Immunolocalization of nitric oxide synthase isoforms in human archival and rat tissues, and cultured cells

Nitric oxide (NO) exerts important physiological and pathological roles in humans. The study of NO requires the immunolocalization of its synthesizing enzymes, neuronal, endothelial and inducible NO synthases (NOS). NOS are labile to formalin-fixation and paraffin-embedding, which are used to prepare human archival tissues. This lability has made NOS immunohistochemical studies difficult, and a detailed protocol is not yet available. We describe here a protocol for the immunolocalization of NOS isoforms in human archival cerebellum and non-nervous tissues, and in rat tissues and cultured cells. Neuronal NOS antigenicity in human archival and rat nervous tissue sections was microwave-retrieved in 50 mM Tris–HCl buffer, pH 9.5, for 20 min at 900 W. Neuronal NOS was expressed in stellate, basket, Purkinje and granule cells in human and rat cerebellum. Archival and frozen human cerebellar sections showed the same neuronal NOS staining pattern. Archival cerebellar sections not subjected to antigen retrieval stained weakly. Antigenicity of inducible NOS in human lung was best retrieved in 10 mM sodium citrate buffer, pH 6.0, for 15 min at 900 W. Inflammatory cells in a human lung tuberculoma were strongly stained by anti-inducible NOS antibody. Anti-endothelial NOS strongly stained kidney glomeruli. Cultured PC12 cells were strongly stained by anti-neuronal NOS without antigen retrieving. The present immunohistochemistry protocol is easy to perform, timeless, and suitable for the localization of NOS isoforms in nervous and non-nervous tissues, in human archival and rat tissues. It has been extensively used in our laboratory, and is also appropriate for other antigens. Keywords Nitric oxide synthases Immunohistochemistry Cerebellum Lung Human and rat nervous system Cell culture 1 Introduction The gaseous signaling molecule nitric oxide (NO) is produced from l -arginine and oxygen by the NO synthases (NOS). NO is involved in several functions, including neurotransmission, regulation of blood vessel tone and immune response. The NOS isoforms, endothelial, neuronal, inducible and mitochondrial, are known to regulate different functions ( Elfering et al., 2002; Garthwaite, 2008 ). In the nervous system, NO synthesis is associated with synaptic plasticity, control of neurosecretion, appetite, sleep, body temperature, and regulation of the non-adrenergic, non-cholinergic relaxation of smooth muscle cells ( Moncada and Higgs, 2006; Garthwaite, 2008 ). On the other hand, excessive amounts of NO and/or the formation of reactive nitrogen species can lead to cellular damage (cf. Calabrese et al., 2007; Hara and Snyder, 2007 ). NO and reactive nitrogen species have been implicated in the pathogenesis of several neurodegenerative disorders (cf. Guix et al., 2005; Pacher et al., 2007 ). The study of NO synthases in normal and pathologic human nervous system usually relies on the immunodetection of the NOS isoforms in formalin-fixed, paraffin-embedded tissue obtained at autopsy. Immunohistochemistry is a powerful tool to detect neurotransmitters and their metabolizing enzymes in different brain structures with high spatial resolution. However, its application to the immunolocalization of NOS in archival nervous tissue is usually a difficult task, as indicated by the small number of reports on this subject. To address this issue, a protocol has been developed that is suitable for the immunohistochemistry of NO synthases in formalin-fixed, paraffin embedded, archival human nervous and non-nervous tissues, as well as for rat nervous tissue, and cultured nervous cells. 2 Results The human cerebellar cortex was strongly stained by the anti-nNOS (TL) antibody after antigen retrieval in 50 mM Tris–HCl buffer, pH 9.5, using a microwave oven at 900 W potency for 20 min ( Fig. 1 A). The morphology of basket and stellate cell processes could be better observed after antigen retrieval in 50 mM Tris–HCl buffer, pH 9.5 than in 10 mM sodium citrate buffer, pH 6.0 (data not shown). A time-course experiment (0–45 min treatment) carried out to determine the extent of microwave treatment, at 900 W potency, for optimal section staining indicated 20, 15 and 20 min as optimum incubation times for nervous, lung and kidney tissues, respectively. A control section of a cerebellar cortex that was formalin-fixed, but cryo-protected and frozen rather than paraffin-embedded, and was not subjected to antigen retrieval ( Fig. 1 B), showed neuronal NOS staining by anti-nNOS (TL) antibody in the same structures that were detected in other cerebellar sections after antigen retrieval in 50 mM Tris–HCl buffer, pH 9.5. In contrast, staining of cell bodies and processes in a cerebellar section adjacent to that in Fig. 1 A, but without using antigen retrieval, was barely detectable using anti-nNOS (TL) ( Fig. 1 C). The substitution of anti-nNOS (TL) or anti-nNOS (FM) for a non-immune serum has led to no staining (not shown). Taken together, the above results indicate that the procedure proposed here is specific. Immunostaining after antigen retrieval was maximal in human cerebellar tissue fixed for up to 2 months ( Table 1 ), using anti-nNOS (TL) antibody, and either 50 mM Tris–HCl buffer, pH 9.5 or 10 mM sodium citrate buffer, pH 6.0. A moderate staining was achieved when the tissue was fixed for 2–8 months. Fixation of a cerebellar cortex for 20 months or more resulted in an essentially complete loss of neuronal NOS IR (not shown). The addition of 2–5% (w/v) urea to the 50 mM Tris–HCl buffer, pH 9.5, incubating solution increased immunostaining, but the morphology of the brain tissue was moderately to severely damaged, and sections frequently detached from slides during antigen retrieval. Strong staining of human cerebral cortex using anti-nNOS (TL) (not shown) and rat cerebellum using anti-nNOS (FM) (granule cells, Fig. 1 D) were also obtained using antigen retrieval in 50 mM Tris–HCl buffer, pH 9.5. Antigen retrieval was not necessary for immunodetecting nNOS in cultured PC12 cells using either confocal fluorescence or transmission light microscopy. An intense staining of cytoplasm and growth cone in PC12 cells was observed ( Fig. 1 E). Inducible NOS was detected in Langhans’ giant cells, among other inflammatory cells, of a human lung tuberculosis granuloma ( Fig. 1 F), after antigen retrieval in 10 mM sodium citrate buffer, pH 6.0, for 15 min at 900 W. Endothelial NOS was strongly expressed in the rat kidney glomerulus, and in afferent and efferent glomerular arterioles ( Fig. 1 G). The present results show that the combined use of a microwave-based antigen retrieval method and the immunostaining procedure described here can specifically detect and enhance the immunostaining of all three NOS isoforms in several formaldehyde-fixed, paraffin-embedded human and rat tissues. 3 Discussion We report here an immunohistochemical protocol that can be used to localize nitric oxide synthase isoforms in human archival nervous, lung and blood tissues, rat nervous and kidney tissues, isolated rat blood cells, and cultured nerve cells. The poor reactivity of nitric oxide synthases in tissues properly fixed in formaldehyde and paraffin-embedded has delayed immunohistochemical studies of these enzymes, mainly in human archival specimens. Using affinity-purified, specific and extensively characterized anti-NOS antibodies against neuronal, endothelial and inducible isoforms of NOS that were commercially available (Transduction Labs) or prepared in our laboratory ( Schmidt et al., 1992 ), we have shown that microwave-based antigen retrieval treatment of formalin-fixed, paraffin-embedded human tissue sections permitted the specific immunolocalization of neuronal NOS in the human hippocampus ( Leite et al., 2002 ), cerebral (Zanella and Martins, unpublished) and cerebellar cortices (the present work). The localization of NOS isoforms, found here in both human and rat tissues, is consistent with findings using both frozen (the present paper) and fixed nervous tissue with anti-NOS antibodies ( Bredt et al., 1990; Springall et al., 1992; Schmidt et al., 1992; Egberongbe et al., 1994 ), and represents a significant methodological improvement for human nervous system research. The use of an antigen retrieval protocol coupled with an accurate immunohistochemical assay for nitric oxide synthases (the present paper), combined with the very large collection of archival human tissues available, permits a sound immunohistochemical approach to the study of NOS isoforms in formaldehyde-fixed, paraffin-embedded tissues. 3.1 Trouble-shooting The handling of tissues during processing for paraffin-embedding, antigen retrieval and immunostaining can affect NOS detection. Fixative type, fixation duration and temperature can affect the intensity of NOS immunoreactivity in the tissue. Indeed, fixation of human cerebellum longer than eight months can substantially decrease nNOS immunoreactivity (the present report). This is probably due to an increased cross-linking of tissue components that can mask the target epitope from antibody detection. Human and rat neuronal, endothelial and inducible NOS isoforms in nervous system, lung and kidney were effectively unmasked using antigen retrieval in 50 mM Tris–HCl buffer, pH 9.5, or 10 mM sodium citrate buffer, pH 6.0 incubation solutions (the present study). It has to be noted that the temperature of the antigen retrieval solution or microwave potency, and the heating time have to be established for each tissue under study, as observed for rat kidney (15 min; Soares et al., 1999 ). The retrieval of a given epitope can require different antigen retrieval solutions (see Shi et al., 2001 for a review), as was the case for the inducible NOS isoform in lung tuberculosis granulomas (the present report) and for purified rat eosinophils ( Zanardo et al., 1997; Ferreira et al., 2002, 2004 ), which required 10 mM citrate buffer, pH 6.0. Microwaving tissue in an aggressive incubation solution, and/or excessive tissue handling, can damage tissue morphology and detach tissue sections from the slides. For example, the addition of 2-5% (w/v) urea to antigen retrieval solutions or the use of more than 10 mM HCl antigen retrieval solution, can severely affect tissue morphology (the present study). The use of a tissue adhesive is necessary to avoid detachment of the tissue from the slides during the antigen retrieval process. Gelatin gave consistently good results as a tissue section adhesive ( Martins et al., 1999 ). The concentration of endogenous biotin can vary in different physiological or physiopathological states of the same tissue. Differences in endogenous biotin concentration can introduce bias when comparing tissues in such different states using biotinylated second antibodies and the ABC technique, because this detection technique is based on avidin–biotin interaction. Therefore, the use of an avidin–biotin blocking reagent is useful to circumvent this bias ( Wood and Warnke, 1981 ). By paying attention to the above points, an accurate immunostaining of NO synthases in several mammalian tissues can be obtained ( Zanardo et al., 1997; Soares et al., 1999; Mendes et al., 2001; Ferreira et al., 2002, 2004; Leite et al., 2002 ). A similar protocol has also been successfully applied to the detection of other antigens in the nervous system and in other tissues of humans and rats [lectin KM + in rat skin ( Ganiko et al., 1998 ), lung ( Ganiko et al., 2005 ) and cerebellum ( Teixeira et al., 2004 ); drebrin isoforms in human cortex (Dombroski and Martins, unpublished); galectin-3 in biopsies of cortical tubers (Donatti and Martins, unpublished); HIV antigens in macrophages of sympathetic ganglia from AIDS patients ( Chimelli and Martins, 2002 ); angiotensin receptors types 1 and 2 in rat CNS ( Pelegrini-da-Silva et al., 2005 )]. 4 Experimental procedure 4.1 Tissues and cultured cells Human nervous tissue was obtained from autopsies performed at the Department of Pathology, Faculty of Medicine of Ribeirão Preto, University of São Paulo, according to protocols approved by the Ethics Committee of the Faculty of Medicine of Ribeirão Preto, University of São Paulo. Brains from twenty-one patients that did not show any evidence of disease on the basis of systematic neuropathological examination, and the lungs from two patients, were studied here. The patients’ age, causa mortis, and interval between dead and fixation are given in Table 1 . Ten-day-old male Wistar rats were obtained from the Animal House of the Faculty of Medicine of Ribeirão Preto University of São Paulo. Rats received food and water ad libitum, and were kept under controlled temperature (22 °C) and illumination (12/12 h light/dark cycle). Animals were handled according to the Ethics Committee of the Faculty of Medicine of Ribeirão Preto, University of São Paulo. PC12 cells were obtained from American Type Culture Collection (ATCC) and were grown as described by Ho and Raw (1992) . 4.2 Chemicals and reagents A first polyclonal rabbit anti-neuronal NOS antibody (generated in the laboratory of one of us, FM) was raised in rabbits against purified rat cerebellar NOS. The resulting immune sera containing IgG was purified by Protein A-Sepharose chromatography ( Schmidt et al., 1992 ), and will be further referred as anti-nNOS antibody (FM). A second polyclonal anti-neuronal NOS antibody (lot number 13, cat. No. 31030, 250 μg/ml; Transduction Laboratories, Lexington, KY) was generated in rabbits against the C-terminal 22 kD amino acid sequence 1095–1289 of human nNOS as immunogen. It was purified from immune serum by negative adsorption and affinity chromatography (Technical Data Sheet by Transduction Laboratories), and will be further referred to as anti-nNOS (TL). Anti-nNOS antibody (TL) recognized only neurons in human brain ( Coers et al., 1998; Ohyu and Takashima, 1998; Cobbs et al., 1995 ) and hippocampus ( Leite et al., 2002 ). A single band at about 155 kDa was detected on Western blots of extracts from mouse brain ( Cobbs et al., 1995 ), human frontal ( Ohyu and Takashima, 1998 ) and temporal lobes ( Leite et al., 2002 ), and of a rat pituitary lysate (Transduction Laboratories) used as an immunoblot control ( Coers et al., 1998; Leite et al., 2002 ). A mouse monoclonal anti-endothelial NOS antibody (clone 3, IgG1 isotype, lot number 5, catalog no. N30020, 250 μg/ml; Transduction Laboratories) was generated against the region 1030–1209 of human endothelial NOS. A single band at 140 kDa was detected in a human endothelial lysate derived from an aortic endothelium cell line (Transduction Laboratories) used as a positive control ( Coers et al., 1998 ), and in rat kidney homogenates ( Han et al., 2005 ). It stained glomeruli, endothelial cells of glomerular capillaries and blood vessels in the kidney ( Soares et al., 1999; Han et al., 2005 ). A mouse monoclonal anti-inducible NOS antibody (clone 6, IgG2a isotype, lot number 16, catalog no. N32020, 250 μg/ml;Transduction Laboratories) was generated against the region 961–1144 of inducible NOS protein. A single band at 130 kDa was detected in a mouse macrophage lysate prepared from the murine RAW 264.7 cell line (Transduction Laboratories), used as an immunoblot control ( Coers et al., 1998; Zhao et al., 1998 ), and in homogenates of B-cell chronic lymphocytic leukemia ( Zhao et al., 1998 ). Other materials used were: biotinylated anti-rabbit and anti-mouse IgG antibodies (Dako); Alexa 488-conjugated anti-rabbit IgG antibody (Molecular Probes); avidin biotin blocking kit (Vector Laboratories); Vectastain Elite ABC kit (Vector Laboratories); 3,3′-diaminebenzidine hydrochloride (Pierce); hydrogen peroxide (Merck); Hematoxylin (Fisher); adhesive (Amazonas); formalin (Merck); paraformaldehyde (Sigma); paraffin, melting point 56–58 °C (Histosec, Merck); glass slides (Thomas); cover glass (Corning); gelatin purified grade, 275 Bloom (Fisher Scientific); chromium potassium sulfate (Fisher Scientific); Permount (Fisher Scientific); defatted dry milk (Molico, Nestle); normal donkey serum (obtained from the Animal House, Faculty of Medicine of Ribeirão Preto, University of São Paulo); Triton X-100, Tris base and sodium citrate (Sigma); sodium hydroxide (Merck); ethanol (Merck); xylene (Merck). All tissue culture reagents were from Sigma: bovine serum albumin; Dullbecco's Modified Eagle's Medium, high glucose; poly- l -lysine hydrobromide; Cell Freezing Medium, DMSO; fetal bovine serum; Fibroblast Growth Factor-Basic human; dibutiryl-cAMP; gentamycin solution; streptomycin sulfate salt. MilliQ water was used to prepare all aqueous solutions. All solutions used were freshly prepared, and filtered through a 0.22 μm filter (Millipore). 4.3 Equipment Optical microscope model BX-60 (Olympus); all objectives were UPlanApo, and were: 4×/0.16, 10×/0.40, 20×/0.70, 40×/0.85, 100×/1.35; inverted microscope DM IL (Leica); confocal microscope, Leica model TCS_NT; tissue processor model TP 1020 (Leica); microtome model 2065 (Leica); cryostat model CM 1850 (Leica); Histotemp-3 plate (Leica); Duo-Vac oven model 3620 (Labline); microwave oven model RB-4A33 (Sharp); SteriCult CO2 Incubator model 3862 (ThermoForma); laminar flow hood (Veco); vacuum/pressure pump model WP611560 (Millipore); Millipore Sterifil 47 mm Aseptic Vacuum Filter System and Holder (Millipore); four-cartridge MilliQ System (Millipore). 5 Detailed procedure 5.1 Coating slides with gelatin-chromealum Preparation of gelatin subbing solution : Two-and-one half g gelatin powder were added slowly to 500 ml distilled water maintained at 50 °C, while stirring with a magnet. After the solution was cooled to 30 °C, 0.25 g chromium potassium sulfate (chromealum) was added. Chromealum must be nice purple crystals. The solution was filtered through medium coarseness filter paper into a large staining dish avoiding foam formation. Slide coating : Pre-cleaned glass slides (Thomas) were placed in a stainless steel rack (30 slides per rack). Racks were lowered slowly into and out of the subbing solution, taking care to avoid bubble formation. Excess solution was blotted on a paper towel by the etched end while separating slides in the rack, such that no air bubbles or gelatin drops remained on slides. Slides were allowed to dry at 37 °C for 24 h. Coated slides were removed from racks and stored in a clean, dust-free box. 5.2 Tissue preparation Human nervous tissue was obtained at autopsy, and fixed by immersion in 4% (w/v) formalin (Merck) buffered with 0.1 M sodium phosphate buffer, pH 7.4, at room temperature for up to 28 months ( Table 1 ). Rat brains were perfused through the left heart ventricle with 300 ml 4% (w/v) paraformaldehyde buffered with 0.1 M sodium phosphate buffer, pH 7.4, at room temperature and at a hydrostatic pressure equivalent to 100 cm of water. Human and rat tissues were cut in sagittal and coronal planes into 4-mm-thick slices by up to 18 mm × 40 mm. Tissue slices were dehydrated with a series of graded ethanol solutions, cleared with xylene, and embedded in paraffin. Six μm thick sections were cut using a rotary microtome, mounted on gelatin–chromium potassium sulfate coated glass slides, and oven-dried at 37–40 °C for 4 h. On the day of the immunohistochemical staining, the sections were dewaxed using xylene, and hydrated through graded alcohols and water. PC12 cells were cultured in Dullbecco's Modified Eagle's Medium (DME) supplemented with 10% (w/w) fetal bovine serum, in a humid atmosphere containing 5% CO 2 and 95% air, at 37 °C. Cells were plated on a poly- d -lysine (10 mg/ml)-treated glass coverslips (1.0 cm diameter) at 4.0 × 10 4 cells/well in a 24 wells plate. Cells were grown in the presence of 1 mM dibutiryl-cAMP with 5 ng/ml FGF-2 for 4 days to enhance process outgrowth. PC12 cells were fixed by immersion in 4% (w/v) paraformaldehyde in 0.1 M sodium phosphate buffer, pH 7.4, for 15 min at room temperature, washed with PBS, and then used for immunohistochemistry without any further treatment. Slides containing PC12 cells were not allowed to dry-out before immunohistochemistry. 5.3 Immunohistochemistry of nNOS in routine formalin-fixed, paraffin-embedded autopsy tissues All operations were carried out at room temperature, except for antigen retrieval. Step 1. Inhibition of endogenous peroxidase activity Dewaxed and hydrated tissue sections were incubated with 4.5 vol.% H 2 O 2 in 50 mM sodium phosphate buffer, pH 7.4, containing 0.9% (w/v) sodium chloride (phosphate buffered saline, PBS) for 15 min. Sections were washed 3 × 2 min using MilliQ water. After dewaxing, sections were not allowed to dry. Step 2. Antigen retrieval The retrieval of nNOS-like immunoreactivity from formalin-fixed, paraffin embedded specimens was carried out using a microwave technique ( Shi et al., 1991, 2001; Martins et al., 1999 ). Neuronal NOS IR was enhanced by treating the tissue in a Sharp model RB-4A33 microwave oven, with an internal capacity of 34 l, and a working frequency of 2.45 GHz, at 900 W nominal power. Slides were placed in plastic Coplin jars (70 ml total capacity) containing 50 mM Tris–HCl buffer, pH 9.5. Since the number and position of the Coplin jars placed in the microwave oven might influence temperature, three jars containing the same fluid volume were located symmetrically on the rotary oven plate, in order to obtain reproducible inter-assay results. The jars were covered with a PVC film that was hold in place by a thin rubber band. After boiling for 10 min in the microwave oven, the buffer level was verified and completed if necessary, and the sections were heated for a further 10 min. The Coplin jars were removed and allowed to equilibrate to room temperature for about 30 min. Controls for the retrieval of nNOS IR were carried out by treating sections with the same solutions and for the same time, except that microwave boiling was omitted. Sections were then rinsed in distilled water and immunostained. Step 3. Delimitation of tissue sections with craftsman glue A circle was inscribed on the slide around the tissue section using a diamond pencil. Glue was placed onto this circle so as to form an elevated line, which prevented the incubation solutions to flow out of the section, and allowed the use of smaller amounts of antibodies. This procedure must be carried out quickly, and care must be taken to keep sections wet. Step 4. Quenching of unreacted aldehyde groups Microwave-treated sections were incubated with 50 mM glycine buffered at pH 7.4 with 0.1 M Tris base, for 30 min, to quench unreacted aldehyde groups. Sections were then dipped twice in 20 mM sodium phosphate buffer, pH 7.4, containing 0.45 M NaCl and 0.3% (w/v) Triton X-100 (Triton buffer). Step 5. Blockade of non-specific interactions Sections were incubated in Triton buffer containing 5% (w/v) defatted dry milk, and 15% (v/v) normal donkey serum (i.e., a non-immune serum, preferably but not necessarily from the same animal species from which the second antibody was obtained) (blocking buffer), for 4 h. Blocking buffer was centrifuged at 10,000 × g for 10 min before use, to precipitate particulate matter. Sections were not washed before step 6. Step 6. Incubation with anti-nNOS primary antibodies Excess blocking buffer was carefully blotted away. Sections were incubated overnight with a rabbit anti-nNOS antibody, either prepared in the laboratory of one of us ( Schmidt et al., 1992 ) or from Transduction Laboratories, diluted 1:100 (v/v) or 1:25 (v/v) in blocking buffer, respectively. Step 7. Washing Sections were washed in Triton buffer, 5 × 10 min. Step 8. Blockade of endogenous biotin Endogenous biotin was blocked ( Wood and Warnke, 1981 ) using a biotin blocking system (Vector Laboratories), according to the manufacturer's instructions. Briefly, sections were: (a) incubated with an avidin solution for 15 min; (b) washed with Triton buffer, 3 × 5 min; (c) incubated with a biotin solution for 15 min; (d) washed with Triton buffer, 3 × 5 min. Step 9. Binding of the second antibody Tissue sections were incubated for 1 h with biotinylated swine anti-rabbit IgG (Dako, code number E0353) diluted 1:100 (v/v) in blocking buffer. Step 10. Washing Sections were washed in Triton buffer, 5 × 10 min. Step 11. Detection of second antibody The biotinylated second antibody was detected using the ABC technique ( Heitzmann and Richards, 1974 ) and the Elite ABC kit, according to Vector Laboratories’ instructions. Briefly, 10 μl reagent A (avidin) and 10 μl reagent B (biotinylated horseradish peroxidase) were added to 500 μl PBS, and the mixture was incubated for 30 min before its addition to tissue sections. The avidin–biotinylated peroxidase complex thus formed was incubated with tissue sections for 1 h. Step 12. Washing Initially, with Triton buffer, 2 × 5 min, followed by 3 × 5 min washes with 50 mM Tris–HCl buffer, pH 7.6. Step 13. Peroxidase reaction-color development Peroxidase was detected using 0.015 vol.% hydrogen peroxide (Merck) and 0.66 mg/ml 3,3′-diaminebenzidine tetrahydrochloride as the chromogen, for up to 12 min. The reaction was stopped with water. Tissue sections were dehydrated in ethanol, cleared in xylene and coverslipped with Permount. The same procedure was used for the detection of endothelial and inducible NOS isoforms, except that their dilutions were 1:100 and 1:50 (v/v), respectively, and inducible NOS retrieval has been carried out in 10 mM sodium citrate buffer, pH 6.0. Several neural and non-neural tissues were used as positive controls, including rat and human cerebellum (granule cells and molecular layer interneurons for neuronal NOS, blood vessels for the endothelial NOS), and lung specimens from tuberculosis patients (for the inducible NOS expression). Primary antibody substituted for a non-immune serum from the same animal species as the secondary antibody (method specificity control) ( Pool et al., 1983 ). A tissue control was obtained omitting both primary and secondary antibodies, in order to check for the level of background endogenous peroxidase activity. The specificity of the antibodies used here has been extensively documented in the literature and in our laboratories (see Section 4 ). 5.4 Time required Immunohistochemical staining takes up to 8 h at the bench: one morning is spent on the first day and another one on the second day, after an overnight incubation in primary antibody. Counterstaining and coversliping takes up to 2 h. These time estimates were based on staining 24 slides at a time. 5.5 Quick procedure A quick immunohistochemical procedure can permit the immunostaining to be carried out in one working day, provided that this quick method has been previously tested for a given tissue. It consists on reducing the incubation times of all steps of the regular protocol above. Washings can be reduced to one-half the time, blocking to 1 h, incubation with anti-NOS and with second antibody to 1 h and to 30 min, respectively. 5.6 Immunohistochemistry of PC12 cells Antigen retrieval is not required for immunostaining cultured cells. These cells can be best processed using the quick procedure above. Acknowledgements We thank Mr. Hildeberto Caldo, Ms. Vani A. Corrêa and Ms. Marcia S.Z. Graeff for skillful technical assistance. Supported by FAPESP, FAEPA, CNPq and CAPES. References Bredt et al., 1990 D.S. Bredt P.M. Hwang S.H. Snyder Localization of nitric oxide synthase indicating a neural role for nitric oxide Nature 347 1990 768 770 Calabrese et al., 2007 V. Calabrese C. Mancuso M. Calvani E. Rizzarelli D.A. Butterfield A.M. Stella Nitric oxide in the central nervous system: neuroprotection versus neurotoxicity Nat Rev Neurosci 8 2007 766 775 Chimelli and Martins, 2002 L. Chimelli A.R. Martins Degenerative and inflammatory lesions in sympathetic ganglia: further morphological evidence for an autonomic neuropathy in AIDS J NeuroAIDS 2 2002 67 82 Cobbs et al., 1995 C.S. Cobbs J.E. Brenman K.D. Aldape D.S. Bredt M.A. Israel Expression of nitric oxide synthase in human central nervous system tumors Cancer Res 55 1995 727 730 Coers et al., 1998 W. Coers W. Timens C. Kempinga P.A. Klok H. Moshage Specificity of antibodies to nitric oxide synthase isoforms in human, guinea pig, rat and mouse tissues J Histochem Cytochem 46 1998 1385 1391 Egberongbe et al., 1994 Y.I. Egberongbe S.M. Gentleman P. Falkai B. Bogerts J.M. Polak G.W. Roberts The distribution of nitric oxide synthase immunoreactivity in the human brain Neuroscience 59 1994 561 578 Elfering et al., 2002 S.L. Elfering T.M. Sarkela C. Giulivi Biochemistry of mitochondrial nitric-oxide synthase J Biol Chem 277 2002 38079 38086 Ferreira et al., 2004 H.H. Ferreira R.A. Costa J.M. Jacheta A.R. Martins M.V. Medeiros M.F. Macedo-Soares Modulation of eosinophil migration from bone marrow to lungs of allergic rats by nitric oxide Biochem Pharmacol 68 2004 631 639 Ferreira et al., 2002 H.H. Ferreira M.L. Lodo A.R. Martins L. Kandratavicius A.F. Salaroli N. Conran Expression of nitric oxide synthases and in vitro migration of eosinophils from allergic rhinitis subjects Eur J Pharmacol 442 2002 155 162 Ganiko et al., 1998 L. Ganiko A.R. Martins E.M. Espreafico M.C. Roque-Barreira Neutrophil haptotaxis induced by the lectin KM+ Glycoconj J 15 1998 527 530 Ganiko et al., 2005 L. Ganiko A.R. Martins E. Freymuller R.A. Mortara M.C. Roque-Barreira Lectin KM+-induced neutrophil haptotaxis involves binding to laminin Biochim Biophys Acta 1721 2005 152 163 Garthwaite, 2008 J. Garthwaite Concepts of neural nitric oxide-mediated transmission Eur J Neurosci 27 2008 2783 2802 Guix et al., 2005 F.X. Guix I. Uribesalgo M. Coma F.J. Munoz The physiology and pathophysiology of nitric oxide in the brain Prog Neurobiol 76 2005 126 152 Han et al., 2005 K.H. Han J.M. Lim W.Y. Kim H. Kim K.M. Madsen J. Kim Expression of endothelial nitric oxide synthase in developing rat kidney Am J Physiol Renal Physiol 288 2005 F694 F702 Hara and Snyder, 2007 M.R. Hara S.H. Snyder Cell signaling and neuronal death Annu Rev Pharmacol Toxicol 47 2007 117 141 Heitzmann and Richards, 1974 H. Heitzmann F.M. Richards Use of the avidin–biotin complex for specific staining of biological membranes in electron microscopy Proc Natl Acad Sci USA 71 1974 3537 3541 Ho and Raw, 1992 P.L. Ho I. Raw Cyclic AMP potentiates bFGF-induced neurite outgrowth in PC12 cells J Cell Physiol 150 1992 647 656 Leite et al., 2002 J.P. Leite L. Chimelli V.C. Terra-Bustamante E.T. Costa J.A. Assirati G. De Nucci Loss and sprouting of nitric oxide synthase neurons in the human epileptic hippocampus Epilepsia 43 Suppl. 5 2002 235 242 Martins et al., 1999 A.R. Martins M.M. Dias T.M. Vasconcelos H. Caldo M.C. Costa L. Chimelli Microwave-stimulated recovery of myosin-V immunoreactivity from formalin-fixed, paraffin-embedded human CNS J Neurosci Methods 92 1999 25 29 Mendes et al., 2001 R.V. Mendes A.R. Martins G. De Nucci F. Murad F.A. Soares Expression of nitric oxide synthase isoforms and nitrotyrosine immunoreactivity by B-cell non-Hodgkin's lymphomas and multiple myeloma Histopathology 39 2001 172 178 Moncada and Higgs, 2006 S. Moncada E.A. Higgs The discovery of nitric oxide and its role in vascular biology Br J Pharmacol 147 Suppl. 1 2006 S193 S201 Ohyu and Takashima, 1998 J. Ohyu S. Takashima Developmental characteristics of neuronal nitric oxide synthase (nNOS) immunoreactive neurons in fetal to adolescent human brains Brain Res 110 1998 193 202 Pacher et al., 2007 P. Pacher J.S. Beckman L. Liaudet Nitric oxide and peroxynitrite in health and disease Physiol Rev 87 2007 315 424 Pelegrini-da-Silva et al., 2005 A. Pelegrini-da-Silva A.R. Martins W.A. Prado A new role for the renin-angiotensin system in the rat periaqueductal gray matter: angiotensin receptor-mediated modulation of nociception Neuroscience 132 2005 453 463 Pool et al., 1983 C.H.R.W. Pool R.M. Buijs D.F. Swaab G.J. Boer F.W. Van Leeuwen On the way to a specific immunocytochemical localization A.C. Cuello Immunohistochemistry. IBRO hoandbook series: methods in the neurosciences vol. 3 1983 John Wiley & Sons New York 1 46 Schmidt et al., 1992 H.H. Schmidt G.D. Gagne M. Nakane J.S. Pollock M.F. Miller F. Murad Mapping of neural nitric oxide synthase in the rat suggests frequent co-localization with NADPH diaphorase but not with soluble guanylyl cyclase, and novel paraneural functions for nitrinergic signal transduction J Histochem Cytochem 40 1992 1439 1456 Shi et al., 2001 S.R. Shi R.J. Cote C.R. Taylor Antigen retrieval immunohistochemistry and molecular morphology in the year 2001 Appl Immunohistochem Mol Morphol 9 2001 107 116 Shi et al., 1991 S.R. Shi M.E. Key K.L. Kalra Antigen retrieval in formalin-fixed, paraffin-embedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections J Histochem Cytochem 39 1991 741 748 Soares et al., 1999 T.J. Soares T.M. Coimbra A.R. Martins A.G. Pereira E.C. Carnio L.G. Branco Atrial natriuretic peptide and oxytocin induce natriuresis by release of cGMP Proc Natl Acad Sci USA 96 1999 278 283 Springall et al., 1992 D.R. Springall V. Riveros-Moreno L. Buttery A. Suburo A.E. Bishop M. Merrett Immunological detection of nitric oxide synthase(s) in human tissues using heterologous antibodies suggesting different isoforms Histochemistry 98 1992 259 266 Teixeira et al., 2004 S.A. Teixeira M.S. Viapiano L. Ganiko M.C. Roque-Barreira A.R. Martins The novel lectin KM+ detects a specific subset of mannosyl-glycoconjugates in the rat cerebellum Glycoconj J 20 2004 501 508 Zanardo et al., 1997 R.C. Zanardo E. Costa H.H. Ferreira E. Antunes A.R. Martins F. Murad Pharmacological and immunohistochemical evidence for a functional nitric oxide synthase system in rat peritoneal eosinophils Proc Natl Acad Sci USA 94 1997 14111 14114 Zhao et al., 1998 H. Zhao N. Dugas C. Mathiot A. Delmer B. Dugas F. Sigaux B-cell chronic lymphocytic leukemia cells express a functional inducible nitric oxide synthase displaying anti-apoptotic activity Blood 92 1998 1031 1043 Wood and Warnke, 1981 G.S. Wood R. Warnke Suppression of endogenous avidin-binding activity in tissues and its relevance to biotin-avidin detection systems J Histochem Cytochem 29 1981 1196 1204