Immunohistochemical Localization of the VIP 1 Receptor (VPAC 1 R) in Rat Cerebral Blood Vessels: Relation to PACAP and VIP Containing Nerves

Abstract

The two structurally related peptides, vasoactive intestinal polypeptide (VIP) and pituitary adenylate cyclase activating polypeptide (PACAP), are present in cerebral vascular nerve fibers. Biologic actions of VIP are exerted through two receptors, VPAC₁ and VPAC₂, having similar binding affinity for both VIP and PACAP. In the current study, the authors have developed a specific antibody against the rVPAC₁ receptor to examine the localization of rVPAC₁ immunoreactivity in cerebral arteries and arterioles of the rat by immunohistochemistry using fluorescence confocal microscopy. Specificity of the antiserum was ensured by immunoblotting and immunocytochemistry of cells transfected with cDNA encoding the different PACAP-VIP receptor subtypes. The rVPAC₁ receptor immunoreactivity was localized to the plasmalemma of circularly orientated smooth muscle cells on superficial cerebral arteries and arterioles taken from the basal surface of the brain. By double immunostaining VIP immunoreactive nerve fibers and, to a lesser extent, those containing PACAP were shown to have intimate contact with the receptor protein. Vasoactive intestinal polypeptide and PACAP containing cerebrovascular nerve fibers were found in separate nerve populations with different distribution pattern and density. In brain sections processes of cortical VIP-, but not PACAP-, containing neurons seemed to innervate the rVPAC₁ receptor of pial arterioles on the brain surface. The current findings provide the neuroanatomical substrate for a role of VIP and maybe PACAP in the regulation of cerebral blood flow.

Keywords

Cerebral blood vessels Immunohistochemistry Innervation Vascular smooth muscle

Vasoactive intestinal polypeptide (VIP) is a 28 amino acid pleiotropic peptide with a widespread occurrence in neurons of the central and peripheral nervous system (Fahrenkrug, 1993). A rich supply of nerve fibers containing VIP has been described around cerebral blood vessels in a number of species including mouse, gerbil, rat, cat, dog, pig, cow, monkey, and human (Edvinsson et al., 1980; Edvinsson and McCulloch, 1985; Gibbins et al., 1984; Hara et al., 1985,1989; Larsson et al., 1976; Lee et al., 1984; Matsuyama et al., 1983; Miao and Lee, 1990; Suzuki et al., 1988; Uemura et al., 1988; Zhang et al., 1991). In rat and dog, the nerve cell bodies giving rise to the VIP immunoreactive fibers of the large cerebral arteries are contained in the ipsilateral sphenopalatine and otic ganglia and in the internal carotic miniganglion (Hara et al., 1989; Suzuki et al., 1988; Uemura et al., 1988).

Pituitary adenylate cyclase activating polypeptide (PACAP), recently identified as a member of the VIP family of neuropeptides, exists in two biologically active forms, PACAP-38 and PACAP-27, both derived from the same precursor. Like VIP, PACAP is widely expressed in the nervous system having a broad spectrum of biologic functions including neurotransmitter and neuroprotective actions (Arimura, 1998). Nerve fibers containing PACAP have been demonstrated in the adventitia and adventitia media border of cat cerebral arteries (Uddman et al., 1993), and PACAP and VIP can dilate cerebral blood vessels suggesting a role in the regulation of cerebral circulation (Dacey et al., 1988; Edvinsson and McCulloch, 1985; Jansen-Olesen et al., 1994; Larsson et al., 1976; McCulloch and Edvinsson, 1980; Seki et al., 1995; Uddman et al., 1993; Wei et al., 1980). Because PACAP-27, PACAP-38, and VIP result in a similar concentration-dependent vasorelaxation, and a VIP receptor antagonist was able to inhibit the VIP and PACAP-38 provoked increase in vertebral artery blood flow, it is possible that the peptides are working through a common receptor (Anzai et al., 1995; Seki et al., 1995). Vasoactive intestinal polypeptide and the two alternative processing forms of PACAP are recognized by 3 G-protein coupled receptors with a large extracellular N-terminal domain and 7 transmembrane helices: the PACAP (or PAC₁) receptor that has a much higher affinity for the 2 forms of PACAP than for VIP, and the VIP₁ (or VPAC₁) and VIP₂ or (VPAC₂) receptors that have a similar binding affinity for VIP and PACAP (Harmar et al., 1998).

In the current study, the authors have developed an antiserum against the N-terminus of the rVPAC₁ receptor and used it for studying rat cerebral blood vessels by immunohistochemistry. The localization of the rVPAC₁ receptor immunoreactivity and the relationship between the receptor protein and VIP or PACAP immunoreactive nerve fibers are described.

MATERIALS AND METHODS

rVPAC1 receptor antibody preparation

Polyclonal antibodies were generated against the rVPAC₁ receptor N-terminal fusion protein in random bred white Danish rabbits. Each animal received 1 mg of the receptor fusion protein per immunization emulsified in double volume Freunds adjuvans.

The 363 bp SacII/BbsI fragment of the rVPAC₁ receptor complementary deoxyribonucleic acid (cDNA) containing the N-terminal was subcloned as a blunt end fragment in the BamHI site of pGEX2T (Amersham Pharmacia Biotech, Allerød, Denmark). The fusion protein product, with glutathione S-transferase followed by the rVPAC₁ receptor N-terminal, was termed VNG270. Fusion protein production in UT5600 E. coli cells, grown in Luria-Bertani broth (10 g bacto-tryptone/L, 5 g bacto-yeast extract/L, and 5 g sodium chloride/L), was initiated by 0.5 mmol/L Isopropyl beta D-thiogalactopyranoside and incubated at 15°C overnight. Cells (7 g) were pelleted at 8,000 g for 10 minutes and resuspended in 320 mL 40 mmol/L 2-amino-2-(hydroxymethyl)-1,3-propandiol (Tris), 20 mmol/L hydrochloric acid (HCl), 10 mmol/L dithiothreitol (DTT), 1 μmol/L leupeptin, 1 mmol/L phenylmethylsulfonylflorid and sonicated (150 W) for 10 minutes in an ice bath. The homogenate was centrifuged at 30,000 g for 10 minutes at 4°C. Sodium phosphate was added to the supernatant to a final concentration of 50 mmol/L and pH 7.0 followed by the addition of 2 mL glutathione Sepharose (Amersham) and incubated for 2 hours at 4°C. The glutathione Sepharose was pelleted by centrifugation for 10 minutes at 500 g at 4°C. The pellet was washed with 3 × 40 mL 20 mmol/L 2-(bis(2-hydroxyethyl)imino)-2- (hydroxymethyl)-1,3-propanediol (Bis-Tris), 10 mmol/L HCl, 10 mmol/L DTT, and 4 mL 50% glycerol was added to the pellet and stored at −20°C. Glutathione Sepharose, with VNG270 attached, was washed twice with 20 mmol/L Bis-Tris, 10 mmol/L HCl, 10 mmol/L DTT, and VNG270 was detached from the Sepharose by incubation with 50 mmol/L Tris, 15 mmol/L acetic acid, 10 mmol/L glutathione, 5 mmol/L DTT for 30 minutes at room temperature. The suspension was centrifuged for 2 minutes at 500 g at 20°C and iodoacetamide was added to the supernatant to a final concentration of 30 mmol/L and incubated for 30 minutes at room temperature.

Antiserum, code No. 4G28, was selected, and the antibody specificity was determined by using Chinese hamster ovary cells (CHO) stably transfected with rVPAC₁ and human embryonic kidney cells 293 (HEK 293) transiently transfected with rPAC₁ and rVPAC₂ receptor.

Purification of rVPAC1 receptor N-terminal

Sodium dodecyl sulfate (SDS) with a final concentration of 0.01% was added to the purified product, VNG270, used for immunization and incubated overnight at room temperature with 0.1 g/L thrombin. Acetonitrile was added to a final concentration of 32% and pH was adjusted to pH 2 with trifluoroacetic acid, and the digest was separated at 50°C by C-4 reverse phase HPLC on a 4.6 × 250 mm column using a linear gradient from 32% to 45% acetonitrile in 0.1% trifluoroacetic acid at 1 mL/min for 60 minutes. Sodium dodecyl sulfate polyacrylamide gel electrophoresis was used to identify the peak containing the rVPAC₁ receptor N-terminal, and amino acid analysis verified the expected content of amino acids.

Immunoblot analysis

Confluent monolayers of transfected cells were washed with phosphate buffered saline (PBS) and detached from their plastic support using a cell scraper. Cells were solubilized in 20 mL 25 mmol/L N-2-hydroxyethylpiperazine-N′-2-ethane-sulfonic acid (HEPES), 2.5 mmol/L CaCl₂, 1.0 mmol/L MgCl₂, 50 mg/L bacitracin, pH 7.4 per plate and disrupted using a polytron for 30 seconds. The homogenate was spun for 20 minutes at 30,000 g at 4°C. The resulting pellet was resuspended in 15 mL 25 mmol/L HEPES, 2.5 mmol/L CaCl₂, 1.0 mmol/L MgCl₂, 50 mg/L bacitracin, pH 7.4 per plate. The preparation was aliquoted and stored at −80°C.

Membrane protein samples were electrophoresed on 10% Tricine SDS-polyacrylamide gels according to the methods of Schägger and von Jagow (1987). After electrophoresis the proteins were transferred to polyvinylidendifluorid (PVDF) membranes (NOVEX, San Diego, CA, U.S.A.). The PVDF membranes were incubated in Tris buffer saline (TBS; 20 mmol/L TrisCl, 140 mmol/L NaCl, pH 7.4) containing 10% Tween-20 overnight at room temperature followed by an incubation with 0.01 mg/mL avidin and the antireceptor antiserum diluted to 1:2,000 in TBS containing 0.1% Tween-20 (TBS-T). After two quick washes with TBS the washing was extended with the addition of 0.01 mg/mL biotin for 3 × 10 minutes. Hereafter the PVDF membranes were incubated with biotinylated donkey anti-rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA, U.S.A.) diluted to 1:20,000 in TBS-T for 30 minutes at room temperature. After two quick and three 10-minute washes with TBS-T, the PVDF membranes were incubated with StreptABComplex/HRP (Dako A/S, Copenhagen, Denmark) diluted to 1:40 in TBS-T for 30 minutes at room temperature. After two quick and three 10-minute washes with TBS-T, the PVDF membranes were incubated with enhanced chemiluminescence reagents according to the manufacturer's instructions (Amersham) and apposed to Hyperfilm (Amersham) and developed for 5 seconds to 20 minutes.

Preabsorption of rVPAC1 receptor antiserum

Glutathione Sepharose with VNG270 attached, 0.5 mL, was applied on a poly-prep chromatography column, 0.8 × 4 cm (Bio-Rad Laboratories, Hercules, U.S.A.), and washed with 3 × 5 mL 20 mmol/L sodium phosphate buffer, 1 mmol/L DTT, pH 7.0. One mL 20 mmol/L sodium phosphate buffer, 10 mmol/L iodoacetamide, pH 7.0 were applied on the column and washed with 5 mL 20 mmol/L sodium phosphate buffer, pH 7.0. One mL 4G28 antiserum was applied on the column and the run through was used as preabsorbed antiserum.

Immunocytochemical characterization of the rVPAC1 receptor antiserum

The transfected cells were grown in eight-chamber glass slides washed with PBS and fixed for 1 hour in 2% paraformaldehyde, 0.2% picric acid in 0.1 mol/L sodium phosphate buffer, pH 7.2. After washing and drying the cells were immunostained with anti-rVPAC₁ receptor antiserum (code No. 4G28) in a dilution of 1:640 using a biotinylated donkey anti-rabbit secondary antiserum (711-065-152 Jackson ImmunoResearch, diluted 1:1,000) and the avidin-biotin complex/streptavidin horseradish peroxidase complex as previously described (Fahrenkrug and Hannibal, 1998).

Animal preparation

Male Wistar rats weighing 180 to 200 g and housed under standard laboratory conditions in a 12-hour light—dark cycle with free access to food and water were used. On the day of fixation the animals were anesthetized with tribromoethanol (20 mg/100 g, BW) and perfused through the heart with a room temperature solution of saline 0.9% to which heparin (15,000, IU/L) was added 75 to 100 mL over 3 minutes. The perfusion was followed by 2% paraformaldehyde, 0.2% picric acid in 0.1 mol/L sodium phosphate buffer, pH 7.2, 300 mL over 15 minutes. In 10 rats, brains were rapidly removed from the skull and the major extracerebral arteries on the base of the skull were removed by careful dissection under a stereomicroscope. The vessels were postfixed in the same fixative for 4 hours at 4°C, and subsequently the vessel preparations were washed in PBS overnight. Fixed brains from 6 additional rats were postfixed for 24 hours at 4°C and subsequently equilibrated in PBS 0.05 mol/L, pH 7.4 containing 30% sucrose for 48 hours and 4°C and then sectioned coronally at 40 μm in a cryostat. The resulting sections were collected as free floating sections in PBS.

Immunohistochemistry

A previously characterized mouse monoclonal PACAP antibody (Code No. MabJHH1, diluted 1:5) directed against an epitope of the amino acid sequence 6–16 of PACAP recognizing both PACAP-38 and PACAP-27 was used for PACAP immunostaining (Hannibal et al., 1995). The monoclonal mouse anti-VIP antibody (Code No. 55, diluted 1:500) was kindly donated by John H. Walsh (Wong et al., 1996). The rabbit anti-VIP antiserum (Code No. 291E, diluted 1:1,000) has been characterized previously (Fahrenkrug et al., 1995). As controls, sections were routinely incubated with antibodies preabsorbed with the respective antigen (20 μg/mL) that abolished all staining. The rVPAC₁ receptor antibody was used in a dilution of 1:500. Whole mounts of cerebral arteries and arterioles and free floating brain sections were processed for single or double antigen fluorescence immunohistochemistry as previously described (Fahrenkrug and Hannibal, 1998). Briefly, the whole mounts or sections were incubated with combinations of primary antibodies for 12 to 18 hours at 4°C. After washing for 3 × 10 minutes in 0.25% bovine serum albumin + 0.1% Triton X-100 (PBS-BT) the whole mounts or sections were incubated with a combination of Cy2 conjugated donkey anti-rabbit (diluted 1:100 for visualization of VIP and rVPAC₁ receptor), or Texas Red® conjugated goat anti-mouse antisera (diluted 1:200 for visualization of VIP), or both, and biotinylated donkey anti-mouse antiserum (diluted 1:600 for visualization of PACAP). All secondary antisera were from Jackson Immunoresearch Laboratories. After washing for 3 × 5 minutes in PBS-BT and incubation for 30 minutes at room temperature in ABC-Streptavidin horse radish peroxidase complex (diluted 1:150; Vector, Burlingame, CA, U.S.A.), the sections or whole mounts were incubated in biotinylated tyramide (diluted 1:50 in application buffer; Tyramide System Amplification, DuPont NEN, Boston, MA, U.S.A.). After washing for 3 × 15 minutes in PBS-BT, the material was incubated for 30 minutes in streptavidin Texas Red conjugated complex (diluted 1:200; Amersham) followed by 3 × 10 minutes washes in PBS-BT and mounting in glycerol/PBS 1:1). For absorption control the anti-rVPAC₁ receptor antiserum was incubated with 2 × 10⁻⁵ mol/L N-terminal receptor fusion protein for 24 hours at room temperature before immunohistochemistry. In addition, control experiments with omission of anti-rVPAC₁ receptor antibody from the first incubation did not detect any staining imputable to the second antibody.

Whole mounts and brain sections were examined in a Leitz DMRB fluorescence microscope equipped with standard filter settings (13, code No. 513 808, excitation BP 450–490 nm, and N2.1, code No. 513 803, excitation PB 515–560 nm) and a red—green filter (code No. 513 803) allowing simultaneous visualization of red and green fluorescence. Photomicrographs were obtained using a Leica DC 200 camera and software (Leica, Herlev, Denmark), or a confocal laser scanning Zeiss LSM 510 microscope, or both. Sections obtained by the confocal microscope were 1.5 to 5.5-μm thick. Image editing software (Adobe Photoshop and Adobe Illustrator, Adobe, San José, CA, U.S.A.) was used to combine the obtained images into plates, and figures were printed on a Tektronix (Wilsonville, OR, U.S.A.) Phase 450 dye sublimation printer.

RESULTS

Immunoblot analysis

The antiserum 4G28 reacted strongly with a band of 45 kDa in CHO cells stably transfected with a rVPAC₁ receptor expression plasmid, whereas the immunoreactivity was completely abolished after preabsorption of the antiserum with the rVPAC₁ receptor N-terminal fusion protein (Fig. 1A). The antiserum did not react with the parental CHO cells that do not express the rVPAC₁ receptor nor with membrane preparations from cells transfected with the rVPAC₂ or rPAC₁ receptor subtypes (Fig. 1A). The antiserum detected the purified rVPAC₁ receptor N-terminal expressed in E. coli as a band of 14 kDa, and it was estimated that the antiserum was able to detect the protein above 3 pmol (Fig. 1B).

FIG. 1.

(A) Western blot analysis of 4G28 immunoreactivity in transfected cells. Membranes (50 μg) from either Chinese hamster ovary (CHO) cells transfected with cDNA encoding rVPAC₁ receptor (lane 1) or parental CHO cells (lane 2). Preabsorbed 4G28 antiserum is shown in lane 3 (rVPAC₁). Membranes (50 μg) from HEK293 cells transfected with cDNA encoding rat VPAC₂ receptor (lane 4) or rat PAC₁ receptor (lane 5). (B) Quantitative estimation of 4G28 immunoreactivity in Western blot analysis. Purified rVPAC₁ receptor N-terminal expressed in E. coli is applied in lanes 1, 2, and 3 with 300, 30, and 3 pmol, respectively, using 4G28 antiserum.

Visualization of the rVPAC1 receptor in transfected cells

The CHO cells stably transfected with the cDNA encoding the rVPAC₁ receptor were immunostained with the 4G28 antiserum (Fig. 2A), whereas no immunoreaction was detected in nontransfected CHO cells (Fig. 2B) or cells transfected with the cDNA encoding the other receptor subtypes (data not shown).

FIG. 2.

Immunocytochemical staining of the rVPAC₁ receptor in Chinese hamster ovary (CHO) cells using antiserum 4G28. Specific staining is seen in the cells stably transfected with cDNA encoding the rVPAC₁ receptor (A), whereas no labeling is observed in nontransfected CHO cells (B). Immunofluorescence photomicrograph of whole mount preparation of the basilar artery showing immunostaining of the rVPAC₁ receptor in the circularly oriented smooth muscle cells (C). Incubation with antibody 4G28 preabsorbed with the N-terminal rVPAC₁ receptor fusion protein completely abolished immunostaining (D). Scale bars = 100 μm.

Visualization of the rVPAC1 receptor in cerebral arteries and arterioles

Whole mount preparation of the basal cerebral arteries showed a distinct pattern of immunostaining that was absent after preabsorption of the rVPAC₁ receptor antibody with the N-terminal receptor protein (Fig. 2C and 2D). The rVPAC₁ receptor immunoreactivity was observed in smooth muscle cells both in superficial cerebral arteries and in arterioles taken from the basal surface of the brain. The strongly immunostained circularly orientated smooth muscles cells were seen in the basilar artery (Fig. 3A and 3B), the arteries of the circle of Willis, the middle cerebral artery, and the anterior and posterior cerebral arteries and their branches (Figs. 3G, 4A, and 4C). In the confocal microscope the rVPAC₁ receptor immunoreactivity was exclusively detected on the plasmalemma of the vascular smooth muscle cells (Fig. 3A, 3B, and 3E) and no immunostaining of the endothelial cell layer was observed.

FIG. 3.

(A and B) Confocal photomicrographs showing rVPAC₁ receptor immunoreactivity of whole mounts of the same artery scanned at two different levels. (C and D) Photomicrographs showing double immunostaining of rVPAC₁ receptor immunoreactivity (green) and vasoactive intestinal polypeptide (VIP) (red) in free floating cross-section of the anterior cerebral artery. (C) Low power photomicrograph showing VIP immunoreactive nerve fibers surrounding the vessel at the adventitia media border. (D) High power confocal photomicrograph showing the rVPAC₁ on the surface of the smooth muscle cells is intimately associated with VIP immunoreactive nerve fibers (arrows). (E) Confocal photomicrograph showing double immunostaining of rVPAC₁ receptor immunoreactivity (green) and VIP (red) of whole mount of the basilar artery. VIP immunoreactive nerve fibers with close contact to rVPAC₁ receptor on the plasmalemma of the cross-sectioned smooth muscle cells are indicated by arrows. (F) Confocal photomicrograph showing double immunostaining for VIP (green) and pituitary adenylate cyclase activating polypeptide (PACAP) (red) in whole mounts of the basilar artery. Note that the two peptides have different distribution patterns and are rarely colocalized. (G) Confocal photomicrograph showing double immunostaining for rVPAC₁ immunoreactivity (green) and PACAP (red) in whole mount of the posterior cerebral artery. (H) High power confocal photomicrograph of double immunostaining of rVPAC₁ receptor immunoreactivity (green) and PACAP (red) in free floating cross-section of the anterior cerebral artery showing the close contact between the rVPAC₁ receptor and PACAP containing nerve fibers (arrows). Scale bars = 50 μm (A, B, E, and G), 100 μm (C), 20 μm (D and H), 200 μm (F). *indicates vascular lumen.

FIG. 4.

Immunofluorescence photomicrographs of arteries and arterioles isolated from the basal surface of rat brain. rVPAC₁ staining along the surface of circularly orientated smooth muscle cells in a large (A) and small (C) arteriole. (B) A single longitudinally orientated pituitary adenylate cyclase activating polypeptide (PACAP) immunoreactive nerve fiber innervating the same arteriole as shown in A (double immunofluorescence). (D and E) Showing double immunostaining for PACAP (D) and vasoactive intestinal polypeptide (VIP) (E) in a first branch of the middle cerebral artery. Note the intense plexus of VIP immunoreactive nerve fibers, whereas PACAP is observed in a few longitudinally orientated nerve fibers. Scale bars = 50 μm (A, B, D, and E), and 100 μm (C).

Examination of free floating brain sections confirmed the strong VIP receptor immunoreactivity found in isolated arteries and arterioles (Fig. 3C, 3D, and 3H). Based on visualization the intensity of immunostaining seemed to be similar in all arteries and arterioles.

Double immunolabeling

A well-developed plexus of perivascular VIP containing nerve fibers was observed in the wall of all major cerebral arteries and their branches (Figs. 3E, 3F, and 4E). Pituitary adenylate cyclase activating polypeptide immunoreactive nerve fibers were much less abundant, and the density of PACAP positive nerve fibers decreased as the diameter of the arterioles became smaller (Figs. 3F, 3G, 4B, and 4D). Although VIP formed a dense network of varicose nerve fibers mainly circularly orientated, the PACAP containing nerve fibers ran primarily along the long axis of the vessels. By double immunostaining, the two peptides were generally found not to be colocalized in the same nerve fibers (Fig. 3F). In free floating sections delicate varicose PACAP or VIP containing nerve fibers were encountered in the adventitia media border of the arteries (Fig. 3C). Confocal microscopical analysis at high magnification of free floating sections and whole mount arterial preparations after double immunostaining with the receptor antibody disclosed that PACAP and VIP immunoreactive nerve fibers had intimate contact with the receptor protein on the surface of the vascular smooth muscle cells (Fig. 3D, 3E, and 3H).

In coronal sections of the cortex processes arising from the superficial poles of VIP containing cell bodies were often found to extend into the pial tissue on the brain surface. By double immunostaining, these processes, which sometimes dichotomized, were shown to innervate the rVPAC₁ receptor of the pial arterioles (Fig. 5A and 5B). Pituitary adenylate cyclase activating polypeptide immunoreactive perikarya, however, did not give rise to projections which extended to the pial arterioles. As previously reported VIP immunoreactivity in the parietal cortex was localized in a homogenous population of radially oriented bipolar neurons in cortical layers II-IV and their processes (Fig. 5C) (Lorén et al., 1979; Morrison et al., 1984; Zhang et al., 1991). The distribution pattern of the PACAP immunoreactive cell bodies differed from that of VIP because they were present in all layers (Fig. 5D), although mainly concentrated in the deeper cortical layers. The number of the PACAP immunoreactive nerve cell bodies seemed to outnumber those of VIP (Fig. 5C and 5D). Generally VIP and PACAP were observed to be stored in separate cortical neurons, but occasionally a few cortical perikarya were shown to contain both peptides (Fig. 5E to 5G).

FIG. 5.

(A and B) Confocal photomicrographs showing double immunostaining of a rVPAC₁ receptor positive pial arteriole (green) seemingly innervated by vasoactive intestinal polypeptide (VIP) immunoreactive processes (red) originating from intracortical perikarya. (B) High magnification of the framed area in A that demonstrates close apposition of the VIP containing nerve fibers to the rVPAC₁ receptor on the pial vessel (arrows). (C, D, E, and F) Double immunostaining showing the distribution of VIP (green in C and E) and pituitary adenylate cyclase activating polypeptide (PACAP) (red in D and F) immunoreactive cell bodies and processes in free floating sections of the rat parietal cortex. In the low power magnification (C and D) the difference in distribution pattern of the two peptides is seen. E, F, and G are confocal photomicrographs. (G) Fluorescence photomicrograph of the same section as E and F visualizing VIP in green, PACAP in red, and coexpression of the two peptides in yellow-orange. Note the colocalization of the two peptides is limited to a single cell body (arrow). Scale bars = 25 μm (A, E, F, and G), 5 μm (B), and 200 μm (C and D). *indicates vascular lumen.

DISCUSSION

The current study demonstrates for the first time the localization of rVPAC₁ receptor immunoreactivity in rat cerebral arteries and arterioles using a polyclonal antibody generated against the N-terminal receptor fusion protein. By Western blot analysis of the expressed receptor N-terminal, the antiserum was found to detect the protein in amounts down to 3 pmol. Through immunoblotting it was also shown that the antibody detected the full length receptor protein expressed in cell lines recognized as a band of the expected 45 kDa, whereas no signal was observed with the related VIP-PACAP receptors. Further support for the specificity of the antibody comes from the immunocytochemical staining of transfected cells and the complete disappearance of rVPAC₁ immunoreactivity in cerebral arteries and arterioles after preincubation of the antibody with the rVPAC₁ receptor N-terminal fusion protein.

rVPAC₁ receptor immunoreactivity was identified in the wall of cortical pial arteries and arterioles and in the major arteries at the base of the brain and their branches. The density and intensity of the receptor immunostaining seemed to be the same in differently sized cerebral vessels. The immunohistochemical receptor localization accords well with a previous autoradiographic study in rat (Amenta et al., 1991). However, in the current study quantitative determination of ¹²⁵I-VIP binding site density in the different arteries of the circles of Willis demonstrated a higher accumulation of silver grains in the anterior than in the posterior cerebral arteries; furthermore, grain density was higher in pial arteries than in circle of Willis arteries. Autoradiographic studies, however, could reflect binding of radiolabeled VIP to other receptor subtypes than the rVPAC₁ receptor.

The rVPAC₁ receptor was found to be localized exclusively to the surface cell membrane of the circularly oriented vascular smooth muscle cells and no immunostaining of the endothelial cell layer was observed. This observation accords well with the unaffected ability of VIP to relax cerebral arteries after removal of endothelial cells (Lee et al., 1984). Through double immunostaining a rich supply of VIP immunoreactive nerve fibers were found to have intimate contact to the receptor protein indicating innervation. Vasoactive intestinal polypeptide immunoreactivity has previously been detected in large granular vesicles in nerve terminals and axons in cerebral arterial walls (Lee et al., 1984). Vasoactive intestinal polypeptide released from these autonomic nerve fibers could directly regulate tonus in agreement with the general view on the role of VIP in cerebral vascular control (Edvinsson et al., 1988). Although VIP, agreeing with previous studies, was identified in nerve fibers and dense nerve plexuses of cerebral arterioles and the major arteries and their brances, the number of the longitudinally oriented PACAP containing nerve fibers was much more limited and their density decreased as the vessels became smaller. Some PACAP immunoreactive nerve fibers were, like VIP, found in close apposition to the rVPAC₁ receptor protein indicating that this peptide also could have a functional role in the regulation of cerebral blood flow. In accordance with the different distribution pattern of the VIP and PACAP containing nerves, double immunostaining disclosed that the two peptides were localized in separate populations of nerve fibers. This finding disagrees with a previously reported study on cat cerebral blood vessels in which PACAP immunoreactivity was demonstrated in a subpopulation of VIP containing nerve fibers (Uddman et al., 1993). Whether this discrepancy reflects a species difference between rat and cat or could merely be explained by different specificities of the antibodies used remains to be clarified. Although the VIP containing fibers to the cerebral circulation primarily originate in the sphenopalatine and otic ganglia and in microganglia located along the carotic artery (Suzuki et al., 1988), the source of the PACAP containing cerebral vascular fibers remains to be clarified. Pituitary adenylate cyclase activating polypeptide containing cell bodies have been demonstrated in a number of auto nomic ganglia including the two parasympathetic ganglia, the sphenopalatine, and the otic ganglia (Sundler et al., 1996). Pituitary adenylate cyclase activating polypeptide is also present in cell bodies in the trigeminal ganglion, which are sensitive to capsaicin (Moller et al., 1993), suggesting the possibility that PACAP fibers could represent sensory cerebral vascular nerves.

In the current study we also found that intracortical VIP containing processes penetrated the cortical surface and appeared to innervate the rVPAC₁ receptor protein on pial arterioles. This innervationlike pattern was also observed after the arterioles had entered the superficial layers of the brain. It is likely that increased activity of the cortical VIP containing neurons causes release, which in turn triggers vasodilation serving as a coupling agent between nervous activity and blood flow (Ingvar and Lassen, 1975). However, processes from the PACAP immunoreactive cortical cells bodies did not project to the surface of the brain and, in general, the distribution pattern of the PACAP and VIP immunoreactive cell bodies in the parietal cortex was quite different suggesting different functional roles.

In conclusion, we have provided direct morphologic evidence that the rVPAC₁ receptor is a postsynaptic receptor uniformly present in smooth muscle cells in cerebral arteries and arterioles. The receptor protein was found to be innervated by VIP containing nerve fibers, and to a lesser extent PACAP containing nerve fibers, providing the neuroanatomic substrate for the role of VIP and maybe PACAP in the regulation of cerebral blood flow circulation.

Footnotes

Acknowledgments

The authors gratefully acknowledge the skillful technical assistance of Anita Hansen and Yvonne Søndergaard. The authors thank Dr. Niels Vrang at the Panum Institute for his help in preparing the confocal photomicrographs. The VIP antibody (code MoAb #55) was donated by Dr. J.H. Walsh, CURE, Gastroenteric Biology Center, Department of Medicine, UCLA, Los Angeles, CA, U.S.A.

References

Amenta

Cavalotti

De Michele

De Vincentis

Rossodivita

(1991) Vasoactive intestinal polypeptide receptors in rat cerebral vessels: an autoradiographic study. J Auton Pharmacol 11:285–293

Anzai

Suzuki

Takayasu

Kajita

Mori

Seki

Saito

Shibuya

(1995) Vasorelaxant effect of PACAP-27 on canine cerebral arteries and rat intracerebral arterioles. Eur J Pharmacol 285:173–179

Arimura

(1998) Perspectives on pituitary adenylate cyclase activating polypeptide (PACAP) in the neuroendocrine, endocrine and nervous system. Jpn J Physiol 48:301–331

Dacey

Bassett

Takayasu

(1988) Vasomotor responses of intracerebral arterioles to vasoactive intestinal peptide, substance P, neuropeptide Y, and bradykinin. J Cereb Blood Flow Metab 8:254–261

Edvinsson

Fahrenkrug

Hanko

Owman

(1980) VIP (vasoactive intestinal polypeptide)-containing nerves of intracranial arteries in mammals. Cell Tissue Res 208:125–142

Edvinsson

McCulloch

(1985) Distribution and vasomotor effects of peptide HI (PHI) in feline cerebral blood vessels in vitro and in situ. Regul Pept 10:345–356

Edvinsson

McCulloch

Kelly

PAT

Tuor

(1988) Role of vasoactive intestinal peptide and peptide histidine isoleucine in the cerebral circulation. In: Vasoactive intestinal peptide and related peptides. ( Said

Mutt

, eds), Ann N Y Acad Sci 527:378–392

Fahrenkrug

(1993) Transmitter role of vasoactive intestinal peptide. Pharmacol Toxicol 72:354–363

Fahrenkrug

Buhl

Hannibal

(1995) Prepro-PACAP derived peptides of occur in VIP producing tumors and co-exist with VIP. Regul Pept 58:89–98

10.

Fahrenkrug

Hannibal

(1998) Pituitary adenylate cyclase activating polypeptide immunoreactivity in capsaicin-sensitive nerve fibers supplying the rat urinary tract. Neuroscience 83:1261–1272

11.

Gibbins

Brayden

Bevan

(1984) Perivascular nerves with immunoreactivity to vasoactive intestinal polypeptide in cephalic arteries of the cat: distribution, possible origins and functional implications. Neuroscience 13:1327–1346

12.

Hannibal

Mikkelsen

Clausen

Holst

Wulff

Fahrenkrug

(1995) Gene expression of pituitary adenylate cyclase activating polypeptide (PACAP) in the rat hypothalamus. Regul Pept 55:133–148

13.

Hara

Hamill

Jacobowitz

(1985) Origin of cholinergic nerves to the rat major cerebral arteries: coexistence with vasoactive intestinal polypeptide. Brain Res Bull 14:179–188

14.

Hara

Jansen

Ekman

Hamel

MacKenzie

Uddman

Edvinsson

(1989) Acetylcholine and vasoactive intestinal peptide in cerebral blood vessels: effect of extirpation of the sphenopalataine ganglion. J Cereb Blood Flow Metab 9:204–211

15.

Harmar

Arimura

Gozes

Journot

Laburthe

Pisegna

Rawlings

Robberecht

Said

Sreedharan

Wank

Waschek

(1998) International Union of Pharmacology. XVIII. Nomenclature of receptors for vasoactive intestinal peptide and pituitary adenylate cyclase-activating polypeptide. Pharmacol Rev 50:265–270

16.

Ingvar

Lassen

(1975) Brain work. Copenhagen and New York, NY: Munksgaard and Academic Press

17.

Jansen-Olesen

Goadsby

Uddman

Edvinsson

(1994) Vasoactive intestinal peptide (VIP)-like peptides in the cerebral circulation of the cat. J Autonom Nerv Syst 49:S97–S103

18.

Larsson

L-I

Edvinsson

Fahrenkrug

Håkanson

Owman

Schaffalitzky de Muckadell

Sundler

(1976) Immunohistochemical localization of a vasodilatory polypeptide (VIP) in cerebrovascular nerves. Brain Res 113:400–404

19.

Lee

TJ-F

Saito

Brezin

(1984) Vasoactive intestinal polypeptide-like substance: the potential transmitter for cerebral vasodilatation. Science 224:898–900

20.

Lorén

Emson

Fahrenkrug

Björklund

Alumets

Håkanson

Sundler

(1979) Distribution of vasoactive intestinal polypeptide in the rat and mouse brain. Neuroscience 4:1953–1976

21.

Matsuyama

Shiosaka

Matsumoto

Yoneda

Kimura

Abe

Hayakawa

Inoue

Tohyama

(1983) Overall distribution of vasoactive intestinal polypeptide-containing nerves on the wall of cerebral arteries: an immunohistochemical study using whole-mounts. Neuroscience 10:89–96

22.

McCulloch

Edvinsson

(1980) Cerebral circulatory and metabolic effects of vasoactive intestinal polypeptide. Am J Physiol 238:H449–H456

23.

Miao

J-PF

Lee

TJ-F

(1990) Cholinergic and VIPergic innervation in cerebral arteries: sequential double-labeling immunohistochemical study. J Cereb Blood Flow Metab 10:32–37

24.

Moller

Zhang

Håkanson

Luts

Sjölund

Uddman

Sundler

(1993) Pituitary adenylate cyclase activating peptide is a sensory neuropeptide—immunocytochemical and immunochemical evidence. Neuroscience 57:725–732

25.

Morrison

Magistretti

Benoit

Bloom

(1984) The distribution and morphological characteristics of the intracortical VIP-positive cell: an immunohistochemical analysis. Brain Res 292:269–282

26.

Schägger

von Jagow

(1987) Tricine-sodium dodecyl sulfatepolyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem 166:368–379

27.

Seki

Suzuki

Baskaya

Kano

Saito

Takayasu

Shibuya

Sugita

(1995) The effects of pituitary adenylate cyclase-activating polypeptide on cerebral arteries and vertebral artery blood flow in anesthetized dogs. Eur J Pharmacol 275:259–266

28.

Sundler

Ekblad

Hannibal

Moller

Zhang

Y-Z

Mulder

Elsås

Grunditz

Danielsen

Fahrenkrug

Uddman

(1996) Pituitary adenylate cyclase-activating peptide in sensory and autonomic ganglia: localization and regulation. In: VIP, PACAP, and related peptides. ( Arimura

Said

, eds), Ann N Y Acad Sci 805:410–428

29.

Suzuki

Hardebo

Owman

(1988) Origins and pathways of cerebrovascular vasoactive intestinal polypeptide-positive nerves in rat. J Cereb Blood Flow Metab 8:697–712

30.

Uddman

Goadsby

Jansen

Edvinsson

(1993) PACAP, a VIP-like peptide: immunohistochemical localization and effect upon cat pial arteries and cerebral blood flow. J Cereb Blood Flow Metab 13:291–297

31.

Uemura

Sugimoto

Kikuchi

Mizuno

(1988) Possible origins of cerebrovascular nerve fibers showing vasoactive intestinal polypeptide-like immunoreactivity: an immunohistochemical study in the dog. Brain Res 448:98–105

32.

Wei

Kontos

Said

(1980) Mechanism of action of vasoactive intestinal polypeptide on cerebral arterioles. Am J Physiol 239:H765–H768

33.

Wong

Sternini

Lloyd

De Giorgio

Walsh

(1996) Monoclonal antibody to VIP: production, characterization, immunoneutralizing activity, and usefulness in cytochemical staining. Hybridoma 15:133–139

34.

Zhang

E-T

Mikkelsen

Fahrenkrug

Møller

Kronborg

Lauritzen

(1991) Prepro-vasoactive intestinal polypeptide-derived peptide sequences in cerebral blood vessels of rats: On the functional anatomy of metabolic autoregulation. J Cereb Blood Flow Metab 11:932–938