Differential effects of deep brain stimulation on reinstatement of cocaine seeking in male and female rats

.


Introduction
Cocaine use disorder, which lacks FDA-approved treatment options, is defined by cyclical patterns of drug use, abstinence, and relapse.Although the prevalence of cocaine use is higher in men (SAMHSA 2021), women demonstrate more rapid progression to problematic cocaine use and an increased likelihood of relapse [18,25,31].Women also experience higher levels of cocaine craving [4], and levels of craving can vary across the menstrual cycle [33].Clinical and preclinical substance use disorder studies have historically focused primarily on males.Inclusion of females is clearly necessary to generate and evaluate potential treatment options for cocaine use disorder.
One promising therapeutic already in use for several neuropsychiatric disorders is deep brain stimulation (DBS).Interestingly, sex differences in the effect of DBS have been observed such that fewer women than men respond to DBS treatment for obsessive compulsive disorder [24] but women respond to a greater extent to DBS for treatmentresistant depression [26].DBS is also under exploration to curb drug craving [28,35,39].Cocaine craving, a key precipitator of relapse, can be modeled in rodents using the cocaine self-administration/ reinstatement paradigm.DBS of the nucleus accumbens shell attenuated cocaine-primed reinstatement of cocaine seeking in male rats following either extinction [37,38] or a period of forced abstinence [9].Accumbens DBS also suppressed cue-evoked cocaine reinstatement [7].However, delivery of DBS after the escalation of cocaine selfadministration actually increased cocaine-intake [13].Preclinical studies investigating the effect of DBS on drug seeking, including our own, have primarily been performed in males.We recently demonstrated that mimicking DBS selectively in neuronal populations of the nucleus accumbens shell using optogenetic stimulation attenuated cocaine seeking in male, but not female, rats [34,36].Whether electrical DBS differentially alters cocaine seeking in female vs. male rats has not yet been explored.
A rich literature has identified consequential sex differences in cocaine self-administration and reinstatement [2,16].Female rats display faster acquisition of cocaine self-administration [22,32] and increased motivation for cocaine compared to their male counterparts [10,40].Cocaine-primed reinstatement of cocaine seeking behavior is also potentiated in female relative to male rats [21].Estrous cycle phase can drive differential cocaine behaviors in female rats.Studies show greater reinstatement of cocaine seeking behavior when female rats are in the estrus phase of their estrous cycle, a time when the level of circulating estradiol is high and progesterone is low [5,14,15].Female rats exhibit more protracted cocaine seeking across extended periods of abstinence, which is exacerbated in cocaine-primed reinstatement tests in females evaluated in estrus but not non-estrus phases [14].Estrogen also heightens the acquisition of cocaine self-administration and the reinstatement of cocaine seeking behavior in rats, while progesterone attenuates such effects [5,11,23].The present study is designed to determine if the effect of DBS on cocaine reinstatement in female rats is dependent on the estrus cycle phase.

Animals and housing
Adult female and male Sprague Dawley rats weighing 250-300 g were obtained from Taconic Laboratories (Germantown, NY).Rats were single housed in individually ventilated cages in the same colony room on a 12 h light/dark cycle with lights on at 6:00 A.M. All experimental procedures were performed during the light cycle.Food and water were available ad libitum.All experiments were conducted according to National Institute of Health guidelines and approved by Rutgers Institutional Animal Care and Use Committee.

Drugs
Cocaine hydrochloride was obtained from the National Institute on Drug Abuse and dissolved in 0.9 % sterile saline.

Materials
All experiments used Med Associates (East Fairfax, VT, USA) operant chambers enclosed within ventilated, sound-attenuating chambers.The apparatus was equipped with two response levers (active and inactive), stimulus lights, food pellet dispensers, and injection pumps for intravenous drug delivery through tubing encased in a rotating swivel.

Surgery
Rats were anesthetized with 80 mg/kg ketamine and 12 mg/kg xylazine (i.p.) prior to surgery.A silastic catheter (Strategic Applications Inc.; Libertyville, Il) was fed subcutaneously over the shoulder and placed into the right jugular vein and sutured in place.The catheter was attached to a mesh platform secured dorsally between the shoulder blades with an exterior port, which was sealed with a plastic cap when not in use.Catheters were flushed daily with antibiotic (Timentin, 0.93 mg/ml) dissolved in heparinized saline.Immediately following the completion of the catheter surgery, rats were mounted in a stereotaxic apparatus (Kopf Instruments; CA, USA).Bipolar stainless steel electrodes (P1 Technologies; Roanoke, VA, USA) were bilaterally implanted into the nucleus accumbens shell according to the following coordinates, relative to bregma [27]: +1.0 mm anteroposterior, ±3.0 mm mediolateral, − 7.3 mm dorsoventral, 17 • angle.Electrodes were held in place by affixing dental acrylic to three steel screws fastened to the skull cap.

Cocaine self-administration and extinction
Following a 7-day recovery period, rats were placed in operant chambers where they were allowed to lever press for intravenous cocaine infusions (0.254 mg of cocaine dissolved in 59 µL of sterile saline) daily for two hours.A 20 s timeout period during which responses had no scheduled consequences followed each cocaine infusion.Rats were limited to 30 infusions per 2-hour cocaine self-administration session.The self-administration period lasted for a total of 21 days.Rats began on a fixed ratio 1 (FR1) schedule until they attained stable responding (<15 % variation in responses for three consecutive days) at which point they were switched to a FR5 schedule for the remainder of the experiment.Rats were switched to an FR5 schedule no later than day 12 of self-administration to allow a minimum of three days on this schedule prior to tether acclimation.Beginning on day 15 of selfadministration and daily throughout the remainder of the experiment, electrodes were tethered to dummy cables to acclimate rats to this procedure before reinstatement testing.After a total of 21 days, rats underwent extinction and reinstatement sessions during which cocaine was replaced with saline.Daily 2-hour extinction sessions continued until responses were < 20 % of the response rate during selfadministration and they achieved stable responding for two consecutive days.

Reinstatement and deep brain stimulation
Following 21 days of self-administration, female rats were assigned to undergo reinstatement testing in either the estrus or non-estrus phases, balanced for performance over the last three cocaine selfadministration sessions.Cocaine seeking was assessed when the extinction criterion was met, and in female rats during one predetermined estrous phase.Immediately prior to each reinstatement session, rats were given a priming injection of cocaine (10 mg/kg, i.p.) to reinstate cocaine seeking behavior.Rats were then tethered to electrical cables and placed in the operant chamber for a 2-hour reinstatement session, during which 0 µA (sham) or DBS at 150 µA stimulation intensity and 160 Hz frequency with a 60 µs pulse width was delivered through intra-accumbens shell electrodes, as previously described [7,8,37,38].DBS was delivered for the entirety of the 2-hour reinstatement session.Active lever presses resulted in an infusion of saline rather than cocaine.Each rat underwent a DBS and sham reinstatement test, in a counterbalanced within-subjects design.Extinction of lever pressing was re-established between reinstatement by daily extinction sessions until responding was again < 20 % of the response rate maintained by cocaine self-administration.

Vaginal swabbing and determination of estrous phase
Vaginal swabs were collected weekly during cocaine selfadministration to acclimate female rats to this procedure.To verify estrous cycle phase during extinction and reinstatement testing, vaginal swabs were collected the day before, day of, and day after reinstatement test sessions.All swabbing occurred at the same time in the morning, prior to the session.A cotton-tipped swab was dipped in sterile water, then rolled against the vaginal wall.The swab was then transferred to a dry glass slide.Samples were stained with methylene blue dye and viewed under a light microscope for identification.Estrous phase was determined by cellular composition, including proportion of leukocytes, cornified epithelial cells and nucleated epithelial cells.

Verification of electrode placement
After the experiment was completed, rats were euthanized with CO 2 inhalation.The brain was removed, fixed in 10 % formalin and coronally sectioned (100 μm) at the level of the nucleus accumbens with a VT1000S vibratome (Leica, Weltzar, Germany).Electrode placement was verified under a light microscope.Only rats that maintained catheter patency throughout the duration of cocaine self-administration and had correct bilateral electrode placement in the nucleus accumbens shell were included in the analyses.

Statistical analysis
Statistical analysis was performed in Prism 10.0 with alpha set at p < 0.05.Self-administration data were analyzed with two-way mixed model ANOVAs with group (estrus females, non-estrus females, and males) as the between-subjects factor and day as the within-subjects factor.Pairwise analyses were made with Bonferroni post-tests (p < 0.05).Extinction duration was analyzed by two-way mixed model ANOVAs with group (estrus females, non-estrus females, and males) as the between-subjects factor and test number (extinction sessions prior to the first or second reinstatement session, which were counterbalanced sham and DBS tests) as the within-subjects factor.Pairwise analyses were made with Bonferroni post-tests (p < 0.05).Lever presses during extinction and reinstatement were analyzed by three-way mixed measures ANOVA with group as the between-subjects factors and with session type and stimulation as the repeated measures factors.A significant three-way interaction was followed by individual two-way repeated measures ANOVAs in each group.Post-hoc analyses were made with Fisher's LSD tests.

Cocaine self-administration was similar in estrus female, non-estrus female, and male rats
Rats lever-pressed for intravenous cocaine for 21 daily 2-hour sessions.To acclimate rats to the tethers used to deliver DBS, rats were attached to dummy tethers beginning on day 15 of cocaine selfadministration.Female rats were balanced into groups and assigned to subsequent reinstatement testing in a designated estrous cycle phase based on lever pressing over the final 3 days of cocaine selfadministration.Two-way mixed measures ANOVA revealed that there was a main effect of day (F 40, 499 = 72.45,p < 0.0001), no main effect of group (F 2,25 = 0.91, p = 0.2127), and a day x group interaction (F 40,499 = 1.63, p = 0.0105) for active lever presses across the selfadministration phase (Fig. 1A).Bonferroni post-hoc analysis indicated that although there were no sustained differences across days of cocaine self-administration, lever presses were higher in females designated for reinstatement testing in non-estrus phases than males on selfadministration day 6, and higher in males than non-estrus females on day 19 (p < 0.05).However, two-way mixed measures ANOVA revealed that there was a main effect of day (F 40, 499 = 3.94, p = 0.0037), no main effect of group (F 2,25 = 1.58, p = 0.2254), and no day x group interaction (F 40,499 = 1.23, p = 0.1611) for cocaine infusions earned across the self-administration phase (Fig. 1B).Thus, under the experimental parameters of this study, there was no effect of sex on cocaine intake, and no difference in cocaine self-administration among female rats that were subsequently designated for reinstatement testing in the estrus phase or non-estrus phases.

Days to reach extinction criterion was similar in estrus female, nonestrus female, and male rats
Rats underwent extinction training, where cocaine was replaced with saline, to extinguish lever responding to less than 20 % of the response rate maintained on an FR5 schedule over the last three days of cocaine self-administration.Extinction baseline was defined as average active lever responses on the two days prior to a cocaine-primed reinstatement session, during which rats would either receive sham stimulation or DBS.To determine whether there was a sex or estrous cycle effect on the duration of extinction training required, the number days to reach extinction criteria prior to each reinstatement test were analyzed.Two-way mixed measures ANOVA revealed that there was a main effect of test number (first or second test, which were counterbalanced sham and DBS tests)(F 1,25 = 16.85,p = 0.0004), no main effect of group (F 2,25 = 1.64, p = 0.2138), and no test number x group interaction (F 2,25 = 1.15, p = 0.3323) for non-cumulative days to reach extinction criteria (estrus females: test 1 = 12.17 ± 1.35, test 2 = 2.25 ± 0.18; non-estrus females: test 1 = 12.67 ± 4.72, test 2 = 5.67 ± 1.54; males: test 1 = 6.57± 0.65, test 2 = 2.86 ± 0.86).Fewer extinction sessions required prior to the second reinstatement test reflects that extinction baseline is reestablished, whereas extinction sessions prior to the first reinstatement sessions also include extinction learning.Thus, females and males similarly extinguished lever responding.Male rats underwent reinstatement testing when they met extinction criteria; additional extinction sessions were sometimes required to ensure female rats were evaluated for cocaine seeking in the predetermined estrous cycle phase.Two-way mixed measures ANOVA revealed a main effect of test number (F 1,25 = 5.52, p = 0.0270), a main effect of group (F 2,25 = 4.31, p = 0.0246), and no test number x group interaction (F 2,25 = 0.34, p = 0.7125) for extinction sessions prior to each reinstatement session (estrus females: test 1 = 17.75 ± 3.52, test 2 = 8.5 ± 1.82; non-estrus females: test 1 = 13.44 ± 4.70, test 2 = 7.0 ± 2.14; males: test 1 = 6.57± 0.65, test 2 = 2.86 ± 0.86).Female rats evaluated for cocaine seeking during the estrus phase underwent more extinction sessions than males, but not females tested in non-estrus phases, prior to the first test (Bonferroni, p < 0.05).This difference reflects an artifact of the experimental design which tested female rats in a balanced, predetermined Fig. 1.Cocaine self-administration was similar in male rats and female rats subsequently evaluated for cocaine seeking in estrus or non-estrus phases.A) Mean (±SEM) active lever presses over 21 days of cocaine self-administration by male rats (n = 7) and female rats subsequently tested for cocaine seeking in estrus (n = 12) or non-estrus phases (n = 9) did not differ.B) Mean (±SEM) cocaine infusions earned over 21 days of cocaine self-administration by male rats and female rats tested in estrus or non-estrus phases did not differ.Arrows denote the beginning of tethering electrodes to dummy cables on Day 15 and during each subsequent self-administration and extinction session to acclimate rats prior to reinstatement testing.phase of the estrous cycle, rather than a difference in extinction learning.

DBS of the nucleus accumbens shell attenuates cocaine seeking in male, but not female, rats
Cocaine seeking, operationally defined by active lever responding, was reinstated by administration of a cocaine prime (10 mg/kg, i.p.).Deep brain stimulation (150 µA) was administered bilaterally to the nucleus accumbens shell or rats received sham stimulation (0 µA) throughout the reinstatement session in a counterbalanced withinsubjects design.Three-way mixed measures ANOVA revealed a significant group (estrus females vs. non-estrus females vs. males) by session type (extinction vs. reinstatement) by stimulation (DBS vs. sham) interaction (F 2,25 = 4.89, p = 0.0160; Fig. 2A).For estrus females, twoway repeated measures ANOVA revealed that there was a main effect of session type (F 1,11 = 17.85, p = 0.0014), no main effect of stimulation (F 1,11 = 1.29, p = 0.2799), and no session type by stimulation interaction (F 1,11 = 2.19, p = 0.1667; Fig. 2A).For non-estrus females, two-way repeated measures ANOVA revealed a main effect of session type (F 1,8 = 17.37, p = 0.0031), no main effect of stimulation (F 1,8 = 0.003p = 0.9563), and no session type by stimulation interaction (F 1,8 = 0.089, p = 0.7725; Fig. 2A).For males, two-way repeated measures ANOVA revealed a main effect of session type (F 1,6 = 19.19,p = 0.0047), a main effect of stimulation (F 1,6 = 6.30p = 0.0459), and a session type by stimulation (F 1,6 = 6.15, p = 0.0479; Fig. 2A).Thus, all groups displayed reinstatement of cocaine seeking indicated by significantly higher active lever responding during reinstatement tests relative to the extinction baseline established by the two extinction sessions immediately prior to each reinstatement test.However, cocaine seeking was significantly attenuated by DBS relative to sham stimulation only in male rats (Fisher's LSD, p < 0.05).
Electrode placement verification for estrus females (Fig. 2B), nonestrus females (Fig. 2C), and males (Fig. 2D) are shown.Overall, DBS of the nucleus accumbens shell attenuated cocaine seeking in male rats, similar to our prior studies [37,38].Surprisingly, DBS of the nucleus accumbens shell did not alter cocaine seeking in female rats tested either during estrus or non-estrus phases, indicating that the effect of DBS to attenuate cocaine seeking may be sex specific.

Discussion
The present study highlights a previously unknown sex difference in the efficacy of electrical DBS to attenuate cocaine seeking in rats.Administration of DBS to the nucleus accumbens shell significantly attenuated cocaine-primed reinstatement of cocaine seeking in male rats, as previously described [37,38].In stark contrast, DBS of the Fig. 2. Deep brain stimulation of the nucleus accumbens shell attenuated cocaine-primed reinstatement of cocaine seeking in male rats, but not female rats evaluated in estrus or non-estrus phases.A) Mean (±SEM) active lever presses during extinction baseline sessions immediately prior to each reinstatement test and counterbalanced reinstatement sessions during which rats received sham stimulation or high frequency DBS.Active lever responding was higher in reinstatement sessions than the preceding two extinction sessions for sham or DBS tests and across estrus females, non-estrus females, and males, indicating that cocaine seeking was sufficiently reinstated by a cocaine priming injection.DBS significantly attenuated cocaine seeking in male rats (*p < 0.05), but not female rats that were evaluated for cocaine seeking in estrus or non-estrus phases.B-D) Placement of electrodes in the nucleus accumbens shell for B) estrus females, C) non-estrus females and D) males.nucleus accumbens shell did not suppress cocaine seeking in female rats, consistent with the failure of DBS-like optogenetic stimulation of specific accumbens medium spiny neuron subpopulations to alter cocaine seeking in female rats [34,36].Estrous cycle phase is known to influence cocaine-related behaviors in female rats such that reinstatement of cocaine seeking may be higher in the estrus phase vs. non-estrus phases of the estrous cycle [5,14,15].However, nucleus accumbens shell DBS was similarly ineffective at altering cocaine seeking in female rats when they were tested in estrus and non-estrus phases.Overall, these results suggest that the ability of nucleus accumbens shell DBS to suppress cocaine seeking is specific to male, but not female, rats.
The mechanisms by which accumbens DBS attenuates cocaine seeking in male rats but fails to do so in female rats are unclear.Inherent differences in neuronal excitability among female and male rats may contribute to this discrepancy.Females rats display elevated excitability compared to males in a number of brain regions, including the nucleus accumbens shell [12] and core subregions [30] as well as regions that send glutamatergic afferents to the accumbens such as the ventral hippocampus [41] and prefrontal cortex [17].Importantly, female mice display inherent hyperexcitability in the nucleus accumbens shell, which is associated with a significantly lower response to glutamate release from infralimbic cortex projections than their male counterparts [12].It is possible that hyperexcitability in the nucleus accumbens shell in females contributed to the lack of DBS to suppress cocaine seeking, perhaps via the occlusion of neuroplasticity in this region.Cocaineevoked changes in neuroplasticity could also be dependent on sex and influence the ability of DBS to alter cocaine reinstatement.For example, low frequency (12 Hz) stimulation elicited long-term potentiation in nucleus accumbens medium spiny neurons in both male and female experimentally-naïve rats; however, in cocaine-experienced rats, this stimulation pattern failed to evoke long-term potentiation in females and only evoked long-term potentiation in D2DR-containing neurons in males [36].
Although not yet explored, high frequency accumbens DBS may elicit similar sex-dependent changes in neuroplasticity which govern cocaine seeking.
Both the reinstatement of cocaine seeking and nucleus accumbens neuronal excitability are influenced by gonadal hormones, which vary across phases of the estrous cycle.Cocaine seeking behavior is highest during the estrus phase, when circulating estrogen levels are high and progesterone levels are low [5,14,15].In contrast, there is an inverse relationship between progesterone and cocaine seeking [6].Estrogen also heightens the acquisition of cocaine self-administration and the reinstatement of cocaine seeking behavior in female rats, while progesterone attenuates such effects [5,11,23].These effects are attributable to gonadal hormones.In ovariectomized females, experimenterdelivered estradiol potentiates cocaine-primed reinstatement [19,20], which is dampened by co-administration of progesterone [1].Discrete phases of the estrous cycle are also associated with altered neuronal excitability in the nucleus accumbens.Specifically, electrophysiological recordings from the accumbens core show that properties associated with extrinsic excitatory input are increased and intrinsic excitability is dampened during proestrus and estrus phases relative to the diestrus phase [30].Estradiol replacement in ovariectomized rats reduced excitability in medium spiny neurons in the nucleus accumbens core, consistent with rats evaluated in the estrus phase, but did not influence the response to excitatory synaptic inputs [29].Recent findings show that the effect of cocaine on neuroplasticity in the nucleus accumbens shell depends on estrous cycle phase and is distinct in D1DR-containing vs. D2DR-containing medium spiny neurons.In contrast to males and females in diestrus, which have lower excitability in D1DR-vs.D2DRneurons when treated with saline, neuronal excitability is similar in D1DR-and D2DR-containing neurons when females are in the estrus phase [3].When excitability was evaluated following abstinence from repeated experimenter-delivered cocaine, males exhibited greater hypoexcitability in D1DR-neurons, there was no change in diestrus females, and estrus females actually expressed enhanced excitability in D2DR-containing neurons [3].The cocaine-mediated increase in glutamatergic synaptic activity was sex specific in that it occurred in both neuronal subtypes in males but was selective to D2DR-neurons in females, regardless of estrous cycle phase [3].Clearly, the effects of gonadal hormones and cocaine on accumbens neuronal excitability are complex.Disentangling these interactions and gleaning a deeper understanding of the way DBS influences excitability and plasticity within the nucleus accumbens shell may provide insight into the sex-specific mechanisms of DBS to reduce cocaine seeking in rats.
There are limitations of the present study that should be considered.The experiments were performed identically to our prior DBS studies in male rats [8,37,38].It is possible that factors including cocaine dose or DBS parameters (e.g., frequency, amplitude, timing or length of delivery) may need to be optimized for female rats.Female and male rats were procured in the same weight range and self-administered similar amounts of cocaine (Fig. 1), but the dose of cocaine was not weight adjusted throughout the study.It is unlikely small differences in the unit dose of cocaine (mg/kg/infusion) drive the differential effect of DBS on cocaine seeking given that DBS and sham reinstatement tests are performed within the same rats.We measured estrous cycle phase and tested female rats for cocaine reinstatement during the same phase for both sham and DBS tests; whether manipulating gonadal hormone levels specifically would modulate the effect of DBS on cocaine seeking is not known.Because female rats were evaluated for cocaine seeking during a predetermined phase of the estrous cycle, females tested in the estrus phase had more extinction sessions than male rats that underwent reinstatement testing as soon as they met extinction criteria.This concern is somewhat ameliorated by using a within-subjects design for sham and DBS testing.Previous work indicated that reinstatement of cocaine seeking is more robust in estrus vs. non-estrus females [5,14,15].We did not observe this effect, which may be due to technical details such as the tethering of intracranial electrodes to cables and repeated counterbalanced reinstatement testing.In the present study, female rats were grouped into estrus and non-estrus phases of the estrous cycle, rather than distinguishing among each of the non-estrus phases − metestrus, diestrus, and proestrus.Since the estrus phase exerts the largest effect on cocaine reinstatement in female rats [5,14,15], it seems unlikely that this design obfuscated nuanced differences the effect of DBS on cocaine seeking across non-estrus phases.
Support for DBS as a potential treatment for substance use disorders is growing [35].Case studies and clinical trials investigating DBS as substance use disorder treatment and which include women are limited, and there have been no systematic efforts to explore sex differences in the efficacy of DBS for substance use disorders in humans.Although caution should be taken when translating preclinical findings to clinical implications, our data indicate that clinical studies designed to specifically determine whether DBS is an effective treatment for cocaine use disorder in women are necessary.

Significance Statement
Deep brain stimulation of the nucleus accumbens attenuated cocaine seeking in male but not female rats, regardless of estrous cycle phase, indicating there may be an important sex difference in the efficacy of DBS treatment for substance use disorders.