Mechanisms of Metabonomic for a Gateway Drug: Nicotine Priming Enhances Behavioral Response to Cocaine with Modification in Energy Metabolism and Neurotransmitter Level
Nicotine Enhances Behavioral Response to Cocaine
National Chengdu Center for Safety Evaluation of Drugs, State Key Lab of Biotherapy, Sichuan University, Chengdu, Sichuan, China
Analytical and Testing Center, Sichuan University, Chengdu, Sichuan, China
Chiba University Center for Forensic Mental Health, Japan
* E-mail: xbcenalan@vip.sina.comAbstract
Nicotine, one of the most commonly used drugs, has become a major concern because tobacco serves as a gateway drug and is linked to illicit drug abuse, such as cocaine and marijuana. However, previous studies mainly focused on certain genes or neurotransmitters which have already been known to participate in drug addiction, lacking endogenous metabolic profiling in a global view. To further explore the mechanism by which nicotine modifies the response to cocaine, we developed two conditioned place preference (CPP) models in mice. In threshold dose model, mice were pretreated with nicotine, followed by cocaine treatment at the dose of 2 mg/kg, a threshold dose of cocaine to induce CPP in mice. In high-dose model, mice were only treated with 20 mg/kg cocaine, which induced a significant CPP. 1H nuclear magnetic resonance based on metabonomics was used to investigate metabolic profiles of the nucleus accumbens (NAc) and striatum. We found that nicotine pretreatment dramatically increased CPP induced by 2 mg/kg cocaine, which was similar to 20 mg/kg cocaine-induced CPP. Interestingly, metabolic profiles showed considerable overlap between these two models. These overlapped metabolites mainly included neurotransmitters as well as the molecules participating in energy homeostasis and cellular metabolism. Our results show that the reinforcing effect of nicotine on behavioral response to cocaine may attribute to the modification of some specific metabolites in NAc and striatum, thus creating a favorable metabolic environment for enhancing conditioned rewarding effect of cocaine. Our findings provide an insight into the effect of cigarette smoking on cocaine dependence and the underlying mechanism.
Introduction
Nicotine and cocaine are two of the most widely abused stimulant drugs in the world. Many surveys have shown that there is a statistical and dramatic association between the use of licit drugs (alcohol or cigarettes) and other illicit drugs [1], [2]. Cigarette smoking acts as a precursor of later illicit drug use [3]. It is reported that 90.4% cocaine users had smoked cigarettes before they began to use cocaine [4]. Behavioral experiments have also proved that nicotine produces an effect on the response to other drugs. For example, nicotine exposure enhances cocaine-induced locomotor activity in mice [5]. Nicotine pre-exposure increases cocaine-induced place preferences in rats [6]. Additionally, several studies have been conducted to explore the mechanism by which nicotine reinforces the response to cocaine. It is reported that nicotine primes the response to cocaine by increasing its ability to induce transcriptional activation of FosB gene through inhibiting histone deacetylase [4]. Nicotine induces dopamine (DA) release in nucleus accumbens (NAc), potentially reinforcing cocaine’s behavioral effects [7], [8]. Nicotine exposure may alter the maturation of central nerve system, and results in a change in reward threshold, thus facilitating the vulnerability for drug dependence [9]. However, these studies only focused on certain genes or neurotransmitters which have already been known to participate in drug addiction, lacking endogenous metabolic profiling in a global view.
Currently, metabonomics has been widely applied in neuropsychiatric research fields, such as motor neuron disease, schizophrenia, Parkinson’s disease and drug addiction [10], [11], [12]. In schizophrenia field, metabonomics findings show systematic changes in pathways of glutamate metabolism and Krebs cycle in the cortex and hippocampus of rats treated with MK-801 [13]. Metabolomics acts as a powerful tool for detecting variations in a range of intracellular compounds upon drug exposure [14]. Unlike genomics, transcriptomics or proteomics, metabonomics shows what indeed happened and could detect the state of multiple metabolites, thus having a potential to identify the related biomarkers. Additionally, Nuclear magnetic resonance (NMR) spectroscopy technique, one of the most commonly used analytical methods in metabonomic study, has been extensively used to investigate the variation of whole metabolites in brain tissues [15]. By using 1H NMR-based on metabonomics, some researches showed that neurotransmitter pathways and energy metabolism are affected by addictive substances, such as morphine and cocaine [16], [17].
In this study, we developed a threshold dose model (50 µg/ml nicotine for 7 days, following 2 mg/kg cocaine for 3 days) and a high-dose model of cocaine (water for 7 days, following 20 mg/kg cocaine for 3 days) in mice to explore the mechanism underlying the effects of nicotine on the response to cocaine. We found that nicotine pretreatment dramatically enhanced the behavioral response to subsequent cocaine. Interestingly, with nicotine pretreatment, CPP induced by low dose of cocaine (2 mg/kg) was increased significantly, which was similar to the CPP induced by 20 mg/kg cocaine alone. More excitingly, the modified metabolites in above two models showed considerable overlap, including neurotransmitter, energy substances and membrane components. These results reflect neurotransmitter disturbance, energy metabolism imbalance as well as membrane disruption in NAc and striatum. Our study using behavioral models coupled with global metabolic profiling provides a new insight into the mechanism by which nicotine as a gateway drug reinforces cocaine’s rewarding effects. Moreover, our findings indicate that reprogramming of metabolites may affect cocaine abuse.
Materials and Methods
1. Drugs
Nicotine hydrogen tartrate was purchased from Sigma. Nicotine was dissolved in distilled water and administrated through the drinking water, which was stored in dark bottles and renewed every 2 days. Cocaine hydrochloride was purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China), and was dissolved in sterile saline before use.
2. Animal Models and Administration
Male C57BL/6J mice (8–12 weeks old) were kept in clear plastic cages with five per cage at under a 12/12 h light-dark cycle in room temperature (21±5°C) with an air change rate of 8–10 changes/hour and a relative humidity of 55±15%, and given food and water ad labium. The animals were acclimatized for 7 days before experiment. This study was carried out in accordance with the guidelines established by the Association for Assessment and Accreditation of Laboratory Animal Care. The protocols were approved by the Institutional Animal Care and Use Committee of the Institute (Protocol number IACUC-S200904-P001). All surgeries were performed under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering. The mice were randomly assigned to 5 groups: drinking water+saline; drinking water +2 mg/kg cocaine; drinking water +20 mg/kg cocaine (high-dose model); 50 µg/ml nicotine+saline; 50 µg/ml nicotine +2 mg/kg cocaine (threshold dose model). Table 1 summarizes the details of the experimental procedure of each group. Cocaine and saline were administrated by intraperitoneal injection (i.p) and nicotine by water drinking. All animal experiments were performed in accordance with the provisions of the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC).
| Groups | Pretreatment | After-treatment | |
| (drink, 7 d) | (i.p., 2 times/d, 3 d) | ||
| a.m. | p.m. | ||
| Saline (control) | water | saline | saline |
| 2 mg/kg cocaine | water | saline | 2 mg/kg cocaine |
| 20 mg/kg cocaine | water | saline | 20 mg/kg cocaine |
| Nicotine | nicotine | saline | saline |
| Nicotine +2 mg/kgcocaine | nicotine | saline | 2 mg/kg cocaine |
| Nicotine +20 mg/kgcocaine | nicotine | saline | 20 mg/kg cocaine |
3. Conditioned Place Preference (CPP)
CPP studies were conducted by using a shuttle box which was composed of two large conditioning chambers and a small central start chamber. One large conditioning chamber had black walls, and the other had white walls. Two groups (nicotine+saline group; nicotine +2 mg/kg cocaine group) were pretreated with nicotine (50 µg/ml) for 7 days continuously; the other three groups (drinking water+saline; drinking water +2 mg/kg cocaine; drinking water +20 mg/kg cocaine) were treated with water for 7 days. On day 8, all mice were placed in the central chamber and allowed to move freely in the apparatus for 30 min to determine the initial preference. Mice with a chamber bias greater than 75% were dropped from studies. On days 9–11 all mice in the study were given one injection of saline and placed in the preferred chamber for 30 min. Four hours later, mice were treated as follows and placed in the non-preferred chamber for 30 min: one injection of saline in drinking water+saline group or nicotine+saline group; one injection of 2 mg/kg cocaine in drinking water +2 mg/kg cocaine group or nicotine +2 mg/kg cocaine group; one injection of 20 mg/kg cocaine in drinking water +20 mg/kg cocaine group. On day 12, CPP test were conducted for all mice. Each mouse was allowed to move freely in all the chambers for 30 min. Time spent in the previously non-preferred and preferred chambers were recorded. Data were analyzed as time spent in the preferred chamber minus time spent in the non-preferred chamber. Each group was presented as mean ± SD, and one-way analysis of variance (ANOVA) followed by Tukey post hoc test was used to determine statistical significance.
4. Preparation of Brain Extracts
At the end of CPP test, mice were sacrificed, and brain NAc and striatum (30∼100 mg) were rapidly dissected and stored at −80°C. The preparation of brain samples was based on previous studies [18], [19]. Briefly, the frozen tissue with 0.8 ml of ultrapure water was homogenized by using Branson Sonifier (Japan) for 2 min (200 W) at 4°C, and 0.8 ml ice-cold chloroform was added into the homogenate. The mixture was mixed for 2 min and kept for 10 min on ice, followed by centrifugation at 13,000×g for 10 min at 4°C. The supernatant (∼0.5 ml) was preserved to lyophilize for about 36 hours. After lyophilization, the powder was added into 0.52 ml D2O (heavy water) including 0.01 mg/ml sodium (3-trimethylsilyl)-2, 2, 3, and 3-tetradeuteriopropionate (TSP). The supernatant (∼0.5 ml) was shifted into a 5 mm NMR tube for 1H NMR detection after centrifugation at 13, 000×g for 5 min at 4°C [20].
5. Solution 1H NMR Spectroscopy
All the spectral data were obtained on a Bruker-Av II 600 MHz spectrometer (Bruker Co., Germany) at 300 K. A one-dimensional spectrum was acquired by using a standard (1D) Carr–Purcell–Meibom–Gill (CPMG) spin-echo pulse sequence, which suppressed the water signals. The free induction decays were weighted by an exponential function with a 0.3-Hz Gaussian Maximum position 0.1, prior to Fourie transformation.
6. Data Reduction and Pattern Recognition Analysis
All NMR spectra were automatically reduced to 440 segments, each with a 0.02 ppm width ranging from 0.2 to 4.6 ppm and 5.1 to 9.4 ppm using MestRe-c2.3 software (http://qobrue.usc.es/jsgroup/MestRe-c). The area for each segmented region was calculated. The region of the spectrum (δ 4.6–5.1 ppm) was removed to exclude the influence of water signal. To account for dilution or bulk mass differences between samples, each spectral intensity data set was normalized to the total sum of the spectral regions following exclusion of the water resonance. The datasets were mean-centered prior to partial least squares (PLS), principal component analysis (PCA), orthogonal signal correction (OSC) analysis by the SIMCA package. Two-dimensional score plots were used to visualize the separation of the samples and the corresponding loading plots were applied to identify the altered contribution to the position of spectra. PCA distinguished the characteristic variable (metabolic signals) or the outlier from the group by statistical method. PLS, a supervised pattern recognition (PR) method, was subsequently applied to enhance this separation. OSC, a spectral filtering method, was applied to optimize the separation. The variable importance (VIP) could inform the important values to separate the cluster and the corresponding loadings for PLS models after application of OSC were applied to identify the altered contribution to the position of spectra that was changed after drug treatment. VIP>1 of multivariate were identified distinguishing metabolites. 1H NMR chemical shifts and assignments of endogenous metabolites were conducted according to the previous literatures and the Human Metabolome Database [21] (http://www.hmdb.ca/), a web-based bioinformatic/cheminformatic resource with detailed information about metabolites and metabolic enzymes. All analyses were carried out by SPSS 11.5, and P<0.05 was considered statistically.
Results
1. Exploring Threshold Dose of Cocaine-induced CPP in Mice
Each group spent almost the same time on the initially non-preferred side, indicating that there were no basal differences among groups. To investigate the threshold dose of cocaine-induced CPP in mice, we designed two doses of cocaine (5 mg/kg and 2 mg/kg) to treat mice separately. After administration, it was observed that 90% mice in 5 mg/kg cocaine group (n = 12) spent significantly more time in the non-preferred chamber, and only 10% still spent more time in the initially preferred chamber on the posttest day than the pretest day (P<0.05). However, 50% mice in 2 mg/kg cocaine group (n = 12) moved to the non-preferred chamber for more time, and 50% still stayed in the initially preferred chamber for more time on the post-test day than the pre-test day (P<0.05) (Figure 1A.). These results showed that 2 mg/kg is a threshold dose to the development of CPP induced by cocaine in C57BL/6J mice.
2. Nicotine Priming Increases Cocaine-conditioned Place Preference
Cocaine (20 mg/kg dose) has been widely used to induce the development of CPP in mice [22], [23]. Either 2 mg/kg or 20 mg/kg cocaine alone could increase CPP in mice compared with saline control. With nicotine pretreatment mice receiving 2 mg/kg cocaine displayed a 150% further increase of the time in cocaine-paired chamber, which was very close to the time of 20 mg/kg cocaine group (Figure 1B.). These results indicated that nicotine priming can enhance behavioral response to cocaine.
3. NMR Spectra and OSC-PLS Analysis
Representative 1H NMR spectra of the water extracts of striatum from five groups were shown and major metabolites in the integrate regions were assigned in Figure 2. Visual inspection of 1H-NMR spectra indicated the differences in these groups. We further utilized PCA, PLS and OSC to gain insights into biochemical information at the molecular level from 1H NMR spectra. As an unsupervised PR method, PCA was initially used to analyze the data sets of the 1H NMR spectra. Subsequently, PLS, a supervised PR method, was applied to increase this separation. However, there was no clear separation in the brain NMR spectra for the first two principal components (PCs) in each treated group when these two PR methods were used. Then, PLS model following OSC was performed to separate NMR spectra among treated groups. After application of OSC-PLS model, the PLS scores plots displayed a significant differentiation among treated groups (Figure 3. and Figure 4.).
In order to confirm the quality and effectiveness of OSC-PLS model, a permutation method was used. The original data points (0∼1) in the training set were mathematically reproduced, and 1 indicates a model with a perfect fit. Q2Y values >0.5 and >0.9 express good and excellent predictive abilities, respectively. High values of R2Y and Q2Y in PLS models in our study indicated that OSC-PLS model was valid (Figure 5. and Figure 6.). For example, R2Y and Q2Y values of NAc samples from 20 mg/kg cocaine group were 0.99 and 0.959, respectively (Figure 5.).
4. Metabolic Changes Induced by Drugs in Brain NAc and Striatum of Mice
To investigate the changes of metabolites in NAc and striatum, PLS loading plots after application of OSC data filter were applied to analyze 1H NMR data (Figure 7. and Figure 8.). In NAc, the metabolites that predominantly contributed to the separation of drug-treated groups were glutamate (Glu), tryptamine, glucose, lactate (Lac), creatine (Cre), 1-methylhistidine, glutamine (Gln) and profine (Figure 7. and Table 2). In striatum, the metabolites that predominantly contributed to the separation were Glu, glucose, a-ketogultaric acid, Lac and Gln (Figure 8. and Table 3). Interestingly, most of these modified metabolites in 20 mg/kg cocaine group (high-dose model) were overlapped with those in combinational drugs group (threshold dose model), revealing an analogous outcome at metabolic profiling (Table 2 and Table 3). Several identical metabolites modified in these two models were also observed in 2 mg/kg cocaine or nicotine alone groups. However, a few metabolites were not significantly changed in 2 mg/kg cocaine group, but were modified clearly when pretreated with nicotine, suggesting that these metabolites were supplied by nicotine priming (Table 2 and Table 3). Correspondingly, according to the VIP values (VIP≥1) in OSC-PLS model, these changed metabolites in drug groups were identified and listed in Table 2 and Table 3, respectively.
| Metabolitesa | Chemicalshift (ppm) | Control vs. 20 mg/kgcocaine | Control vs. nicotine+2 mg/kg cocaine | 2 mg/kg cocaine vs.nicotine +2 mg/kg cocaine | Control vs.2 mg/kg cocaine | ||||
| VIPb | Loadingc | VIPb | Loadingc | VIPb | Loadingc | VIPb | Loadingc | ||
| Glutamate | 3.72 | 14.95 | +0.728 | 14.91 | +0.715 | 15.19 | +0.726 | 1.16 | +0.057 |
| Glutamate | 3.74 | 10.23 | +0.498 | 9.45 | +0.453 | 9.57 | +0.457 | / | / |
| Tryptamine | 7.70 | 4.54 | +0.171 | 2.05 | −0.098 | 4.52 | +0.216 | 9.55 | −0.487 |
| Tryptamine | 7.68 | 2.99 | +0.112 | 1.43 | −0.068 | 3.64 | +0.174 | 7.34 | −0.374 |
| Glucose | 3.76 | 2.41 | +0.117 | / | / | / | / | 1.99 | −0.102 |
| Lactate | 1.34 | 2.05 | −0.089 | 1.23 | −0.059 | 5.26 | −0.249 | 6.48 | +0.288 |
| * | 3.66 | 1.92 | +0.094 | 2.66 | +0.128 | 1.73 | +0.083 | 1.84 | +0.081 |
| Acetylcholine | 3.24 | 1.83 | −0.089 | / | / | / | / | 3.32 | −0.159 |
| 1-Methylhistidine | 3.70 | 1.57 | +0.076 | 1.51 | +0.073 | 1.52 | +0.072 | 1.02 | +0.014 |
| L-glutamine | 2.14 | 1.45 | −0.071 | 2.37 | −0.113 | / | / | 2.28 | −0.114 |
| L-methionine | 2.16 | 1.37 | −0.065 | 2.38 | −0.114 | / | / | 2.75 | −0.139 |
| Creatine | 3.04 | 1.27 | −0.062 | 1.38 | −0.065 | 2.37 | −0.113 | 1.29 | +0.066 |
| α-Ketogultaric acid | 2.48 | 1.14 | −0.055 | 1.47 | −0.071 | / | / | 1.96 | −0.09 |
| 1-Methylhistidine | 3.68 | 1.08 | +0.052 | / | / | / | / | / | / |
| α-Ketogultaric acid | 2.46 | 1.07 | −0.051 | 1.53 | −0.073 | / | / | 2.05 | −0.099 |
| * | 3.56 | 1.06 | −0.051 | / | / | / | / | 1.01 | +0.012 |
| Phosphocholine | 3.22 | 1.03 | −0.048 | / | / | / | / | 1.22 | +0.049 |
| Proline | 4.12 | 1.01 | +0.047 | 1.24 | +0.058 | 1.68 | −0.08 | 4.15 | +0.211 |
| Metabolitesa | Chemicalshift (ppm) | Control vs.20 mg/kg cocaine | Control vs.Nicotine +2 mg/kg cocaine | 2 mg/kg cocaine vs.nicotine +2 mg/kgcocaine | Control vs.nicotine | Control vs.2 mg/kg cocaine | |||||
| VIPb | Loadingc | VIPb | Loadingc | VIPb | Loadingc | VIPb | Loadingc | VIPb | Loadingc | ||
| Glutamate | 3.72 | 17.04 | +0.816 | 15.98 | +0.767 | 15.61 | +0.746 | 14.28 | +0.684 | 9.76 | +0.506 |
| Glutamate | 3.74 | 10.52 | +0.504 | 10.49 | +0.503 | 10.20 | +0.487 | 10.35 | +0.496 | 6.51 | +0.335 |
| Lactate | 1.34 | 2.92 | −0.138 | / | / | 6.12 | −0.291 | 1.26 | −0.024 | 7.06 | +0.332 |
| * | 3.66 | 2.12 | +0.102 | 2.35 | +0.113 | 1.62 | +0.076 | 1.69 | +0.081 | 2.23 | +0.116 |
| 1-Methylhistidine | 3.70 | 1.62 | +0.077 | / | / | / | / | / | / | / | / |
| Glucose | 3.76 | 1.24 | +0.059 | / | / | 2.05 | +0.098 | 3.34 | +0.159 | 1.67 | −0.084 |
| 1-Methylhistidine | 3.68 | 1.21 | +0.058 | / | / | / | / | / | / | / | / |
| α-Ketogultaric acid | 2.46 | 1.08 | −0.051 | 1.62 | −0.074 | / | / | / | / | 2.87 | −0.139 |
| Lactate | 1.36 | 1.03 | −0.049 | / | / | 1.53 | −0.073 | / | / | 1.80 | +0.095 |
| L-glutamine | 2.14 | 1.00 | −0.048 | 3.08 | −0.146 | / | / | / | / | 4.60 | −0.242 |
5. Neurotransmitter Changes
We found that Glu, tryptamine, Gln and 1-methylhistidine were markedly altered in NAc and striatum after treatment of nicotine or cocaine. Glu (3.72 and 3.74 ppm) in NAc and striatum was increased by 2 mg/kg cocaine. With nicotine pretreatment, 2 mg/kg cocaine significantly increased the level of Glu. Interestingly, Glu was also significantly elevated by 20 mg/kg cocaine (Table 2 and Table 3). Markedly changed neurotransmitters in NAc and striatum are listed in Table 2 and Table 3.
We then carefully analyzed the altered metabolites among treatment groups. We found that tryptamine (7.7 and 7.68 ppm) clearly decreased in 2 mg/kg cocaine group, but only slightly reduced in nicotine +2 mg/kg cocaine group. Moreover, tryptamine was slightly increased by 20 mg/kg cocaine (Table 4). Additionally, Gln in NAc and striatum were markedly lowered by 2 mg/kg cocaine alone, but only showed a little decrease with nicotine preconditioning. Interestingly, a little decrease of Gln also displayed in 20 mg/kg cocaine alone group (Table 4 and Table 5). These results showed that after nicotine pretreatment, neurotransmitter profiles induced by subsequent 2 mg/kg cocaine shifted toward that of 20 mg/kg cocaine group.
| Metabolitesa | Chemicalshift (ppm) | Control vs.20 mg/kg cocaine | Control vs.nicotine +2 mg/kg cocaine | 2 mg/kg cocaine vs.nicotine +2 mg/kg cocaine | Control vs.2 mg/kg cocaine |
| Loadingb | Loadingb | Loadingb | Loadingb | ||
| Glutamate | 3.72 | ++ | ++ | ++ | + |
| Glutamate | 3.74 | ++ | ++ | ++ | / |
| Tryptamine | 7.70 | ++ | − | + | – |
| Tryptamine | 7.68 | ++ | − | + | – |
| Glucose | 3.76 | + | / | / | − |
| Lactate | 1.34 | – | − | − | ++ |
| * | 3.66 | + | ++ | ++ | + |
| Acetylcholine | 3.24 | − | / | / | − |
| 1-Methylhistidine | 3.70 | ++ | ++ | ++ | + |
| L-glutamine | 2.14 | − | − | / | − |
| L-methionine | 2.16 | − | − | / | − |
| Creatine | 3.04 | − | − | − | + |
| α-Ketogultaric acid | 2.48 | − | − | / | − |
| 1-Methylhistidine | 3.68 | + | / | / | / |
| α-Ketogultaric acid | 2.46 | − | − | / | − |
| * | 3.56 | − | / | / | + |
| Phosphocholine | 3.22 | − | / | / | + |
| Proline | 4.12 | + | + | − | ++ |
| Metabolitesa | Chemical shift(ppm) | Control vs.20 mg/kg cocaine | Control vs. nicotine+2 mg/kg cocaine | 2 mg/kg cocaine vs.nicotine +2 mg/kg cocaine | Control vs.Nicotine | Control vs.2 mg/kg cocaine |
| Loadingb | Loadingb | Loadingb | Loadingb | Loadingb | ||
| Glutamate | 3.72 | ++ | ++ | + | ++ | + |
| Glutamate | 3.74 | ++ | ++ | + | ++ | + |
| Lactate | 1.34 | − | / | − | – | + |
| * | 3.66 | ++ | ++ | + | ++ | + |
| 1-Methylhistidine | 3.70 | + | / | / | / | / |
| Glucose | 3.76 | + | / | + | ++ | − |
| 1-Methylhistidine | 3.68 | + | / | / | / | / |
| α-Ketogultaric acid | 2.46 | − | − | / | / | − |
| Lactate | 1.36 | − | / | − | / | + |
| L-glutamine | 2.14 | − | − | / | / | − |
7. Disruption of Membrane and Amino Acids
Phosphocholine (3.22 ppm), a membrane ingredient, was elevated by 2 mg/kg cocaine alone; however, it showed no obvious alteration in nicotine +2 mg/kg cocaine group or nicotine alone group. A somewhat decline in phosphocholine was induced by 20 mg/kg cocaine (Table 2). Proline (4.12 ppm) in NAc displayed a slightly low level after nicotine treatment, whereas it was markedly elevated by 2 mg/kg cocaine alone (Table 2). Interestingly, increased proline level in threshold dose model was also found in high-dose model. Additionally, L-methionine (2.16 ppm) was slightly declined by nicotine pretreatment, whereas it was significantly decreased by following 2 mg/kg cocaine. Surprisingly, a decline of L-methionine in threshold model was also observed in high-dose model (Table 4 and Table 5). These results indicate that after nicotine pretreatment, reprogramming of metabolites induced by a low dose of cocaine (2 mg/kg cocaine) is extremely comparable with that induced by a high dose of cocaine (20 mg/kg cocaine) alone.
8. Discovery of Key Metabolites Related to Rewarding Effects of Cocaine Elicited by Nicotine
To further gain an insight into the key metabolites associated with rewarding effects of cocaine primed by nicotine pretreatment, we used box-and-whisker plots to clarify altered levels of those identified metabolites. We obtained 12 representative metabolites among these five treatment groups (saline, 2 mg/kg cocaine, nicotine, nicotine +2 mg/kg cocaine and 20 mg/kg cocaine). Interestingly, 10 identical metabolites in NAc showed a similar change between 20 mg/kg cocaine group and nicotine +2 mg/kg cocaine group, including Glu, tryptamine, Lac, 1-methylhistidine, Gln, L-methionine, Cre, α-ketoglutaric acid, phosphocholine and proline (Figure 9.). Some modified metabolites in above two models were overlapped with those in 2 mg/kg cocaine group or nicotine group, such as Glu and 1-methylhistidine. Four metabolites, Gln, L-methionine, α-ketoglutaric acid and phosphocholine, showed a consistent trend only in 2 mg/kg cocaine group. Tryptamine, Lac, Cre and proline were altered only in nicotine group (Figure 10A.). The change tendency of above metabolites in NAc was also observed in striatum. These results show that with nicotine pretreatment threshold dose of cocaine could reprogram and shift the metabolic profiles in NAc and striatum to that of high dose cocaine. These modified metabolites mainly included neurotransmitter, energy substances, membrane components and amino acids. For example, Cre in NAc showed somewhat high level in 2 mg/kg cocaine group; however, after nicotine pretreatment, 2 mg/kg cocaine clearly decreased its level to that of 20 mg/kg cocaine group. Meanwhile, Lac displayed a similar alteration to Cre. The α-ketoglutaric acid in NAc and striatum was significantly decreased by 2 mg/kg cocaine, whereas it displayed a slight decline in combinational drugs group, which was consistent with the findings in 20 mg/kg cocaine group (Table 5). Taken above, after nicotine pretreatment, metabolic profiling of threshold dose of cocaine shifts to that of high-dose of cocaine. We consider that nicotine priming may provide a favorable metabolite environment in brain, facilitating the rewarding effect of subsequent cocaine.
Discussion
Addictive drugs, including heroin, marijuana, nicotine and cocaine, exert their addictive effects in part by increasing the level of dopamine, a pivotal neurotransmission in the control of initial drug use in the ventral striatum [24], [25]. Nicotine and cocaine have interactive neurochemical effects, particularly with regard to dopamine transmission [26]. Many studies have been performed to explore the effect of nicotine on cocaine. For example, nicotine exposure in adolescence has been shown to alter the rewarding effect of cocaine in adulthood [27]. Nicotine enhances locomotor activity to cocaine and activates other indirect dopamine agonists [28], [29]. However, these studies mainly focus on certain genes or neurotransmitters, and cannot understand the underlying mechanism from systems biology framework. It is well known that NAc and striatum are the two key components of brain’s reward circuitry [30], [31]. Therefore, we applied 1H NMR-based on metabonomic in NAc and striatum to study the effect of nicotine preconditioning on cocaine.
With nicotine pretreatment,low dose of cocaine (2 mg/kg) dramatically increased CPP of mice. Interestingly, the mice receiving 20 mg/kg cocaine or nicotine +2 mg/kg cocaine displayed not only analogous CPP but also extremely similar metabolic profiling. In other words, nicotine pretreatment moved the metabolic profiling of threshold dose of cocaine to that of high dose. We consider that the effect of nicotine on reinforcing behavioral response to cocaine may be attributed to nicotine-induced modification of some specific metabolites, thus creating a favorable environment of metabolites for conditioned rewarding effects of cocaine. The changes of metabolites in NAc and striatum involved in neurotransmitter disturbance, energy metabolism imbalance, membrane and amino acids disruptions.
1. Disturbance in Neurotransmitters Specific to Nicotine and Cocaine Treatment
Interactions between nicotine and cocaine have been studied in rodents, and it has been found that nicotine can enhance cocaine’s behavioral effects [28]. Drug-related learning, memory and sensitization are linked to a glutamate-mediated LTP induction and maintenance at these synapses of smokers [32], [33], [34]. In this study, Glu in NAc and striatum was slightly increased by 2 mg/kg cocaine alone. Surprisingly, after nicotine pretreatment, Glu was dramatically increased by low-dose cocaine (2 mg/kg), reaching to the level of high dose (20 mg/kg). Gln in NAc was decreased similarly in threshold dose and high-dose model of cocaine. We guess that the reduction in Gln could be attributed to being hydrolyzed to Glu. Previous studies have shown that cocaine indirectly influences Glu transmission in the limbic system, producing persistent changes in neuronal function that alters behavioral effect of cocaine [35], [36], [37]. Similarly, reinstatement of cocaine seeking is linked to the increased Glu release in NAc [38]. Thinking above, in the present study, increase of Glu induced by nicotine should contribute to the induction of glutamatergic synaptic plasticity, likely through activation of N-methyl-D-aspartic acid receptor (NMDAR) [39]. Glu increases glutamate release onto DA neurons in ventral tegmental area, and initially activates NMDA receptor-mediated signaling cascades, which, in turn, upregulate α-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid receptors and induce glutamatergic synaptic plasticity [40]. We speculate that nicotine’s reinforcing effect on cocaine CPP, at least in part, owe to the increase of Glu release, thus leading to Glu receptor activation and modification of glutamatergic synaptic plasticity.
Additionally, we found that nicotine pretreatment increased tryptamine level in NAc, which was further elevated by subsequent 2 mg/kg cocaine. Interestingly, the elevated level of tryptamine in threshold dose model was similar to that in high-dose model. It is known that 5-HT, a derivative of tryptamine, plays a role in initiation and maintenance of addictive behavior [41]. Therefore, elevated tryptamine may play a part in the reinforcing effect of nicotine on cocaine. Collectively, our results, together with previous studies, suggest that nicotine exerts its effect on cocaine though creating a specific environment of metabolite, thus reinforcing conditioned rewarding effects of cocaine.