains to be confirmed (Sperlágh et al., 1995, 1998; Kanjhan et al., 1999; Morán-Jiménez and Matute, 2000; Price et al., 2003; Florenzano et al., 2008; Guo et al., 2008; Song et al., 2011). Thus, we used GRABATP sensor imaging to directly monitor real-time ATP release, which is reflected by the change of fluorescence signal (Fig. 1A–F). In the MHb, we did not detect changes in GRABATP sensor fluorescence under endogenous conditions, suggesting undetectable, if any, extracellular ATP release at baseline (Fig. 1C). Thus, we next sought to determine whether acute nicotine would alter ATP extracellular activity. Perfusion of nicotine induced transient and localized increases in fluorescence signal, indicative of nicotine-mediated ATP release (Fig. 1D). Given that the MHb is highly involved in regulating nicotine withdrawal (Görlich et al., 2013; Dao et al., 2014; Fowler and Kenny, 2014; Shih et al., 2014; Antolin-Fontes et al., 2015), we next hypothesized that signaling dynamics may be altered following chronic nicotine exposure. In brain sections from mice exposed to nicotine across 10 d, we observed limited ATP release activity in the MHb at baseline, similar to that found in the naive tissue (Fig. 1E). Interestingly, re-exposure to nicotine during nicotine abstinence increased the occurrence of fluorescent transients across the duration of recording (Fig. 1F). These observations indicate that nicotine induces ATP release in the MHb.
Nicotine exposure induces extracellular ATP release in the MHb. MHb sections expressing the GRABATP sensor were imaged to monitor ATP release activity (n = 4 subjects). Individual traces show fluorescence signals. A, Illustration of GRABATP sensor AAV bilateral injection into the MHb. B, Illustration of GRABATP sensor expressed in the MHb of the brain, in which the presence of extracellular ATP induces a change in fluorescence signal. C, At baseline with aCSF perfusion, little to no changes in fluorescence signal were detected. D, During 1 μm nicotine aCSF perfusion, changes in fluorescence signal were observed in the MHb. E, Mice were treated with 1.5 mg/kg nicotine for 10 d, and 24 h later, brain sections were examined for GRABATP fluorescence. During abstinence with aCSF perfusion, few to no changes in fluorescence signal were detected. F, During 1 μm nicotine aCSF perfusion, nicotine re-exposure induced changes in fluorescence signal in the MHb.
Antagonism of purinergic signaling demonstrates interaction with nicotine
Next, we sought to examine whether endogenous ATP signaling alters neuronal firing in the MHb at baseline. Brain sections were perfused with aCSF containing the following: (1) the P2 general receptor antagonist suramin, to block endogenous purinergic signaling; (2) nicotine alone; or (3) coperfusion of nicotine and suramin. Data analysis revealed significant differences among the groups (repeated-measures, one-way ANOVA: F(3,21) = 8.178, p = 0.0009, R2 = 0.5388; Fig. 2A–C). No significant difference in the firing frequency was found between baseline and suramin perfusion (p = 0.9998). This finding is consistent with the results of the GRABATP sensor imaging, in which limited ATP release was observed without the presence of nicotine. Therefore, given that nicotine exposure appeared to elevate ATP signaling with the GRABATP sensor, we then investigated the effect of blocking the action of nicotine-mediated ATP release on habenular neuronal activity by coperfusing suramin and nicotine. Nicotine induced a significant increase in neuronal firing, compared with baseline levels (p = 0.002) and when suramin only is administered (p = 0.0024). Interestingly, suramin prevented the nicotine-mediated elevation in firing frequency, which was significantly lower from the level of firing found during nicotine alone (p = 0.0082; Fig. 2B,C). It should be noted that both males and females were used in these studies, but no significant differences were found between sexes in the effect of suramin on nicotine-mediated firing (Table 1). Moreover, one prior study suggested that higher concentrations of suramin may act on NMDA and AMPA receptors (Suzuki et al., 2004), but no statistically significant effects were found at 30 μm in their study, which was the dose used herein. Nevertheless, we conducted a control experiment to assess suramin alone, and, as expected, no differences were observed in neuronal firing [paired t test: t(4) = 0.5663, p = 0.6015, R2 = 0.07421; baseline (mean ± SEM), 2.682 ± 0.6243; suramin, 2.840 ± 0.7833]. Therefore, these findings indicate an interaction between purinergic signaling and cholinergic signaling in MHb cholinergic neurons.
Statistical analyses of the change from baseline values for each treatment, comparing between males and females
Perfusion of the P2 general antagonist prevented the nicotine-mediated firing frequency increase in MHb neurons. The general P2 antagonist suramin was perfused to examine the effects of endogenous ATP signaling on neuronal firing in the MHb (n = 8 cells from 5 subjects). A, Recordings were obtained from neurons located in the ventral part of the MHb. B, Representative traces during suramin and nicotine perfusion and coperfusion. C, Suramin alone did not elicit any significant effect on the firing frequency, whereas nicotine substantially increased firing frequency. When suramin and nicotine were coapplied, blockade of the P2 receptors prevented the ability of nicotine to increase firing frequency in the MHb neurons. **p < 0.001, compared with other conditions indicated. Data are presented as the mean ± SEM.
ATP facilitates neuronal firing via P2X and P2Y1 receptors
To determine whether ATP alone can alter neuronal activity, we next examined the effects of the P2X and P2Y1 receptor agonists on firing frequency. It was previously demonstrated that activation of P2X receptors leads to influx of Na+ and Ca2+ (Pankratov et al., 2009; Yang and Liang, 2012; Burnstock, 2015). Therefore, activating the P2X receptors would be expected to induce an excitatory effect on cholinergic neuronal activity in the MHb. For all of the studies, ventral MHb neurons were validated as cholinergic based on an increase in neuronal activity with nicotine application following the experimental recordings, given the known nAChR expression in this cellular population (Fowler et al., 2011; Görlich et al., 2013; Dao et al., 2014; Shih et al., 2014). As for P2Y1 receptors, it has been shown that activation of these membrane receptors increases intracellular Ca2+ through G-protein-coupled signaling (Abbracchio et al., 2006; Guzman and Gerevich, 2016). Furthermore, in mesencephalic trigeminal neurons, ATP was demonstrated to act on P2Y1 receptors to facilitate HCN channel currents (Huang et al., 2010), so such an interaction may underlie the findings observed in the current study. In sum, the current evidence from the field leads to the hypothesis that activation of both P2X and P2Y1 receptors would result in an elevation of habenular neuronal firing frequency. As a first control measure to validate the approach, we applied vehicle (0.9% saline) with picospritzer application, and no significant change was found in the firing frequency (paired t test: t(4) = 1.730, p = 0.1587, R2 = 0.4280; Fig. 3A,B). No differences were observed between males and females in this assessment (Table 1). Next, to investigate the effects of the P2X receptor, we applied the general agonist, αβ-methylene ATP and found a significant increase in firing frequency with the P2X receptor activation in the habenular cholinergic neurons (paired t test: t(13) = 3.886, p = 0.0019, R2 = 0.5374; Fig. 3C,D). Thereafter, we examined the function of the P2Y1 receptor with the specific agonist MRS2365 and observed a significant increase in firing frequency in the cholinergic neurons (paired t test: t(12) = 5.545, p = 0.0001, R2 = 0.7193; Fig. 3E,F). Together, these data indicate that activation of either the P2X or P2Y1 receptors can induce an increase in MHb activity independent of the presence of nicotine.
Activation of the P2X or P2Y1 receptors increases firing frequency in cholinergic neurons. Ventral MHb neuron activity was recorded, and then the agonists for the P2X receptor and P2Y1 receptor were applied to examine changes in neuronal firing. A, Representative trace of neuronal activity during vehicle application. The arrow indicates the time point of vehicle application. B, Vehicle application did not alter firing frequency across time (n = 5 cells from 4 subjects). C, Representative trace of the firing frequency change with P2X agonist application. The arrow indicates the time point of αβ-methylene ATP application. D, The P2X receptor agonist significantly increased firing frequency in neurons (n = 14 cells from 8 subjects). E, Representative trace of the firing frequency change during P2Y1 agonist application. The arrow indicates the time point of MRS2365 application. F, The activation of P2Y1 receptors significantly increased firing frequency (n = 13 cells from 10 subjects). **p < 0.01, ***p < 0.001, compared with baseline. Data are presented as the mean ± SEM.
To further investigate whether ATP differentially affects neuronal activity during nicotine abstinence, we then examined the effects of the agonists in mice treated with nicotine across 10 consecutive days prior (nicotine abstinence condition). Specifically, brain sections were examined at 24 h after the last nicotine treatment, in the absence of nicotine. Application of the P2X receptor agonist (paired t test: t(11) = 4.377, p = 0.0011, R2 = 0.6352; Fig. 4A,B) or the P2Y1 receptor agonist (paired t test: t(6) = 3.679, p = 0.0103, R2 = 0.6929; Fig. 4C,D) significantly increased firing frequencies in cholinergic neurons during nicotine abstinence. Of note, no difference was found in these studies between males and females for both the effects of P2X or P2Y1 agonists under naive or nicotine abstinence conditions (Table 1). To further investigate whether the effects were greater with P2X or P2Y1 agonist application, we then derived the change in neuronal firing from baseline and examined it across datasets, but no statistically significant differences were found (mean ± SEM; nicotine naive state: P2X agonist, 2.586 ± 0.6656; P2Y1 agonist, 3.888 ± 0.7603; nicotine abstinence state: P2X agonist, 2.835 ± 0.6477; P2Y1 agonist, 3.644 ± 0.9903; one way ANOVA: F(3,43) = 0.7546, p = 0.5258, R2 = 0.05,001).
Activation of the P2X or P2Y1 receptors during nicotine abstinence increases firing frequency. MHb neuronal activity was recorded 24 h after the last nicotine injection, and the agonists of the P2X or P2Y1 receptors were applied. A, Representative trace of the firing frequency change during P2X agonist application. The arrow indicates the time point of αβ-methylene ATP application. B, The P2X receptor agonist significantly increased firing frequency during nicotine abstinence (n = 12 cells from 8 subjects). C, Representative trace of the firing frequency change during P2Y1 agonist application. The arrow indicates the time point of MRS2365 application. D, The P2Y1 receptor agonist significantly increased firing frequency during nicotine abstinence (n = 7 cells from 7 subjects). *p < 0.05, **p < 0.01, compared with baseline. Data are presented as the mean ± SEM.
Interaction between nicotinic and ATP signaling during acute nicotine exposure
A prior study reported a potential mutual inhibition between P2X receptors and cholinergic signaling in HEK-293 cells and cultured myenteric neurons (Decker and Galligan, 2009). Specifically, application of ATP or acetylcholine induced independent effects on inward cell currents, but when both were applied together, a summative effect was not found in the cellular response (Decker and Galligan, 2009), leading to the conclusion of mutual inhibition. Given this and our findings above, we hypothesized that the mutual inhibition between P2X and cholinergic signaling would limit the effects of the P2X agonist on neuronal activity. Consistent with this expectation, application of αβ-methylene ATP did not induce a significant change in firing frequency during nicotine perfusion (repeated-measures, one-way ANOVA: F(2,24) = 14.29, p < 0.0001, R2 = 0.3191; Fig. 5A,B). Post hoc test results indicate that while nicotine perfusion increased firing frequency above baseline levels (baseline vs nicotine, p = 0.0091; baseline vs nicotine with αβ-methylene ATP, p < 0.0001), no difference was found between nicotine and coexposure to nicotine and αβ-methylene ATP (p = 0.4550). Furthermore, no significance differences were found between males and females during nicotine perfusion with the P2X agonist (Table 1). We next examined whether a similar effect could be found between cholinergic and P2Y1 receptor mediated signaling, and significant differences were found among the treatment groups (repeated-measures, one-way ANOVA: F(2,20) = 21.14, p < 0.0001, R2 = 0.6789; Fig. 5C,D). The post hoc analysis revealed a significant increase in firing frequency during nicotine exposure (p = 0.0446), and a further statistically significant increase with application of the P2Y1 agonist MRS2365 during nicotine perfusion (p = 0.0026). The cotreatment group also differed from baseline firing frequency levels (p < 0.0001). No sex differences were found with coadministration of the P2Y1 receptor and nicotine (Table 1). Together, these data support the notion of mutual inhibition between the P2X receptor and nicotinic receptors, but not the P2Y1 receptor.
Coexposure to acute nicotine and the P2Y1 receptor agonist potentiates neuronal firing in cholinergic MHb neurons. The interaction between purinergic signaling and nicotinic signaling was examined by applying the purinergic receptor agonists to MHb cholinergic neurons during nicotine perfusion. A, Representative trace of neuronal firing when the P2X agonist was applied during nicotine perfusion. The arrow indicates the time point of αβ-methylene ATP application. B, The P2X receptor agonist did not further elevate the firing frequency during nicotine perfusion (n = 13 cells from 11 subjects). C, Representative trace of the neuronal firing change with the P2Y1 agonist application during nicotine perfusion. The arrow indicates the time point of MRS2365 application. D, P2Y1 receptor agonist application further elevated the firing frequency during nicotine perfusion (n = 11 cells from 5 subjects). *p < 0.05, **p < 0.01, ***p < 0.001, compared with other conditions indicated. Data are presented as the mean ± SEM.
Chronic nicotine exposure alters the interactive effects of purinergic signaling and nicotine
Given the importance of cholinergic MHb neurons in mediating nicotine withdrawal and previously documented changes in cellular activity following chronic nicotine exposure (Görlich et al., 2013; Dao et al., 2014), we sought to determine whether the interaction between purinergic and cholinergic signaling mediates the altered state of circuit function. Following 10 d of nicotine exposure, the effects of P2X agonist were examined with coapplication of nicotine and αβ-methylene ATP (repeated-measures, one-way ANOVA: F(2,28) = 39.11, p = 0.0151, R2 = 0.2555). Nicotine increased the firing frequency from baseline (mean ± SEM; baseline, 0.330 ± 0.162; nicotine, 3.79 ± 0.615; p = 0.0020), and this effect was further potentiated in the presence of αβ-methylene ATP and nicotine (baseline, 0.331 ± 0.162; nicotine and αβ-methylene ATP, 8.35 ± 1.22; p < 0.0001). Coexposure of nicotine and αβ-methylene ATP was also significantly different from baseline (nicotine, 3.791 ± 0.615; nicotine and αβ-methylene ATP, 8.347 ± 1.22; p < 0.0001). However, when these data were separated based on sex, we observed relevant sex-dependent differences (Fig. 6). Specifically, the potentiated firing frequency in the presence of nicotine and αβ-methylene ATP was strikingly observed in females (repeated-measures, one-way ANOVA: F(2,14) = 34.03, p < 0.0001, R2 = 0.8302; post hoc, baseline vs nicotine, p = 0.0494; baseline vs nicotine and αβ-methylene ATP, p < 0.0001; nicotine vs nicotine and αβ-methylene ATP, p = 0.0002; Fig. 6A,B). However, a significant potentiated effect was not found in males (repeated-measures, one-way ANOVA: F(2,12) = 13.04, p = 0.0001, R2 = 0.6848; post hoc, baseline vs nicotine, p = 0.0232; baseline vs nicotine plus αβ-methylene ATP, p = 0.0008; nicotine vs nicotine plus αβ-methylene ATP, p = 0.1636; Fig. 6C,D). Thus, we next compared the change in firing frequency by subtracting the values for nicotine and αβ-methylene ATP from the values for nicotine alone. The change in firing frequency was significantly higher in females compared with males (unpaired t test: t(13) = 3.422, p = 0.0045, R2 = 0.4739; Fig. 6E), thereby demonstrating a notable sex difference in cholinergic neuron response. Finally, the effects of the P2Y1 receptor agonist were also examined with nicotine re-exposure during abstinence (repeated-measures, one-way ANOVA: F(2,24) = 9.927, p = 0.0007, R2 = 0.4527; Fig. 6F,G). The post hoc analysis revealed a significant increase in neuronal firing frequency with nicotine (p = 0.0054) and coexposure of nicotine and MRS2365 (p = 0.0010). However, there was no difference in the response to nicotine, either with or without MRS2365 (p = 0.7789). Further, no significant difference was found between males and females during abstinence with the P2Y1 receptor agonist (Table 1). Together, these results reveal an ATP receptor and sex-specific interactive effect with nicotinic signaling in the MHb.
During nicotine abstinence, coexposure to nicotine and the P2X receptor agonist potentiates neuronal firing in female, but not male, cholinergic MHb neurons. The interaction between purinergic signaling and cholinergic signaling during nicotine abstinence (24 h post-nicotine exposure) was examined by applying purinergic receptor agonists during nicotine perfusion. A, Representative trace of neuronal firing change in females when the P2X agonist was applied during nicotine perfusion. The arrow indicates the time point of αβ-methylene ATP application. B, In females, the P2X receptor agonist significantly potentiated the firing frequency during nicotine perfusion (n = 8 cells from 6 subjects). C, Representative trace of neuronal firing in males when the P2X agonist was applied during nicotine perfusion. The arrow indicates the time point of αβ-methylene ATP application. D, In males, application of the P2X receptor agonist did not significantly elevate the firing frequency during nicotine perfusion (n = 7 cells from 6 subjects). E, When comparing the P2X agonist effects with nicotine during abstinence in the male and female adolescent mice, the females exhibited a significantly higher firing frequency elevation than that found in males. F, Representative trace of the neuronal firing when the P2Y1 receptor agonist was applied during nicotine perfusion. The arrow indicates the time point of MRS2365. G, Application of P2Y1 receptor agonist did not elevate the firing frequency during nicotine perfusion in males and females (n = 13 cells from 12 subjects). *p < 0.05, **p < 0.01, ***p < 0.001, ***p < 0.0001, compared with other conditions indicated. Data are presented as the mean ± SEM.
Discussion
These studies reveal that purinergic signaling selectively modulates neuronal activity in the MHb and signaling dynamics become dysregulated with chronic nicotine exposure in the adolescent brain. The effects nicotine appear to be interactive with purinergic signaling in MHb cholinergic neurons. Both the P2X and P2Y1 receptors independently increased cholinergic neuron firing during a naive or nicotine abstinence state. However, in the presence of nicotine, a dramatically different effect was observed, which was based on sex and a history of drug exposure. Specifically, for the P2X receptor, the agonist had no observable effect on nicotine-mediated activity, but during nicotine abstinence, exposure to nicotine and the P2X receptor agonist induced a potentiated effect, but only in adolescent females. In contrast, with activation of the P2Y1 receptor, a potentiated effect of the agonist and nicotine was found in nicotine-naive subjects, but not following chronic nicotine exposure. This suggests that mutual inhibition between purinergic and nicotinic signaling may occur during the first exposure to nicotine with P2X receptor activation and following chronic nicotine exposure with the P2Y1 receptor. Together, these data provide a complex understanding of the interactions between purinergic and nicotinic signaling in habenular cholinergic neurons during adolescence and further reveal the impact of nicotine on extracellular ATP signaling and subsequent altered circuit function.
Independent effects of P2X and P2Y1 receptors in the MHb
Activation of P2X receptors induced an excitatory effect on cholinergic neuronal activity in the MHb in naive subjects. These studies are the first to reveal this functional role of purinergic signaling in the MHb on action potential firing. Previous studies have investigated P2X receptors in synaptic transmission in other brain regions. For instance, in the hippocampus, a P2X receptor agonist inhibited NMDA receptor-mediated excitatory postsynaptic currents, whereas a P2 antagonist facilitated the formation of long-term potentiation (Pankratov et al., 2002). It has been proposed that the influx of Ca2+ through the P2X channel leads to calcium-dependent NMDA inactivation/desensitization (Legendre et al., 1993; Zhang et al., 1998). Of relevance, recent cell-specific investigations have identified functional habenular NDMA receptors (Otsu et al., 2018), suggesting a potential link. However, in our studies, we examined the immediate effects of P2X receptor activation on cellular firing, which would have occurred in a more discrete time frame than that required for NMDA receptor modulation through downstream intracellular signaling processes. Furthermore, we observed an increase in neuronal activity with the agonist, not a decrease, as found in the prior study (Pankratov et al., 2002), thereby suggesting cell type-specific effects. Next, activation of presynaptic P2X receptors could have facilitated glutamate release, as demonstrated in the hippocampus (Sperlágh et al., 2002; Khakh et al., 2003; Rodrigues et al., 2005), which would thereby increase AMPA and/or NMDA receptor activation. However, activation of presynaptic P2Y1 receptors has been shown to dampen glutamatergic transmission in the hippocampus (Rodrigues et al., 2005), whereas activation of postsynaptic P2Y1 receptors inhibited NMDA receptor-mediated currents in the prefrontal and parietal cortex (Luthardt et al., 2003). Given these opposing effects, one would expect that each agonist for the P2X or P2Y receptors would have opposite effects on habenular activity. However, we found increased neuronal activity with either the P2X or P2Y1 agonist, suggesting that different mechanisms likely predominate in the MHb. ATP has been shown to act on P2Y1 receptors to facilitate HCN channel currents in mesencephalic trigeminal neurons (Huang et al., 2010). Given the function of HCN channels in maintaining spontaneous MHb neuronal activity (Görlich et al., 2013), the increased neuronal firing evidenced with activation of the P2Y1 receptor could involve interaction with habenular HCN channel signaling via downstream calcium-mediated mechanisms, a possible interaction that will need to be tested further.
Interaction between purinergic signaling and acute nicotine
While the P2X receptor agonist and nicotine independently increased neuronal firing, we did not observe an additive effect on firing frequency during coexposure. This finding could potentially be because of either a ceiling-related effect with cellular activity, or mutual inhibition. Given that higher firing frequencies have been observed in habenular neurons with other experiments (e.g., during nicotine abstinence herein and as shown previously; Görlich et al., 2013), a putative ceiling-related effect can be excluded. Interestingly, mutual inhibition between P2X and cholinergic signaling has been suggested (Khakh et al., 2000; Decker and Galligan, 2009). Coapplication of ATP and acetylcholine led to a significantly lower current than the predictive summation when ATP or acetylcholine were individually applied, as demonstrated in HEK293 cells and Xenopus oocytes expressing α3β4 nAChRs and P2X receptors (Khakh et al., 2000; Decker and Galligan, 2009). Another study in acutely dissociated sympathetic neurons also demonstrated mutual inhibition between P2X receptors and nAChR-mediated signaling (Searl et al., 1998). However, it should be noted that such inhibitory effects have not been documented in all cell types that express both purinergic and nicotinic receptors, including visceral sensory ganglion (Reyes et al., 2006) and cultured sympathetic neurons (Rogers et al., 1997). Such discrepancies are likely because of the different expression profile of P2X receptor and nAChR subtypes across cellular populations. Since α3β4 nAChRs and P2X2 receptors are both expressed in the MHb (Zoli et al., 1995; Kanjhan et al., 1999), consistent with the cellular populations demonstrating mutual inhibition (Khakh et al., 2000; Decker and Galligan, 2009), these findings support the conclusion of mutual inhibition for the effects observed in our study.
Interaction between purinergic signaling and nicotine following chronic nicotine treatment
The P2X agonist induced a potentiated effect on neuronal firing during re-exposure to nicotine in female subjects undergoing nicotine abstinence. One of the possible mechanisms for this finding is the interaction between purinergic and neurokinin signaling. Neurokinin receptor activation can modulate nicotine-induced firing frequency in MHb neurons and has been linked to nicotine withdrawal (Dao et al., 2014). Further, neurokinin has been shown to potentiate ATP-mediated currents in dorsal root ganglion neurons (Wang et al., 2001). However, while this prior study included both males and females (Dao et al., 2014), potential sex differences were not examined, and as such, this would not explain why we only observed a potentiated response in females. Moreover, it should be noted that neurokinin would be expected to be present during the acute nicotine application, for which we did not find a potentiated effect. Thus, it remains to be determined whether chronic nicotine induces changes in neurokinin signaling that could explain the differences found during the abstinence state, or whether such nicotine exposure induces sex-dependent effects on the expression of ATP and/or nicotinic signaling mechanisms independently. For instance, it has been demonstrated that, in the interpeduncular nucleus during nicotine withdrawal, females exhibit an increase in expression of the α5 nAChR subunit, while males have an increase in the expression of the α2 and α3 nAChR subunits (Correa et al., 2019). Females also exhibit greater withdrawal-induced cellular activation in the central region of the interpeduncular nucleus (Correa et al., 2019; Matos-Ocasio et al., 2021) and differ in the response to the aversive properties of nicotine (Torres et al., 2009). Thus, these changes may lead to differences in circuit function, thereby altering nicotine dependence susceptibility. Along these lines, it will be interesting in future studies to examine whether purinergic signaling is also involved in directly mediating nicotine self-administration. In humans, women suffer more from adverse symptoms during nicotine withdrawal and are more likely to relapse during quit attempts (Swan et al., 1993; Bjornson et al., 1995; Rojas et al., 1998). Thus, the current findings provide mechanistic insight into a putative sensitized state in the female brain during nicotine withdrawal, in which nicotine re-exposure can lead to altered neuronal activity to thereby promote relapse.
Interestingly, for P2Y1 receptor-mediated signaling, we did not observe any potentiation in neuronal activity with nicotine and the agonist during nicotine withdrawal, contrary to that observed during acute nicotine exposure. This provides evidence that chronic nicotine exposure altered the cellular response, leading to an effect that is reminiscent of the mutual inhibition observed with the P2X receptor in naive subjects. However, given that the P2X is an ionotropic receptor and P2Y1 is a metabotropic receptor, the molecular interactions resulting in the effect on neuronal firing likely differ. Thus, it will be of interest in future studies to examine downstream intracellular signaling mechanisms.
Considerations
In these studies, we reveal a sex difference in potentiated firing frequency with the activation of both the nAChRs and P2X receptors, which was significantly higher in females compared with males. Given that a complex experimental design to investigate sex differences across treatment groups would require high power demands to reach statistical significance with reasonable sample sizes (e.g., for two-way or three-way ANOVAs), we used a normalization approach to allow us to directly compare between sexes. Specifically, the values for nicotine and αβ-methylene ATP were subtracted from the values of nicotine alone to derive the potentiated difference in firing frequency. These values were then statistically examined with a t test, allowing for direct comparison to determine a sex difference. These studies also examined one dose of nicotine to induce withdrawal with controlled, subcutaneous injections. It will be informative in future studies to examine across a variety of nicotine doses to determine dose-dependent effects, as well as examine with other routes of nicotine exposure, such as e-cigarette vape inhalation.
Conclusions
The findings reported herein have multiple implications beyond that discussed above. For instance, the endogenous neurotransmitter acetylcholine acts on nAChRs in the MHb, and, as such, activation of nAChRs with the receptor agonist nicotine provides further potential insight into endogenous circuit function. Second, humans consume nicotine recreationally, and, as such, these findings add an important mechanistic understanding of the effects of nicotine on brain circuit activity. We focused our studies in adolescent mice since increased e-cigarette use in teenagers has become prevalent. Adolescents are more prone to develop nicotine dependence, and the impact of nicotine on the adolescent developing brain may lead to long-term changes in neural function and mental health (Luna et al., 2010; National Center for Chronic Disease Prevention and Health Promotion, 2016; Siqueira, 2017; Grant et al., 2019; Ren and Lotfipour, 2019; Mahajan et al., 2021). It will be interesting in future studies to examine whether these effects of nicotine on extracellular ATP signaling mechanisms persist into adulthood. Together, these findings reveal an interaction between extracellular ATP signaling and the actions of nicotine and further identify specific purinergic receptors that could be targeted for future therapeutic development with receptor subtype-selective drugs.
Footnotes
This work was supported by National Institutes of Health National Institute on Drug Abuse Grants DP1-DA-039658 and R01-DA-051831 (to C.D.F.) and Tobacco-Related Disease Research Program Grant T31R1767 (to C.D.F.).
The authors declare no competing financial interests.
- Correspondence should be addressed to Christie D. Fowler at cdfowler{at}uci.edu
