Rats

In the sexual behavior experiment (carried out in Tromsø), 36 subject male Wistar rats, 10 stimulus female Wistar rats, and 4 stimulus male Wistar rats were acquired from Janvier Labs (France), and were pair housed (unless otherwise indicated) in Macrolon IV® cages on a reversed 12 h light/dark cycle (lights on between 23:00 and 11:00) in a room with controlled temperature (21 ± 1 °C) and humidity (55 ± 10%), with ad libitum access to standard rodent food pellets and tap water throughout the experiment. In the sucrose self-administration experiment (carried out in Utrecht), a total number of 32 adult male Wistar rats, acquired from Charles River (Sulzfeld, Germany), was used. They were pair housed until brain surgery, and individually housed for the rest of the experiment in Macrolon III sawdust bedded cages on a reversed 12 h light/dark cycle (lights on between 07:00 and 19:00) in a room with controlled temperature (21 ± 2 C° and humidity (50–70%).

This study only investigated male behavior, because males and females show different kinds of sexual behaviors making it difficult to compare. We hope that future studies will address the same questions in female rats.

Viral constructs and drugs

For sexual behavior, conducted at UiT The Arctic University of Norway (NO), AAV5-CaMKIIa-hM4D-mCherry (Gi; inhibitory DREADD, University of North Carolina Vector Core, Chapel Hill, USA), AAV5-CaMKIIa-hM3D-mCherry (Gq; stimulatory DREADDs, University of North Carolina Vector Core, Chapel Hill, USA) and AAV5-CaMKIIa-EYFP (Sham; no DREADDs, University of North Carolina Vector Core, Chapel Hill, USA) were used in this experiment. For sucrose self-administration, conducted at University of Utrecht (NL), the same viral constructs were used as for sexual behavior, but now obtained from Addgene, USA (pAAV5-CaMKIIa-hM4D(Gi)-mCherry and pAAV5-CaMKIIa-hM3D(Gq)-mCherry were gifted from Bryan Roth (Addgene viral prep # 50477 and # 50,476; pAAV-CaMKIIa-mCherry was a gift from Karl Deisseroth (Addgene plasmid # 114,469). Viral constructs were used as is or diluted in saline before use, with a final viral titer between 1.4 and 4.3 × 1012 vg/mL. Clozapine N-oxide (CNO) (BML-NS-105; Enzo Life Sciences, Farmingdale, USA) was dissolved in ddH2O at a concentration of 1 mg/mL and kept at -20 °C until use. For experiments, rats were injected intraperitoneally with 1 mL/kg of the 1 mg/mL CNO solution or vehicle (ddH2O). Silastic capsules (medical grade Silastic tubing, 0.0625 in. inner diameter, 0.125 in. outer diameter; Degania Silicone, Degania Bet, Israel) for females were 5 mm long and contained 10% 17β-estradiol (E8875, Sigma, St. Louis, USA) in cholesterol (C3292, Sigma, St. Louis, USA). The silastic tubing was closed off by inserting pieces of toothpick into both ends and sealed with medical grade adhesive silicone (A-100; Factor II Incorporated, Arizona, USA). Progesterone (P0130, Sigma, St. Louis, USA) was dissolved in peanut oil (Apotekproduksjon, Oslo, Norway) at a concentration of 5 mg/mL.

Surgical proceduresBrain surgerySexual behavior experiment

Brain surgery consisted of subsequent bilateral infusions of the viral vector solution into the BNST. Subject male rats were anesthetized with isoflurane and placed in a stereotaxic apparatus (Stoelting Europe, Ireland). A 10 uL Hamilton syringe with a 30G blunt needle was mounted in a Hamilton injector (Stoelting) on the stereotaxic apparatus and inserted into each brain hemisphere sequentially at AP -0.4 mm, ML ± 3.80 mm, and DV -6.60 mm in reference to bregma, at a 20° angle with respect to the DV-axis, inserting the needle tip in the coronal plane from the lateral aspect toward the medial aspect of each hemisphere. Per hemisphere, 250 nL of the viral construct solution was injected into the BNST at an infusion rate of 150 nL/min. Following infusion, the needle was left in place for 10 min before withdrawal and closing of the skin with a continuous intradermal suture (Vicryl Rapide 40, Ethicon, Cincinnati, USA). After surgery, rats were immediately pair housed. Analgesic treatment consisted of 0.05 mg/kg buprenorphine and 2 mg/kg meloxicam pre-operative and post-operative after 24 and 48 h.

Sucrose self-administration experiment

The same volume of the same viral constructs was infused bilaterally into the BNST at the same coordinates during this experiment. Anesthesia consisted of a mixture of 75 mg/kg ketamine hydrochloride (Narketan 10%; VETOQUINOL, France ) and 0.25 mg/kg dexmedetomidine (Dexdomitor; Pfizer Animal Health B.V., the Netherlands) injected intraperitoneally. The pressure injection was conducted using a 30G blunt cannula needle mounted in a holder on the stereotaxic apparatus (Kopf Instruments), attached to a piece of tubing (Plastics One, Roanoke, USA) connected to a Hamilton syringe mounted in a minipump. After viral construct infusions, a double guide cannula (26G, length 8.2 mm, ML distance 1.6 mm; Plastics One, Roanoke, USA) was implanted at AP -5.05 mm, ML ± 0.80 mm, DV -7.15 mm, targeting the ventral tegmental area, for intracerebral infusion of CNO to target the BNST→VTA projection. The guide cannula was fixed to the skull using stainless steel screws and surgical cement (Simplex™ P bone cement with tobramycin, Stryker Nederland B.V., The Netherlands). Anesthesia was terminated through subcutaneous administration of 1.0 mg/kg atipamezole (Antisedan; Pfizer Animal Health B.V., the Netherlands). After post-experimental brain processing revealed that the VTA guide cannulas were not in the target area. Therefore, the intracranial CNO infusion data were not further analyzed and omitted from this manuscript. After surgery, rats were single housed in individually ventilated cages for 7 days, and subsequently single housed in regular cages for the remainder of the experiment. Analgesic treatment consisted of 0.05 mg/kg buprenorphine and 2 mg/kg meloxicam pre-operative and post-operative after 24 and 48 h.

Ovariectomy

Stimulus females for the sexual behavior experiment were ovariectomized under isoflurane anesthesia as previously described (Ågmo 1997). Briefly, a medial dorsal incision of the skin of about 1 cm, and small incisions in the muscle layer on each side, were made. The ovaries were located and extirpated, and the muscle layer sutured. A silastic capsule containing β-estradiol was placed subcutaneously. The skin was closed with a wound clip. Females were allowed at least one week of recovery before use in behavioral tests. Analgesic treatment consisted of 0.05 mg/kg buprenorphine and 2 mg/kg meloxicam pre-operative and post-operative after 24 and 48 h.

Behavioral assessmentSexual incentive motivation

The sexual incentive motivation (SIM) test (see also: (Huijgens et al. 2023) apparatus is a rectangular arena (100 × 50 × 45 cm) with rounded corners. At each long side, in opposite corners, a closed incentive stimulus cage is attached to the arena and separated from the arena by wire mesh (25 × 25 cm). Testing always takes place in a dimly lit (ca. 5 lx) room. During testing, a social stimulus (intact male rat) was placed in one of the stimulus cages and a sexual stimulus (receptive female rat; ovariectomized, implanted with a 10% 17β-estradiol silastic capsule, and injected with 1 mg progesterone 4 h before use in the test (standard procedure in our laboratory (Huijgens et al. 2021a, c; Ågmo and Snoeren 2017) was placed in the other stimulus cage. The position of the stimulus cages and the stimulus rats were randomly changed throughout each experimental session. To start the test, the subject male was introduced to the middle of the arena and then video-tracked by Ethovision software (Noldus, Wageningen, the Netherlands) for 10 min. Virtual stimulus zones in (30 × 21 cm) were defined in the arena in front of each stimulus cage in Ethovision. The subject male was considered to be in one of the zones whenever its point of gravity was. Frequency of entering, and time spent in each of the zones, distance moved, and mean velocity, were output measures of Ethovision. The preference score was calculated by dividing the time spent in the female incentive zone by the total time spent in incentive zones. Between tests, feces and urine were removed from the arena, and the SIM arena was cleaned with diluted acetic acid between experimental days.

Copulation

The copulation test was conducted in rectangular boxes (40 × 60 × 40 cm) with a Plexiglas front, in a dimly lit (ca. 5 lx) room. At the time of testing, the male subject was allowed 5 min of habituation to the copulation box, after which a receptive female (injected with 1 mg progesterone 4 h before use) was introduced, which started the 30 min test. All copulation tests were recorded on camera and behavior was later assessed from video. Behavioral assessment consisted of scoring behavioral events in Observer XT software (Noldus, Wageningen, the Netherlands). For the entire 30 min, 100% of the elapsed time was behaviorally annotated, similar to previous experiments (Huijgens et al. 2021a, c). The ethogram consisted of the copulatory behaviors mount (pelvic thrusting while being mounted on the female), intromission (mounting the female with pelvic thrusting and penile insertion into the vagina; characterized by a more vigorous dismount) and ejaculation (characterized by a longer intromission and the female moving away from the male), as well as genital grooming (autogrooming of the genital region), other grooming (autogrooming in other regions than the genitals), chasing (running after the female), anogenital sniffing (sniffing the anogenital region of the female), head towards female (head oriented in the direction of the female while not engaging in other behavior, also includes sniffing the female), head not towards female (any behavior that is not oriented towards the female except grooming, e.g. walking, sniffing the floor, standing still with head direction away from female). From these data points, the following outcome measures were determined: number of ejaculations, number of mounts, intromissions, and mount bouts, intromission ratio (number of intromissions divided by number of mounts + intromissions), average mounts and intromissions per mount bout, mean time-out duration, post-ejaculatory interval duration (time from ejaculation to first mount or intromission), latency to ejaculation (time from first mount or intromission to ejaculation), latency to first behavior (time from start of the test to first mount or intromission), and percentage of time spent on behaviors. The outcome measures were calculated for the first ejaculation series (referred to as “Series 1”), except for the total number of ejaculations in the 30 min test. A mount bout is defined as “a sequence of copulatory behaviors (one or more), uninterrupted by any behavior (other than genital autogrooming) that is not oriented towards the female” (Huijgens et al. 2021b; Sachs and Barfield 1970). Mount bouts were identified by reviewing the events between each copulatory behavior (i.e. mount or intromission) using a python script (available upon request). Whenever “other grooming” or “head not towards female” occurred in between copulatory behaviors, this marked the end of the previous mount bout (end time was then set on the end of the last copulatory behavior) and the beginning of the next mount bout (start time of the next copulatory behavior), and the time in between these mount bouts as a time out. Time spent on non-copulation oriented behavior was defined as time spent on (non-genital) grooming + head not towards female.

Operant sucrose self-administrationApparatus

Subject male rats were trained and tested in operant conditioning chambers (29.5 × 24 × 25 cm, Med Assiociates Inc., USA) in light- and sound-attenuating boxes with a metal grid floor (bars 1.57 cm apart), a white house light (28 V, 100 mA) and a ventilation fan, controlled by Med-PC IV software (version 4.2). The chambers were equipped with two retractable levers (4.8 × 1.9 cm), a white cue light (28 V, 100 mA) above each lever, and a sucrose pellet receptacle (equipped with an infrared beam for nose poke detection) underneath a sucrose pellet dispenser in between the two levers. One lever was designated as “active”, responding on which was reinforced with a sucrose pellet (45 mg; TestDiet, USA), the other lever was designated as “inactive”, responding on which was recorded but initiated nothing. The position of the active and inactive levers was counterbalanced between rats and within groups.

Fixed ratio 1

Subject male rats were trained to respond for sucrose pellets during 30-minute operant sessions, once daily, 2–4 times a week. All rats had a minimum of 12 training sessions before testing. Some rats received 1 or 2 extra training sessions to achieve stable active lever pressing performance. Performance was considered stable at group level when the average number of active lever presses on individual days fell within a 90-110% variability range of the total average of active lever presses over the last 3 training days. During training and testing, pressing the active lever turned on the cue light, dispensed a single sucrose pellet into the receptacle, retracted the levers, and turned off the house light. As soon as the animal nose-poked into the receptacle to retrieve the sucrose pellet, detected by interruption of the infrared beam, a new trial was started. Upon the start of a new trial, the levers were reintroduced, the cue light was turned off, and the house light was turned on again. The software output contained timestamps of each active lever response, inactive lever response, and first nosepoke after active lever response. This enabled analysis of response frequencies as well as latencies.

Progressive ratio

After FR1 training and testing, rats were trained under a progressive ratio (PR) schedule of reinforcement, where progressively additional active lever presses were required (response requirements: 1, 2, 4, 6, 9, 12, 15, 20, 25 etc. (Richardson and Roberts 1996)) for the subsequent single sucrose pellet reward. All subjects were trained once daily on 6 days over a period of 8 days. Performance was considered stable at group level when the average number of obtained rewards on individual days fell within a 90-110% variability range of the total average of obtained rewards over the last 3 training days. A PR session ended when a rat failed to obtain a subsequent reward within 30 min.

Brain processing, immunostaining and imagingBrain processing

At the end of the experiments, rats were injected intraperitoneally with a lethal dose of pentobarbital (100 mg/kg), and transcardially perfused with 0.1 M phosphate buffered saline (PBS; pH 7.4) followed by 4% formaldehyde in 0.1 M PBS. Brains were removed and post-fixed in 4% formaldehyde for 48 h. Subsequently, brains were kept in 20% sucrose in 0.1 M PBS until sunken, and then in 30% sucrose in 0.1 M PBS until sunken. Brains were then snap frozen in isopentane cooled by dry ice and kept in -80 °C until sectioning on either a cryostat (Cryostar NX70, Thermo Fisher Scientific, Waltham, USA), or a vibratome (Leica VT1200s) into 40 μm thick coronal sections and stored in cryoprotectant solution (30% sucrose w/v, 30% ethylene glycol v/v in 0.1 M PBS, pH 7.4) until further use.

Immunostaining

For immunohistochemistry staining of the DREADD-fused fluorophore, 1 in every 5th free-floating brain section throughout and surrounding the BNST was washed in 0.1 M Tris-buffered-saline (TBS), blocked for 30 min in 0.5% BSA in TBS, and incubated on an orbital shaker for 24 h at room temperature + 24 h at 4 °C in polyclonal rabbit anti-mCherry (1:30 000, Abcam, cat. Ab167453) or polyclonal chicken anti-EYFP (1:100 000), Abcam, cat. Ab13970) antibody solution containing 0.1% Triton-X and 0.1% BSA in TBS. Sections were then incubated in biotinylated goat anti-rabbit (1:400, Abcam, cat. Ab6720) or biotinylated goat anti-chicken (1:400, Abcam, cat. Ab6876) antibody solution containing 0.1% BSA in TBS for 30 min, avidin-biotin-peroxidase complex (VECTASTAIN ABC-HRP kit, Vector laboratories, cat. PK-6100, dilution: 1 drop A + 1 drop B in 10 mL TBS) solution for 30 min, and 3,3′-diaminobenzidine solution (DAB substrate kit (HRP), Vector laboratories, cat. SK-4100, dilution: 1 drop R1 (buffer solution) + 2 drops R2 (3,3′-diaminobenzidine solution) + 1 drop R3 (hydrogen peroxide solution) in 5 mL water) for 5–10 min, with TBS washes between all steps. Slides were dehydrated, cleared, and coverslipped using Entellan mounting medium (Sigma, St. Louis, USA). Immunohistochemical staining of c-Fos was carried out separately, on different brain sections than the DREADD fluorophore staining. Macroscopically matching AP anatomical location across brains, six sections (2 sections containing anterior and ventral BNST, 2 sections containing lateral BNST, and 2 sections containing posterior BNST) were selected for each animal. The same staining protocol as described above was employed, but a rabbit anti-c-Fos polyclonal antibody (1:4 000, Cell signaling, cat. 9F6) was used for primary antibody incubation.

Imaging

For imaging, slides were loaded into an Olympus VS120 virtual slide microscope system. High resolution image scans of the entire brain sections were obtained using a 10x objective using automatic focusing. For DREADD images, automatic exposure settings were used, whereas one and the same manual exposure time was used for c-Fos images across all animals and sections.

Image analysisDREADD

Using OlyVIA online database software (Olympus, Tokyo, Japan), the extent and location of DREADD expression was determined in each rat. DREADD expression in the BNST and in surrounding regions was categorized based on the amount of DREADD+ cells per region in the sample of sections (1 in 5 throughout the BNST). We qualified expression using a scoring system per brain region per hemisphere: 0 (no expression in the brain region); 1 (low expression; i.e. no more than a few positive cells in the region), 2 (medium expression throughout the brain region, typically > 10 and < 30 DREADD+ cells per section, or high expression in part of the brain region); and 3 (high expression throughout the brain region, typically > 30 positive cells per section). A second observer validated the qualifications in 5 brains with various expression patterns. We then added the scores for each hemisphere and excluded rats with a total score (left + right hemisphere) smaller than 3 for the BNST from further analysis. Some rats in the Sham group (n = 6 in the total experiment) had very low DREADD expression and did not meet inclusion criteria. These animals were not excluded from data analysis, reasoning that lack of DREADD expression does not disqualify their control condition.

Furthermore, besides extensive expression of DREADD throughout the anterior, lateral, ventral, and posterior BNST (Fig. 1C), DREADD expression was also observed in brain regions ventro-posterior to the BNST in all animals, namely the reticular thalamic nucleus, the sublenticular extended amygdala, posterior lateral hypothalamus, and the lateral preoptic area. As the overall weight of expression was in the BNST, no animals were excluded from analysis based on DREADD expression.

c-Fos+ cell analysis

Images were loaded into FIJI ImageJ and 4 regions of interest were defined: anterior BNST (between AP 0.0 mm and − 0.12 mm, dorsal of anterior commissure), ventral BNST (between AP 0.0 mm and − 0.12 mm, ventral of anterior commissure), lateral BNST (between AP -0.24 mm and − 0.36 mm, lateral of anterior commissure), and posterior BNST (between AP -0.72 mm and − 0.9 mm, between internal capsule and fornix). Boxes of always the same shape, size, and location were placed within these regions on both sides of the brain section images (one section per region of interest per brain, Suppl. Figure 6). The image was converted to 8-bit and then to binary by thresholding to the same threshold for each image in same experiment. The threshold was determined by testing on a wide range of images with high and low c-Fos+ density. The ImageJ particle analyzer (size range 5-2000 pixels, circularity range 0.5-1.0) was used to count the number of particles (c-Fos+ cells) within each box on the binary thresholded Image (Suppl. Figure 7). For each animal, the number of c-Fos+ cells were averaged across hemispheres for each region of interest. The automated counts were validated by comparing to manual counting of c-Fos+ cells of a sample of boxes with a wide range of c-Fos+ density.

Experimental design and interventionsSexual behavior experimentPair housing

Subject males were first sexually trained in the copulation test set-up. Sexual training sessions consisted of allowing the males to copulate with a female until ejaculation, or for 30 min if no ejaculation was reached. When males had reached ejaculation in 2 sessions, a final full 30 min copulation test (allowing for more than 1 ejaculation) was conducted for complete sexual experience. This ensured that each subject had ejaculated at least 3 times before experimental behavioral tests were conducted. Based on the number of ejaculations in the final training session, males were divided over 3 homologous phenotype groups for brain surgery. After brain surgery and recovery for 3 weeks, in the week before the first behavioral tests, subject males were habituated to the SIM arena without stimulus rats present in 3 sessions of 10 min. On subsequent testing days, subject males were first tested in the SIM test, 30 min after i.p. injection with 1 mg/kg CNO or vehicle, and then in the copulation test 5–15 min after the SIM test. Testing occurred once a week, following a Latin square within-subject design for balanced order of CNO and vehicle administration.

Single housing intervention

In the second part of the experiment, all males were single housed for 1 week before each testing day (vehicle and CNO) in order to study the effects of short-term social deprivation. For single housing, males were housed in smaller Macrolon cages with metal wire lids, still allowing for odors and sounds of rats in the same room to enter the cages. After single housing and the first test day, the subject males were rehoused in pairs for 1 week before being single housed for 1 week again for the second test day. Order of vehicle and CNO testing was counterbalanced within groups.

All behavioral tests were conducted during lights-off time.

Sucrose self-administration experimentAd libitum food access

After brain surgery and recovery, subject males were trained under FR1 until stable performance. Animals were subsequently tested 4 times under FR1; twice after CNO and twice after vehicle following a Latin square within-subject design. On testing days, rats were injected intraperitoneally with vehicle or CNO, and put in the operant box 30 min later. There was a minimum of 3 days between testing days, and each test was preceded by a baseline FR1 training session (no injections administered) the day before testing. After FR1 testing was completed, rats were trained under PR and tested 2 times; once after CNO and once after vehicle (counterbalanced). There was one day between the tests, and each test was preceded by a baseline PR training session the day before testing.

Food restriction

Rats underwent regular PR training sessions until re-testing under food restriction conditions. Food restriction consisted of removing homecage food access for 24 h before testing under PR. Animals were tested twice after food restriction; once after CNO and once after vehicle (counterbalanced). There were 5 days between the tests, and each test was preceded by a baseline PR training session two days (to allow for the 24-hour food restriction) before testing.

All behavioral tests were conducted during lights-off time.

Perfusion and DREADD validation

All rats were injected with 1 mg/kg CNO intraperitoneally on the day of euthanasia. Gq-animals, and half of the Sham-animals remained in the homecage after CNO injection and underwent cardiac perfusion with 4% formaldehyde 90 min later. In the sexual behavior experiment, all Gi-animals, and the other half of the Sham-animals were allowed to copulate until one ejaculation 30 min after CNO injection and were perfused 45 min after ejaculation. In the sucrose self-administration experiment, all Gi-animals, and the other half of the Sham-animals were allowed a 30-minute FR1 session in the operant box 30 min after CNO injection, were then returned to the homecage and perfused 30 min later. Brains of all animals were harvested for immunohistochemical analysis of DREADD expression and c-Fos protein positive cells.

Data analysis and statistics

During copulation behavioral annotation from video, it was noticed that one of the females was not receptive on one of the test days. This female was paired with 3 subject males in the copulation test and no copulation occurred in these tests due to the lack of receptivity. The data from these 3 copulation tests have therefore been removed from the analysis. If a rat achieved no ejaculations in a 30-minute copulation test, the latency to ejaculation was scored as 1800 s (30 min). Similarly, if a rat did not perform any copulatory behaviors, the latency to first behavior was scored as 1800 s.

Because the sucrose self-administration outcome measures under FR1 were quite variable over sessions within rats, each rat was tested twice after VEH and CNO treatment, and the data presented here are the comprised of the average of the two tests for each rat. Custom python scripts were used to extract the timestamps and latencies, and to calculate and create the cumulative curves of obtained rewards under FR1 and PR. In FR1, “mean reward interval” was calculated as (time of last reward – time of first reward)/ (number of obtained rewards − 1). In PR, mean lever press interval was calculated as (total test duration)/(total number of active lever presses).

All data was analyzed in SPSS statistical software (IBM, version 29, Armonk, USA) and GraphPad Prism (version 9.5). A linear mixed model comprised of only the factor virus*treatment interaction term was run for each of the separate outcome measures of the FR1 sucrose self-administration under ad libitum food conditions. In case of a significant interaction effect (alpha 0.05), Bonferroni-corrected post hoc tests were conducted to identify significant within- (CNO vs. vehicle) and between- (only Gi vs. Sham and Gq vs. Sham) group differences. For the sucrose self-administration under PR and the sexual behavior outcome measures, a second interaction term factor of virus*treatment*condition (the latter being food restriction or single housing) was added to the linear mixed model. Appropriate Bonferroni corrected post hoc tests were run upon significant interaction (alpha 0.05). The SIM preference score was compared to chance (0.5) with a one-sample t-test for each condition within each treatment within each group. Time spent in female zone was compared to time spent in male zone, and female zone frequency was compared to male zone frequency, with a paired t-test.

In addition, we identified FR1 lever pressing bouts by assuming that a distribution of all intra-bout interval durations has a smaller range and lower variance than inter-bout-intervals within animals. Hence, we ordered and plotted all lever press interval durations under vehicle conditions as scree plots for each animal (see Suppl. Figure 5 for example). Every scree plot was then visually examined to identify the approximate inflection point of the plot. This inflection point was considered the threshold duration for inter-bout-intervals. Since this threshold was around 10 s for most animals, we decided to use a 10 s threshold on all of the data for lever press bout analysis.

Densities of c-Fos+ cells were compared by means of 3 one-tailed independent samples t-tests (Gi vs. its Sham control, Gq vs. its Sham control, and the 2 Sham groups vs. each other, alpha = 0.05), or one-tailed Mann-Whitney U test in case an F-test revealed unequal variances. For 2 animals in the sexual behavior experiment, data from the anterior and ventral BNST were excluded because of damaged sections prohibiting reliable counting of c-Fos+ cells.