Address correspondence and reprint requests to Dr Ana P. Silva, Laboratory of Pharmacology and Experimental Therapeutics, Faculty of Medicine, Subunit 1, Polo 3, Azinhaga de Santa Comba, Celas, 3000-354 Coimbra, Portugal. E-mail: email@example.com.
Methamphetamine (METH) is a psychostimulant drug that causes irreversible brain damage leading to several neurological and psychiatric abnormalities, including cognitive deficits. Neuropeptide Y (NPY) is abundant in the mammalian central nervous system (CNS) and has several important functions, being involved in learning and memory processing. It has been demonstrated that METH induces significant alteration in mice striatal NPY, Y1 and Y2 receptor mRNA levels. However, the impact of this drug on the hippocampal NPY system and its consequences remain unknown. Thus, in this study, we investigated the effect of METH intoxication on mouse hippocampal NPY levels, NPY receptors function, and memory performance. Results show that METH increased NPY, Y2 and Y5 receptor mRNA levels, as well as total NPY binding accounted by opposite up- and down-regulation of Y2 and Y1 functional binding, respectively. Moreover, METH-induced impairment in memory performance and AKT/mammalian target of rapamycin pathway were both prevented by the Y2 receptor antagonist, BIIE0246. These findings demonstrate that METH interferes with the hippocampal NPY system, which seems to be associated with memory failure. Overall, we concluded that Y2 receptors are involved in memory deficits induced by METH intoxication.
Methamphetamine (METH) is a highly addictive psychostimulant that may lead to neurological and psychiatric abnormalities. Indeed, METH abusers show structural brain abnormalities, specifically in the hippocampus and prefrontal cortex, which are associated with cognitive deficits (Thompson et al. 2004; Salo et al. 2009). Similarly, contingent and non-contingent METH regimens lead to impairment of spatial, short- and long-term recognition and perceptual memories in rodents (Cherng et al. 2007; Lee et al. 2011; O'Dell et al. 2011; Reichel et al. 2012). In accordance with these observations, we previously showed that an acute METH administration alters the expression of several rat hippocampal ionotropic glutamate receptor subunits, which seems to be correlated with the hippocampal-dependent memory impairment observed (Simões et al. 2007). It is known that METH induces neuropathology via several mechanisms, including monoaminergic system damage, excitotoxicity, and neuroinflammation (Silva et al. 2010). In fact, METH intoxication triggers hippocampal gliosis and cytokine production (Gonçalves et al. 2008, 2010), together with significant cytoskeleton, synaptic, and axonal protein alterations (Gonçalves et al. 2010), but without evidence of cell death.
Neuropeptide Y (NPY) is a neuromodulator widely distributed in the hippocampus (Gehlert 2004) and acts via NPY Y1, Y2, and Y5 receptor subtypes (Silva et al. 2005a; Xapelli et al. 2006). This peptide has several important functions such as the regulation of appetite and circadian rhythms (Berglund et al. 2003), cognitive processing (Thorsell et al. 2000; Karl et al. 2008; Sørensen et al. 2008a,b), and neuroprotection (Silva et al. 2005b). Moreover, seizures up-regulate hippocampal NPY levels, and it has been consistently considered an endogenous antiepileptic agent (Woldbye et al. 1996; Silva et al. 2003a, 2005a; Xapelli et al. 2007). However, increased NPY levels could also affect other hippocampal functions, including learning and memory. In fact, some authors have clearly demonstrated that hippocampal NPY over-expression was accompanied by hippocampal activity-dependent plasticity reduction in excitatory synapses, which was associated with acquisition and retention deficits in spatial memory (Thorsell et al. 2000; Sørensen et al. 2008a). Furthermore, in vitro studies showed that hippocampal NPY reduces calcium influx and glutamatergic transmission mainly via presynaptic Y2 receptors (Silva et al. 2003b), and these effects result in the suppression of long-term potentiation (LTP) (Sørensen et al. 2008a,b). There are limited data available regarding the effects of METH on NPY and its receptors. It is known that multiple high doses of METH produce an increase of rat striatal pre-pro-NPY mRNA-expressing neurons (Horner et al. 2006). Similarly, Thiriet et al. (2005) demonstrated an up-regulation of striatal NPY mRNA, together with a down-regulation or biphasic changes in Y1 or Y2 receptor expression, respectively, following METH administration. The authors also reported neuroprotective effects of NPY against METH-induced striatal neurotoxicity mediated via Y1 and Y2 receptors (Thiriet et al. 2005). Nevertheless, as far as we know, the role of the hippocampal NPY system under METH consumption has never been addressed before.
Thus, the aim this study was to investigate the possible changes in mouse hippocampal NPY expression, as well as the expression and functionality of NPY receptors triggered by METH intoxication. Moreover, memory performance and the signaling pathway underlying such alterations were also evaluated. We concluded that METH increases hippocampal NPY levels and differentially affects the levels and functionality of NPY receptors. Moreover, we suggested that augmented Y2 receptor activation may be implicated in METH-induced memory impairment. Overall, our findings demonstrate that METH interferes with the hippocampal NPY system that may be associated with memory impairment.
Materials and methods
Animal procedures and drug treatments
Three-month-old male C57BL/6J mice (20–30 g body weight, Charles Rivers Laboratories, Barcelona, Spain) were housed with access to food and water ad libitum on a 12-h light/dark cycle in a temperature-controlled room. Behavioral experiments were performed in a sound-attenuated room, where mice were habituated for 1 h. Experimental procedures were performed according to the guidelines of the European Community Council Directives (2010/63/EU), the Portuguese law for the care and use of experimental animals (DL nº129/92), and the Animal Experimentation Inspectorate, Denmark. The authors attest that all efforts were made to minimize animal suffering and the number of animals used.
Animals were anesthetized with Avertin® [0.013 mL/g body weight, intraperitoneal administration (i.p.)]. Afterward, a stainless-steel guide cannula (Plastic One, Roanoke, VA, USA) was implanted unilaterally above the right lateral ventricle, according to the coordinates described by Decressac et al. (2010), and was fixed to the skull with dental cement. Mice were individually housed for at least 3 days, followed by drug treatments using an injector cannula (Plastics One). Dummy cannulas (Plastics One) were used during experiments, except when injection was conducted.
METH [(+)-methamphetamine hydrochloride, Sigma-Aldrich, St Louis, MO, USA; 30 mg/kg body weight, i.p.] was administered and mice were killed at 1 h, 24 h, 3 days and 7 days post-injection. Importantly, this paradigm provides a good relevance to intravenous and smoked routes of METH exposure in humans and simulates the toxic effects of METH in non-tolerant users (Krasnova and Cadet 2009; Marshall and O'Dell 2012). After a single dose, amphetamines persist in the human body for nearly a day (Cook et al. 1993), and this long duration of action is thought to contribute to the brain injury and cognitive impairments seen in METH users (Marshall and O'Dell 2012). In addition, the METH intoxication protocol selected in this study has been successfully used by us and others (Zhu et al. 2006; Gonçalves et al. 2010; Martins et al. 2011; Tulloch et al. 2011) and did not cause seizures in treated animals. Nevertheless, it is important to notice that it is always critical to make a correlation between an animal model and human condition. So, taking into consideration all the limitations in this type of studies, we selected a well-known and -established animal model of METH intoxication. Control animals were administered with 0.9% NaCl (saline group) and killed at the same time points as abovementioned. NPY Y2 receptor antagonist was dissolved in 1% dimethylsulfoxide/saline solution and injected intracerebroventricularly (i.c.v., 2 μL of 1.0 nmol BIIE0246; Tocris Bioscience, Bristol, UK) as previously described (Redrobe et al. 2002; Thorsell et al. 2002), 45 min before METH treatment. Control animals received 1% dimethylsulfoxide/saline (i.c.v.). Importantly, we have considered the use of knockout mice to complement this study, but we identified several problems both in Y2 (Zambello et al. 2011) and AKT (Chen et al. 2001; Woulfe et al. 2004; Easton et al. 2005) knockout mice that led us to exclude its use for this specific purpose. Thus, we decided to use a pharmacological approach by applying the Y2 receptor antagonist, BIIE0246 (Dumont et al. 2000).
Brain sectioning for in situ hybridization and binding studies
Mice were killed 1 h, 24 h, 3 days and 7 days post-METH treatment, and brains were removed, frozen on dry ice and stored at −80°C. Coronal sections (12 μm) were cut on a cryostat (Thermo Shandon Inc., Pittsburgh, PA, USA) across the hippocampus, thaw-mounted onto slides, and stored at −80°C until further use.
In situ hybridization studies
In situ hybridization procedure was performed as previously described (Woldbye et al. 2005; Christensen et al. 2006). Briefly, following fixation in 4% paraformaldehyde solution, brain sections were incubated at 42°C overnight (ON) with synthetic DNA complementary oligonucleotide probes (DNA Technology A/S, Aarhus, Denmark) labeled with [α35S]dATP (1250 Ci/mmol, Amersham, GE Healthcare UK Ltd, Little Chalfont, Buckinghamshire, UK) using terminal deoxynucleotidyl transferase (Roche Diagnostics, Mannheim, Germany). To evaluate NPY and Y1, Y2, and Y5 receptor mRNA levels, we used a sequence of probes previously described by Woldbye et al. (2005, 2010). Afterward, slides were washed, air-dried, and exposed to 35S-sensitive Kodak BioMax MR films together with 14C-microscales (both from Amersham) for 4 weeks before being developed in Kodak GBX developer. The mRNA quantification was performed using Scion Image computer analysis software (NIH, Bethesda, MD, USA) by measuring bilaterally over the dentate gyrus (DG), Cornu Ammonis field 1 (CA1) and Cornu Ammonis field 3 (CA3) subregions, and optical densities (nCi/g) were based on 14C-microscales calibration curves. Right- and left-side values were averaged per section and, after subtracted background, used to calculate the mean for each animal. In control experiments, the specificity of the antisense oligoprobes was confirmed by adding corresponding ‘non-labeled’ antisense probes.
NPY receptor binding
This protocol was performed as previously described (Woldbye et al. 2005, 2010; Christensen et al. 2006). Briefly, slices were pre-incubated in binding buffer, and then incubated in the same buffer containing 0.1 nM [125I][Tyr36]mono-iodo- peptide YY (PYY) (4000 Ci/mmol, porcine synthetic; Amersham) alone or together with l μM NPY (mouse synthetic; Bachem AG, Bubendorf, Switzerland) to visualize total binding to all NPY receptor subtypes or non-specific binding, respectively. Afterward, slices were exposed for 4 days to 125I-sensitive Kodak Biomax MS films with 125I-microscales (both from Amersham). Calibration was based on the 125I-microscales, and optical densities were measured in hippocampal subregions, using Scion Image computer software (NIH, Bethesda, MD, USA). Total specific NPY-sensitive binding was calculated by subtracting non-specific binding from total binding.
NPY-stimulated [35S]GTPγS binding autoradiography
As previously described (Christensen et al. 2006; Silva et al. 2007; Woldbye et al. 2010; Gøtzsche et al. 2012), brain sections were pre-incubated in assay buffer with NPY receptor antagonists (if applicable) to shift all G-proteins into the inactive state. Then, slices were incubated in assay buffer + 40 pM [35S]-Guanine-5′-triphosphate gamma S (GTPγS) (1250 Ci/mmol; PerkinElmer, Skovlunde, Denmark) with 10−6 M NPY (Bachem AG) plus different combinations of NPY receptor antagonists (10−6 M of each) as follows: BIBP3226, Y1 receptor antagonist (Bachem AG); BIIE0246, Y2 receptor antagonist (Tocris Cookson); L-152,804, Y5 receptor antagonist (Tocris Cookson). Basal, non-specific binding and total blocking were performed as control experiments. Sections were exposed to Kodak BioMax MR films (Sigma-Aldrich) for 5 days together with 14C standards (Amersham). Functional binding was quantified by measuring optical densities bilaterally in the DG (molecular layer), CA1 and CA3 (fiber layers), using Scion Image® analysis program. Right and left values were averaged per section and used to calculate the mean of each animal, after subtraction of the background.
Western blot analysis
Mice were killed at 1 h or 24 h post-METH administration, brains were removed and the hippocampi were then homogenized in lysis buffer supplemented with a protease inhibitor cocktail. Protein concentrations were determined using the BCA assay kit (Pierce, Rockford, IL, USA), and protein samples (50 μg) were separated onto sodium dodecyl sulfate–polyacrylamide gel electrophoresis, before being transferred onto polyvinylidene difluoride membrane (PVDF; Millipore, Madrid, Spain). After blocking, membranes were incubated, ON at 4°C, with the following primary antibodies: phospho-Akt-Ser473 (1 : 1000), phospho- mammalian target of rapamycin (mTOR)-Ser2448 (1 : 1000), phospho-p70S6 kinase-Thr421/Ser424 (1 : 1000; Cell Signaling Technology, Danvers, MA, USA), and anti-neuropeptide Y (1 : 1000; Sigma-Aldrich). Membranes were then incubated with secondary antibodies, at 23°C during 1 h, and immunoreactive bands were visualized by enhanced chemofluorescent detection (ECF kit, Amersham). The levels of phosphorylation for AKT, mTOR and p70S6 kinase were normalized by reprobing stripped membranes with antibodies raised against the respective total proteins. Moreover, the membranes were reprobed with anti-Glyceraldehyde 3-phosphate dehydrogenase GAPDH (1 : 1000; Abcam, Cambridge, UK) as a loading control. The immunoblots were analyzed with ImageQuant software (NIH, Bethesda, MD, USA) to measure the optical density of the bands.
Animals were anesthetized with sodium pentobarbital (Sigma-Aldrich) at 1 h or 24 h post-METH administration, and intracardially perfused with phosphate-buffered saline (PBS) (10 mL) followed by 4% paraformaldehyde (20 mL). Brains were removed and post-fixed in the same solution, followed by immersion in 20% sucrose. Afterward, coronal sections were cut (30 μm) along its anterior–posterior axis (between −1.28 mm and −2.30 mm from Bregma; Paxinos and Franklin 2004) and collected in PBS until further used. Double-labeling immunofluorescence was performed for NPY and neuron-specific class III beta-tubulin (Tuj-1). Slices were blocked with 10% fetal bovine serum (FBS)/0.5% Triton X-100 in PBS, incubated with polyclonal anti-NPY (1 : 200; ON at 4°C), washed and then incubated for 90 min at 23°C with Alexa Fluor 488 antibodies (1 : 200; Invitrogen, Inchinnan Business Park, UK). Afterward, slices were again blocked with 1% FBS/0.3% Triton X-100 in PBS followed by Tuj-1 incubation (1 : 1000; ON at 4°C; Covance, Ermeryville, California, USA). After rinses, slices were incubated with Alexa Fluor 594 secondary antibodies (1 : 200; Invitrogen), and the nuclei were counterstained with Hoechst 33342 (Sigma-Aldrich). Additionally, other brain sections were blocked with 5% goat serum/0.3% Triton X-100 or 10% FBS/0.5 Triton X-100 in PBS, followed by incubation with phospho-AKT (1 : 200; Cell Signaling) or Y2 receptor (1 : 100; Alomone Labs, Jerusalem, Israel) primary antibodies, respectively, ON at 4°C. Then, slices were incubated with the secondary antibody Alexa Fluor 488 (1 : 200; Invitrogen) for 90 min at 23°C, stained with Hoechst 33342 (Sigma-Aldrich) and mounted in Dako fluorescence medium (Dako North America, Carpinteria, USA). The fluorescent images were recorded using a LSM 710 Meta Confocal microscope (Carl Zeiss, Oberkochen, Germany).
Spatial working memory was accessed using the Y-maze apparatus with three identical black horizontal arms (5 cm wide × 36 cm long × 16 cm high; Panlab, Barcelona, Spain) symmetrically disposed at 120º to each other. In the first trial, 5 min after METH or saline injection, mice were placed in the end of a random-assigned arm (start arm) and allowed to explore only two arms (start and familiar arms) for 8 min. The second trial was performed at 24 h post-treatment, to evaluate long-term memory (O'Dell et al. 2011; Reichel et al. 2011), and each mouse was placed again in the maze with free access to all three arms during 8 min. To discriminate if METH impairs acquisition rather than memory retention, other group of animals were administered with METH immediately after the exposure to start and familiar arms (8 min), and after 24 h, the mice were allowed to explore the three arms. The exploration time and the percentage number of entries into the arms were recorded. The total number of arms entries was measured as an index of locomotor activity to rule out the possible interference of changes in motility with the parameters of learning and memory. The recognition of the novel arm from the other two familiar arms is considered a memory improvement effect (Cherng et al. 2007).
Novel object recognition test
The novel object recognition test was conducted in a white open-field box with the floor divided into nine equal squares (Bura et al. 2007), and two objects, a plastic animal figure and a Lego® (LEGO Company Ltd., GB-Slough, Berks, UK) toy, were used. Before experiments, mice were individually habituated to the open field for 40 min. In the next day, 5 min after or before METH administration (30 mg/kg; i.p.), animals were submitted to a 10-min acquisition trial during which they were placed in the box in the presence of the object A. We recorded locomotor activity (number of crossings) and latency to explore object A. After 24 h, a 10-min test trial was performed to assess long-term recognition and retention memories (O'Dell et al. 2011; Reichel et al. 2011), in the presence of objects A and B. Once again, locomotor activity and latency to explore object A (tA) and object B (tB) were recorded, and recognition index was expressed as (tB/(tA+tB))x100. Objects A and B were counterbalanced so that half of the animals in each experimental group were first exposed to object A and then to object B or vice versa. Importantly, no mice showed preference for the different objects (data not shown).
Measurements of in situ hybridization and receptor-binding studies were performed by a person blinded to treatments. Data were analyzed by one-way anova followed by Bonferroni's multiple comparison test, except in Fig. 5b and d where analyses were performed using the Mann–Whitney test. Data were considered statistically significant when p < 0.05.
METH increases hippocampal NPY, Y2 and Y5 receptor levels
The NPY system seems to be altered under several pathological conditions (see review Silva et al. 2005b), but limited information is available regarding this issue (Goodman and Slovitter 1993; Cannizzarro et al. 2003; Thiriet et al. 2005). Nevertheless, Thiriet et al. (2005) showed that multiple high doses of METH increases striatal NPY mRNA and changed the Y1 and Y2 receptor mRNA levels. Thus, we hypothesized that METH intoxication (30 mg/kg; i.p.) could also trigger significant alterations in the hippocampal NPY system. We found that NPY mRNA levels were up-regulated at 1 h post-METH administration (163.3 ± 10.5% of control, n = 9; p =0.0007; F5,45 = 5.23) in the DG, whereas in the CA3 subfield, this increase was only significant after 24 h (175.9 ± 22.7% of control, n = 7; p =0.0099; F5,40 =3.52; Fig. 1a). In the CA1 subregion, we observed an increase of NPY mRNA levels at both 1 h (190.1 ± 5.8% of control, n = 6) and 24 h post-injection (187.9 ± 22.9% of control, n = 7; p <0.0001; F5,40 = 7.22; Fig. 1a). Importantly, the NPY mRNA in hippocampal formation returned to basal values 3 days post-METH treatment (Fig. 1a). In Fig. 1b, representative in situ hybridization autoradiograms shows the increase of NPY mRNA expression in the mouse hippocampus at 1 h and 24 h post-METH administration. Additionally, we showed that METH increases NPY protein levels at 1 h (133.3 ± 3.8% of control, n = 9) and 24 h post-injection (166.9 ± 13.3% of control, n = 9; p <0.0001; F2,16=22.63; Fig. 1c). These results were confirmed by immunohistochemical analysis (Fig. 1d), which demonstrates that neurons are the main source of NPY, since Tuj-1 immunoreactivity (neuronal marker) co-localized with NPY labeling (Fig. 1d), whereas there were no co-localization of NPY with specific markers for astrocytes or microglia (GFAP or CD11b labeling, respectively; data not shown).
As NPY in the hippocampus acts predominantly via Y1, Y2, and Y5 receptors (see review Silva et al. 2005b; Xapelli et al. 2006), we also examined whether METH was able to induce changes in the receptor mRNA levels. We concluded that METH did not induce significant changes in Y1 receptor mRNA levels (Fig. 2a). However, there was a significant increase of Y2 mRNA levels in the CA3 (120.6 ± 1.7% of control, n = 6; p <0.0001; F5,42 = 10.08) and CA1 (127.6 ± 5.0% of control, n = 7; p <0.0001; F5,45 = 9.95) pyramidal cells at 1 h post-injection (Fig. 2b), and in dentate granule cells after 24 h (124.0 ± 7.3% of control, n = 8; p <0.0001; F5,45 = 6.24; Fig. 2b). Accordingly, we also observed an up-regulation of Y2 receptor protein levels after 1 h in both CA3 and CA1 subregions (Fig. 2d). At 3 days and 7 days post-drug administration, the values returned to basal levels. Moreover, we observed an increase of Y5 receptor mRNA levels at 1 h after METH treatment in all hippocampal subregions analyzed as follows: DG, 159.8 ± 17.8% of control (n = 7; p <0.0001; F5,44 = 7.45); CA3, 146.3 ± 16.7% of control (n = 7; p <0.0001; F5,43 = 3.95); CA1, 158.7 ± 15.2% of control (n = 6; p =0.059; F5,44 = 10.09) (Fig. 2c).
Changes in NPY receptor binding induced by METH
The previous findings led us to hypothesize that the activity of NPY receptors could be also affected by METH. In fact, we demonstrated that total [125I]PYY binding, which allows us to visualize the total NPY binding sites (Sóvágó et al. 2001), was significantly increased in the CA1 subregion at 1 h following METH injection (1074.0 ± 82.5 pCi/mol, n = 7; p = 0.043; F5,41 = 2.40) when compared with saline (736.0 ± 52.1 pCi/mol, n = 9; Fig. 3a and b). However, no significant changes were detected in the DG (p = 0.32) and CA3 subfield (p = 0.74; Fig. 3a and b). Additionally, no differences were detected in non-specific binding between saline and METH-treated mice (data not shown).
Then, we evaluated the effect of METH on the functional binding of Y1, Y2, and Y5 receptors using the [35S]GTPγS-binding assay. We concluded that METH administration induced a down-regulation of specific Y1 receptor-stimulated binding in the hippocampal formation at 1 h as follows: DG, −6.2 ± 4.3 nCi/g (n = 7, p =0.0374; F5,29 = 2.75); CA3, −21.1 ± 8.3 nCi/g (n = 8, p =0.0139; F5,37 = 3.33); CA1, −27.0 ± 0.7 nCi/g, (n = 8, p =0.005; F5,38 = 4.03; Fig. 4a). On the other hand, METH induced a significant increase in Y2 receptor binding to 119.9 ± 14.8 (n = 10), 154.1 ± 9.3 (n = 7), or 148.3 ± 11.0 (n = 7) pCi/g (p <0.0001; F5,43 = 7.07) at 1 h, 24 h, or 3 days post-injection in CA3 subregion, respectively, without significant changes at 7 days (Fig. 4b). Furthermore, in the CA1 subfield, METH led to an up-regulation of Y2 receptor binding at 24 h (143.7 ± 12.5 pCi/g, n = 7) that remained until 3 days post-drug (143.5 ± 5.3 pCi/g, n = 7; p < 0.0001; F5,37 = 15.97; Fig. 4b). In Fig. 4d, it is possible to observe Y2 receptor functional binding at 24 h post-METH and the blockade by the Y2 receptor antagonist, BIIE0246. No changes were detected in Y5 receptor functional binding (Fig. 4c). Taken together, these results indicate that METH not only interferes with the expression of NPY receptors but also with their functionality. Interestingly, Y1, Y2, and Y5 are also differentially affected by this drug of abuse.
Selective blockade of Y2 receptors prevents memory impairment induced by METH
As NPY plays an important role in several physiological functions including memory processing (Thorsell et al. 2000; Karl et al. 2008; Sørensen et al. 2008a) and considering the changes of the hippocampal NPY system induced by METH, our next goal was to investigate if this drug of abuse led to memory deficits. For that, we examined spatial working memory using a Y-maze test and verified that METH-treated mice spent less time in the novel arm (81.5 ± 7.0 s, n = 15; p = 0.0001; F3,40 = 10.65) as compared with the saline-treated group (123.7 ± 10.6 s, n = 14; Fig. 5a). Taking into consideration that the integrity of the hippocampus is essential not only to spatial working memory but also to recognition memory (Broadbent et al. 2004), we also performed the novel object recognition test, and similarly to the previous results, the preference toward a novel object in the METH-treated mice was decreased (44.2 ± 3.7 recognition index, n = 15; p =0.0002; F3,37 = 8.53) when compared to the saline group (66.3 ± 2.1 recognition index, n = 13; Fig. 5c). As in these studies we administered METH immediately before the first trial, we further performed another set of experiments to specifically evaluate the effect of METH on memory retention without interfering with acquisition. We concluded that METH-injected animals display a poor long-term retention (Y-maze test: 79.8 ± 6.2 s, n = 8; p =0.0031; Fig. 5b; novel object recognition test: 39.1 ± 4.4 recognition index, n = 8; p =0.0006; Fig. 5d) when compared with control animals (Y-maze test: 128.1 ± 10.0 s, n = 7; Fig. 5b; novel object recognition test: 64.7 ± 3.7 recognition index, n = 8; Fig. 5c). So, the results obtained with both administration protocols are very similar. Additionally, it is important to notice that during the familiarization phase in the Y-maze and novel object recognition test, the experimental groups showed no significant differences regarding the number of entries into the arms and the interaction with the objects, respectively, as well as no stereotypic behavior or differences in the exploratory activity and motivation (Figure S1 and S2). These observations support the conclusion that acute METH treatment leads to memory deficits rather than impairment in acquisition or a lack of interest in novelty.
To disclose the possible correlation between NPY system changes, namely the NPY and Y2 receptors up-regulation, and the memory deficits observed in the METH-treated mice, we evaluated if the blockade of Y2 receptors could prevent the recognition memory impairment. In fact, we observed that Y2 receptor antagonist, BIIE0246, completely prevented METH-induced spatial working (113.7 ± 8.6 s, n = 10; p =0.008; F3,40 = 10.65) and recognition (61.8 ± 4.9 recognition index, n = 7; p =0.002; F3,37 = 8.53) memory impairment, suggesting that METH-induced memory deficits involve the activation of Y2 receptors. Importantly, neither horizontal nor vertical exploratory activity was affected by treatment with METH plus BIIE0246 (Figure S1 and S2), as previously demonstrated by Thorsell et al. (2002).
Modulation of AKT/mTOR signaling pathway by METH
Long-term synaptic plasticity and memory require protein synthesis, which is regulated by various signaling pathways (Kandel 2001), including the AKT/mTOR pathway (Hoeffer and Klann 2010). Considering the memory impairment induced by METH, we further investigated if this drug could also disrupt the AKT/mTOR pathway. We observed that METH significantly decreased the phosphorylation levels of AKT, mTOR, and its down-stream p70S6k (Fig. 6). Specifically, at 1 h post-METH, there was a decrease in AKT phosphorylation levels (76.6 ± 3.2% of control, n = 10; p =0.0001; F4,44 = 28.28; Fig. 6a) that was confirmed by immunohistochemical analysis, showing phospho-AKT (p-AKT) immunolabelling reduction in the CA3 (Fig. 6b), DG, and CA1 (data not shown) subfields. Accordingly, we also observed a significant decrease of p-mTOR and p-p70S6k levels at 1 h (57.2 ± 6.1% of control, n = 13; p =0.0044; F4,33 = 8.93; Fig. 6c) and 24 h (73.3 ± 4.1% of control, n = 17; p =0.0001; F4,50 = 6.22; Fig. 6d), respectively, post-METH administration. Here, we also verified that Y2 receptor blockade prevented the down-regulation of the phosphorylated levels of AKT (188.3 ± 16.6% of control, n = 5; p =0.0020; F4,44 = 28.28), mTOR (105.3 ± 5.8, n = 5; p =0.0055; F4,33 = 8.93), and p70S6k (92.7 ± 3.7, n = 7; p =0.0047; F4,50 = 6.22), as shown in Fig. 6a–d. Altogether, these findings show that METH-induced memory impairment was accompanied by a disruption of the AKT/mTOR signaling pathway, which was once again prevented by the Y2 receptor blockade.
The present work reveals that hippocampal NPY system alterations are centrally involved in memory impairment induced by METH intoxication. Specifically, METH-induced memory deficit was abolished by Y2 receptor blockade, and we further unraveled the involvement of the AKT/mTOR signaling pathway in these effects.
METH is an illicit psychostimulant drug that affects several brain regions, leading to neurological changes that include damage to monoaminergic neurons, gray-matter loss, white-matter hypertrophy, and excitotoxicity (reviewed by Cadet and Krasnova 2009). Recently, we verified that an acute METH treatment leads to astrogliosis, microglia activation, and tumor necrosis factor-∝ production, accompanied by neuronal dysfunction, namely disturbance at cytoskeletal, synaptic and axonal protein levels, but without evidence of hippocampal cell death (Gonçalves et al. 2010). Furthermore, the same METH regimen increases the blood–brain barrier permeability in the hippocampus, without affecting the striatum and frontal cortex (Martins et al. 2011), which led us to conclude that hippocampus is particularly susceptible to METH. Concerning the impact of METH on the NPYergic system, Thiriet et al. (2005) showed that METH (4 × 10 mg/kg, each 2 h) increases striatal NPY mRNA levels after 3 and 7 days. Moreover, the same study reported that METH induced a down-regulation of Y1 receptor mRNA and a contrasting up-regulation of Y2 receptor mRNA levels (Thiriet et al. 2005). Nonetheless, these authors did not show if changes in the striatal NPY system led to behavioral alterations. Here, we hypothesized that the observed changes in the hippocampal NPY system could underlie METH-induced mnemonic deficits. In fact, neuroplastic changes in the NPY system are prominent in several brain pathologies, such as epilepsy, Parkinson's disease, brain ischemia, or drug abuse (Goodman and Slovitter 1993; Cannizzarro et al. 2003; Duszczyk et al. 2009; Olling et al. 2009). Here, we verified that expression levels of NPY and its receptors, as well as receptor function, were differentially altered by METH, depending on hippocampal subregion and the time-point analyzed. However, further studies are needed to clarify if these transient effects are benign or will lead to long-lasting brain injury. Nevertheless, other studies have demonstrated that distinct insults differentially modulate the NPYergic system. Olling et al. (2009) described that ethanol-treated rats (5× 20% ethanol daily, during 4 days) only changed Y2 receptor mRNA levels in the DG subregion. In contrast, repeated electroconvulsive seizures led to an up-regulation of Y1, Y2, and Y5 receptor mRNA levels in the DG subfield together with a decrease in Y2 receptor mRNA levels in the CA3 of mouse hippocampus (Christensen et al. 2006). NPY receptor localization in the hippocampal subregions may be one possible explanation for these differences. In fact, Y1 receptors are highly expressed in the DG, Y2 receptors show strong expression in CA3 and CA1 pyramidal cell layers, and Y5 receptors are mainly expressed in DG and CA1 subfields (Naveilhan et al. 1998). Several authors have suggested that NPYergic system changes can be an adaptive mechanism for counteracting degeneration and/or cell death (reviewed by Vezzani and Sperk 2004; Xapelli et al. 2006), but NPY signaling is not only involved in neuroprotection. Indeed, there are studies showing that NPY receptors regulate voluntary ethanol consumption and its toxic effects (Thiele et al. 2002; Schroeder et al. 2003; Rimondini et al. 2005). Moreover, the activation of Y1 receptors can have proconvulsive effects (Olesen et al. 2012), while Y2 receptors have an important role in the generation of anxiety- and stress-related behaviors (Tschenett et al. 2003).
Multiple strands of evidence indicate an important role for the hippocampus in the formation and retrieval of episodic and contextual memories in humans and animals (Wiltgen et al. 2010; Vann and Albasser 2011), and it is well established that hippocampal dysfunction produces pronounced amnesia for newly acquired information (reviewed by Frankland and Bontempi 2005). Our group previously described that acute subcutaneous injection of METH (30 mg/kg) impairs rat spatial working memory, one of the hallmarks of hippocampal integrity (Simões et al. 2007). Besides, it has been demonstrated that acute and chronic METH exposure leads to both spatial and recognition memory deficits in rodents (Belcher et al. 2005, 2008; O'Dell et al. 2011; Reichel et al. 2012). Additionally, human studies also proved that METH use leads to memory impairment (Scott et al. 2007). In fact, a single dose of amphetamines has a long-lasting effect in the human body, which may contribute to memory deficits shown in METH abusers (Marshall and O'Dell 2012). It is also known that the storage and update of memory also require hippocampal plasticity mechanisms, such as LTP (reviewed by Bruel-Jungerman et al. 2007; Neves et al. 2008). In fact, it was shown recently that rats (Hori et al. 2010) and mice (Swant et al. 2010) chronically exposed to METH exhibited a decrease of both membrane potential and LTP magnitude in the CA1 pyramidal cell layer. So, it is plausible to hypothesize that defective hippocampal synaptic plasticity could trigger METH-induced cognitive performance deficits.
It is well established that METH consumption interferes with several neurotransmitters, including glutamate (reviewed by Krasnova and Cadet 2009; Silva et al. 2010) that plays a pivotal role in LTP (reviewed by Bliss and Collingridge 1993). In fact, METH-injected rats (7.5 mg/kg every 2 h over a period of 6 h; i.p.) exhibited augmented hippocampal extracellular glutamate levels (Raudensky and Yamamoto 2007), and we previously demonstrated that an acute METH treatment (30 mg/kg, s.c.) up-regulates both N-methyl-D-aspartate and α-amino-3-hydroxy-5-methylisoxazole-4-propionate receptors in the rat hippocampus (Simões et al. 2007). Additionally, it is well known that METH causes pronounced increases in extracellular levels of monoamines that may lead to activation of an array of brain circuits, such as the perirhinal cortex-prefrontal cortex-hippocampal circuitry, which is essential for memory tasks (Reichel et al. 2012). Specifically, Reichel et al. (2012) suggested that alterations in the function of transporters for serotonin in the object-in-place circuitry may underlie memory deficits independently of overt neurotoxic effects. Here, we describe that METH increases the levels of hippocampal NPY and Y2 receptors, as well as Y2-mediated functional activity. Additionally, it was previously reported that the activation of metabotropic glutamate receptors stimulates NPY mRNA expression in granule cells and interneurons of rat dentate gyrus (Schwarzer and Sperk 1998). Moreover, presynaptic Y2 receptor activation mediates NPY-induced inhibition of hippocampal glutamate release (Silva et al. 2001), which can explain LTP impairment at CA1-subicular cell synapses (Sørensen et al. 2008a). Thus, we hypothesized that the observed changes in the hippocampal NPY system could underlie METH-induced mnemonic deficits. To clarify this question, we examined spatial working memory using a Y-maze test (Cherng et al. 2007), and taking into consideration that the integrity of the hippocampus is essential not only to spatial working memory but also to recognition memory (Broadbent et al. 2004), we also performed the novel object recognition test. The contribution of NPY and its receptors to cognitive behaviors has been poorly explored. Nevertheless, Thorsell et al. (2000) demonstrated that transgenic rats over-expressing NPY in the hippocampus show spatial memory impairment. Supporting this evidence, it was shown that vector-mediated hippocampal NPY over-expression in rats impaired LTP in the CA1 subfield and, consequently, the animals exhibited a delay in hippocampal-dependent spatial discrimination learning (Sørensen et al. 2008b). Moreover, we observed an early up-regulation in both mRNA and protein levels of Y2 receptors, as well as in their functional activity. These alterations persisted until 3 days post-METH injection and were significant in the CA1 subregion that has a prominent role in hippocampal plasticity. Additionally, it is well established that METH triggers the release of monoamines, but the relationship between both monoamine and NPY systems is poorly understood, with no references to cognitive functions. Nevertheless, data from Meurs et al. (2007) indicate that NPY-induced increases in hippocampal dopamine may be mediated via sigma 1 receptors and NPY effects occur via increased activation of hippocampal D2 dopamine receptors. On the other hand, hippocampal D2 dopamine receptor binding showed positive linear correlations not only with memory function but also with frontal lobe functions, which supports the conclusion that D2 dopamine receptors contribute to local hippocampal functions (long-term memory) (Takahashi et al. 2008). Here, we did not explore the possible involvement of the dopamine system, but we clearly showed that the blockade of Y2 receptors completely prevented METH-induced memory impairment.
Synaptic plasticity, and consequently memory formation, requires new protein synthesis (Kandel 2001; Bruel-Jungerman et al. 2007; Costa-Mattioli et al. 2009). Furthermore, protein synthesis underlying the formation of long-lasting memories is highly regulated at the translation level, where the AKT/mTOR pathway regulates the translation rate and the integration of information from diverse synaptic inputs (reviewed by Hoeffer and Klann 2010). Indeed, the AKT/mTOR signaling cascade was identified as being crucial to the induction of protein synthesis-dependent synaptic plasticity required for hippocampus-dependent learning and memory processes (Cammalleri et al. 2003; Opazo et al. 2003). Hence, to unravel the signaling pathways mediating the anti-mnemonic effects of METH, we explored the potential involvement of the AKT/mTOR pathway. We found that an acute dose of METH disrupted the AKT/mTOR pathway in the hippocampus through Y2 receptor activation. Moreover, we have previously shown that NPY via Y2 receptor activation inhibits Ca2+ influx (Silva et al. 2003b), which, in turn, can restrain the AKT/mTOR cascade compromising new protein synthesis and, consequently, inhibiting LTP (Opazo et al. 2003). Indeed, the impairment of protein synthesis machinery and LTP results in memory deficits. Thus, it is reasonable to postulate that our observations could be explained by Ca2+ depression mediated by activation of the hippocampal NPY system.
In summary, this study provides a new explanation for the memory deficits induced by METH intoxication. Our work supports the potential importance of the hippocampal NPY system, and the functional relationship between Y2 receptors and AKT/mTOR pathway activation in memory impairment induced by METH. Thus, we may suggest that targeting this specific pathway or the modulation of Y2 receptor activity could provide interesting therapeutic approaches.
This work was supported by Grants PTDC/SAU-FCF/67053/2006 and PTDC/SAU-FCF/098685/2008 (COMPETE and FEDER funds) and Fellowship SFRH/BD/35893/2007 from Foundation for Science and Technology (Portugal), co-financed by QREN. The authors thank Vanessa Coelho-Santos for her collaboration in the Y-maze and novel object recognition tests, and both Áurea Castilho and Andreia Gonçalves for their collaboration in animal manipulation for i.c.v. injection. All authors report no biomedical financial interests or potential conflicts of interest.