ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS

Vol. 255, No. 2, June, pp. 329-336,1987

Negative and Positive Assays of Superoxide Dismutase

Based on Hematoxylin Autoxidation’-*

JOSEPH P. MARTIN, JR.,~ MICHAEL DAILEY, AND ELLIOTT SUGARMAN

Department of Biology, William Marsh Rice University, P.O. Box 1892, Houstmz, Texas 77005

Received December 1,1986, and in revised form February 25,1987

Hematoxylin, a natural dye commonly used as a histological stain, generates super-

oxide upon oxidation to its quinonoid product, hematein. The parameters affecting this
reaction were assessed in developing a new and versatile assay for superoxide dismutase.
The autoxidation of hematoxylin to hematein was accompanied by an increase in ab-
sorbance between 400 and 670 nm. The autoxidation rate was proportional to hematoxylin
concentration and increased with pH above 6.55. Trace metals accelerated the autoxi-
dation and this effect was eliminated by EDTA. Superoxide dismutase inhibited the
autoxidation 90-95% below pH 7.8, but above pH 8.1 the rate was augmented by super-
oxide dismutase. The rate inhibition at low pH was proportional to the superoxide
dismutase concentration up to 70% inhibition. The rate acceleration at high pH was
proportional to superoxide dismutase concentration up to approximately 200% accel-
eration. The autoxidation rate was not significantly affected by ethanol, cyanide, azide,
hydrogen peroxide, or catalase. However, the reaction was inhibited by the reducing
agents NADH, reduced glutathione, ascorbate, and dithiothreitol, and by undialyzed
extracts of Escherichia coli B. When cell extracts were dialyzed prior to assay, the degree
of inhibition observed was proportional to the concentration of superoxide dismutase
in the extract. These observations form the basis for negative and positive assays of
superoxide dismutase which are inexpensive and simple to perform. The negative assay
has the added advantage of being applicable at physiological pH. o 1987 Academic press, inc.

Superoxide dismutase (SOD)4 catalyzes A variety of spectrophotometric assays

the reaction
0,+02+2H+-)H202+02. PI
i A preliminary communication of these results ap-
peared in (1983) Fed Proc 44,687.
2 This work was supported by Public Health Service
Research Grant AI-19695 from the National Institute
of Allergy and Infectious Diseases; Grant C-900 from
the Robert A. Welch Foundation; and a grant from
the American Heart Association, Texas Affiliate.
3 To whom correspondence should be addressed.
4 Abbreviations used: BSA, bovine serum albumin;
NADH, B-nicotinamide adenine dinucleotide, reduced
form; CuZnSOD, copper zinc superoxide dismutase;
SOD, superoxide dismutase; 01, superoxide; EDTA,
ethylenediaminetetraacetic acid, A,,, absorbance at
560 nm.
329
for the enzyme have been described. One
category of assays combines superoxide
generating systems, such as xanthine ox-
idase (l), alkaline dimethyl sulfoxide (2),
and photochemical oxidation of reduced
substrates (3), with an indicating scaven-
ger of superoxide. These scavengers include
cytochrome c (l), hydroxylamine (4), and
nitroblue tetrazolium (3). In these assays
superoxide dismutase activity is quanti-
tated by determining the ability of the en-
zyme to compete with the scavenger for the
available 0;. The assays generally require
the use of one or more proteins. In contrast,
autoxidation assays can be performed in
which the oxidizing species is both a source
of 0; and an indicating scavenger for 0;.
0003-9861/87 $3.00
Copyright 0 1987 by Academic Press, Inc.
All rights of reproduction in any form reserved.
330 MARTIN, DAILEY,
Assays based on the autoxidation of py-
rogallol (5, 6), hydroxylamine (7), 6-hy-
droxydopamine (8), and epinephrine (9,10)
are simple to perform, involve few reaction
components, and are easily monitored in
the visible region of the absorption spec-
trum. One disadvantage of autoxidation
assays is that most of them must be carried
out at high pH where the specific activity
of manganese and iron SODS are low (11).
Only the 6-hydroxydopamine assay is ap-
plicable at pH values in the physiological
range (8).
In this paper we report the development
of a new autoxidation assay based on the
transformation of hematoxylin to its two
electron-oxidized product, hematein. The
assay is sensitive, simple, and inexpensive.
The reaction product, hematein, is rela-
tively stable and has an absorption maxi-
mum and extinction coefficient similar to
those of the commonly used 0; scavenger,
cytochrome c. The reaction is inhibited by
SOD and can be performed in the physio-
logical pH range of 6.8-7.8 where SOD in-
hibits the reaction 90-95%. In addition, at
pH values above 8.1, the autoxidation re-
action is accelerated by added SOD and the
reaction becomes a positive assay for the
enzyme, similar in principle to SOD assays
based upon the photochemical and enzy-
matic oxidation of dianisidine (12, 13).
MATERIALS AND METHODS
Materials. BSA, catalase, CuZnSOD, cytochrome c
type III, NADH, ascorbate, glucose, reduced gluta-
thione, dithiothreitol, hematoxylin, and hematein
were purchased from Sigma Chemical Co., St. Louis,
Missouri. Phosphate buffer salts, potassium cyanide,
hydrogen peroxide, sodium azide, and metal salts were
purchased from Fisher Scientific Co. Trypticase soy
broth and yeast extract were from Difco.
Cell preparation. Escherichia coli B (ATCC-23226)
provided by Irwin Fridovich was used throughout this
study. Cultures were grown in M9 salts plus glucose
(3 g NaaHP04, 0.5 g NaCl, 1 g NH&l, 495 mg MgSOI,
10.1 mg CaClz; per liter, pH 7.2) or in TSY medium
(3% trypticase soy, 0.5% yeast extract). Bacteria were
grown in 50-ml cultures in 250-ml nephelo flasks at
37°C with vigorous shaking (200 rpm) in a New
Brunswick G-76 rotary platform shaker. Preinduction
of E. coli MnSOD was achieved by the addition of 100
pM MnClz and 100 PM FeCls to the M9 medium followed
by 4 to 6 h of growth (14). Homogenates of E. coli B
AND SUGARMAN
were prepared by harvesting late log cells by centrif-
ugation at 7OOOgat 4°C for 20 min followed by washing
and resuspension in 2 ml of 0.05 M potassium phos-
phate, 0.1 mM EDTA, pH 7.4. Resuspended cells were
sonicated by six, 1-min bursts at 100 W in the cup
horn of a Heat Systems Model 370 sonicator. Lysates
were centrifuged at 15OOOgfor 20 min at 4°C and the
supernatants were dialyzed for 48 h against 0.01 M
potassium phosphate, 0.1 mM EDTA, pH 7.4. SOD ac-
tivity was determined by the xanthine oxidase-cy-
tochrome c method (1). Xanthine oxidase was isolated
from unpasteurized cream (15). General protein was
determined by the Bradford assay using BSA as a
standard (16).
Hematoxylin assay. Hematoxylin was made up as
a 5 mM stock solution in 0.05 M monobasic potassium
phosphate buffer. Reactions were started by adding
aliquots of this stock solution to 3 ml of 0.05 M potas-
sium phosphate, 0.1 mM EDTA, at the indicated pH
at 25°C. Autoxidation was monitored as an increase
in the absorbance at 560 nm in a Gilford Model 2000
spectrophotometer, an IBM Model 9420 uv-visible
spectrophotometer or a Cary Model 15 uv-visible
spectrophotometer. The rates reported represent those
obtained over the first 4 min of reaction. Anaerobic
measurements and measurements under 100% oxygen
were made in anaerobic quartz cuvettes (17). The
buffer was scrubbed for 25 min with ultrapure nitro-
gen or oxygen and hematoxylin was tipped into the
buffer from a sidearm. SOD inhibition studies were
carried out by adding aliquots of bacterial extract or
of purified CuZnSOD solution to the hematoxylin au-
toxidation assay. The SOD solution and bacterial ex-
tracts were precalibrated using the xanthine oxidase-
cytochrome c method (1).
RESULTS
Hematoxylin a&oxidation. When he-
matoxylin (Fig. 1) dissolved in an acidic
stock solution was diluted into 0.05 M po-
tassium phosphate, 0.1 mM EDTA, pH 7.5,
25°C its autoxidation was observed as an
increase in absorbance between 400 and 670
nm (Fig. 2). The spectra of the oxidized
substrate were identical to those of the ox-
idation product, hematein, and showed an
absorption maximum at 558 nm. The an-
aerobic spectrum, measured after 3 h of
incubation, was similar to the aerobic
spectrum obtained immediately following
dilution. The rate of autoxidation observed
in solutions bubbled with 100% Ozwas 2.49-
fold greater than that seen under ambient
oxygenation (data not shown). The absor-
bance of 22 PM hematein in 0.05 M potas-
 SUPEROXIDE DISMUTASE ASSAYS USING HEMATOXYLIN 331

Hematoxylin Hematein

FIG. 1. The structures of hematoxylin and its oxi-

dation product, hematein.

sium phosphate, 0.1 mM EDTA, pH 7.5, in-

dicated an ASGOextinction coefficient of
27,000 M-l cm-‘.
The autoxidation rate, estimated at 560
nm, was first order.with respect to hema- J
toxylin concentration (Fig. 3) and was pH 0 50 100 150 200 250
dependent (Fig. 4). At pH 6.55 the autoxi-
dation of 66 PM hematoxylin was nearly
undetectable but the rate steadily in- FIG. 3. The dependence of the increase in Am per
creased with increasing pH, and at pH 8.9 minute on hematoxylin concentration. All rates were
averaged over the first 4 min of reaction. Reactions
it was 0.293 AsBO/min.Hematoxylin autox-
were carried out at pH 7.5 in 0.05 M potassium phos-
idation at the low end of the pH range was phate buffer, 0.1 mM EDTA, as described under Ma-
easily observed by increasing the hema- terials and Methods.
toxylin concentration in the assay. For ex-
ample, at pH 7.2 a rate of 0.022 As&min
was obtained at a hematoxylin concentra- pH 8.05. Although initial rates of autoxi-
tion of 165PM, whereas 72 PM gave the same dation were slightly autocatalytic over the
reaction rate at pH 7.5. first 4 min of reaction (Fig. 5), SOD inhi-
Eflects of superoxide dismutase. SOD in-
hibited hematoxylin autoxidation below

0.6 I I

120 min
0.6 -

0.5 -
z
5
$ 0.4 -
a
4
0.3

0.2 .

0.1 .

0.0 ' '

400 450 500 550 600 650 6.00 7.00 8.00 9.00
Wavelength (nm) PH

FIG. 2. Repetitive scanning spectra of 22 PM he- FIG. 4. The pH dependence of hematoxylin autoxi-
matoxylin undergoing autoxidation in 0.05 M potas- dation. Hematoxylin (66 pM) was dissolved in 0.05 M
sium phosphate, 0.1 mM EDTA, pH 7.5, 25°C. Scans potassium phosphate buffer, 0.1 mM EDTA, as de-
were initiated at the indicated times. Line 1 represents scribed under Materials and Methods, and the rate of
the spectrum after 3 h of anaerobic incubation and autoxidation was measured as the increase in ASSOper
the aerobic spectrum at zero time. minute averaged over the first 4 min.
332 MARTIN, DAILEY, AND SUGARMAN

7.8 maximum inhibition of the autoxidation

1 by SOD was greater than 90%) both in the
presence and absence of 0.1 mM EDTA.
Hematoxylin autoxidation was stimulated
by Cu2+ and Mn3+ (Table I), and the rate
enhancements caused by these metal cat-
ions were not reversed by SOD addition.
Cu2+-EDTA and Mn3+-EDTA complexes
did not enhance the autoxidation rate.
Ferric iron inhibited the autoxidation re-
action, but this inhibition was eliminated
in buffer containing 0.1 ITIM EDTA. Maxi-
mum inhibition by SOD was greater than
90% in the presence of the three metal-
Minutes
EDTA complexes. Thus, free metal cations
catalyzed hematoxylin autoxidation by an
FIG. 5. The increase of Am during hematoxylin au- O;-independent pathway, and EDTA re-
toxidation and the effect of dialyzed E. coli B extracts directed the reaction to an O;-dependent
on the shape and slope of the reaction curve. The re-
pathway which is sensitive to SOD inhi-
actions were carried out at pH 7.5 as described in Fig.
2. The hematoxylin concentration was 72 pM and the
bition. Similar effects of EDTA have been
uninhibited autoxidation rate was 0.022 Admin. The observed in the autoxidation of epinephrine
amount of extract added to each assay is indicated in (9), hydroxylamaine (7), and pyrogallol(5).
the figure in terms of the protein content of the ex- Potential interferingfactors. Compounds
tract. commonly added to SOD assays include
cyanide, to inhibit CuZnSOD and cyto-
chrome oxidase (18), or sodium azide and
bition could be quantitated by comparing
the rates obtained over this interval in the
presence of varying concentrations of SOD.
The inhibitory effect of a dialyzed extract loo’
of E. coli B containing SOD is shown in
Fig. 5. Figure 6 shows that hematoxylin
autoxidation was progressively inhibited
by increasing concentrations of CuZnSOD
at pH 7.5 up to a maximum of 92%. When
the hematoxylin concentration was ad-
justed to yield an initial AAh60of O.OZ/min,
the reaction was inhibited 50% by the ad-
dition of 0.55 pg of CuZnSOD. This corre-
sponds to two units of SOD activity in the
xanthine oxidase-cytochrome c assay.
Thus, the hematoxylin assay is one-half as
sensitive as the cytochrome c method under
these conditions. The sensitivity of the he-
matoxylin assay could be varied by ad-
justing the hematoxylin concentration
(Fig. 7). For example, the assay was inhib- SOD units
ited 50% by one cytochrome c unit of SOD FIG. 6. Inhibition of hematoxylin autoxidation by
at an autoxidation rate of 0.015 ABBO/min. CuZnSOD. The SOD solution was precalibrated using
Transition metal eflects. Hematoxylin the xanthine oxidase-cytochrome c method (1). The
autoxidation was slightly faster in potas- uninhibited rate of hematoxylin autoxidation was
sium phosphate buffer without EDTA than 0.021 Admin. Reactions were carried out at pH 7.5
in the presence of 0.1 mM EDTA. Below pH as described in Fig. 2.
 SUPEROXIDE DISMUTASE ASSAYS USING HEMATOXYLIN 333

both ferricytochrome c and nitroblue tet-

A 56o/min * 1000
FIG. 7. The activity of CuZnSOD required to inhibit
hematoxylin autoxidation by 50% as a function of the
hematoxylin at&oxidation rate. The indicated amounts
of CuZnSOD were added to 3 ml of 0.05 M potassium
phosphate, 0.1 mM EDTA, pH 7.8. The hematoxylin
autoxidation rate was varied by adjusting the con-
centration of hematoxylin in the assay.
razolium at significant rates. In addition,
these reductants completely suppressed the
autoxidation of epinephrine at pH 10.2 and
pyrogallol at pH 8.2. Mercaptoethanol, 2
mM, and 0.2 mM NADH partially sup-
pressed epinephrine and pyrogallol autox-
idation. Thus, most SOD assays are sus-
ceptible to interferences by reductants.
Biological reductants can be removed from
crude cell extracts by dialysis. When ex-
tracts of E. coli B were dialyzed prior to
assay, the inhibition of hematoxylin au-
toxidation by the dialyzed extracts was de-
pendent on the amount of SOD activity
they contained (Fig. 8).
pH efects on SOD inhibition. Although
SOD inhibited hematoxylin autoxidation
more than 90% below pH 7.9, the observed
inhibition diminished precipitously be-
tween pH 7.9 and 8.05 (Fig. 9A). Above pH
TABLE I
THE EFFECTSOFTRANSITIONMETALSANDEDTA
METAL COMPLEXESONTHEAUTOXIDATION
OFHEMATOXYLIN

Relative
hydrogen peroxide to inhibit CuZnSOD and
rate of
iron SOD (19,20). Ethanol and ammonium autoxidation
sulfate are often added to SOD assays as Additons (WI
extract components during purification of
the enzyme (1, 18). Catalase can be added Hematoxylin control rate’ 100
to oxidation assays of pyrogallol, 6-hy- 8 units CuZnSOD 10
droxydopamine, and dianisidine to sup-
75 fiM MnClz 489
press enzymatic peroxidation of these sub- 125pM MnC& 1130
strates (5, 6,8, 12,13). None of these com- 125 PM MnCla + 25 units CuZnSOD 1000
pounds significantly influenced the rate of 66 yM Mns+-EDTA 99
hematoxylin autoxidation (Table II). How- 100 PM Mn3+-EDTA 99
ever, the intracellular reducing agents,
NADH, ascorbate, and reduced glutathi- 50 /LMCUSO~ 200
150 /AM CUSO, 425
one, and the reducing agent, dithiothreitol, 150 PM CuSO, + 25 units CuZnSOD 470
completely inhibited the autoxidation re- 75 fiM CL?+-EDTA 99
action, as did a crude undialyzed extract of
E. coli B. The effects of reducing agents 100 pM FeCla 75
were also examined in other assays of SOD 100 FM Fea’-EDTA 100
including those employing 10 PM ferricy-
a The control rate was determined by addition of 40
tochrome c or 1 mM nitroblue tetrazolium FM hematoxyIin to 0.05 M potassium phosphate, pH
as indicating scavengers (l-3,7) and in the 7.8,25’C, in a 3-ml assay. Under these conditions the
epinephrine (9,lO) and pyrogallol(56) au- autoxidation rate was 0.020 A&min. Additions of
toxidation assays (data not shown). Re- metals were made prior to the addition of hematox-
duced glutathione, 0.5 mM, 66 j&M ascorbate ylin. Metal concentrations given in the table represent
and 166 I.LM dithiothreitol directly reduced final assay concentrations.
334 MARTIN, DAILEY, AND SUGARMAN

TABLE II

THE EFFECTS OFPOTENTIALINTERFERINGCOM-

POUNDSONTHERATEOFHEMATOXYLIN
AUTOXIDATION

Relative
rate of
autoxidation
Additions (%I

Hematoxylin control rate’ 100

8 units CuZnSOD 10
10 mM sodium azide 100
1 mM potassium cyanide 95
1 mM ammonium chloride 98
350 mM ethanol 107
0 20 40 60 80 100 120 140
60 units catalase 106
60 PM hydrogen peroxide 103 Proteiqug
2 mM hydrogen peroxide 117
FIG. 8. The dependence of inhibition of hematoxylin
0.2 mM reduced glutathione 0
autoxidation on the concentration of dialyzed cell ex-
0.2 mM ascorbate 0
tract. E. coli B was grown either in TSY medium (H)
1.0 mM dithiothreitol 0
or in glucose minimal medium supplemented with 100
0.5 mM NADH 0
PM Fe’+ and 100 j&M Mna+ (0) as described under Ma-
terials and Methods. Extracts were added to a 3-ml
a Assay conditions were identical to those described
assay volume in the amounts indicated. Reactions
in Table I. The autoxidation rate was 0.02 &,,/min.
were performed at pH 7.5 at an uninhibited autoxi-
Concentrations given in the table represent final assay
dation rate of 0.02 AW/min. The specific activities of
concentrations.
SOD in the extracts, determined by the xanthine ox-
idase-cytochrome c assay (l), were 21 units/mg (m)
and 53 units/mg (0).
8.05 SOD actually enhanced the autoxida-
tion rate (Fig. 9B), and this enhancement
was maximal at pH 8.3. The degree of max- thine oxidase acting upon 50 PM xanthine
imum rate enhancement was dependent on (data not shown). The abrupt transition
the level of SOD (Fig. 10). Approximately from SOD rate inhibition to rate stimu-
34 cytochrome c units of CuZnSOD were lation over a pH range of 0.3 pH units is
required to double the reaction rate, and a probably due to the ionization of hematox-
maximum enhancement of 5.3-fold was ob-
tained upon addition of 480 SOD units to
the reaction. CuZnSOD, inactivated by B
boiling, had no effect on the autoxidation 300

rate at pH 8.3. Reaction rate acceleration

was proportional to SOD concentration up 200
to approximately 200% (Fig. 10, inset).
Similar rate enhancements by SOD have
been observed during the autoxidation of 100
the antitumor antibiotic, 9-hydroxyellip-
ticene (21) and the enzymatic (13) and pho-
0 ,:
tochemical (12) oxidation of dianisidine. :.5 7.0 7.5 8.0 8.5 8.0 8.2 8.4 8.6 0.8 9.0
Inhibition of hematoxylin autoxidation PH PH

at low pH by SOD indicates that 0, par-
ticipates in the autoxidation as an oxidizing
species. In fact, autoxidation of 72 PM he-
matoxylin was accelerated 50% by 0; in-
troduced through the action of 6 nM xan-
FIG. 9. The effect of CuZnSOD on the hematoxylin
autoxidation rate as a function of pH. 80 cytochrome
c units of CuZnSOD were added to reactions contain-
ing 66 pM hematoxylin. Other assay conditions were
as described under Materials and Methods.
 SUPEROXIDE DISMUTASE ASSAYS USING HEMATOXYLIN 335

At high pH:

HTH- + Me” --f HTH’ + Men-’ 181

Men-’ + Oz--, Me” + 0; [31
HTH- + O2--* HTH’ + 0, PI
HTH’+ 0, + HTH- + Oz WI
HTH’+HTH’-*HTH-+HT+H+ [ll]
HTH’+O,-*HT+O;+H+ WI
2H++0,+0;+HZOZ+02 PI

Trace metal contaminants, chelated by

EDTA, may initiate autoxidation at both
low and high pH in reactions [2] and [8].
Alternatively, hematoxylin or its anion
may directly reduce oxygen to form 0; (re-
FIG. 10. The enhancement of hematoxylin autoxi- actions [4] and [9]). Superoxide formed in
dation by CuZnSOD. CuZnSOD was added to reaction reactions [3] and [4] may both initiate and
assays containing 66 ELM hematoxylin at pH 8.3. Other propagate the autoxidation reaction (re-
assay conditions were as described under Materials actions [5] and [6]) at low pH. At high pH,
and Methods. The reaction rate obtained after the 0, reduces the hematoxylin semiquinone
addition of 480 units of SOD inactivated by boiling in reaction [lo], and dismutation of the or-
(+). Inset: activation of hematoxylin autoxidation by ganic radicals plays a significant role in the
up to 28 units of SOD.
autoxidation (reaction [ll]). SOD acceler-
ates the autoxidation rate by preventing
ylin’s oxidizable hydroxyl group at position reaction [lo] and by pulling reactions [9]
9. The rate enhancement by SOD at high and [12] to the right. The high concentra-
pH suggests that 0; reduces an interme- tion of SOD required to accelerate the au-
toxidation above pH 8.1 suggests an excep-
diate of the autoxidation reaction, thereby
preventing the formation of hematein. The tionally high reaction rate between 0; and
following two reaction schemes are pro- the hematoxylin semiquinoid intermediate.
posed to account for the foregoing obser-
vations. In these schemes, HTHz and HT DISCUSSION
represent hematoxylin and hematein re-
spectively. Me” and Men-’ represent oxi- Hematoxylin autoxidation forms the
dized and reduced transition metals che- basis for two simple and convenient assays
lated by EDTA. for SOD activity which complement exist-
At low pH: ing methods. The reaction does not require
purified proteins such as horseradish per-
HTHz + Me” --) HTH’ + Men-’ + H+ [2] oxidase, xanthine oxidase, catalase, or cy-
Men-’ + O2+Me”+O, [31 tochrome c. Hematein, the immediate ox-
idation product, possesses a bright red
HTHz + O2+ HTH’ + 0; [41 chromophore with a well defined absorp-
tion maximum. Although hematein even-
H+ + HTHz + 0; + HTH’+ HzOz PI tually decomposesinto secondary products,
H++HTH’+O;-+HT+H,O, the time course for its disappearance is
161 slow by comparison to its rate of formation
2H++0;+O;-+H202+02 m through autoxidation. Other autoxidation
336 MARTIN, DAILEY,
assays (e.g., epinephrine and pyrogallol)
generate multiple, relatively unstable in-
termediates which exhibit broad visible
absorption maxima. The sensitivity of the
hematoxylin method may be adjusted sim-
ply by changing the concentration of he-
matoxylin in the assay. The negative assay
is useful between pH 6.8 and 7.8, and the
maximum inhibition that may be obtained
with SOD is constant and greater than 90%
within this range. The 6-hydroxydopamine
assay is the only other autoxidation
method that is effective in this pH range.
Unlike 6-hydroxydopamine, a neurotoxin,
pyrogallol, and hydroxylamine, hematox-
ylin is not toxic. Moreover, it is widely
available, chemically stable at room tem-
perature, and inexpensive. Hematoxylin
autoxidation is not inhibited by compo-
nents frequently added to SOD assays in
order to distinguish among the three types
of SOD. Although the autoxidation is in-
hibited by reducing agents, this problem is
shared by other autoxidation assays and
by assay methods which employ ferricy-
tochrome c and nitroblue tetrazolium as
indicating scavengers. This difficulty may
be avoided by dialyzing protein extracts
before assaying them. Hematoxylin autox-
idation provides a sensitive negative assay
at low pH, and at high pH it can be used
as a positive assay for SOD activity in
which the autoxidation rate is stimulated
in proportion to the level of SOD. The pos-
itive method is an order of magnitude less
sensitive than the positive assays based on
dianisidine oxidation but it may be usefully
applied in situations in which the SOD
concentration is high, such as are encoun-
tered during purification of SOD.
AND SUGARMAN
REFERENCES
1. MCCORD, J. M., AND FRIDOVICH, I. (1969) J. Biol.
Chem. 244,6049-6055.
2. HYLAND, K., VOISIN, E., BANOUN, H., AND AUCLAIR,
C. (1983) Anal B&hem. 135,280-287.
3. BEAUCHAMP, C., AND FRIDOVICH, I. (19’71) Anal.
Biochem. 44,276-287.
4. ELSTNER, E., AND HEUPEL, A. (1976) Anal.
Biochem. 70,616-620.
5. MARKLUND, S., AND MARKLUND, G. (1974) Eur. .I
B&hem. 47,469-474.
6. MARKLUND, S. (1985) in CRC Handbook of Oxygen
Radical Research (Greenwald R., Ed.), pp. 243-
247, CRC Press, Boca Raton, FL.
7. KONO, Y. (1978) Arch. Biochem. Biophys. 186,189-
195.
8. HEIKKELA, R. E., AND CABBAT, F. S. (1976) Anal.
Biochem. 75,356-362.
9. MISRA, H. P., AND FRIDOVICH, I. (1972) J. Biol.
Chem. 247.3170-3175.
10. SUN, M., AND ZIGMAN, S. (1981) Anal. B&hem. 90,
81-89.
11. FORMAN, H. J., AND FRIDOVICH, I. (1973) Arch.
Biochem. Biophys. 159,396-400.
12. MISRA, H. P., AND FRIDOVICH, I. (1977) Arch.
Biochem Biophys. l&31,308-312.
13. MISRA, H. P., AND FRIDOVICH, I. (1977) Anal.
Biochem. 79,553-560.
14. PUGH, S. Y. R., AND FRIDOVICH, I. (1985) J. Bac-
teriol. 162, 196-202.
15. WAUD, W. O., BRADY, F. O., WILEY, R. D., AND
RAJAGOPALAN, K. V. (1975) Arch. B&hem.
Biophys. 169,695-701.
16. BRADFORD, M. (1976) Anal Biochem. 72,248~255.
17. HODGSON, E. K., MCCORD, J. M., AND FRIDOVICH,
I. (1973) Anal Biochem. 5,470-473.
18. WEISIGER, R. A., AND FRIDOVICH, I. (1973) J. Biol.
Chem, 248,3582-3592.
19. HODGSON, E. K., AND FRIDOVICH, I. (1975) Bio-
chemistry 14,5294-5299.
20. MISRA, H. P., AND FRIDOVICH, I. (1978) Arch.
Biochem Biophys. 189,317-322.
21. AUCLAIR, C., HYLAND, K., AND PAOLETTI, C. (1983)
J Med. Chem. 26,1438-1444.