Abstract

The ventrolateral orbital cortex (VLO) is part of an endogenous analgesic system, consisting of the spinal cord–thalamic nucleus submedius–VLO periaqueductal gray (PAG)–spinal cord loop. The present study examined morphological connections of GABAergic (γ-aminobutyric acidergic) neurons and serotonergic projection terminals from the dorsal raphe nucleus (DR), as well as the relationship between GABAergic terminals and VLO neurons projecting to the PAG, by using anterograde and retrograde tracing combined with immunofluorescence, immunohistochemistry, and electron microscopy methods. Results indicate that the majority (93%) of GABAergic neurons in the VLO also express the 5-HT1A (5-hydroxytryptamine 1A) receptor, and serotonergic terminals originating from the DR nucleus made symmetrical synapses with GABAergic neuronal cell bodies and dendrites within the VLO. GABAergic terminals also made symmetrical synapses with neurons expressing GABAA receptors and projecting to the PAG. These results suggest that a local neuronal circuit, consisting of 5-HTergic terminals, GABAergic interneurons, and projection neurons, exists in the VLO, and provides morphological evidence for the hypothesis that GABAergic modulation is involved in 5-HT1A receptor activation-evoked antinociception.

Introduction

Anatomical studies in rats and cats have demonstrated that the prefrontal ventrolateral orbital cortex (VLO) receives ascending projections from neurons in lamina I of the trigeminal subnucleus caudalis and spinal dorsal horn, via the thalamic nucleus submedius (Sm) (Craig and Burton 1981; Craig et al. 1982; Yoshida et al. 1991, 1992; Coffield et al. 1992). The VLO contains neurons that project to the midbrain periaqueductal gray (PAG) (Hardy and Leichnetz 1981; Craig et al. 1982), a region intensively involved in descending modulation of nociception (Fields and Basbaum 1999). Behavioral studies have indicated that electrically or chemically induced neuronal activation in the VLO depresses tail-flick and jaw-opening reflexes, and these effects are eliminated by lesioning or microinjecting inhibitory neurotransmitter γ-aminobutyric acid (GABA) into the PAG (Zhang et al. 1997a, 1997b, 1998), suggesting that VLO-induced antinociception is produced by activation of the VLO–PAG brainstem descending inhibitory system.

Furthermore, microinjection of serotonin (5-hydroxytryptamine, 5-HT) or 5-HT1A receptor (5-HT1AR) agonist into the VLO produces an obvious antinociceptive effect in the rat tail-flick test; this effect can be blocked by 5-HT1AR antagonist (Qu et al. 2008; Huo et al. 2008). It has been suggested that 5-HT1AR activation-induced antinociception in the VLO may be induced by blocking the inhibitory action of GABAergic interneurons, via the GABAA receptor (GABAAR), on output neurons projecting to the PAG. This disinhibitory effect leads to activation of the brainstem descending inhibitory system and depression of nociceptive inputs at the spinal/trigeminal level (Huo et al. 2008). There is evidence that GABA- (Huo et al. 2005), GABAAR- (Pirker et al. 2000; Princivalle et al. 2001), and 5-HT1AR–positive neurons (Wright et al. 1995; Hossein et al. 1996; Barnes and Sharp 1999; Santana et al. 2004), as well as 5-HTergic terminals originating from the dorsal raphe (DR) nucleus, are distributed in the VLO (Li et al. 1993; Matsuzaki et al. 1993). However, the connection between GABAergic neurons, GABAAR, 5-HTergic terminals, 5-HT1AR, and projecting neurons in the VLO remains unclear. The aim of the present study was to examine the connections between 5-HTergic terminals and GABAergic neurons that may express 5-HT1AR, as well as the connections between GABAergic terminals and output neurons projecting to the PAG that may express GABAAR in the VLO, through the use of anterograde and retrograde tracing, combined with immunohistochemistry, immunofluorescent, and electron microscopy methods. Some of the data reported here have been previously presented in abstract form (Huo et al. 2007).

Materials and Methods

Animal Preparation

The experiments were performed on 36 male Sprague–Dawley rats, weighing 230–260 g, provided by the Experimental Animal Center of the Fourth Military Medical University (Xi'an, China). The experiments were performed in accordance with ethical guidelines of the International Association for the Study of Pain (Zimmermann 1983) and were approved by the Institutional Animal Care Committee of Xi'an Jiaotong University and the Fourth Military Medical University. All efforts were made to minimize the number of animals used and their suffering.

5-HT1AR Immunohistochemistry

The rats (n = 8) were deeply anesthetized with sodium pentobarbital (100 mg/kg body weight, i.p.) and perfused transcardially with 100 mL of 0.01 M phosphate-buffered saline (PBS, pH 7.4), followed by 500 mL of 0.1 M phosphate buffer (PB, pH 7.4) containing 4% (w/v) paraformaldehyde, and 75% (v/v) saturated picric acid. Subsequent to perfusion, the brains were rapidly removed and placed into the same, fresh fixative to be fixed for additional 4 h at 4 °C. Subsequently, the brains were soaked in 0.1 M PB (pH 7.4) containing 25% (w/v) sucrose solution overnight at 4 °C, and cut serially into 20-μm thick coronal sections on a freezing microtome (Kryostat, 1720; Leitz, Mannheim, Germany). The forebrain sections, including the VLO, were collected serially into 5 separate dishes containing 0.01 M PBS (pH 7.4) according to the numerical order while cutting (e.g., sections 1, 6, and 11 were placed in dish 1, sections 2, 7, and 12 in dish 2, and sections 3, 8, and 13 in dish 3, etc.). Each dish typically contained 16–18 sections. All sections were washed carefully with 0.01 M PBS.

The sections in the first dish were used for immunohistochemistry staining of 5-HT1AR using the avidin–biotin–peroxidase (ABC) method (Hsu et al. 1981). Briefly, the sections were washed in 0.01 M PBS (pH 7.4), and incubated sequentially with 1) guinea pig antiserum against 5-HT1AR polyclonal antibody (AB5406, 1:1000 dilution; Chemicon, Temecula, CA) in 0.01 M PBS containing 5% (v/v) normal goat serum (NGS), 0.3% (v/v) Triton X-100, 0.05% (w/v) NaN3, and 0.25% (w/v) carrageenan (PBS-NGS, pH 7.4) for 48–72 h at 4 °C; 2) biotinylated goat anti-guinea pig IgG (1:200 dilution; Vector, Burlingame, CA) in PBS-NGS overnight at 4 °C; and 3) ABC Elite complex (Vector: 1:100) in 0.01 M PBS (pH 7.4) containing 0.3% (v/v) Triton X-100 (PBS-X) for 2 h at room temperature. Bound peroxidase was visualized by incubation with 0.05% 3,3-diaminobenzidine tetrahydrochloride (DAB; Dojin, Kumamoto, Japan) and 0.003% H2O2 in 0.05 M Tris–HCl buffer (pH 7.6) for 20–30 min. The sections were rinsed at least 3 times in 0.01 M PBS (pH 7.4) for at least 10 min after each incubation. The sections were then mounted onto gelatin-coated glass slides, air dried, dehydrated and cleaned, coverslipped with DPX, and observed under light microscope (BX-60; Olympus, Tokyo, Japan). The microphotographs were taken with a digital camera (DP-70; Olympus) attached to a microscope.

The sections in the second dish were mounted onto gelatin-coated glass slides and processed for Nissl staining. The sections in the third dish were used for control tests. In the control experiments, the primary antibodies were omitted or replaced with normal guinea pig serum; no positive staining for the omitted or replaced antibodies was detected.

GABA and/or 5-HT1AR Immunofluorescent Staining

The sections in the fourth dish were used for immunofluorescent double staining for GABA and 5-HT1AR. In brief, the sections were incubated in PBS-NGS (pH 7.4) at 4 °C sequentially with 1) a mixture of rabbit antiserum against GABA polyclonal antibody (A2052, 1:2000 dilution; Sigma, St Louis, MO) and guinea pig antiserum against 5-HT1A R polyclonal antibody (AB5406, 1:1000 dilution; Chemicon) for 48–72 h; 2) biotinylated goat anti-guinea pig IgG (1:200 dilution; Vector) and donkey anti-rabbit IgG conjugated with fluorescein isothiocyanate (FITC; 1:200 dilution; Jackson Immuno Research, West Groove, PA), overnight; and 3) Cy3-conjugated avidin D (1:1000 dilution; Vector) in PBS-X (pH 7.4) for 2 h. The sections were rinsed at least 3 times in 0.01 M PBS (pH 7.4) for at least 10 min following each incubation. The sections were then mounted onto clean glass slides, air-dried, and coverslipped with a mixture of 50% (v/v) glycerin and 2.5% (w/v) triethylene diamine (antifading agent) in 0.01 M PBS. Finally, the sections were observed with a confocal laser-scanning microscope (LSM, FV1000; Olympus; 488 nm for FITC and 543 nm for Cy3) and appropriate emission filters (500–535 nm for FITC and 570–654 nm for Cy3). Digital images were captured using an attached digital camera and were modulated for better light/contrast using Adobe Photoshop software (Version 7.0, Adobe Systems Inc., Singapore).

The fifth series of the sections was used for control tests. In the control experiments, primary antibodies were omitted or replaced with normal rabbit and guinea pig sera; no immunofluorescent staining for the omitted or replaced antibodies was detected.

For each rat (n = 8), the number of GABA-positive, 5-HT1AR–positive neurons, and GABA/5-HT1AR double-labeled neurons were counted (20× or 40× magnification) in every ninth coronal section from the fourth dish. The same VLO region was counted, according to the Nissl-stained sections and rat brain atlas (Paxinos and Waston 1986). The ratios of GABA/5-HT1AR double-labeled neurons to the total number of GABA-positive and 5-HT1AR–positive neurons were calculated. Careful focusing through the thickness of all sections determined that immunofluorescent labeling had penetrated the entire thickness of the sections, and only neuronal cell bodies with obvious light emission were counted. Because emission from some immunopositive neurons might have been too weak to detect, the number of GABA-positive neurons and/or 5-HT1AR–positive neurons were regarded to represent the minimum number of immunopositive neurons within the sections. In addition, to avoid possible double counting of immunopositive neurons, the sections were carefully moved across the stage and analyzed up and down. Although some GABA/5-HT1AR double-labeled neurons in both sides of the observed areas were captured with the camera, only the GABA/5-HT1AR double-labeled neurons on the top side of the VLO were counted.

Anterograde Tracing Combined with Immunofluorescent Triple Staining of BDA, GABA, and 5-HT

The rats (n = 8) were anesthetized with sodium pentobarbital (50 mg/kg body weight, i.p.), and received a 0.2-μL microinjection of 10% biotinylated dextranamine (BDA, 10 000 MW; Molecular Probes, Eugene, OR) in 0.01 M PB (pH 7.2) into the DR. The BDA was stereotaxically pressure-injected into the DR (8.0 mm posterior to bregma, midline, 6.5 mm from cerebral surface) (Paxinos and Watson 1986) over 20 min through a glass micropipette (internal tip diameter of 10–20 μm), which was attached to a 1 μL of Hamilton microsyringe. Seven days later, the rats were deeply anesthetized and transcardially perfused. The brains were removed and cryoprotected according to the above-mentioned protocol. The brains were subsequently serially cut into 20-μm-thick coronal sections using a freezing microtome (Kryostat 1720; Leitz). The midbrain and pons regions, including the DR, were serially collected into the first and the second dishes, and the forebrain sections, including the VLO, were serially collected into the remaining 3 dishes (dish 3–5) containing 0.01 M PBS. All sections were carefully washed with 0.01 M PBS.

The BDA injection sites on the midbrain and pons sections were revealed by incubating the sections for 4 h at room temperature in 0.01 M PBS (pH 7.4) solution containing ABC Elite complex (1:100 dilution; Vector) and 0.3% (v/v) Triton X-100. Finally, the peroxidase-bound ABC complex was visualized by incubating the sections with 0.05% DAB (Dojin), 0.003% H2O2, and 0.04% Ni(NH4)2SO4 in 0.05 M Tris–HCl buffer (pH 7.6) for 20–30 min. The sections were mounted onto gelatin-coated glass slides, air-dried, and coverslipped with DPX. Microphotographs of injection sites were taken with a digital camera (DP-70, Olympus) attached to the BX-60 microscope (Olympus). The sections in the second dish were mounted onto gelatin-coated glass slides and processed for Nissl staining.

The forebrain sections, including the VLO in the third dish, were used for immunofluorescent triple staining of BDA, GABA, and 5-HT. In brief, the sections were incubated at 4 °C sequentially with 1) a mixture of rabbit antiserum against GABA polyclonal antibody (A2052, 1:2000 dilution; Sigma) and rat antiserum against 5-HT polyclonal antibody (MAB352, 1:200 dilution; Chemicon) in PBS-NGS (pH 7.4) for 48–72 h; 2) Cy3-conjugated avidin D (1:1000 dilution; Vector), donkey anti-rabbit IgG-FITC (1:200 dilution; Jackson), and goat anti-rat IgG-Cy5 (1:200 dilution; Jackson) in PBS-X (pH 7.4) for 4–6 h. The sections were rinsed at least 3 times in 0.01 M PBS (pH 7.4) for a minimum of 10 min following each incubation. The sections were then mounted onto clean glass slides, air-dried, and coverslipped with a mixture of 50% (v/v) glycerin and 2.5% (w/v) triethylene diamine in 0.01 M PBS. Finally, the sections were observed with a confocal LSM (FV1000; Olympus; 488 nm for FITC, 543 nm for Cy3, and 650 nm for Cy5) and appropriate emission filters (500–535 nm for FITC, 570–654 nm for Cy3, and 660–685 nm for Cy5). The digital images were captured using an attached digital camera and were modulated for better light/contrast using Adobe Photoshop software (Version 7.0).

The forebrain sections, including the VLO in the fourth dish, were used as controls. The primary antibodies were omitted or replaced with a mixture of normal rabbit serum and normal rat sera. The remaining incubation steps were the same as those used for VLO sections in the third dish. No immunofluorescent staining was detected. The forebrain sections, including the VLO in the fifth dish, were mounted onto clean glass slides, air-dried, and used for Nissl staining.

Retrograde Tracing Combined with Immunofluorescent Triple Staining of the Tracer, GABAAR, and GABA

The rats (n = 8) were anesthetized with sodium pentobarbital (50 mg/kg body weight, i.p.) and placed onto a stereotaxic frame. They were then stereotaxically injected with 10% tetramethyl rhodamine (TMR, 3000 MW; Molecular Probes) dissolved in 0.1 M citrate-NaOH (pH 3.0). A volume of 0.2 μl of 10% TMR solution was made and slowly injected by pressure over 20 min unilaterally into the ventrolateral PAG subregion (vlPAG) according to the rat brain atlas (−8.0 mm from bregma, 0.9 mm lateral to the midline, and 5.9 mm deep from the surface of the brain) (Paxinos and Watson 1986). The injection took place through a glass micropipette (internal tip diameter: 15–25 μm), which was filled with TMR and attached to a 1-μL Hamilton microsyringe. After stereotaxic injection of the tracer, the glass micropipette was left in place for 20 min.

At 10 days after injection, all rats were deeply anesthetized and transcardially perfused. The brains were removed and cryoprotected according to the above-mentioned procedures. Subsequently, all brains were serially cut into 20-μm thick frontal sections using a freezing microtome (Kryostat 1720; Leitz). The midbrain sections, including the TMR injection site (vlPAG), were collected into the first dish, and the forebrain sections, including the VLO, were serially collected into the remaining 4 dishes (dish 2–5). The dishes will filled with 0.01 M PBS (pH 7.4) and the sections were dividing according to the numerical order while cutting (e.g., sections 1, 4, and 7 were place in dish 2, sections 2, 5, and 8 in dish 3, and sections 3, 6, and 9 in dish 4, etc.). Each dish typically contained 20–25 sections. All sections were washed carefully with 0.01 M PBS. The serial sections from the first and second dishes were mounted onto clean, glass slides to examine the TMR injection sites in the vlPAG, as well as distributions of the TMR retrograde-labeled neurons in the VLO.

The forebrain sections, including VLO from the third dish, were subjected to immunofluorescent staining for GABAAR and GABA. In brief, the sections were incubated at 4 °C sequentially in 3 steps: 1) a mixture of goat antiserum against GABAAR antibody (sc-7348, 1:200 dilution; Santa Cruz Biotech, CA) and mouse antiserum against GAD67 antibody (MAB5406, 1:200 dilution; Chemicon) in PBS-NGS (pH 7.4) for 72 h; 2) biotinylated donkey anti-goat IgG (1:200 dilution; Vector) and donkey anti-mouse IgG-Cy5 (1:200 dilution; Vector) in PBS-NGS (pH 7.4) overnight; and 3) FITC-avidin D (1:500 dilution; Vector) in PBS-X (pH 7.4) for 2 h. Between each step, the sections were carefully washed with 0.01 M PBS. Subsequent to immunofluorescent staining, the sections were rinsed in 0.01 M PBS, mounted onto clean, glass slides, air-dried, coverslipped with a mixture of 50% (v/v) glycerin and 2.5% (w/v) triethylene diamine in 0.01 M PBS, and observed with a confocal LSM (FV1000; Olympus; 488 nm for FITC, 570 nm for TMR, and 650 nm for Cy5) using appropriate emission filters (500–535 nm for FITC, 600–654 nm for TMR, and 660–685 nm for Cy5). The digital images were captured using an attached digital camera and were modulated for better light/contrast using Adobe Photoshop software (Version 7.0).

The number of TMR-labeled neurons, GABAAR-positive neurons, and TMR/GABAAR double-labeled neurons in the VLO were counted in the twelfth coronal section from the third dish at the same VLO region according to Nissl-stained sections and the rat brain atlas (Paxinos and Waston 1986). To assure that collected data were reliable, only TMR-labeled, GABAAR-positive, and TMR/GABAAR double-labeled neurons lacking nuclear staining were counted.

The fourth set of the serial sections was mounted onto gelatin-coated glass slides and processed for Nissl staining. The fifth set of serial sections was used for controls. In the control experiments, the primary antibodies were omitted or replaced with a mixture of normal goat and mouse sera, and the remaining steps were performed identically to the sections from the third dish. No immunofluorescent staining was detected.

Connections between GABAergic Neurons/Terminals and 5-HTergic Terminals or 5-HT1AR in the VLO—Revealed by Dual-Labeling Electron Microscopy

Rats (n = 6) were deeply anesthetized with sodium pentobarbital and transcardially perfused with 100 mL of 0.01 M PBS (pH 7.4), followed by 500 mL of 4% (w/v) paraformaldehyde, 0.05% glutaraldehyde, and 15% (v/v) saturated picric acid in 0.1 M PB (pH 7.4). Subsequent to perfusion, the brains were quickly removed and placed into the same, fresh fixative without glutaraldehyde to be postfixed for additional 2 h at 4 °C. The forebrain regions containing VLO were then serially cut into 50-μm-thick frontal sections using a microslicer (DTK-100; Dosaka, Kyoto, Japan). The sections were consecutively collected into 3 dishes and placed in 0.05 M PB that contained 25% (w/v) sucrose and 10% (v/v) glycerol for 30 min for cryoprotection. Subsequently, the sections were freeze-thawed in liquid nitrogen to enhance antibody penetration during the immunohisochemistry reaction. The sections were then placed in 0.05 M Tris-buffered saline (TBS, pH 7.4) containing 20% NGS for 1 h to block nonspecificity.

The sections in the first dish were used for GABA/5-HT dual-label staining. Briefly, the sections were incubated for 24 h at room temperature with a mixture of rabbit antiserum against GABA polyclonal antibody (A2052, 1:1000 dilution; Sigma) and rat antiserum against 5-HT polyclonal antibody (MAB352, 1:100 dilution; Chemicon), which was diluted in 0.05 M TBS (pH 7.4) containing 2% (v/v) NGS (TBS-NGS). The sections were washed in TBS and further incubated overnight at room temperature with a mixture of 1.4-nm gold particles conjugated to anti-rabbit IgG (1:100 dilution, Nanoprobes, Stony Brook, NY) and biotinylated donkey anti-rat IgG (1:100 dilution; Jackson) diluted in TBS-NGS.

The sections in the second dish were used for GABA/5-HT1AR dual-label staining. The sections were incubated for 24 h at room temperature with a mixture of rabbit antiserum against GABA polyclonal antibody (A2052, 1:1000 dilution; Sigma) and guinea pig antiserum against 5-HT1AR polyclonal antibody (AB5406, 1:1000 dilution; Chemicon) in TBS-NGS, and then incubated with 1.4-nm gold particles conjugated to anti-guinea pig IgG (1:100 dilution; Nanoprobes) and biotinylated goat anti-rabbit IgG (1:100 dilution; Vector) overnight at room temperature.

After incubating, the sections in the first and second dishes were washed in 0.05 M TBS 3 times and subsequently processed for: 1) postfixation with 1% glutaraldehyde in 0.1 M PB for 10 min; 2) after washing in distilled water, silver enhancement with HQ Silver Kit (Nanoprobes) in the dark; 3) incubation with ABC Elite Kit (Vector) diluted at 1:100 in 0.05 M TBS for 2 h at room temperature; 4) visualization of 5-HT–like immunoreactivity by incubating with 0.02% (w/v) DAB (Dojin) and 0.0003% H2O2 for 20–30 min; 5) osmification with 1% osmium tetroxide (OsO4) in 0.1 M PB at room temperature for 45 min; 6) counterstaining with 1% uranyl acetate in 70% ethanol in the dark for 1 h; and 7) flat-embedding in Durcupan (Fluka Chemie, Buchs, Switzerland) after dehydration in a graded series of ethanol and mounting on silicon-coated slide glass. Once the resin polymerized, the sections were examined with a light microscope, and VLO regions that contained many silver-stained neuronal cell bodies and 5-HT–positive axons (in the first set of sections) or GABA-positive neuronal cell bodies (in the second set of sections) were selected and cut out from the flat-embedded sections with razor blades under a dissection microscope. The tissue samples of the selected regions were cut into ultrathin sections with a diamond knife mounted on a ultramicrotome (Reichert-Nissei Ultracut S; Lecia, Vienna, Austria). The ultrathin sections were mounted on single-slot grids coated with pioloform membrane (Agar Scientific, Stansted, UK), stained with 1% (w/v) lead citrate, and then examined with an electron microscope (H-7100; Hitachi, Tokyo, Japan) as described elsewhere (Li et al. 2000).

Triple-Label Electron Microscopy Shows Connections between GABAergic Terminals and Retrogradely Labeled Neurons Expressing GABAAR in the VLO

The following experiment was performed on the remaining 6 male Sprague–Dawley rats. A total of 0.2 μL of 2% (w/v) wheat germ agglutinin-conjugated horseradish peroxidase (WGA-HRP) (Toyobo, Tokyo, Japan) dissolved in 0.1 M Tris–HCl (pH 7.0) was injected in the vlPAG, similar to the TMR injections described above. After the WGA-HRP injection, the rats were housed for an additional 48–60 h. The rats were then deeply anesthetized and transcardially perfused with 100 ml of 0.01 M PBS (pH 7.4), followed by 500 ml of 0.1 M PB (pH 7.4) containing 4% (w/v) paraformaldehyde, 0.1% (w/v) glutaraldehyde, and 15% (v/v)–saturated picric acid. After the brains were removed, the mesencephalic regions containing the injection sites were serially cut into 30-μm-thick frontal sections using a freezing microtome, whereas the prefrontal cortex regions containing the VLO were cut into 50-μm-thick cross-sections using a microslicer (DTK-100; Dosaka). The sections were collected in 0.01 M PBS (pH 7.4) and processed for WGA-HRP histochemistry staining, using tetramethylbenzidine (TMB). Sodium tungstate was used as a stabilizer, and the WGA-HRP reaction products were intensified with DAB/cobalt/H2O2 solution. The sections containing the injection sites were then mounted onto gelatinized glass slides. The sections were placed on glass slides, and sections containing WGA-HRP–labeled neurons in the VLO were selected under a light microscope, collected in vials containing a mixture of 25% (w/v) sucrose and 10% (v/v) glycerol in 0.05 M PB, and treated by the freeze–thaw method to enhance antibody penetration for the following immunohistochemistry reaction. Subsequently, the sections were washed 3 times in 0.05 M TBS (pH 7.4) and blocked with 20% (v/v) NGS in 0.05 M TBS for 30 min. For further double labeling of GABA and GABAAR, the sections were stained using the immunoperoxidase method (GABA) and the immunogold silver labeling method (GABAAR). The sections were incubated with a mixture of rabbit antiserum against GABA polyclonal antibody (A2052, 1:2000 dilution; Sigma) and goat antiserum against GABAA receptor antibody (sc-7348, 1:100 dilution; Santa Cruz) in 0.05 M TBS-NGS for 48 h at 4 °C. Then, the sections were washed in 0.05 M TBS 3 times and incubated with a mixture of biotinylated donkey anti-rabbit IgG (1:200 dilution; Vector) and 1.4-nm gold particles conjugated with anti-goat IgG (1:100 dilution; Nanoprobes) in TBS-NGS overnight.

After incubation, the sections were subsequently processed at room temperature similarly to the above-mentioned detailed procedures for dual-label electron microscopy. In brief, 1% glutaraldehyde postfixation; washing in distilled water; silver enhancement with HQ Silver Kit (Nanoprobes) in the dark; and incubation with the ABC Kit (Vector). And finally, the sections were placed in 0.05 M Tris–HCl (pH 7.6) containing 0.02% DAB and 0.0003% H2O2 for 10–20 min. The immunolabeled sections were postfixed in 1% OsO4, counterstained with 1% uranyl acetate in 70% ethanol, flat-embedded in Durcupan (Fluka Chemie) after dehydration and polymerized. Small pieces of the VLO areas containing many labeled cells were selected and cut out from the flat-embedded sections. The selected tissue pieces were cut into serial ultrathin sections and mounted on single-slot grids coated with pioloform membrane (Agar Scientific), stained with 1% lead citrate, and then examined by electron microscopy (H-7100; Hitachi) (Li et al. 2000).

Results

Distribution of GABAergic, 5-HT1A Receptor-Expressing, and GABA/5-HT1AR Double-Labeled Neurons in the VLO

5-HT1AR–positive neuronal cell bodies and terminals were distributed extensively throughout VLO layers II–VI, but not layer I, with a greater number in layers II and III than in layers V and VI (Fig. 1A). Most 5-HT1AR–positive neuronal cell bodies exhibited staining in the perikarya and cell membrane, but were void of staining in the nuclei and processes. Some of the processes were covered with bead-shaped varicosities. 5-HT1AR–positive neuronal cell bodies were rounded, triangular, or multipolar in shape. The 5-HT1AR–positive neuronal cell bodies were about 10–20 μm in diameter (Fig. 1B).

Figure 1.

Light field images of distribution of 5-HT1A receptor (5-HT1AR)–positive neurons in the VLO. (A) Low-magnification image of 5-HT1AR–positive neurons in the VLO, (B) High-magnification view of the region outlined by rectangular box on (A). 5-HT1AR–positive neurons were distributed throughout layers II–VI, but not in layer I, of the VLO (A). 5-HT1AR–positive neuronal cell bodies were rounded, triangular, or multipolar in shape (B). The diameters of 5-HT1AR–positive neuronal cell bodies were 10–20 μm. Abbreviations: AI, agranular insular cortex; fmi, forceps minor corpus callosum; LO, lateral orbital cortex. Scale bar: A = 100 μm, B = 25 μm.

Figure 1.

Light field images of distribution of 5-HT1A receptor (5-HT1AR)–positive neurons in the VLO. (A) Low-magnification image of 5-HT1AR–positive neurons in the VLO, (B) High-magnification view of the region outlined by rectangular box on (A). 5-HT1AR–positive neurons were distributed throughout layers II–VI, but not in layer I, of the VLO (A). 5-HT1AR–positive neuronal cell bodies were rounded, triangular, or multipolar in shape (B). The diameters of 5-HT1AR–positive neuronal cell bodies were 10–20 μm. Abbreviations: AI, agranular insular cortex; fmi, forceps minor corpus callosum; LO, lateral orbital cortex. Scale bar: A = 100 μm, B = 25 μm.

In addition, 5-HT1AR–positive neuronal cell bodies and terminals were also sparsely distributed in the lateral orbital cortex and agranular insular cortex, where the density and distribution pattern were similar to the staining pattern observed in the VLO. With exception to the forceps minor corpus callosum and olfactory ventricle, 5-HT1AR–positive neuronal cell bodies and terminals were observed throughout the neocortex as well.

GABA-positive neuronal cell bodies and processes exhibited similar distribution patterns, morphological features, and diameter range as the 5-HT1AR–positive neurons. These results were in good accordance with our previous studies (Huo et al. 2005).

The majority of 5-HT1AR–positive neurons also exhibited GABA-positive in the VLO area. GABA/5-HT1AR double-labeled neuronal cell bodies and processes were observed in layers II to VI, with the majority in layers II and III. In layers V and VI, a considerable number of neurons with GABA/5-HT1AR dual-labeling were observed, but there were no double-labeled neurons in layer I. In the VLO, a total of 716 (n = 8) neuronal cells were GABA-positive; 865 (n = 8) were 5-HT1AR–positive, and 663 (n = 8) were GABA/5-HT1AR–positive. Of the GABA-positive neurons, 93.0% (663/716) were immunopositive for 5-HT1AR. Of the 5-HT1AR–positive neurons, 76.6% (663/865) were immunoreactive for GABA (Fig. 2).

Figure 2.

Confocal laser-scanning digital images of GABAergic- (A) and 5-HT1A receptor (5-HT1AR)–positive neurons (B) in the VLO. GABAergic neurons were stained with FITC and 5-HT1AR–positive neurons were labeled with Cy3. The field in (A) is the same as that in (B), taken under different filters. Double arrowheads indicate neurons showing both GABA- and 5-HT1AR–positive labeling; arrowhead indicates a GABAergic neuron, whereas an arrow indicates a 5-HT1AR–positive neuron in (C), respectively. The majority of the GABAergic neurons also exhibited 5-HT1AR–positive labeling in the VLO. Scale bars = 20 μm.

Figure 2.

Confocal laser-scanning digital images of GABAergic- (A) and 5-HT1A receptor (5-HT1AR)–positive neurons (B) in the VLO. GABAergic neurons were stained with FITC and 5-HT1AR–positive neurons were labeled with Cy3. The field in (A) is the same as that in (B), taken under different filters. Double arrowheads indicate neurons showing both GABA- and 5-HT1AR–positive labeling; arrowhead indicates a GABAergic neuron, whereas an arrow indicates a 5-HT1AR–positive neuron in (C), respectively. The majority of the GABAergic neurons also exhibited 5-HT1AR–positive labeling in the VLO. Scale bars = 20 μm.

In addition, GABA/5-HT1AR double-labeled neurons were widely found in the lateral orbital and the agranular insular cortices, as well as throughout the neocortex.

On the ultrathin sections double-labeled for GABA and 5-HT1AR, the immunoperoxidase/DAB method was used for GABA labeling, and immunogold method was used for 5-HT1AR labeling. Both GABA- and 5-HT1AR–positive structures contained somata, dendrites, and axon terminals. DAB staining was observed in the cytoplasm, and immunogold particles were primarily located on or near the cytoplasma and intracellular membrane. A total of 76 neuronal cells exhibited GABA- and/or 5-HT1AR–positive. Of these, 56 (73.7%) were GABA- and 5-HT1AR–positive (Fig. 3), 4 (5.3%) were GABA–positive only, and 16 (21.1%) were 5-HT1AR–positive only. Taken together, 93.3% of the 5-HT1AR–positive neuronal cells were GABA positive, whereas 77.8% of the GABA-positive neuronal cells were 5-HT1AR–positive. GABA/5-HT1AR double-labeled neurons, which were visualized with both DAB reaction products and silver-gold particles, were also observed in the lateral orbital and the agranular insular cortices.

Figure 3.

Dual-label electron microscopy observation showing GABA-immunoreactive (electron-dense, peroxidase-immunoreactive product) dendritic profiles (Den) of a VLO neuron expressing 5-HT1A receptors (immunogold silver grains). Scale bar = 0.6 μm.

Figure 3.

Dual-label electron microscopy observation showing GABA-immunoreactive (electron-dense, peroxidase-immunoreactive product) dendritic profiles (Den) of a VLO neuron expressing 5-HT1A receptors (immunogold silver grains). Scale bar = 0.6 μm.

Synaptic Connections between Ascending 5-HTergic Fibers and GABAergic Neurons in the VLO

Five rats were successfully injected with anterograde tracer BDA into the DR (Fig. 4). The BDA anterogradely labeled fibers and terminals were bilaterally distributed in all VLO layers. Finely beaded fibers with varicosities appeared to penetrate the VLO from rostral toward caudal levels. Moreover, the predominant BDA-labeled finely beaded fibers and terminals were 5-HT positive, and close connections between BDA/5-HT double-labeled terminals and GABA-positive neurons were widely observed (Fig. 5). The close connections between BDA/5-HT double-labeled terminals and GABA-positive neurons were present in the lateral orbital and the agranular insular cortices, as well as adjacent neocortical areas.

Figure 4.

Light field photomicrograph of coronal sections showing BDA injection site in the DR of rat. Abbreviations: Aq, aqueduct; mlf, medial longitudinal fasciculus. Scale bar = 500 μm.

Figure 4.

Light field photomicrograph of coronal sections showing BDA injection site in the DR of rat. Abbreviations: Aq, aqueduct; mlf, medial longitudinal fasciculus. Scale bar = 500 μm.

Figure 5.

Confocal laser-scanning digital images of GABAergic-positive neurons (A), BDA (B), and serotonergic terminals (C) in the VLO. GABAergic neurons were stained with FITC, BDA was labeled with Cy3, and serotonergic terminals were labeled with Cy5. Arrows indicate the predominant, BDA-labeled, finely beaded fibers and terminals that were 5-HT positive. The BDA anterograde-labeled serotonergic terminals were closely connected with GABAergic neurons. Arrowheads indicate the close connection between serotonergic terminals and GABAergic neurons (D). Scale bars = 20 μm.

Figure 5.

Confocal laser-scanning digital images of GABAergic-positive neurons (A), BDA (B), and serotonergic terminals (C) in the VLO. GABAergic neurons were stained with FITC, BDA was labeled with Cy3, and serotonergic terminals were labeled with Cy5. Arrows indicate the predominant, BDA-labeled, finely beaded fibers and terminals that were 5-HT positive. The BDA anterograde-labeled serotonergic terminals were closely connected with GABAergic neurons. Arrowheads indicate the close connection between serotonergic terminals and GABAergic neurons (D). Scale bars = 20 μm.

Under electron microscopy, connections between 5-HT–positive axon terminals and GABA-positive neurons were confirmed. In the VLO, DAB reaction products labeled 5-HT–positive axon terminals, and silver-gold labeled GABA-positive somata and neurites were observed. DAB reaction products in the 5-HTergic axon terminals accumulated on the synaptic vesicles and were heavily stained. In addition to the GABA-positive neuronal cell bodies and dendrites, the silver-gold particles were scattered in the cytoplasm (Fig. 6). A total of 61 5-HTergic axon terminals were observed to form synaptic connections. These 5-HTergic axon terminals primarily made symmetric axon-dendritic synaptic connections (72.1%, 44/61) with GABA-positive dendrites, and also formed symmetric axon-somatic synapses (24.6%, 15/61) with GABA-positive somatic profiles (Fig. 6). Occasionally, 5-HTergic axon terminals were comprised of asymmetric axon-dendritic synapses (3.3%, 2/61) with GABA-positive dendrites. The synaptic connections of GABA/5-HT, which were visualized with silver-gold particles and DAB reaction products, were also observed in the lateral orbital and the agranular insular cortices.

Figure 6.

Dual-label electron microscopy observation showing serotonergic (electron-dense, peroxidase-immunoreactive product) axon terminals (Ax) in symmetric synaptic contact with GABA-positive (immunogold silver grains) dendritic profiles (Den) of a VLO neuron. Arrowheads indicate the synaptic site. Scale bar = 0.6 μm.

Figure 6.

Dual-label electron microscopy observation showing serotonergic (electron-dense, peroxidase-immunoreactive product) axon terminals (Ax) in symmetric synaptic contact with GABA-positive (immunogold silver grains) dendritic profiles (Den) of a VLO neuron. Arrowheads indicate the synaptic site. Scale bar = 0.6 μm.

Synaptic Connections between GABAergic Terminals and Descending Projection Neurons Expressing GABAAR

Four rats were successfully injected with retrograde tracer TMR into the vlPAG (Fig. 7). TMR retrograde-labeled neuronal cell bodies were distributed predominantly in layer V of the VLO (82%, 141/172, n = 4), ipsilateral to the injection sites. Few TMR-labeled neurons were observed in layer V of the contralateral VLO (18%, 31/172, n = 4).

Figure 7.

Light field photomicrograph of coronal sections showing TMR injection site in the vlPAG of the rat. Abbreviations: Aq, aqueduct. Scale bar = 500 μm.

Figure 7.

Light field photomicrograph of coronal sections showing TMR injection site in the vlPAG of the rat. Abbreviations: Aq, aqueduct. Scale bar = 500 μm.

GABAAR-positive neurons were extensively distributed throughout layers II–VI of the VLO, with a greater number of cells in layers II, III, and V, compared with layer VI. There were no positive neurons observed in layer I.

A considerable number (27.3%, 141/516, n = 4) of TMR retrograde-labeled neuronal cell bodies were GABAAR-positive in the ipsilateral VLO. Moreover, almost all TMR/GABAAR double-labeled neurons made close connections with the GABAergic terminals (Fig. 8).

Figure 8.

Confocal laser-scanning digital images of the TMR retrograde-labeled neurons (A), GABAAR-positive neurons (B) and GABAergic-positive terminals (C) in the VLO. GABAAR-positive neurons were stained with FITC and GABAergic terminals were labeled with Cy5. GABAergic terminals were closely connected with TMR retrograde-labeled neurons that were also GABAAR positive. Arrowheads indicate the closely connected sites. Scale bars = 20 μm.

Figure 8.

Confocal laser-scanning digital images of the TMR retrograde-labeled neurons (A), GABAAR-positive neurons (B) and GABAergic-positive terminals (C) in the VLO. GABAAR-positive neurons were stained with FITC and GABAergic terminals were labeled with Cy5. GABAergic terminals were closely connected with TMR retrograde-labeled neurons that were also GABAAR positive. Arrowheads indicate the closely connected sites. Scale bars = 20 μm.

Electron microscopy also further confirmed that TMR-labeled output neurons were GABAAR-positive in the VLO. Four rats were successfully injected with retrograde tracer WGA-HRP into the vlPAG (Fig. 9). In the VLO, TMB reaction products indicated positive dendritic and somatic profiles of WGA-HRP retrograde-labeled neurons, DAB reaction products indicated GABAergic axon terminals, and silver-gold reaction products indicated GABAAR-positive neuronal cell bodies and dendritic processes. Dark clusters of crystal-like TMB reaction products were located in the cytoplasm and intracellular membrane of WGA-HRP retrograde-labeled neuronal cell bodies and dendritic processes. DAB reaction products in GABAergic axon terminals were accumulated in a relatively homogeneous population within the synaptic vesicles. GABAAR-positive neuronal cell bodies and dendritic profiles exhibited silver-gold particles scattered throughout the cytoplasm, particularly in the intracellular membrane. The majority of WGA-HRP retrograde-labeled neuronal cell bodies and dendrites made symmetric synaptic connections with GABAergic axon terminals. Fifty-eight GABAergic axon terminals made symmetric axon-somatic synapses (32.8%; 19/58) and axon-dendritic synapses (67.2%, 39/58), respectively, with WGA-HRP retrograde-labeled neuronal cell bodies and dendrites that were also GABAAR-positive. No axo-axon synapses, asymmetric axon–somatic, or axon–dendritic synapses were observed between the GABAergic axon terminals and the WGA-HRP retrograde-labeled neurons, regardless of whether the cells were GABAAR-positive or not (Fig. 10).

Figure 9.

Light field photomicrograph of coronal sections showing WGA-HRP injection site in the vlPAG of rat. Abbreviations: Aq, aqueduct. Scale bar = 500 μm.

Figure 9.

Light field photomicrograph of coronal sections showing WGA-HRP injection site in the vlPAG of rat. Abbreviations: Aq, aqueduct. Scale bar = 500 μm.

Figure 10.

Triple-label electron microscopy observation showing GABAergic (electron-dense, peroxidase-immunoreactive product) axon terminals (Ax) in symmetric synaptic contact with WGA-HRP retrograde-labeling (electron-dense, cobalt-intensified peroxidase-immunoreactive product) dendritic profile (Den) in the VLO neuron, indicating GABAA receptor expression (immunogold silver grains); arrowheads indicate the synaptic site. Scale bar = 0.6 μm.

Figure 10.

Triple-label electron microscopy observation showing GABAergic (electron-dense, peroxidase-immunoreactive product) axon terminals (Ax) in symmetric synaptic contact with WGA-HRP retrograde-labeling (electron-dense, cobalt-intensified peroxidase-immunoreactive product) dendritic profile (Den) in the VLO neuron, indicating GABAA receptor expression (immunogold silver grains); arrowheads indicate the synaptic site. Scale bar = 0.6 μm.

Discussion

Previous studies have indicated that serotonergic neurons in the DR and adjacent PAG region project their axons bilaterally to the VLO (Li et al. 1993; Matsuzaki et al. 1993). In addition, 5-HT1AR–positive neurons (Wright et al. 1995; Hossein et al. 1996; Barnes and Sharp 1999; Santana et al. 2004), GABAergic neurons (Huo et al. 2005), and GABAAR-positive neurons (Pirker et al. 2000; Princivalle et al. 2001) are distributed in the frontal cortex, including the VLO. The present study not only confirmed these previous results, but also demonstrated local intrinsic connections among serotonergic terminals, 5-HT1AR, GABAergic neurons, GABAAR, and projection neurons in the VLO. Results from immunohistochemistry stainings indicated that GABAergic neuronal cell bodies and 5-HT1AR–positive neuronal cell bodies with similar morphological characteristics were distributed in the same layers (II–VI) of the VLO. Furthermore, dual-labeling immunofluorescent staining indicated that the majority of GABAergic neurons express 5-HT1A receptors, and vice versa. These results strongly suggest the coexistence of GABA and 5-HT1AR in the same neurons in the VLO. Following BDA injections into the DR, many BDA anterograde-labeled fibers and terminals were 5-HT positive, and made close connections with GABAergic neurons. These results indicated that serotonergic terminals from the DR might regulate activities of GABAergic neurons in the VLO. Electron microscopic observations further indicated that serotonergic terminals formed symmetric synapses with the majority of GABAergic neurons where 5-HT1AR–positive expression was also detected. These results further suggest that serotonergic fibers originating from the DR may modulate activities of GABAergic neurons through presynaptic 5-HT1AR in the VLO.

The present results also demonstrated that VLO output neurons (TMR retrograde-labeled neurons in the VLO) projecting to the vlPAG express GABAAR, and also made close connections with GABAergic terminals. The triple-labeled electron microscopic observations further indicated that GABAergic terminals formed symmetric synapses with the majority of the VLO output neurons to the PAG, which expressed GABAAR. These results suggest that inhibitory GABAergic interneurons or terminals via GABAAR may regulate the function of the VLO–PAG–brainstem descending inhibitory system within the VLO.

Together with the results described above, the present study demonstrated a local neuronal circuit within the VLO, comprised of serotonergic afferent terminals, 5-HT1AR, GABAergic interneurons, GABAAR, and projection neurons to the PAG (Fig. 11). These results also provide morphological evidence for the hypothesis that GABAergic modulation may be involved in the VLO 5-HT1A receptor activation-induced descending antinociception.

Figure 11.

Diagram showing a local neuronal circuit in the VLO demonstrated in the present study, which may modulate 5-HT1A receptor activation-evoked descending antinociception through a GABAergic disinhibition mechanism. +, excitation; −, inhibition, 5-HT1A-R, 5-HT1A receptor; 5-HT t, 5-HTergic terminal; GABA n, GABAergic neuron; GABAA-R, GABAA receptor.

Figure 11.

Diagram showing a local neuronal circuit in the VLO demonstrated in the present study, which may modulate 5-HT1A receptor activation-evoked descending antinociception through a GABAergic disinhibition mechanism. +, excitation; −, inhibition, 5-HT1A-R, 5-HT1A receptor; 5-HT t, 5-HTergic terminal; GABA n, GABAergic neuron; GABAA-R, GABAA receptor.

Previous studies have shown that the VLO, as a higher center, is involved in an endogenous analgesic system, which consists of the spinal cord–Sm–VLO–PAG–spinal cord loop (Zhang et al. 1997a, 1997b, 1998). Microinjections of 5-HT or 5-HT1AR agonist (±)-8-hydroxy-DPAT hydrobromide (8-OH-DPAT) into the VLO depress the tail-flick reflex in the rat, and this effect is antagonized by a microinjection of 5-HT1AR antagonist NAN-190 hydrobromide (NAN-190) into the same site. These results suggest that 5-HT, via 5-HT1AR, might be involved in mediating VLO-induced descending antinociception (Huo et al. 2008; Qu et al. 2008). It is well known that 5-HT1AR is an inhibitory G-protein–coupled receptor and that activation of this receptor inhibits neuronal activity via cell membrane hyperpolarization, an effect mediated by the pertussis toxin (PTX)–sensitive G-protein–coupled opening of 4-AP–sensitive K+ channels in the frontal cortex, hippocampus, and septum (Aghajanian 1995; Albert et al. 1997). It has been suggested that the 5-HT1AR activation-induced excitatory effect may be induced by blocking an inhibitory GABAergic interneuron (disinhibition), thereby leading to the activation of projection neurons (Koyama et al. 1999, 2002; Okuhara and Beck 1994; Susana et al. 2003). Therefore, it is reasonable to propose that 5-HT, via 5-HT1AR–mediated antinociception in the VLO, might be produced by blocking the tonically inhibitory action of GABAergic interneurons on the output neurons projecting to the PAG. This disinhibitory effect would lead to enhanced activity of the VLO–PAG–brainstem descending inhibitory system and depression of the nociceptive inputs at the spinal dorsal horn. Evidence to support this hypothesis has been obtained in behavioral studies using microinjections of the GABAA receptor antagonists, bicuculline or picrotoxin, into the VLO, resulting in dose-dependent depression of the tail-flick reflex (Qu et al. 2006). A smaller dose of bicuculline or picrotoxin enhances 5-HT1A receptor agonist 8-OH-DPAT–induced inhibition of the tail-flick reflex, whereas microinjections of the GABAA receptor agonists, muscimol or HITP, attenuate the effect of 8-OH-DPAT (Huo et al. 2008). The present study provides morphological evidence for the above-mentioned hypothesis.

In addition, this study also determined that the serotonergic terminals, GABAergic neurons, and 5-HT1AR–positive neurons are also distributed in the lateral orbital and agranular insular cortices. These appeared frequently, with a distribution pattern and synaptic connection very similar to what was observed in the VLO, suggesting that these forebrain structures may also be implicated in 5-HT1AR–mediated antinociception, because both regions have been implicated in modulation of nociception (Burkey et al., 1996, 1999; Zhang et al. 1997b, 1998). In fact, behavioral and electrophysiological studies have indicated that GABAergic disinhibition mechanisms induced by 5-HT1AR activation may exist in many other brain regions, such as the cerebral cortex, basolateral amygdala, hippocampus, thalamus, locus coeruleus, and PAG. (Okuhara and Beck 1994; Matsuyama et al. 1997; Koyama et al. 1999, 2002; Kishimoto et al. 2001; Susana et al. 2003; Xiao et al. 2005a, 2005b). Therefore, the mechanism of disinhibition induced by 5-HT via 5-HT1AR may be a universal phenomenon in the central nervous system. Morphological methods in the present study demonstrated a local neuronal circuit that may also be applicable to other brain regions in order to better understand GABAergic modulation of 5-HT1AR activation-induced excitatory effects.

Conclusion

In conclusion, the present study suggests that a local neuronal circuit, consisting of 5-HT terminals, 5-HT1AR, GABAergic interneurons, GABAAR, and projection neurons, exists in the VLO. These results provide morphological evidence for the hypothesis that GABAergic modulation is involved in 5-HT1AR activation-induced descending antinociception within the VLO.

Funding

National Natural Science Foundation of China grants (No. 30570592) for JS. Tang; Innovation Research Team Program of the Ministry of Education of China (IRT0560); and National Program of Basic Research of China (G2006CB500808) for YQ. Li.

We wish to thank Liwen Bianji (Edanz Editing China) for expert help in preparing the manuscript. Conflict of Interest: None declared.

References

Aghajanian
GK
Bloom
FR
Kupfer
DG
Electrophysiology of serotonin receptor subtypes and signal transduction pathways
Psychopharmacology: the fourth generation of progress
 , 
1995
New York
Raven
(pg. 
1451
-
1459
)
Albert
PR
Lembo
P
Storring
JM
Charest
A
Saucier
C
The 5-HT1A receptor: signaling, desensitization, and gene transcription
Neuropsychopharmacology.
 , 
1997
, vol. 
14
 (pg. 
19
-
25
)
Barnes
NM
Sharp
T
A review of central 5-HT receptors and their function
Neuropharmacology.
 , 
1999
, vol. 
38
 (pg. 
1083
-
1152
)
Burkey
AR
Carstens
E
Jasmin
L
Dopamine reuptake inhibition in the rostral agranular insular cortex produces antinociception
J Neurosci.
 , 
1999
, vol. 
19
 (pg. 
4169
-
4179
)
Burkey
AR
Carstens
E
Wenniger
JJ
Tang
J
Jasmin
L
An opioidergic cortical antinociception triggering site in the agranular insular cortex of the rat that contributes to morphine antinociception
J Neurosci.
 , 
1996
, vol. 
16
 (pg. 
6612
-
6623
)
Coffield
JA
Bowen
KK
Miletic
V
Retrograde tracing of projections between the nucleus submedius, the ventrolateral orbital cortex and the midbrain in the rat
J Comp Neurol.
 , 
1992
, vol. 
321
 (pg. 
488
-
499
)
Craig
AD
Burton
H
Spinal and medullary Lamina I projection to nucleus submedius in medial thalamus: a possible pain center
J Neurophysiol.
 , 
1981
, vol. 
45
 (pg. 
443
-
465
)
Craig
AD
Wiegand
SJ
Price
JL
The thalamocortical projection of the nucleus submedius in the cat
J Comp Neurol.
 , 
1982
, vol. 
206
 (pg. 
28
-
48
)
Fields
HL
Basbaum
A
Wall
PD
Melzack
R
Central nervous system mechanisms of pain modulation
Textbook of pain
 , 
1999
4th ed
New York
Churchill Livingstone
(pg. 
309
-
329
)
Hardy
SGP
Leichnetz
GR
Frontal cortex projections to the periaqueductal gray in the rat: a retrograde and orthograde horseradish peroxidase study
Neurosci Lett.
 , 
1981
, vol. 
23
 (pg. 
13
-
17
)
Hossein
KK
Marie-Christine
M
Marie-Jeanne
B
Genevieve
D
Mustapha
R
Salah
EM
Michel
H
Daniel
V
Immunocytochemical location of seretonin1A receptors in the rat central nervous system
J Comp Neurol.
 , 
1996
, vol. 
365
 (pg. 
289
-
305
)
Hsu
SM
Raine
L
Fanger
H
Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures
J Histochem Cytochem.
 , 
1981
, vol. 
29
 (pg. 
577
-
580
)
Huo
FQ
Chen
T
Wang
J
Li
YQ
Tang
JS
The relationship of serotoninergic terminals and GABAergic neurons in the ventrolateral orbital cortex of rat: a light and electron microscopic study
Neurosci Bull.
 , 
2007
, vol. 
23
 (pg. 
154
-
155
)
Huo
FQ
Qu
CL
Li
YQ
Tang
JS
Jia
H
Forthcoming 2008.
GABAergic modulation is involved in the ventrolateral orbital cortex 5-HT1A -induced antinociception in the rat
Pain.
 
Huo
FQ
Wang
J
Li
YQ
Chen
T
Han
F
Tang
JS
GABAergic neurons express μ-opioid receptor in the ventrolateral orbital cortex of rat
Neurosci Lett.
 , 
2005
, vol. 
382
 (pg. 
265
-
268
)
Kishimoto
K
Koyama
S
Akaike
N
Synergistic μ-opioid and 5-HT1A presynaptic inhibition of GABA release in rat periaqueductal gray neurons
Neuropharmacology.
 , 
2001
, vol. 
41
 (pg. 
529
-
538
)
Koyama
S
Kubo
C
Rhee
JS
Akaike
N
Presynaptic serotonergic inhibition of GABAergic synaptic transmission in mechanically dissociated rat basolateral amygdala neurons
J Physiol.
 , 
1999
, vol. 
518
 (pg. 
525
-
538
)
Koyama
S
Matsumoto
N
Murakami
N
Kubo
C
Nabekura
J
Akaike
N
Role of presynaptic 5-HT1A and 5-HT3 receptors in modulation of synaptic GABA transmission in dissociated rat basolateral amygdala neurons
Life Sci.
 , 
2002
, vol. 
72
 (pg. 
375
-
387
)
Li
JL
Wang
D
Kaneko
T
Shigemoto
R
Nomura
S
Mizuno
N
Relationship between neurokinin-1 receptor and substance P in the striatum: light and electron microscopic immunohistochemical study in the rat
J Comp Neurol.
 , 
2000
, vol. 
418
 (pg. 
156
-
163
)
Li
YQ
Takada
M
Matsuzaki
S
Shinonaga
Y
Mizuno
Identification of periaqueductal gray and dorsal raphe nucleus neurons projecting to both trigeminal sensory complex and forebrain structures: a fluorescent retrograde double-labeling study in the rat
Brain Res.
 , 
1993
, vol. 
623
 (pg. 
267
-
277
)
Matsuyama
S
Nei
K
Tanaka
C
Regulation of GABA release via NMDA and 5-HT1A receptors in guinea pig dentate gyrus
Brain Res.
 , 
1997
, vol. 
761
 (pg. 
105
-
112
)
Matsuzaki
S
Takada
M
Li
YQ
Tokuno
H
Mizuno
N
Serotonergic projections from the dorsal raphe nucleus to the nucleus submedius in the rat and cat
Neuroscience.
 , 
1993
, vol. 
55
 (pg. 
403
-
416
)
Okuhara
DY
Beck
SG
5-HT1A Receptor linked to inward-rectifying potassium current in hippocampal CA3 pyramidal cells
J Neurophysiol.
 , 
1994
, vol. 
71
 (pg. 
2161
-
2167
)
Paxinos
G
Watson
C
The rat brain in stereotaxic coordinates
 , 
1986
2nd ed.
Sydney
Academic Press
Pirker
S
Schwarzer
C
Wieselthaler
A
Sieghart
W
Sperk
G
GABA(A) receptors: immunocytochemical distribution of 13 subunits in the adult rat brain
Neuroscience.
 , 
2000
, vol. 
101
 (pg. 
815
-
850
)
Princivalle
AP
Pangalos
MN
Bowery
NG
Spreafico
R
Distribution of GABA (B (1a)), GABA (B (1b)) and GABA (B2) receptor protein in cerebral cortex and thalamus of adult rats
Neuroreport.
 , 
2001
, vol. 
12
 (pg. 
591
-
595
)
Qu
CL
Huo
FQ
Huang
FS
Li
YQ
Tang
JS
Jia
H
The role of 5-HT receptor subtypes in the ventrolateral orbital cortex of 5-HT-induced antinociception in the rat
Neuroscience.
 , 
2008
, vol. 
152
 (pg. 
487
-
494
)
Qu
CL
Tang
JS
Jia
H
Involvement of GABAergic modulation of antinociception induced by morphine microinjected into the ventrolateral orbital cortex
Brain Res
 , 
2006
, vol. 
1073-1074c
 (pg. 
281
-
289
)
Santana
N
Bortolozzi
A
Serrats
J
Mengod
G
Artigas
F
Expression of serotonin1A and serotonin2A receptors in pyramidal and GABAergic neurons of the rat prefrontal cortex
Cereb Cortex.
 , 
2004
, vol. 
14
 (pg. 
1100
-
1109
)
Susana
A
Zhao
XQ
Reshma
S
Birgitte
R
Gitte
MK
The 5-HT1A serotonin receptor is located on calbindin- and parvalbumin-containing neurons in the rat brain
Brain Res.
 , 
2003
, vol. 
959
 (pg. 
58
-
67
)
Wright
DE
Seroogy
KB
Lundgren
KH
Davis
BM
Jennes
L
Comparative localization of serotonin 1A, 1C, and receptor subtype mRNAs in rat brain
J Comp Neurol.
 , 
1995
, vol. 
351
 (pg. 
357
-
373
)
Xiao
DQ
Zhu
JX
Tang
JS
Jia
H
5-hydroxytryptamine1A (5-HT1A) but not 5-HT3 receptor is involved in mediating the nucleus submedius 5-HT-induced antinociception in the rat
Brain Res.
 , 
2005
, vol. 
1046
 (pg. 
38
-
44
)
Xiao
DQ
Zhu
JX
Tang
JS
Jia
H
GABAergic modulation mediates antinociception produced by serotonin applied into thalamic nucleus submedius of the rat
Brain Res.
 , 
2005
, vol. 
1057
 (pg. 
161
-
167
)
Yoshida
A
Dostrovsky
JO
Chiang
CY
The afferent and efferent connections of the nucleus submedius in the rat
J Comp Neurol.
 , 
1992
, vol. 
324
 (pg. 
115
-
133
)
Yoshida
A
Dostrovsky
JO
Sessle
BJ
Chiang
CY
Trigeminal projections to the nucleus submedius of the thalamus in the rat
J Comp Neurol.
 , 
1991
, vol. 
307
 (pg. 
609
-
625
)
Zhang
S
Tang
JS
Yuan
B
Jia
H
Involvement of the frontal ventrolateral orbital cortex in descending inhibition of nociception mediated by the periaqueductal gray in rats
Neurosci Lett.
 , 
1997
, vol. 
224
 (pg. 
142
-
146
)
Zhang
YQ
Tang
JS
Yuan
B
Jia
H
Inhibitory effects of electrically evoked activation of ventrolateral orbital cortex on the tail-flick reflex are mediated by periaqueductal gray in rats
Pain.
 , 
1997
, vol. 
72
 (pg. 
127
-
135
)
Zhang
S
Tang
JS
Yuan
B
Jia
H
Inhibitory effects of glutamate- induced activation of thalamic nucleus submedius are mediated by ventrolateral orbital cortex and periaqueductal gray in rats
Eur J Pain.
 , 
1998
, vol. 
2
 (pg. 
153
-
163
)
Zimmermann
M
Ethical guideline for investigations of experimental pain in conscious animals
Pain.
 , 
1983
, vol. 
16
 (pg. 
109
-
110
)

Author notes

Fu-Quan Huo and Tao Chen contributed equally to this work.