Sex differences in contextual pattern separation, neurogenesis, and functional connectivity within the limbic system

Subjects

Sixteen 8-week-old Sprague-Dawley rats (males: n = 8; females: n = 8) were purchased from Charles River Canada (St-Constant, QC, Canada). Rats were pair-housed in opaque polysulfone bins (432 mm × 264 mm × 324 mm) with paper towels, a single polycarbonate hut, virgin hardwood chip bedding, and free access to food and water. Males and females were housed in separate colony rooms that were maintained under a 12:12-h light/dark cycle (lights on at 07:00 h). All animals were handled every day for two minutes beginning one week after arrival for two weeks. All experiments were carried out in accordance with the Canadian Council for Animal Care guidelines and were approved by the animal care committee at the University of British Columbia. All efforts were made to reduce the number of animals used and their suffering during all procedures.

Apparatus

Behavioral testing for all experiments was conducted in four operant chambers (30.5 × 24 × 21 cm; Med-Associates, St Albans, VT) enclosed in sound-attenuating boxes. The boxes were equipped with a fan to provide ventilation and to mask extraneous noise. All behaviors were monitored and recorded by a single video camera mounted on the ceiling of each box. The chambers were equipped with a single 100-mA houselight located in the top center of a wall and the chamber floor consisted of 23 metal grid bars (0.4 cm in diameter) that ran parallel to the shorter wall of the chamber, which connected to a shock generator. Two chambers had wide vertical black (18 mm width) and white (12 mm width) stripe patterns on the walls and wiped with vinegar before and after each animal. The other two chambers had narrow vertical black (12 mm width) and white (12 mm width) stripe patterns on the walls and wiped with 70% isopropanol before and after each animal (see Fig. 1). All the chambers were connected to a computer through a digital interface that recorded all experimental settings.

Fig. 1figure 1

Experimental design. A Schematic illustration of experimental timeline: subjects received one intraperitoneal injection of 5-chloro-2'-deoxyuridine on Experimental Day 1 and one intraperitoneal injection of 5-iodo-2'-deoxyuridine on Experimental Day 8. Then, subjects were tested in the contextual pattern separation task for a total of 12 days (Experimental Day 16–28), followed by an activation trial in which the rats were placed in the context previously paired with shock but received no shock (Experimental Day 29). During the contextual pattern separation task, subjects were exposed to two different contexts each day; context A a shock-paired context (context paired with foot shocks) and context A’ a neutral context (context with no foot shock). B Brain regions that were examined for functional connectivity using zif268. ACC cingulate cortex (Cg1), PrL prelimbic cortex, IL infralimbic cortex, lDS lateral dorsal striatum, mDS medial dorsal striatum, NAc nucleus accumbens core, NAs nucleus accumbens shell, LA lateral amygdala, BLA basolateral amygdala, CeA central amygdala, dDG dorsal dentate gyrus, vDG ventral dentate gyrus, dCA1 dorsal cornu ammonis 1, vCA1 ventral cornu ammonis 1, dCA3 dorsal cornu ammonis 3, vCA3 ventral cornu ammonis 3

ProceduresExperimental timeline

Subjects received one injection of 5-chloro-2'-deoxyuridine (CldU:171 mg/kg; intraperitoneal (i.p.), MP Biomedicals, Santa Ana, CA, USA) on Experimental Day 1 and one injection of 5-iodo-2'-deoxyuridine (IdU: 56.75 mg/kg; i.p., Cayman Chemical, Ann Arbor, MI, USA) on Experimental Day 8, thus ensuring that we examined 4 week old and 3 week old cells, respectively. A previous study demonstrates that adult-born neurons reach the full maturation four weeks after BrdU injection in rats [31], the present study examined zif268 expression of 4-week-old neurons as activation of fully matured neurons. Thymidine analogs, IdU and CldU, incorporate into DNA during synthesis phase of cell proliferation, which can be distinguished from one another using respective antibodies [32,33,34,35]. Subjects were tested in the contextual pattern separation task (modified from [36]) for 12 days (Experimental Days 16–28, which are referred to as Trial Days 1–12), followed by a day of activation test trial that is described below (Experimental Day 29; see Fig. 1A).

Behavioral testing for contextual pattern separation

Subjects were exposed daily for 5 min each to two different contexts (4–5 h interval between contexts), a shock-paired context (Context A) and a neutral context (Context A’), for a total of 12 days. The contexts for Context A trials and Context A’ trials were counterbalanced across subjects and remained the same for each subject throughout the entire experiment. During the shock-paired trial in Context A, subjects were allowed to explore the chamber for three minutes followed by three 1-s foot shocks (0.6 mA) with 30-s intervals between each shock. The subjects returned to their home cage one minute after the third shock. During the neural trial in Context A’, the subjects explored a different context from Context A for five minutes without receiving a foot shock and returned to their home cage. The order of two contexts that subjects were exposed each day for the first 6 days followed AA’–A’A–A’A–AA’–AA’–A’A design, and the order was reversed for the remaining days [36] (see Fig. 1A).

The duration of freezing during the first 3 min of each trial (prior to any shocks) was examined as the conditioned fear response, and the percentage of freezing was calculated by dividing the duration of freezing by 180 s. A discrimination index (DI) was calculated with the following formula on the last two days of training:

$$\rm=\frac-\rm})}+\rm})}.$$

As a previous study found sex differences in darting, an active fear response, in a cued fear-conditioning task [37], darting behavior was also recorded.

Activation trial and perfusion

On the day after Training Day 12, the Activation Test Trial was conducted to examine fear memory. Subjects were exposed to the Context A for 5 min without a foot shock and returned to their home cage. Video recordings were analyzed for active fear behavior (darting), passive fear behavior (freezing), or other behaviors (rearing, grooming and non-specific behaviors; see Additional file 1). However, no darting in our paradigm was observed. Ninety minutes after the Activation trial, subjects were administered an overdose of sodium pentobarbital (500 mg/kg, i.p.) and perfused transcardially with 60 mL of 0.9% saline followed by 120 mL of 4% paraformaldehyde (Sigma‐Aldrich).

Tissue processing

Extracted brains were postfixed in 4% paraformaldehyde overnight, then transferred to 30% sucrose (Fisher Scientific, Ottawa, ON, Canada) solution for cryoprotection and remained in the solution until sectioning. Brains were sliced into 30-μm coronal sections using a Leica SM2000R microtome (Richmond Hill, ON, Canada). Sections were collected in series of 10 throughout the entire rostral‐caudal extent of the forebrain (Bregma 5.64 to − 7.56 mm) and stored in antifreeze solution consisting of ethylene glycol, glycerol, and 0.1 M PBS at − 20 °C.

Immunohistochemistry

Brain tissue was double-stained for the immature neuronal protein, doublecortin (DCX), and the immediate early gene, zif268 (see details of antibodies and reagents used in Additional file 1: Tables S1 and S2). A majority (70% or more) of adult-born granule cells express DCX within 24 h after mitosis for up to two weeks, with maximal expression at 4 days after mitosis, and DCX expression is rapidly reduced 3 weeks after mitosis (less than 20%) in both male and female rats [20, 31, 38]. Therefore, we used DCX to examine a cell population of new neurons that were larger 2 weeks old or younger. In addition, tissue was triple-stained for IdU, CldU, and zif268 to examine neural activation of 3-week-old (IdU) cells and 4-week-old (CldU) cells in the dentate gyrus.

Doublecortin/zif268 double labeling

Free-floating sections were prewashed three times for 10 min with 0.1 M Tris buffer saline (TBS; Sigma-Aldrich, Oakville, ON, Canada). Sections were then incubated in a primary antibody solution containing 1:500 rabbit anti-zif268 (Santa Cruz Biotechnology, Dallas, TX, USA), 1:500 goat anti-doublecortin (Santa Cruz Biotechnology, Dallas, TX, USA) 0.3% Triton-X (Sigma-Aldrich) and 3% normal donkey serum (NDS; MilliporeSigma, Burlington, MA, USA) in 0.1 M TBS for 24 h at 4 °C. Sections were washed three times for 10 min in TBS and a further incubation of sections commenced in a secondary antibody solution containing 1:500 donkey anti-rabbit ALEXA 594 (Invitrogen, Burlington, ON, Canada), 1:500 donkey anti-goat ALEXA 488 (Invitrogen, Burlington, ON, Canada), 3% NDS and 0.3% Triton-X in 0.1 M TBS for 24 h at 4 °C. Following three final rinses with TBS, the sections were mounted onto microscope slides and cover-slipped with PVA DABCO.

IdU/CldU/zif268 triple labeling

Two different thymidine analogues (CldU and IdU) were visualized with CldU-specific (rat monoclonal, clone BU1/75) and IdU-specific (mouse monoclonal, clone B44) antibodies [39], coupled with labeling using the immediate early gene, zif268 antibody (rabbit polyclonal). Briefly our protocol was as follows: free-floating sections were prewashed three times for 10 min with 0.1 M TBS. Sections were then incubated in a primary antibody solution containing 1:500 rabbit anti-zif268 (Santa Cruz Biotechnology, Dallas, TX, USA), 0.3% Triton-X (Sigma-Aldrich) and 3% NDS in 0.1 M TBS for 24 h at 4 °C. Next, sections were incubated in a secondary antibody solution containing 1:250 donkey anti-rabbit ALEXA 647 (Invitrogen, Burlington, ON, Canada), 0.3% Triton-X, and 3% NDS in 0.1 M TBS, for 18 h at 4 °C. After being rinsed three times for 10 min with TBS, sections were washed with 4% paraformaldehyde for 10 min, and rinsed twice in 0.9% NaCl for 10 min, followed by incubation in 2 N HCl (Fisher Scientific, Waltham, Massachusetts, USA) for 30 min at 37 °C. Sections were then rinsed three times in TBS for 10 min each and incubated in a CldU primary antibody solution consisting of 1:1000 rat anti-BrdU (BU1/75; Abcam; Toronto, ON, Canada), 3% NDS, and 0.3% Triton-X in 0.1 M TBS for 24 h at 4 °C. Sections were then incubated in an IdU primary antibody solution consisting of 1:500 mouse anti-BrdU (B44; BD Biosciences, San Jose, CA, USA), 0.3% NDS, and 0.3% Triton-X in 0.1 M TBS for 24 h at 4 °C. Sections were then washed twice for 10 min each in a high stringency wash solution consisting of 32 mM Tris buffer, 50 mM NaCl and 0.5% tween (pH 8.0) at 37 °C. Following three washes in TBS, sections were incubated in a secondary antibody solution containing 1:500 donkey anti-rat ALEXA 594 (Invitrogen, Burlington, ON, Canada), 1:500 donkey anti-mouse ALEXA 488 (Invitrogen, Burlington, ON, Canada), 3% NDS and 0.3% Triton-X in 0.1 M TBS for 24 h at 4 °C. Following three final rinses with TBS, the sections were mounted onto microscope slides and cover-slipped with PVA DABCO.

Cell counting

All counting was conducted by an experimenter blind to the group assignment of each animal using an Olympus FV1000 confocal microscope and/or Zeiss Axio Scan.Z1 (Carl Zeiss Microscopy, Thornwood, NY, USA). Density of immunoreactive cells was calculated by dividing the total immunoreactive (ir) cells by volume (mm3) of the corresponding region. Volume estimates were calculated by multiplying the summed areas by thickness of sections (0.03 mm, using Cavalieri’s principle [40]). Area measurements for the region of interest were obtained using digitized images on Zen 3.0 software (blue edition; Carl Zeiss Microscopy, Thornwood, NY, USA).

Brain regions were defined according to a standard rat brain atlas [41]. Location of immunoreactive cells in the dorsal or ventral hippocampus was examined using the criterion defined by Banasr et al. [42] with sections 7.20–4.48 mm from the interaural line (Bregma − 1.80 to − 4.52 mm) defined as dorsal and sections 4.48–2.20 mm from the interaural line (Bregma − 4.52 to − 6.80 mm) as ventral [42]. Cells were counted separately in each region because the different regions are associated with different functions (reviewed in [43]) and different maturation timelines of neurogenesis [13, 20]. The dorsal hippocampus is associated with spatial reference memory, whereas the ventral hippocampus is associated with working memory, stress and anxiety [14, 15].

IdU and CldU counting

Thymidine analogue immunoreactive (IdU-ir and CldU-ir) cells were counted under a 40 × objective lens using Olympus FV1000 confocal microscopy. Every 20th section of the granule cell layer (GCL) that includes the subgranular zone (SGZ) was counted. The SGZ was defined as a narrow layer of cells within 30 μm (equivalent to the width of three granule cell bodies) away from the innermost edge of GCL (Redila and Christie, 2006).

The percentages of IdU/zif268 and CldU/zif268-ir cells were obtained by randomly selecting 200 IdU-ir or 200 CldU-ir cells (100 cells from dorsal and 100 cells from ventral DG) and calculating the percentage of cells that were double-labeled with zif268 under a 40 × objective lens using Olympus FV1000 confocal microscopy. Density of DCX/zif268-ir, IdU/zif268-ir or CldU/zif268-ir cells were calculated by multiplying the density of IdU-ir or CldU-ir cells by the percentage of double-labeled cells.

Doublecortin counting

Doublecortin immunoreactive (DCX-ir) cells were counted on digitized images on Zen 3.0 software (blue edition). Photomicrographs were taken from four dorsal and four ventral hippocampi using a ZEISS Axio Scan.Z1 slidescanner with a 40 × objective lens. The percentages of DCX/zif268-ir cells were obtained by randomly selecting 200 DCX-ir cells (100 cells from dorsal and 100 cells from ventral DG) and calculating the percentage of cells that were double-labeled with zif268 on Zen 3.0 software. Density of DCX/zif268-ir cells were calculated by multiplying the density of DCX-ir cells by the percentage of DCX/zif268-ir cells.

Estrous cycle determination. Vaginal cells suspended in water were obtained using a glass pipette, transferred onto microscope slides, stained with Cresyl Violet (Sigma), and analyzed using a 20 × objective. Proestrus stage was determined when 70% of the cells were nucleated epithelial cells [44].

zif268 counting

Photomicrographs of coronal sections containing the frontal cortex, amygdala, hippocampus, dorsal striatum, nucleus accumbens were obtained from ZEISS Axio Scan.Z1 slidescanner with a 20 × objective lens (four images from each region of interest: see Fig. 1B). Zif268-ir cells in the infralimbic cortex (IL), prelimbic cortex (PrL), anterior cingulate cortex (ACC: Cg1), medial part of dorsal striatum (mDS), lateral part of dorsal striatum (lDS), nucleus accumbens core (NAc), nucleus accumbens shell (NAs), central nucleus of amygdala (CeA), basolateral nucleus of the amygdala (BLA), lateral nucleus of the amygdala (LA), dorsal(d) hippocampus (dCA1, dCA3, dDG) and ventral(v) hippocampus (vCA1, vCA3, vDG) were counted automatically from the digitized images using a code developed by JEJS (see [20] for details) on MATLAB (MathWorks; Natick, Massachusetts, USA).

Estrous cycle determination

Daily lavage samples were taken from all females after behavioral procedures (see Additional file 1: Methods section). Estrous cycle determination was done as the estrous cycle stage can affect long-term potentiation and IEG expression in the hippocampus [45, 46]. There was one female in the proestorus stage during the Activation Trial and thus, estrous cycle phase was used as a covariate for all analyses.

Statistical analyses

All analyses were conducted using Statistica (Statsoft Tulsa, OK) unless otherwise stated, and significance level was set at α = 0.05. Repeated-measures or factorial analysis of variance (ANOVA), with sex (male and female) as between-subject variables were conducted on our variables of interest (freezing, zif268 expression). Post-hoc tests used the Newman–Keuls procedure. A priori comparisons were subjected to Bonferroni corrections. Effect sizes are given with Cohen’s d or partial η2. Pearson product-moment calculations and principal component analyses on zif268 expression across regions were also performed.

The percentage of freezing during the Context A trials and Context A’ trials in the contextual pattern separation task was analyzed using repeated-measures analysis of variance (ANOVA), with sex (male and female) as between-subject variables and context (Context A and Context A’) and trial day (1st–12th day) as within-subject factors. The discrimination index of the last trial block and percentage of freezing during the activation trial were analyzed using one-way ANOVA with sex as between-subject variable. The density of adult-born cells (DCX-ir, IdU-ir or CldU-ir cells) and those double-labeled with zif268 in the dentate gyrus were each analyzed using repeated-measures ANOVA with sex as between-subject variable and region (dorsal and ventral) as within-subject variable. The density of zif268-ir cells in each region (frontal cortex, dorsal striatum, nucleus accumbens, amygdala) was analyzed separately using repeated-measures ANOVA with sex as between-subject variables and subregions (frontal cortex: IL, PrL, ACC; dorsal striatum: lateral, medial; nucleus accumbens: core, shell; amygdala: central, lateral, basal; hippocampus: dorsal and ventral CA1, CA3 and DG) as within-subject variables.

Pearson product–moment correlations between the percentage of freezing and the density of zif268-ir cells were calculated in the regions of interest. For functional connectivity, Pearson product–moment correlations were calculated with the density of zif268-ir cells between each brain region. To examine the functional connectivity of adult-born cells in the dentate gyrus with the other brain regions, correlations were also calculated between the density of IdU/zif268-ir, CldU/zif268-ir or DCX/zif268-ir cells and the density of zif268-ir cells in each region. Inter-regional correlations were compared between the two sexes (male and female) using the single-sided observed Fisher z-test statistic.

Principal component analyses were conducted to assess brain networks that explain variances of zif268-ir cell density in the regions of interest. PCA data analyses were conducted using Statistica and R (3.4.3) statistical analysis software with the “FactoMineR” package. Horn’s parallel analysis was used to determine which component factors were retained for further analyses [47]. Horn’s parallel analysis was conducted using R (3.4.3) statistical analysis software with the “psych” package. Following the principal component analysis, a repeated-measures ANOVA was conducted with the principal component factor scores (1st, 2nd, 3rd) as the within-subject variable and sex (male, female) as the between-subject variable.

留言 (0)

沒有登入
gif