Involvement of the dopaminergic system in the reward-related behavior of pregabalin

There has been an increase in cases of drug addiction and prescription drug abuse worldwide. Recently, pregabalin abuse has been a focus for many healthcare agencies, as highlighted by epidemiological studies. We previously evaluated the possibility of pregabalin abuse using the conditioned place preference (CPP) paradigm. We observed that a 60 mg/kg dose could induce CPP in mice and that pregabalin-rewarding properties were mediated through glutamate neurotransmission. Notably, the dopaminergic reward circuitry is also known to play a crucial role in medication-seeking behavior. Therefore, this study aimed to explore the possible involvement of dopaminergic receptor-1 in pregabalin-induced CPP. Mice were randomly allocated to receive saline or the dopamine-1 receptor antagonist SKF-83566 (0.03 mg/kg, intraperitoneal). After 30 min, the mice received either saline or pregabalin (60 mg/kg) during the conditioning phase. Among the control groups that received saline or SKF-83566, the time spent in the two conditioning chambers was not significantly altered. However, among the pregabalin-treated group, there was a marked increase in the time spent in the drug-paired chamber compared to the time spent in the vehicle-paired chamber. Notably, blocking dopamine-1 receptors with SKF-83566 completely prevented pregabalin-induced place preference, thus demonstrating the engagement of the dopaminergic system in pregabalin-induced reward-related behavior.

www.nature.com/scientificreports/ data using ANY-maze software to determine the baseline values. No mouse had a chamber preference exceeding 67% of the total time of the preconditioning phase. During the conditioning phase (days [4][5][6][7][8][9][10][11], the mice received pretreatment intraperitoneal injection (i.p.) of saline or the D 1 receptor antagonist SKF (0.03 mg/kg) 30 min before the administration of saline (10 ml/kg) or pregabalin (PGB) (60 mg/kg, i.p.). Each mouse was administered pregabalin or the vehicle four times every other day for eight conditioning days. Subsequently, we conducted the postconditioning test on day 12, with each mouse being placed in the CPP apparatus without restriction between the chambers for 30 min. Finally, the time spent in each chamber was assessed using the same method that was used to investigate the pretreatment behavior.
Procedure. We randomly assigned the mice to one of four groups, as shown in Fig. 1B. Group 1 (Saline-Saline) mice were administered saline (10 ml/kg, i.p.) 30 min before receiving the same dose of saline (10 ml/kg, i.p.) for eight sessions (n = 6). Group 2 (SKF-Saline) mice were administrated SKF (0.03 mg/kg, i.p.) 30 min before receiving saline (10 ml/kg, i.p.) for four sessions (eight sessions in total), with alternating saline sessions during the conditioning phase (n = 8). Group 3 (saline-pregabalin) mice received saline (10 ml/kg, i.p.) 30 min before receiving pregabalin (60 mg/kg) for four sessions (eight sessions in total), with alternating saline sessions during the conditioning phase (n = 6). We selected this dose based on our previous finding that 60 mg/kg pregabalin can induce CPP 18 . Group 4 (SKF-pregabalin) mice received SKF (0.03 mg/kg, i.p.) 30 min before receiving pregabalin (60 mg/kg) for four sessions (eight sessions in total), with alternating saline sessions during the conditioning phase (n = 7). Subsequently, the place preference was assessed after all the conditioning sessions had been completed.

Statistical analysis.
For all CPP behavioral studies, the time that each mouse spent in each chamber pretest and posttest was analyzed using two-way repeated-measures analysis of variance (RM ANOVA). We performed the Newman-Keuls multiple comparisons test using GraphPad Prism. A p-value of < 0.05 was the chosen level of significance.
Effects of pregabalin and SKF-83566 on pregabalin-induced place preference. In the salinepregabalin group (group 3), we observed a significant effect on the phase (F (1, 5) = + infinity, p < 0.0001) and chamber (F (1, 5) = 24.90, p = 0.0041), as well as a significant interaction between the phase and chamber (F (1, 5) = 28.55, p = 0.0031). The post hoc analysis revealed a significant increase in the time spent in the pregabalin-paired chamber compared to that spent in the saline-paired chamber during the postconditioning test (p < 0.0100; Fig. 3A). Moreover, there was an increase in the time spent in the pregabalin-paired chamber during www.nature.com/scientificreports/ the postconditioning test compared to that during the preconditioning test (p < 0.0500). Notably, pretreatment with SKF attenuated the pregabalin-induced CPP. We observed a significant effect on the phase (F (1, 6) = 15,210, p < 0.0001), no significant effect on the chamber (F (1, 6) = 0.08476, P = 0.7807), and no significant interaction between the phase and chamber (F (1, 6) = 1.242, p = 0.3077) (Fig. 3B).

Discussion
There is empirical proof that the dopamine mechanism plays an essential role in substance-related reward and motivation [41][42][43] . Moreover, the association between the dopaminergic receptors and the reward signals in the midbrain has revealed that blocking dopamine receptors could attenuate the reward circuitry involved in different drugs of abuse [44][45][46] . The addictive properties of pregabalin remain controversial; however, several reports have shown an association of pregabalin administration with euphoric effects and abuse potential 3,47 . A previous study demonstrated that pregabalin increased dopamine release in the NAc in a rat model of neuropathic pain 48 . This indicates that exposure to pregabalin could potentiate this connection and subsequently increase dopamine levels. Several reports have shown that SKF is able to interfere with cocaine-evoked dopamine release in vitro, suggesting that this compound may be a potential candidate in attenuating the effects of cocaine in vivo 49 . Moreover, it was reported that pretreatment with SKF blocked amphetamine-induced locomotor stereotypy and hyperlocomotion 39 . Furthermore, SKF blocked the effect of propofol in inducing glutamate neurotransmission in rat midbrain slices via presynaptic D 1 receptors 50 . Therefore, the D 1 receptor has been proposed as an important target for testing the behavioral effects related to dopaminergic and glutamatergic neurotransmissions 51,52 . Notably, in the current study, we found that pretreatment with SKF attenuated the reward effects of pregabalin in CPP. Accumulating evidence indicates that dopamine transport is important for behavioral reward regulation in the NAc [53][54][55] . Pavlovian conditioning using the CPP model has been known to assess the rewarding effects of drugs 56 and be dependent on NAc dopaminergic neurotransmission 57 . The mesolimbic dopaminergic pathway, including the ventral tegmental area (VTA) and the NAc, is considered a critical neural region underlying reward and drug-seeking behavior 58 . Activation of D1 receptors, but not D2 receptors, in the NAc is essential for longterm potentiation and positive reinforcement behavior 59 . However, blocking D1 or D2 receptors has shown to impair locomotion and rearing effects associated with dopaminergic neuron inhibition in the NAc core and shell 60 . Several reports have investigated the role of substance abuse in the release of dopamine in rewarding circuits 61 . Importantly, repeated morphine treatment for four doses has been shown to induce CPP in mice 62 . The latter study suggested the development of receptor supersensitivity for postsynaptic dopamine in mice. In confirming this effect, apomorphine (a dopamine agonist) produced stimulated response in ambulatory activity. Moreover, administration of the NMDA antagonist, Mk-801 inhibited the morphine-induced CPP behavior and the development of postsynaptic dopamine receptor supersensitivity. In addition, this effect was observed with other drugs of abuse including cocaine and methamphetamine in mice 63,64 . This is consistent with our findings that blocking D1 receptors attenuated pregabalin-induced CPP. Therefore, pregabalin might induce rewarding effects through the activation of postsynaptic D1 receptors in the NAc. Further studies are needed to examine the effects of pregabalin on dopamine release in the NAc.
Additionally, the glutamatergic system is significantly involved in mediating the drug-seeking effects of several abuse drugs. Drug-seeking behavior has been linked to glutamatergic imbalance in the NAc and downregulation of the glial excitatory amino acid transporter (GLT-1), which is the main regulator of glutamatergic www.nature.com/scientificreports/ homeostasis in the brain [65][66][67][68] . For example, cocaine-seeking behavior has been linked to the downregulation of GLT-1 expression 69 . Furthermore, the disturbance in the glutamatergic system is linked with the spillover of glutamate when the uptake of glutamate by a synapse is decreased, which in turn overactivates the postsynaptic receptors that mediate drug-seeking behavior 70 . Moreover, presynaptic glutamate receptors such as metabotropic glutamate receptors Type 2 (mGlu2/3) have been shown to regulate glutamate release in the NAc and prefrontal cortex and be involved in reward and drug-seeking behavior 71 . Interestingly, these two systems of dopamine and glutamate have been shown to be interconnected and to influence each other in brain regions. It has been proposed that D 1 receptors are located in the presynaptic glutamatergic terminal of VTA 72 . Activating D 1 receptors facilitates the release of glutamate in the VTA 52 . Additionally, it has been mentioned that ethanol-induced spontaneous excitatory postsynaptic currents (sEPSCs) via glutamate alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) receptors are suppressed by SKF in the VTA 73,74 . Several studies have shown that ceftriaxone, via the upregulation of GLT-1, can attenuate ethanol intake and relapse in rats 75,76 . The involvement of glutamatergic neurotransmission in pregabalin-induced CPP has been previously reported 18 and could be due to the activation of presynaptic D 1 receptors located within glutamatergic synapses. Where, activating the D1 dopamine receptor could augment AMPA receptor transmission as shown in the NAc cell cultures prepared from rat pups 77 . Of note, the interaction between dopamine and glutamate is complex in the NAc. The glutamatergic activation in the VTA has been shown to increase dopaminergic activity and release in the NAc 78,79 . Moreover, glutamate, at presynaptic level in the NAc, can facilitate dopamine release 80,81 . Dopamine can also modulate glutamatergic firings in the NAc that originate from the hippocampus amygdala 82 . Interestingly, this effect has been shown to be mediated through D1 receptors.
Several studies have suggested that repeated dopaminergic activation during behavioral conditioning performs an essential function in the cue stimuli, leading to drug-seeking behavior [83][84][85] . Reportedly, dopaminergic neurotransmission is associated with glutamate release 86 . In fact, dopamine terminals within the NAc cross on single dendrites with glutamatergic terminals across several key brain regions such as the hippocampus, prefrontal cortex, and amygdala 87 . Furthermore, reports indicate a strong similarity between the activation of the glutamate receptor N-methyl-d-aspartate (NMDA) and the D 1 receptor in drug reward paradigms 88,89 . Moreover, it has been shown that glutamate is released upon dopaminergic neurotransmission in the midbrain region in in vitro and in vivo models [90][91][92][93][94][95] . For example, in the NAc shell, dopaminergic terminals were found to release glutamate when activated with channelrhodopsin-2 96 . In methamphetamine seeking, both the NAc and dorsomedial prefrontal cortex showed high levels of glutamate and dopamine when analyzed by microdialysis 97 , whereas systemic administration of a D 1 receptor antagonist (SCH 23390), but not a dopamine receptor-2 antagonist (eticlopride), attenuated methamphetamine seeking 98 . Although standard receptor binding tests have revealed that pregabalin is not bound to the D 1 receptor 99 , it might increase dopamine levels and glutamate release with the euphoric mental state being achieved. Therefore, the D 1 receptor antagonist counteracted the D 1 receptor effects in glutamate release regulation and inhibited the dopamine effects. This could lead to a decrease in glutamate and dopamine release 86,100 . This is in line with our earlier results which confirmed an association between glutamatergic neurotransmission and the rewarding effects of pregabalin 18 . These interconnections between dopamine and glutamate in key brain regions support our previous findings that ceftriaxone, a known regulator of glutamate homeostasis, attenuates pregabalin-induced CPP 18 . Blocking the D1 receptors with SKF in the current study also blocked pregabalin induced CPP. Together, dopaminergic and glutamatergic neurotransmissions in key brain regions might play a significant role in pregabalin-induced CPP. This is consistent with several known drugs of abuse where both dopaminergic and glutamatergic neurotransmissions have been shown to be involved in their rewarding effects [61][62][63][64][65][66][67][68] . Studies are warranted to investigate the neurochemical interactions between dopaminergic and glutamatergic systems in pregabalin induced reward.
The fact that there are no biological studies to confirm these findings is one of the limitations of the present study. Measuring the level of dopamine and glutamate in key brain regions during the conditioning phases, as well as the posttest phase, could provide insights into the mechanisms of pregabalin rewarding properties. A recent study revealed that acute administration of pregabalin did not affect the dopamine level in the NAc 101 . However, the study did not investigate other brain areas such as the NAc subregions (core and shell), prefrontal cortex, or the VTA. Thus, testing only a single area may not be enough to determine the effects of pregabalin on the dopaminergic system as a whole. Another limitation of this study was not assessing the impact of SKF on pregabalin-induced CPP in a dose-dependent manner. This should be considered for examination in future studies. Moreover, it may be worthwhile to assess whether the effect of SKF in blocking pregabalin-induced CPP may also affect other brain neurotransmitters.
A previous in vitro study on human neocortical slices assessed different neurotransmitters and reported that pregabalin modulates acetylcholine, serotonin, and norepinephrine release without changing dopamine release 102 . However, acetylcholine-mediated activation of the D 1 receptor by SKF-38393 in striatal cells has been shown [103][104][105] , and this activation was inhibited by SKF 103 . Similarly, SKF appears to partially act against serotonin receptor-1c 106 . Pregabalin interaction with the alpha-2 delta subunit of the calcium channel remains only partially understood. Previous findings regarding pregabalin indicate an essential role of GABAergic neurotransmission in the reward and dependence effects of drugs of abuse 107 . Since pregabalin is a GABA analog, its abuse liability might involve GABA-modulating properties. Specifically, pregabalin administration has been found to slightly increase the extracellular GABA levels in the brain 7,108-110 . Therefore, the weak GABAergic activity of pregabalin may induce GABA-mimetic activity that influences the euphoria and relaxation described by some patients and drug abusers. Thus, future investigations of the effects of SKF on several neurotransmitter systems are needed to understand the mechanistic effect of pregabalin-induced CPP.
In conclusion, findings after pretreatment with SKF indicated that the D 1 receptors might play a crucial role in the ability of pregabalin to induce behavioral sensitization through the dopamine reward system. However, www.nature.com/scientificreports/ there is a need for further neurochemical studies to identify similarities in the abuse liability mechanism between pregabalin and other defined addictive drugs.

Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.