Sage Journals: Discover world-class research

Abstract

Superoxide anions are highly reactive radicals overproduced in many pathological situations such as inflammation and ischemia. One of the major factors in the protection from superoxide anions is the enzyme superoxide dismutase (SOD), which catalyzes the dismutation of superoxide to hydrogen peroxide. This study presents a quantitative histochemical method to estimate SOD activity in rat brain tissue sections. This method is based on the cerium capture method and 3,3'-diaminobenzidine amplification of transition cerium compounds. Substrate for SOD was provided by reduction of oxygen during the autoxidation of riboflavin in the presence of UV light. This histochemical method reveals the overall activity of the three different forms of SOD described in mammalian tissues: cytosolic copper-zinc SOD, mitochondrial manganese SOD, and the high molecular weight extracellular SOD. Eventually, this method can be used to quantify SOD activity in tissue sections by image analysis.

Keywords

free radicals cerium DAB quantitative histochemistry image analysis

Superoxide anions (O₂ ⁻) are highly reactive radicals (ROS) generated in cells and tissues by enzymes such as xanthine oxidase and the mitochondrial electron chain (Freeman and Crapo 1982; Kehrer 1993). Their excess has been linked to tissue damage in different pathologies (Halliwell et al. 1992; Kontos et al. 1992; Cheeseman and Slater 1993). The production of O₂ ⁻. and related radicals is, in fact, enhanced in conditions such as inflammation, hypoxic damage, and UV ray exposure (Hammond et al. 1985).

Different mechanisms protect cells from O₂ ⁻. and related radicals such as H₂O₂ and OH⁻. The most important one is the dismutation of O₂ ⁻. to H₂O₂ + O₂ and the subsequent dismutation of H₂O₂ to O₂ and H₂O. The first reaction is catalyzed by superoxide dismutase (SOD), the second one by catalase.

In the nervous system, SOD activity is unbalanced during neuronal apoptosis (Troy and Shelanski 1994; Luetjens et al. 2000), cerebral ischemia (Murakami et al. 1998; Fujimura et al. 1999), and in a familiar form of amyotrophic lateral sclerosis (Deng et al. 1993; Rosen et al. 1993). As a consequence, SOD is considered to be a major factor in protection of the nervous tissue against excitotoxic and ischemic/hypoxic lesions (Nelson et al. 1992).

The localization and quantification of SOD activity in tissues is of great interest and can give information about the tissue responses to oxidative stresses. At present it is possible to localize and quantify the amount of SOD enzyme immunohistochemically by use of specific antibodies, but not its activity (Moreno et al. 1997).

Previous experience led to histochemical techniques able to localize SOD activity in situ, but these techniques could not be used to quantify the signal (Frederiks and Bosch 1997; Okabe et al. 1998). Therefore, we developed a new quantitative histochemical method to evaluate SOD activity in the rat CNS.

This method is based on riboflavin-UV as a source of O₂ ⁻. (Segura-Aguilar 1993; Okabe et al. 1998) and cerium capture method to reveal H₂O₂ production (Frederiks and Bosch 1997). In the presence of UV light, riboflavin reduces oxygen to superoxide radical (Beauchamp and Fridovich 1971). The resulting H₂O₂ produced by SOD reacts with cerium ions, giving rise to an insoluble reaction product (cerium per-hydroxide: Ce[OH]₂OOH) (Briggs et al. 1975) which, in a second step, oxidizes diaminobenzidine (Angermuller and Fahimi 1988). The resulting brown product was quantitatively analyzed using a computer-assisted image analysis system to provide an estimate of SOD activity in situ. This histochemical method reveals the sum of the activities of the three different forms of SOD described in mammalian tissues: cytosolic copper–zinc SOD (Cu/Zn–SOD), mitochondrial manganese SOD (Mn–SOD), and the high molecular weight extracellular SOD (EC-SOD) (Weisiger and Fridovich 1973; Marklund 1982).

Materials and Methods

Animals

Twelve adult male Sprague–Dawley rats from our own colony (200–250 g body weight) were housed in groups of two on standard conditions with a 12-hr light–dark cycle (lights from 0700 to 1900) and kept on a laboratory diet and water ad lib. The animals were handled daily to reduce stress due to the manipulation. All other parameters fulfilled the requirements of the “Guide for the Care and Use of Laboratory Animals” by the National Research Council, implemented by EU and local rules.

Brain Removal, Freezing, and Sectioning

The animals were sacrificed by decapitation during the light phase of the cycle to minimize circadian variation. The brain was quickly removed and either directly frozen by immersion in 2-methylbutane cooled in dry ice or first homogenized on ice for 2–3 min (tissue standards) and then transferred into small tubes and frozen in dry ice. Frozen brains and brain pastes were then stored at −80C until sectioning with the cryostat. Cryostat sections (50 μm thick) were cut at −20C and mounted on precleaned glass slides (single frosted; Sigma, Milan, Italy).

Histochemical Staining Procedure

1. Fixation step. The sections were fixed in cold buffered 0.5% glutaraldehyde in 0.1 M, pH 7.4, PBS for 5 min, washed two times in 0.1 M phosphate buffer, pH 7.4, for 5 min each, and then in 0.1 M Tris buffer, pH 8.0, for 5 min.

2. Incubation step. Sections were incubated with a thin film of freshly prepared incubation medium [0.5 mM riboflavin + 10 mM cerium chloride (Sigma) + 0.5% Triton X-100 in Tris-HCl, pH 8] and irradiated under a UV lamp for 30 min.

3. Intensification step. After incubation, sections were rinsed in 100 mM Tris-HCl, pH 8.0, and incubated for 30 min in 1.4 mM 3,3'-diaminobenzidine (DAB; Sigma) + 100 mM NaN₃ + 0.6 mM H₂O₂ in 100 mM Tris-HCl, pH 7.6.

After washing in PBS, sections were dehydrated in ethanol 85%, 95%, and 100% for 5 min each, cleared in xylene, and coverslipped with Permount.

Immunohistochemistry

Alternate sections were stained with the histochemical procedure as above or with immunohistochemistry for the Zn/Cu-SOD isoform as detailed here.

Sections were fixed for 5 min in formaldehyde 4% in PBS. After washing in PBS, sections were treated for 30 min with H₂O₂ 0.3% in methanol, washed again in PBS, and incubated in mouse anti-SOD (Sigma) 1:400 in PBS + 10% normal bovine serum (NBS) overnight at 4C. After three washes in PBS, sections were incubated in biotinilated anti-mouse antibody (Vector; Burlingame, CA) 1:100 + 10% NBS in PBS for 1 hr at room temperature (RT). Sections were then washed three times in PBS and then incubated with ABC (Vector) for 1 hr according to the supplier's instructions. After three washes in PBS, the signal was visualized with 0.1% DAB + 0.02% H₂O₂ in PBS for 10 min in the dark at RT. The reaction was then stopped with cold PBS, the sections dehydrated in alcohol, cleared in xylene, and coverslipped with Permount.

Control Experiments

The following variables were changed during the histochemical staining of sections of tissue standards to test the specificity and the optimal conditions for SOD histochemistry.

Fixation. Sections were stained without fixation in glutaraldehyde.

UV Irradiation. To reduce the formation of superoxide, sections were not exposed to UV irradiation during the incubation with riboflavin.

DAB. The intensification step with DAB was omitted to evaluate the intensity of the signal deriving from cerium ions.

Diethyldithiocarbamate. Superoxide dismutase activity was inhibited adding 10 mM diethyldithiocarbamate (Carlo Erba; Milan, Italy) in the first incubation medium (Frederiks and Bosch 1997).

Methodological Validation of SOD Histochemistry

To validate the histochemical assay as a quantitative tool, the relationship between time of incubation, section thickness, and the densitometrically measured reaction product was studied to fulfill general quantitative methodological requirements detailed by Stoward (1980). The relationship between section thickness and optical density (OD) was analyzed using tissue sections of increasing thickness (25, 50, 75, and 100 μm) and the standard staining procedure.

The relationship between time of incubation and OD was analyzed using 50-μm-thick tissue sections incubated at increasing times during the first step of UV exposure (0, 0.5, 1, 2, 5, 10, 15, and 30 min).

Image Analysis

Brain sections and tissue standards were analyzed at low magnification (obj. X0.5) with a Zeiss Axioskop 20 equipped with high-resolution analogic monochrome camera Hamamatsu C5405. Images were digitized (box size 1024 × 1024; pixel size 1.5 μm) using an MCID-2 board and analyzed with AIS imaging (Imaging Research; St. Catherines, Ontario, Canada). The regions of interest were outlined and the relative optical density [ROD = log (256/gray level)] calculated. The background ROD (unstained tissue) was also calculated. All data were then reported after subtraction of background OD to get an absolute scale (0 OD = unstained tissue).

The light intensity (Köler illumination) was adjusted to avoid saturation in blank fields and in the regions of interest. A blue filter (λ = 440 ± 10 nm) was used to filter out green and red visible light. Light stability was checked over time using the illuminator device of an MCID system. Neutral glasses of defined transmission were used to calibrate the system and check the densitometric linearity.

The camera gain was set to 0 to minimize noise. Slices were always in focus during measurement and each image was averaged over 16 digitized images. Four different measurements were taken on each standard section and four different standard sections were used for each time/section thickness point: therefore, 16 measurements were obtained for each standard.

A blank field was digitized before every session for shading correction. All other parameters were adjusted according to the guidelines in Hardie et al. (2002), Jonker et al. (1997), and Böcking et al. (1997).

Statistics

Each measurement was repeated four to ten times on different sections. Measurements were statistically analyzed calculating the regression coefficient for thickness-OD and time of incubation–OD relationships, or testing different conditions in control experiments by ANOVA and planned comparisons with Student's t-test for non-paired data. All data in the graphs have been reported as mean ± SE.

Results

Histochemical Staining Procedure

The three-step histochemical procedure described here is easy to perform and gives reproducible results. Moreover, the use of standards allows comparison of experiments executed at different times.

This technique requires a first step in glutaraldehyde to preserve morphological details and to fix SOD activity in situ. After the fixation, repeated washes of the tissue sections in PBS before the incubation step effectively remove red blood cells that could interfere during the intensification step with DAB.

Sections are then incubated in a solution containing an O₂ ⁻. -producing system (riboflavin–UV) and a capture system for H₂O₂ (cerium). Oxidized riboflavin is reduced under UV irradiation and then undergoes autoxidation to release superoxide radicals (Segura-Aguilar 1993). These are highly reactive and diffusible species and, in turn, are locally dismutated in H₂O₂ by SOD in brain sections. H₂O₂ reacts locally with cerium ions, giving rise to an insoluble product (Frederiks and Bosch 1997). Eventually, the reaction is intensified by DAB, which is oxidized in presence of H₂O₂ and oxidized cerium, giving rise to an insoluble brown product (Frederiks and Bosch 1997). Sodium azide was added to prevent peroxidase-dependent DAB oxidation (Angermuller and Fahimi 1988).

The morphological details of the neurons, due to the prevalent localization of the enzyme in the soma, are very good (Figure 1). SOD activity is localized mainly in neuronal soma, whereas the neuropil is less reactive (Figure 1). The staining is more evident in the neuronal cytoplasm than in the nucleus (Figure 1). Myelin and white matter appear unstained (Figure 2).

Positive neurons are widespread in the nervous system, across different regions and different neuronal types in the same region. However, some neurons are more intensely stained than others. For example, pyramidal cells of the hippocampus, granule cells in the cerebellum, deep cerebellar nuclei, and some neuronal subsets in the reticular nuclei of the brainstem neurons showed the highest intensity (Figures 2 and 3). Endothelial and ependymal cells, and cells of choroid plexus, are also intensely stained. White matter was generally unstained.

Immunohistochemistry

Alternate sections treated with the histochemical reaction (SOD activity) or immunostained for SOD protein displayed the same staining pattern (Figure 3). The immunohistochemical localization of the cytoplasmic SOD isoenzyme shows a higher definition of cell contours, whereas the histochemical SOD reaction shows some signal also in the neuropil, possibly due to the identification of the extracellular SOD isoenzymes (see also Frederiks and Bosch 1997).

Figure 1

SOD histochemistry in rat cerebral cortex. Note intense staining of pyramidal and granular cells. The neuropil is less intensely stained and the nuclei appear unstained. Bar = 100 μm.

Figure 2

SOD histochemistry at different levels of the rat brain (coronal sections). The activity appears to be widely expressed in the gray matter, the white matter being unstained. Some regions are more densely stained, such as the choroid plexuses, pyramidal cells of the hippocampus, granule cells in the cerebellum, deep cerebellar nuclei, and some nuclei in the brainstem. Insets show the approximate plane position of the sections. Bar = 1 mm.

Specificity and Optimal Conditions of the Histochemical Reaction

A balance is needed between optimal enzyme activity and good tissue preservation, because a totally unfixed brain shows several artifacts and is poorly stained (control = 0.0969 ± 0.008 OD; no fixation = 0.0169 ± 0.008 OD; t-test p<0.05; Figure 4) due to the loss of SOD enzyme in the reaction buffer. The best conditions were no perfusion-fixation and post-fixation on slides for 5 min with 0.5% glutaraldehyde.

The reaction product was significantly reduced without UV irradiation in the incubation step, although residual staining persisted in both paste standards and brain sections (control = 0.0969 ± 0.008 OD; no UV = 0.0437 ± 0.004 OD; t-test p<0.05; Figure 4). The absence of the intensification step resulted in a significant decrease of the reaction product (control = 0.0969 ± 0.008 OD; no DAB = 0.0290 ± 0.003 OD; p<0.05, Figure 4).

Superoxide dismutase inhibitors such as diethyldithiocarbamate (DTC) significantly inhibited the signal (control = 0.0969 ± 0.008 OD; DTC = 0.02 ± 0.006 OD; p<0.05).

Figure 3

Comparison between SOD immunohistochemistry (left) and histochemistry (right) in the cerebellar cortex (A,B), deep cerebellar nuclei (C,D), and trigeminal tract (E,F). Note the overlap between the two staining techniques. Bars: A–D = 100 μm; F,G = 50 μm.

Methodological Validation

As shown in Figure 4, there is a high correlation between section thickness and OD (r=0.98; p≪0.05) and between time of incubation and OD (r=0.935; p≪0.05).

Discussion

A great deal of work in the literature is based on the immunolocalization of SOD enzyme, a few papers focusing on the localization of SOD activity in tissue sections.

Previous work by Frederiks and Bosch (1997) demonstrated SOD activity using the hypoxanthine–xanthine oxidase system to produce O₂ ⁻. The final reaction product was then visualized using the cerium capture method. Unfortunately, the originally described xanthine–hypoxanthine method (Frederiks and Bosch 1997; Ukeda et al. 1997) required a very dense solution so that the sections were severely damaged during the histochemical procedure. Moreover, an aspecific signal was always present outside the section, making this technique not useful for quantitative purposes.

Another histochemical technique was based on blotting the tissue sections on a filter and then staining the filter using the principle of inhibition of autoxidation of nitroblue tetrazolium in the presence of SOD (Okabe et al. 1998). Unfortunately the blot itself does not allow display of the morphological details.

The present study shows that SOD activity can be determined and quantified by autoxidation of riboflavin in the presence of UV light as the source of substrate (superoxide radicals) for SOD activity (Traber et al. 1982). One product of this enzymatic activity, H₂O₂, reacts with cerium ions, giving rise to insoluble cerium perhydroxide that is visualized by the brown reaction product formed by oxidized 3,3'-diaminobenzidine. Using OD analysis, it is then possible to quantify this product and therefore SOD activity.

Figure 4

Effect of section thickness (A) (representative images of stained cerebellum of increasing thickness are shown), time of incubation (B), and different control conditions (C: without fixation, with diethylthiocarbamate (DTC), without UV irradiation, without intensification step (DAB) and normal conditions; see text for details) on the OD of standard brain homogenates. The standard procedure gave the best results in terms of signal intensity. The increase in OD was linear with respect to increasing incubation time and section thickness.

The advantage of the technique reported here is the possibility of visualizing SOD activity with high resolution and of quantifying the reaction product. Our technique showed high SOD activity in hippocampal pyramidal and granular cells and in the granular layer of cerebellum, in agreement with previous data (Moreno et al. 1997; Okabe et al. 1998). The specificity of this new method has been established by the use of diethyldithiocarbamate, a specific SOD inhibitor.

It is interesting to note that without UV irradiation (therefore without an external source of superoxide anions) there is weak residual staining in both paste standards and brain sections, possibly due to ROS production in the section (Wentworth et al. 2001) or to lipid hydroperoxides that might react with cerium ions (Halliwell and Chirico 1993).

The optical density (OD) of the sections was highly correlated with the duration of incubation time and the section thickness, fulfilling general quantitative methodological requirements detailed by Stoward (1980): (a) there must be a way to subtract non-specific staining (background subtraction); (b) the change in OD must be linear with the incubation period and with (c) tissue section thickness.

However, because cerium ions have a low degree of penetration, detergents (Triton X-100 in our case; Robinson 1985) must be used to obtain fully linear results also for thicker sections (see also Van Noorden and Frederiks 1993 for alternative methods to allow cerium ions to penetrate).

Some differences between SOD histochemistry and immunohistochemistry, particularly from a quantitative point of view, may depend on the activity of different forms of SOD enzymes (Cu/Zn–SOD, Mn–SOD, and extracellular SOD) with the histochemical method, whereas immunolocalization allows the determination of one type of SOD isoform due to the specificity of the antibody (see also Okabe et al. 1998 for a discussion of the similarity between the histochemical and immunohistochemical techniques).

Moreover, the histochemical method could be sensitive to the enzymatic activity/inactivity of SOD in the section, modulated by oxidative substances (Jewett et al. 1989). It is therefore possible that the ratio of active SOD to inactive SOD results in a quantitatively different signal using SOD immunolocalization and SOD histochemistry.

Footnotes

Acknowledgements

We are grateful to anonymous referees for useful comments.

References

Angermuller

Fahimi

(1988) Light microscopic visualization of the reaction product of cerium used for localization of peroxisomal oxidases. J Histochem Cytochem 36:23–28

Beauchamp

Fridovich

(1971) Superoxide dismutase: improved assays and an assay applicable to acrylamide gels. Anal Biochem 44:276–287

Böcking

Giroud

Reith

(1997) Consensus report of the ESCAP task force on standardization of diagnostic DNA image cytometry. European Society for Analytical Cellular Pathology (ESCAP). http://www.biochem.mpg.de/valet/esacimag.html

Briggs

Drath

Karnovsky

(1975) Localization of NADH oxidase on the surface of human polymorphonuclear leukocytes by a new cytochemical method. J Cell Biol 67:566–586

Cheeseman

Slater

(1993) An introduction to free radical biochemistry. Br Med Bull 49:481–493

Deng

Hentati

Tainer

Iqbal

Cayabyab

Hung

Getzoff

. (1993) Amyotrophic lateral sclerosis and structural defects in Cu/Zn superoxide dismutase. Science 261:1047–1051

Frederiks

Bosch

(1997) Localization of superoxide dismutase activity in rat tissues. Free Radic Biol Med 22:241–248

Freeman

Crapo

(1982) Biology of disease: free radicals and tissue injury. Lab Invest 47:412–426

Fujimura

Morita-Fujimura

Kawase

Copin

Calagui

Epstein

Chan

(1999) Manganese superoxide dismutase mediates the early release of mitochondrial cytochrome C and subsequent DNA fragmentation after permanent focal cerebral ischemia in mice. J Neurosci 19:3414–3422

10.

Halliwell

Chirico

(1993) Lipid peroxidation: its mechanism, measurement, and significance. Am J Clin Nutr 715S–724S

11.

Halliwell

Gutteridge

Cross

(1992) Free radicals, antioxidants, and human disease: where are we now? J Lab Clin Med 119:598–620

12.

Hammond

Kontos

Hess

(1985) Oxygen radicals in the adult respiratory distress syndrome, in myocardial ischemia and reperfusion injury, and in cerebral vascular damage. Can J Physiol Pharmacol 63:173–187

13.

Hardie

Gregory

Hebert

PDN

(2002) From pixels to picograms: a beginners' guide to genome quantification by Feulgen image analysis densitometry. J Histochem Cytochem 50:735–749

14.

Jewett

Cushing

Gillespie

Smith

Sparks

(1989) Reaction of bovine-liver copper-zinc superoxide dismutase with hydrogen peroxide. Evidence for reaction with H₂O₂ and HO^2- leading to loss of copper. Eur J Biochem 180:569–575

15.

Jonker

Geerts

WJC

Chieco

Moorman

AFM

Lamers

Van Noorden

(1997) Basic strategies for valid cytometry using image analysis. Histochem J 29:347–364

16.

Kehrer

(1993) Free radicals as mediators of tissue injury and disease. Crit Rev Toxicol 23:21–48

17.

Kontos

Wei

Williams

Kontos

Povlishock

(1992) Cytochemical detection of superoxide in cerebral inflammation and ischemia in vivo. Am J Physiol 263:H1234–1242

18.

Luetjens

Bui

Sengpiel

Munstermann

Poppe

Krohn

Bauerbach

. (2000) Delayed mitochondrial dysfunction in excitotoxic neuron death: cytochrome c release and a secondary increase in superoxide production. J Neurosci 20:5715–5723

19.

Marklund

(1982) Human copper-containing superoxide dismutase of high molecular weight. Proc Natl Acad Sci USA 79:7634–7638

20.

Moreno

Nardacci

Ceru

(1997) Regional and ultrastructural immunolocalization of copper-zinc superoxide dismutase in rat central nervous system. J Histochem Cytochem 45:1611–1622

21.

Murakami

Kondo

Kawase

Sato

Chen

Chan

(1998) Mitochondrial susceptibility to oxidative stress exacerbates cerebral infarction that follows permanent focal cerebral ischemia in mutant mice with manganese superoxide dismutase deficiency. J Neurosci 18:205–213

22.

Nelson

Wei

Povlishock

Kontos

Moskowitz

(1992) Oxygen radicals in cerebral ischemia. Am J Physiol 263:H1356–1362

23.

Okabe

Saito

Ito

Kimura

Niioka

Kurasaki

(1998) Histochemical localization of superoxide dismutase activity in rat brain. Free Radic Biol Med 24:1470–1476

24.

Robinson

(1985) Improved localization of intracellular sites of phosphatases using cerium and cell permeabilization. J Histochem Cytochem 33:749–754

25.

Rosen

Siddique

Patterson

Figlewicz

Sapp

Hentati

Donaldson

. (1993) Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature 362:59–62

26.

Segura-Aguilar

(1993) A new direct method for determining superoxide dismutase activity by measuring hydrogen peroxide formation. Chem Biol Interact 86:69–78

27.

Stoward

(1980) Criteria for the validation of quantitative histochemical enzyme techniques. Trends in Enzyme Histochemistry and Cytochemistry. Amsterdam, CIBA Foundation, Excerpta Medica 11–31

28.

Traber

Kramer

Hemmerich

(1982) Mechanism of light-induced reduction of biological redox centers by amino acids. A flash photolysis study of flavin photoreduction by ethylenediaminetetraacetate and nitrilotriacetate. Biochemistry 21:1687–1693

29.

Troy

Shelanski

(1994) Down-regulation of copper/zinc superoxide dismutase causes apoptotic death in PC12 neuronal cells. Proc Natl Acad Sci USA 91:6384–6387

30.

Ukeda

Maeda

Ishii

Sawamura

(1997) Spectrophotometric assay for superoxide dismutase based on tetrazolium salt 3′–1–(phenylamino)-carbonyl–3,4-tetrazolium]-bis(4-methoxy-6- nitro)benzenesulfonic acid hydrate reduction by xanthine-xanthine oxidase. Anal Biochem 251:206–209

31.

Van Noorden

CJF

Frederiks

(1993) Cerium methods for light and electron microscopical histochemistry. J Microsc 171:3–16

32.

Weisiger

Fridovich

(1973) Mitochondrial superoxide simutase. Site of synthesis and intramitochondrial localization. J Biol Chem 248:4793–4796

33.

Wentworth

Jr Jones

Wentworth

Zhu

Larsen

Wilson

. (2001) Antibody catalysis of the oxidation of water. Science 293:1806–1811

Quantitative Histochemical Assay for Superoxide Dismutase in Rat Brain

Abstract

Keywords

Materials and Methods

Animals

Brain Removal, Freezing, and Sectioning

Histochemical Staining Procedure

Immunohistochemistry

Control Experiments

Methodological Validation of SOD Histochemistry

Image Analysis

Statistics

Results

Histochemical Staining Procedure

Immunohistochemistry

Specificity and Optimal Conditions of the Histochemical Reaction

Methodological Validation

Discussion

Footnotes

Acknowledgements

References