|Home | About | Journals | Submit | Contact Us | Français|
Identifying the neurological mechanisms underlying nicotine reinforcement is a healthcare imperative if society is to effectively combat tobacco addiction. The majority of studies of the neurobiology of addiction have focused on dopamine (DA)-containing neurons of the ventral tegmental area (VTA). However, recent data suggests that neurons of the laterodorsal tegmental (LDT) nucleus which sends cholinergic, GABAergic- and glutamatergic-containing projections to DA-containing neurons of the VTA are critical to gating normal functioning of this nucleus. The actions of nicotine on LDT neurons are unknown. We addressed this issue by examination of nicotinic effects on identified cholinergic and non-cholinergic LDT neurons using whole-cell patch clamp and Ca2+ imaging methods in brain slices from mice (P12-P45). Nicotine applied via puffer pipette or bath superfusion elicited membrane depolarization that often induced firing and TTX-resistant inward currents. Nicotine also enhanced sensitivity to injected current; and, baseline changes in intracellular calcium were elicited in the dendrites of some cholinergic LDT cells. Additionally, activity-dependent calcium transients were increased, suggesting that nicotine exposure sufficient to induce firing may lead to enhancement of levels of intracellular calcium. Nicotine also had strong actions on glutamate and GABA-releasing presynaptic terminals since it greatly increased the frequency of miniature EPSCs and IPSCs to both cholinergic and non-cholinergic neurons. Utilization of nAChR subunit antagonists revealed that presynaptic, inhibitory terminals on cholinergic neurons were activated by receptors containing α7, β2 and non-α7 subunits; whereas, presynaptic glutamatergic terminals were activated by nAChRs comprised of non-α7 subunits. We also found that direct nicotinic actions on cholinergic LDT neurons were mediated by receptors containing α7, β2 and non-α7 subunits. These findings lead us to suggest that nicotine exposure from smoking will enhance both the excitability and synaptic modulation of cholinergic and non-cholinergic LDT neurons and increase their signature neurotransmitter outflow to target regions including the VTA. This may reinforce the direct actions of this drug within reward circuitry and contribute to encoding stimulus saliency.
Nicotine is the addictive ingredient that drives increasing numbers of people to tobacco dependence. Understanding the mechanisms by which nicotine acts in the brain is fundamental to developing effective therapies for nicotine addiction, and may also have implications for understanding the neural systems underlying addiction to other drugs of abuse. Investigations of the neurobiology of addiction to nicotine have so far largely focused on the mechanisms that underlie its reinforcing effects within the mesolimbic dopaminergic pathway which encompasses the ventral tegmental area (VTA; Corrigall et 1994; Laviolette and van der Kooy, 2003; Nisell et al, 1994) along with its projections to the nucleus accumbens (NAcc; Di Chiara and Imperato, 1988; Koob, 1996; Nestler, 2001; Redgrave et al, 1999) and the pre-frontal cortex (Arnsten, 1997; Brozoski et al, 1979; Carr and Sesack, 2000a; Goldman-Rakic, 1995). This pathway, traditionally viewed as reward circuitry, responds to salient stimuli, and a major factor contributing to tobacco addiction is nicotine-induced modulation of dopamine (DA) release within this pathway (Dani and Heinemann, 1996; Di Chiara and Imperato, 1988; Koob, 1992; Nestler, 1992; Nisell et al, 1994; Picciotto, 1998; Picciotto et al, 1998; Pidoplichko et al, 1997). While activation of nicotinic acetylcholine receptors (nAChRs) located in the VTA is critical for nicotine addiction (Calabresi et al, 1989; Clarke, 1993; Maskos et al, 2005; Picciotto et al, 1998; Pidoplichko et al, 1997; Pons et al, 2008; Wada et al, 1989), synaptic mechanisms both upstream and downstream of DA release from the VTA likely contribute to the etiology of nicotine addiction (Maskos, 2008; Mena-Segovia et al, 2008; Williams and Adinoff, 2008).
One likely important upstream regulator of the VTA is the pontine laterodorsal tegmental nucleus (LDT), which is a major source of the endogenous cholinergic input to the nAChRs and muscarinic receptors within the established VTA reward circuitry (Forster et al, 2002b; Mansvelder et al, 2002). Recent studies have demonstrated that substantial direct input from cholinergic afferents targets DA-containing cells of the mesoaccumbal pathway (Omelchenko and Sesack, 2006) and the morphology and synaptology of cholinergic axons within the VTA are similar to those observed for the projections arising from the LDT (Omelchenko and Sesack, 2005). While terminals that contain acetylcholine make up a substantial portion of the LDT-derived innervation to the VTA, data suggest that GABA and glutamate afferents to DA neurons also arise from the LDT (Omelchenko and Sesack, 2006). Neurochemical studies indicate these projections are functional since stimulation of the LDT results in dopamine efflux in the NAcc via cholinergic and glutamatergic mechanisms (Forster and Blaha, 2000). Moreover, a normally functioning LDT is required for the burst firing of dopaminergic VTA neurons (Lodge and Grace, 2006), the functionally relevant firing pattern that signals stimulus salience (Berridge and Robinson, 1998; Grace, 1991; Schultz, 1998). Finally, the LDT is involved in development of behavioral sensitization to amphetamine in rats, a finding which suggests the occurrence of a drug-induced, neuronal plasticity in the LDT (Nelson et al, 2007). Based on these and other studies (Alderson et al, 2005; Forster and Blaha, 2000; Forster et al, 2002a; Forster et al, 2002b), the LDT is implicated in the neurobiology of addiction, in addition to its apparent role in arousal (Hobson and Pace-Schott, 2002; Jones, 2004; Pace-Schott and Hobson, 2002).
However, the electrophysiological responses of cells within this nucleus to drugs of abuse had not been previously described. Since mRNAs for nAChR subunits are present in LDT neurons (Azam et al, 2003), we investigated the direct and indirect actions of nicotine on membrane potential, intracellular calcium levels, and synaptic activity in cholinergic and non-cholinergic LDT cells in mouse brain slices. We found that nicotine has excitatory actions in the majority of cells within the LDT, and that these actions varied depending on cell type and on anatomical location of the cell. As it has been shown that exogenous stimulation of this pathway stimulates DA release from VTA neurons (Forster and Blaha, 2000), we hypothesize that nicotine exposure from smoking will enhance both the excitability and synaptic modulation of cholinergic LDT neurons and thereby increase cholinergic outflow to their target regions including the VTA. These, and the actions on non-cholinergic LDT neurons could, therefore, play a role in the neuronal underpinnings of nicotine addiction.
All procedures complied with NIH guidelines for ethical use of animals. Additionally, all procedures were reviewed and approved by the Institutional Animal Care and Use Committee of New York Medical College.
Brain slices were prepared from 12 to 45 day old male and female C57BL/6NTac mice (Taconic, New York) as previously described (Kohlmeier and Leonard, 2006). Animals were decapitated following induction of deep anesthesia with isofluorane. A block of the brain containing the LDT was rapidly removed and incubated in ice-cold ACSF, which contained (in mM) 124 NaCl, 5 KCl, 1.2 NaH2PO4, 2.7 CaCl2, 1.2 MgSO4, 26 NaHCO3 and 10 dextrose (295-305 mOsm), and was oxygenated by bubbling with carbogen (95% O2, 5% CO2). The brainstem was then blocked in a coronal plane and sectioned at 250 μm on a Leica vibrotome (VT1000S). Slices containing the LDT were incubated at 35 °C for 15 min and then stored at room temperature. Recordings were obtained from slices submerged in a recording chamber, which was perfused at 1–3 ml/min with the continuously oxygenated ACSF at room temperature.
Nicotine bitartrate (Sigma, USA) was prepared with minimal light exposure on the day of the experiment to final concentration. Nicotine was applied via two different methods: via the bath at 1-10 μM for 1-5 minutes, or, via a “puff” with controlled pressure pulses (10 mM, 35 - 1000 ms, 137.9 kPa; Picospritzer II, General Valve Corp.) from a patch pipette positioned just above the slice but downstream from the cell from which recordings were being obtained. 1,1-Dimethyl-4-phenylpiperazinium iodide (DMPP, 50μM, Sigma), a specific nAChR agonist (Phelan and Gallagher, 1992) was applied via a puff (100ms, 137.9kPa). In those cases where calcium imaging was being conducted, to minimize any artifact due to moving of the dendrite, a lower puff pressure of 34.4kPa was utilized. ACh was applied utilizing the latter method (1 mM, 100 - 300 ms, 137.9 kPa) thus allowing very local application of drug with a quick wash in and wash out time.
Atropine, an irreversible muscarinic receptor antagonist, was obtained from Sigma and applied at a final concentration 5 μM. Atropine occasionally has actions at nAChRs (Zwart and Vijverberg, 1997); therefore, the lowest possible concentration of this antagonist should be utilized to try to limit blocking actions to muscarinic antagonism. In a separate series of experiments, we found that 0.5 μM failed to completely block ACh-mediated outward currents in cholinergic LDT neurons (2.4%-78.4% of outward current remaining, n=7/7), putatively mediated by activation of muscarinic receptors (Luebke et al, 1993: Leonard and Llinas, 1994). We therefore bath applied 5 μM atropine to assure that we completely blocked muscarinic responses, which is consistent with reports in these and other cells (Kohlmeier and Leonard, 2006; Zhang and Warren, 2002). Methyllycaconitine citrate (MLA, 10 nM, Sigma), a selective antagonist of α7 subunit-containing nAChRs (Alkondon et al, 1992; McGehee and Role, 1995; Seguela et al, 1993), mecamylamine (MEC, 1 μM, Sigma), a selective antagonist of non-α7 subunit-containing nAChRs (McGehee and Role, 1995) and dihydro-β-erythroidine hydrobromide (DHβE, 500 nM, Sigma), another non-α7 antagonist which is selective for β2-containing nAChRs (Alkondon and Albuquerque, 1993; Luetje et al, 1990) were dissolved in ACSF at a stock concentration and dissolved to final concentration in ACSF the day of the experiment. TTX (500 nM, Alamone) was dissolved in the superfused ACSF to block voltage gated sodium channels. The ionotropic receptor antagonists 6,7-Dinitroquinoxaline-2,3(1H,4H)-dione (DNQX,15 μM, Sigma), D(−)-2-Amino-5-phosphonopentanoic acid (APV, 50 μM, Sigma), bicuculline (10 μM, Sigma) or SR-95531 (Gabazine, 20 μM, Sigma) and strychnine (2.5 μM, Sigma) were added to ACSF in some recordings (Burlet et al, 2002). For low calcium solution recordings, the [Ca2+] of the ACSF was buffered to < 20 μM by the addition of 2.7 mM EGTA (calculated with Patcher's Power Tools XOP for Igor Pro).
Micropipettes (2-4 MΩ) used for patch clamp recordings (Borosilicate, Catalogue number 8050, AM systems) were pulled on a horizontal puller (Sutter Instruments, P-2000). In those cases where calcium imaging was not being conducted, pipettes were filled with a recording solution containing (in mM) 144 K-gluconate, 0.2 EGTA, 3 MgCl2, 10 HEPES, 0.3 NaGTP and 4 Na2ATP (310 mOsm), or, in those cases where calcium imaging was being conducted, pipettes contained the potassium salt of Fura-2 (bis-fura 2, 50 μM, Molecular Probes) which was dissolved in the patch pipette solution of, (in mM) 144 K-gluconate, 3 MgCl2, 10 HEPES, 0.3 NaGTP, and 4 Na2ATP. Recordings conducted to examine IPSC activity were performed with a high chloride recording solution to optimize detection of inhibitory events by increasing the drive for, and reversing the polarity of, chloride-mediated currents (in mM) 144 KCl, 0.2 EGTA, 3 MgCl2, 10 HEPES, 0.3 NaGTP and 4 Na2ATP. In all cases, biotinylated Alexa-594 (25 μM) was also included in the patch solution in order that recovered cells could be histochemically identified following recordings.
Neurons were visualized for whole cell recordings at 160× magnification with visible-light, differential interference contrast optics, using a Nuvicon tube camera (Dage VE-1000) mounted on a fixed stage microscope (Olympus BX50WI). Cells for recording were chosen within the nucleus of the LDT which were identified using a 4× objective; however, locations of recorded neurons within the LDT was verified after the tissue had been processed for cell identification.
Seals of the pipette to the cellular membrane exceeding 1 gigaohm in resistance were obtained under visual control using an Axopatch 200B amplifer (Axon Instruments) operated in voltage-clamp mode and filtered at 2 or 5 KHz with a four-pole Bessel filter at the amplifier output and sampled at 4, 10 or 20 KHz. After establishing the whole-cell recording configuration, cells were filled with both the calcium indicator (when present) and biotinylated Alexa-594 by either passive diffusion or brief, hyperpolarizing pulses. Data were not collected until at least 20 minutes had passed following break-through to allow dye equilibration. Neurons were imaged through a 40× objective using a cooled, CCD camera equipped with a back illuminated EEV 57 frame transfer chip with an imaging area of 512 × 512 pixels (field size=160 μm/side; MicroMax, Roper Scientific). bis-Fura 2 and Alexa-594 were excited at 380 nm or 594nm, respectively, with light from a 75 watt Xenon lamp that was shuttered to reduce light exposure to the tissue in-between data acquisition. Recordings were either conducted in voltage clamp or “I-clamp fast” current clamp mode, following appropriate compensation for the pipette capacitance; quality of the cells was assayed by monitoring the holding current and the input resistance as determined by the voltage or current response to a brief, negative-going step. Recordings were uncorrected for liquid junction potentials which were calculated to be ~12mV. Current and voltage traces were digitized and command pulses were generated by custom designed software (TIWB; Inoue et al, 2001) executed by a Mac OS/G4 computer, which controlled an ITC-18 interface (Instrutech). TIWB software also controlled the camera and shutter, which allowed precise synchronization between electrophysiological and optical signals. The camera was read out through a 1 MHz, 14 bit A/D converter. Images were binned (4 × 4) on the chip and acquired in two different modes : 1) “discontinuous”, a low speed imaging mode utilized to monitor changes in calcium associated with depolarizations or inward currents during which an image was collected every 4 seconds or, 2) “continuous”, a high-speed imaging mode in which an image was collected every 50 ms, a rate fast enough to monitor changes in calcium accompanying rapid alterations of the membrane potential. Changes in fluorescence (dF/F) were quantified by the average pixel values within regions of interest (ROI) that were positioned on background- and baseline-subtracted fluorescent images. The background was determined from a ROI that was positioned at a location remote from the filled cell or its processes. Baseline fluorescence was determined as the fluorescence measured in the first few frames of each sequence prior to stimulation. Because dF/F traces recorded from dendrites have a low signal to noise ratio, each data point in this analysis represents the average of 10 frames. bis-Fura 2 fluorescence decreases when calcium increases and is bound by the dye; however, for purposes of clarity, dF/F responses have been inverted in figures such that positive dF/F traces indicate rises in calcium.
Current and voltage waveforms were analyzed and figures were prepared using Igor Pro Software (Wavemetrics, Lake Oswego, OR). Holding currents were calculated from one-second averages of the holding current at -60mV. Differences between means were compared using a two-sample, two-tailed t test corrected for multiple comparisons (when necessary) and repeated-measures ANOVA. Detection of PSCs and some analyses were done using Mini Analysis software (Synaptosoft, Decatur, GA). To compare the effects of nicotine on the number and amplitudes of PSCs, epochs of at least 30 seconds were used before application of nicotine (control) and following effects of nicotine (drug). Cumulative distributions of EPSCs or IPSCs were then compared for significant differences between the control and nicotine condition in individual cells using Kolmogorov-Smirnov (K-S test) statistics. Mean increases (% elevation over control) are reported for the population of cells that showed individual significance utilizing the K-S statistic. Since only effects on increasing PSC frequency were observed, a paired, one-tailed t-test was used to quantify changes in this parameter for histograms. A two-tailed, paired t-test was utilized to quantify changes in PSC amplitude for these data reported in histograms. Statistical significance was determined by p< 0.05. All numerical results are reported as mean ± SEM.
Cholinergic LDT neurons express high levels of neuronal nitric oxide synthase (nNOS) and presence of this enzyme can serve to identify cholinergic neurons in the LDT (Bredt et al, 1991; Vincent and Kimura, 1992; Vincent et al, 1983) and although by no means a homogeneous nucleus (Wang and Morales, 2009), presence of the cholinergic neurons has been proposed to delineate the nucleus (e.g. Leonard et al, 1995). During recordings, neurons were filled with Alexa-594, which passively diffused into the cell. Following the electrophysiological recordings, the tissue was fixed in 4% paraformaldehyde at 4°C and then stored in 0.01 phosphate buffered saline (PBS) and 30% sucrose for cryoprotection. Slices were then resectioned at 40 μm and incubated during gentle shaking at 4°C for 24 hours in darkness with a nNOS antibody (Sigma, Catalogue number N7280, rabbit polyclonal 1:400) followed by incubation with the FITC conjugated secondary antibody (nNOS A11008, goat, anti-rabbit, Molecular Probes). Further histochemical processing was not necessary to visualize the recorded, Alexa-594 filled cell which was viewed under a Texas-red filter-cube set.
All data in this report were obtained from neurons located within a region delineated by the cholinergic neurons in the LDT. As in previous studies (Kohlmeier and Leonard, 2006; Kohlmeier et al, 2008), cholinergic neurons from which recordings were obtained were those identified as both nNOS+ (Fig. 1A1) and Alexa-594+ (Fig. 1A2) under fluorescence optics (Fig. 1A, arrowhead). Non-cholinergic neurons from which recordings were obtained were identified as Alexa-594+ and nNOS-.
Bath application (1 min) of 10 μM nicotine resulted in a depolarization in cholinergic cells (6.8 ±1.5 mV, n=10/10) and non-cholinergic neurons (6.9 ± 1.5 mV, n=8/8) when recorded in current clamp mode. Depolarization was accompanied by a decrease in input resistance that was significant in both cell types (nNOS+ cells, 13.5 ± 6.6%, n=7; nNOS- cells, 26.4 ± 4.8%, n=7). Repeat actions of nicotine could be elicited if at least 15 minutes elapsed between applications. Often, bath application of nicotine produced sufficient depolarization to induce action potentials (n=10/18). Nicotine never induced a membrane hyperpolarization.
While delivery of nicotine via the bath may more accurately mimic the exposure profile of a smoker's brain by slowly ramping up and remaining constant for some time at the final drug concentration (Henningfield et al, 1993), the relatively gradual exposure to an increasing nicotine concentration may result in desensitization of the receptors, thus blunting the potential maximal response elicited by the agonist. Additionally, nicotine is known to have presynaptic actions in other nuclei where this has been studied. To facilitate detection of maximal responses and to limit responses to postsynaptic receptor activation, we examined actions of nicotine on membrane potential in the presence of blockers of GABA and ionotropic, glutamatergic receptors while utilizing a more rapid introduction of nicotine to the slice. Accordingly, 10 mM of nicotine was puffed onto cells (40 - 60 ms) via a picospritzer in the presence of blockade of the NMDA, AMPA, glycine and GABAA receptors with inclusion in the ACSF of APV, DNQX, strychnine and bicuculline, respectively. Nicotine, applied in this manner, induced depolarization in cholinergic (14.7 ± 2.2 mV, n=21/21; Fig. 1B) and non-cholinergic neurons (9.7 ± 4.4 mV, n=2/2). In some cases, depolarization was sufficient to induce action potentials (nNOS+ cells, n=15/21, in all cells recorded, n=16/23; Fig. 1B1). The nicotine-induced depolarization after puffer application was never reduced by TTX (21.8 ± 5.3 mV, n=9; unpaired t-test, p > 0.05; Fig. 1B, C); however, action potentials were abolished when TTX was included in the bath even in those cases where similar levels of depolarization of the membrane were achieved by nicotine (n=9; Fig. 1B2). Taken together, these data suggest that brief, local application of nicotine causes depolarization via activation of nicotinic receptors on the postsynaptic cell and that spikes elicited are mediated by activation of voltage-activated sodium channels, rather than voltage-operated calcium channels, which have been shown to be activated in these cells by other excitatory neuroactive substances (Kohlmeier et al, 2008).
In addition to a direct depolarizing effect, nicotine may also alter excitability of LDT neurons. To test this, we examined whether the sensitivity to injected current was altered by nicotine in these neurons. Cholinergic LDT neurons contain A-type outwardly rectifying potassium conductances (Kamondi et al, 1992; Leonard and Llinas 1990; Sanchez et al, 1998) which nicotine has been shown to inhibit (Hamon et al, 1997; Wang et al, 2000), resulting in heightening the excitability of the cell via an NMDA-involved mechanism as shown in cultured hippocampus neurons (Kim et al, 2007). To avoid this possible complication, we tested the ability of nicotine to alter the sensitivity to injected current by looking at firing rates when the cell was stepped to a more depolarized potential by the injection of current in the presence of blockade of excitatory and inhibitory transmission with APV, DNQX, strychnine and bicuculline. Using the same current step, and using DC current to bring the drug-treated cell back to baseline membrane potential to avoid contamination by voltage-activated conductances and the heightening of excitability induced by depolarization, we found that a slightly higher, but significant firing rate was achieved in the presence of nicotine (22.5% increase in firing frequency, n=9; nNOS+ cells; paired t-test, p < 0.05) (Fig. 1D, E). In other cell types, nicotine has been shown to lower the threshold for firing which may be one mechanism altering nicotine-induced sensitivity to injected current. However, the threshold for firing was not significantly altered during nicotine exposure in LDT neurons (control, -40.7 ± 0.8 mV; nicotine, -41.1 ± 1.1 mV; n=9; p > 0.05). These data indicate that nicotine does have specific actions on cellular excitability of LDT neurons. We conclude that the mechanism of nicotine-induced enhancement is not dependent on activation of the NMDA receptor, nor induction of EPSP or IPSC activity in the postsynaptic cell and did not depend on alteration of the threshold of action potential induction as has been one reported action of nicotine in other cell types (Kawai et al, 2007; Mansvelder et al, 2002).
In voltage clamp mode in the presence of bicuculline, APV, DNQX, strychnine and TTX, puffer-applied nicotine induced an inward current in all recorded cells (nNOS+ cells, 551.5 ± 113.6 pA, n=16; nNOS- cells, 598.1 ± 224.2 pA, n=6; Fig. 2A). Amplitudes of nicotine-induced inward currents were not statistically different between nNOS+ and nNOS- cells (p > 0.05). An inward current was also induced by the specific nAChR agonist DMPP. Puffer-applied DMPP (50 μM; 100ms; same recording conditions as above) elicited inward currents in all cholinergic LDT neurons recorded (33.6 ± 3.45 pA, n=9/9, p <0.05; Fig. 2B). Since nicotine-induced inward currents persisted in the presence of low calcium solution (1.5 mM Ca, 2.7 mM EGTA, n=5) these data indicate that nicotine-induced depolarization/inward currents resulted from a post-synaptic action rather than from stimulation of nicotinic receptors on presynaptic terminals.
We also found that bath application of 10 μM nicotine in conjunction with TTX and blockers of excitatory and inhibitory transmission, also resulted in inward currents in LDT neurons (15.9 ± 6.8 pA, n=12, data not shown). The nicotine-induced inward current was not significantly different between cell phenotypes (nNOS+ cells, 12.2 ± 4.7 pA, n=5; nNOS- cells, 22.7 ± 19.6 pA, n=4; p > 0.05). Nicotine-induced inward currents exhibited a reversal potential of -21.6 ± 3.2mV in nNOS+ cells (n=6) and -24.6 ± 2.5 mV in nNOS- cells (n=4) consistent, in both cell types, with the opening of cationic channels permeable to Na+, K+ and/or Ca2+ as has been reported for nAChRs.
The action of acetylcholine (ACh), the endogenous ligand for nAChRs was also examined. We and others have shown that cholinergic stimulation of LDT neurons results in activation of muscarinic receptors which activates an inwardly rectifying potassium conductance resulting in outward currents (Leonard and Llinas, 1994; Luebke et al, 1993; Kohlmeier Leonard 2006). ACh (Puff; 1 mM) induced a rapid inward current followed by an outward current in nNOS+ cells (Fig. 2C). To prevent ACh-induced outward current, we utilized atropine, which blocks muscarinic receptors. As expected, in the presence of 5 μM atropine, ACh elicited only an inward current in the same cholinergic cells in which in the absence of atropine an outward current was seen (78.3 ± 10.9 pA, n=30; Fig. 2C, D). The amplitude of the ACh-induced inward current was not statistically different from that elicited by ACh when atropine was absent (80.9 ± 16.8 pA, n=16; unpaired t-test, p > 0.05, Fig. 2D), suggesting, that atropine did not block ACh-mediated responses at nAChRs. Taken together, along with our DMPP and nicotine findings that these two nAChR agonists elicited only inward currents, we conclude 1) that ACh-induced outward currents resulted from stimulation of muscarinic receptors known to be present on these cells and 2) because the amplitude of ACh-induced inward current was not significantly different between atropine and control conditions, that atropine was not blocking ACh activation of nAChRs at the concentration we utilized. Given the caveat that equipotent concentrations may not have been utilized, the amplitudes of ACh-induced inward currents were not compared to those elicited by nicotine. Comparison of inward currents elicited by subsequent puffs of ACh or nicotine revealed attenuation in inward current induced by second applications to nicotine across a population of cells (second responses to nicotine were 83.4 ± 6.3% of the first response, n=16), although individual cells could show second responses that were nearly or more than 100% of the first (n=7/16; Fig. 2E, G). ACh responses to second applications were not significantly different than initial responses (second responses to ACh were 100.6 ± 1.7 % of the first response, n=30; Fig. 2F, G), indicating that ACh evokes little receptor desensitization or use-dependent actions at the nicotinic postsynaptic receptor (Fig. 2G).
Cholinergic neurons within the LDT have been reported to express mRNA for α7 and β2 subunits; mRNA for the α4 subunit was shown to be present in non-cholinergic neurons (Azam et al, 2003). However, functional evidence regarding the nicotinic subunits comprising nAChRs in the LDT is lacking. Therefore, the subunits of the nAChRs mediating inward currents on postsynaptic cells were explored pharmacologically within the LDT. Since nicotine effects were not consistently repeatable on induction of inward currents as has also been shown in other cells types (Giniatullin et al, 2000), but ACh actions on induction of inward currents were repeatable, we utilized ACh for this series of studies. Antagonists were determined to be effective if the second application of ACh in the presence of antagonist resulted in inward currents reduced by two standard deviations (18%) of those elicited by the first application of agonist alone.
Using these criteria in ACSF containing TTX, bicuculline, APV, strychnine and DNQX, DHβE, a β2 nAChR subunit antagonist, at a concentration shown by others to fully antagonize responses mediated by β2 nAChR subunits (500 nM; Mansvelder et al, 2002), significantly reduced the nicotine-induced inward currents in nNOS+ cells by 37.9 ± 7.5% (ANOVA, p < 0.05; n=9). This suggests that in cholinergic neurons, nAChRs containing the β2 nAChR subunit contributed to the current response but that other subtypes are probably involved. Consistent with this interpretation, MLA, at a concentration specific for antagonism of the α7 receptor, also significantly reduced postsynaptic responses by 37.1 ± 6.1% (ANOVA, p < 0.05; n=8; Fig. 2H2, I). Responses resistant to either MLA or DHβE could be mediated by nAChRs devoid of the β2 or α7 subunit and we found that MEC at a concentration specific for antagonism of non-α7 subunits reduced the postsynaptic inward current by 28.4 ± 6.9% (ANOVA, p < 0.01; n=8). A cocktail of DHβE, MEC and MLA reduced postsynaptic induced currents by 95.0± 1.1 % (ANOVA, p < 0.01; n=15; Fig. 2H3, I). These data are summarized in Figure 2H and show that β2 and α7 subunits are present in nAChRs located on cholinergic neurons, providing pharmacological evidence that receptors containing these subunits are functionally expressed. Additionally, our results provide the first functional data that activation of nAChRs in LDT neurons results in induction of inward currents mediated by the β2 and α7 subunits and that subunits other than the β2 or α7 also comprise nAChRs on these cells.
The calcium permeability of nAChRs depends on the composition of their subunits. Homopentameric receptors containing the α7 subunit exhibit a high relative permeability to calcium (Bertrand et al, 1993; Seguela et al, 1993); whereas, nAChRs comprised of the α4β2 subunits are less permeable to calcium. The presence of α7 subunit in nAChRs on cholinergic neurons led us to ask whether nicotinic activation of these cells results in increases in intracellular calcium ([Ca2+]i). Accordingly, we assayed nicotine-mediated changes in [Ca2+]i by monitoring changes in baseline and activity-dependent [Ca2+]i using a calcium sensitive dye, bis-Fura 2. After the cell was filled passively with bis-fura 2 via the recording electrode, a region of interest (ROI) overlying the soma or dendrite(s) was selected and changes in fluorescence monitored within this area by collection of images of the ROI every 50 ms. We found that nicotine failed to significantly increase baseline calcium levels in somata of LDT cells (n=18). These data did not rule out calcium-induced changes in the dendrites and interestingly, we found that in about half of the cells recorded, all of which were identified as nNOS+, nicotine induced a significant increase in the amplitude of baseline calcium in dendrites (all cells: n=10/18; nNOS+ identified: n=10/10, 1.6% rise in dF/F in identified cholinergic neurons, p < 0.05; Fig. 3B1). This effect was repeatable (nNOS+, n=6/6, Fig. 3B2). These rises were significantly attenuated by application of MLA to the slice prior to nicotine exposure (n=7/7; Fig. 3B3). These data indicate that there are α7 subunit-containing nAChRs in the dendrites of cholinergic neurons and that activation of these receptors stimulates increases in dendritic [Ca2+]i. Several sources of Ca2+ might have contributed to these increases including influx through α7-subunit containing nAChRs, activation of voltage-gated Ca2+ channels and Ca2+-evoked Ca2+ release.
In a second series of experiments, we monitored nicotine-induced changes in spike-evoked calcium levels. Positive current pulses sufficient to trigger 5 action potentials were injected via the pipette into the soma (5 Hz) and changes in fluorescence resulting from these spikes were monitored before and after bath application of nicotine (5 min). In the majority of neurons, when glutamatergic and GABAergic input were blocked by inclusion of bicuculline, strychnine, APV and DNQX in the bath, nicotine significantly enhanced the amplitude of activity-dependent [Ca2+]i changes in cholinergic cells (108.0 ± 0.2%, n=5; Fig. 3C, D). Additionally, nicotine significantly and reversibly enhanced the amplitude of dF/F measured during the decay phase by 138.7 ± 10.4% (66% of the peak during the decay, p < 0.05, n=5; Fig. 3C), an effect which recovered. Taken together, these data indicate that while nicotine does not alter baseline calcium levels at the soma, if firing occurs, resultant somatic calcium levels will be enhanced and remain higher for a longer period of time than in the absence of nicotine, which could have important implications for nicotine-induced, activity-dependent, synaptic plasticity.
Nicotine also induced robust excitatory synaptic activity that was sensitive to blockers of fast, glutamatergic transmission (Fig. 4A). Second applications of nicotine failed to induce an increase in synaptic activity in which first applications had induced a significant enhancement of EPSC events. Since repeat application of nicotine induced inward currents, albeit sometimes these currents were attenuated, this suggests differences in desensitization kinetics between mechanisms involved in nicotine-induced inward current and synaptic activity. In a striking difference in effect of agonists, puffer application of ACh sufficient to activate inward currents did not induce synaptic activity in naïve cells, whereas in the same cells, subsequent application of nicotine in the presence of atropine induced synaptic activity (n=3; Fig. 4C). Several possible explanations for this lack of effect exist including rapid hydrolysis of ACh or differences in activation or desensitization between ACh and nicotine acting at differing combinations of subunits comprising the nAChRs (Albuquerque et al, 1997; Alkondon and Albuquerque, 2006; Alkondon et al, 1997b; Bradaia and Trouslard, 2002; Fenster et al, 1997; Gopalakrishnan et al. 1995; Virginio et al. 2002) mediating inward currents and synaptic activity, although nicotine and ACh have been demonstrated in other cell types to have equivalent actions on induction of PSCs (Mansvelder et al, 2002; Sharma et al, 2008). More rapid catabolization of ACh may explain this difference in cholinergic LDT neurons since, while ACh applied in the presence of 1 μM eserine with or without atropine did not result in differences in inward current (ACh: 80.9 ± 16.7pA, n=16, ACh in atropine, 72.2 ±18.3 n=7), ACh in eserine with atropine 85.6 ± 17.4pA, n=6, unpaired t-test; p>0.05), an increase in synaptic activity was sometimes seen during acetycholinesterase blockade (n=6). Our data suggest that more rapid breakdown of ACh may, in part, account for our observed differences in agonist effect, and raise the possibility that pre-synaptic sites of nAChRs are protected from exogenous ACh. Regardless in order to focus on nicotine, the drug of abuse, and to avoid the use of atropine, only first applications of nicotine were utilized for quantification of spontaneous synaptic activity induced by activation of nAChRs.
In the presence of bicuculline and strychnine to block inhibitory input, in the majority of cells, puffer application of nicotine induced a significant increase in the frequency of spontaneous EPSCs (sEPSCs; K-S test on individual cells, p < 0.05, n=13/14). The mean frequency increase was 438.7 ± 118.8% (n=13, Fig. 4A). Mean amplitude was significantly increased in 8/14 cells (K-S test, p < 0.05, n=8/14, Fig. 4A) by an average of 118.4 ± 3.2%. In all nNOS+ identified cells, application of nicotine induced a significant increase in sEPSC frequency (K-S test, p < 0.05, n=11/11). The average change in mean frequency in these cells was 456.5 ± 138.5%; and mean amplitude was increased in 6/11 cells (K-S test, p < 0.05, n=6/11) by 118.5 ± 4.2%. In one of the two nNOS- cells recovered and identified, nicotine induced a significant increase in the frequency of sEPSCs (156.4%; K-S test, p < 0.05, n=1/2), as well as an increase in sEPSC amplitude (122.5%; K-S test, p < 0.05; n=1/2).
We also examined the actions of nicotine on miniature EPSCs (mEPSCs). In the presence of TTX, nicotine still produced large changes in the distribution of mEPSC frequency in almost all cells examined (K-S test, p < 0.05, n=28/29; Fig. 4B). The average change in mean frequency was 407.2 ± 45.0%. Nicotine significantly increased the amplitude of mEPSCs in fifteen of these cells (K-S test, p < 0.05, n=15/29) by 137.8 ± 10.1%. In 93.8% of the cells identified as nNOS+, nicotine significantly increased the frequency of mEPSCs (K-S test, p < 0.05, n=15/16). The average increase in mean sEPSC frequency was 488.5 ± 70.9%. In 43.8% of the cells, the amplitude distribution was significantly different (K-S test, p < 0.05, n=7/16) with an average increase in mean amplitude of 153.1 ± 19.8%. In all of the nNOS- cells recorded, nicotine induced a significant increase in mEPSC frequency of 268.4 ± 41.6 % (K-S test, p < 0.05, n=10/10) and in the majority of the cells, nicotine significantly increased their amplitude (K-S test, p < 0.05, n=6/10) by 122.0 ± 7.0%. These actions were not significantly different from the actions of nicotine on these parameters in cells recorded without TTX in the bath (unpaired t-test, p > 0.05; Fig. 4F). The increase in nicotine-induced frequency of mEPSCs in nNOS+ cells was significantly greater than the increase induced in nNOS- cells (unpaired t-test, p < 0.05; Fig. 4E), suggesting the possibility that exposure to nicotine would result in a greater glutamatergic excitation of cholinergic neurons by presynaptic excitatory terminals than that induced in non-cholinergic cells. In low calcium solution, nicotine failed to induce mEPSC activity (n=4/4). The success of low calcium solution, coupled with the failure of TTX, in abolishment of the nicotine-induced changes in frequency of EPSPs indicate that nicotine is acting on presynaptic, glutamatergic terminals.
To verify that nicotine had similar excitatory actions on synaptic transmission when delivered in a manner more consistent with that experienced by the brain of a smoker, nicotine was bath applied. Bath-applied nicotine (10 μM) in presence of bicuculline and strychnine induced a significant increase in frequency of sEPSCs in 66% of the cells recorded (K-S test, p < 0.05, n=8/12) with an average change of 334.9.6 ± 85.0%; and in 42% of these cells, amplitude was significantly increased (K-S test, p < 0.05; n=5/12) by 113.3 ± 7.0. In the majority of cells identified as nNOS+ (Fig 4G1), nicotine applied via the bath induced a significant increase in mean sEPSC frequency (K-S test, p < 0.05, n=4/5) of 403.0 ± 253.9%; but failed to induce a significant increase in amplitude in the majority of cells (p > 0.05, n=4/5) with an average amplitude change of 98.5 ± 6.2%. In 50% of the cells identified as nNOS- cells (Fig 4G2), similarly-applied nicotine induced a significant increase in sEPSC frequency (K-S test, p < 0.05, n=2/4) with an average change of 196.3 ± 90.5%, and failed to significantly increase the amplitude in the majority of cells (p > 0.05, n=3/4) with an average change of 85.6 ± 8.0%. These data support the conclusion that synaptic activity resulting from bath-applied nicotine results from activation of presynaptic glutamatergic neurons. In two nNOS+ cells, nicotine also significantly increased the mEPSC frequency (K-S test, p < 0.05), but failed to affect the amplitude distribution, indicating that bath delivery of nicotine also activates presynaptic glutamatergic terminals on cholinergic neurons.
As second applications of nicotine failed to induce EPSC activity, we were not able to compare the effects of nAChRs subunits antagonists on nicotine-induced increases in EPSC frequency in the same cells. Therefore, comparisons were done across two separate cell populations. The increase in mean frequency of mEPSCs in nNOS+ cells was not significantly affected by inclusion of MLA (10nM; 95.5 ± 17.4% of control, n=16, p > 0.05; Fig. 5A3, B) or DHβE (500nM; 135.8 ± 27.6% of control, n=14, p > 0.05; Fig. 5A2, B) in the recording solution. However, MEC, an antagonist of the non-α7 nAChR subunit, blocked the nicotine-induced increase in mEPSC frequency (20.7 ± 4.8% of control, n=10, p < 0.01; Fig. 5A4, B). Taken together, this pharmacological profile suggests that the β2 and α7 subunit are not major components of nAChRs located on glutamatergic terminals presynaptic to cholinergic LDT neurons, but rather that non-α7-containing nAChRs mediate the increase in glutamatergic EPSCs on LDT neurons.
Puffer application of nicotine also induced inhibitory synaptic activity that was bicuculline and strychnine-sensitive. Nicotine-induced actions on inhibitory synaptic activity were quantified in recordings with high chloride pipettes to reverse the chloride gradient and with APV and DNQX to block excitatory synaptic transmission. Under these conditions in the vast majority of cells recorded (87.5%), nicotine induced a significant increase in spontaneous IPSC (sIPSC) frequency (K-S test, p < 0.05, n= 21/24) with an average change in mean frequency of 280.0 ± 28.5% from control; and, only a minority of these cells showed a significant increase in their amplitude (K-S test, n=7/24, p < 0.05) of 131.3 ± 13.5%. In 94% of the nNOS+ cells recovered, nicotine significantly increased the frequency of IPSCs (K-S test, p < 0.05, n=15/16; Fig. 6A) with an average change of 292.9 ± 30.7%. In 75% of the identified nNOS- cells, nicotine also induced a significant increase in frequency of IPSCs (K-S test, p < 0.05, n=3/4) with an average change of 308.0 ± 131.1%, in these cells. The change in amplitude of spontaneous IPSCs increased significantly in 4/16 nNOS+ cells (K-S test, p < 0.05, n=14/16) by an average of 148.0 ± 20.4% and in half of the identified nNOS- cells (K-S test, p < 0.05, n=2/4) with an average change of by 107.1 ± 4.7%.
To determine if these changes could be accounted for by actions at presynaptic terminals, the effect of nicotine on mIPSCs was examined. In the majority of cells recorded with TTX in the ACSF, nicotine significantly increased mIPSCs (K-S test, p < 0.05, n=17/21) by an average of 747.9 ± 212.7%; mIPSC amplitude was also significantly enhanced in 47.6% of the cells (K-S test, p < 0.05, n=10/21) with an average change of 168.5 ± 14.6%. In 71% of the nNOS+ cells identified, and in all of the nNOS- cells, nicotine significantly increased the frequency of mIPSCs (K-S test, p < 0.05; nNOS+ : n=10/14; nNOS-: n=5/5, Fig. 6B) by an average of 1092.2 ± 321.0% and 290.3 ± 82.1%, respectively. The amplitude of nicotine-induced mIPSCs was not significantly different in the majority of nNOS- cells (K-S test, p > 0.05, n=4/5); however, in 50% of the nNOS+ cells recorded (n=7/14), the amplitude of mIPSCs did increase significantly following nicotine administration (K-S test, p < 0.05, n=7/14) by an average of 186.0 ± 16.9%. The difference in frequency of mIPSCs induced by nicotine was not significantly different between nNOS+ cells and nNOS- cells (p > 0.05; Fig. 6C). Although the change in IPSC frequency in the population of nNOS+ cells appeared to be greater when TTX was included in the ACSF from that evoked by nicotine in its absence, the increase in nicotine-induced synaptic activity was not significantly different between the two conditions in cholinergic neurons, probably due to the variability in nicotine-enhanced frequency changes in these cells in the TTX condition (p > 0.05; Fig. 6D). Combined with the increase in amplitude of mIPSCs, these data suggest that some nicotine-stimulated inhibition of GABAergic cells presynaptic to nNOS+ cells occurs and that this inhibition is blocked when action potentials are inhibited, thereby allowing a barrage of inhibition to be directed to cholinergic cells that may summate in time, or in spatial point of occurrence. However, this possibility was not investigated further. In the presence of low calcium ACSF, nicotine failed to induce mIPSCs (n=4, data not shown). Since nicotine-induced inhibitory synaptic currents require extracellular calcium but are not dependent on generation of action potentials, we conclude that nicotine excites GABA and/or glycinergic terminals impinging on cholinergic neurons as well as inhibitory terminals directed to non-cholinergic cells.
The maximum frequency change in IPSPs when nicotine was bath-applied (10 μM) was significantly greater than that induced when nicotine was puffed onto cells. With APV and DNQX in the bath to block excitatory amino acid transmission, the frequency of sIPSCs increased in all cells (K-S test, p < 0.05, n=4/4) by 1399.8 ± 722.8%. One cell was identified as nNOS+ and increased in frequency by 3097.9%; one cell was identified as nNOS- and increased by 2103.8%, suggesting that this phenomenon extended to both cholinergic and non-cholinergic cells. When TTX was included in the ACSF, in the majority of recorded cells, nicotine increased the frequency of mIPSPs (K-S test, p < 0.05, n=8/11) by an average of 1008.4 ± 343.6%. In the majority of nNOS+ cells tested, nicotine increased the frequency of mIPSCs (K-S test, p < 0.05, n=4/5) by 1502.6 ± 572.2% and in 66% of the nNOS- cells tested, miniature inhibitory events increased (K-S test, p < 0.05, n=2/3) by an average of 335.9 ± 91.2%. The data obtained using either bath or puffer applications suggest that nicotine can induce GABA release from terminals located presynaptic to cholinergic and non-cholinergic LDT cells.
The increase in mIPSCs induced by nicotine in nNOS+ cells was significantly attenuated by inhibition of the α7 subunit with MLA (10nM) (75.4 ± 11.9 % reduction from control, n=12, p < 0.05; Fig. 7A1, 7A3, B). DHβE (500nM) also significantly reduced the nicotine-induced increase in mIPSC frequency (80.4 ± 18.7 % reduction from control, n=10, p < 0.05; Fig. 7A2, B). MEC (1μM), an antagonist of non-α7 subunit containing receptors, abolished the nicotine-induced increase in mIPSC frequency (98.0 ± 1.3 % reduction from control, n=11, p < 0.01; Fig. 7A4, B). These data indicate that nAChRs on GABAergic terminals presynaptic to cholinergic cells contain α7, β2 and non-α7 subunits, and our data taken together with the effects of nAChRs antagonists on mEPSCs, suggest that the subunit composition of nAChRs on GABA-containing presynaptic terminals and those on glutamate-containing presynaptic terminals within the LDT is different.
In this study, we show that nicotine has profound direct and indirect actions on both cholinergic and non-cholinergic LDT neurons. In almost all LDT neurons examined, regardless of presence or absence of nNOS, nicotine elicited membrane depolarization with a concurrent decrease in input resistance. Often membrane potential changes were sufficient to induce sodium channel-dependent action potentials. Interestingly, nicotine activation of nAChRs resulted in calcium rises in dendrites of cholinergic LDT neurons, whereas similar rises were not observed in the somata. However, nicotine potentiated intracellular calcium changes at the soma arising from action potentials and increased cellular excitability as demonstrated by higher firing rates produced by constant current injection. Nicotine also induced a TTX-resistant increase in frequency of excitatory glutamatergic, and inhibitory GABAergic synaptic activity in nNOS+ and nNOS- cells. Utilization of nAChR subunit antagonists provides the first functional evidence for the presence of β2, α7 and non-α7 subunits on cholinergic neurons and revealed a differential distribution of subunits comprising nAChRs located on glutamatergic and GABAergic terminals presynaptic to cholinergic LDT neurons. Further, antagonist-specific actions on calcium elevations induced by nicotine in dendrites of nNOS+ neurons indicated that dendritic nAChRs in these cells contain α7 subunits. Taken together, our data suggest that exposure of the LDT to nicotine would excite all cell types in this nucleus, albeit likely to different degrees. Moreover, owing to the differences in nAChR subunit distribution within different cell types, nicotine would be expected to differentially regulate neuronal function within the heterogeneous LDT nucleus.
Although to our knowledge, this report represents the first study of the electrophysiological actions of nicotine on mouse LDT neurons, a selective nicotinic receptor agonist (1,1-dimethyl-4-phenyl-piperazinium iodide; DMPP) has been shown to have effects on presumed cholinergic neurons of the rat pedunculopontine tegmental (PPT) nucleus. The PPT is in close anatomical proximity to the LDT and includes another large collection of mesopontine cholinergic neurons. PPT cells recorded from animals between ages P12-17, and identified as cholinergic based on presence of A-type current, responded to bath-applied DMPP with depolarization (Good et al, 2007) which agrees with our findings in the mouse. However, by age 21 days, in a striking response reversal, DMPP was reported to elicit membrane hyperpolarization (Good et al, 2007). In our study, we never observed a hyperpolarization in nNOS+ or nNOS- neurons in response to nicotine even though we recorded from animals up to 45 days of age and used two methods of agonist application. Further, DMPP did not elicit outward currents in LDT cells recorded from animals between the ages of 21 days and 23 days. Thus, it seems unlikely we overlooked a hyperpolarizing response to nicotine in LDT neurons. With the caveat that different species, route of administration, recording methods and slice temperatures were utilized in the two studies, it is possible that these different responses are authentic and should be further investigated. Such a difference might reflect the disparate functions predicted to be subserved by the LDT and PPT based on their differential innervation of reward circuitry (Omechencko and Sesack, 2005, 2006; Geisler and Zahm, 2005) and the different addiction-related behavioral outcomes of lesioning these two nuclei (Alderson et al, 2008; Laviolette et al, 2000; Forster et al, 2002a; Winn, 2008).
Previous studies have shown a nicotinic receptor-mediated, presynaptic enhancement of fast, ionotropic glutamatergic transmission (Girod et al, 2000; Gray et al, 1996; McGehee al, 1995). Consistent with these reports, we found in the current study that nicotine enhanced the frequency of APV- and DNQX-sensitive EPSCs in cholinergic and non-cholinergic LDT neurons. This appeared to be due to the action of nicotine in promoting Ca2+-influx into glutamate-releasing presynaptic terminals since the increase in mEPSC frequency was insensitive to TTX and was attenuated by lowering extracellular Ca2+ as described in other neurons (Fisher et al, 1998; Lu et al, 1999; Radcliffe et al, 1999). Interestingly, nicotine induced a greater increase in mean frequency of EPSCs in cholinergic neurons than in non-cholinergic neurons. The cell-specific degree of nicotinic activation of EPSCs represents a potentially important functional difference. While immunohistochemical data suggested that ACh-containing LDT neurons may co-localize GABA and glutamate (Charara et al, 1996; Omelchenko and Sesack, 2006), recent in situ evidence contradicts this conclusion (Wang and Morales, 2009). Very few cholinergic LDT neurons were found to contain mRNA transcripts encoding proteins for the vesicular transports of glutamate or for the synthesis of GABA, suggesting that glutamate and GABA are unlikely to be co-released with ACh from LDT neurons (Wang and Morales, 2009). Further, the cholinergic neurons appeared not to be the major neuronal phenotype within the LDT with vGluT2- and GAD-containing neurons being equal to or twice as prevalent, respectively. Accordingly, the pool of non-cholinergic neurons in our study were likely to be in part comprised of GABAergic and glutamatergic cells. Our data, indicating that nicotine elicits a relatively larger excitatory glutamatergic drive onto cholinergic as compared to non-cholinergic cells taken in combination with the in situ data indicating segregation of neurotransmitter phenotype across LDT cells, raises the possibility of relatively enhanced release of ACh over glutamate and GABA at LDT projection targets following identical nicotinic exposure.
GABAergic transmission has also been shown to be enhanced by nicotine in many different cell types (Alkondon et al, 1997a; Alkondon et al, 2000; Fisher et al, 1998; Lena and Changeux, 1997; Lena et al, 1993; Radcliffe et al, 1999). Nicotine induced a TTX-resistant, low calcium solution-sensitive increase in frequency of strychnine- and bicuculline-sensitive IPSCs directed to nNOS+ and nNOS- LDT neurons. The increase in frequency of IPSCs was not significantly different between cholinergic and non-cholinergic cells. These data suggest nAChRs are present on GABA-containing terminals presynaptic to cholinergic and non-cholinergic LDT neurons and indicate that following nicotine exposure, inhibition would be comparably augmented in both cholinergic and non-cholinergic LDT neurons. Our data showing that in the presence of blockers of excitatory transmission, bath application induced significantly more sIPSCs than did puff application, whereas elicitation of EPSCs was similar regardless of method of drug application, suggest that bath-applied nicotine excited nAChRs located at more distal sites than did the more locally applied puff and, accordingly, that there may be a more local presence of nAChRs on glutamate-containing terminals impinging on recorded cells. Accordingly, our findings suggest a difference between distribution of nAChRs on those GABA-containing and glutamate neurons that terminate on LDT neurons.
Nicotinic receptors are pentameric, ligand-gated ion channels composed of different combinations of α (2-10) and/or β (2-4) subunits (Gotti et al, 1997; Jones et al, 1999; McGehee and Role, 1995; Role and Berg, 1996; Salamone and Zhou, 2000). Nicotinic receptors can be loosely divided into two classes based on their subunit composition. The first class is homopentameric, composed of the α7 subunit and highly calcium permeable (Fucile, 2004). The second class is heteropentameric formed by both α and β subunits and shows low calcium permeability (Fucile, 2004). Azam et al, (2003) found expression of β2 and α7 mRNAs in almost all cholinergic neurons of the LDT which suggested that α7-subunit containing nAChRs may mediate nicotinic actions in cholinergic neurons with or without involvement of the β2 subunit. Non-cholinergic neurons expressed mRNA for the α4 nAChR subunit, suggesting the possibility of nAChRs containing α4β2 subunits in these cells (Azam et al, 2003). Our data utilizing nAChR antagonists indicate that inward currents induced by nicotine in cholinergic neurons were mediated by nAChRs containing α7 and β2 as well as non-α7 subunits. While DHβE at the concentrations we utilized can have actions on nAChRs containing α4β2 and α4β4 subunits in other cell populations (Harvey et al, 1996), cholinergic neurons appear to be devoid of mRNA for the α4 subunit (Azam et al, 2003), suggesting the antagonistic effect of DHβE was mediated by β2 subunit-containing nAChRs. Failure to detect changes in baseline calcium in the somata of cholinergic neurons could indicate that nAChRs located on the somas are heteropentameric, consisting of both α and β subunits. However, nicotine did elicit dendritic calcium rises. These data suggest that if there are nAChRs at the soma, there may be differences in their subunit composition from those at the dendrites of cholinergic LDT neurons, although it can not be ruled out that the larger surface to volume ratio at the dendrite may have precluded detection of somatic calcium rises. Regardless, such intracellular subunit-specific targeting has been reported (Champtiaux et al, 2002; Lena and Changeux, 1999; Lena et al, 1999). Further, nicotine-induced increases in dendritic calcium were sensitive to MLA indicating the presence of the α7 subunit within dendritic nAChRs as has been shown for hippocampal neurons (Fayuk and Yakel, 2007), and suggesting, although we did not rule out a contribution of VGCCs to nicotine-induced dendritic calcium changes, that these rises were mediated at least partly by α7-subunit containing receptors. Calcium influx via α7-containing nAChRs is linked to several calcium-dependent processes, including developmental-related switching of polarity of electrophysiological actions following GABAA receptor activation (Liu et al, 2006), transcription and gene regulation (Chang and Berg, 2001), and amplitudes of GABAA mediated currents (Zhang and Berg, 2007). Further, in other cell types, nicotine increases amplitudes of mEPSCs via calcium-dependent, plastic changes mediated by α7-containing nAChRs in the presynaptic terminal. These changes lead to a coordinated release of quanta, resulting in heightened excitatory neurotransmitter concentration at the postsynaptic cell (Sharma et al, 2008). To sum, elevations of intracellular calcium have been linked to a variety of distinct signaling functions as well as synapse formation which makes the spatial and temporal rises in baseline calcium in dendrites of cholinergic neurons following nAChR activation a topic of potential interest.
Ability of MEC, but failure of MLA and DHβE, to reduce the nicotine-induced frequency increase of APV and DNQX-sensitive mEPSPs indicates that nAChRs on glutamate-containing terminals synapsing on cholinergic LDT neurons contain neither α7 nor β2 subunits. In contrast, blockade of bicuculline and strychnine-sensitive IPSCs with MEC, MLA and DHβE suggests the presence of α7, β2 as well as non-α7 subunit-containing receptors on terminals of GABAergic inhibitory neurons synapsing on cholinergic LDT cells. Accordingly, our data suggest differences in subunit compositions of nAChRs located on excitatory and inhibitory presynaptic terminals on cholinergic LDT cells which may confer differing rates of receptor activation and desensitization (Albuquerque et al, 1997; Alkondon and Albuquerque, 2006; Alkondon et al, 1997b; Fenster et al, 1997). Differing rates of desensitization of nAChRs on terminals directed to DA-containing VTA cells has been elegantly shown to underlie a long-lived enhancement (over one hour following nicotine exposure) of excitation of glutamate-containing VTA cells with concurrent depression of GABA-containing local neurons within the VTA (Mansvelder et al, 2002; Mansvelder and McGehee, 2000). Differing rates of nicotinic activation and desensitization have also been shown to mediate differential actions of nicotine on DA-containing neurons of the VTA and those of the substantia nigra pars compacta (Pidoplichko et al, 2004; Wooltorton et al, 2003). Accordingly, the observed differences in subunits comprising nAChRs in GABAergic and glutamatergic terminals may be physiologically significant for LDT functioning by imparting different regulatory actions of ACh and nicotine on these presynaptic sites.
Although drugs of abuse target the entire brain, enhanced dopamine release from the VTA to the NAcc is a key element in rewarding properties of drugs. Burst firing in DA-containing VTA neurons, resulting in DA release in the NAcc, conveys the physiologically relevant stimulus for assigning reward to behavior, making it critical to understand synaptic input to the VTA, which could regulate behaviorally-relevant firing patterns. Within the VTA, cholinergic agonists, including nicotine, increase the activity of DA- and GABA- containing cells, can induce burst firing, and influence the release of DA in the NAcc (Corrigall et al, 1994; Grillner and Svensson, 2000; Gronier et al, 2000; Nisell et al, 1994; Westerink et al, 1998; Westerink et al, 1996). Very recent data indicate that ACh fluctuations within the VTA underlie cocaine-seeking behavior (You et al, 2008), providing further evidence that cholinergic inputs within the VTA are part of important segments of the reward circuitry.
In the most comprehensive anatomical study to date of direct, cholinergic innervation of the VTA, Omelchenko and Sesack (2006) expand upon findings reported by Garzon et al, (1999) and demonstrate cholinergic input to NAcc-projecting DA-containing VTA neurons and conclude that vesicular acetylcholine transporter (VAChT)-containing terminals more frequently synapsed on this set of neurons and much less commonly on DA-containing, mesoprefrontal-projecting neurons, contradicting a previous hypothesis (Garzon et al, 1999). Further, cholinergic axons, were found to contribute assymetrical (presumed excitatory) synapses most commonly on DA-containing, NAcc-projecting VTA cells; whereas, symmetrical (presumed inhibitory) synapses were more common on GABA-containing cells (Omelchenko and Sesack, 2006). The source of endogenous cholinergic input derives in large part from the LDT, based on the morphology and synaptology of cholinergic axons within the VTA (Omelchenko and Sesack, 2005). Anatomical findings suggest that endogenous cholinergic input within the VTA from the LDT would most likely be excitatory to NAcc-projecting DA-containing cells both via direct excitation and via inhibition of local circuit GABA cells. In addition to cholinergic projections, the LDT also sends glutamate and GABAergic input to the VTA (Hallanger and Wainer, 1988; Oakman et al, 1995). The synapse types of cholinergic terminals within the VTA suggested co-localization of other neuroactive substances (Omelchenko and Sesack, 2006), in particular glutamate (Bevan and Bolam, 1995; Clarke et al, 1997) and GABA (Jia et al, 2003). However, a very recent study demonstrated that the vast majority of cholinergic LDT neurons do not contain GAD or VGluT2 transcripts (Wang and Morales, 2009); further, in a preliminary report, the majority of axon terminals in the VTA, some of which certainly derive from the LDT, that contained VAChT did not contain immunolabel for VGluT2 nor GAD (Wang et al, 2008).
The LDT influences DA release within the VTA, which demonstrates functional connectivity between these two nuclei. Electrical stimulation of the LDT results in two phases of increases in DA levels within the VTA; the first phase is due to activation of ionotropic glutamatergic and nicotinic receptors within the VTA; whereas the second and longer-lived phase depends upon metabotropic M5 receptors (Forster and Blaha, 2000; Forster et al, 2002b). Microdialysis studies in the VTA and NAcc in amphetamine-injected rats revealed an increase in glutamate release within these sites following AMPA injections in the LDT indicating a heightened excitability, and subsequent facilitation, of synaptic transmission within the LDT-VTA-NAcc pathway, which suggested for the first time that drug-induced plasticity can occur within the LDT (Nelson et al, 2007). Heightened transmission from the LDT to the VTA may underlie the sensitization of post-stereotypy locomotive behaviors as lesions of the LDT attenuated this addiction-associated phenotype developed in response to repeated amphetamine injections (Nelson et al, 2007). Loss of LDT projections from the LDT to the VTA were speculated to underlie the failure of nicotine to induce normal locomotor actions in LDT lesioned animals (Alderson et al, 2005). Additional evidence for the role excitatory projections from the LDT to the VTA play in addiction is provided by the finding that burst-firing of DA-VTA neurons, which results in large increases in synaptic DA and therefore is believed to be the functionally relevant mode of transmission elicited by exposure to behaviorally relevant stimuli, is entirely dependent upon a physiologically-intact, functioning LDT (Lodge and Grace, 2005). Accordingly, it has been proposed that the LDT serves as a gate that allows VTA-DA neurons to switch to behaviorally relevant, phasic patterns of firing and a role for cholinergic transmission in this gating is indicated (Grace et al, 2007. While GABA-containing afferents from the LDT synapse on DA-containing VTA cells (Charara et al, 1996; Omelchenko and Sesack, 2005, 2006), the role they play in the modulation of activity of these cells is unclear. Taken together, the anatomical and physiological data suggest that the LDT via cholinergic and glutamatergic projections to the VTA provides a major excitation of NAcc-projecting DA-containing neurons and that these projections play a vital role in communicating the saliency properties of natural and drug stimuli serving to initiate, as well as maintain, motivated behavior.
An accumulating body of evidence suggests neurons of the LDT are important substrates for signaling reward and for modulation of goal-driven behavior. As the LDT appears to exert excitatory regulation of NAcc-projecting, DA-containing neurons, direct excitation of cholinergic and non-cholinergic LDT neurons by nicotine would likely result in heightened excitation of these VTA cells. Additionally, input to the LDT from non-VTA areas also included in reward circuitry, such as the pre-frontal cortex (Carr and Sesack, 2000a, b), may be scaled differently following nicotine exposure of the LDT. Nicotine-induced increases in GABAergic IPSCs directed to cholinergic neurons might selectively dampen cholinergic output; however, phasic GABAA-mediated inhibition does not alter excitability induced by tonic depolarization (Nusser and Mody, 2002), suggesting the possibility that the directly-mediated depolarizing effect of nicotine on LDT neurons would not be greatly opposed by inhibitory synaptic events. Further, our data show non-cholinergic neurons, some of which are almost certainly local GABA-containing cells (Ford et al, 1995; Jones, 2004), are subject to nicotine-induced inhibitory GABA input, thus potentially limiting their inhibitory actions on cholinergic neurons and other target cells. Nicotine enhanced intracellular calcium levels induced by action potentials and increased baseline calcium levels in dendrites. Somatic and dendritic calcium rises may be proximal to calcium-dependent effectors suggesting the possibility that nicotine could engender synaptic plasticity as shown to be induced by amphetamine and suggested to be mediated by induction of LTP (Nelson et al, 2007). Taken together, our data provide physiological evidence that nicotine alters excitability in LDT circuitry and suggest that the LDT may be involved in imparting stimulus saliency via nicotine-induced alteration of throughput of this nucleus to circuitry important for signaling reward. The LDT may, therefore, be critically involved in the development of nicotine dependence and maintenance of addiction.
The pontine cholinergic system is also considered to be a mediator of cortical arousal via rostrally-directed projections (Semba and Fibiger, 1989, 1992; Semba et al, 1990) and a positive effector of REM sleep via its cholinergic caudal projections to a pontine site known as the REM-sleep induction zone due to the ability of cholinomimetics to induce a REM-like state when locally injected (Baghdoyan et al, 1993). Studies examining nicotine's actions on sleep and wakefulness are contradictory, perhaps due to differences in species, routes of administration and dose (Gillin et al, 1994; Salin-Pascual et al, 1999; Vazquez et al, 1996; Wetter and Young, 1994). Although the precise action of nicotinic transmission in the LDT on behavioral state remains to be elucidated, given the importance of cholinergic and non-cholinergic LDT neurons in the expression of sleep and wakefulness, our data showing strong electrophysiological actions of this drug on cells in this nucleus suggests that exposure of the LDT to nicotine, either acutely in the form of occasional use of tobacco products, or chronically as experienced by an addicted smoker, could lead to profound changes in behavioral state. Moreover, withdrawal of nicotine from the brain of a recurrent nicotine user, such as during a sleep bout of 7-8 hours, would also be expected to influence LDT functioning which may, in part, contribute to deviations from the normal expression of behavioral state patterns in smokers (Gillin et al, 1994; Salin-Pascual et al, 1999; Vazquez et al, 1996; Wetter and Young, 1994).
The LDT contributes to amphetamine, nicotine and morphine-induced behaviors associated with drug abuse (Forster et al, 2002a; Laviolette et al, 2000; Nelson et al, 2007) and an appreciation is developing that it may be critically involved in expression of addiction to a wide range of drugs of abuse (Maskos, 2008; Mena-Segovia et al, 2008). An important step for incorporation of the LDT into the neural circuitry involved in addiction to nicotine was determination of whether this drug has actions on cells within this nucleus. By demonstrating that nicotine has pre- and postsynaptic actions on virtually every cholinergic and non-cholinergic LDT cell, our data contributes to understanding the mechanisms that may underlie the likely role the LDT, via its projections to reward circuitry plays in the development, and maintenance of, addiction to nicotine. It will now be important to determine whether nicotine-dependent synaptic plasticity occurs in the LDT as has been suggested by the data of Nelson et al (2007) and demonstrated in the VTA (Mansvelder and McGehee, 2000). Actions of nicotine within the LDT incurred in a first time smoker, or those in a chronic smoker, may be long lasting and may result in alteration of cholinergic and non-cholinergic output to the VTA and to other terminal fields. Accordingly, gaining information about the basic action of nicotine on LDT neurons represents a necessary step towards refinement of the current model of the neurobiology of addiction to include involvement of the LDT.
The authors would like to gratefully acknowledge Dr. Iryna Gumenchuk for her assistance with the immunohistochemistry presented in this report and Dr. Morten P. Kristensen for comments on previous versions of this manuscript.
Research described in this article was supported by Philip Morris USA Inc. (KAK) and NIH grant HL64150 (CSL).
CS Leonard and M Ishibashi are supported by the National Institute of Health (NIH) for projects unrelated to the work presented in this paper. KA Kohlmeier is supported by the Benzon Foundation (Denmark) for an unrelated project.
Disclosure/Conflict of Interest: Masaru Ishibashi, Christopher S. Leonard and Kristi A. Kohlmeier have no financial conflicts of interest with any data or original ideas presented in this report.