Abstract

The rat prefrontal cortices participate in cognitive, affective and mnemonic functions. The importance of dopamine innervation for these computations is illustrated in studies showing that both supranormal levels and chemical lesions of prefrontal dopamine induce severe behavioral deficits. Observed hormone effects on some of these same behaviors suggest that the prefrontal cortices are also sensitive to gonadal steroids. These two influences seem to converge in recent evidence of increased dopamine axon density in representative prefrontal but not sensory or motor cortices in gonadectomized adult male rats. The seeming selectivity of these effects was further explored here using immunocytochemistry for tyrosine hydroxylase, dopamine-b-hydroxylase, serotonin and choline acetyltransferase to label neurochemically identified afferents in remaining, unstudied prefrontal fields of rat cortex in animals that were sham-operated or gonadectomized and given placebo, testosterone propionate, estradiol or dihydrotestosterone 28 days before being killed. Group comparisons revealed that across prefrontal zones, gonadectomy produced androgen-sensitive increases in presumed dopamine axon density, but did not affect the other afferents. These findings thus bolster evidence for a targeted gonadal steroid influence involving the prefrontal cortices and a neurotransmitter essential for their normal operations and implicated in their dysfunction in disorders such as schizophrenia as well.

Introduction

The prefrontal cortices are a constellation of areas that mediate highest order cognitive, affective and mnemonic processes. Identified largely by inputs from the mediodorsal nucleus of the thalamus (Leonard, 1969; Krettek and Price, 1977; Van Eden, 1986; Groenewegen, 1988), in rats the prefrontal areas include the medial precentral shoulder cortex (PrCm), a presumed homologue of the frontal eye fields in primates (Sesack et al., 1989), the prelimbic (PL) and anterior cingulate (ACd) areas which orchestrate working memory functions (Eichenbaum et al., 1983; Kolb, 1990; Kesner et al., 1996) and the anterior insular areas (AId, AIv) that organize affective behaviors (Eichenbaum et al., 1983; Kolb, 1990). The infralimbic (IL) cortex, which participates in sequential or temporal ordering tasks (Kolb, 1990), receives only a sparse mediodorsal thalamic input (Divac et al., 1978; Sarter and Markowitsch, 1983), but is nonetheless included among prefrontal areas in rats for the rich dopamine (DA) innervation it shares with other mediodorsal-recipient prefrontal zones. Area IL also shares with other prefrontal fields a strong functional reliance on the integrity of these DA afferents. Indeed, each of the functions enumerated above are vulnerable to either selective chemical lesions of prefrontal DA afferents (Kalsbeek et al., 1990) or supranormal levels of DA stimulation induced by local administration of DAreceptor agonists (Zahrt et al., 1997), or pharmacologic stimulation of prefrontal DA turnover (Murphy et al., 1996) or release (Verma and Moghaddam, 1996). In addition, however, sex differences and/or hormone malleability of the acquisition and performance of working memory and other cognitive tasks not only in rats (Beatty, 1984; Van Hest et al., 1988; Kritzer et al., 2001), but also in humans (Christiansen and Knussman, 1987; Hampson, 1990) and primates (Clark and Goldman-Rakic, 1989), indicate that the prefrontal cortices are also influenced by gonadal hormones. The present studies pursue recent evidence of a functional link between these two prefrontal influences — namely, gonadal hormone stimulation of prefrontal DA innervation in adult male rats.

Although multiple cortical endpoints of gonadal steroid stimulation have been identified, those that may have particular relevance for the prefrontal cortices are recently identified effects on cortical DA innervation. Thus, long-term gonadectomy in adult male rats has been shown to increase DA but not noradrenalin (NA) axon density in representative prefrontal but not sensory or motor areas (Kritzer et al., 1999; Kritzer, 2000). Such selective actions could have relevance for normal operations of the frontal lobes and may even have implications for disorders such as schizophrenia, where prefrontal hypodo-paminergia has been linked to its negative symptoms, e.g. anhedonia, which are also those most commonly observed in males (Goldstein, 1988; Goldstein and Tsuang, 1990; Seeman and Lang, 1990). However, the issue of selectivity remains uncertain, mainly because hormone effects have only been explored in two prefrontal areas. This limitation is significant not only because the rat prefrontal cortices are structurally and functionally distinct, but also because these areas are also distinguished by anatomical and physiological differences in their DA afferents. Thus, from prefrontal region to region, DA afferents not only show unique distributions, densities and morphologies (Slopsema et al., 1982; Berger et al., 1985; Van Eden et al., 1987; Kalsbeek et al., 1988), but they can also differ with respect to metabolic rates, developmental histories, co-localization patterns with neuropeptides and physiological responses to stress and/or pharmacologic stimulation (Berger et al., 1991, 1992; Deutch, 1993). This diversity thus limits the predictive power of previous analyses (in supragenual areas ACd and AId) for outcomes in remaining, unstudied prefrontal fields. The present studies addressed this deficiency by expanding investigations of the effects of long-term gonadectomy to prefrontal areas IL, PL, pregenual ACd and PrCm. Together with the previously examined regions, these areas round out lists of both major prefrontal areas and DA-enriched regions of the rat cerebrum. Analyses were also expanded to include assessments of indolaminergic and cholinergic as well as catecholaminergic afferents. Using immunoreactivity for the catecholamine synthesizing enzymes tyrosine hydroxylase (TH) and dopamine-β-hydroxylase (DBH), for serotonin (5-HT), or the acetylcholine synthesizing enzyme choline acetyltransferase (CHAT) as markers, axon systems were assessed in identified layers in left and right hemifields of each of four prefrontal areas (see above) in rats that had been sham-operated or gonadectomized and supplemented with placebo, testosterone propionate, estradiol or dihydro-testosterone 28 days prior to being killed. Detailed comparisons of corresponding innervations across animal groups were used to identify prefrontal afferent systems vulnerable to the effects of long-term gonadectomy and to gain insights into the hormone signaling pathways active in their regulation. As described below, the results obtained suggest a highly selective hormone modulation of axon density that is focused on the DA afferents and perhaps specifically on the so-called Class 1 DA afferents (Berger et al., 1991) of the prefrontal cortices in adult male rats.

Materials and Methods

Animal Subjects

Twenty-five adult male Sprague–Dawley rats (Taconic Farms, Germantown, NY) were used. All procedures involving animals had approval by the Institutional Animal Care and Use Committee, SUNY at Stony Brook and were designed to minimize their use and discomfort. Animals were housed with food and water freely available under a 12 h light/dark cycle. Five rats were sham-operated and implanted with placebo; the remaining 20 were gonadectomized and implanted with pellets containing placebo, testosterone propionate (TP), 17-β-estradiol (E), or 5-α-dihydrotestoster-one (DHT).

Surgical Procedures

Gonadectomies were performed under aseptic conditions using ketamine (0.09 ml/100 g) plus xylazine (0.05 ml/100 g) for anesthesia. To begin gonadectomy and sham surgeries, the sac of the scrotum and underlying tunica were incised; for gonadectomies, the vas deferens was then bilaterally ligated and the testes were removed. For both gonadectomy and sham procedures, slow-release pellets were implanted (below) within the scrotal sac, and incisions were closed using 6-0 silk sutures.

Hormone Treatments

Animals were implanted with slow release pellets that contained TP, E or DHT in biodegradable matrix, or placebo (pl), i.e. matrix only (cholesterol, microcrystalline cellulose, α-lactose, di- and tri-calcium phosphate, calcium and magnesium stearate and stearic acid; Innovative Research of America, Toledo, OH). The TP-containing pellets implanted released 0.5–2 ng of TP per milliliter of blood per day; the E-containing pellets released between 10 and 40 pg of E per milliliter of blood per day; and the DHT-containing pellets released 0.5–2 ng of DHT per milliliter blood per day.

Euthanasia

Twenty-eight days after surgery, rats were deeply anesthetized using i.m. injection of ketamine (0.09 ml/100g) and xylazine (0.05 ml/100 g). After corneal reflexes could no longer be elicited, the chest cavity was opened, samples of trunk blood were collected and rats were intracardially injected with heparin (0.2 ml of a 1000 U/ml solution). Rats were then transcardially perfused first with 50–100 ml of 0.1 phosphate-buffered saline (PBS), then with 4% paraformaldehyde in 0.1 M PB, pH 6.5 (flow rate 30 ml/min, duration 5 min) and, finally, with 4% paraformaldehyde, in 0.1 M borate buffer, pH 9.5 (flow rate 35 ml/min, duration 20 min). After perfusion, brains were removed, the frontal lobes were blocked and cryoprotected in 0.1 M PB containing 30% sucrose prior to rapid freezing in powdered dry ice and storage at −80°C. The bulbocavernosus muscles were also dissected out and weighed (Table 1).

Serum Hormone Level Assays

Samples of trunk blood collected from anesthetized animals sat in open microfuge tubes at room temperature for ∼15 min to allow the blood to coagulate. The samples were then centrifuged (6000 r.p.m., 15 min) to separate and save serum, which was snap-frozen and stored at −80°C prior to direct solid-phase enzyme immunoassay for levels of TP, DHT and/or E; the results of the assay are shown in Table 2 (Testosterone EIA, Dihydrotestosterone EIA, Estradiol EIA; American Laboratory Products Co. Ltd, Windham, NH).

Immunocytochemistry

Tissue blocks containing the frontal lobes were coronally sectioned on a freezing microtome (40 μm) from roughly mid-levels of the olfactory bulb through the genu of the corpus callosum; left hemispheres were marked with identifying artifacts. Adjacent rostral to caudal series of sections from each animal were then immunoreacted using antibodies recognizing the synthesizing enzymes TH (Chemicon International Inc., Temecula, CA), DBH (Protos Biotech Corp., New York) or CHAT (Chemicon International Inc.), or recognizing the indolamine neurotransmitter 5-HT (Protos Biotech Corp., New York). For each antigen, all tissue sections to be used for comparative analyses were processed together as a group under identical conditions. For all experiments, sections were rinsed in 0.1 M PB, endogenous peroxidase activity was eliminated using 1% H2O2 in PB (45 min) and free aldehydes were bound using 1% sodium borohydride in PB (45 min). Sections were then rinsed extensively in 50mM Tris-buffered saline (TBS, pH 7.4) and placed in blocking solution [50 mM TBS containing 10% normal swine serum (NSS) for anti-TH, anti-DBH and anti-CHAT; 50 mM TBS containing 10% NSS and 0.25% Triton X-100 for anti-5-HT] for 2 h, prior to incubation in primary antibodies (TH — 3 days, diluted 1:1000 in TBS containing 1% NSS at 4°C; CHAT — 4 days, diluted 1:100 in TBS containing 1% NSS at 4°C; DBH — 3 days, diluted 1:100 in TBS containing 1% NSS and 0.25% Triton X-100 at 4°C; 5-HT — 4 days, diluted 1:1000 in TBS containing 1% NSS and 0.25% Triton X-100 at 4°C). The tissue sections were then rinsed thoroughly in TBS, incubated in biotinylated secondary antibodies (Vector, Burlingame, CA) overnight at 4°C and a working dilution of 1:100, rinsed again in TBS and then placed in avidin–biotin-complexed horseradish peroxidase (ABC; Vector) for 2 h at room temperature. After this final incubation, sections were rinsed in Tris buffer (pH 7.6) and reacted using 0.07% 3,3′-diaminobenzidine (DAB) as chromagen. Subsets of sections from representative animals were also processed as above, but with omission of primary antisera as controls.

Silver/Gold Intensification

The brown DAB reaction product was silver-intensified to a deep black using published methods (Kitt et al., 1988). First, DAB-reacted, slide-mounted sections were incubated in 1% silver nitrate (pH 7.0, 50 min, 55°C, in the dark). After a brief rinse in running distilled H2O, slides were incubated in 0.2% gold chloride (15 min, room temperature, in the dark). Additional rinses in distilled H20, were followed by fixing the sections in a 5% sodium thiosulfate solution (10 min, room temperature), more rinses in distilled H2O and then counterstaining using 0.1% cresyl violet. As a final step, sections were placed under coverslips and coded for analysis. Control sections (see above) were silver-intensified in parallel with normally immunoreacted slides and all sections labeled with a given antibody to be used for comparative analyses were processed together as a group under identical conditions.

Qualitative Evaluation

In each group of animals, the laminar distribution, orientation, approximate density and the morphology of TH-, DBH-, CHAT- and 5-HT-immunoreactive (ir) axons were assessed in representative sections throughout the rostrocaudal extent of the left and right hemifields of the pregenual anterior dorsal cingulate cortex (area ACd), PL, IL and the PrCm of the medial frontal cortical wall. These areas and their layers were positively identified by matching features of cytoarchitecture revealed in the counterstained sections to published descriptions of these divisions of the rat frontal lobe (Van Eden and Uylings, 1985).

Quantitative Evaluation

Quantitative estimates of axon density were obtained for cytoarchitectonically identified regions and layers at a single, matched anteroposterior level. In most regions, layers II/III and V were analyzed; due to difficulty in identifying boundaries of infragranular layers in area IL, especially in sections in which immunostaining was dense, labeling here was quantified only in layers II/III. All tissue sections used for these analyses were immunoreacted (by antigen) as a group and all analyses were carried out by a single observer (M.F.K.).

Fiber density values were derived from camera lucida drawings of ir fibers visualized under brightfield illumination using a Zeiss Axioskop and a 63× oil immersion objective. At the outset, section thickness was measured by roll-focusing from surface to surface, using the calibrated fine-focus to ensure uniformity in breadth from section to section. Then, individual drawings were made, subtending regions of ∼300–600 mm across (measured parallel to the pial surface), with heights dictated by the thickness of the layer and depths corresponding to the 40 mm thickness of the sections. For every animal, area, layer, hemisphere and antigen, data were derived from two non-overlapping drawings. In selecting areas for analysis, only physically damaged regions were disqualified.

Note that the analysis of the cholinergic afferents presented a special set of problems because ir profiles included both afferent axons and interneurons that were especially numerous in the supragranular layers. Although it was not possible to distinguish afferents arising from basal forebrain versus intrinsic cholinergic cells, the thick, often varicose primary and secondary dendrites of these local circuit neurons were readily distinguished from the thin, more threadlike cholinergic axons and were thus excluded from analysis. Accordingly, it is presumed that the density measures reported here largely reflect axonal profiles, both intrinsic and afferent, but may also include some measure of distal dendritic processes. The ranges of the numbers of ir somata that appeared in total drawing areas from subject to subject appear in Table 6.

Camera lucida drawings were then scanned (DeskScan 2.0), digitized as binary images of black and white pixels (black and white drawings) and skeletonized (NIH Image 1. 58) — a process with replaces lines of potentially varying thickness with lines of uniform diameter. From these processed images, measures for mean pixel density were used to yield fiber density estimates that were directly proportional to two-dimensional fiber length (Kritzer and Kohama, 1998).

Statistical Analyses

The sample sizes of all data sets compared in this study were equal. Descriptive statistical analyses were performed on each set (Stat View 4.5) to evaluate sample distributions, means and variances. Body and bulbocavernosus muscle weights and serum hormone levels were then compared across groups using analyses of variance (ANOVA), followed by Student–Newman–Keuls post hoc comparisons where indicated (Super ANOVA 1.11). Mean pixel density measures (axon density) were compared across groups on a per-region basis using ANOVAs with repeated-measures designs (Super ANOVA 1.11); the individual factors included in these analyses were: animal subjects, cortical hemisphere, cortical layer and the two measures (drawings) obtained per layer. These ANOVAs were also followed by Student–Newman–Keuls post hoc comparisons where indicated. For all statistical tests, a level of P < 0.05 was accepted as significant.

Photodocumentation

All photomicrographs appearing in this study were obtained using a Zeiss Axioplan microscope and a 63× oil immersion objective interfaced with a digital camera (RC SNAP Color Camera; Diagnostic Instruments Inc., Sterling Heights, MI) and Image Pro Plus morphometric software. All images were captured using identical settings for gain (7) exposure time (800 ms) and binning (1). Images were then saved as TIFF files and imported into Canvas 8.0 to be grouped into plates and labeled. Some adjustments were made to the brightness and/or contrast of the images using the Canvas software to optimize visualization of fine processes.

Results

Efficacy of Hormone Treatments

The procedures used to reduce and replenish circulating gonadal steroids in adult male rats were gonadectomy followed by implantation of slow-release pellets containing either placebo or hormone (TP, E or DHT). Two means of evaluation were used to confirm that these procedures changed circulating hormone levels. First, the androgen-sensitive bulbocavernosus muscles (Wainman and Shipounoff, 1941) were weighed and compared for expected group differences in an ANOVA. These analyses revealed significant effects of hormone treatment on muscle weights (P < 0.0001), with post hoc comparisons (Student– Newman–Keuls) showing that that rats subjected to chronic (28 day) gonadectomy and implanted with placebo or estrogen had absolute and proportionate muscle weights that were significantly smaller than controls (asterisks, Table 1). In contrast, there were no significant differences in muscle mass between control and GDX rats supplemented with TP or DHT (Table 1). Direct solid phase enzyme immunoassay measures of serum levels of TP, DHT and E in the control, gonadectomized and hormonally replaced groups measured at the time of death also revealed expected patterns of hormone deficiencies and levels of replaced gonadal steroids that either approximated or exceeded values measured in intact adult males (TP, DHT) or reported in the literature for intact adult females (E, Table 2) (Belanger et al., 1981). An ANOVA also revealed significant effects of hormone treatment (P < 0.0001) on serum levels and post hoc comparisons showed specifically that levels of TP and DHT were significantly lower in the GDX-pl group than in controls (asterisks, Table 2).

Immunoreactivity for Catecholamine Synthesizing Enzymes and for 5-HT

The present studies used commercially available antisera that recognize TH, CHAT, DBH and 5-HT to label dopaminergic, cholinergic, noradrenergic and serotonergic axons, respectively. That all four antibodies produced patent staining of TH, CHAT, DBH or 5-HT fibers in this study was evinced mainly by the similarities in the morphology, distribution and apparent densities of the axons labeled in the control animals of this study (see below) to descriptions of these afferents found in the literature for TH (Berger et al., 1985; Van Eden et al., 1987), DBH (Levitt and Moore, 1978; Morrison et al., 1978; Lewis et al., 1979), 5-HT (Lidov et al., 1980; Kosofsky and Molliver, 1987) and CHAT (Houser et al., 1985). Antibody specificity was also reflected in the elimination of obvious patterned staining in control experiments in which primary antisera were omitted from labeling protocols.

Gonadal Steroid Stimulation of TH Innervation in Rostromedial Prefrontal Cortex

The present studies analyzed cortical TH-ir afferents in chronically gonadectomized (28 day) adult male rats in four previously unexplored subdivisions of the prefrontal cortex — areas IL, PL, ACd and PrCm. From initial qualitative analyses alone it was evident that TH-ir axons in all but area PrCm responded vigorously to this long-term hormone deprivation. Specifically, in the GDX-pl cohort, in all layers and hemifields of areas IL, PLand ACd, what should have been moderate groupings of TH-iraxons (Figs 1A, 2A and 3A) were, instead, unusually dense accumulations of ir fibers (Figs 1B, 2B and 3B). Closer inspection revealed, however, that these denser than usual afferent meshworks were composed of axons of essentially normal appearance and orientation. For example, in the control animals, in areas IL (Fig. 1A) and ACd (Fig. 3A), TH-ir axons corresponded to fields of thin to medium caliber axons that were oriented more or less randomly across a range of angles, whereas in area PL (Fig. 2A), axons included both relatively thick, smooth fibers that were oriented perpendicular to the pial surface and thinner axons that ran at more oblique angles. These characteristics were also evident in the GDX-pl rats, albeit on denser scales (Figs 1B, 2B and 3B). In area PrCm, axons were also qualitatively similar in the GDX-pl and control groups (Fig.4A,B). However, unlike areas IL, PL and ACd, there were no discernible differences in axon density. Rather, the characteristic thick and thin axons that run mainly perpendicular to cortical lamination in this region seemed to be present in similar numbers in the GDX-pl and control rats alike (Fig. 4A,B).

Analyses were also carried out on gonadectomized rats supplemented with TP. These studies revealed patterns of TH-ir innervation that were not appreciably different from controls in morphology, distribution, or apparent density in any region evaluated (Figs 1E, 2E, 3E and 4E). Similar findings were also obtained in gonadectomized rats supplemented with DHT, where, again, patterns of TH-ir innervation were in all areas similar to those present in the controls (Figs 1C, 2C, 3C and 4C). In marked contrast, analyses in gonadectomized rats treated with E revealed patterns of immunolabeling that resembled those in the GDX-pl rats. Thus, while the cortices of the GDX-E animals contained TH-ir axons that were morphologically similar to controls, their apparent densities were similar to controls only in area PrCm (Fig. 4D) and were clearly higher than normal in areas IL (Fig. 1D), PL (Fig. 2D) and ACd (Fig. 3D).

The apparent stimulatory effects of long-term gonadectomy for TH-ir afferents in the GDX-pl and GDX-E cohorts and their attenuation in GDX-TP and GDX-DHT rats were substantiated in quantitative assessments of axon density. For these studies, calculations of mean axon densities (mean pixel densities — see Materials and Methods) were compared across animal groups in left and right hemifields of layers II/III in area IL and layers II/III and V/VI of areas PL, ACd and PrCm (Table 3). These analyses confirmed visual impressions (see above) and showed that in areas IL, PL and ACd, axon densities in the GDX-pl and GDX-E groups were typically between 50 and 100% higher than corresponding values in controls in every layer and hemisphere evaluated (Table 3). Quantitative estimates also showed that in area PrCm of the GDX-pl and GDX-E groups and in all prefrontal areas in the GDX-TP and GDX-DHT cohorts, what appeared to be qualitatively normal levels of innervation corresponded to axon density measures that differed from control values often by <5% (Table 3). Statistical comparisons in region-by-region ANOVAs including all five animal groups confirmed the reliability of axon density data by identifying significant main effects of hormone treatment on TH innervation in areas IL (P < 0.0018), PL (P < 0.0023) and ACd (P < 0.0001) and no significant variance in the data attributable to individual animals for any area. The failure to find significant interactions between hormone treatment and cortical hemisphere or layer within these areas further indicated the similarity of hormone effects across layers and hemifields. Finally, permitted post hoc comparisons (Student–Newman– Keuls) showed that all axon density measures from areas IL, PL and ACd in the GDX-pl and GDX-E groups — and none of them from the GDX-TP or GDX-DHT cohorts — were significantly different from controls (asterisks, Table 3). The ANOVA carried out for area PrCm, on the other hand, identified no main effect of hormone treatment (P < 0.89), no significant interactions between hormone treatment and cortical hemisphere or layer and no significant variance in the data attributable to individual animals.

Gonadal Steroid Stimulation of DBH, CHAT and 5-HT Innervation in Rostromedial Prefrontal Cortex

Analyses of DBH-, CHAT- and 5-HT-ir axons were also carried out in prefrontal areas IL, PL, ACd and PrCm. In contrast to the immediately evident effects of gonadectomy on TH-ir axons, qualitative comparisons of control and GDX-pl rats showed that in each region and for each antigen, signature patterns of labeling identified in hormonally intact rats were largely preserved in the hormonally deprived cohort. Thus, in the controls, immunoreactivity for 5-HT corresponded to dense meshworks of axons that, upon closer inspection, were found to include both thick, relatively smooth fibers and thinner, more beaded axons in all the prefrontal areas examined (Figs 5–8, panels AC). The patterns of CHAT immunoreactivity in control animals also corresponded to modest (area IL) to dense (e.g. area PL) arrays of thin fibers that were uniquely dispersed among populations of CHAT-ir interneurons (Figs 5–8 , panelsGI). Axons immunoreactive for DBH, on the other hand, were less dense than 5-HT-ir or CHAT-ir processes and tended to be thicker and more prominently beaded than the 5-HT-ir or CHAT-ir processes (Figs 5–8 , panelsDF). These and other characteristics identified for immunolabeling in the controls (Figs 5–8, panels A,D,G) — including the apparent densities of 5-HT, CHAT and DBH-ir afferents, were repeated without discernible difference in the GDX-pl rats (Figs 5–8, panels B,E,H). Analyses of gonadectomized animals supplemented with TP produced similar results, with patterns of immunoreactivity that were in all areas and for all antigens not appreciably different from controls (Figs 5–8, panels C,F,I).

Similarities in innervation across groups were also supported at a quantitative level in calculated densities for all three classes of axon in the GDX-pl and GDX-TP groups that were found to lie within ∼10% of corresponding values of controls (Tables 4–6). Statistical evaluations of these data also failed to identify significant main effects of hormone treatment for any of the four cortical areas examined (for DBH-ir axons, P-values ranged from 0.034 in area PL to 0.80 in area ACd; for 5-HT-ir afferents, P-values were between 0.17 in area PrCm and 0.48 in area IL; for CHAT-ir axons, P-values were between 0.224 in area PL and 0.654 in area PrCm). There were also no significant interactions identified between hormone treatment and cortical hemisphere or layer and no significant variance in the data attributable to individual animals for any region or antigen evaluated. In view of the negative findings in the GDX-pl and GDX-TP animals, additional analyses in GDX-E or GDX-DHT groups were not pursued.

Discussion

The prefrontal areas in rats are a functional mosaic of cortical zones, each making unique as well as orchestrated contributions to cognitive information processing. The importance of DA innervation for these computations in rats, as in primates, is illustrated in studies showing that either supranormal stimulation (Murphy et al., 1996; Verma and Moghaddam, 1996; Zahrt et al., 1997) or chemical lesions of prefrontal DA inputs (Stam et al., 1989; Kalsbeek et al., 1990; Wilcott and Xuemei, 1990) can induce behavioral deficits as severe as those produced by cortical ablations. These findings suggest that any influence thatmoves DA away from baseline can have negative functional consequences. Previous studies showing that long-term gonadectomy in adult male rats significantly increases the density of TH-ir but not DBH-ir axons in representative association but not sensory or motor areas (Kritzer et al., 1999; Kritzer, 2000) and induces deficits in working memory tasks (Beatty, 1984; Van Hest et al., 1988; Kritzer et al., 2001) tentatively identify gonadal steroids as such an influence. The present analyses of catecholamine, indolamine and cholinergic innervation in four additional prefrontal areas of the rostral frontal lobe confirm and extend previous findings indicating that in adult male rats, long-term gonadectomy may also be a highly selective means of modulating the functionally critical prefrontal DA afferents.

Some Methodological Considerations

The present data were derived from qualitative and quantitative analyses of tissue sections immunocytochemically labeled for either a neurotransmitter (5-HT), or for enzymes involved in neurotransmitter synthesis (TH, DBH, CHAT). In order to equate changes seen in these profiles with hormone effects on particular neurochemically identified cortical afferent systems, the antibodies used must specifically label appropriate cortical elements. Whereas CHAT, 5-HT and DBH are, respectively, selective markers for cholinergic, serotonergic and noradrenergic axons in the rat cerebrum, the issue is less certain for TH-ir, as TH itself is present in both DA- and NA-containing cortical axons. Importantly, however, the axons labeled in our studies by the anti-TH antibody were similar in morphology, distribution and/or density to cortical afferents identified in rats using TH-ir in animals in which NA afferents had been selectively depleted (Berger et al., 1974). Further, these TH-ir axons were clearly distinguished from DBH-ir axons by their morphologies, densities and, seemingly, by their responses to gonadectomy. Thus, it is tentatively concluded that TH-ir in this study largely represents cortical DA innervation (Adler et al., 1999; Kritzer, 2000). Because gonadectomy altered the density of these TH-ir but not DBH-ir afferents, it can be further assumed that any cross-reactivity that might occur would, at worst, produce an underestimate of effects on the presumed DA afferents.

While the methods used in this study may also be less direct than, for example, high-performance liquid chromatography (HPLC) measures of neurotransmitter content, they also held certain advantages. For example, in addition to a higher level of spatial resolution, the anatomical approach obviated the need to pool samples across animals, thus enabling assessments of intra- and inter-animal variance to be made and statistically eliminated as significant sources of variability in group data. Because afferent systems were also amenable to qualitative evaluation, some inference was possible regarding contributions of, for example, axon outgrowth or dying back to observed changes in axon density (see below). Finally, there were also transmitter-specific advantages to examining immunolabeled profiles in intact sections. For example, the 5-HT innervation of the rat cortex is derived from distinct, dual sources — the dorsal and medial raphae nuclei (Kosofsky and Molliver, 1987) — which are differentially sensitive to perturbations such as amphetamine-induced neurotoxicity (Mamounas et al., 1991). Because axons arising from these two nuclei are also morphologically distinct, both types could be tentatively identified in this study. Further, qualitative evaluations suggested that both axon subpopulations were present in all animal groups in seemingly similar proportions. Thus, it seems that neither the fine, relatively smooth 5-HT-ir fibers that arise from the dorsal raphae, nor the larger diameter, more conspicuously beaded axons of presumed median raphae origin (Kosofsky and Molliver, 1987) were appreciably affected by hormone manipulations. It was not possible, however, to distinguish between afferent and intrinsic cholinergic axons in the cortical areas investigated. Because there were no obvious qualitative or quantitative differences in the appearance of CHAT-ir across animal groups, it is presumed that neither axon population was affected by long-term gonadectomy. However, the perhaps unlikely possibility that there were offsetting shifts in axon densities between these two fiber populations cannot be excluded. In addition, the basic measure of axon density utilized throughout this study should be considered in relation to contexts of prefrontal cortical volume, cell size and cell density, which if affected by gonadal hormone manipulations to a large enough degree would confound interpretation of the axon density differences and similarities reported. Unbiased quantification of these parameters was not possible in the animals included in this study. However, as previous stereological analyses of the volume, cell number, cell size and cell density of prefrontal (CG1), sensory (Par 1) and motor (AGl) areas found no significant effects of hormone treatment in adult male rats gonadectomized on the day of birth (Venkatesan, 2000), appreciable differences in corresponding parameters in the prefrontal areas examined here under conditions of much shorter hormone manipulations seem unlikely.

A Note on Possible Sex Differences

Among studies describing effects of gonadal steroids on the cholinergic, catecholaminergic and indolaminergic innervations of the rat cerebral cortex are separate studies that explore effects in male or female subjects. Although these studies are often also distinguished by differences in experimental design and/or endpoints of neurotransmitter systems measured, comparisons nonetheless offer some indication that cortical neurotransmitters may be more responsive to changes in gonadal hormones in females. For example, whereas cholinergic innervation levels are essentially unchanged by gonadectomy in males (present study), in females ovariectomy decreases CHAT activity and high affinity choline uptake in the cerebral cortex (Luine, 1985; O’Malley et al., 1987; Singh et al., 1994) and CHAT immunoreactivity in nuclei of the basal forebrain nuclei, including those housing the cells of origin of cholinergic cortical afferents (Gibbs, 1997). Likewise, cortical 5-HT levels, which are also largely unresponsive to gonadectomy in males — as seen in the present study and elsewhere (Battaner et al., 1987; Fink et al., 1998) — are negatively correlated with circulating progesterone levels and, to a lesser extent, with circulating E levels in pregnant and postpartum female rats (Glaser et al., 1990). Disparities also seem to apply to the case of cortical DA innervation, which is significantly increased by chronic gonadectomy in adult male rats — as seen in the present study and elsewhere (Battaner et al., 1987) — and significantly decreased by long-term ovariectomy in females (DuPont et al., 1981). Thus, while far from proven, there are intriguing hints in the literature suggesting the possibility for sex differences in the hormonal regulation of cortical neuro-transmitter systems, including most of those investigated here. Accordingly, in placing the findings of this study within more detailed contexts of the existing literature, discussions have been limited largely to studies carried out in males.

Hormone Regulation of Cortical Neurochemistry in Adult Male Rats: Comparisons with Previous Studies

Gonadal hormones influence aspects of neurotransmitter systems in the prefrontal association cortices of adult male rats ranging from effects on stimulated turnover (Handa et al., 1997) to receptor binding (Cyr et al., 1998; Sumner and Fink, 1998). However, the studies that provide closest technical matches for the present analyses are those measuring cortical transmitter levels. These include one study in adult male rats that had been gonadectomized 3 weeks prior to the study, a hormone manipulation also similar to our own 4 week paradigm. In this study, HPLC analyses of cortical homogenates showed that chronic gonadectomy significantly increased levels of DA and its metabolite 3,4-dihydroxy phenylacetic acid (DOPAC) in a TP-reversible manner, but that measures of 5-HT, its major metabolite 5-hydroxyindolacetic acid (5-HIIA) and of NA were unchanged. Although the present studies used a different approach to assessing cortical transmitters, there nonetheless seems to be good agreement that long-term gonadectomy stimulates levels of DA, but has no appreciable effect on the noradrenergic or serotonergic afferents of the adult male rat cerebrum. Other studies suggest that these transmitter-specific patterns of hormone sensitivity and insensitivity may become established early on during postnatal life (Siddiqui and Gilmore, 1988; Stewart et al., 1991; Stewart and Rajabi, 1994; Siddiqui and Shah, 1997).

The present studies also extend earlier evidence for a regional selectivity of the transmitter-specific effects of long-term gonadectomy in the male rat cerebrum. Thus, in earlier work, the effects of chronic hormone deprivation and replacement on TH-ir and DBH-ir axons were assessed within the primary motor, primary somatosensory, premotor, anterior cingulate and anterior insular cortices at the level of the mid-septal nucleus (Adler et al., 1999; Kritzer et al., 1999; Kritzer, 2000). These studies revealed that long-term gonadectomy only affected presumed dopaminergic afferents and only in the deep layers of the two prefrontal areas examined (Adler et al., 1999; Kritzer et al., 1999; Kritzer 2000). The present study extended findings of the insensitivity of DA afferents to hormone manipulation among premotor areas, as afferents in area PrCm, the suspected homologue of the frontal eye fields in primates (Sesack et al., 1989), were not appreciably different between the gonadectomized and hormonally intact animals. They also affirmed and extended the vulnerability of prefrontal DA afferents by revealing significantly higher than normal TH-ir axon densities in areas IL, PL and ACd in the gonadectomized group. However, a major difference emerged in the laminar patterns of the hormone sensitivity of DA afferents in pregenual versus supragenual prefrontal fields. Specifically, whereas effects in the supragenual areas were limited to infragranular cortical layers (Adler et al., 1999; Kritzer et al., 1999), in the more rostral prefrontal areas axons in upper and lower layers were equally affected by long-term gonadectomy (Table 3). As discussed below, these regional and laminar patterns of hormone sensitivity and insensitivity show striking parallels with the distributions of the so-called class 1 and class 2 DA afferents (Berger et al., 1991).

Hormone Sensitivity and DA Heterogeneity

Specific aspects of anatomy and/or physiology differentiate the DA innervations of different cytoarchitectonic regions of the cerebral cortex. In rats, one of the classification schemes used for dividing up these afferents identifies DA inputs as either class1 afferents, distinguished by prenatal development, colocalization with neurotensin and relatively low rates of DA utilization, or class 2 axons that develop postnatally, do not colocalize neurotensin and have higher rates of DA metabolism (Berger et al., 1991; Gaspar et al., 1991; Berger, 1992). Mapping studies have shown that these two axon classes are also differentially localized within the cortical mantle. Thus, class 2 afferents innervate all layers of motor, premotor and sensory areas, as well as the supragranular layers of supragenual cingulate cortex. In each of these areas, DA axons have been shown tobe insensitive to long-term gonadectomy, both here and elsewhere (Adler et al., 1999; Kritzer et al., 1999). Class 1 afferents, on the other hand, innervate the deep layers of the supragenual cingulate cortex, as well as all layers of the more rostral prefrontal fields — in essence, those areas where TH-ir afferents respond most vigorously to hormone manipulation (Adler et al., 1999; Kritzer et al., 1999). Thus, like the examples of differences in hormone sensitivity among incertohypothalamic and tuberoinfundibular hypothalamic DA systems (Lookingland and Moore, 1984), the class 1 and class 2 DA afferents of the rat cerebral cortex seem to represent differentially hormone sensitive subsets of mesocortical DA neurons. Other physiological divisions that have been identified among prefrontally projecting midbrain DA neurons, however, do not have obvious parallels in differential hormone response. The DA afferents of area IL, for example, which show unique increases in transmitter release upon exposure to mild stress (Deutch, 1991, 1993) or administration of β-carbolines (Murphy et al., 1996), exhibited patterns of hormone sensitivity in this study that were not appreciably different from those of other pregenual prefrontal afferent fields that are less responsive to stressful stimuli.

Possible Mechanisms

A main finding of this study was that the density of specific subsets of presumed DA afferents projecting to the prefrontal cortices in adult male rats was significantly and, at least among 5-HT, ACh and NA afferents, selectively increased by long-term gonadectomy. However, other than an upward shift in density, TH-ir axons seemed to be oriented, shaped and distributed much as they were in controls. With no obvious signs of, for example, growth-cone-like profiles, aberrant axon trajectories, or axon spinules, it seems most likely that gonadectomy is acting to increase levels of TH within existing arbors, thus rendering them more detectable by immunocytochemistry than in controls. The attenuation of the density effects by supplementing animals with exogenous steroid also identifies changes in hormone levels as probable causative factors. This, in turn, focuses mechanistic possibilities on the genomic and/or non-genomic actions of androgen and estrogen signaling pathways. The effectiveness of supplementing animals with DHT, a non-aromatizable derivative of testosterone — and the contrasting ineffectiveness of E — may point more specifically to roles for androgens. In the absence of any available evidence for receptor subtypes or non-genomic actions, the intracellular androgen receptors identified within subsets of midbrain DA neurons in the ventral VTA, dorsomedial substantia nigra and retrorubral fields of the rat (Kritzer, 1997) remain likely candidates as target sites of relevant hormone stimulation. However, despite a negative outcome with E replacement, it may be a mistake to rule out contributions from estrogen signaling altogether. It is possible, for example, that the continuous, relatively constant levels of estrogen exposure experienced by the hormonally replaced animals down-regulated estrogen receptors. This possibility should be explored using some sort of pulsatile administration of EB, particularly as significant numbers of DA neurons in midbrain nuclei that give rise to mesocortical afferents in rats have recently been shown tocontain the β-subtype of the intracellular estrogen receptor (Creutz and Kritzer, 2002).

Summary and Conclusions

Gonadal hormones are increasingly identified as factors important in the biology of cortical information processing. The studies presented here combine with earlier work to suggest that one means of stimulation relevant for the cognitive functions of the prefrontal cortices in particular may be a robust and highly selective influence over one of its most functionally critical inputs — its DA innervation. Thus, studies now more comprehensive in terms of identified prefrontal divisions of the rat cortex and inclusive of serotonergic, cholinergic, noradrenergic and dopaminergic afferents, indicate that long-term changes in circulating hormone levels produce significant increases in axon density that seem to target the so-called class 1 DA afferents in a seemingly androgen-sensitive manner. Although questions remain regarding the sensitivity of, for example, the amino acid transmitters and the myriad of interneurons that populate the cerebral cortices, the present findings nonetheless build toward rather than erode evidence for a highly selective influence of gonadal steroids within the prefrontal cortices that targets a transmitter nearly synonymous with their normal operations (Brozoski et al., 1979; Stam et al., 1989) and frequently implicated in their dysfunction in disorders such as schizophrenia as well (Davis et al., 1991).

Table 1

Mean percentage whole body wt represented by bulbocavernosus muscle (BCN) mass (±SD)

Group (n = 5) BCN (g)/whole body wt (g) (mean ± SD) 
The androgen-sensitive BCN shrink by >50% after long-term gonadectomy (GDX) in animals supplemented with placebo (pl) or estradiol (E) and are proportionately smaller than hormonally intact controls (CTRL) in these groups. Supplementing GDX animals with TP or its non-aromatizable derivative DHT prevents BCN shrinkage. Asterisks denote proportionate muscle masses that are significantly different from control (ANOVA, Student–Newman–Keuls, P < 0.05). 
CTRL 0.0033 ± 0.00016 
GDX-pl 0.0012 ± 0.00008* 
GDX-E 0.0015 ± 0.00013* 
GDX-DHT 0.0041 ± 0.00056 
GDX-TP 0.0039 ± 0.00046 
Group (n = 5) BCN (g)/whole body wt (g) (mean ± SD) 
The androgen-sensitive BCN shrink by >50% after long-term gonadectomy (GDX) in animals supplemented with placebo (pl) or estradiol (E) and are proportionately smaller than hormonally intact controls (CTRL) in these groups. Supplementing GDX animals with TP or its non-aromatizable derivative DHT prevents BCN shrinkage. Asterisks denote proportionate muscle masses that are significantly different from control (ANOVA, Student–Newman–Keuls, P < 0.05). 
CTRL 0.0033 ± 0.00016 
GDX-pl 0.0012 ± 0.00008* 
GDX-E 0.0015 ± 0.00013* 
GDX-DHT 0.0041 ± 0.00056 
GDX-TP 0.0039 ± 0.00046 
Table 2

Mean serum levels (±SE) of TP, DHT or E

Group (n = 5) TP (ng/ml; mean ± SD) DHT (pg/ml; mean ± SD) E (pg/ml; mean ± SD) 
All serum levels are expressed as mean ng/ml or pg/ml. Circulating levels of TP and DHT in the gonadectomized animals (GDX-pl) are significantly lower than controls (asterisks, ANOVA, Student–Newman–Keuls, P < 0.05). Levels of E in the GDX-pl cohort are also lower than in controls, but not significantly so. Levels of TP, DHT and E in gonadectomized animals replaced with these steroids (GDX-TP, GDX-DHT, GDX-E) approximate values in control males (TP, DHT), or as reported for females in the literature (E). 
CTRL  9.08 ± 3.23 764 ± 240.8  2.2 ± 0.58 
GDX-pl  0.38 ± 0.07*  18.6 ± 6.06*  1.4 ± 0.51 
GDX-E – – 52.4 ± 13.9 
GDX-DHT –  529 ± 35.2 – 
GDX-TP 12.86 ± 2.66 1410 ± 357.2 11.6 ± 3.29 
Group (n = 5) TP (ng/ml; mean ± SD) DHT (pg/ml; mean ± SD) E (pg/ml; mean ± SD) 
All serum levels are expressed as mean ng/ml or pg/ml. Circulating levels of TP and DHT in the gonadectomized animals (GDX-pl) are significantly lower than controls (asterisks, ANOVA, Student–Newman–Keuls, P < 0.05). Levels of E in the GDX-pl cohort are also lower than in controls, but not significantly so. Levels of TP, DHT and E in gonadectomized animals replaced with these steroids (GDX-TP, GDX-DHT, GDX-E) approximate values in control males (TP, DHT), or as reported for females in the literature (E). 
CTRL  9.08 ± 3.23 764 ± 240.8  2.2 ± 0.58 
GDX-pl  0.38 ± 0.07*  18.6 ± 6.06*  1.4 ± 0.51 
GDX-E – – 52.4 ± 13.9 
GDX-DHT –  529 ± 35.2 – 
GDX-TP 12.86 ± 2.66 1410 ± 357.2 11.6 ± 3.29 
Table 3

Mean pixel densities (±SD) of TH-ir axons

 Layer II/III, left Layer V, left Layer I/III, right Layer V, right 
Axon density measures are derived from camera lucida drawings in layers II/III and V in left and right hemifields of areas PrCm, ACd, PL and IL of control (CTRL) and gonadectomized rats given placebo (GDX-pl), E (GDX-E), DHT (GDX-DHT), or TP (GDX-TP). Densities in GDX-pl and GDX-E rats are significantly higher than controls in areas ACd, PL and IL (asterisks, ANOVA, Student–Newman–Keuls, P < 0.05), but not in area PrCm. All values from GDX-TP and GDX-DHT rats are similar to controls. 
Area PrCm     
CTRL  6.48 ± 1.13  6.77 ± 1.30  7.17 ± 1.65  7.09 ± 1.11 
GDX-pl  6.98 ± 1.71  7.00 ± 1.25  7.54 ± 2.18  6.92 ± 1.56 
GDX-E  7.06 ± 1.80  6.78 ± 1.32  6.74 ± 2.91  6.29 ± 2.65 
GDX-DHT  7.28 ± 1.52  6.31 ± 1.26  7.55 ± 2.43  6.06 ± 1.54 
GDX-TP  7.66 ± 1.97  6.77 ± 1.45  7.73 ± 1.81  7.51 ± 1.03 
Area ACd     
CTRL  8.18 ± 2.68  8.40 ± 1.93  8.93 ± 2.19  7.84 ± 2.38 
GDX-pl 16.34 ± 2.09* 15.13 ± 4.79* 16.49 ± 2.45* 14.50 ± 2.75* 
GDX-E 14.90 ± 2.93* 14.41 ± 1.18* 11.99 ± 3.86* 11.28 ± 1.81* 
GDX-DHT  9.08 ± 1.47  7.51 ± 1.86  9.25 ± 2.41  8.14 ± 1.43 
GDX-TP 10.21 ± 2.35  9.36 ± 1.18 10.71 ± 4.08  9.13 ± 1.06 
Area PL     
CTRL  8.28 ± 1.74  8.50 ± 2.32  8.05 ± 2.44  7.91 ± 1.70 
GDX-pl 13.11 ± 2.49* 15.75 ± 2.06* 13.30 ± 2.85* 15.78 ± 2.86* 
GDX-E 13.02 ± 2.87* 13.99 ± 2.02* 12.68 ± 1.03* 12.88 ± 3.32* 
GDX-DHT  8.58 ± 1.90  8.67 ± 0.74  7.31 ± 1.04  8.88 ± 0.71 
GDX-TP  8.54 ± 1.41  8.99 ± 1.59  8.41 ± 0.96  8.86 ± 1.71 
Area IL     
CTRL  9.31 ± 1.47   7.91 ± 1.50  
GDX-pl 13.35 ± 2.95*  13.02 ± 2.13*  
GDX-E 12.99 ± 2.94*  11.73 ± 3.00*  
GDX-DHT  7.61 ± 1.61   7.92 ± 1.32  
GDX-TP  9.68 ± 2.33   9.75 ± 2.20  
 Layer II/III, left Layer V, left Layer I/III, right Layer V, right 
Axon density measures are derived from camera lucida drawings in layers II/III and V in left and right hemifields of areas PrCm, ACd, PL and IL of control (CTRL) and gonadectomized rats given placebo (GDX-pl), E (GDX-E), DHT (GDX-DHT), or TP (GDX-TP). Densities in GDX-pl and GDX-E rats are significantly higher than controls in areas ACd, PL and IL (asterisks, ANOVA, Student–Newman–Keuls, P < 0.05), but not in area PrCm. All values from GDX-TP and GDX-DHT rats are similar to controls. 
Area PrCm     
CTRL  6.48 ± 1.13  6.77 ± 1.30  7.17 ± 1.65  7.09 ± 1.11 
GDX-pl  6.98 ± 1.71  7.00 ± 1.25  7.54 ± 2.18  6.92 ± 1.56 
GDX-E  7.06 ± 1.80  6.78 ± 1.32  6.74 ± 2.91  6.29 ± 2.65 
GDX-DHT  7.28 ± 1.52  6.31 ± 1.26  7.55 ± 2.43  6.06 ± 1.54 
GDX-TP  7.66 ± 1.97  6.77 ± 1.45  7.73 ± 1.81  7.51 ± 1.03 
Area ACd     
CTRL  8.18 ± 2.68  8.40 ± 1.93  8.93 ± 2.19  7.84 ± 2.38 
GDX-pl 16.34 ± 2.09* 15.13 ± 4.79* 16.49 ± 2.45* 14.50 ± 2.75* 
GDX-E 14.90 ± 2.93* 14.41 ± 1.18* 11.99 ± 3.86* 11.28 ± 1.81* 
GDX-DHT  9.08 ± 1.47  7.51 ± 1.86  9.25 ± 2.41  8.14 ± 1.43 
GDX-TP 10.21 ± 2.35  9.36 ± 1.18 10.71 ± 4.08  9.13 ± 1.06 
Area PL     
CTRL  8.28 ± 1.74  8.50 ± 2.32  8.05 ± 2.44  7.91 ± 1.70 
GDX-pl 13.11 ± 2.49* 15.75 ± 2.06* 13.30 ± 2.85* 15.78 ± 2.86* 
GDX-E 13.02 ± 2.87* 13.99 ± 2.02* 12.68 ± 1.03* 12.88 ± 3.32* 
GDX-DHT  8.58 ± 1.90  8.67 ± 0.74  7.31 ± 1.04  8.88 ± 0.71 
GDX-TP  8.54 ± 1.41  8.99 ± 1.59  8.41 ± 0.96  8.86 ± 1.71 
Area IL     
CTRL  9.31 ± 1.47   7.91 ± 1.50  
GDX-pl 13.35 ± 2.95*  13.02 ± 2.13*  
GDX-E 12.99 ± 2.94*  11.73 ± 3.00*  
GDX-DHT  7.61 ± 1.61   7.92 ± 1.32  
GDX-TP  9.68 ± 2.33   9.75 ± 2.20  
Table 4

Mean pixel densities (±SD) of DBH-ir fibers

 Layer II/III, left (mean ± SD) Layer V, left (mean ± SD) Layer II/III, right (mean ± SD) Layer V, right (mean ± SD) 
Axon density measures are derived from camera lucida drawings in layers II/III and V, or II/III only in left and right hemifields of medial prefrontal areas PrCm, ACd, PL and IL of control (CTRL) and gonadectomized animals given placebo (GDX-pl), E (GDX-E). DHT (GDX-DHT), or TP (GDX-TP). Density values across groups typically remained within a few % of corresponding control values; no significant group differences were seen for density measures in any region, layer or hemisphere evaluated. 
Area PrCm     
CTRL 6.62 ± 0.76 5.72 ± 0.55 6.51 ± 1.28 5.58 ± 1.15 
GDX-pl 5.83 ± 1.14 5.54 ± 1.09 6.25 ± 0.67 5.35 ± 0.91 
GDX-TP 6.42 ± 1.31 6.29 ± 0.94 6.69 ± 0.68 5.92 ± 0.61 
Area ACd     
CTRL 6.49 ± 0.99 5.33 ± 0.73 6.18 ± 1.60 5.45 ± 0.87 
GDX-pl 6.22 ± 1.18 5.31 ± 0.80 5.92 ± 0.44 4.99 ± 0.64 
GDX-TP 5.77 ± 1.16 6.45 ± 1.48 6.09 ± 0.83 5.44 ± 0.64 
Area PL     
CTRL 7.68 ± 0.98 7.04 ± 0.73 7.30 ± 1.21 6.67 ± 0.80 
GDX-pl 6.86 ± 1.08 7.14 ± 0.86 6.53 ± 1.16 7.17 ± 1.09 
GDX-TP 7.44 ± 0.78 7.95 ± 1.34 7.58 ± 1.24 7.56 ± 1.13 
Area IL     
CTRL 8.70 ± 1.30  9.38 ± 1.52  
GDX-pl 9.07 ± 1.48  8.75 ± 0.81  
GDX-TP 9.33 ± 0.77  9.78 ± 1.04  
 Layer II/III, left (mean ± SD) Layer V, left (mean ± SD) Layer II/III, right (mean ± SD) Layer V, right (mean ± SD) 
Axon density measures are derived from camera lucida drawings in layers II/III and V, or II/III only in left and right hemifields of medial prefrontal areas PrCm, ACd, PL and IL of control (CTRL) and gonadectomized animals given placebo (GDX-pl), E (GDX-E). DHT (GDX-DHT), or TP (GDX-TP). Density values across groups typically remained within a few % of corresponding control values; no significant group differences were seen for density measures in any region, layer or hemisphere evaluated. 
Area PrCm     
CTRL 6.62 ± 0.76 5.72 ± 0.55 6.51 ± 1.28 5.58 ± 1.15 
GDX-pl 5.83 ± 1.14 5.54 ± 1.09 6.25 ± 0.67 5.35 ± 0.91 
GDX-TP 6.42 ± 1.31 6.29 ± 0.94 6.69 ± 0.68 5.92 ± 0.61 
Area ACd     
CTRL 6.49 ± 0.99 5.33 ± 0.73 6.18 ± 1.60 5.45 ± 0.87 
GDX-pl 6.22 ± 1.18 5.31 ± 0.80 5.92 ± 0.44 4.99 ± 0.64 
GDX-TP 5.77 ± 1.16 6.45 ± 1.48 6.09 ± 0.83 5.44 ± 0.64 
Area PL     
CTRL 7.68 ± 0.98 7.04 ± 0.73 7.30 ± 1.21 6.67 ± 0.80 
GDX-pl 6.86 ± 1.08 7.14 ± 0.86 6.53 ± 1.16 7.17 ± 1.09 
GDX-TP 7.44 ± 0.78 7.95 ± 1.34 7.58 ± 1.24 7.56 ± 1.13 
Area IL     
CTRL 8.70 ± 1.30  9.38 ± 1.52  
GDX-pl 9.07 ± 1.48  8.75 ± 0.81  
GDX-TP 9.33 ± 0.77  9.78 ± 1.04  
Table 5

Mean pixel densities (±SD) of 5-HT-ir fibers

 Layer II/III, left (mean ± SD) Layer V, left (mean ± SD) Layer II/III, right (mean ± SD) Layer V, right (mean ± SD) 
Axon density measures are derived from camera lucida drawings in layers II/III and V, or II/III only in left and right hemifields of medial prefrontal areas PrCm, ACd, PL and IL of control (CTRL) and gonadectomized animals given placebo (GDX-pl), E (GDX-E). DHT (GDX-DHT), or TP (GDX-TP). Density values across groups typically remained within a few % of corresponding control values; no significant group differences were seen for density measures in any region, layer or hemisphere evaluated. 
Area PrCm     
CTRL 20.43 ± 2.16 16.68 ± 2.69 19.85 ± 2.57 17.74 ± 3.55 
GDX-pl 20.10 ± 1.54 17.33 ± 1.44 16.04 ± 2.10 19.88 ± 2.42 
GDX-TP 21.94 ± 2.64 20.29 ± 3.04 20.46 + 1.31 21.76 ± 2.62 
Area ACd     
CTRL 18.68 ± 2.58 17.27 ± 2.78 19.76 ± 1.28 16.05 ± 2.34 
GDX-pl 18.30 ± 1.56 16.81 ± 0.76 18.35 ± 1.99 16.98 ± 2.65 
GDX-TP 21.49 ± 2.84 19.68 ± 1.23 19.88 ± 1.67 17.05 ± 0.96 
Area PL     
CTRL 21.03 ± 2.23 18.87 ± 2.20 20.51 ± 1.05 18.57 ± 1.18 
GDX-pl 20.84 ± 0.75 18.24 ± 1.34 21.20 ± 2.97 19.13 ± 2.44 
GDX-TP 24.78 ± 4.96 21.64 ± 2.89 19.81 ± 2.86 18.86 ± 1.10 
Area IL     
CTRL 22.58 ± 2.10  24.82 ± 2.42  
GDX-pl 22.74 ± 2.56  24.77 ± 2.07  
GDX-TP 26.63 ± 5.00  23.50 ± 0.86  
 Layer II/III, left (mean ± SD) Layer V, left (mean ± SD) Layer II/III, right (mean ± SD) Layer V, right (mean ± SD) 
Axon density measures are derived from camera lucida drawings in layers II/III and V, or II/III only in left and right hemifields of medial prefrontal areas PrCm, ACd, PL and IL of control (CTRL) and gonadectomized animals given placebo (GDX-pl), E (GDX-E). DHT (GDX-DHT), or TP (GDX-TP). Density values across groups typically remained within a few % of corresponding control values; no significant group differences were seen for density measures in any region, layer or hemisphere evaluated. 
Area PrCm     
CTRL 20.43 ± 2.16 16.68 ± 2.69 19.85 ± 2.57 17.74 ± 3.55 
GDX-pl 20.10 ± 1.54 17.33 ± 1.44 16.04 ± 2.10 19.88 ± 2.42 
GDX-TP 21.94 ± 2.64 20.29 ± 3.04 20.46 + 1.31 21.76 ± 2.62 
Area ACd     
CTRL 18.68 ± 2.58 17.27 ± 2.78 19.76 ± 1.28 16.05 ± 2.34 
GDX-pl 18.30 ± 1.56 16.81 ± 0.76 18.35 ± 1.99 16.98 ± 2.65 
GDX-TP 21.49 ± 2.84 19.68 ± 1.23 19.88 ± 1.67 17.05 ± 0.96 
Area PL     
CTRL 21.03 ± 2.23 18.87 ± 2.20 20.51 ± 1.05 18.57 ± 1.18 
GDX-pl 20.84 ± 0.75 18.24 ± 1.34 21.20 ± 2.97 19.13 ± 2.44 
GDX-TP 24.78 ± 4.96 21.64 ± 2.89 19.81 ± 2.86 18.86 ± 1.10 
Area IL     
CTRL 22.58 ± 2.10  24.82 ± 2.42  
GDX-pl 22.74 ± 2.56  24.77 ± 2.07  
GDX-TP 26.63 ± 5.00  23.50 ± 0.86  
Table 6

Mean pixel densities (±SD) of CHAT-ir fibers

 Layer II/III, left (mean ± SD) Layer V, left (mean ± SD) Layer II/III, right (mean ± SD) Layer V, right (mean ± SD) 
Axon density measures are derived from camera lucida drawings in layers II/III and V, or II/III only in left and right hemifields of medial prefrontal areas PrCm, ACd, PL and IL of control (CTRL) and gonadectomized animals given placebo (GDX-pl), E (GDX-E). DHT (GDX-DHT), or TP (GDX-TP). Density values across groups typically remained within a few % of corresponding control values; no significant group differences were seen for density measures in any region, layer or hemisphere evaluated. Numbers in parentheses indicate the range across animals for the number of immunopositive somata present in the counting field. 
Area PrCm     
CTRL 29.51 ± 0.54 (5–7) 26.248 ± 1.33 (2–3) 30.37 ± 4.18 (1–7) 26.58 ± 0.52 (0–1) 
GDX-pl 28.75 ± 1.09 (3–4) 31.33 ± 3.46 (0–1) 27.78 ± 0.43 (2–4) 28.59 ± 1.78 (0–2) 
GDX-TP 31.94 ± 3.92 (3–6) 29.28 ± 2.71 (0–1) 29.71 ± 1.03 (4–6) 26.74 ± 2.50 (1–2) 
Area ACd     
CTRL 24.98 ± 5.84 (4–7) 26.40 ± 3.95 (1–3) 27.34 ± 4.54 (5–6) 24.41 ± 3.23 (1–2) 
GDX-pl 33.36 ± 5.09 (6–7) 33.69 ± 5.16 (1–3) 32.47 ± 1.56 (5–6) 29.97 ± 1.51 (0–1) 
GDX-TP 29.85 ± 1.81 (6–9) 30.47 ± 1.32 (0–1) 34.24 ± 2.53 (5–6) 30.92 ± 3.09 (1–2) 
Area PL     
CTRL 29.89 ± 2.32 (7–9) 29.63 ± 2.71 (1–2) 26.47 ± 5.59 (6–9) 28.75 ± 1.43 (0–1) 
GDX-pl 34.52 ± 1.61 (6–7) 32.12 ± 4.15 (0–1) 34.95 ± 3.01 (4–5) 32.74 ± 0.82 (0–1) 
GDX-TP 33.25 ± 4.04 (7–9) 34.24 ± 1.73 (1–2) 30.12 ± 0.48 (8–9) 32.53 ± 4.26 (1–3) 
Area IL     
CTRL 30.88 ± 2.32 (3–6)  28.05 ± 4.42 (6–7)  
GDX-pl 30.07 ± 2.72 (2–5)  30.50 ± 1.59 (3–7)  
GDX-TP 33.09 ± 3.78 (5–6)  32.17 ± 0.81 (4–8)  
 Layer II/III, left (mean ± SD) Layer V, left (mean ± SD) Layer II/III, right (mean ± SD) Layer V, right (mean ± SD) 
Axon density measures are derived from camera lucida drawings in layers II/III and V, or II/III only in left and right hemifields of medial prefrontal areas PrCm, ACd, PL and IL of control (CTRL) and gonadectomized animals given placebo (GDX-pl), E (GDX-E). DHT (GDX-DHT), or TP (GDX-TP). Density values across groups typically remained within a few % of corresponding control values; no significant group differences were seen for density measures in any region, layer or hemisphere evaluated. Numbers in parentheses indicate the range across animals for the number of immunopositive somata present in the counting field. 
Area PrCm     
CTRL 29.51 ± 0.54 (5–7) 26.248 ± 1.33 (2–3) 30.37 ± 4.18 (1–7) 26.58 ± 0.52 (0–1) 
GDX-pl 28.75 ± 1.09 (3–4) 31.33 ± 3.46 (0–1) 27.78 ± 0.43 (2–4) 28.59 ± 1.78 (0–2) 
GDX-TP 31.94 ± 3.92 (3–6) 29.28 ± 2.71 (0–1) 29.71 ± 1.03 (4–6) 26.74 ± 2.50 (1–2) 
Area ACd     
CTRL 24.98 ± 5.84 (4–7) 26.40 ± 3.95 (1–3) 27.34 ± 4.54 (5–6) 24.41 ± 3.23 (1–2) 
GDX-pl 33.36 ± 5.09 (6–7) 33.69 ± 5.16 (1–3) 32.47 ± 1.56 (5–6) 29.97 ± 1.51 (0–1) 
GDX-TP 29.85 ± 1.81 (6–9) 30.47 ± 1.32 (0–1) 34.24 ± 2.53 (5–6) 30.92 ± 3.09 (1–2) 
Area PL     
CTRL 29.89 ± 2.32 (7–9) 29.63 ± 2.71 (1–2) 26.47 ± 5.59 (6–9) 28.75 ± 1.43 (0–1) 
GDX-pl 34.52 ± 1.61 (6–7) 32.12 ± 4.15 (0–1) 34.95 ± 3.01 (4–5) 32.74 ± 0.82 (0–1) 
GDX-TP 33.25 ± 4.04 (7–9) 34.24 ± 1.73 (1–2) 30.12 ± 0.48 (8–9) 32.53 ± 4.26 (1–3) 
Area IL     
CTRL 30.88 ± 2.32 (3–6)  28.05 ± 4.42 (6–7)  
GDX-pl 30.07 ± 2.72 (2–5)  30.50 ± 1.59 (3–7)  
GDX-TP 33.09 ± 3.78 (5–6)  32.17 ± 0.81 (4–8)  
Figures 1–3.

Representative digital photomicrographs showing the morphology, distribution and apparent density of axons immunoreactive for TH in layers II/III of area IL (Fig. 1), area PL (Fig. 2) and area ACd (Fig. 3) in control (A, CTRL) and gonadectomized rats supplemented with placebo (B, GDX-pl), DHT (C, GDX-DHT), E (D, GDX-E) or TP (E, GDX-TP). Images are all from left hemispheres and are oriented to match the inset diagrams showing the location of these areas, i.e. with the pial surface to the right (IL, PL) or up and to the right (ACd). Axons in all animals in all areas include both thick, relatively smooth fibers and thinner, more varicose axons, that are oriented either randomly (IL, ACd) or more perpendicularly to the pial surface (PL). However, axon density in each of these areas is noticeably higher than normal in sections from GDX-pl and GDX-E rats. Scale bars = 20 μm.

Figures 1–3.

Representative digital photomicrographs showing the morphology, distribution and apparent density of axons immunoreactive for TH in layers II/III of area IL (Fig. 1), area PL (Fig. 2) and area ACd (Fig. 3) in control (A, CTRL) and gonadectomized rats supplemented with placebo (B, GDX-pl), DHT (C, GDX-DHT), E (D, GDX-E) or TP (E, GDX-TP). Images are all from left hemispheres and are oriented to match the inset diagrams showing the location of these areas, i.e. with the pial surface to the right (IL, PL) or up and to the right (ACd). Axons in all animals in all areas include both thick, relatively smooth fibers and thinner, more varicose axons, that are oriented either randomly (IL, ACd) or more perpendicularly to the pial surface (PL). However, axon density in each of these areas is noticeably higher than normal in sections from GDX-pl and GDX-E rats. Scale bars = 20 μm.

Figure 4.

Representative digital photomicrographs showing the morphology, distribution and apparent density of axons immunoreactive for TH in layers II/III of area PrCm in control (A, CTRL) and gonadectomized rats supplemented with placebo (B, GDX-pl), DHT (C, GDX-DHT), E (D, GDX-E), or TP (E, GDX-TP). Images are all from left hemispheres and are oriented to match the inset diagram (lower left) showing the location of area PrCM, i.e. with the pial surface towards the top. Axons in area PrCm in control (A) and gonadectomized rats (BE) alike include both thick, relatively smooth axons and thinner more beaded processes, all of which run mainly perpendicular to the pial surface. Axons were also present in similar densities across all animal groups. Scale bars = 20 μm.

Figure 4.

Representative digital photomicrographs showing the morphology, distribution and apparent density of axons immunoreactive for TH in layers II/III of area PrCm in control (A, CTRL) and gonadectomized rats supplemented with placebo (B, GDX-pl), DHT (C, GDX-DHT), E (D, GDX-E), or TP (E, GDX-TP). Images are all from left hemispheres and are oriented to match the inset diagram (lower left) showing the location of area PrCM, i.e. with the pial surface towards the top. Axons in area PrCm in control (A) and gonadectomized rats (BE) alike include both thick, relatively smooth axons and thinner more beaded processes, all of which run mainly perpendicular to the pial surface. Axons were also present in similar densities across all animal groups. Scale bars = 20 μm.

Figures 5–8.

Representative digital photomicrographs showing the morphology, distribution and apparent density of axons immunoreactive for 5-HT (AC), DBH (DF) and CHAT (GI) in layers II/III of area IL (Fig. 5), area PL (Fig. 6) area ACd (Fig. 7), or area PrCm (Fig. 8) in control (A, D, G, CTRL) and gonadectomized rats supplemented with placebo (B, E, H, GDX-pl), or testosterone propionate (C, F, I, GDX-TP). Images are all from left hemispheres and are oriented to match the locations of these areas in the frontal lobe, i.e. with the pial surface to the right (IL, PL) or up and to the right (ACd, PrCm). In each of these areas, 5-HT axons are dense and randomly distributed, DBH axons are relatively sparse and criss-cross the section at near right angles to one another and CHAT immunoreactivity corresponds to a loose meshwork of extremely thin, threadlike axons that are scattered among populations of CHAT-ir interneurons (G, H, I, arrows). These patterns and the apparent densities of all axon types were similar in all areas and in all animal groups, regardless of hormone treatment. Scale bars = 30 μm.

Figures 5–8.

Representative digital photomicrographs showing the morphology, distribution and apparent density of axons immunoreactive for 5-HT (AC), DBH (DF) and CHAT (GI) in layers II/III of area IL (Fig. 5), area PL (Fig. 6) area ACd (Fig. 7), or area PrCm (Fig. 8) in control (A, D, G, CTRL) and gonadectomized rats supplemented with placebo (B, E, H, GDX-pl), or testosterone propionate (C, F, I, GDX-TP). Images are all from left hemispheres and are oriented to match the locations of these areas in the frontal lobe, i.e. with the pial surface to the right (IL, PL) or up and to the right (ACd, PrCm). In each of these areas, 5-HT axons are dense and randomly distributed, DBH axons are relatively sparse and criss-cross the section at near right angles to one another and CHAT immunoreactivity corresponds to a loose meshwork of extremely thin, threadlike axons that are scattered among populations of CHAT-ir interneurons (G, H, I, arrows). These patterns and the apparent densities of all axon types were similar in all areas and in all animal groups, regardless of hormone treatment. Scale bars = 30 μm.

The outstanding technical support of Mr Alex Adler, Mr James Tsiakos and Mr Michael Ukaegbu is gratefully acknowledged. Ms Meaghan Casey is also thanked for her assistance in initiating the studies using serotonin antibodies. This work was supported by an R01 Award (NS4196607) to M.F.K.

References

Adler A, Vescovo P, Robinson JK, Kritzer MF (
1999
) Gonadectomy in adult life increases tyrosine hydroxylase immunoreactivity in the prefrontal cortex and decreases open field activity in male rats.
Neuroscience
 
89
:
939
–954.
Battaner E, Del Casillo AR, Guerra M, Mas M (
1987
) Gonadal influences on spinal cord and brain monoamines in male rats.
Brain Res.
 
425
:
391
–394.
Beatty WW (
1984
) Hormonal organization of sex differences in play fighting and spatial behavior.
Prog Brain Res
 
61
:
315
–329.
Belanger A, Cusan L, Caron S, Barden N, Dupont A (
1981
) Ovarian progestins, androgens and estrogen throughout the 4-day estrous cycle in the rat.
Biol Reprod
 
24
:
591
–596.
Berger B (
1992
) Dopamine innervation of the frontal cerebral cortex: evolutionary trends and functional implications.
Adv Neurol
 
57
:
525
–544.
Berger B, Tassin JP, Blanc G, Moyne M, Thierry T (
1974
) Histochemical confirmation for dopaminergic innervation of the rat cerebral cortex after destruction of the noradrenergic ascending pathways.
Brain Res
 
81
:
332
–337.
Berger B, Verney C, Febvret A, Vigny A, Helle KB (
1985
) Postnatal ontogenesis of the dopaminergic innervation in the rat anterior cingulate cortex (area 24). Immunocytochemical and catecholamine fluorescence histochemical analysis.
Dev Brain Res
 
21
:
31
–47.
Berger B, Gaspar P, Verney C (
1991
) Dopaminergic innervation of the cerebral cortex: unexpected differences between rodents and primates.
Trends Neurosci
 
14
:
21
–27.
Brozoski TJ, Brown RM, Rosvold HE, Goldman PS (
1979
) Cognitive deficit caused by regional depletion of dopamine in prefrontal cortex of rhesus monkey.
Science
 
205
:
929
–932.
Christiansen K, Knussman R (
1987
) Sex hormones and cognitive functioning in men.
Neuropsychobiology
 
18
:
27
–36.
Clark AS, Goldman-Rakic PS (
1989
) Gonadal hormones influence the emergence of cortical function in nonhuman primates.
Behav Neurosci
 
103
:
1287
–1295.
Creutz LM, Kritzer MF (
2002
) Estrogen receptor-β immunoreactivity in the midbrain of adult rats: regional, subregional and cellular localization in the A10, A9 and A8 dopamine cell groups.
J Comp Neurol
 
446
:
288
–300.
Cyr M, Bosse R, Di Paolo T (
1998
) Gonadal hormones modulate 5-hydroxytryptamine 2A receptors: emphasis on the rat frontal cortex.
Neuroscience
 
83
:
829
–836.
Davis KL, Kahn RS, Ko G, Davidson M (
1991
) Dopamine in schizophrenia: a review and reconceptualization.
Am J Psychiatry
 
148
:
1474
–1486.
Deutch AY (
1991
) Heterogeneity of the prefrontal cortical dopamine system in responsiveness to stress.
Soc Neurosci Abstr
 
17
:
529
.
Deutch AY (
1993
) Prefrontal cortical dopamine systems and the elaboration of functional corticostriatal circuits: implications for schizophrenia and Parkinson’s disease.
J Neural Transm
 
91
:
197
–221.
Divac I, Bjorklund A, Lindvall O, Passingham RE (
1978
) Converging projections from the mediodorsal thalamic nucleus and mesencephalic dopaminergic neurons in the neocortex of three species.
J Comp Neurol
 
180
:
59
–72.
DuPont A, diPaolo T, Gagne B, Barden N (
1981
) Effects of chronic estrogen treatment on dopamine concentrations and turnover in discrete brain nuclei of ovariectomized rats.
Neurosci Lett
 
22
:
69
–74.
Eichenbaum H, Clegg RA, Feeley A (
1983
) Reexamination of functional subdivisions of the rodent prefrontal cortex.
Exp Neurol
 
79
:
434
–451.
Fink G, Sumner BE, McQueen JK, Wilson H, Rosie R (
1998
) Sex steroid control of mood, mental state and memory.
Clin Exp Pharmacol Physiol
 
25
:
764
–775.
Gaspar P, Duyckaerts C, Alvarez C, Javoy-Agid F, Berger B (
1991
) Alternations of dopaminergic and noradrenergic innervations of motor cortex in Parkinson’s disease.
Ann Neurol
 
30
:
365
–374.
Gibbs RB (
1997
) Effects of estrogen on basal forebrain cholinergic neurons vary as a function of dose and duration of treatment.
Brain Res
 
757
:
10
–16.
Glaser J, Russell VA, de Villiers AS, Searson JA, Taljaard JJ (
1990
) Rat brain monoamine and serotonin S2 receptor changes during pregnancy.
Neurochem Res
 
15
:
949
–956.
Goldstein JM (
1988
) Gender differences in the course of schizophrenia.
Am J Psychiatry
 
145
:
684
–689.
Goldstein JM, Tsuang MT (
1990
) Gender and schizophrenia: an introduction and synthesis of findings.
Schizophrenia Bull
 
16
:
179
–183.
Groenewegen HJ (
1988
) Organization of the afferent connections of the mediodorsal thalamic nucleus in the rat, related to the mediodorsal-prefrontal topography.
Neuroscience
 
24
:
379
–431.
Hampson E (
1990
) Variations in sex-related cognitive abilities across the menstrual cycle.
Brain Cogn
 
14
:
26
–43
Handa RJ, Hejna GM, Lorens SA (
1997
) Androgen inhibits neuro-transmitter turnover in the medial prefrontal cortex of the rat following exposure to a novel environment.
Brain Res
 
751
:
131
–138.
Houser CR, Crawford GD, Salvaterra PM, Vaughn JE (
1985
) Immuno-cytochemical localization of choline acetyltransferase in rat cerebral cortex: a study of cholinergic neurons and synapses
J Comp Neurol
 
234
:
17
–34.
Kalsbeek A, Voorn P, Buys RM, Pool CW, Uylings HBM (
1988
) The development of the dopaminergic innervation in the prefrontal cortex of the rat.
J Comp Neurol
 
269
:
58
–72.
Kalsbeek A, DeBruin JPC, Feenstra MGP, Uylings HBM (
1990
) Age-dependent effects of lesioning the mesocortical dopamine system upon prefrontal cortex morphometry and PFC-related behaviors.
Prog Brain Res
 
85
:
257
–283.
Kesner RP, Hunt ME, Williams JM, Long JM (
1996
) Prefrontal cortex and working memory for spatial response, spatial location, and visual object information.
Cereb Cortex
 
6
:
311
–318.
Kitt CA, Levey AI, Friedmont DP, Walker LC, Koliatsos VE, Raskin LS, Price DL (
1988
) Immunocytochemical visualization of cholinergic fibers in monkey neocortex: enhanced visualization using silver nitrate.
Soc Neurosci Abstr
 
14
:
631
.
Kolb B (1990) Prefrontal cortex. In: The cerebral cortex of the rat (Kolb B, Tees RC, eds), pp. 437–458. Cambridge, MA: MIT Press.
Kosofsky BE, Molliver ME (
1987
) The serotonergic innervation of cerebral cortex: different classes of axon terminals arise from dorsal and median raphae nuclei.
Synapse
 
1
:
153
–168.
Krettek JE, Price JL (
1977
) The cortical projections of the mediodorsal nucleus and adjacent thalamic nuclei in the rat.
J Comp Neurol
 
172
:
157
–192.
Kritzer MF (
1997
) Selective colocalization of immunoreactivity for intracellular gonadal hormone receptors and tyrosine hydroxylase in the ventral tegmental area, substantia nigra and retrorubral fields in the rat.
J Comp Neurol
 
379
:
247
–260.
Kritzer MF (
2000
) Effects of acute and chronic gonadectomy on the catecholamine innervation of the cerebral cortex in adult male rats: insensitivity of axons immunoreactive for dopamine-β-hydroxylase to gonadal steroids, and differential sensitivity of axons immunoreactive for tyrosine hydroxylase to ovarian and testicular hormones.
J Comp Neurol
 
427
:
617
–633.
Kritzer MF, Kohama SG (
1998
) Ovarian hormones influence the morphology, distribution, and density of tyrosine hydroxylase immunoreactive axons in the dorsolateral prefrontal cortex of adult rhesus monkeys.
J Comp Neural
 
395
:
1
–17.
Kritzer MF, Adler A, Marotta J, Smirlis T (
1999
) Regionally selective effects of gonadectomy on cortical catecholamine innervation in adult male rats are most disruptive to afferents in prefrontal cortex.
Cereb Cortex
 
9
:
507
–518.
Kritzer MF, McLaughlin PJ, Smirlis T, Robinson JK (
2001
) Gonadectomy impairs T-maze acquisition in adult male rats.
Hormones Behav
 
39
:
167
–174.
Leonard CM (
1969
) The prefrontal cortex of the rat. I. Cortical projections of the mediodorsal nucleus. II. Efferent connections.
Brain Res
 
12
:
321
–343.
Levitt P, Moore RY (
1978
) Noradrenaline neuron innervation of the neocortex in the rat.
Brain Res
 
139
:
219
–231.
Lewis MS, Molliver ME, Morrison JH, Lidow HGW (
1979
) Complementarity of dopaminergic and noradrenergic innervation in anterior cingulate cortex of the rat.
Brain Res
 
164
:
328
–333.
Lidov HGW, Grzanna R, Molliver ME (
1980
) The serotonin innervation of the cerebral cortex in the rat — an immunohistochemical analysis.
Neuroscience
 
5
:
207
–227.
Lookingland KI, Moore KE (
1984
) Effects of estradiol and prolactin on incertohypothalamic dopaminergic neurons in the male rat.
Brain Res
 
323
:
83
–91.
Luine VN (
1985
) Estradiol increases choline acetyltransferase activity in specific basal forebrain nuclei and projection areas of female rats.
Exp Neurol
 
89
:
484
–490.
Mamounas LA, Mullen CA, O’Hearn E, Molliver ME (
1991
) Dual serotonergic projections to forebrain in the rat: morphologically distinct 5-HT axon terminals exhibit differential vulnerability to neurotoxic amphetamine derivatives.
J Comp Neurol
 
314
:
558
–586.
Morrison JH, Grzanna R, Molliver ME, Coyle JT (
1978
) The distribution and orientation of noradrenergic fibers in the neocortex of the rat. An immunofluorescence study.
J Comp Neurol
 
181
:
17
–40.
Murphy BL, Arnsten AFT, Goldman-Rakic PS, Roth RH (
1996
) Increased dopamine turnover in the prefrontal cortex impairs spatial working memory performance in rats and monkeys. Proc Natl Acad.
Sci USA
 
93
:
1325
–1329.
O’Malley CA, Hautamaki RD, Kelley M, Meyer EM (
1987
) Effects of ovariectomy and estradiol benzoate on high affinity choline uptake, ACh synthesis, and release from rat cerebral cortical synaptosomes.
Brain Res
 
403
:
389
–392.
Sarter M, Markowitsch HJ (
1983
) Convergence of basolateral amygdaloid and mediodorsal thalamic projections in different areas of the frontal cortex in the rat.
Brain Res Bull
 
10
:
607
–622.
Seeman MV, Lang M (
1990
) The role of estrogens in schizophrenia gender differences.
Schizophrenia Bull
 
16
:
185
–191.
Sesack SR, Deutch AY, Roth RH, Bunney BS (
1989
) Topographical organization of the efferent projections of the medial prefrontal cortex in the rat: an anterograde tract-tracing study with phaseolus vulgaris leucoagglutinin.
J Comp Neurol
 
290
:
213
–242.
Siddiqui A, Gilmore DP (
1988
) Regional differences in the catecholamine content of the rat brain: effects of neonatal castration and androgenization.
Acta Endocrinol
 
118
:
483
–494.
Siddiqui A, Shah BH (
1997
) Neonatal androgen manipulation differentially affects the development of monoamine systems in rat cerebral cortex, amygdala, and hypothalamus.
Dev Brain Res
 
98
:
247
–252.
Singh M, Meyer EM, Millard WJ, Simpkins JW (
1994
) Ovarian steroid deprivation results in a reversible learning impairment and compromised cholinergic function in female Sprague–Dawley rats.
Brain Res
 
644
:
305
–312.
Slopsema JS, van der Gugten J, De Bruin JPC (
1982
) Regional concentration of noradrenaline and dopamine in the frontal cortex of the rat: dopaminergic innervation of prefrontal subareas and lateralization of prefrontal dopamine.
Brain Res
 
250
:
197
–200.
Stam C, deBruin J, vanHaelst A, van der Gugten J, Kalsbeek A (
1989
) Influence of the mesocortical dopaminergic system on activity, food hoarding, social-agonistic behavior, and spatial delayed alternation in male rats.
Behav Neurosci
 
103
:
24
–35.
Stewart J, Rajabi H (
1994
) Estradiol derived from testosterone in prenatal life affects the development of catecholamine systems in the frontal cortex in the male rat.
Brain Res
 
646
:
157
–160.
Stewart J, Kuhnemann S, Rajabi H (
1991
) Neonatal exposure to gonadal hormones affects the development of monoamine systems in rat cortex.
J Neuroendocrinol
 
3
:
85
–93.
Sumner BE, Fink G (
1998
) Testosterone as well as estrogen increases serotonin 2A receptor mRNA and binding site densities in the male rat brain.
Brain Res Mol Brain Res
 
59
:
205
–214.
Van Eden CG (
1986
) Development of connections between the mediodorsal nucleus of the thalamus and the prefrontal cortex in the rat.
J Comp Neurol
 
244
:
349
–359.
Van Eden CG, Uylings HBM (
1985
) Cytoarchitectonic development of the prefrontal cortex in the rat.
J Comp Neurol
 
241
:
253
–267.
Van Eden CG, Hoorneman EMD, Bujis RM, Matthijssen MAH, Geffard M, Uylings HBM (
1987
) Immunocytochemical localization of dopamine in the prefrontal cortex of the rat at the light and electron microscopical level.
Neuroscience
 
22
:
849
–862.
Van Hest A, Van Kempen M, Van Haaren F, Van de Poll NE (
1988
) Memoryin male and female Wistar rats: effects of gonadectomy and stimulus presentations during the delay interval.
Behav Brain Res
 
29
:
103
–110.
Venkatesan C (
2000
) The influence of postnatal gonadal hormone exposure on circuit organization in the cerebral cortex. Doctoral dissertation, State University of New York, Stony Brook.
Verma A, Moghaddam B (
1996
) NMDA receptor antagonists impair prefrontal cortex function as assessed via spatial delayed alternation performance in rats: modulation by dopamine.
J Neurosci
 
16
:
373
–379.
Wainman P, Shipounoff GC (
1941
) The effects of castration and testosterone propionate on the striated perineal musculature in the rat.
Endocrinology
 
29
:
975
–978.
Wilcott R, Xuemei Q (
1990
) Delayed response, preoperative overtraining, and prefrontal lesions in the rat.
Behav Neurosci
 
104
:
74
–83.
Zahrt J, Taylor JR, Mathew RG, Arnsten AFT (
1997
) Supranormal stimulation of D1 dopamine receptors in the rodent prefrontal cortex impairs spatial working memory performance.
J Neurosci
 
17
:
8528
–8535.