Single-dose methamphetamine administration impairs ORM retrieval in mice via excessive DA-mediated inhibition of PrL^Glu activity

Abstract

Methamphetamine (METH), an abused psychostimulant, impairs cognition through prolonged or even single-dose exposure, but animal experiments have shown contradictory effects on memory deficits. In this study we investigated the effects and underlying mechanisms of single-dose METH administration on the retrieval of object recognition memory (ORM) in mice. We showed that single-dose METH administration (2 mg/kg, i.p.) significantly impaired ORM retrieval in mice. Fiber photometry recording in METH-treated mice revealed that the activity of prelimbic cortex glutamatergic neurons (PrL^Glu) was significantly reduced during ORM retrieval. Chemogenetic activation of PrL^Glu or glutamatergic projections from ventral CA1 to PrL (vCA1^Glu-PrL) rescued ORM retrieval impairment. Fiber photometry recording revealed that dopamine (DA) levels in PrL of METH-treated mice were significantly increased, and micro-infusion of the D2 receptor (D2R) antagonist sulpiride (0.25 μg/side) into PrL rescued ORM retrieval impairment. Whole-cell recordings in brain slices containing the PrL revealed that PrL^Glu intrinsic excitability and basal glutamatergic synaptic transmission were significantly reduced in METH-treated mice, and the decrease in intrinsic excitability was reversed by micro-infusion of Sulpiride into PrL in METH-treated mice. Thus, the impaired ORM retrieval caused by single-dose METH administration may be attributed to reduced PrL^Glu activity, possibly due to excessive DA activity on D2R. Selective activation of PrL^Glu or vCA1^Glu-PrL may serve as a potential therapeutic strategy for METH-induced cognitive dysfunction.

Introduction

The abuse of methamphetamine (METH), an illegal psychostimulant, has always been a severe global public health problem [1]. Numerous clinical studies have found that chronic METH administration causes severe psychological and physical disorders, including a diverse range of cognitive deficits [2, 3], and even single-dose METH administration can elicit memory impairment in clinical trial subjects [4, 5]. Notably, animal experiments on memory have shown contradictory results. For instance, separate studies reported that single-dose METH administration impaired or did not affect the recognition memory in mice [6–8], while another study indicated that single-dose METH administration enhanced spatial memory in mice [9]. This contradiction may be due to variations in memory types as well as distinct experimental protocols [10, 11]. Furthermore, different types of memory engage distinct brain regions and neural circuits. For example, hippocampus (CA3, CA1, dentate gyrus (DG)), medial prefrontal cortex (mPFC) and perirhinal cortex (PRh) may be involved in object recognition memory (ORM) [12, 13]. Recent studies have reported that optogenetic silencing of ventral CA1 (vCA1) input to the mPFC disrupted ORM retrieval in rats [14], and chemogenetic activation of central amygdala (CeA) input to the locus coeruleus enhanced ORM in mice [15]. Thus, more research is required to determine the effects of single-dose METH administration on the ORM as well as the key brain regions and neural circuits involved.

It is widely recognized that METH can elevate the extracellular level of dopamine (DA) in the brain by inhibiting the dopamine transporter [16, 17]. Maintaining optimal levels of dopaminergic signaling in the brain is crucial for memory. For example, DA receptor antagonist administration significantly impaired ORM in mice [18–20]; and elevating dopaminergic signaling in the brain via exogenous psychostimulants or overexpression of DA receptors resulted in memory impairment [21, 22]. In addition, DA is significantly associated with neuronal synaptic strength and neuronal excitability, which contribute to memory acquisition, consolidation and retrieval [23–25]. Both exogenous DA and D1 receptor (D1R) agonists were found to reduce the release of presynaptic transmitters from mPFC pyramidal neurons [26, 27]. Similarly, low doses of DA increased mPFC pyramidal neuron excitability while high doses of DA decreased the excitability [28]. It has been reported that chronic administration of METH significantly reduced the release of presynaptic glutamate from mPFC pyramidal neurons [29] accompanied by impairment of ORM in mice [6, 30]. However, it remains unclear how single-dose METH modulates dopaminergic signaling in key brain regions, whether it further affects neuronal excitability and synaptic strength, and whether it ultimately affects the ORM of mice.

In the present study, using the NOR test combined with immunofluorescence, fiber photometry, chemogenetic manipulation, pharmacological methods, electrophysiology and retrograde tracing, we found that ORM retrieval was impaired following single-dose METH administration. Importantly, the impairment of ORM retrieval may be attributed to the reduced activity of prelimbic cortex glutamatergic neurons (PrL^Glu), possibly through excessive DA-induced D2 receptor (D2R) overactivation. Selective activation of PrL^Glu and glutamatergic projections from the vCA1 to the PrL (vCA1^Glu-PrL) rescued the ORM retrieval impairment, which may serve as a potential therapeutic strategy for METH-induced cognitive impairment.

Materials and methods

Animals

Eight-week-old male C57BL/6 mice were provided by the Experimental Animal Center of Shenyang Pharmaceutical University. A total of 5–6 mice were maintained in each cage following a 12-h light-dark cycle with unrestricted availability of food and water. Every attempt was made to minimize the number of animals used and reduce their suffering. Every experimental procedure was carried out in adherence to the Regulations of the Chinese Administration of Affairs Concerning Experimental Animals and approved (Registration nos. SYPU-IACUC-GZR2020-04.13-110) by the Experimental Animal Research Committee of Shenyang Pharmaceutical University (Shenyang, China).

Drug administration

For systemic drug delivery, METH (purity >98%, acquired from Liaoning Institute of Crime Detectives) and Clozapine-N-oxide (CNO) (Sigma-Aldrich, C0832, St. Louis, MO, USA) were dissolved in 0.9% saline. For behavioral assessment or fiber photometry recording, all drugs were delivered intraperitoneally (i.p.). For cannula infusion in freely behaving mice, D1R antagonist SCH23390 (Sigma-Aldrich, D054, St. Louis, MO, USA) and D2R antagonist Sulpiride (MedChemExpress, HY-B1019, Shanghai, China) were dissolved in 0.9% saline.

Stereotaxic surgery

The surgical methods were similar to those described previously [31]. To establish deep anesthesia, mice were given 4% isoflurane (w/v), which was then maintained at 1.5%. The mice were then placed on the stereotaxic equipment manufactured by RWD (Shenzhen, China) for skull operations. All adeno-associated viruses (AAVs) were purchased from Braincase (Shenzhen, China).

For chemogenetic manipulation of PrL^Glu and glutamatergic neurons in dorsal dentate gyrus (dDG^Glu), 200 nL of AAV-CaMKIIα-EGFP, AAV-CaMKIIα-hM3Dq-EGFP or AAV-CaMKIIα-hM4Di-EGFP was injected into the PrL (AP, 1.9 mm; ML, ±0.3 mm; DV, 2.4 mm) or dDG (AP −1.9 mm; ML ± 1.0 mm; DV 2.2 mm) of mice bilaterally. To activate specific neural circuits, 300 nL of Retro-AAV-CaMKIIα-Cre was bilaterally injected into the PrL (AP, 1.9 mm; ML, ±0.3 mm; DV, 2.4 mm), and 300 nL of AAV-hSyn-DIO-hM3Dq-EGFP was bilaterally injected into the vCA1 (AP, −3.16 mm; ML, ±3.2 mm; DV, 3.5 mm) or the basal lateral amygdala (BLA) (AP, −1.22 mm; ML, ±3.0 mm; DV, 4.5 mm). Three weeks subsequent to the virus injection, the novel object recognition test was performed.

For fiber photometry recordings, 300 nL of AAV-CaMKIIα-GCaMP6s or AAV-hSyn-DA2m was injected unilaterally into the PrL (AP, 1.9 mm; ML, 0.3 mm; DV, 2.4 mm) of mice. Immediately after the AAV injection, the optical fiber [200 μm in diameter, 0.37 numerical aperture (NA), Hangzhou Inper, China] was implanted 0.15–0.2 mm above the viral injection sites. After that, dental cement and skull screws were applied to fix the optical fibers for further experiments. Three weeks subsequent to the virus injection, the novel object recognition test or dopamine release recording was performed.

For retrograde tracing of the PrL glutamatergic axons, 200 nL of AAV11-CaMKIIα-EGFP was injected unilaterally into the PrL (AP, 1.9 mm; ML, 0.3 mm; DV, 2.4 mm) of mice. Three weeks subsequent to the virus injection, the mice were euthanized and their brain tissues were processed for imaging.

For specific electrophysiology recording, 200 nL of AAV-CaMKIIα-EGFP were injected bilaterally into the PrL (AP, 1.9 mm; ML, ±0.3 mm; DV, 2.4 mm) of mice. Three weeks subsequent to the virus injection, mice were perfused for electrophysiology recording following the completion of the novel object recognition test.

For intra-PrL injection of antagonist, bilateral cannulas (RWD Life Science Co., Ltd, Shenzhen, China) were implanted into the PrL (AP, 1.9 mm; ML, ±0.3 mm; DV, 2.4 mm) and fixed with dental cement and skull screws. Three weeks subsequent to the cannula implantation, the novel object recognition test was performed.

In all surgeries, the AAV was injected via a micro-syringe (Hamilton, USA), and the antagonist was injected via the bilateral cannulas which were attached to a micro-syringe (Hamilton, USA) via polyethylene tubing. The rate of the micro-injection pump (KD Scientific, USA) was programmed to 200 nL/min, and the injector remained stationary for an additional 5 min following the injection to facilitate the diffusion of viral particles or antagonists.

Novel object recognition (NOR) test

The NOR task was conducted in a plexiglass open-field arena (50 cm× 50 cm× 60 cm). Mice were handled and placed in the arena facing one of the corners. They were then permitted to investigate the arena freely, object-free, for 30 min/d for 3 d (habituation). One day after the last habituation session, two novel objects (objects A and B) were placed symmetrically in the arena. Following this, mice were permitted unrestricted access to the objects for 5 min (training). In this session, mice acquired memories about these two objects. After different time intervals, mice were re-introduced into the arena containing the familiar object A, located in its previous position, together with a novel object C, placed where object B was previously located (test). The test session lasted 5 min. Objects were made of wood and had no innate appeal to mice. The open-field arena and the objects were cleaned with ethanol (75%) before each trial to ensure that any odor did not affect other mice. The behaviors of mice were automatically tracked using EthoVision XT 8.0 (Noldus, Wageningen, Netherlands). The preference index was calculated as: time exploring novel object C/total object exploration time × 100% [32]. Mice were excluded from data analyses if they did not explore one or both of the objects.

To investigate the effects of single-dose METH administration on ORM retrieval [14], mice experienced habituation during days 1–3. On day 4, training was carried out. After 1 h or 2 d intervals, the ORM test was performed. Thirty minutes before the test, mice were injected with saline or METH (1, 2, 4 mg/kg, i.p.). To investigate the effects of chemogenetic activation of specific brain regions (PrL^Glu and dDG^Glu) or neural circuits (vCA1^Glu-PrL and BLA^Glu-PrL) on ORM retrieval after injury by single-dose METH administration, mice were injected with saline or CNO (1 mg/kg, i.p.), and simultaneously all the mice were injected with METH (2 mg/kg, i.p.), 30 min before the test. Other procedures were carried out as described above. To investigate the effects of chemogenetic inhibition of PrL^Glu or dDG^Glu on ORM retrieval, normal mice were injected with saline or CNO (1 mg/kg, i.p.) 30 min before the test. Other procedures were carried out as described above. To investigate the effects of D1R antagonist SCH23390 on ORM retrieval after injury by single-dose METH (2 mg/kg, i.p.) administration, mice received intra-PrL administration of saline or SCH23390 (0.05, 0.1, 0.2 μg/side), and simultaneously all the mice were injected with METH (2 mg/kg, i.p.), 30 min before the test. Other procedures were carried out as described above. To investigate the effects of D2R antagonist Sulpiride on ORM retrieval after injury by single-dose METH (2 mg/kg, i.p.) administration, mice received intra-PrL administration of saline or Sulpiride (0.0625, 0.25, 1 μg/side), and simultaneously all the mice were injected with METH (2 mg/kg, i.p.), 30 min before the test. Other procedures were carried out as described above.

To investigate the duration of memory maintenance in this training protocol, mice experienced habituation during days 1–3. On day 4, training was performed. After 1 h, 1 d, 2 d, 3 d or 4 d intervals, the mice were subjected to the ORM test without any treatment.

Conditioned place preference (CPP) test

The CPP apparatus was comprised of three rectangular-based compartments, each divided by guillotine doors, all made of plexiglass. Two main chambers (20 cm× 15 cm× 15 cm) were distinguished by different wall patterns (alternating vertical or horizontal patterns of black and white) and floor patterns (smooth and coarse). In brief, mice were handled and given unrestricted access to the CPP apparatus for 15 min on day 1 and 2 (Habituation). On day 3, the pre-CPP test was performed. As in the habituation phase, after 15 min of unrestricted access to the CPP apparatus, the duration of time that the mice spent in the two main chambers was recorded (Pre-CPP). Mice that spent over 66% of their entire time in one chamber were identified and eliminated. To investigate whether single-dose METH (1, 2, 4 mg/kg, i.p.) administration induces reward memory in mice, the mice were injected with METH or saline on day 4 and confined in the drug-paired chamber for 1 h. In this phase, fifty percent of the mice in each group were assigned to the chamber equipped with horizontal stripes, whereas the remaining 50% were assigned to the chamber equipped with vertical stripes. On day 5, mice were administered with saline and placed in the opposite chamber for a duration of 1 h (Conditioning). On day 6, the mice were given unrestricted access to the entire apparatus for 15 min in order to measure the amount of time they spent in the chambers (Post-CPP). To prove that three doses of METH (2 mg/kg, i.p.) can induce reward memory, mice were allowed three cycles of conditioning and the other processes were consistent with those described above. Throughout the CPP procedure, the mice were captured on video using a camera, and an analysis was conducted on the behavior of every mouse using the EthoVision XT 8.0 software (Noldus, Wageningen, Netherlands). The CPP scores were presented as Post-Pre values, which were calculated as [(Post-CPP value) − (Pre-CPP value)] to indicate the METH-associated reward memory [33, 34].

HE staining

Following the NOR task, mice were anesthetized by administration of sodium pentobarbital (100 mg/kg, i.p.), and then a 4% paraformaldehyde perfusion was performed. The brains were immediately separated from the skull and submerged in 4% paraformaldehyde for 2 days. Then, the brains were fixed with paraffin and sectioned into 3-μm coronal slices using a microtome (Leica, RM2235, Wetzlar, Germany). The slices were deparaffinized with xylene, rehydrated with an ethanol gradient, and stained using the HE stain kit (Solarbio, G1120, Beijing, China) in proper order. The morphological changes of the hippocampus and prefrontal cortex were visualized using a microscope (Nikon, Tokyo, Japan) at 100× and 400× magnification [35].

Immunofluorescence

Following the NOR task, mice were perfused as described for HE staining. Then, using a vibratome (Leica, VT1200S, Wetzlar, Germany), the brains were sectioned into 40-m coronal sections. For immunofluorescence, the sections were rinsed with 0.01 M PBS for 3 × 10 min, then the antigen was re-exposed by incubating brain sections that were free-floating in solution (ZSGB, ZLI-9065, Beijing, China) at 80 °C for 30 min. Afterwards, the sections were rinsed with 0.01 M PBS for 3 × 10 min, then incubated with 5% PBS-pig serum (Solarbio, S9060, Beijing, China) containing 0.3% Triton X-100 for 1 h. Next, the sections were subjected to an overnight incubation at 4 °C with primary antibodies. The following primary antibodies were used: c-Fos (Cell Signaling Technology, 2250S, 1:500) and CaMKIIα (Cell Signaling Technology, 50049 S, 1:200). The next day, after rinsing with PBS-0.3% Triton X-100 for 3 × 10 min, the sections were incubated with a goat anti-rabbit IgG antibody conjugated with Alexa 488 (Abcam, ab150077 1:500), a donkey anti-rabbit IgG with Alexa 555 (Abcam, ab150074 1:500) or a donkey anti-mouse IgG with Alexa 647 (Abcam, ab150107 1:500) for 1 h at room temperature. For nuclear staining, the sections were incubated with DAPI (Beyotime, C1006) following the secondary antibodies. Following this procedure, the sections were rinsed with PBS-0.3% Triton X-100 for 3 × 10 min. Finally, the sections were mounted using anti-fading medium (Beyotime, P0128M) and subjected to imaging by a confocal microscope (A1R, Nikon, Tokyo, Japan). To estimate the volume of a specific brain region, the boundaries were first drawn according to the Allen brain atlas and then the positive cells were manually counted with ImageJ. In each analysis, three sections were selected at random from each animal. The density of the positive cells was calculated using the formula: Density = n/[Area (μm²) × Thickness (μm)/10⁹], where n is the number of counted cells [34, 36].

Fiber photometry recording

Three weeks following recovery, fiber photometry recordings were performed on the mice. Calcium signals and DA release signals were obtained with a fiber photometry system with excitation lasers operating at 470 and 410 nm (Inper, Hangzhou, China) [31, 37]. To record the calcium signals of freely moving mice during the novel object recognition test, 470-nm laser was employed to induce fluorescence from the genetically encoded Ca sensor GCaMP6s. The 410-nm laser was employed to regulate bleaching and movement. The calcium signals and object recognition behavior of mice were simultaneously recorded using the software Inper Studio (Inper, Hangzhou, China). The 470- and 410-nm signals were independently processed and normalized to baseline signals to determine delta F/F, where delta F/F = (F–F₀)/F₀. For each trial, the object interactions were delineated by the time at which the mice detected the object with their noses touching them and were manually added as markers. The markers were defined as “0 s”, that is, the onset of the event. The averaged baseline signals between −3 s to −2 s (2–3 s before the object interaction) were considered as “F₀”. “F” represents the corrected 470-nm signal, which was obtained by subtracting the 410-nm trace from the 470-nm trace. Data were analyzed using the software Inper Plot (Inper, Hangzhou, China). The area under the curve (AUC) during the interaction time window (3 s prior to and 3 s subsequent to the object interaction) was quantified and compared for analysis of significant differences. Mice with incorrect optical fiber location were excluded from analysis. Before each recording, mice were habituated to the fiber for 5 min in their home cage.

To record the dopamine release signals of freely moving mice in their home cage, 470-nm laser was used to induce fluorescence from the genetically encoded DA sensor (DA2m). The time of METH administration (2 mg/kg, i.p.) was defined as “0 min”. The averaged baseline signals between −2 min and 0 min were considered as “F₀”. Other parameters are consistent with the description above.

Brain slice electrophysiology

After three weeks of recovery, mice underwent the novel object recognition test. Afterwards, they were anesthetized with sodium pentobarbital (100 mg/kg, i.p.) and perfused transcardially with ice-cold oxygen-saturated (95% O₂/5% CO₂) NMDG-based artificial cerebrospinal fluid (NMDG-ACSF) solution containing (in mM): 93 NMDG, 2.5 KCl, 1.2 NaH₂PO₄·H₂O, 10 MgSO₄·7H₂O, 30 NaHCO₃, 25 glucose, 20 HEPES, 3 Na-pyruvate, 5 Na-ascorbate, 2 thiourea, 12 NAC and 0.5 CaCl₂·2H₂O, adjusted to pH 7.35 with HCl. After perfusion, the brain was promptly dissected and immediately immersed in the ice-cold oxygen-saturated NMDG ACSF solution. The brains were sectioned into 300-μm coronal slices in the same NMDG-ACSF buffer solution with a vibratome (Leica, VT1200S, Wetzlar, Germany). The brain sections containing the PrL were collected and incubated in oxygen-saturated NMDG-ACSF at 32–33 °C for 15 min, then transferred to a normal ACSF solution containing (in mM): 126 NaCl, 2.5 KCl, 1.25 NaH₂PO₄·H₂O, 2 MgSO₄·7H₂O, 25 NaHCO₃, 10 glucose and 2 CaCl₂·2H₂O, pH 7.35 at room temperature bubbled with 95% O₂ and 5% CO₂. They were incubated for at least 1 h for the subsequent electrophysiology recording. The osmolarity of all solutions was maintained at 280–300 mOsm. The chemicals used for slice preparation were purchased from Sigma-Aldrich (St. Louis, MO, USA) [38].

For whole-cell patch-clamp recordings, PrL sections were transferred to the recording chamber perfused with the ACSF solution described above at a rate of 2 mL/min at room temperature. For PrL glutamatergic neuron-specific recording, neurons were identified with a 40× water-immersion objective (Nikon, Tokyo, Japan) fitted on a fluorescence microscope (Nikon, FN1, Tokyo, Japan). PrL glutamatergic neurons expressed EGFP fluorescence. Then, target neurons were visualized using infrared video microscopy and differential interference contrast optics (IR-DIC). The recording pipettes (4–6 MΩ) were prepared with a micropipette puller (Sutter Instrument, P-97, USA).

To record action potentials (AP), the pipettes were filled with intracellular solution containing (in mM): 140 K-gluconate, 10 HEPES, 2.5 MgCl₂, 5 KCl, 4 Mg-ATP, 0.4 Na-GTP, 10 Na Phosphocreatine and 0.6 EGTA, adjusted to pH 7.35 with KOH, 290 mOsm. Neurons were held at around –70 mV. In whole-cell current clamp mode, a current-step protocol (from −50 to +200 pA, with 25 pA increment and 500 ms duration) was run and repeated for AP induction. The threshold, amplitude, half-width and decay time were measured following the first AP spike. The threshold was defined as the first point on the rising phase of the AP spike where the change in voltage surpassed 10 V/s. The amplitude was defined as the difference between the threshold and the peak voltage. Half-width was defined as the time interval between the rising and falling branches when the amplitude is half of the AP amplitude. Decay time was defined as the time interval between 90% of peak voltage and 10% of peak voltage.

To record miniature excitatory postsynaptic currents (mEPSC), ACSF solution was used in the presence of 100 μM picrotoxin and 1 μM tetrodotoxin. The intracellular solution was the same as described above. The neurons were held at −70 mV in voltage clamp mode to record mEPSC for 5 min. The peak amplitude and frequency were analyzed.

Recordings were obtained through a patch-clamp amplifier (HEKA, EPC-10 USB double, Stuttgart, Germany) controlled by the software Patchmaster (HEKA, Stuttgart, Germany). The signals were sampled at 10 kHz and low-pass filtered at 2 kHz. Compensation of the offset potential and pipette capacitive currents was performed using the automatic mode “C-fast” of the amplifier. Whole-cell capacitance was compensated by using the “C-slow” of the amplifier. Series resistance (Rs) was monitored continuously for each neuron. Only neurons with Rs < 30 MΩ were recorded, and recordings were terminated when a significant (>15%) increase in Rs occurred. The raw data were analyzed offline using the software Clampfit 10.2 (Molecular Devices, Union City, CA, USA) and Mini Analysis (Synaptosoft Inc, Fort Lee, NJ, USA).

Statistical analysis

The data are reported as the mean ± SEM, and the statistical analyses were performed using GraphPad Prism 9. Statistical significance was determined using the unpaired t test for comparisons between two groups, while analysis of variance (ANOVA) was used for comparisons among four groups. In addition, the action potential electrophysiological data were analyzed using two-way ANOVA. One-way ANOVA and two-way ANOVA were followed by the Bonferroni post hoc test. For all results, a significance level of P < 0.05 was considered to be statistically significant.

Results

Single-dose METH administration impairs ORM retrieval in mice

Although the evidence showed that single-dose METH administration did not affect the ORM, the effects on the retrieval of ORM remain unclear [6]. It has been demonstrated that METH affects short- and long-term ORM differently [6], probably due to the differences in the neural mechanisms underlying short- and long-term memory [39, 40]. In this study, using the 5-min training protocol, we found that ORM can be maintained for 3 d in mice (Supplementary Fig. 1). Therefore, we specifically examined the potential impact of single-dose METH administration on the retrieval of short- (1 h) and long-term (2 d) ORM in mice. In brief, to study the effects of single-dose METH on ORM retrieval, mice were intraperitoneally injected with saline or METH (1, 2, 4 mg/kg, i.p.) 30 min before the test (Fig. 1a, d). One hour or 2 days following the 5-min training session, the NOR test was performed, and mice were allowed to explore the novel object C and the familiar object A for 5 min. The amount of time spent exploring each object was recorded and the preference index (PI; percentage of total exploration time spent exploring novel object C) was calculated. During the training session, there was no significant difference in the exploration time of objects A and B, demonstrating that mice had no inherent preference for the objects (Fig. 1b, e).

Fig. 1. Single-dose METH administration impairs ORM retrieval in mice.

a, d Experimental timeline and schematic of NOR test to examine the effects of single-dose METH administration on short-term (1 h) and long-term (2 d) ORM retrieval. Saline or METH (1, 2, 4 mg/kg, i.p.) was injected 30 min before the test. b, e Exploration time for objects A and B in memory acquisition (NOR training) and exploration time for objects A and C in the memory test (NOR test). c, f Preference index (PI), calculated as the percentage of time exploring the novel object C over the total time of object exploration in the memory test (NOR test). b, c Administration of METH (2, 4 mg/kg) inhibits short-term ORM retrieval. Compared with object A, the exploration time of object C is significantly increased in the Saline and METH (1 mg/kg) groups, but not in the METH (2, 4 mg/kg) groups (n = 6 in Saline group; n = 8 in METH 1 mg/kg group; n = 9 in METH 2 mg/kg group; n = 8 in METH 4 mg/kg group; unpaired t test, t = 2.360, P = 0.04 for Saline group; t = 2.174, P = 0.05 for METH 1 mg/kg group; t = 0.3132, P = 0.76 for METH 2 mg/kg group; t = 0.6537, P = 0.52 for METH 4 mg/kg group; *P < 0.05 versus object A). PI is significantly decreased in the METH (2, 4 mg/kg) groups (one-way ANOVA, F = 8.009, P < 0.001; *P < 0.05 and **P < 0.01 versus the Saline group). e, f Administration of METH (1, 2, 4 mg/kg) inhibits long-term ORM retrieval. Compared with object A, the exploration time of object C is significantly increased in the Saline group, but not in the METH (1, 2, 4 mg/kg) groups (n = 9 in each group; unpaired t test, t = 2.314, P = 0.03 for Saline group; t = 0.7288, P = 0.48 for METH 1 mg/kg group; t = 0.3913, P = 0.7 for METH 2 mg/kg group; t = 0.3563, P = 0.73 for METH 4 mg/kg group; *P < 0.05 versus object A). PI is significantly decreased in the METH (1, 2, 4 mg/kg) groups (one-way ANOVA, F = 6.088, P = 0.002, **P < 0.01 versus the Saline group). Data are presented as mean ± SEM.

During the short-term ORM retrieval test, the exploration time of the novel object C compared with the familiar object A was significantly increased after administration of saline or METH (1 mg/kg), but not METH (2, 4 mg/kg) (Fig. 1b). Compared with saline, the PI was significantly decreased after administration of METH (2, 4 mg/kg) but not METH (1 mg/kg) (Fig. 1c). These results indicated that METH (2, 4 mg/kg) administration 30 min before the test impaired the retrieval of ORM.

Furthermore, we also examined whether single-dose METH administration affects long-term (2 d) ORM. The findings indicated that METH (1, 2, 4 mg/kg, i.p.) administration impaired long-term ORM retrieval (Fig. 1d–f). These results demonstrated that single-dose METH administration impaired ORM retrieval at both short (1 h) and long-term (2 d) intervals in mice.

Single-dose METH administration inhibits the activation of PrL^Glu during ORM retrieval

METH is not only highly addictive but also neurotoxic, and its properties are dose-dependent [16, 41]. Here, we performed the conditioned place preference (CPP) test and found that single-dose METH (1, 2, 4 mg/kg, i.p.) administration cannot induce reward memory (Supplementary Fig. 2a, b), but three doses of METH (2 mg/kg, i.p.) induced it successfully (Supplementary Fig. 2c, d). Next, HE staining was performed to observe the morphological changes of hippocampus and mPFC, two crucial brain regions related to learning and memory. The results showed that administration of single-dose METH (1, 2, 4 mg/kg, i.p.) did not alter the structure of neurons in those regions (Supplementary Fig. 2e). These findings suggested that single-dose METH (1, 2, 4 mg/kg, i.p.) administration is unlikely to cause obvious reward and neurotoxic effects in mice.

Previous studies suggested that ORM retrieval can increase the expression of the neuronal activity marker c-Fos in specific brain regions [13]. In order to examine the effects of METH on the activation of specific brain regions during ORM retrieval, the mice were sacrificed at 1.5 h after the NOR test (Fig. 2a). During ORM retrieval, the number of c-Fos-positive neurons was significantly increased in dorsal CA1 (dCA1), lateral amygdala (LA), BLA and central amygdala (CeA) in the METH group compared with the Saline group (Fig. 2b). Notably, the number of c-Fos-positive neurons was significantly decreased in the PrL and dDG, indicating that single-dose METH administration reduced the activation of these two brain regions (Fig. 2b). A previous study reported that the reduction of c-Fos expression in a specific brain region induced ORM impairment [42]. Thus, we speculate that PrL and dDG may be involved in ORM retrieval impairment. Next, the neuronal types were examined using the glutamatergic marker CaMKIIα (calcium/calmodulin-dependent protein kinase IIα). The results indicated that the number of c-Fos and CaMKIIα co-labeled neurons was significantly decreased in PrL (Fig. 2c–e) and dDG of mice (Supplementary Fig. 3) after METH (2 mg/kg, i.p.) administration.

Fig. 2. Single-dose METH administration inhibits the activation of PrL^Glu in mice.

a Experimental timeline and schematic of NOR test to examine the effects of single-dose METH administration on the activation of specific brain regions during ORM retrieval. Saline or METH (1, 2, 4 mg/kg, i.p.) was injected 30 min before the test, and the mice were perfused 1.5 h after the test for c-Fos immunofluorescence staining. b Representative images of c-Fos immunofluorescence staining and quantification of the number of c-Fos positive (c-Fos⁺) cells in dDG, dCA2, dCA3, dCA1, ACC, PrL, IL, LA, BLA, CeA, PRh and EC (n = 3, 3 slices per mouse; one-way ANOVA, F = 10.56, P < 0.001 for dDG; F = 0.2617, P = 0.85 for dCA3; F = 0.5458, P = 0.65 for dCA2; F = 8.633, P < 0.001 for dCA1; F = 3.294, P = 0.03 for ACC; F = 8.987, P < 0.001 for PrL; F = 1.952, P = 0.1295 for IL; F = 6.845, P < 0.001 for LA; F = 3.232, P = 0.03 for BLA; F = 7.444, P < 0.001 for CeA; F = 0.9176, P = 0.44 for PRh; F = 1.085, P = 0.36 for EC; *P < 0.05, **P < 0.01, ***P < 0.001 versus the Saline group). Scale bar = 100 μm. c Representative immunofluorescence images of c-Fos⁺/CaMKIIα⁺ cells in PrL from saline and METH (2 mg/kg)-treated mice. White arrows point to c-Fos⁺/CaMKIIα⁺ co-labeled neurons. Scale bar = 100 μm. d, e Quantification of the number of c-Fos⁺ cells and c-Fos⁺/CaMKIIα⁺ cells in PrL. METH (2 mg/kg) administration significantly reduces the abundance of c-Fos⁺ cells and c-Fos⁺/CaMKIIα⁺ cells in PrL (n = 3, 3 slices per mouse; unpaired t test, t = 4.752, P < 0.001 for c-Fos⁺ cells; t = 4.847, P < 0.001 for c-Fos⁺/CaMKIIα⁺ cells; ***P < 0.001 versus the Saline group). f Experimental timeline and schematic of NOR test to verify the calcium responses of PrL^Glu to single-dose METH administration. Three weeks after virus injection and fiber implantation, mice were subjected to the NOR test, and calcium signals were simultaneously recorded using fiber photometry. g Typical image of viral expression and fiber location within PrL. Scale bar = 1 mm. h, j Heatmaps of calcium signals of PrL^Glu during object interactions from mice treated with saline or METH (2 mg/kg). i, k, l Average plots across different trials of individual animals show that object interactions activate PrL^Glu in Saline (gray) and METH (red) groups. m Quantification of the area under the curve (AUC) in (l) (n = 5 in Saline group; n = 6 in METH group; unpaired t test, t = 2.893, P = 0.02, *P < 0.05 versus the Saline group) shows that AUC is significantly reduced in METH (2 mg/kg)-treated mice. Data are presented as mean ± SEM.

To further verify the response of PrL^Glu to single-dose METH administration during ORM retrieval, calcium signals in PrL^Glu were monitored during the NOR test in freely-moving mice. Adeno-associated virus (AAV)-CaMKIIα-GCaMP6s was injected into the PrL of mice and an optical fiber was implanted (Fig. 2f, g). Three weeks later, the mice were subjected to the NOR task, and the calcium signal was recorded at the same time. After saline or METH (2 mg/kg, i.p.) treatment, interaction with the objects during the NOR test triggered PrL^Glu activities in both groups (Fig. 2h–k). However, the intensity of the calcium signal in the METH group was weaker, indicating that the PrL^Glu response during ORM retrieval was decreased in the METH group compared with the Saline group (Fig. 2l, m). These results suggested that the activation of PrL^Glu was reduced following single-dose METH administration during ORM retrieval in mice. In addition, for both the saline- or METH-treated mice, there was no significant difference in the AUC of calcium signals when mice explored objects A and C. These results suggested that there was no significant difference in the activation of PrL^Glu when mice explored these two objects, even after single-dose METH administration (Supplementary Fig. 4).

Activation of PrL^Glu or vCA1^Glu-PrL rescues the ORM retrieval impairment induced by single-dose METH administration

To determine whether PrL^Glu has a role in single-dose METH administration-induced ORM retrieval impairment, we activated PrL^Glu using chemogenetic techniques. Without virus injection, CNO did not affect ORM retrieval (Supplementary Fig. 5). HM3Dq virus or EGFP (Control) virus selectively targeting glutamatergic neurons was infused into the PrL bilaterally. Three weeks later, the mice were subjected to the NOR task. Thirty minutes before the NOR test, mice were simultaneously injected with METH (2 mg/kg, i.p.) and CNO (1 mg/kg, i.p.) to activate PrL^Glu. One and a half hours after the NOR test, mice were sacrificed for immunofluorescence assays (Fig. 3a). The precise location of the viral expression was confirmed by EGFP fluorescence, and the specificity of virus targeting was confirmed by colocalization of EGFP with CaMKIIα (Fig. 3b). The results showed that the number of c-Fos/EGFP co-labeled neurons in PrL was significantly increased in the hM3Dq group compared with the EGFP group, indicating the successful implementation of the DREADD (designer receptors exclusively activated by designer drugs) system (Fig. 3c). The behavioral test results showed that the exploration time of novel object C compared with familiar object A was increased in the hM3Dq group, and PI was increased compared with the EGFP group (Fig. 3d, e). These results indicated that activating PrL^Glu during the NOR test significantly alleviated the ORM retrieval impairment induced by single-dose METH administration. Without CNO, hM3Dq did not affect ORM retrieval (Supplementary Fig. 6a). Additionally, without METH, chemogenetic activation of PrL^Glu did not affect ORM retrieval (Supplementary Fig. 6b). We also found that activating PrL^Glu significantly alleviated the impaired long-term ORM (2 d) retrieval (Supplementary Fig. 7a, b). However, activating dDG^Glu, whose activity was also inhibited by single-dose METH administration, failed to rescue the ORM retrieval impairment (Supplementary Fig. 8a-e). These results suggested that activating PrL^Glu, but not dDG^Glu, can rescue the single-dose METH-induced ORM retrieval impairment in mice.

Fig. 3. Chemogenetic activation of PrL^Glu rescues the single-dose METH-induced ORM retrieval impairment in mice, and inhibition of PrL^Glu induces ORM retrieval impairment in normal mice.

a Experimental timeline and schematic of NOR test to examine the effects of chemogenetic activation of PrL^Glu on single-dose METH-induced ORM retrieval impairment. Three weeks after virus injection, mice were subjected to the NOR test. METH (2 mg/kg, i.p.) and CNO (1 mg/kg, i.p.) were injected 30 min before the test. b Left panel: typical image of viral expression within PrL. Scale bar = 1 mm. Right panel: representative immunofluorescence images of EGFP⁺/CaMKIIα⁺ cells in PrL from EGFP and hM3Dq mice. Scale bar = 100 μm. These images show that EGFP is specifically targeted to the glutamatergic neurons in PrL. c Left panel: representative immunofluorescence images of EGFP⁺/c-Fos⁺ cells in PrL from EGFP and hM3Dq mice. Scale bar = 100 μm. Right panel: Quantification of the number of EGFP⁺/c-Fos⁺ cells in PrL. DREADD significantly increases the abundance of EGFP⁺/c-Fos⁺ cells in PrL (n = 3, 3 slices per mouse; unpaired t test, t = 8.905, P < 0.001, ***P < 0.001 versus the EGFP group). These results indicate that DREADD is successfully implemented and PrL^Glu are successfully activated in the hM3Dq group. d Nose tracking (top) and exploratory behavior heat maps (bottom) during the NOR test of one representative mouse in the EGFP and hM3Dq groups. e Chemogenetic activation of PrL^Glu rescues the single-dose METH-induced ORM retrieval impairment. Left panel: compared with object A, the exploration time of object C is significantly increased in the hM3Dq group, but not in the EGFP group (n = 14 in each group; unpaired t test, t = 0.0518, P = 0.96 for EGFP group; t = 2.479, P = 0.02 for hM3Dq group; *P < 0.05 versus object A). Right panel: PI is significantly increased in the hM3Dq group (unpaired t test, t = 3.739, P < 0.001, ***P < 0.001 versus the EGFP group). f Experimental timeline and schematic of NOR test to examine the effects of chemogenetic inhibition of PrL^Glu on ORM retrieval. Three weeks after virus injection, mice were subjected to the NOR test. CNO (1 mg/kg, i.p.) was injected 30 min before the test. g Left panel: typical image of viral expression within PrL. Scale bar = 1 mm. Right panel: representative immunofluorescence images of EGFP⁺/CaMKIIα⁺ cells in PrL from EGFP and hM4Di mice. Scale bar = 100 μm. These images show that EGFP is specifically targeted to the glutamatergic neurons in PrL. h Left panel: representative immunofluorescence images of EGFP⁺/c-Fos⁺ cells in PrL from EGFP and hM4Di mice. Scale bar = 100 μm. Right panel: quantification of the number of EGFP⁺/c-Fos⁺ cells in PrL. DREADD significantly decreases the abundance of EGFP⁺/c-Fos⁺ cells in PrL (n = 3, 3 slices per mouse; unpaired t test^, t = 7.910, P < 0.001, ***P < 0.001 versus the EGFP group). These results indicate that DREADD is successfully implemented and PrL^Glu are successfully inhibited in the hM4Di group. i Nose tracking (top) and exploratory behavior heat maps (bottom) during the NOR test of one representative mouse in the EGFP and hM4Di groups. j Chemogenetic inhibition of PrL^Glu impairs ORM retrieval. Left panel: compared with object A, the exploration time of object C is significantly increased in the EGFP group, but not in the hM4Di group (n = 11 in each group; unpaired t test, t = 2.514, P = 0.02 for EGFP group; t = 0.6414, P = 0.53 for hM4Di group; *P < 0.05 versus object A). Right panel: PI is significantly decreased in the hM4Di group (unpaired t test, t = 4.146, P < 0.001, versus the EGFP group). Data are presented as mean ± SEM.

Next, in order to further clarify the role of PrL^Glu in ORM retrieval, we examined whether inhibiting PrL^Glu can induce ORM retrieval impairment in normal mice. HM4Di virus or EGFP (Control) virus selectively targeting glutamatergic neurons was infused into the PrL bilaterally. Thirty minutes before each NOR test, mice were injected with CNO (1 mg/kg, i.p.) to inhibit PrL^Glu (Fig. 3f). The precise location of the viral expression was confirmed by EGFP fluorescence, and the specificity of virus targeting was confirmed by colocalization of EGFP with CaMKIIα (Fig. 3g). Immunofluorescence data showed that the number of c-Fos/EGFP co-labeled neurons of PrL was significantly decreased in the hM4Di group compared with the EGFP group, indicating successful implementation of the DREADD system (Fig. 3h). The behavioral test results showed that the exploration time of novel object C compared with familiar object A was not significantly different in the hM4Di group, while PI was significantly decreased compared with the EGFP group (Fig. 3i, j). Without CNO, hM4Di did not affect ORM retrieval (Supplementary Fig. 6c). These results indicated that inhibiting PrL^Glu during the NOR test is able to induce significant ORM retrieval impairment in normal mice. The same effects were also found in long-term ORM (2 d) retrieval (Supplementary Fig. 7c, d). However, inhibiting dDG^Glu failed to induce the ORM retrieval impairment (Supplementary Fig. 8f–i). These results suggested that neural activity of PrL^Glu is essential for ORM retrieval.

We then investigated possible circuit-level connections between PrL neurons and neurons from other upstream brain regions that might mediate single-dose METH administration-induced impairment of ORM retrieval. Firstly, AAV11-CaMKIIα-EGFP, an emerging retrograde viral tracer, was infused into the PrL. EGFP-positive neurons were observed in vCA1, BLA, mediodorsal and intermediodorsal thalamic nucleus (MD and IMD), and retrosplenial cortex (RSC), regions known to provide glutamatergic input to the PrL [43] (Supplementary Fig. 9). Next, retro-AAV-CaMKIIα-Cre was infused into the PrL and AAV-hSyn-DIO-hM3Dq-EGFP was infused specifically into upstream brain regions. Three weeks later, the mice were subjected to the NOR task. The results showed that activating vCA1^Glu-PrL (Supplementary Fig. 10) but not BLA^Glu-PrL (Supplementary Fig. 11) during the NOR test significantly alleviated the ORM retrieval impairment induced by single-dose METH administration. These results suggested that vCA1^Glu-PrL mediates the single-dose METH-induced ORM retrieval impairment in mice.

Single-dose METH administration impairs the ORM retrieval through overactivation of D2R in PrL

Numerous studies have reported that METH increases the DA levels of the central nervous system by inhibiting the DA transporter [16, 17]. To investigate the potential molecular mechanism of the single-dose METH-induced ORM retrieval impairment in mice, we further verified the effect of METH on the dopaminergic system in PrL. Firstly, to monitor in vivo dopamine levels, we performed continuous recording with fiber photometry using a genetically encoded dopamine probe (AAV-hSyn-DA2m) in PrL. The precise location of the fiber and viral expression was confirmed (Fig. 4a). After a stable DA2m signal baseline was acquired, the mice were injected with saline or METH (2 mg/kg, i.p.). Compared with saline, METH administration significantly increased the DA levels in the PrL (Fig. 4b). To identify the specific DA receptors involved, thirty minutes before the NOR test, mice were injected with METH (2 mg/kg, i.p.), and different doses of D1R antagonist (SCH23390) or D2R antagonist (Sulpiride) were infused bilaterally into the PrL (Supplementary Fig. 12a, c; Fig. 4c). The results showed that neither SCH23390 (0.05, 0.1, 0.2 μg/side) nor Sulpiride (0.0625,1 μg/side) could rescue the single-dose METH-induced ORM retrieval impairment (Supplementary Fig. 12b, d). Nevertheless, the exploration time of novel object C compared with familiar object A was increased in the METH + Sulpiride (0.25 μg/side) group, and PI was increased compared with the METH + Saline group (Fig. 4d; Supplementary Fig. 12d). These results indicated that bilateral infusion of Sulpiride (0.25 μg/side) effectively reversed the ORM retrieval impairment. In addition, the exploration time of novel object C compared with familiar object A was not significantly different in the Saline + Sulpiride (0.25 μg/side) group, while PI was decreased compared with the Saline + Saline group (Fig. 4d). These results indicated that bilateral infusion of Sulpiride (0.25 μg/side) into the PrL impaired ORM retrieval in normal mice. Collectively, these results suggested that single-dose METH administration could increase the DA levels of PrL, followed by overactivation of D2R in PrL and finally impairment of ORM retrieval in mice; furthermore, moderate inhibition of D2R may reverse the impairment [44].

Fig. 4. Single-dose METH administration impairs ORM retrieval in mice through overactivation of D2R in PrL.

a Left panel: schematic of fiber photometry recording in PrL. Right panel: typical image of viral expression and fiber location within PrL. Scale bar = 1 mm. b Left panel: DA2m sensor signals in PrL from mice treated with saline (gray line) and METH (red line). Light gray and light red shading indicate SEM. Vertical dotted line represents the timepoint of saline and METH injection. Right panel: quantifications of DA2m sensor signals 20–50 min after saline or METH administration (n = 5 in Saline group; n = 6 in METH group; unpaired t test, t = 2.905, P = 0.02, *P < 0.05 versus the Saline group). Statistical analysis shows that the DA levels of PrL in the METH group are significantly enhanced compared with the Saline group. c Experimental timeline and schematic of NOR test to examine the role of Sulpiride (D2R antagonist) on ORM retrieval in saline- and METH-treated mice. Saline or METH (2 mg/kg, i.p.) was injected 30 min before the test; simultaneously, saline or Sulpiride (0.25 μg/side) was microinfused bilaterally into PrL. d Microinfusions of Sulpiride (0.25 μg/side) into PrL rescue the single-dose METH-induced ORM retrieval impairment, and impair ORM retrieval in normal mice. Left panel: compared to object A, the exploration time of object C is significantly increased in the Saline + Saline group and the METH + Sulpiride group, but there is no significant difference in the Saline + Sulpiride group and the METH + Saline group (n = 10 in Saline + Saline group; n = 9 in Saline + Sulpiride and METH + Sulpiride groups; n = 8 in METH + Saline group; unpaired t test, t = 3.623, P = 0.002 for Saline + Saline group; t = 0.621, P = 0.54 for Saline + Sulpiride; t = 0.085, P = 0.93 for METH + Saline; t = 3.236, P = 0.005 for METH + Sulpiride; **P < 0.01 versus object A). Right panel: PI is significantly decreased in the Saline + Sulpiride and METH + Saline groups compared with the Saline + Saline group. PI is significantly increased in the METH + Sulpiride group compared with the METH + Saline group (two-way ANOVA, F = 9.617, P < 0.001; *P < 0.05 versus METH + Saline group; **P < 0.01 versus Saline + Saline group). Data are presented as mean ± SEM.

Single-dose METH administration inhibits the intrinsic excitability of PrL^Glu mediated by D2R

Studies have shown that DA is strongly correlated with neuronal excitability and synaptic strength [26, 28], which play an important role in the process of memory retrieval [24]. To investigate whether single-dose METH administration affects PrL^Glu function in mice, AAV-CaMKIIα-EGFP was infused into the PrL to label PrL^Glu. Three weeks later, the mice were subjected to the NOR task. Thirty minutes before the NOR test, mice were injected with saline or METH (2 mg/kg, i.p.). After the NOR test, mice were perfused and brains were sliced for electrophysiological recordings, which were performed specifically on the PrL neurons expressing CaMKIIα-EGFP (Fig. 5a, b). To determine whether single-dose METH administration affects the intrinsic excitability of PrL^Glu, whole-cell current-clamp recordings were performed. The results showed that the action potential (AP) frequency of the METH group was significantly lower compared with the Saline group (Fig. 5c, d). The threshold was increased, but there were no significant changes in the amplitude, half-width and decay time (Fig. 5e, f; Supplementary Fig. 13a–d). These results suggested that the intrinsic excitability of PrL^Glu was decreased by single-dose METH administration. We subsequently assessed whether single-dose METH administration affects the basal glutamatergic synaptic transmission by measuring miniature excitatory postsynaptic currents (mEPSC) using whole-cell voltage-clamp recordings. In the METH group, the mEPSC amplitude remained unchanged, but the mEPSC frequency was significantly decreased compared with the Saline group (Fig. 5g–i). These results suggested that PrL^Glu basal glutamatergic synaptic transmission was decreased by single-dose METH administration.

Fig. 5. Single-dose METH administration inhibits PrL^Glu intrinsic excitability and basal glutamatergic synaptic transmission in mice, and inhibition of D2R in PrL rescues the intrinsic excitability of PrL^Glu.

a Experimental timeline and schematic of NOR test to examine the effects of single-dose METH administration on the function of PrL^Glu. Three weeks after virus injection, mice were subjected to the NOR test. Subsequently, the mice were perfused for electrophysiological recording. b Representative immunofluorescence images of EGFP⁺ cells in PrL from saline and METH-treated mice. The white dotted lines represent recording pipettes. Scale bar = 50 μm. c Representative action potential (AP) traces of PrL^Glu from the Saline and METH groups, in response to a series of 500-ms currents stepping from −50 to 200 pA with increments of 25 pA. The red traces represent an input current of 50 pA. Scale bars = 20 mV, 200 ms. d The average number of APs generated in response to depolarizing current pulses (n = 12 cells from 6 Saline group mice; n = 12 cells from 5 METH group mice; two-way ANOVA, F = 2.2, P = 0.03; *P < 0.05 and **P < 0.01 versus the Saline group). e Representative APs of PrL^Glu from the Saline and METH groups. Scale bars = 25 mV, 2 ms. f Statistical analysis shows that the threshold is significantly enhanced in the METH group (unpaired t test, t = 2.163, P = 0.04; *P < 0.05 versus the Saline group). g Representative miniature excitatory postsynaptic current (mEPSC) traces of PrL^Glu from the Saline and METH groups. Scale bars = 10 pA, 5 s. h Quantifications of mEPSC amplitude and cumulative distributions of the mEPSC inter-event intervals. Inset: statistical analysis shows that the mEPSC amplitude of PrL^Glu is not significantly different between the Saline and METH groups. i Quantifications of mEPSC frequency and cumulative distributions of the mEPSC inter-event intervals. Inset: statistical analysis shows that the mEPSC frequency of PrL^Glu is significantly decreased in the METH group (n = 17 cells from 6 mice in each group; unpaired t test, t = 1.078, P = 0.29 for mEPSC amplitude; t = 2.648, P = 0.01 for mEPSC frequency; *P < 0.05 versus the Saline group). j Experimental timeline and schematic of NOR test to examine the effects of D2R antagonist Sulpiride on single-dose METH administration-induced inhibition of intrinsic excitability of PrL^Glu. Three weeks after virus injection and cannula implantation, mice were subjected to the NOR test. Subsequently, the mice were perfused for electrophysiological recording. k The average number of APs generated in response to depolarizing current pulses (n = 8 cells from 3 mice in Saline + Saline and METH + Sulpiride groups; n = 9 cells from 3 mice in Saline + Sulpiride and METH + Saline groups; two-way ANOVA, F = 2.78, P < 0.001; ***P < 0.001 versus the Saline + Saline or METH + Saline group). l Statistical analysis shows that the threshold is significantly enhanced in the METH + Saline group, and Sulpiride rescues this phenomenon (one-way ANOVA, F = 8.8, P < 0.001; **P < 0.01 versus the Saline + Saline or METH + Saline group). Data are presented as mean ± SEM.

We also examined the effect of the D2R antagonist Sulpiride on the AP of PrL^Glu. The results showed that microinjection of Sulpiride (0.25 μg/side) into the PrL significantly rescued the decrease in AP frequency and increase in threshold of PrL^Glu caused by single-dose METH administration (Fig. 5j–l). Compared with the METH + Saline group, the half-width and decay time were significantly increased in the METH + Sulpiride group (Supplementary Fig. 13e-g). These results indicated that single-dose METH administration inhibited PrL^Glu mediated by D2R, and inhibition of D2R reversed this phenomenon.

Discussion

In the present study, we found that single-dose METH administration impaired the retrieval of ORM in mice. Immunofluorescence and fiber photometry results showed that the activation of PrL^Glu was inhibited in METH-treated mice during ORM retrieval, and chemogenetic activation of PrL^Glu or vCA1^Glu-PrL rescued the ORM retrieval impairment. Fiber photometry results showed that the DA level of the PrL was increased in METH-treated mice, and micro-infusion of the D2 receptor antagonist Sulpiride rescued the ORM retrieval impairment. Electrophysiological recordings in brain slices revealed that the intrinsic excitability of PrL^Glu and basal glutamatergic synaptic transmission were both decreased by single-dose METH administration, and inhibition of D2R reversed the decrease in intrinsic excitability of PrL^Glu.

Various clinical studies have demonstrated that chronic use of METH leads to a diverse range of significant cognitive impairments, including deficits in learning and memory, attention, and executive function [2, 3]. In a clinical study, subjects who received single-dose METH (0.3 mg/kg, intramuscular) displayed an increase in incorrect responses during memory recall tests [4]. Previous animal studies mainly used binge METH regimens (such as 4 × 10 mg/kg every 2 h within a day, i.p.), resulting in impaired ORM and spatial memory in rats accompanied by a loss of monoamine neurotransmitter [45]. Animals given chronic METH treatment (1–2 mg·kg⁻¹·d⁻¹ for 7 d, i.p.) showed impaired ORM [6], spatial memory [46] or working memory [47]. Furthermore, it was reported that chronic METH administration (2 mg·kg⁻¹·d⁻¹ for 14 d, i.p.) resulted in impaired ORM and spatial memory as well as a significant decrease in the number of hippocampal neurons in young rats during adulthood [48]. To avoid neurotoxicity in the animal model and concentrate more on the low-dose METH-induced psychiatric symptoms, some studies employed acute low-dose METH regimens. Single-dose METH (1 mg/kg, i.p.) was reported to impair the visual discrimination memory [7] and conditioned discrimination memory [8] of mice. However, in another study, single-dose METH injected subcutaneously (1, 4 mg/kg, s.c.) before the three-day habituation did not affect the ORM [6]. We hypothesize that these contradictory results may be due to variations in memory types. A study reported that single-dose METH (1 mg/kg, i.p.) improved spatial memory consolidation, but did not affect the retrieval and reconsolidation [9]. However, the effects of single-dose METH on the retrieval of ORM remain unclear. The neural mechanisms underlying short- and long-term memory exhibit variations. For instance, certain clinical studies have observed that individuals with severe short-term memory impairments might maintain relatively intact long-term memory. Conversely, patients with temporal lobe damage may experience impaired long-term memory while retaining relatively intact short-term memory [39, 40]. A study demonstrated that METH (1 mg·kg⁻¹·d⁻¹ for 7 d, s.c.) impaired long-term (1 d) ORM in mice, but had no significant effect on short-duration (1 h) ORM, suggesting that chronic administration of METH may have no effect on the acquisition of ORM and short-term ORM, but can impair long-term ORM [6]. In our study, by altering the interval between NOR training and test, we found that both short-term (1 h) and long-term (2 d) ORM retrieval was impaired by single-dose METH administration (Fig. 1). Our hypothesis posits that this phenomenon might be attributed to the effect of single-dose METH administration on the function of specific brain regions and neural circuits that are crucial for both short-term and long-term ORM.

Different types of memory engage distinct brain regions. The key brain regions underlying the single-dose METH administration-induced ORM retrieval impairment have not been previously identified. To address this, we performed c-Fos-based neural activity mapping. A previous study found that there was an increase in c-Fos expression in the DG, mPFC, and PRh of rats that underwent ORM retrieval, as compared with rats that stayed in their home cage, indicating that these regions may be critical for ORM retrieval [13]. In our study, we found that the number of c-Fos-positive neurons and c-Fos-positive glutamatergic neurons were both significantly decreased in the PrL and dDG in METH-treated mice, indicating that single-dose METH administration decreased the activation of PrL^Glu and dDG^Glu (Fig. 2; Supplementary Fig. 3). In addition, the number of c-Fos-positive neurons was significantly increased in the majority of brain regions, including dCA1, LA, BLA, and CeA (Fig. 2b). We speculate that these phenomena may be attributed to the extensive neuronal excitability induced by METH. It was reported that METH, functioning as a psychostimulant, increased the expression of c-Fos in hypothalamic-pituitary-adrenal axis-associated regions including the hypothalamus, amygdala, hippocampus and cortex [49].

A study reported that the c-Fos induction in the PRh of Kv7.2 knock-in mice was significantly lower than that of wild-type mice after NOR training, and the transgenic mice exhibited ORM impairment. Interestingly, ORM impairment was also rescued after pharmacological intervention restored the c-Fos expression in PRh [42]. Therefore, we hypothesized that decreased neuronal activity in specific brain regions may be critical in causing ORM impairment. In our study, we focused on two key brain regions in which c-Fos expression was decreased following the administration of single-dose METH. We found that the selective activation of dDG^Glu through chemogenetic methods failed to rescue the ORM retrieval impairment, and inhibition of dDG^Glu could not impair the ORM retrieval of normal mice (Supplementary Fig. 8). The DG is a subregion of the hippocampus, and it is well acknowledged that the hippocampus is essential for the development of episodic memory [50]. A study reported that lesions in the DG did not induce ORM impairment in rats when the context was the same as the training phase. However, the ORM was impaired when the experimental arena changed [51]. These results indicated that DG may play an important role in combining object and spatial information; however, DG is not crucial to ORM which does not involve changes in spatial information, or other brain regions can compensate for DG function when the DG is impaired. In addition, a previous study reported that chemogenetic activation of the parvalbumin interneurons of DG did not affect ORM retrieval, but impaired social recognition memory [52]. Optogenetic inhibition of the neural circuit from entorhinal cortex to DG impaired social recognition memory but did not affect ORM [53]. The above evidence suggested that dDG^Glu may not be involved in the ORM retrieval impairment induced by single-dose METH administration, and it remains to be examined whether METH impairs ORM when the environment changes and whether dDG^Glu is involved in this process. Subsequently, we found that the specific activation of PrL^Glu effectively rescued the ORM retrieval deficits, and inhibition of PrL^Glu significantly impaired the ORM retrieval of normal mice (Fig. 3; Supplementary Fig. 7). The PrL is a subregion of the mPFC, and it is widely recognized that the mPFC is essential for cognition, attention, decision-making, reward and emotion [54, 55]. For instance, pharmacological inhibition with muscimol or lesion of the mPFC before NOR training did not affect NOR performance, indicating that mPFC may not be engaged in ORM acquisition [56, 57]. Nevertheless, microinfusions into the mPFC of anisomycin after NOR training or muscimol before the test impaired NOR performance, indicating that mPFC is engaged in ORM consolidation and retrieval [12, 57]. The above evidence suggested that PrL^Glu may be involved in the ORM retrieval impairment induced by single-dose METH administration.

Our study demonstrated that the decreased level of PrL^Glu activation may play a crucial role in the ORM retrieval impairment induced by single-dose METH administration. The ORM is dependent not only on specific brain regions but also on complex neural circuits that connect the cortical regions and temporal lobe structures [58].To investigate whether the impairment could be reversed, we indirectly activated PrL^Glu by activating glutamatergic projections from the upstream brain regions of PrL^Glu. The PrL is abundant in glutamatergic neurons that receive long-range inputs from many other brain regions, including the thalamus, BLA, hippocampus and other cortical regions. Our retrograde tracing results also confirm the existence of these projections (Supplementary Fig. 9). Hippocampal inputs mainly arise from glutamatergic neurons in vCA1 and are largely restricted to glutamatergic neurons in PrL layer 5 [59]. Extensive research has been conducted on the interactions between the vCA1 and the mPFC, with particular emphasis on their role in episodic memory [60]. More importantly, a recent study has reported that optogenetic silencing of vCA1 input to the mPFC disrupts ORM retrieval in rats [14]. In our study, we found that chemogenetic activation of the vCA1^Glu-PrL circuit significantly alleviated the ORM retrieval impairment (Supplementary Fig. 10). Additionally, we investigated the role of the BLA^Glu-PrL circuit, and found that activating this circuit failed to rescue the ORM retrieval impairment (Supplementary Fig. 11). This could be attributed to the fact that this circuit primarily mediates emotional memory [61]. Taken together, our results show that activation of vCA1^Glu-PrL specifically rescues the ORM retrieval impairment induced by single-dose METH administration.

It is well established that DA has an essential influence on PFC cognition functions [62–64]. Using microdialysis, it was shown that METH induces an elevation in DA release in the mPFC [65, 66], and this was further confirmed in our study using fiber photometry (Fig. 4a, b). Previous studies reported that microinjection of the D1R antagonist SCH23390 [18, 19] or the D2R antagonist L741,626 [20] into the mPFC before NOR training significantly impaired ORM acquisition. Microinjection of SCH23390 or L741,626 into the PFC after NOR training also significantly impaired ORM consolidation [44]. For the D1R antagonist, but not the D2R antagonist, systemic administration instead of infusion into the specific brain region before each METH administration rescued the chronic METH (1 mg·kg⁻¹·d⁻¹ for 7 d) administration-induced ORM impairment [6, 67]. However, it is unclear how the excess DA induced by single-dose METH administration further modulates the DAergic signaling within the PrL and ultimately induces ORM retrieval impairment in mice. To examine the specific DA receptor subtypes within PrL modulated by single-dose METH administration, different doses of D1R or D2R antagonists were microinjected into PrL and the NOR task was performed [8, 68]. We found that infusion of the D1R antagonist SCH23390 (0.05, 0.1, 0.2 μg/side) bilaterally into the PrL before the NOR test failed to reverse the single-dose METH-induced ORM retrieval impairment (Supplementary Fig. 12). Interestingly, the D2R antagonist Sulpiride produced an ‘inverted-U’ dose-response on ORM: 0.25 μg, but not 0.0625 μg or 1 μg, rescued the ORM retrieval impairment (Supplementary Fig. 12). In addition, we found that Sulpiride (0.25 μg/side) impaired the ORM retrieval of normal mice (Fig. 4d). This phenomenon has been observed in many studies, wherein either too little or too much stimulation of dopamine receptors impairs cognitive performance [69–71]. We speculated that Sulpiride (0.25 μg/side) may potentially alleviate the overactivation of D2R caused by excessive DA, dampening the activity to a more optimal level for ORM retrieval. For normal mice, Sulpiride (0.25 μg/side) may dampen the activity to a sub-optimal level, causing impaired ORM retrieval. Several studies have also reported that D2R is associated with METH-induced memory impairment. For example, it was reported that pretreatment with Stepholidine, an alkaloid from the Chinese herb Stephania intermedia, conferred a protective effect against METH (10 mg/kg, i.p., once per day for 7 consecutive days)-induced ORM deficits, possibly through the rescue of the METH-induced overactivation of the dopaminergic signaling pathway mediated by D2R [72]. In another study, single-dose METH (1 mg/kg, i.p.) was reported to impair the conditioned discrimination memory of mice, and microinjection of a D2R antagonist into the NAc significantly rescued the spine enlargement and memory impairment caused by METH [8]. Additionally, it is unclear whether Sulpiride affects DA release in PrL. However, previous studies found that administration of the D2R antagonists Sulpiride and Haloperidol did not change the DA levels in the mPFC of rats [73, 74]. Therefore, we speculate that microinjection of Sulpiride into PrL may not affect the release of DA in the PrL. Of course, it is important to detect the DA levels in the PrL to clarify the effect of Sulpiride on DA release in the PrL. Taken together, our results prove that single-dose METH administration increases DAergic signaling in PrL through overactivation of D2R, thereby impairing the ORM retrieval.

DA is strongly correlated with neuronal synaptic strength and excitability [26–28]. Ever since Donald Hebb’s hypothesis in 1949 on the role of synaptic plasticity as a fundamental mechanism underlying the development of memory, the focus of research on memory has mostly centered around synapses and the process of synaptic transmission [75, 76]. Our study found that the mEPSC frequency of PrL^Glu was significantly decreased following METH administration (Fig. 5i). This result is consistent with a previous study, which reported that chronic METH administration (1 mg·kg⁻¹·d⁻¹ for 7 d) decreased mEPSC frequency in mPFC pyramidal neurons [29]. It was reported that single-dose METH (1.0 mg/kg) administration induced the paired-pulse facilitation of EPSC in mPFC pyramidal neurons [77]. The above evidence indicates that the administration of METH results in a reduction in the presynaptic probability of glutamate release from PrL^Glu. In addition to synaptic changes, it has been proposed that the excitability of neurons plays a role in the processes of memory acquisition, consolidation, and retrieval [23–25, 78]. In particular, during memory retrieval, the excitability of engram cells was increased, which may contribute to an animal’s ability to recognize contexts more precisely and more effectively [24]. A recent study reported that chronic METH administration (1 mg·kg⁻¹·d⁻¹ for 14 d) impaired the temporal order memory, and this effect may be associated with the enhancement of the intrinsic excitability of parvalbumin-positive interneurons in mPFC [79]. It was reported that METH affected the intrinsic excitability of dopaminergic neurons [80] and GABAergic neurons in striatum [81]. In our study, we measured the AP of PrL^Glu and observed a significant decrease in the AP frequency and an increase in the AP threshold following METH administration (Fig. 5d, f). This indicated that single-dose METH administration decreases the intrinsic excitability of PrL^Glu. In addition, we found that infusion of the D2R antagonist Sulpiride (0.25 μg/side) bilaterally into the PrL before the NOR test reversed the single-dose METH-induced decrease in intrinsic excitability of PrL^Glu (Fig. 5k, l). Taken together, these results suggested that single-dose METH administration may ultimately impair ORM retrieval by increasing DAergic signaling in PrL mediated by D2R, which in turn decreases excitability in PrL^Glu.

Our study found that the D2R antagonist Sulpiride, when microinfused into the PrL, rescues the ORM retrieval impairment. Future studies are needed to comprehensively evaluate the effect and expression of D2R in PrL^Glu, especially in PrL neurons receiving vCA1^Glu projections, as well as other types of PrL neurons on the single-dose METH administration-induced ORM retrieval impairment. In addition, we found that single-dose METH administration decreased the glutamate release of PrL^Glu. However, because of technological limitations, we are unable to measure synapse function between vCA1^Glu and PrL^Glu.

Our data show that single-dose METH administration impairs ORM retrieval, and this may be attributed to reduced PrL^Glu activity which is possibly associated with excessive DA activity on D2R. Selective activation of PrL^Glu or vCA1^Glu-PrL rescues the ORM retrieval impairment (Fig. 6). Our study offers an experimental foundation for investigating the underlying mechanisms of cognitive impairment caused by single-dose METH administration, and provides insights into potential therapeutic approaches.

Fig. 6. Summary schematic.

Single-dose METH administration impairs ORM retrieval, and this may be attributed to reduced PrL^Glu activity which is possibly associated with excessive DA activity on D2R. Selective activation of PrL^Glu or vCA1^Glu-PrL rescues the ORM retrieval impairment.

Author contributions

JCM designed the study; JCM carried out the study, analyzed data and wrote the manuscript; XHC, XNZ, and YTL supervised the study and helped in discussing the data; JCM performed the experiments; AXR, YH, CLY, and ZTX participated in the animal experiments; CFW and JYY supervised this study and reviewed the manuscript. All authors read and approved the manuscript.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41401-024-01321-9.