Comparison of indirect peroxidase and avidin‐biotin‐peroxidase complex (ABC) immunohistochemical staining procedures for c‐fos in rat brain

Abstract

c‐Fos is the product of a gene expressed within neurons in the brain that serves as an anatomical marker of cellular activation. Immunohistochemical staining for c‐fos allows a characterization of the effects of many different types of experimental manipulations on neuronal activity, making it a powerful technique for understanding brain, drug and behavior relationships. This study compared visualization of an anti‐c‐fos primary antibody in 40‐μm‐thick cryostat sections of formaldehyde‐fixed rat brainstem using either a peroxidase enzyme‐conjugated secondary antibody (indirect peroxidase) or the peroxidase‐conjugated avidin‐biotin complex (ABC) method. All sections were treated with H₂O₂ to quench endogenous peroxidase enzyme and sodium borohydride to enhance permeability of the tissue and improve staining quality. Every other section was used to examine either the indirect peroxidase or the ABC method. Sections for the indirect peroxidase method were treated with Triton X‐100 detergent to increase tissue permeability, goat serum to reduce non‐specific binding of the secondary antibody and, in some cases, bovine serum albumin (BSA) to reduce non‐specific binding of the primary antibody. Sections for the ABC method were treated with dilute normal serum, and avidin and biotin solutions and, in some cases BSA. Alternate sections were incubated for 72 h in either rabbit anti‐c‐fos primary antibody (1 : 20 000) or its vehicle (negative control). For the indirect peroxidase protocol, tissues were treated with peroxidase‐conjugated goat anti‐rabbit secondary antibody. For the ABC protocol, tissues were treated with biotinylated goat anti‐rabbit secondary antibody and ABC peroxidase complex. All sections were reacted with 3,3′‐diaminobenzadine (DAB) and H₂O₂, mounted and coverslipped. Both methods produced specific staining of c‐fos‐containing neurons, relative to the negative control sections. The indirect peroxidase protocol produced clear staining of c‐fos‐containing neurons, with very little background in the negative control sections. Staining for c‐fos was enhanced using the ABC method in that c‐fos stained neurons were darker and more clearly visible after shorter treatment with DAB. However, negative control sections showed a greater amount of non‐specific staining with the ABC method. Thus, the ABC method was more sensitive but showed reduced specificity, with BSA treatment slightly reducing the level of non‐specific staining. Overall, the ABC method produced better visualization and contrast of c‐fos‐containing neurons against the background color of the tissue.

Introduction

c‐Fos is the protein product of a gene characterized as a proto‐oncogene because of its role in tumorigenesis and cell proliferation (Stiles, 1985; see Feldman & Yarden (2014) for a review). c‐Fos is part of an intracellular biochemical cascade that activates genetic transcription induced by growth factors and other substances (Greenberg & Ziff, 1984; Muller et al. 1984). Because c‐fos protein expression is rapidly evoked by a variety of stimuli, the c‐fos gene is categorized as an immediate early gene (Greenberg & Ziff, 1984; Muller et al. 1984).

Within behavioral neuroscience research, c‐fos protein expression is often used as an anatomical marker of neuronal activation in the brain (Curran & Morgan, 1995). Immunohistochemical staining for c‐fos allows for a widespread characterization of the effects of many different types of chemical, synaptic and behavioral manipulations that activate neurons (see Flavell & Greenberg 2008, Loebrich & Nedivi 2009 and Minatohara et al. 2016 for reviews). One use for c‐fos staining is to determine the effects of drugs throughout the brain (Hale et al. 2010; Zahm et al. 2010; Davoodi et al. 2014), making it a powerful technique for increasing information about brain, drug and behavior relationships.

Immunohistochemical staining methods often use an unlabeled primary antibody that binds to the antigen of interest in the tissue (reviewed in Sternberger, 1986). Subsequently, the primary antibody is visualized by the attachment of another (secondary) antibody conjugated with a molecule that can be visualized or produces a visible product. A common strategy involves the attachment to the secondary antibody of a single peroxidase enzyme whose activity can create a visible label (Nakane & Pierce, 1967). An alternative strategy employs a secondary antibody complexed with avidin, biotin and multiple peroxidase enzymes, the avidin‐biotin‐complex (ABC) method (Hsu et al. 1981). Because the greater number of enzymes creates a more visible label, the ABC method generally has a higher sensitivity and produces more intense tissue staining (Hsu et al. 1981).

Although the ABC method is more sensitive, it can lead to enhanced background staining because of non‐specific binding of avidin and/or biotin to the tissue (reviewed in Sternberger, 1986). Additional steps may be necessary to reduce the non‐specific staining produced. Moreover, immunohistochemical staining using different primary antibodies may vary idiosyncratically with the different visualization methods and reagents used, such that no particular procedure will always and universally produce superior results (reviewed in Sternberger, 1986).

To optimize our immunohistochemical procedure for visualizing c‐fos expression in formaldehyde‐fixed rat brains, we compared the ABC method for visualizing c‐fos with the indirect peroxide method. Staining protocols for the latter are simpler and entail fewer steps and reagents, saving time and expense, while generally producing good results. We made a direct comparison of the two visualization procedures to determine their relative strengths and weaknesses regarding specific staining and non‐specific background labeling.

Methods

Chemicals were obtained from Fisher Scientific (Fair Lawn, NJ, USA) or Sigma Chemical (St. Louis, MO, USA) except where noted. Rat brains were used from a research project that was approved by the Illinois State University Institutional Animal Care and Use Committee. Care of the rats met the standards in the US National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Male Sprague‐Dawley rats (Harlan, Indianapolis, IN, USA) weighing 275–400 g were anesthetized with urethane (1.5 g kg⁻¹ i.p.) and prepared for electrophysiological recording as in previous work (Heidenreich & Rebec, 2000; Heidenreich et al. 2004; Greco et al. 2006). Briefly, each rat was placed in a stereotaxic instrument (David Kopf Instruments, Tujunga, CA, USA); the scalp was incised, a burr hole was drilled in the skull and the dura excised, and a glass microelectrode was lowered into the forebrain. After completion of the electrophysiological experiment 4–6 h later, rats were given an overdose of urethane i.v. until breathing ceased. After thoracotomy, the pericardium was removed and a transcardial perfusion was performed with 300 mL of 0.9% saline followed by 300 mL of 4% para‐formaldehyde in 0.1 m phosphate buffer (PB). The brain was removed from the skull, placed in paraformaldehyde for 18–24 h for further fixation and stored in 20% sucrose in 0.1 m PB at 4 °C for at least 4 days to cryoprotect the brain.

A cryostat microtome was used to cut 40‐μm‐thick sections of rat brainstems (medulla, pons, midbrain). Sections were washed three times in 10 mm phosphate‐buffered saline (PBS) for 10 min each and then treated with 0.4% H₂O₂ in methanol for 20 min to quench endogenous peroxidase activity in the tissue. After three 10‐min washes with PBS, tissues were treated for 30 min with a solution of 1% sodium borohydride in 0.1 m PB to enhance the permeability of the tissue and improve staining quality, as in previous work (Muma et al. 2001; Greco et al. 2006). The sections were washed twice in 0.1 m PB for 5 min each and PBS for 30 min.

Every other section was used to examine either the peroxidase enzyme‐conjugated secondary antibody (indirect peroxidase) or the peroxidase‐conjugated avidin‐biotin complex (ABC) method. The sections for the indirect peroxidase method were treated three times for 20 min each with PBS containing 0.2% Triton X‐100 detergent to increase tissue permeability and 1% goat serum to reduce non‐specific binding of the secondary antibody used later. The sections for the ABC method, however, were treated with diluted normal blocking serum (Vector Elite ABC kit, Vector Laboratories, Burlingame, CA, USA) and solutions of avidin and biotin (Vector blocking kit), all in PBS. For some brains, sections were then treated with 0.8% bovine serum albumin (BSA) in PBS for 40 min to reduce non‐specific binding of the primary antibody to the tissue. For both methods, alternate sections were incubated for approximately 72 h at 4 °C in either rabbit polyclonal anti‐c‐fos primary antibody in PBS (Ab‐5, 1 : 20 000, Calbiochem, EMD Chemicals, San Diego, CA, USA) with 0.25% Triton X‐100 or the same solution without primary antibody (omit primary antibody negative control). The activity and specificity of this anti‐c‐fos antibody has been verified by the manufacturer and it does not react with the related c‐jun protein. This antibody has been used successfully in previous studies of c‐fos in formaldehyde‐fixed rat brain using the ABC method (Hale et al. 2010; Zahm et al. 2010).

For the indirect peroxidase protocol, sections were removed from incubation solutions, washed three times in PBS for 10 min each and treated for 60 min with peroxidase‐conjugated goat anti‐rabbit secondary antibody (1 : 1000, Jackson Immunologicals, West Grove, PA, USA). After three 10 min washes in PBS, the tissues were treated with 0.0005% 3,3′‐diaminobenzadine (DAB) and 0.0001 H₂O₂ in PBS for 45–90 s, creating the insoluble brown chromogen. After three 10‐min washes in PBS, the sections were mounted on gelatin‐coated slides from 0.033 m PB and allowed to dry overnight. Slides were coverslipped with DPX (Electron Microscopy Sciences, Hatfield, PA, USA) after dehydration through a series of increasingly concentrated ethanol solutions and xylene.

For the ABC protocol, tissues were removed from incubation solutions, washed three times in PBS for 10 min each and treated for 60 min with biotinylated goat anti‐rabbit secondary antibody (Vector Elite ABC kit). After three 10‐min washes in PBS, sections were treated for 60 min with a solution of ABC peroxidase complex (Vector Elite ABC kit). After three 10‐min washes in PBS, sections were treated with DAB and H₂O₂, washed in three times in PBS for 10 min each and mounted on slides and coverslipped as above.

Stained and negative control sections from both methods were examined and compared using standard light microscopy (Leica Microsystems MZ9.5 and DMRBE microscopes, Buffalo Grove, IL, USA). Photomicrographs were taken using a CCD camera (Leica Microsystems DFC300FX) and the image‐pro express MC program (MediaCybernetics, Silver Spring, MD, USA) running on a Windows‐based PC and were not altered or enhanced with image editing software.

To compare the two visualization methods, the number of c‐fos immunoreactive neurons was determined for the lateral parabrachial nucleus, an area that showed numerous stained cells in all brains. A single section was chosen with the greatest extent and density of c‐fos positive neurons for each side of the brain, corresponding to 0.16 mm posterior to the interaural line according to the atlas of Paxinos & Watson (1986). Manual counts of stained cells were made bilaterally using image‐pro express and averaged for each brain.

To determine the depth of background color from the staining procedures, we examined the superior colliculus, a large area of gray matter that stained homogeneously but with few immunoreactive cells. The center of the superior colliculus, free of any tissue damage, processing artifact or large blood vessels, was analyzed unilaterally in a single section from each brain under 200× magnification. The image‐pro express program was used to determine a measure of luminance, i.e. the amount of light transmitted through the tissue. Depth of color in the tissue from staining is the inverse of luminance; darker colored sections yield smaller luminance values.

Data are presented as the mean ± SEM. Statistical analyses were performed using paired t‐tests with the level of significance set at P < 0.05.

Results and Discussion

Both the indirect peroxidase (Fig. 1) and ABC (Fig. 2) methods produced specific staining of c‐fos (Figs 1a,c,e and 2a,c,e), relative to the negative control sections (Figs 1b,d,f and 2b,d,f). The protocol using the indirect peroxidase‐conjugated secondary antibody produced clear staining of c‐fos‐containing neurons (Fig. 1c,e) with very little background staining (color) in the negative control sections (Fig. 1b,d,f). Analysis of the luminance of the omit primary sections and the anti‐c‐fos antibody treated sections yielded comparable values (Table 1; t(2) = 3.2, n.s.); tissue darkness and background staining did not differ significantly with inclusion of the primary antibody using the indirect peroxide method.

Figure 1.

c‐Fos immunostaining using the indirect peroxidase method. Top: Photomicrographs of entire sections of the pons immunostained using the indirect peroxidase method (a) or the negative control procedure (b). Note the higher level of staining in the section using the indirect method relative to the background color of the negative control. Magnification: 8×; scale bars: 1000 μm. Middle: Photomicrographs of the lateral parabrachial nucleus of the pons (arrow) immunostained using the indirect peroxidase method (c) or the negative control procedure (d). Note the numerous stained neurons in the section using the indirect method relative to the absence of cells in the negative control. Magnification: 60×; scale bars: 250 μm. Bottom: Photomicrographs showing detail of c‐fos immunoreactive cells in the lateral parabrachial nucleus of the pons (e) and their absence in the negative control section (f). Magnification: 400×; scale bars: 50 μm.

Figure 2.

c‐Fos immunostaining using the ABC peroxidase method. Top: Photomicrographs of entire sections of the pons immunostained using the ABC peroxidase method (a) or the negative control procedure (b). Note the higher level of staining in the section using the ABC method relative to the background color of the negative control. Magnification: 8×; scale bars: 1000 μm. Middle: Photomicrographs of the lateral parabrachial nucleus of the pons (arrow) immunostained using the ABC peroxidase method (c) or the negative control procedure (d). Note the numerous stained neurons in the section using the ABC method relative to the absence of cells in the negative control. Magnification: 60×; scale bars: 250 μm. Bottom: Photomicrographs showing detail of c‐fos immunoreactive cells in the lateral parabrachial nucleus of the pons (e) and their absence in the negative control section (f). Magnification: 400×; scale bars: 50 μm.

Table 1.

Tissue luminance after staining procedures

	Omit primary controls	Anti‐c‐fos antibody
Indirect peroxidase method
With BSA	228.8 ± 0.3	226.4 ± 1.1
Without BSA	229.9*	221.7*
ABC method
With BSA	222.3 ± 2.9	196.9 ± 4.0a,b
Without BSA	224.4 ± 0.3	190.1 ± 4.7c,d

BSA, bovine serum albumin.

Data are presented as the mean ± SEM; n = 3/group except *n = 1.

^a P < 0.05 vs. ABC method omit primary controls with BSA.

^b P < 0.05 vs. indirect peroxide method anti‐c‐fos antibody with BSA.

^c P < 0.05 vs. ABC method omit primary controls without BSA.

^d P < 0.05 vs. ABC method anti‐c‐fos antibody with BSA.

Staining for c‐fos was enhanced using the ABC method, in that c‐fos stained neurons were darker and more clearly visible in the tissue (Fig. 2c,e) after shorter treatment with DAB, the visualization chromogen used. However, the number of c‐fos immunoreactive cells in the lateral parabrachial nuclei seen with the ABC method (155.3 ± 30.2 cells section⁻¹; n = 4) did not differ from the number produced by the indirect peroxide method (111.4 ± 23.2 cells section⁻¹; n = 4; t(3) = 1.4, n.s.).

Luminance values for the omit primary control sections and the anti‐c‐fos antibody‐treated sections with the ABC method differed significantly (Table 1; t(2) = 18.2, P < 0.05); tissue darkness and background staining was greater with inclusion of the primary antibody. Control sections using the ABC procedure (Fig. 2b,d,f) appeared slightly darker than the corresponding sections using the indirect peroxide method (Fig. 1b,d,f) but did not differ significantly in luminance values (Table 1; t(2) = 2.2, n.s.). However, with inclusion of the anti‐c‐fos primary antibody, luminance was significantly lower after ABC treatment relative to the indirect peroxide method (Table 1; t(2) = 7.7, P < 0.05); background staining was greater using the ABC procedure.

Thus, the ABC method was more sensitive than the indirect peroxidase method, in terms of the intensity of color of c‐fos, but this was accompanied by an increase in the background staining of the tissue. It is not known why this background staining was greater, although treatment with H₂O₂, normal serum, avidin, biotin and, in some cases, BSA, should have removed likely sources. Although it is possible that reducing the incubation time in the anti‐c‐fos antibody could reduce the level of non‐specific staining using the ABC method, the 72 h incubation posed no problem using the indirect peroxidase protocol. For the purposes of this particular study, the ABC method produced better overall results visualizing c‐fos immunoreactive neurons as compared with the indirect peroxidase method. Similar protocols using the ABC method have been used with good results (Hale et al. 2010; Zahm et al. 2010).

We examined the effect of treatment with BSA prior to incubation in the primary antibody or the diluent using either the ABC protocol (Fig. 3) or the indirect peroxidase method (not shown). For the ABC method, treatment with BSA had no effect on luminance values for the omit primary control sections (Table 1; t(2) = 0.7, n.s.). In contrast, luminance values were significantly greater for the sections treated with BSA prior to incubation in anti‐c‐fos antibody compared with those not treated with BSA (Table 1; t(2) = 4.5, P < 0.05); BSA reduced the darkness of background staining in sections with clear c‐fos immunoreactivity. This result is consistent with the intent of the BSA treatment to reduce non‐specific binding of the primary antibody to the tissue, to reduce background color.

Figure 3.

Preincubation treatment with BSA slightly reduced background staining. Photomicrographs of entire sections of the pons immunostained using the ABC peroxidase method (a,b) or the negative control procedure (c,d). Note the slight effect of BSA treatment (a,c) on reducing the level of background staining relative to sections not treated with BSA (b,d). Magnification: 8×; scale bars: 1000 μm.

Using both staining methods, c‐fos immunoreactive neurons were found throughout the brainstem. There was no noticeable difference between the indirect peroxidase and ABC methods regarding the general pattern of neuronal immunoreactivity visible. Structures with c‐fos positive neurons included the dorsal raphe nucleus, an area of serotonin‐containing cell bodies, the locus ceruleus, an area of norepinephrine‐containing cell bodies, and the lateral parabrachial nuclei (shown). The lateral parabrachial nuclei play a role in regulating respiration (Chamberlin & Saper, 1994), heart rate, blood pressure and fluid levels (Davern, 2014). The brains in this study came from urethane‐anesthetized rats; neuronal activity, thus c‐fos staining, in the lateral parabrachial nuclei likely was high to maintain these critical homeostatic systems, even as activity in other parts of the brain may have been relatively low due to anesthesia. Interestingly, the lateral parabrachial nuclei show low basal levels of c‐fos immunostaining in rats in a drug self‐administration paradigm (Zahm et al. 2010).

In conclusion, the ABC method produced better visualization of the c‐fos‐containing neurons against the background color of the tissue in this study. This resulted in a greater ease of counting stained cells by visual inspection. However, for other purposes, e.g. multiple immunohistochemical labeling procedures or neuronal tract tracing experiments, the indirect method may be preferable because of its clear specific staining and lower background color. Treatment of sections with BSA prior to incubation in the primary antibody had a positive effect overall by reducing non‐specific staining.

Author contributions

B.A.H. and J.L.B. conceived the study, J.L.B. collected the data and B.A.H., J.L.B. and B.J.B. contributed to the preparation of the manuscript.