Upregulation of nAChRs and Changes in Excitability on VTA Dopamine and GABA Neurons Correlates to Changes in Nicotine-Reward-Related Behavior

Abstract Previous reports indicate that nicotine reward is mediated through α4β2*, α6β2*, and α4α6β2* nicotinic acetylcholine receptors (nAChRs; * indicates that additional nAChR subunits may be present). Little is known about α4α6β2* nAChR involvement in reward and reinforcement because of a lack of methods that allow the direct investigation of this particular nAChR subtype. Here, we use male and female mice that contain α4-mCherry and α6-GFP nAChR subunits to show that concentrations of nicotine sufficient to evoke reward-related behavior robustly upregulate α4* and α4α6* nAChRs on midbrain dopamine (DA) and GABA neurons. Furthermore, the extent of α4α6* nAChR upregulation on ventral tegmental area (VTA) DA neurons aligns with the magnitude of nicotine reward-related behavior. We also show that the upregulation of nAChRs is accompanied by a functional change in firing frequency of both DA and GABA neurons in the VTA that is directly linked to nicotine reward-related behavior.


Introduction
The use of tobacco or nicotine-containing products remains the leading cause of preventable death. Each year, nearly 6 million people die worldwide because of tobaccorelated health complications (World Health Organization, 2011). In the United States alone, ;480,000 people die annually from tobacco products resulting in $75 billion in medical costs [National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health, 2014]. Nicotine, the primary addictive component of tobacco products, stimulates dopamine (DA) neurotransmission through the activation of nicotinic acetylcholine receptors (nAChRs). Several nAChR subtypes contribute to the multiple stages of nicotine addiction, but many reports highlight a4b 2p, a6b 2b 3, and a4a6b 2p nAChRs as primary mediators of nicotine reward, because these subtypes exhibit the highest sensitivity to acutely applied nicotine (Kuryatov and Lindstrom, 2011;Liu et al., 2012;Engle et al., 2013). Furthermore, studies have shown that these nAChRs on VTA DA neurons are essential for intravenous self-administration of nicotine (Pons et al., 2008).
One well-studied phenomenon of a4p (and in some cases a6p) nAChRs is that they upregulate following longterm exposure to nicotine (Marks et al., 1983;Nashmi et al., 2007;Jasinska et al., 2014;Henderson et al., 2016). These upregulated nAChRs exhibit an increase in sensitivity to nicotine (Kuryatov et al., 2005;Vallejo et al., 2005). In VTA DA neurons, a4b 2p, a6b 2p, and a4a6b 2p nAChRs are upregulated by nicotine and the amount of upregulation depends on the concentration of nicotine used (Nashmi et al., 2007;Henderson et al., 2017). In ventral tegmental area (VTA) and substantia nigra pars reticulata (SNr) GABA neurons, a4b 2 nAChRs are upregulated more robustly than nAChRs present on VTA DA neurons (Nashmi et al., 2007). The upregulated a4b 2 nAChRs on GABA neurons exhibit increased sensitivity to nicotine reward (Nashmi et al., 2007;Ngolab et al., 2015) and are rapidly desensitized during acute nicotine exposure (Mansvelder et al., 2002), resulting in the disinhibition of VTA DA neurons and an increase in DA neurotransmission within the mesolimbic pathway. While the inhibitory inputs to VTA DA neurons are altered by nicotine, there is also an enhancement in excitatory inputs from medial VTA glutamate neurons (Yan et al., 2018(Yan et al., , 2019. Together, these changes in inhibitory and excitatory transmission contribute to nicotine reward. Despite general agreement that long-term nicotine exposure produces nAChR upregulation, there is little understanding regarding how nAChR upregulation is related to behaviors associated with nicotine reward or reinforcement. We sought to fill this gap in knowledge by assessing the magnitude of nAChR upregulation in mice that completed conditioned place preference (CPP) assays with nicotine. To accomplish this we used a previously characterized mouse line that contain a4-mCherry and a6-GFP nAChR subunits (Henderson et al., 2017). In using the same mice for both CPP and nAChR upregulation assays, we were able to establish direct links between nicotine reward-related behavior and nAChR upregulation.

Reagents
(-)-Nicotine dihydrogen ditartrate (product number AC415660100) was obtained from ACROS Organics. All doses and concentrations of nicotine listed here are given as free base.

Mice
All experiments were conducted in accordance with the guidelines for care and use of animals provided by the National Institutes of Health, and protocols were approved by the Institutional Animal Care and Use Committee at Marshall University. Mice were kept on a standard 12/12 h light/dark cycle at 22°C and given food and water ad libitum. At day 21, tail biopsies were taken for genotyping analysis by PCR. Genotyping was outsourced to a commercial lab to detect the presence of fluorescence proteins (Transnetyx). For microscopy assays, only mice that were transgenic for a6-GFP and homozygous for a4-mCherry were used (see CPP assay methods below), with the exception of a6-GFP and a4-mCherry mice used for NFRET controls. For electrophysiology assays we used a6-GFP mice. All experiments used adult (three to five months old) mice. Both male and female mice were used and numbers of each are detailed below in the methods for specific experiments.

CPP assays
CPP assays were completed with a three-chamber spatial place-preference chamber (Harvard Apparatus, PanLab) (Fig. 2). Time in chambers was recorded by video motion tracking software (SMART 3.0). An unbiased protocol was used where nicotine was given in the white/gray chamber on "drug" days, and saline was given in the white/black chamber on saline days. For saline-control cohorts, saline injections were given on both sides. On day 1 of testing, mice were placed in the central chamber and allowed free access for 20 min. Mice that spent .65% of the test in one chamber were judged to show initial biases and were excluded from the experiment. On days 2-5, mice were trained using twice-daily sessions. In morning sessions, mice were given an intraperitoneal injection of drug (saline or 0.5 mg/kg nicotine) immediately before confinement in the drug-paired chamber for 20 min. In the afternoon session, mice were given an intraperitoneal injection of saline immediately before confinement in the saline-paired chamber for 20 min. On day 6, a post-test was completed where the mice were again placed in the central chamber and allowed free access for 20 min. A total of 25 mice were used in CPP assays, and two were excluded because of initial bias. Nearly equal numbers of male and female mice, three to six months old, were used in CPP assays. No sex differences were observed at this dose of nicotine. Data are expressed as a change in baseline preference between the pre-test and post-test. Of these mice, 12 were later studied in confocal microscopy assays (see below).
Between the A.M. and P.M. CPP sessions, each mouse experienced a break of ;4 h. Given the pharmacokinetics of nicotine, injections up to 1 mg/kg are cleared in 1 h and most likely within 30 min (Matta et al., 2007;Li et al., 2013). Thus, nicotine is not present in the brain or plasma during the PM CPP saline conditioning sessions. For each cohort, only mice of the same sex were tested. All drugs used in CPP assays were dissolved in saline at pH 7.4 and injected intraperitoneally.

Confocal imaging of mouse brain slices
Some mice studied in CPP assays were also studied in microscopy assays. Drug delivery for these cohorts is described in the above CPP section. Following the completion of CPP assays, mice were euthanized with CO 2 and subjected to a quick cardiac perfusion with 10-ml icecold saline. We found that this significantly reduced the autofluorescence in the mCherry emission range. The brain was then quickly removed and frozen using acetone and dry ice and then stored at À80°C overnight. At least 24 h later, brains were sectioned at 20 mm using a cryostat onto glass slides (1.0 mm). Coronal brain slices were mounted with Vectashield (Vector Labs, H-1000) and coverslipped (1.5-mm thickness, glass). We targeted bregma À3.1 mm because this region gave the most consistent sections that contained a large portion of the VTA, SNr, substantia nigra pars compacta (SNc), and dentate gyrus in a single slice. In each session, brain slices from mice that received different drug treatments were selected according to bregma (À3.1 mm) so that all imaging sessions were matched between animals and across cohorts.
A Leica SP5 TCSII confocal microscope was used for imaging of the a4-mCherrya6-GFP mouse brain slices. a6-GFP and a4-mCherry were excited at 488 and 561 nm, respectively; 40Â images with a 5Â digital zoom were collected for the quantitative measurements of a4-mCherry and a6-GFP neuron raw integrated density (RID) of the midbrain, while 10Â images with a 5Â digital zoom were collected for the quantitative measurements of a4-mCherry expression of the dentate gyrus. NFRET was calculated using the PixFRET ImageJ plug-in. Control data for NFRET donor and acceptor bleedthrough were obtained by imaging neurons from homozygous a4-mCherry mice (wild-type a6) and a6-GFP (wild-type a4) mice separately. In total, 19 male and 13 female mice were used in confocal assays following CPP (excluding those used for NFRET controls). Nearly equal numbers of male and female mice, three to five months old, were used and no sex differences were observed.
All calculations and analysis of GFP and mCherry fluorescence were made using ImageJ. We used RID as a metric for upregulation to match similar work on nAChR upregulation Cooper et al., 2020). For each mouse, two to three different brain slices were imaged by two to three experimenters blind to drug treatment until all data analysis was completed. For each brain slice, 41-71 VTA DA neurons, 36-50 VTA GABA neurons, or 36-68 SNr GABA neurons were imaged. Data were averaged to provide RID values for each mouse. To minimize biasing, we did not select subpopulations in each brain region and analyzed every GFP and/or mCherry neuron present in each brain area.

Calculation of upregulation score
Following microscopy assays of saline-treated and nicotine-treated mice from CPP assays, we calculated the mean RID for a4p (a4-mCherry), a6p (a6-GFP), and a4a6p (identified by NFRET) nAChRs in saline-treated mice (n = 6 mice). We subtracted these mean RID values from the RID of individual mice used in the nicotine CPP cohort to obtain the upregulation score. For example, the upregulation scores for a hypothetical mouse X are calculated as following: These upregulation scores are plotted as y-axis values with the mouse's respective CPP scores as the x-axis values (see Figs. 3,4). For linear regression analysis, we used GraphPad Prism software with no constraints on the data. This analysis in GraphPad provided R 2 , F, and p values.

Immunofluorescence
For immunofluorescence assays ( Fig. 1 only), we used three-month-old a6-GFP mice. Mice were anesthetized in a manner identical to the method used for microscopy. However, before brain extraction, mice were perfused with 20-ml PBS and 40 ml 4% paraformaldehyde. Brains were extracted and placed in 40% sucrose for 24 h. Brains were sliced at a thickness of 20 mm and mounted on microscope slides. Brain sections on slides were rinsed twice for 10 min with PBS and then permeabilized with 0.5% Triton X-100 in PBS for 1 h. Brain sections were then blocked with 4% goat serum in PBS for 45 min. The primary antibody was diluted in 4% goat serum in PBS and incubated overnight at 24°C [rabbit anti-GFP (A11122, Invitrogen) and sheep anti-TH (AB1542, Millipore) both at concentrations of 1:500]. Brain sections were then washed three times for 15 min with PBS, and the secondary antibody was diluted in 4% goat serum in PBS and incubated for 1 h at 24°C [goat anti-rabbit Alexa Fluor 488 (A11008, Invitrogen) and donkey anti-sheep Alexa Fluor 647 (ab150179, Abcam) both at concentrations of 1:1000]. Brain sections were washed three times for 15 min with PBS and mounted with Vectashield. Finally, brain slices were imaged using the same confocal microscope used for direct imaging of a4-mCherrya6-GFP mouse brain slices as described above.

Whole-cell patch-clamp electrophysiology
Recordings were performed using brain slices prepared from three-month-old male a6-GFP mice. We used a6-GFP mice because a6 nAChR subunits are selectively expressed in DA neurons in the lateral VTA ( L VTA) and SNc and provide a method to identify putative DA (pDA) neurons (see Fig. 1). Previous investigations have also concluded that a6-GFP expression in the L VTA and SNc completely overlaps with the expression of tyrosine hydroxylase (TH; Mackey et al., 2012). To match the investigation of VTA DA neurons in microscopy assays, we targeted bregma À3.1 for electrophysiology assays. Mice were anesthetized with CO 2 and were subjected to cardiac perfusion. The artificial cerebral spinal fluid solutions we used were made in accordance to the protective recovery method (www.brainslicemethods.com). We used standard artificial CSF (ACSF; containing the following: 125 mM NaCl, 2.5 mM KCl, 1.2 mM NaH 2 PO 4 , 1.2 mM MgCl 2 , 2.4 mM CaCl 2 , 26 mM NaHCO 3 , and 11 mM glucose, adjusted to 300-310 mOsm and pH 7.4) for transcardial perfusion and brain slicing (250-mm coronal sections). Following cutting, brain slices containing VTA DA neurons were allowed to recover at 32-34°C in a NMDG ACSF solution (containing the following: 93 mM NMDG, 2.5 mM KCl, 1.2 mM NaH 2 PO 4 , 30 mM NaHCO 3 , 20 mM HEPES, 25 mM glucose, 5 mM Na-ascorbate, 2 mM thiourea, 3 mM Na-pyruvate, 10 mM MgSO 4 , and 0.5 mM CaCl 2 , adjusted to 300-310 mOsm and pH 7.4) for 15 min. Next, brain slices remained in standard ACSF at room temperature for !60 min. Slices were then transferred to the slice chamber and continually perfused with carbogen-saturated standard ACSF (1.5-2.0 ml/min) at 32°C.
Neurons were visualized with an upright microscope (Axio Examiner A1, Zeiss) equipped with an Axiocam 702 mono using DIC near-infrared illumination. Blue illumination was used to visualize a6-GFP presence in pDA neurons. Cell-attached and whole-cell patch-clamp techniques were used to record electrophysiological signals with an Integrated Patch-Clamp Amplifier (Sutter). Patch electrodes had resistances of 4-10 MV when filled with intrapipette solution: 135 mM K gluconate, 5 mM KCl, 5 mM EGTA, 0.5 mM CaCl 2 , 10 mM HEPES, 2 mM Mg-ATP, and 0.1 mM GTP (adjusted to  Research Article: New Research 280-300 mOsm and pH of 7.4). Recordings were sampled at 10-40 kHz. The junction potential between patch pipette and bath solutions was nulled just before gigaseal formation. Series resistance was monitored without compensation throughout experiments using SutterPatch software. The recording sessions for neurons were terminated if the series resistance changed by .20%. Nicotine (hydrogen tartrate salt, dissolved in standard ACSF at pH 7.4) applications were applied using a Picospritzer III (Parker) at a pressure of 5 psi for 20 s at concentration of 300 and 500 mM.

Statistical analysis
All results are presented as mean 6 SEM, and all statistical analyses were performed using GraphPad Prism. For CPP assays, upregulation, and comparison of firing frequencies, we used an unpaired, two-tailed t test to compare saline-treated and nicotine-treated groups. For correlations between CPP scores and nAChR upregulation, we used GraphPad Prism software to perform linear regression analysis. All statistical test results are presented in Table 1.

Results
Nicotine reward-related behavior in a4-mCherrya6-GFP mice Given that our goal was to investigate a link between nAChR upregulation and nicotine reward-related behavior, we required a method to reliably identify DA and GABA neurons in the VTA. To distinguish between DA and GABA neurons in the VTA, we used the presence of a6-GFP nAChRs in a manner that is similar to previous reports that have used fluorescently tagged nAChR subunits (Henderson et al., 2014(Henderson et al., , 2017Cooper et al., 2020). Previous investigations have shown that a6-GFP or b 3-GFP nAChRs that are selectively expressed in VTA and SNc DA neurons exhibit a high co-localization with TH presence (.95%; Mackey et al., 2012;Srinivasan et al., 2016). To confirm this using our own methods, we used immunofluorescence on coronal brain slices (bregma, À3.1) from a4-mCherrya6-GFP mice to examine the co-localization of a6-GFP in this mouse line with TH ( Fig.  1). Here, we observed that 100% of the a6-GFP-labeled neurons in the L VTA and SNc co-localized with TH ( Fig. 1; Table 2). Conversely, 83.9% and 73.9% of the TH-positive neurons contained a6-GFP in the L VTA and SNc, respectively ( Fig. 1; Table 2). We acknowledge that a6 nAChR subunits are expressed on GABAergic boutons in the VTA (Yang et al., 2011). Accordingly, we have assigned neurons positive for a6-GFP expression as pDA neurons.
The a4-mCherrya6-GFP mice enable the examination of a4b 2p, a6b 2p, and a4a6b 2p nAChRs in midbrain DA and GABA neurons ( Fig. 2A,B; Henderson et al., 2017). We tested mice using a CPP assay to measure reward-related behavior (Fig. 2C). In rodents, acute nicotine produces rewarding effects depending on route of administration and concentration. Previous investigations have reported that mice exhibit reward-related behavior to 0.5 mg/kg nicotine in a CPP assay (Tapper et al., 2004;Sanjakdar et al., 2015;Henderson et al., 2016Henderson et al., , 2017. In agreement with these previous reports, male and female a4-mCherrya6-GFP mice exhibited significant reward-related behavior to intraperitoneal injections of 0.5 mg/kg nicotine when compared with saline (p = 0.0011 and p = 0.006, respectively; Fig. 2C). No sex differences were noted with this single dose of nicotine (Table 1), but we acknowledge this may not be the case if the dose range were to be expanded.
Our CPP experiments used intraperitoneal injections of 0.5 mg/kg nicotine. Based on nicotine pharmacokinetics in mice (Matta et al., 2007), these daily injections produced a "peak" brain concentration of ;250 nM nicotine that would be eliminated in ;30 min. This would be applicable to beginning smokers in the early stages of dependence (Tapper et al., 2004). It is important to note that nAChRs on VTA DA neurons would be functionally activated at this concentration of nicotine as previous investigations show that 300 nM nicotine is sufficient to evoke inward currents in VTA DA neurons (Liu et al., 2012;Engle et al., 2013). These previous reports also stress that this low concentration of nicotine only activates nAChRs that contain both a4 and a6 nAChR subunits. To confirm this, we conducted our own whole-cell patch-clamp electrophysiology assays to determine whether 300 and 500 nM nicotine stimulates inward currents in VTA DA neurons. We prepared brain slices from three-month-old male a6-GFP mice (target bregma, À3.1). VTA pDA neurons were identified by presence of a6-GFP (Fig. 2E1), firing frequencies ,10 Hz, and presence of I h (Margolis et al., 2006). pDA neurons were voltage clamped at À65 mV, and 300 or 500 nM nicotine was puffed onto target neurons for 10 s (Fig. 2E2,E3). We noted inward currents in 80% of pDA neurons to both concentrations of nicotine, similar to previous reports (Liu et al., 2012;Engle et al., 2013). This suggests that the concentrations of nicotine used in our CPP assays not only produced reward-related behavior but also were sufficient to activate nAChRs on VTA pDA neurons.
Upregulation of a4a6p, a4p, and a6p nAChRs in VTA DA neurons To determine whether nAChR upregulation correlates to nicotine reward-related behavior, we used confocal microscopy assays similar to previous investigations into nAChR upregulation (Henderson et al., 2014(Henderson et al., , 2017. Coronal mouse brain slices containing the VTA, SNc, and SNr (bregma, À3.1) were prepared from mice subjected to the CPP assay ( Fig. 2A,C).
As discussed above, we used the presence of a6-GFP to distinguish between DA and GABA neurons in the VTA for our assays using inherent fluorescence (no immunostaining; Fig. 2A,B). Using FRET methods that have been described previously (Henderson et al., 2017;Avelar et al., 2019), we identified regions of neurons that contain a4a6b 2p nAChRs. Upregulation of nAChRs was assessed by quantifying the change in RID of a6-GFP or a4-mCherry fluorescence. We observed a significant increase in the RID of a4a6p nAChRs in VTA DA neurons of both male and female mice used to CPP assays (p = 0.019 and p = 0.005 for males and females, respectively; Fig. 2D1). We also noted a significant increase in a4p nAChR RID in VTA pDA neurons (p = 0.05 and p = 0.008 for males and females, respectively; Fig.  2D2). This finding is similar to our previous study, which showed a non-significant increase in a4p nAChR RID using alternating daily injections of 0.5 mg/kg nicotine and saline (Henderson et al., 2017). We observed no change in a6-GFP RID in VTA pDA neurons following nicotine treatment (Fig. 2D3). Similar to our CPP assays, we detected no differences between males and females ( Table 1).
Upregulation of a4a6p and a4p nAChRs in VTA pDA neurons correlates to nicotine reward-related behavior Following the observations described above, we analyzed nAChR upregulation for individual mice as a function of their individual CPP score. For each mouse we calculated upregulation scores for a4p, a6p, and a4a6p nAChRs (defined in Materials and Methods), which we then plotted against the individual mouse's CPP score (Fig. 3A1-A4,B1-B3). We found that a4a6p nAChR upregulation in VTA pDA neurons significantly correlated to changes in nicotine reward-related behavior (CPP score) with a R 2 of 0.67 for males (F (1,5) = 10.1, p = 0.02; Fig. 3A2) and a R 2 of 0.82 for females (F (1,6) = 28.2, p = 0.002; Fig. 3B1). We also observed a  Research Article: New Research significant correlation between a4p nAChR upregulation in VTA pDA neurons and nicotine reward-related behavior. Here, we noted a R 2 of 0.83 (F (1,5) = 24.4, p = 0.004; Fig. 3A3) in males and a R 2 of 0.81 (F (1,6) = 25.6, p = 0.002; Fig. 3B2) in females. While we observed a significant correlation between a6p nAChR upregulation and nicotine reward-related behavior in male mice (R 2 of 0.74, F (1,5) = 14.3, p = 0.01), we note that four of the 6 data points are at the level of zero upregulation. We observed no correlation between a6p nAChR upregulation and reward-related behavior in female mice (Fig. 3B3). Using the same methodology, we examined upregulation of the nAChR subtypes in VTA DA neurons from the saline-treated group (Fig. 3C1-C4). Here, we found no correlation between changes in nAChR RID and reward-related  . Upregulation of a4a6p and a4p nAChRs in VTA pDA neurons correlates with nicotine reward-related behavior. A 1 , C 1 , Representative merged images of L VTA DA neurons in a a4-mCherrya6-GFP brain slice. Scale bar, 10 mm. In nicotine-treated mice, changes in nAChR RID was correlated to CPP score for a4a6p nAChRs, a4p nAChRs, and a6p nAChRs for male (A 2 , A 3 , A 4 ) and female (B 1 , B 2 , B 3 ) mice. Linear fits (red line) with 95% confidence intervals (dotted red lines). In saline-treated mice, changes in nAChR RID was correlated to CPP score for a4a6p nAChRs (C 2 ), a4p nAChRs (C 3 ) and a6p nAChRs (C 4 ). Linear fits (red line) with 95% confidence intervals (dotted red lines) were applied using Graphpad Prism software. Nicotine correlations used 7 male and 8 female a4-mCherrya6-GFP mice and the saline correlations used 5 a4-mCherrya6-GFP mice (3 males and 2 females). For each mouse, 36-71 neurons were imaged. behavior. These correlative data may imply that the upregulation of a4p and a4a6p nAChRs may underlie nicotine reward-related behavior, at least in the context of CPP assays.

Upregulation of a4b2p nAChRs on GABA neurons of the VTA and SNr
The upregulation of a4b 2p nAChRs on GABA neurons may play an important role in nicotine's rewarding effect in the mesolimbic circuit (Mansvelder et al., 2002;Nashmi et al., 2007). Nashmi et al. (2007) showed that a4p nAChRs exhibit a greater degree of upregulation on GABA neurons (VTA and SNr) than they do on DA neurons. In VTA GABA neurons, we observed a significant, 3.4-fold increase in a4-mCherry RID (p = 0.01; Fig. 4A1, A2). In male and female mice, we observed a significant correlation between a4p nAChR upregulation in VTA GABA neurons and nicotine reward-related behavior (males, R 2 = 0.55, F (1,6) = 7.39, and p = 0.04; and females, R 2 = 0.56, F (1,6) = 6.5, p = 0.05; Fig. 4A3,A4). Similar to the case of a6p nAChR upregulation in male mice, we note that the slope of this correlation is relatively flat.
SNr GABA neurons in mice from the nicotine CPP group exhibited a 2.4-fold increase in a4-mCherry RID that was significant compared with saline-treated mice (p = 0.02; Fig. 4B1,B2). While this increase in a4p nAChRs is robust, we observed no correlation between a4p nAChR upregulation in SNr GABA neurons and nicotine reward-related behavior in either male or female mice (Fig. 4B3). These data show that concentrations of nicotine sufficient to evoke reward-related behavior upregulate a4b 2 nAChRs on GABA neurons in the VTA and SNr. Given the absence of a strong correlation between upregulation and nicotine reward-related behavior, these data also suggest that VTA DA neurons may be more critical for the initiation of nicotine reward. However, this finding needs to be followed-up with functional experiments to examine changes in neurophysiology (see below).
We also investigated how nicotine reward-related behavior is linked to nAChR upregulation in the dentate gyrus (ventral hippocampus; bregma, À3.1; Fig. 4C1,C2). We observed no significant upregulation in the dentate gyrus of the mice that were part of the nicotine CPP cohort. In fact, we observed a significant decrease in a4p nAChR density (p = 0.047; Fig. 4D1). Previous investigations did detect upregulation in the dentate gyrus (Nashmi et al., 2007); but this was with a persistent, high dose of nicotine (2 mg/kg/h via osmotic minipumps for 10 d). Thus, while upregulation is noted with a4p nAChRs in the dentate gyrus with high doses of nicotine, we did not notice upregulation with doses used (0.5 mg/kg injections) to produce reward-related behavior.
Changes in VTA pDA and putative GABA (pGABA) neuron firing frequency correlates to reward-related behavior While upregulation may indicate changes in nAChR number on neurons, we acknowledge that our confocal methods do not directly indicate whether the upregulation is indicative of increases in intracellular or plasma membrane bound nAChRs. Thus, we used brain-slice electrophysiology to determine whether changes in VTA pDA and pGABA neuron function are linked to reward-related behavior. Nicotine has been observed to decrease the baseline firing frequency of VTA DA neurons and increase the firing frequency of VTA and SNr GABA neurons (Mansvelder et al., 2002;Nashmi et al., 2007). We used a separate cohort of mice assigned to saline or nicotine in CPP assays (Fig. 5C). Similar to results described above, we noted a significant difference between saline-treated and nicotine-treated mice in our CPP assay (p = 0.02). Similar to microscopy assays, at the end of CPP, mouse brains were extracted but here were used for brain-slice electrophysiology assays. Similar to microscopy assays, we used a6-GFP fluorescence to identify pDA neurons in the L VTA (Fig. 5A,B1,B2). Using cell-attached configurations we examined baseline firing frequency of VTA pDA neurons (Fig. 5D). Here, we observed that nicotine-treated mice exhibited a significant decrease in baseline firing of VTA pDA neurons (p = 0.009; Fig. 5E) that matches previous investigations (Nashmi et al., 2007;Xiao et al., 2009). Next, we correlated the VTA pDA neuron firing frequency of individual nicotine-treated mice (mean of !3 neurons per mouse) to its respective CPP score (Fig. 5F). We observed a significant inverse-correlation between firing frequency of VTA pDA neurons and nicotine reward-related behavior (R 2 = 0.65, F (1,6) = 11.2, p = 0.016).
Using these same mice, we examined the firing frequency of VTA pGABA neurons (Fig. 6A,B1,B2). Similar to previous reports, we noted a significant increase in the baseline firing frequency of VTA pGABA neurons in mice assigned to the nicotine CPP cohort (Fig. 6D). Despite observing no significant correlation between nicotine reward-related behavior and a4b 2 nAChR upregulation in VTA GABA neurons (or SNr GABA neurons), we did observe a significant correlation between nicotine rewardrelated behavior and VTA pGABA neuron firing frequency (R 2 = 0.77, F (1,16) = 20.3, p = 0.004; Fig. 6E).

Discussion
It is important to note that nAChRs on VTA DA neurons would be functionally activated as our data (300 and 500 nM nicotine; Fig. 2) and previous investigations (300 nM; Liu et al., 2012;Engle et al., 2013) show that nicotine concentrations that match those present in CPP assays are sufficient to evoke inward currents in VTA DA neurons. These previous reports also stress that activation with this low concentration of nicotine only activates nAChRs that contain a4 and a6 subunits. In line with these previous studies, we noted a significant link between nicotine reward-related behavior and a4a6p nAChRs, a4(non-a6)p nAChRs, but not a6(non-a4)p nAChRs, in the L VTA. We observed no relationship between a4p nAChR upregulation in the SNr to nicotine reward-related behavior. This finding matches a previous investigation by Pons et al. (2008) that showed nicotine self-administration in mice depends on a4p, a6p, and b 2p nAChRs in the VTA but not in the substantia nigra.
One limitation of the current study is that we only observed a single time point (the final CPP test day). The . Upregulation of a4p nAChRs on SNr and VTA putative GABAergic neurons may not correlate with nicotine reward-related behavior. A 1 , B 1 , and C 1 , Representative images of neurons in the L VTA (A 1 ), SNr (B 1 ), or dentate gyrus (C) in a a4-mCherrya6-GFP brain slice. Scale bar, 10 mm (A 1 and B 1 ), 250 mm (C 1 ), and 25 mm (C 2 , C 3 ). A 2 , B 2 , RID of saline-and nicotine-treated mice from CPP assays for SNr GABA neurons or VTA GABA neurons. In A 2 and B 2 the n of individual male and female mice are indicated by individual dots. In nicotine-treated mice, changes in nAChR RID was correlated to CPP score for a4p nAChRs in SNr GABA neurons or VTA GABA neurons. In A 3 , A 4 , B 3 , and B 4 the n of individual male and female mice are indicated by individual dots. D 1 , RID/area of saline-and nicotine-treated mice from CPP assays for dentate gyrus (n = 6, 3 male and 3 female). D 2 , Changes in nAChR RID/area was correlated to CPP score for a4p nAChRs in the dentate gyrus. Linear fits (red line) with 95% confidence intervals (dotted red lines) were applied using Graphpad Prism software. Data is Mean 6 SEM; pp = 0.05, ppp = 0.01; unpaired, two-tailed t test. problem this poses is that our analysis of upregulation and neuronal firing frequency captures the target nAChRs in a state that is free of nicotine for ;24 h. As a result, even in this case where we observed nAChR upregulation and changes in firing frequency in our assays, we ca not adequately assess how nicotine-induced activation of the nAChRs connects to our observed reward-related behavior. Furthermore, our behavioral assay (CPP) is a non-contingent assay and while it provides information regarding nicotine-related reward, it does not provide insight into reinforcement or the impact of long-term nicotine exposure. One manner to improve on this would be to use a contingent drug delivery assay (intravenous or vapor self-administration). With these assays we would be able to extract the brains immediately after operant responding so that we can capture the behavioral endpoint and the brain samples that are still saturated with nicotine. Thus, we could examine changes to nAChRs while nicotine is present and use a model that more closely matches smoking-related behaviors. These assays (vapor self-administration) are currently being pursued but are only at their early stages.
Previous reports show that nicotine robustly upregulates high-sensitivity a4b 2 nAChRs on SNr and VTA GABAergic neurons (Nashmi et al., 2007;Xiao et al., 2009), and we have reproduced that here (Fig. 4). Interestingly, our data show a link between a4p nAChR upregulation on VTA pGABA neurons and nicotine reward-related behavior but no correlation between upregulation of a4p nAChRs on SNr pGABA neurons and reward-related behavior. For a4b 2 nAChR upregulation on VTA GABA neurons, the R 2 values of 0.55 and 0.56 (males and females, respectively) may indicate their upregulation is less sensitive when compared with what we observed on pDA neurons. Here, we must also take into consideration that pDA neurons have a multitude of nAChR subtypes that assemble in multiple possible stoichiometries [a4b 2 (a4 (3) b 2 (2) and a4 (2) b 2 (3) ), a4a6b 2 (a4 (1) a6 (1) b 2 (3) , a4 (2) a6 (1) b 2 (2) , a4 (1) a6 (2) b 2 (2) , a4 (1) a6 (1) b 2 (2) b 3 (1) , and potentially others), and a6b 2p (mostly a6 (2) b 2 (2) b 3 (1) )]. Thus, we must acknowledge that changes in RID of nAChR subunits may also be the result of a net change in stoichiometry. This is another weakness of our methods as we cannot distinguish between nAChRs that reside on the plasma membrane from those that are intracellular. Additionally, we can detect changes in stoichiometry but not define specific stoichiometries. The former problem (plasma membrane changes) is partially reconciled by our electrophysiology assays (discussed below) but the latter can now be addressed using newly developed single-molecule microscopy assays (Fu et al., 2019).
Our electrophysiological methods detected a correlation between VTA pGABA neurons and reward-related behavior that is greater than the correlation between VTA pDA neuron firing and reward-related behavior. One conclusion for this is that our electrophysiological methods are more sensitive than that of our microscopy assays. While this is likely true, the caveat is that brainslice electrophysiology is lower throughput as we can examine 4-10 neurons per mouse, while there are hundreds of neurons in each brain region that could be measured using microscopy. Consequently, the electrophysiological methods may be a more accurate determination of changes in neuronal populations in relation to reward-related behavior.
Following recent investigations into VTA DA and GABA neurons (Grieder et al., 2019), a second conclusion is that there is a distinct difference in the role that DA and GABA systems play within the VTA in nicotine reward. The work by Grieder et al. (2019) revealed that VTA DA neurons mediate aversion while VTA GABA neurons mediate reward to nicotine. This finding also agrees with another previous report that revealed the selective upregulation of a4b 2p nAChRs on VTA GABA neurons was sufficient to produce CPP in mice and increased sensitivity to nicotine reward (Ngolab et al., 2015). Our observations of an inverse-correlation between VTA pDA neuron firing and nicotine reward but a positive correlation between VTA pGABA neuron firing and nicotine reward may support these previous conclusions. One way to address distinction between aversive and rewarding stimuli within the VTA with our current methodology would be to repeat these microscopy and electrophysiology assays but with a conditioned place aversion paradigm using a higher, aversive dose of nicotine. Figure 6. Increase in VTA pGABA neuron firing frequency correlates with reward-related behavior. A, Schematic of target pGABA neurons within the L VTA (target bregma, À3.1). B 1 , B 2 , Representative images of VTA pGABA neurons in DIC (B 1 ) and GFP (B 2 ) imaging modes. Scale bar: 20 mm. C, Representative cell-attached recordings of VTA pGABA neuron baseline firing frequency from mice assigned to saline or 0.5 mg/kg nicotine CPP cohorts. D, Mean VTA pGABA neuron firing frequency in mice treated with saline or nicotine in CPP assays. Data are mean 6 SEM E, Reward-related behavior (CPP Score) were correlated to baseline firing frequency of VTA pGABA neurons; ppp , 0.01 with; unpaired t test. Exact p values are given in text. While this study focuses on DA and GABA neurons of the midbrain, there are other cell types and brain regions that are critical to nicotine reward and reinforcement. Accordingly, our understanding of how changes in nAChR upregulation and function directly link to reward (or reinforcement) is still in the beginning stages and have yet to examine the recently defined local nicotine-related glutamatergic inputs (Yan et al., 2018(Yan et al., , 2019 or the aversive stimuli that originate from the habenula and interpeduncular nuclei (Fowler et al., 2011;Fowler and Kenny, 2012). The methods of this study should be applied to several additional behavioral aspects of chronic nicotine, namely nicotine self-administration. At present, our data show that the upregulation of a4p and a4a6p nAChRs in VTA neurons is tied to nicotine reward-related behavior.