Mediation of mPFC‐LHb pathway in acupuncture inhibition of cocaine psychomotor activity

The medial prefrontal cortex (mPFC) and the lateral habenula (LHb) play roles in drug addiction and cognitive functions. Our previous studies have suggested that acupuncture at Shenmen (HT7) points modulates mesolimbic reward system in order to suppress drug‐induced addiction behaviours. To explore whether an mPFC‐LHb circuit mediates the inhibitory effects of acupuncture on addictive behaviours, we examined the projection from mPFC to LHb, excitation of mPFC neurons during acupuncture stimulation, the effects of optogenetic modulation of mPFC‐LHb on HT7 inhibition of cocaine‐induced locomotion and the effect of mPFC lesion on HT7 inhibition of nucleus accumbens (NAc) dopamine release. Acupuncture was applied at bilateral HT7 points for 20 s, and locomotor activity was measured in male Sprague–Dawley rats. Although cocaine injection significantly increased locomotor activity, HT7 acupuncture suppressed the cocaine‐induced locomotion. The inhibitory effect of HT7 on cocaine‐enhanced locomotion was blocked by optogenetic silencing of the mPFC‐LHb circuit. In vivo extracellular recordings showed that HT7 acupuncture evoked an increase in the action potentials of mPFC neurons. Optopatch experiment proved glutamatergic projections from mPFC to LHb. HT7 acupuncture suppressed NAc dopamine release following cocaine injection, which was blocked by electrolytic lesion of mPFC. These results suggest the mediation of mPFC‐LHb circuit in the inhibitory effects of acupuncture on cocaine psychomotor activity in rats.


| INTRODUCTION
A 1997 National Institutes of Health consensus statement concluded that acupuncture can relieve certain conditions including drug abuse, stroke rehabilitation and headache. 1Over the past 40 years, there has been a growing interest in acupuncture treatment of substance abuse around the world including cocaine addiction, and currently, over 700 addiction recovery centres in the United States of America (USA) use acupuncture as an adjunctive procedure. 2There is cumulative evidence supporting the effects of acupuncture on drug addiction, and the underlying mechanisms are being elucidated.Our prior studies have demonstrated that stimulation of a specific acupoint Heart 7 (HT7 or Shenmen) over the ulnar nerve suppresses the hyper-locomotor activity induced by cocaine administration and that these effects are mediated by activation of A-fibres at the ulnar nerve trunk and the dorsal columnmedial lemniscus pathway. 3,46][7] However, the central pathway between peripheral acupuncture signals and the mesolimbic reward circuit remains underexplored.
10] Actions of cocaine and other drugs of abuse are known to be closely associated with mPFC. 11The mPFC critically contributes to vulnerability to cocaine relapse provoked by stressful events during periods of abstinence. 12In particular, inhibition of dorsal mPFC (also called prelimbic [PL] cortex) in models of cocaine dependence results in increased punishment-resistant drug seeking. 13The habenula is a complex anatomical region nestled in the epithalamus that is comprised of the medial habenula and lateral habenula (LHb).In particular, the LHb receives aversion-related information from hypothalamic, limbic and basal ganglia brain sites and encodes negative reward prediction errors that modulate decision-making processes. 14,15The mPFC sends projections to many of the same midbrain and hindbrain regions involved in reward seeking to which LHb neurons project; in turn, the LHb glutamatergic neurons innervate the VTA and PFC. 16,17The fasciculus retroflexus (FR) is the primary efferent pathway connecting the LHb to the midbrain and governs the release of glutamate onto the rostromedial tegmental nucleus (RMTg) GABA neurons.The GABA released from RMTg neurons inhibits DA neurons in the VTA.Recent report suggests that the robust inhibition of dopaminergic cells evoked by the LHb is unlikely to arise from a selective innervation of VTA GABAergic neurons. 18Moreover, the LHb is suggested to mediate a direct excitation of dopaminergic cells that is overridden by indirect inhibition originating from an extrinsic source. 17Furthermore, our previous study has shown that stimulation of the HT7 acupoint near the ulnar nerve activates LHb-RMTg circuits and suppresses acute cocaine-induced psychomotor responses. 19sed on these observations, we hypothesized that acupuncture at HT7 would activate the mPFC-LHb pathway and thus reduce druginduced psychomotor activity.To this end, we examined whether

| Acupuncture treatment
As previously described, acupuncture was performed using a mechanical acupuncture instrument (MAI) that was developed by our laboratory. 3Briefly, while an assistant lightly restrained the rat without anaesthesia, needles (0.10 mm in diameter, 10 mm in length of needle; Dongbang Medical Co., Boryeong, Korea) were bilaterally inserted 3 mm deep into the HT7 or Large Intestine 5 (LI5) acupoint, located on the transverse crease of the wrist of the forepaw and stimulated by our MAI that consisted of a custom-made control unit and a mechanical vibrator mated to a needle.The inserted needles were mechanically stimulated for 20 s and 1.3 m/s 2 in intensity, maintained up to 1 min after needle insertion and subsequently withdrawn.
Acupoint HT7 (called Shenmen) is an important acupoint of Heart (HT) meridian, most frequently prescribed for psychological disorders. 5To assess the possibility that the noxious mechanical stimulation impaired locomotor behaviours, LI5, located on the radial side of wrist joint, was mechanically stimulated as a control.

| Cocaine-induced locomotor activity
Locomotor activity was measured by an image analysis system (Ethovision 3.1, Noldus Information Technology BV, Wageningen, Netherlands), as previously described. 3,4Briefly, each animal was placed into a square open field box (40 cm Â 40 cm Â 45 cm) and monitored with an overhead video camera and video tracking software (Ethovision 3.1).After recording baseline activity for 30 min, the animal was given an intraperitoneal injection of cocaine (15 mg/kg) and monitored up to 60 min after injection.The distance travelled during each 10-min period was analysed.
The skull bones were retracted to drill holes bilaterally over mPFC (stereotaxic coordinates: anterior, +3.0 mm; lateral, ±0.7-0.8 mm; deep, À4.5 mm).A 26-gauge Hamilton syringe (Hamilton Company, Reno, NV, USA) filled with ibotenic acid was infused at a rate of 0.5 μL/min using a microinjection pump (Pump 22, Harvard Apparatus, South Natick, MA, USA).The syringe was left in place for at least 5 min to facilitate diffusion after injection.The sham group received the saline into the mPFC instead of ibotenic acid.At the end of the experiments, the rats were sacrificed for histological confirmation of lesion sites.In in vivo fast-scan cyclic voltammetry (FSCV) for DA detection, mPFC lesions were made immediately before experiments by electrolytic method, as performed in our laboratory (Kim et al, 2021).In brief, under urethane anaesthesia (1.5 g/kg, i.p.) a tungsten electrode insulated except at 0.5 mm tip was inserted in the mPFC (stereotaxic coordinates: anterior, +3.0 mm; lateral, ±0.7 mm; deep, À4.5 mm), and the lesions were made by passing ±0.35 mA of DC current for 8 s.

| Optogenetic inhibition of mPFC-LHb circuit
Optogenetic silencing of LHb was accomplished as described in our previous studies. 19,20After placing the animal into a stereotaxic frame under pentobarbital anaesthesia (50 mg/kg, i.p.), a total volume of 1 μL of the viral construct NpHR was administered bilaterally into the mPFC using a 26-gauge Hamilton syringe.The viral vectors were infused at a rate of 0.25 μL/min using a microinjection pump (Pump 22, Harvard Apparatus, South Natick, MA, USA).The syringe was left in place for at least 5 min to facilitate diffusion after injection.Two weeks after the viral injection, the rats were bilaterally implanted with optic fibres directly over the LHb for the optogenetic inhibition at the following coordinates: 10 angle; anterior, À3.4 mm; lateral, ±1.5 mm; deep, À5.5 mm.The rats were then allowed to recover for a week prior to experiments.For optogenetic inhibition of mPFC-LHb projections, mating ferrules were connected to the implanted optic fibre and illuminated (590 nm, constant 0.5 mW yellow light) by using an optogenetic system (Doric Lenses, Quebec, QC, Canada).At the end of the experiments, the rats were sacrificed for histological confirmation of injection sites or viral expression.

| In vivo extracellular recording of the mPFC neurons
Single-unit discharges of mPFC neurons during HT7 stimulation were recorded as performed in our laboratory. 21Briefly, under urethane anaesthesia (1.5 g/kg, i.p.) the animal was positioned on a stereotaxic apparatus, and burr holes were made over the skull to accommodate the recording electrode.Body temperature was kept constant at 37.5 C using a feedback-controlled direct-current heating pad.For recording of mPFC neurons, a carbon-filament glass microelectrode (0.4-12 MW, Carbostar-1, Kation Scientific, Minneapolis, MN, USA) was stereotaxically inserted into the mPFC.Spontaneous discharges were amplified and filtered at 0.1-10 kHz (ISO-80; World Precision Instruments, Sarasota, FL, USA).Single-unit activity was isolated, recorded and analysed via a CED 1401 Micro3 device and Spike2 software (Cambridge Electronic Design, Cambridge, UK).After recording stable baseline, the rats were given acupuncture stimulation at HT7 for 1 min.The mean firing rates for each 1-min period before, during and after acupuncture stimulation were analysed.

| In vivo FSCV for DA detection
Electrically evoked-DA release in the NAc was carried out as performed in our previous study. 22In brief, a triangular cyclic sweep (À4.0 to +1.3 V versus Ag/AgCl at 400 V/s scan rate) was applied to a custom-made carbon fibre electrode (CFE) at every 100 ms by ChemClamp voltage clamp amplifier (Dagan Corporation, Minneapolis, MN, USA).Recording and analysing were performed using a LabVIEW-based (National Instruments, Austin, TX, USA) customized Demon voltammetry software.Under urethane anaesthesia (1.5 g/kg, i.p.), bipolar stainless-steel electrode and CFE were stereotaxically inserted into the medial forebrain bundle (MFB; stereotaxic coordinate: posterior, À2.5 mm; lateral, 1.9 mm; ventral, 8.0-8.5 mm) and the NAc shell (stereotaxic coordinate: posterior, 1.6 mm; lateral, 1.9 mm; ventral, 6.5 mm).The MFB was stimulated with 60-Hz monophasic pulses at 2 min intervals.After a stable baseline was established (<10% in peak height variation of three consecutive collections), the changes of the NAc DA release following cocaine injection (15 mg/kg, i.p.) were monitored for a further 30 min.

| Histological examination of lesions
At the termination of experiments, all rats were sacrificed for histological confirmation of lesions.Animals were perfused with phosphatebuffered saline (PBS) and then with 4% paraformaldehyde.Brains were removed, post-fixed in 4% paraformaldehyde and cryoprotected in 30% sucrose.The tissue was then cryosectioned into 30 μm thick and stained with toluidine blue.Only those rats with correctly placed lesions were included for data analysis.

| Data analysis
All data are presented as mean ± SEM (standard error of the mean) and analysed by one-or two-way repeated measurement analysis of variance (ANOVA) followed by post hoc testing using the Tukey method or paired t-test, where appropriate.Statistical significance was considered at p < 0.05.HT7 or LI5 and mechanically stimulated with our MAI (Figure 1A).

| Acupuncture at HT7 activates the PL-mPFC neurons
To examine whether acupuncture stimulation activates PL-mPFC neurons, we performed single-unit extracellular recordings for activity of PL-mPFC neurons following HT7 acupuncture in anaesthetised rats (Figure 2A,B).Figure 2B shows histograms representing extracellular recordings of PL-mPFC neurons during HT7 acupuncture.The firing rates of PL-mPFC neurons (n = 16) significantly increased to 208.10 ± 22.89% from baseline during acupuncture stimulation at HT7 and quickly returned to baseline after acupuncture stimulation (one-way ANOVA; F (2,45) = 13.298,*p < 0.001; Figure 2C,D).These results showed the excitation of PL-mPFC neurons by HT7 acupuncture.

| mPFC lesion prevents the effects of HT7 acupuncture on acute cocaine-induced NAc DA release
Finally, to explore the mediation of the PL-mPFC in the acupuncture inhibition of cocaine-induced NAc DA release, in vivo FSCV was performed in the rats with mPFC lesions (Figure 5A).Cocaine injection rapidly increased NAc DA release, which lasted up to 30 min, compared with the value before injection, which was significantly suppressed by acupuncture at HT7.The HT7 acupuncture failed to reduce cocaineenhanced DA release in the mPFC-lesioned rats (PFC X + HT7; twoway repeated ANOVA: group factor F (2,14) = 6.9991, p = 0.008; time factor F (17,53) = 15.117,p < 0.001; interaction F (34,238) = 1.760, p = 0.008; HT7 vs. PFC X + HT7, *p < 0.05; Figure 5B-D).The mPFC is thought to play a prominent role in cocaine addiction. 24A number of studies have suggested that activation of PL-mPFC suppresses cocaine addiction-like behaviours, 13,24 although a few studies have shown opposite results according to the methods of cocaine cessation in rats (i.e., cocaine abstinence and withdrawal). 24en et al reported that in the rodent model of cocaine selfadministration, pharmacological or optogenetic activation of mPFC inhibited cocaine seeking under certain conditions and optogenetic inhibition of mPFC increased cocaine seeking. 13It has been reported that the PL-mPFC reveals hypofunction in the loss of inhibitory control over drug seeking. 23,25Prolonged cocaine self-administration decreases intrinsic excitability of the PL neurons, which is more pronounced in compulsive drug-seeking animals. 26Furthermore, compensating for hypoactive PL neurons with in vivo PL cortex stimulation prevents compulsive cocaine seeking; whereas, optogenetic PL inhibition significantly increased compulsive cocaine seeking. 13In the present study, HT7 acupuncture activated PL-mPFC neurons and suppressed cocaine-enhanced locomotion and NAc DA release, which was blocked by artificial lesions of PL-mPFC.Thus, we suggest that activation of PL-mPFC by acupuncture suppresses cocaine-enhanced psychomotor response.
The habenula is a complex anatomical region nestled in the epithalamus that is comprised of the medial and LHb.It is involved in processing of reward and aversive information, cognitive flexibility and emotion. 27,28The LHb receives inputs from several forebrain limbic circuits involved in motivation and value processing and sends outputs to brainstem centres involved in prediction of aversive events and behavioural inhibition. 29The LHb sends glutamatergic inputs to RMTg GABA neurons via the primary efferent pathway of the FR.The GABA released from RMTg neurons in response to LHb activation inhibits DA neurons in the VTA.Through this process, activation of the LHb strongly inhibits VTA DA neurons and suppresses addiction behaviours. 18The mPFC and the LHb may interact to regulate cocaine addiction-like behaviours.Mathis et al reported that mPFC neurons projecting to the LHb regulate the resumption of extinguished cocaine seeking in mice, as measured using the intravenous cocaine self-administration and cocaineconditioned place preference procedures. 30In the present study, from mPFC evoked EPSC in PL-mPFC-projecting LHb neurons that was sensitive to D-AP5 and CNQX.In our previous study, electrolytic inactivation of LHb inhibited cocaine-induced locomotion. 4Taken together, we suggest that acupuncture suppresses cocaine-enhanced psychomotor response by enhancing glutaminergic transmission from PL-mPFC to LHb.Furthermore, the present study showed that acupuncture at HT7 reduced NAc DA release following acute cocaine injection, which was blocked by the PL-mPFC lesions.LHb sends excitatory inputs onto RMTg neurons and its activation promotes behavioural avoidance. 31r previous study showed that LHb neurons projecting to RMTg are activated by the stimulation of ulnar nerve near HT7 acupoint, and acupuncture activates LHb-RMTg circuits leading to inhibition of VTA DA neurons and thus suppression of cocaine enhancement of locomotor activity. 19Thus, we suggest that acupuncture suppresses cocaineinduced behaviours by inhibiting VTA DA neurons via activation of mPFC-LHb-RMTg circuits.
Although our previous and present studies show that acupuncture at HT7 can activate LHb, it is not clear yet how peripheral input such as acupuncture enters the mPFC.Our previous studies revealed that acupuncture at HT7 activates peripheral sensory afferents, such as Pacinian and Meissner corpuscles, which are conveyed via large A-fibres within the trunk of the ulnar nerve 3 and the dorsal-column medial lemniscal pathway and inhibits the cocaine locomotion and seeking behaviours. 4,32r previous study also revealed that lesions of the ventral posterolateral nucleus (VPL) abolish activation of LHb neurons by ulnar nerve stimulation near HT7 acupoint, suggesting a putative VPL-LHb link. 4 Others have shown that the VPL projects to the prefrontal cortex, which sends excitatory inputs to the LHb. 29,33Therefore, it may be possible that peripheral inputs during acupuncture stimulation at HT7 are conveyed via ulnar nerve and the dorsal column medial lemniscal path-

( 1 )
acupuncture at HT7 reduces acute cocaine-induced locomotor activity and the acupuncture effects are blocked by inhibition of the mPFC-LHb circuit; (2) acupuncture excites mPFC neurons, and (3) acupuncture suppresses acute cocaine-induced DA release in NAc via the mPFC-LHb circuits.2 | METHODS AND MATERIALS 2.1 | Animals Male Sprague-Dawley rats (Daehan Animal, Seoul, Korea) weighing 270-350 g were used.All rats had free access to food and water and were maintained on a 12 h light-dark cycle in a temperaturecontrolled room (20-24 C).All procedures were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee (IACUC) at Daegu Haany University (DHU 2022-12) and Yonsei University (# 2023-0006).Each group consisted of six to eight rats, unless stated otherwise.

3 | RESULTS 3 . 1 |
Chemical lesion of the mPFC ablates the inhibitory effect of HT7 acupuncture on acute cocaine-induced locomotor activity First, we examined whether acupuncture could suppress acute cocaine-induced locomotor activity.For acupuncture groups, acupuncture needles were inserted at a depth of 3 mm into acupoints F I G U R E 1 Effect of prelimbic (PL)-medial prefrontal cortex (mPFC) chemical lesions on acupuncture inhibition of acute cocaine-induced locomotor activity.(A-C) Effects of acupuncture on acute cocaine-induced locomotor activity.(A) The acupuncture points Heart 7 (HT7) or Large Intestine 5 (LI5) were stimulated on both sides by using a mechanical acupuncture instrument (MAI).(B) Effects of acupuncture at HT7 or LI5 on acute cocaine-induced locomotor activity.Mechanical acupuncture at HT7 (Cocaine + HT7, n = 8), but not at LI5 (Cocaine + LI5, n = 8), significantly reduced acute cocaine enhancement of locomotor activity, compared with the control group (Cocaine, n = 8).*p < 0.05 vs. Cocaine.(C) Representative movement traces for 60 min after cocaine or saline injection.(D-F) Effect of PL-mPFC lesion on acupuncture-mediated inhibition of cocaine-induced locomotion.(D) Chemical lesions in the PL-mPFC were stained with toluidine blue.(E) Effect of PL-mPFC lesions on HT7 inhibition of cocaine locomotion.Acupuncture at HT7 reduced acute cocaine-induced locomotor activity (Sham + HT7, n = 6; *p < 0.05, vs. PFC X).However, such effects were not observed in the PL-mPFC-lesioned rats (PFC X; n = 6, PFC X vs. PFC X + HT7, n = 6).(F) Representative traces for 60 min following cocaine injection in PFC (PL)-lesioned or sham-operated rats.ACd, dorsal anterior cingulate; IL, Infralimbic cortex.
Activation of prefrontal cortex (PFC) prelimbic (PL) neurons following acupuncture at Heart 7 (HT7).(A) Schematic for in vivo extracellular recordings of PFC (PL) neurons.(B-D) Effect of acupuncture at HT7 on single-unit activities of PFC (PL) neurons.(B) Recording sites stained with toluidine blue.(C) Representative waveforms and peri-stimulus time histograms of in vivo extracellular recordings in PFC (PL) neurons (n = 16) before (Pre), during (HT7) and after (Post) acupuncture stimulation.(D) Single-unit discharges of PFC (PL) neurons increased significantly during HT7 stimulation, compared with the values before acupuncture (Pre) expressed as percentage of pre-treatment values.ACd, dorsal anterior cingulate; IL, Infralimbic cortex.*p < 0.001 vs. Pre.(Figure 3A,B).AAV2-hSyn-ChR2(E123A)-EYFP expressing fibres were found throughout the LHb (Figure 3B), suggesting that the PL-mPFC sends projections to the LHb.To further verify the glutaminergic projections from the PL-mPFC to the LHb, we examined the expression of vGluT2 and performed Optopatch experiments in LHb neurons.vGluT2 positive boutons (red) were observed in close apposition with the axons from PL-mPFC in LHb (Figure 3C).In the Optopatch experiments, we delivered wide-field flashes of blue light (488 nm, 2 mW cm 2 , 5 ms) and recorded the excitatory postsynaptic currents (EPSC) in the LHb neurons.Light illumination evoked the EPSC in LHb neurons.When perfused for 10 min with D-AP5 (50 μM) and CNQX (20 μM), the light-evoked EPSC was completely blocked (Figure 3D,E), indicating glutamatergic projections from mPFC to LHb.

F
I G U R E 3 Glutamatergic projections from the medial prefrontal cortex (mPFC) to the lateral habenula (LHb).(A) Schematics showing the injection of AAV2-hSyn-ChR2(E123A)-EYFP in the prefrontal cortex (PFC) prelimbic (PL) cortices and projection to the LHb.(B) Green fluorescence shows viral expression in PL and LHb.Bar = 500 μm (C) Expression of vesicular glutamate transporter 2 (vGlut2) (red) and EYFPlabelled mPFC projections (green) in LHb.(D,E) Light-evoked excitatory post-synaptic currents (EPSCs) recorded from LHb neurons.(D) Representative traces of EPSC in response to LED light stimulation and the blocking effect by perfusion with D-2-Amino-5-phosphonovaleric acid (D-AP5) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX).(E) Effects of D-AP5 + CNQX on light-evoked EPSC amplitude of the LHb neurons (n = 10), expressed as percentage of baseline (pre-treatment values).Ten minutes perfusion of D-AP5 and CNQX blocked the lightevoked EPSC.*p < 0.05 vs. baseline.In the present study, acupuncture at HT7 reduced cocaine-enhanced locomotor activity and DA release in NAc, which was blocked by the lesions of PL-mPFC or optogenetic inhibition of the PL-mPFC-LHb circuit.Our immunohistochemistry and Optopatch experiments revealed the mPFC glutamatergic projection to LHb.The PL-mPFC neurons were excited during acupuncture stimulation.These findings suggest that the PL-mPFC-LHb circuit mediates the acupuncture inhibition of cocaine-induced psychomotor responses.
way and reach VPL-mPFC-LHb, which activates RMTg GABA neurons and in turn suppresses DA neurons, thereby leading to inhibition of cocaine-induced psychomotor responses.Acupuncture may activate the VPL-mPFC-LHb-RMTg circuit to reduce cocaine-induced psychomotor response, which remains to be elucidated.In conclusion, acupuncture at HT7 reduced cocaine enhancement of locomotor activity and NAc DA release, which were blocked by lesioning of the PL-mPFC and optogenetic inhibition of PL-mPFC-LHb circuits, suggesting that acupuncture recruits the PL-mPFC-LHb circuit to suppress cocaine-induced psychomotor responses.AUTHOR CONTRIBUTIONSSuchan Chang and Hee Young Kim designed the experiment.Suchan Chang, Hyung Kyu Kim, Yeonhee Ryu, Han Byeol Jang, DanBi Ahn, Dong-ho Youn, Bae Hwan Lee and Bong Hyo Lee performed the experiments and analysed the data.Suchan Chang and Hee Young Kim drafted the manuscript.Hee Young Kim was responsible for the overall direction of the project and for the edits to the manuscript.