ANALYTICAL BIOCHEMISTRY 174,33 l-336 (1988)

A Spectrophotometric Method for Determination of Catalase Activity in

Small Tissue Samples

LARS H. JOHANSSON AND L. A. H&AN BORG'

Department of Medical Ce// Biology. University of Uppsala, Box 5 71, S-75 I 23 Uppsala, Sweden

Received February 16, 1988

A simple and rapid method for determination of catalase activity in small tissue samples is
described. Using a new approach, we have exploited the peroxidatic function of catalase for the
determination of enzyme activity. The method was based on the reaction of the enzyme with
methanol in the presence of an optimal concentration of hydrogen peroxide. The formaldehyde
produced was measured spectrophotometrically with 4-amino-3-hydrazino-5-mercapto- 1,2,4-
triazole (Purpald) as a chromogen. With this method, a detection limit of 12.5 ng of purified
catalase from bovine liver was possible, and it was successfully applied to microgram amounts
of mouse liver and pancreatic islet homogenates. The catalase activity in liver was about 50
times higher than that in pancreatic islets. 0 1988 Academic Press. Inc.
KEY WORDS: catalase; spectrophotometry; Purpald; liver; pancreatic islets.

Catalase (EC 1.11.1.6) is able to decom-
pose hydrogen peroxide by two types of reac-
tions. Both reactions include a first step of
formation of an intermediate (compound I)
consisting of the enzyme and hydrogen per-
oxide (1). The catalatic activity, which is a
unique property of catalase, catalyzes a reac-
tion with a second molecule of hydrogen per-
oxide, resulting in the production of oxygen
and water. The peroxidatic activity of the en-
zyme, first described by Keilin and Hartree
(2,3), catalyzes reactions of compound I with
hydrogen donors other than hydrogen perox-
ide. Details of enzyme structure, reaction
mechanisms, kinetic characteristics, and the
physiological role of catalase have been the
subject of numerous reviews (4-12). Obvi-
ously, the peroxidatic activity of catalase is
most evident at relatively low concentrations
of hydrogen peroxide (3,13), and a number of
different hydrogen donors may serve as sub-
strates. Lower alcohols, such as methanol and
ethanol, are particularly reactive (1,3). While
the aliphatic alcohols serve as specific sub-
’ To whom correspondence should be addressed.
strates for catalase, other enzymes with per-
oxidatic activity do not utilize these sub-
strates ( 14).
Current methods for quantitative determi-
nation of catalase activity are based on the
catalatic function of the enzyme. Such meth-
ods involve measurements of either hydrogen
peroxide consumption by titrimetry or spec-
trophotometry or of oxygen production by
manometric techniques or by oxygen-sensi-
tive electrodes. The latter methods are partic-
ularly sensitive but require combersome pro-
cedures or sophisticated equipment. In a new
approach, we have evaluated the possibility
of utilizing the peroxidatic function of cata-
lase for determination of enzyme activity in
small tissue samples. We have used methanol
as the hydrogen donor and measured the pro-
duction of formaldehyde spectrophotometri-
tally with 4-amino-3-hydrazino-5-mercapto-
1,2,4-triazole (Purpald) as a chromogen.
Previous observations indicate that this sub-
stance specifically forms a bicyclic heterocy-
cle with aldehydes, which at oxidation
changes from colorless to a magenta or purple
color ( 15).

331 0003-2697188 $3.00

Copyright 0 1988 by Academic Press, Inc.
All tights of reproduction in any form reserved.
332 JOHANSSON AND BORG

MATERIALS AND METHODS formaldehyde and Purpald was oxidized by

adding 50 ~165.2 mM potassium periodate in
Chemicals. Purified catalase (EC 1. I 1.16) 470 mM potassium hydroxide to each tube.
from bovine liver, 2,2’-azinobis(3-ethylbenz- Any particulate material in the tubes was sed-
thiazoline-6-sulfonic acid) (ABTS), diam- imented by centrifugation at 9500g for 10
monium salt, and 4,4’-dicarboxy-2,2’-bi- min, and using glass standard microcuvettes
quinoline, disodium salt, were purchased of 2-mm width and 1O-mm light path, the ab-
from Sigma Chemical Company, St. Louis, sorbance of the clear liquid was measured at
Missouri; 4-amino-3-hydrazino-5-mercapto- 550 nm in a Zeiss PMQ II spectrophotometer
1,2,4-triazole (Purpald) from Aldrich-Chemie, (Carl Zeiss, Oberkochen, FRG).
Steinheim, FRG; collagenase (EC 3.4.24.3) Preparation of tissue samples.Tissue sam-
from Boehringer-Mannheim, Mannheim, ples were obtained from adult male NMRI
FRG; and bovine serum albumin, fraction V, mice (ALAB, Sollentuna, Sweden). The mice
from Miles Laboratories, Slough, UK. All were killed by cervical dislocation. After
other chemicals, of analytical grade, were decapitation and bleeding the livers were
from E. Merck, Darmstadt, FRG. quickly excised and minced in ice-cold 25
Procedurefor determination of catalase ac- mM KI-12P04-NaOH buffer, pH 7.0, and ap-
tivity. Purified catalase or homogenized tis- proximately 20% (w/v) homogenates were
sues were incubated with methanol and hy- prepared. The homogenates were then di-
drogen peroxide in 250 mM KH2P04-NaOH luted with buffer as appropriate. Pancreatic
buffer, pH 7.0. By the standard procedure, 50 islets were isolated by a collagenase (EC
~1 buffer, 50 ~1 100% (w/v) methanol, giving 3.4.24.3) digestion technique ( 16). Batches of
a final concentration of 5.9 M, and 10 ~1 400-2500 islets were homogenized on ice in
0.27% (w/v) hydrogen peroxide, giving a final 500 ~1 25 mM KH2P04-NaOH buffer, pH
concentration of 4.2 IIIM, were mixed in 7.0. The islet protein content was 36.2 + 0.10
small polystyrene test tubes. The enzymatic pg/ 100 islets (mean + SE; four observations).
reaction was initiated by addition of 100 ~1 of Homogenization was performed in aground-
a catalase-containing sample. For evaluation glass tissue homogenizer (Kontes Glass Co.,
of methodological details, the sample con- Vineland, NJ) of 8-ml capacity for the livers
sisted of 0.125-2.0 pg/ml purified catalase and of 2-ml capacity for the islets. To liberate
from bovine liver, with an activity of 2500 catalase from subcellular particles all homo-
units/mg, dissolved in 25 InM KH2P04- genates were treated in an ultrasonic ice bath
NaOH buffer, pH 7.0. Tissue samples were (35 kHz; 95 W) for 15 s. Extreme care was
prepared as described below. Standard sam- taken to avoid inactivation of catalase by
ples consisted of 26.6- 133 pM formaldehyde overheating the homogenates.
solutions in 25 mM KI-12P04-NaOH buffer, Determination of hemoglobin. Since eryth-
pH 7.0. Pure buffer was used as blanks. The rocytes contain catalase, the presence of re-
reaction mixture was incubated with contin- sidual blood in the tissue preparations was
uous shaking for exactly 20 min at room tem- taken into account. The content of hemoglo-
perature (20°C). The enzymatic reaction was bin in the tissue homogenates was deter-
terminated by addition of 50 ~1 7.8 M potas- mined by the method described by Marklund
sium hydroxide solution to each tube. Imme- (17) and the catalase activity originating
diately thereafter the tubes were supplied from erythrocytes was calculated from mea-
with 100 ~1 each of 34.2 mrvr Purpald in 480 surements performed on mouse blood. The
mM hydrochloric acid, and a second incuba- liver homogenates contained 1250 + 142 ng/
tion with continuous shaking was performed ~1 (mean * SE; three observations) hemoglo-
for 10 min at 20°C. To obtain a colored com- bin, which corresponded to 0.013 + 0.002%
pound, the product of the reaction between (mean f SE; three observations) of their total
 SPECTROPHOTOMETRIC DETERMINATION OF CATALASE ACTIVITY 333

gated system, which probably causes the dis-

FIG. I. Absorbance as a function of the time for reac-
tion between formaldehyde and Purpald. Oxidizing re-
agent was added at different time points. Samples con-
sisted of I &ml purified catalase from bovine liver and
were incubated with 5.9 M methanol and 4.2 mM hydro-
gen peroxide. Means f SE of 20 observations.
catalase activity. The hemoglobin content of
the pancreatic islet homogenates was, how-
ever, below the detection limit of the hemo-
globin assay, which was 10 ng/hl. For the liver
samples, the enzyme activity of the erythro-
cytes was subtracted from the total catalase
activity.
Determination ofprotein. The protein con-
tent of the tissue preparations was deter-
mined by the method of Smith et al. ( 18) us-
ing 4,4’-dicarboxy-2,2’-biquinoline, or 2,2’-
bicinchoninic acid, to enhance the sensitivity
of the biuret reaction. Protein standards con-
sisted of bovine serum albumin.
tinct light absorption in the visible region
( 15). Potassium periodate was used as an oxi-
dizing agent, and full absorbance was devel-
oped almost instantaneously. The colored
product was stable for at least 3 h. By varying
the length of the second incubation step,
when the reaction between formaldehyde
and Purpald took place, we found that the re-
action was completed within 10 min (Fig. l),
and this duration of the incubation was used
in all subsequent experiments. By employing
standard samples containing 2.66- 13.3 nmol
formaldehyde a linear relationship between
the amount of aldehyde and absorbance was
found (Fig. 2). In further experiments, stan-
dard curves, similar to that shown in Fig. 2,
were used to determine the amount of form-
aldehyde formed by catalase.
The catalatic and peroxidatic activities of
catalase are not independent of each other.
Oshino et al. (13) have shown that the rela-
tionship between the two activities of catalase
is a function of the ratio of hydrogen peroxide
and other hydrogen donor concentration.
From their experiments, it also appeared that
neither activity was saturable. Indeed, when
incubating purified catalase from bovine liver
with various concentrations of methanol at a
fixed hydrogen peroxide concentration of 4.2

RESULTS AND DISCUSSION

The present approach to determine cata-
lase activity utilized the peroxidatic function
of the enzyme. Methanol was used as hydro-
gen donor, and Purpald was applied as a re-
agent for formaldehyde produced by the en-
zymatic activity. Purpald reacts very specifi-
cally with aldehydes in alkaline environment
(15). Both the hydrazine and the amino
group of Purpald bind to the aldehyde and
FIG. 2. Absorbance as a function of the formaldehyde
an aminal is formed. Subsequent oxidation of content in standard samples. Samples were incubated
the aminal leads to the formation ofa bicyclic with 5.9 M methanol and 4.2 mM hydrogen peroxide.
heterocycle containing an extensive conju- Means + SE of 20 observations.
334 JOHANSSON
FIG. 3. Relationship between formaldehyde produc-
tion and concentration of hydrogen peroxide. Samples
consisted of I &ml purified catalase from bovine liver
and were incubated with 5.9 M methanol and hydrogen
peroxide as indicated. Means k SE of 20 observations.
mM, we found a linear relationship between
methanol concentration and formaldehyde
production up to 6 M methanol. At even
higher concentrations of methanol there was,
however, a decreased peroxidatic activity of
the enzyme, and this activity was almost
completely inhibited at 7.5 M methanol. This
inhibition of the enzymatic activity probably
resulted from unspecific changes of the en-
zyme in the concentrated methanolic solu-
tion. Variations in hydrogen peroxide con-
centration at a fixed concentration of metha-
nol of 5.9 M had significant effects on the
peroxidatic oxidation of the alcohol (Fig. 3).
At concentrations of hydrogen peroxide up to
2.3 mM there was a very rapid increase in the
peroxidatic activity. Between 2.3 and 4.7 mM
hydrogen peroxide there was little change in
this activity, whereas at higher concentra-
tions of hydrogen peroxide it decreased grad-
ually. The rapid enhancement of the peroxi-
datic activity at increasing hydrogen peroxide
concentrations up to 2.3 mM could be ex-
plained by the production of increasing
steady-state concentrations of compound I.
The enzymatic activity would then reflect the
binding of hydrogen peroxide to the enzyme.
It is known that maximal binding is reached
when 30-40% of catalase hematin is occupied
by hydrogen peroxide (19-21). Under the
present conditions, this apparently occurred
AND BORG
at a concentration of about 2 mM hydrogen
peroxide. The decrease in peroxidatic activ-
ity, when the hydrogen peroxide concentra-
tion was increased above 4.7 mM, would de-
pend on an increasing proportion of catalatic
activity (13). Optimal conditions for utilizing
the peroxidatic function of catalase for deter-
mination of enzyme activity were thus found
at concentrations of methanol and hydrogen
peroxide of 5.9 M and 4.2 mM, respectively.
At these concentrations of methanol and
hydrogen peroxide there was an essentially
constant velocity of formaldehyde produc-
tion for at least 20 min, when purified cata-
lase from bovine liver was incubated at room
temperature (20°C) (Fig. 4). The enzyme ac-
tivity gradually decreased, however, when the
incubation was performed at 37°C. Indeed,
incubations of purified catalase performed
before the determination of enzyme activity
showed that in buffer the enzyme rapidly lost
activity at both 20 and 37°C (Table 1). How-
ever, at 0°C the activity remained constant
for at least 60 min. When 4.2 mM hydrogen
peroxide was added to the incubation buffer,
the decrease in enzyme activity was also more
pronounced at 0°C (Table 1). These findings
emphasize the extreme importance of main-
taining catalase-containing samples at low
temperature. In particular, heating during ul-
; 51 /
FIG. 4. Relationship between formaldehyde produc-
tion and incubation time at two different temperatures.
Samples consisted of 1 &ml purified catalase from bo-
vine liver and were incubated with 5.9 M methanol and
4.2 mM hydrogen peroxide. Means f SE of 20 observa-
tions.
 SPECTROPHOTOMETRIC DETERMINATION OF CATALASE ACTIVITY 335

TABLE 1
STABILITYOFCATALASEINSOLUTION'

Catalase activityb

Without hydrogen peroxide With hydrogen peroxide

Incubation time
(mm) 0°C 20°C 37°C 0-Z 2o’c 37°C

0 100 + 2.3 100 + 1.7 100 f 1.0 100 z!z2.3 100 * 1.7 lOOk 0.5
10 98 f 2.4 43 i- 1.8 17a0.8 94 rtr 1.5 19rt 1.0 13kO.5
30 100 + 1.7 2 z+z0.2 ND’ 73 ?I 0.9 ND ND
60 97 * 1.2 ND ND 52 +- 0.5 ND ND

’ Solutions of I pg/ml purified catalase from bovine liver in 25 IrIM KH2P04-NaOH buffer, pH 7.0, were incubated
in the absence and presence of 4.2 mM hydrogen peroxide at three different temperatures, as indicated. The catalase
activity was determined after the incubations.
b Activity is expressed as percentage of the activity at time zero of incubation. Means ~frSE of 10 observations.
’ No detectable activity.

trasonic disruption of tissue samples should
be avoided. Our findings also demonstrate an
inhibitory effect of hydrogen peroxide on cat-
alase activity, which can occur in tissue ho-
mogenates even at 0°C. This effect can be as-
cribed to conversion of the primary catalase-
hydrogen peroxide intermediate into an inac-
tive form (compound II), which is relatively
stable (22). In contrast to hydrogen peroxide,
aliphatic alcohols are electron donors that do
not promote formation of compound II (23).
Moreover, compound II is decomposed by
such alcohols and catalase activity restored
(22). Methanol is therefore effective both in
prevention of the formation of compound II
and in the decomposition of compound II al-
ready formed (24). High concentrations of
methanol should thus minimize problems
with the presence of compound II during de-
termination of catalase activity. As shown in
Fig. 4, we virtually avoided such problems,
when the enzyme reaction was carried out at
20°C. The gradual decrease in enzyme activ-
ity during incubation in the presence of
methanol at 37°C probably resulted from de-
composition of the enzyme in solution.
The dose-response relationship for puri-
fied catalase from bovine liver was strictly lin-
ear from 12.5 to at least 200 ng of enzyme
sample (Fig. 5A). Homogenates of mouse

0 IO 100 150 200 0 1 2 3 4 0 50 ‘DO IS0 2m

Cslalow In*> !“ro,*in ,po PImel” $0,

FIG. 5. Relationship between formaldehyde production and amounts of purified catalase from bovine
liver (A; means f SE of 20 observations), mouse liver homogenate (B; 9 observations), and mouse pancre-
atic islet homogenate (C, 12 observations). Samples were incubated with 5.9 M methanol and 4.2 mM
hydrogen peroxide.
336 JOHANSSON
liver (Fig. 5B) and pancreatic islets (Fig. 5C)
also displayed a linear relationship between
the amount of tissue and catalase activity
over a wide range. The activity was 2.42
+- 0.060 pkat/pg protein (mean -+ SE; three
observations) in liver and 0.049 +- 0.002
pkat/pg protein (mean + SE; four observa-
tions) in pancreatic islets. Hence, mouse liver
contained about 50 times higher catalase ac-
tivity than mouse pancreatic islets. This find-
ing confirms other observations (25).
In separate experiments, we observed that
methanol under the present conditions was
not oxidized to formaldehyde, when mouse
liver homogenates were incubated in the ab-
sence of hydrogen peroxide. This observation
is in good agreement with findings of no (26)
or very low (27) liver alcohol dehydrogenase
(EC 1.1.1.1) activity with methanol as sub-
strate. Furthermore, when 133 PM formalde-
hyde was added to samples of mouse liver ho-
mogenates in the presence of either hydrogen
peroxide or methanol, the formaldehyde con-
centration remained essentially constant dur-
ing prolonged incubations. Thus, it is un-
likely that other enzyme activities in tissue
homogenates affect the results obtained by
the present method.
In conclusion, we have found that the per-
oxidatic function of catalase is very well
suited to the determination of the enzyme ac-
tivity in small tissue samples. Moreover, the
determination can be performed by simple
methodology and does not require special-
ized equipment.
ACKNOWLEDGMENTS
This work was supported by grants from the Swedish
Diabetes Association, the Family Emfors Fund, the Swe-
dish Hoechst Diabetes Fund, and the Swedish Medical
Research Council (12X-109, 12X-6538).
REFERENCES
1. Chance, B. (1947) Acfa Chem. &and. 1,236-267.
2. Keilin, D., and Hartree, E. F. (1936) Proc. R. Sot.
London Sect. B 119,14 I- 159.
AND BORG
3. Keilin, D., and Hartree, E. F. (1945) Biochem. J. 39,
293-301.
4. Nicholls, P., and Schonbaum, G. R. (1963) in The
Enzymes (Boyer, P. D., Lardy, H., and Myrbkk,
K., Eds.), 2nd ed., Vol. 8, Part B, pp. 147-225,
Academic Press, New York/London.
5. Brill, A. S. (1966) in Comprehensive Biochemistry
(Florkin, M., and Stotz. E. H., Eds.), Vol. 14, pp.
447-479, Elsevier, Amsterdam/London/New
York.
6. Jones, P., and Suggett, A. (1968) Biochem. J. 110,
621-629.
7. Deisseroth, A., and Dounce, A. L. (1970) Physiol.
Rev. SO,3 19-375.
8. Schonbaum. G. R., and Chance, B. (1976) in The
Enzymes (Boyer. P. D., Ed.), 3rd ed.. Vol. 13, Part
C, pp. 363-408, Academic Press, New York/Lon-
don.
9. Jones, P., and Dunford. H. B. (1977) J. Theor. Biol.
69,451-470.
10. Chance, B.. Sies, H., and Boveris, A. (1979) Physiol.
Rev. 59,527-605.
11. Dounce, A. L. (1983) J. Theor. Biol. 105,553-567.
12. Fita, I., and Rossmann, M. G. (1985) J. Mol. Biol.
185,21-37.
13. Oshino, N., Oshino, R., and Chance, B. (1973) Bio-
them. J. 131,555-563.
14. Mason, H. S. (1957) Adv. Enzymol. 19,79-233.
15. Dickinson, R. G., and Jacobsen, N. W. (1970)
Chem. Commun., 17 19- 1720.
16. Schnell, A. H., Swenne, I., and Borg, L. A. H. (1988)
Cell Tissue Res. 252,9- 15.
17. Marklund, S. (1979) Clin. Chim. Acta 92,229-234.
18. Smith, P. K., Krohn, R. I., Hermanson, G. T., Mal-
lia, A. K.. Gartner, F. H., Provenzano, M. D., Fu-
jimoto, E. K., Goeke, N. M., Olson, B. J., and
Klenk, D. C. (1985) Anal. Biochem. 150,76-85.
19. Kremer, M. L. (1970) Biochim. Biophys. Acfa 198,
199-209.
20. Chance, B., and Oshino, N. (197 1) Biochem. J. 122,
225-233.
21. Sies, H.. Biicher, T., Oshino, N., and Chance, B.
(1973)Arch. Biochem. Biophys. 154, 106-I 16.
73 Chance, B. (1950) Biochem. J. 46,387-402.
AL.
23. Keilin, D., and Nicholls, P. (1958) Biochim. Bio-
phys. Acta 29,302-307.
24. Adams, D. H.. and Burgess, E. A. (1959) Enzy-
mologia 20,34 I-354.
25. Grankvist, K., Marklund, S. L., and Tiiljedal, I.-B.
(1981) Biochem. J. 199,393-398.
26. Theorell, H., and Bonnichsen, R. (195 1) Acta Chem.
Stand. 5,1105-l 126.
27. Kini, M. M., and Cooper, J. R. (1961) Biochem.
Pharmacol. 8,207-2 15.