A single dose of cocaine raises SV2A density in hippocampus of adolescent rats

Abstract Objective Cocaine is a highly addictive psychostimulant that affects synaptic activity with structural and functional adaptations of neurons. The transmembrane synaptic vesicle glycoprotein 2A (SV2A) of pre-synaptic vesicles is commonly used to measure synaptic density, as a novel approach to the detection of synaptic changes. We do not know if a single dose of cocaine suffices to affect pre-synaptic SV2A density, especially during adolescence when synapses undergo intense maturation. Here, we explored potential changes of pre-synaptic SV2A density in target brain areas associated with the cocaine-induced boost of dopaminergic neurotransmission, specifically testing if the effects would last after the return of dopamine levels to baseline. Methods: We administered cocaine (20 mg/kg i.p.) or saline to rats in early adolescence, tested their activity levels and removed the brains 1 hour and 7 days after injection. To evaluate immediate and lasting effects, we did autoradiography with [3H]UCB-J, a specific tracer for SV2A, in medial prefrontal cortex, striatum, nucleus accumbens, amygdala, and dorsal and ventral areas of hippocampus. We also measured the striatal binding of [3H]GBR-12935 to test cocaine’s occupancy of the dopamine transporter at both times of study. Results: We found a significant increase of [3H]UCB-J binding in the dorsal and ventral sections of hippocampus 7 days after the cocaine administration compared to saline-injected rats, but no differences 1 hour after the injection. The [3H]GBR-12935 binding remained unchanged at both times. Conclusion: Cocaine provoked lasting changes of hippocampal synaptic SV2A density after a single exposure during adolescence


Introduction
Cocaine is a drug of abuse that can lead to repeated use and addiction in humans (Ward et al., 1997) and animals, including rats (Deroche-Gamonet et al., 2004). As reviewed by Brown and The psychostimulant effects of cocaine mainly relate to interactions with monoaminergic pathways, including dopaminergic synapses. Cocaine inhibits dopamine re-uptake by occupying the pre-synaptic dopamine transporters (DAT), thus raising the extracellular neurotransmitter presence above the physiological level and increasing dopaminergic neurotransmission (reviewed, e.g., by Kalivas (2007), Zhu and Reith (2008), Docherty and Alsufyani (2021)). Dopamine receptor occupancy is an important factor in cocaine addiction that tends to be proportional to the degree of cocaine craving (Wong et al., 2006). The cocaine effects on extracellular dopamine are transient, however, as demonstrated by microdialysis (Hurd and Ungerstedt, 1989), intravenous administration of low dose of cocaine produces immediate and dramatic increases of striatal dopamine levels, peaking at 10 minutes after cocaine administration and returning to baseline in 30 minutes.
Cocaine intake also significantly affects neuronal activity by producing important structural and molecular changes in multiple brain areas. Cocaine-induced neuroplasticity has been detected in structures related to dopaminergic neurotransmission and reward mechanisms (reviewed, e.g. in Uys andReissner (2011), Nyberg (2014)). These adaptations can be variable. For example, changes of N-methyl-D-aspartate receptor signalling induced by cocaine in different structures are heterogeneous and sometimes contradictory, mainly due to the complexity of the signalling pathways and the differences of experimental paradigms associated with cocaine administration, as summarised by Ortinski (Ortinski, 2014). A single dose of cocaine induces long-term potentiation in midbrain dopamine neurons (Ungless et al., 2001), and other studies revealed neuroplastic effects of acute cocaine exposure, indicating molecular, structural and functional changes of the mesolimbic system core (Kozell and Meshul, 2001, Grignaschi et al., 2004, Sarti et al., 2007, Friend et al., 2021. Adolescence is a sensitive period of life in which synapses undergo major maturation and fine-tuning events, reshaping brain connectivity and morphology with distinct regional variations, particularly in the cortex (Khundrakpam et al., 2016, Juraska andDrzewiecki, 2020). It is plausible that in this period, events that interfere with synaptic function (including drug exposure) significantly influence synaptic maturation. In animals, acute and repeated exposures to cocaine during adolescence cause changes of neurotransmission and neuroplasticity, affect the expression of trophic factors and synaptic elements, and induce immediate or persisting risks to cognition, behaviour and brain development (for review, see Caffino et al. (2021)).
The synaptic vesicle glycoprotein 2A (SV2A) commonly reveals synaptic density variations in the brain. Synaptic vesicle glycoproteins 2 constitute a class of transmembrane proteins of all synaptic vesicles (Buckley and Kelly, 1985), and the main isoform SV2A is expressed ubiquitously in all grey matter brain structures, albeit with regional variations (Bajjalieh et al., 1993, Varnäs et al., 2020. SV2A commonly serves as a marker of synaptic density in neuroimaging studies, although some questions remain of the interpretation of the imaging outcomes (Rossi et al., 2022). SV2A seems to have a number of different functions in neurons (for an overview, see Rossi et al. (2022)) but SV2A indubitably has crucial roles in the modulation of neurotransmission (Crowder et al., 1999, Bradberry andChapman, 2022) that make it an interesting target of the study of cocaine addiction. A number of specific tracers of SV2A exist, including UCB-J (Mercier et al., 2014) that proved to be optimally applicable to in vivo neuroimaging by means of positron emission tomography (PET) when radiolabeled with 11 C (Nabulsi et al., 2016). Postmortem imaging techniques such as autoradiography with [ 3 H]UCB-J also served to assess changes of synaptic SV2A density in preclinical studies (Binda et al., 2021, Raval et al., 2021, Thomsen et al., 2021. Here, we administered a single high dose of cocaine to rats during early adolescence, with the aim of proving if cocaine has effects on pre-synaptic SV2A density in different brain areas after the cocaine-induced boost of extracellular dopamine. We assessed changes of SV2A density in dopaminergic pathways by evaluating [ 3 H]UCB-J autoradiograms 1 hour and 7 days after cocaine administration. We did [ 3 H]GBR-12935 autoradiography at both times to verify the occupancy of DAT, as a marker of dopaminergic system functionality (as previously described, e.g. by Wong et al. (1998)) after the extinction of acute effects of cocaine.

Materials and methods
Animals. We conducted all animal procedures according to the FELASA guidelines for animal experimentation with permission from the Danish Animal Experiment Inspectorate (license number 2016-15-0201-01105) and reported experiments according to the ARRIVE guidelines. We used 28 male Sprague-Dawley rats (Taconic Biosciences, Denmark) of average post-natal day (PND) 35 at the time of experimentation with sacrifice on PND 35 or 42, roughly consistent with the peri-adolescence period (Sengupta, 2013). We chose animals of these ages according to previous literature of cocaine exposure in adolescent rats (Giannotti et al., 2015, Caffino et al., 2017, Caffino et al., 2018, Caffino et al., 2020. Before the experiment, the rats underwent 7 days of environmental habituation in a climate-controlled facility (12 h light/ 12 h dark cycle, average temperature of 22°C, average humidity of 55%) with ad libitum access to food and water.
After randomisation, we intraperitoneally injected a single dose of cocaine (20 mg/kg, N = 16) or saline (N = 12) as control. After the assessment of locomotion parameters to evaluate the success of the treatment, we randomly assigned the rats to two groups, each consisting of 14 animals, including cocaine-injected (N = 8) and saline-injected (N = 6) rats. We euthanised animals of the two experimental groups by decapitation at different time points, 1 hour or 7 days after single injection. We rapidly removed and froze the whole brains in powdered dry ice and stored them at −80°C.
Open field test (OFT). We performed OFT in a flat and empty arena with a surface of 200 × 200 cm 2 , divided into 4 equal quadrants with a surface of 100 × 100 cm 2 each and bordered by walls to prevent interactions between conspecifics during the test. We recorded the entire space of movement of the animals with a camera installed above the arena. We tested the animals in the open field for 30 minutes immediately after the cocaine or saline injection. We analysed the acquired videos by EthoVision XT® (Noldus) software. An observer blind to the treatment groups analysed the mean velocity (cm/s) and distance moved (cm).
Anatomy and cryostat sectioning. We sectioned fresh-frozen whole brains with a cryostat (Cryostar NX70, Thermo Scientific) and identified anatomical regions of interest according to the Paxinos and Watson rat brain atlas (Paxinos and Watson, 2004), including the medial prefrontal cortex (Bregma (B) 3.72 mm, Interaural (I) 12.72 mm), striatum and nucleus accumbens (B 1.80 mm, I 10.80 mm), amygdala (B −2.40 mm, I 6.60 mm), dorsal and ventral hippocampus (B −4.80 mm, I 4.20 mm). We cut the brains along the coronal plane, producing sequential brain sections of 20 μm-thickness. We mounted brain sections on poly-L-lysine coated microscope slides (Thermo Scientific). We mounted an average of six brain sections on each slide and stored them at −80°C.
[ 3 H]UCB-J labelling. We thawed the slides at room temperature for 1 hour and then pre-washed them for 15 minutes in Tris-HCl buffer (pH 7.4) containing 50 mM Trizma® base (Sigma) and milliQ water. We divided the slides into two groups to assess total binding (TB) and non-specific (NS) binding. For each animal, we used one slide for TB and one slide for NS for each brain area. We incubated the slides for 1 hour either in TB solution, containing 1 nM [ 3 H]UCB-J (Novandi Chemistry AB; molar activity 78 Ci/mmol) in Tris-HCl buffer, or NS binding solution containing 1 nM [ 3 H]UCB-J and 100 μM levetiracetam (Sigma-Aldrich) in Tris-HCl buffer. We post-washed the slides in Tris-HCl buffer and milliQ water and dried them at room temperature.
[ 3 H]GBR-12935 labelling. We used slides of the striatum to assess the availability of DAT as previously described (Stokholm et al., 2021). For each animal, we used one slide for TB and one slide for NS binding. We thawed the slides at room temperature for 1 hour and then incubated them in a buffer solution containing 50 mM Trizma® base (Sigma), 300 mM NaCl, 0.2% bovine serum albumin (Sigma), 1 μM cis-flupentixol (provided by H. Lundbeck A/S) and milliQ water. We then incubated the slides either in the TB solution containing radiolabeled 2 nM GBR-12935 (PerkinElmer; molar activity 40 Ci/mmol) in the buffer solution or in the NS binding solution containing 2 nM GBR-12935 and 1 μM GBR-12909 (Sigma). We incubated slides overnight at 4°C.
Autoradiography acquisition and analysis. We performed autoradiography acquisitions for 2 hours for [ 3 H]UCB-J and 22 hours for [ 3 H]GBR-12935 with the digital real-time autoradiography BeaQuant system (ai4r, France) (Donnard et al. 2009). An experimenter blind to the treatment groups analysed the autoradiograms with Beamage software (version 3.1.2, ai4r, France) using a pixel size of 100 μm. We manually defined regions of interest, respecting the interindividual anatomical differences, and we averaged values from symmetrical brain structures. We subtracted NS binding values (cp/min/mm 2 ) from TB values (cp/min/mm 2 ) to obtain specific binding values.
Statistical analysis. We statistically analysed data with GraphPad Prism® 9.4.1. The data passed the Shapiro-Wilk test for normality. We compared experimental groups (saline-treated vs cocaine-treated; sacrifice after 1 hour vs sacrifice after 7 days) with unpaired two-tailed t-tests. We considered p-values <0.05 as indicative of statistical significance, with graphical representation of significance by means of asterisks ((*p < 0.05, **p < 0.01, ***p < 0.001). The corresponding author will make complete data collections available upon reasonable request.

Discussion
As described, we tested if a single dose of cocaine administered during adolescence alters the density of the SV2A protein, evaluating both short-(1 hour) and long-(7 days) lasting effects. As cocaine raises dopaminergic neurotransmission (Kalivas, 2007, Zhu and Reith, 2008, Docherty and Alsufyani, 2021, and SV2A is involved in different aspects of the modulation of neurotransmission (reviewed in Rossi et al. (2022)), we tested the hypothesis that changed SV2A levels in brain areas targeted by dopaminergic pathways eventually would relate to synaptic adaptations incurred by the increased dopamine availability. Previous studies already demonstrated that a single dose of cocaine, administered at the same concentration used in this experimental work (20 mg/kg), is sufficient to elicit molecular, structural (Giannotti et al., 2015, Caffino et al., 2017, Caffino et al., 2018 or behavioural changes (Caffino et al., 2020) in early adolescent rats.
We assessed the successful administration of cocaine by testing the locomotion parameters by OFT. Psychostimulants, like cocaine, induce a sharp increase in dopamine neurotransmission with increased locomotor activity (reviewed, e.g. in Beninger (1983)). We performed the OFT for 30 minutes immediately following the injection as befits the timespan of dopamine raised by the effects of cocaine (Hurd and Ungerstedt, 1989). Cocainetreated rats moved with elevated mean velocity and distance compared to control rats. The cocaine-induced increase of locomotor activity previously was reported in the literature (e.g. Yeh and Haertzen (1991)), and early adolescence rats have been shown to have the greatest rise of motor activity following cocaine administration, compared to rats of other ages (Badanich et al., 2008).
We verified the activity of the dopaminergic system by means of [ 3 H]GBR-12935 autoradiography, showing that striatal [ 3 H]GBR-12935 binding was unchanged in cocaine-treated rats compared to controls at both times of analysis. Cocaine persistency in the dopaminergic system has been assessed by different studies; for example, Hurd and Ungerstedt (1989) showed that the striatal cocaine-induced rise of dopamine returns to baseline within 30 minutes from the administration. However, another study (Javaid and Davis, 1993) demonstrated that cocaine levels remain high 1 hour after intraperitoneal injection in different rat brain areas, meaning that cocaine occupancy of DAT may be expected at this time. The present [ 3 H]GBR-12935 autoradiography results suggest that at 1 hour, and 7 days after the treatment, cocaine no longer occupies DAT in the striatum and that dopaminergic transmission is normal at both times.
The [ 3 H]UCB-J autoradiography results imply no significant changes 1 hour after the treatment, but the SV2A density of cocaine-injected rats had increased compared to control animals 7 days after the administration in dorsal and ventral hippocampus (Table 1). It is noteworthy also to mention the trend towards an increase in SV2A density in amygdala 7 days after the treatment ( Table 1). These data suggest that a single administration of cocaine can provoke changes of SV2A density that are evident at least one week after the exposure, much beyond the acute effects of the drug. Since SV2A does not identify a specific agent of neurotransmission (Buckley and Kelly, 1985), it is difficult to associate the [ 3 H]UCB-J autoradiography outcomes with specific types of The data are presented as the mean ± standard deviation (SD); SD is graphically shown as vertical bars.

Acta Neuropsychiatrica 5
neurons and specific phenomena. However, cocaine is known to be associated with molecular and functional changes in synapses; for example, cocaine-facilitated neuroplasticity has been observed in several brain structures, including amygdala (e.g. Goussakov et al. (2006), Fu et al. (2007)) and hippocampus (e.g. Thompson et al. (2002), Thompson et al. (2004), Thompson et al. (2005), Therefore, we suggest the increased hippocampal SV2A density to be linked with cocaine-induced changes in synapses which may occur after a single drug exposure. However, to our knowledge, specific studies of the contribution of SV2A to drugfacilitated neuroplastic processes are lacking. The involvement of the hippocampus in onset, development and maintenance of addiction through drug-associated memory formation has already been hypothesised and investigated (reviewed by Kutlu and Gould (2016)). So, the presence of cocaine-induced changes in hippocampal synapses after a single dose may be linked with the formation of drug-associated memories and may relate to the onset of addiction in adolescence. However, we did not investigate long-term behavioural effects in the present study, so further studies are necessary to fully understand the particular molecular events that are attributable to these changes, and how SV2A may participate in cocaine-induced neuroadaptations. Comparisons with adult animals are also necessary to clarify if these effects are related to adolescence or if they extend to adulthood.
The lack of changes in [ 3 H]UCB-J binding in the medial prefrontal cortex of cocaine-injected rats compared to controls is worth a comment. The literature is equivocal about the effects of cocaine in this region. For example, some studies (e.g. Robinson and Kolb (1999), Muñoz-Cuevas et al. (2013)) reported increased dendritic spine density in rodent frontal cortex after repeated cocaine exposure. Caffino et al. (2018) showed that a single dose of cocaine with the same concentration of 20 mg/kg used here provoked a decrease of dendritic spine density and an impairment of post-synaptic elements of glutamatergic terminals in the medial prefrontal cortex of adolescent rats aged 35 days at the time of the injection, with effects detected 7 days after the drug exposure. A PET study (Angarita et al., 2022) of SV2A density in the medial prefrontal cortex of patients with cocaine use disorder revealed decreased [ 11 C]UCB-J binding in frontal cortices, but in this work, the subjects had a history of repeated cocaine exposures, different from the present experimental setup in which we analysed the effects of a single dose in rat. Overall, further studies are necessary to clarify if and how acute cocaine effectively can modify synapses of the medial prefrontal cortex with short-or long-term effects.
In conclusion, the present experimental work investigated both the immediate and lasting effects of a single high dose of cocaine on a measure of pre-synaptic SV2A density in male adolescent rats by using [ 3 H]UCB-J autoradiography. The results revealed an increased tracer binding in the dorsal and ventral hippocampus 7 days after the injection of cocaine compared to saline-injected rats. We emphasised the possible importance of the detected synaptic changes in hippocampus as possible early indicators of cocaine addiction development in adolescents, but further studies are needed to understand the ultimate causes of such changes.
Acknowledgements. Cis-flupentixol was kindly provided by Benny Bang-Andersen, H. Lundbeck A/S (Valby, Copenhagen). We thank Professor Paul Cumming, PhD student Shokouh Arjmand and Dr Giulia Monti for helpful discussions. Financial support. Financial support for RR's salary was provided by the Bjarne Saxof Fund administered through the Parkinsonforeningen to AML and the Erasmus Program to RR. Running costs were provided by Lundbeck Foundation, Parkinsonforeningen and the Translational Neuropsychiatry Unit (Aarhus University, Denmark).

Conflict of interest.
Gregers Wegener is the Editor-in-Chief of Acta Neuropsychiatrica at the time of submission but was not involved during the review or decision process of this manuscript.