Role of Oxygen Radicals in Experimental
Allergic Uveitis
Narsing A. Rao,* Alex 5evanian,f Mary Ann 5. Fernandez,* Jorge L. Romero,* Jean-Pierre Faure4
Yvonne de Kozak.J Gerd O. Till,§ and George E. Marak, Jr."
The experimental autoimmune disease elicited by a large dose of retinal S antigen in guinea pigs is
characterized by massive necrotizing uveitis and retinitis. Treatment of these animals with the antioxidants
superoxide dismutase, catalase, and sodium benzoate resulted in marked reduction of uveal inflammation.
The attenuated inflammation was characterized by a relatively well-preserved retina and retinal pigment
epithelium, along with a reduction of subretinal exudate and vitreous inflammation. These findings
suggest that reactive oxygen metabolites may play a role in the destruction of ocular tissue and amplification of the inflammatory process in experimental uveitis. Invest Ophthalmol Vis Sci
28:886-892,1987
The authors previously reported considerable differences in the experimental autoimmune disease elicited by retinal S antigen.12 A massive necrotizing intraocular inflammation was produced by an inoculum
of 50 Mg of S antigen, although a mild granulomatous
panuveitis was induced by injecting 5-10 jig of the
antigen. In this article, the authors will report the modulating effects of superoxide dismutase (SOD), catalase
and benzoic acid on severe S antigen-induced ocular
inflammation.
The damaging effects of inflammation are primarily
mediated through chemical substances released by the
infiltrating polymorphonuclear leukocytes (PMNs) or
monocytes. These chemical mediators include neutral
proteases, elastases, cathepsin G, arachidonic acid metabolites, monokines and others.1 Recent studies of the
mechanisms of phagocyte-mediated destruction of microbes have revealed a significant increase in oxygen
consumption by these inflammatory cells that is associated with the generation of reactive oxygen products. The role of these reactive oxygen metabolites in
antimicrobial defense mechanisms has been the subject
of intensive investigation. Recently, attention has been
focused on the potential significance of these metabolites in the inflammatory destruction of cellular components. 23
The primary oxygen metabolites include univalent
and divalent products of oxygen, superoxide, and hydrogen peroxide.2'4'5 The superoxide is the principal
product of the respiratory burst of stimulated neutrophils and mononuclear phagocytes.6 Many of the cytotoxic effects of the superoxide are believed to result
from the formation of more reactive species that include hydrogen peroxide and hydroxyl radicals.2'7"11
Materials and Methods
The investigations described in this study conform
to the ARVO Resolution on the Use of Animals in
Research.
Thirty-six female Hartley strain adult guinea pigs
were given a single hind footpad injection of purified
bovine soluble retinal antigen in complete Freund's
adjuvant. The authors' method of isolation and purification of the S antigen was similar to the method
described by Dorey et al.13 Each animal received 50
tig of the antigen; all were examined daily for limbal
injection, anterior chamber reaction, and vitreous exudate. The guinea pigs were divided into six groups of
six animals each and treated as follows. (1) Group 1
animals received daily intraperitoneal (IP) injections
of 0.5 ml of normal saline. (2) Group 2 animals received
daily IP injections of 2,000 units of polyethylene-glycol-modified superoxide dismutase (PEG-SOD) (Enzon, Inc.; Piscataway, NJ). (3) Group 3 animals received similar daily IP injections of heat-inactivated
polyethylene-glycol-modified SOD, (PEG-SOD, 2,000
units). (4) Group 4 animals received IP injections of
polyethylene glycol-modified catalase, (PEG-catalase),
From the *Estelle Doheny Eye Foundation, and fDepartments of
Ophthalmology and Pathology, Institute for Toxicology, University
of Southern California, Los Angeles, California; the JUnite'de Recherche D'Ophtalmologie, Paris, France; §University of Michigan,
Ann Arbor, Michigan; and "Georgetown University, Washington,
D.C.
Supported in part by research grant EY 05662 and EY 03040 from
the National Eye Institute, National Institutes of Health, Bethesda,
Maryland.
Submitted for publication: July 24, 1985.
Reprint requests: Narsing A. Rao, MD, Estelle Doheny Eye Foundation, 1355 San Pablo Street, Los Angeles, CA 90033.
886
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
No. 5
OXYGEN RADICALS IN UVEITIS / Rao er al.
2,400 units/kg (Enzon, Inc.; Piscataway, NJ). (5) Group
5 animals received daily IP injections of heat-inactivated polyethylene glycol-modified catalase (PEG-catalase, 2,400 units/kg). (6) Group 6 animals received
daily IP injections of sodium benzoate, 100 mg/kg
(Aldrich Chemical Co.; Milwaukee, WI).
The IP administrations were begun on the tenth day
after the injection of S antigen, and all animals were
killed on day 21. The eyes were enucleated, fixed in
formalin, then processed for paraffin embedding and
hematoxylin-eosin (H-E) staining.
The eyes were coded to ensure that the pathologist
was unaware of the treatment at the time of evaluation.
One representative slide of each eye from every animal
was selected for morphometric analysis. The slide was
then magnified XI00 and the outline of the choroid
was traced using a camera lucida system.
The choroidal tracing was then analyzed for total
choroidal length (perimeter) and total choroidal area,
using a Videoplan computer analyzer (Carl Zeiss Inc;
West Germany). The average choroidal thickness ("h")
of each eye was calculated by dividing the total choroidal length into the total choroidal area. This "h"
value correlated (correlation coefficient = 0.9986), with
99.9% confidence limits, to a mean value that was obtained by measuring the thickness of the choroid at
80-100 locations at regular intervals along its entire
length. On completion of the morphometry, the code
was broken. Mean "h" values of the choroids in control
animals, as well as those treated with active and inactive
SOD, catalase and sodium benzoate, were then compared using Student's t-test.
Determination of SOD in Ocular Tissue
To determine the levels of SOD in ocular tissue following IP injection of this enzyme a group of six normal
Hartley strain guinea pigs received IP injections of
PEG-SOD (2,000 units); a second group of six received
similar IP injections of the same dose of heat-inactivated PEG-SOD; and a third group of six received IP
injections of normal saline. Twenty-four hours following this procedure, the animals were killed. The freshly
enucleated globes were analyzed for SOD activity essentially as described by Crapo et al.14
Briefly each eye was homogenized in 2 ml of detergent (0.625 mg cholic acid in 100 ml phosphate buffer
with 0.1 M EDTA). Tubes were then centrifuged and
the supernatant was collected. To 100 n\ undialyzed
supernatant in a 3-ml cuvette with a light path of 1.0
cm at 24°C, 0.3 ml of 0.1 mM Ferricytochrome C, 0.3
ml of 0.5 mM xanthine and 2.3 ml buffer were added
and mixed. The reaction was initiated with 20 n\ xanthine oxidase (diluted 1:50 from a stock solution). The
rate of increase in absorbance per minute was recorded
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
887
at 550 nm. In another reaction mixture, the sample
was replaced with an equal volume of buffer and the
rate recorded after xanthine oxidase was added. The
contribution of nonspecific activity to the reduction
rate also was taken into consideration. The supernatant
first was boiled and then processed as described. The
change in absorbance was plotted against SOD concentration. SOD concentration was calculated in units
based on the suppressed reduction of Ferricytochrome
C in the reaction mixture. One unit is defined as the
quantity of SOD required to inhibit the reduction rate
of cytochrome C by 50% under the specific condition.14
Protein concentration was determined by the standard
Biuret method. Samples obtained from the 18 animals
mentioned above were divided into three groups according to the treatments administered. The mean of
each group was compared and analyzed using Student's
t-test.
Determination of Catalase in Ocular Tissue
A group of six Hartley strain guinea pigs received
IP injection of 2,400 units of PEG-catalase; a second
group of six animals similarly received heat-inactivated
PEG-catalase (2,400 units); and a third group in this
study received IP injection of normal saline. Twentyfour hours later, the animals were killed and the globes
were enucleated. Catalase activity was determined by
the method of Aebi.15
Each eye was homogenized and sonicated in 2 ml
of 50 mM phosphate buffer (pH, 7.0) containing 0.125
gm cholic acid/50 ml buffer. Tubes were then centrifuged and the supernatant collected. One hundred microliters of supernatant was diluted in 1.9 ml of 50
mM phosphate buffer (pH, 7.0). The solution containing the enzyme and H2O2 was read against a control
blank solution containing the enzyme and phosphate
buffer. To start the reaction, 1 ml of 30 mM H2O2 was
added. The estimation of catalase activity was based
on measurements of a first order rate constant (k) as
recommended by Aebi.15 This is derived by measuring
the decrease in absorbance over a set time interval.
The authors found the decrease in absorbance at
AA240 nM to be linear for at least 1 min and thus
catalase activities were measured over a 1-min interval
after initiating the reaction with H2O2. The following
relationship was used for our calculations:
k = (2.3/60) (log Ao/A.)
The results are expressed on the basis of protein content
in the samples. The protein content of each eye was
determined by the Biuret assay method. The mean of
each group was compared and analyzed using Student's
t-test.
888
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / May 1987
Vol. 28
Fig. 1. There is a marked thickening of the choroid (C) from inflammatory cell infiltration in untreated animals. Note the disruption of
retinal pigment epithelium and outer retinal (R) layers. (Hematoxylin-eosin stain, X200.)
Results
Clinically, all of the experimental and control animals developed signs of uveitis on days 12 or 13 after
injection of the S antigen. Histopathologically, the eyes
from the control animals (given normal saline, heatinactivated SOD and heat-inactivated PEG-catalase
injections) showed massive infiltration of mononuclear
cells, neutrophils, and eosinophils, mainly in the uveal
tract but with an extension of cells into the retina and
vitreous. Diffuse infiltration was evident throughout
the entire uveal tract. The retina at the site of inflammatory cell infiltration was necrotic; the retinal pigment
epithelium was both necrotic and disrupted (Fig. 1).
In the vitreous, most of the inflammatory cells were
found near the pars plana (Fig. 2).
In contrast to these changes, the globes of animals
treated with SOD, catalase, and sodium benzoate re-
vealed less inflammatory cell infiltration, with the infiltrate restricted predominantly to the uveal tract (Fig.
3). Although there were occasional foci of inflammation
involving the retina, the extent of retinal necrosis and
inflammatory cell infiltration was minimal. The retinal
pigmented epithelium was relatively uninvolved in the
inflammatory process, except at sites where the inflammatory cells extended into the retina. Moreover, in the
vitreous, inflammatory cellular exudate was minimal
near the pars plana (Fig. 4).
Morphometric analysis of the inflammatory process
in the six groups of animals revealed an average choroidal thickness of 194 ± 83 nm in the control (normal
saline-treated) group; 124 ± 12 and 188 ± 9 2 /mi in
the heat-inactivated SOD-treated and heat-inactivated
catalase group, respectively; 80 ± 7 pm in the SODtreated group; 60 ± 33 nm in the catalase-treated animals; and 93 ± 93 pm in guinea pigs treated with
Fig. 2. Severe inflammation involving the ciliary body (B) and the vitreous (V). (Hematoxylin-eosin stain, X180.)
Fig. 3. The inflammation is restricted mainly to the choroid (C) in animals treated with superoxide dismutase. (Hematoxylin-eosin stain,
X200.)
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
No. 5
889
OXYGEN RADICALS IN UVEITI5 / Rao er al.
T zv$i$.wassr
4.
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
890
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / May 1987
*
*
*\
•'
•
Vol. 28
t
Fig. 4. Minimal inflammatory cell infiltration in the ciliary body (B) and vitreous (V) of animal treated with catalase. (Hematoxylin-eosin
stain, XI80.)
sodium benzoate. In comparison with the control animals, the difference in choroidal thickness in three of
the treated groups (SOD, catalase, and sodium benzoate) was uniformly significant at P < 0.001 (Table
1). A similarly significant reduction (P < 0.001) in
choroidal thickness was noted when the SOD-treated
group was compared with the group treated with heatTable 1. Results of morphometric analysis of
choroidal thickness in animals treated with
antioxidants and hydroxyl radical scavenger
Treatment groups*
Normal saline
Superoxide dismutase
Heat-inactivated SOD
Catalase
Heat-inactivated catalase
Sodium benzoate
Choroidal
thickness
(urn)
Significance}
194 ±83
80± 7
J>< 0.001
124 ± 12
60 ±33
188 ±92
93 ±93
P< 0.001
P< 0.001
* Six animals in each group.
t Experimental groups were compared with the normal saline group using
Student's t-test.
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
inactivated SOD. Similar significant reduction was
noted when PEG-catalase group was compared with
the group treated with heat-inactivated PEG-catalase.
Superoxide Dismutase in Ocular Tissue
Eyes obtained from the injected animals revealed
the following mean values of SOD activity: 2.73 ± 0.26
units/mg of protein (active SOD group); 1.61 ±0.12
units/mg (heat-inactivated SOD group); and 1.62
±0.13 units/mg (normal saline control group). There
were significantly higher levels of SOD in the eyes obtained from SOD-injected animals when compared
with the heat-inactivated SOD and normal saline
groups (P < 0.001) (Table 2). Based on the values of
boiled homogenates, the authors found that nonspecific
activity in no way contributed to the reduction of cytochrome C.
Catalase in Ocular Tissue
Catalase activity in three groups of animals is summarized in Table 3. There was no significant difference
No. 5
891
OXYGEN RADICALS IN UVEITIS / Rao er al.
in the activity of this enzyme in oculartissueof animals
injected with normal saline (k = 2.20 ± 0.07/mg) when
compared with the group treated with heat-inactivated
PEG-catalase (k = 2.29 ± 0.12/mg). The animals injected with PEG-catalase (k = 3.28 ± 0.24/mg) showed
significantly higher levels of catalase activity when
compared with the other two groups (P < 0.001).
Discussion
Administration of antioxidant enzymes or a hydroxyl radical scavenger reduced the severity of ocular
inflammation and the extent oftissuenecrosis in severe
experimental allergic uveitis (EAU). Although there is
no direct evidence that reactive oxygen metabolites are
generated in EAU, the reduction of inflammatory
components achieved by the antioxidant enzymes and
hydroxyl radical scavenger indicates that reactive oxygen metabolites play a role in the inflammatory tissue
destruction evidenced in severe EAU.
It has been demonstrated that superoxide can generate chemotactic factors for neutrophils.16"18 Such
generation can lead to the recruitment of polymorphonuclear leukocytes, which in turn perpetuates and
amplifies the inflammatory response. The reduced severity of ocular inflammation and tissue destruction
in the animals treated with SOD may be due to partially
reduced recruitment of inflammatory cells.
The leukocytic neutral proteases released during inflammation can inflict damage on the extracellular
components; normally such damage is prevented by
antiproteases that are present in the tissue. Because
superoxide radicals are known to inactivate these antiproteases,1 it follows that the reduced tissue damage
and inflammation observed in animals treated with
SOD also could have resulted from the antiprotease
protection mechanism.
In the experimental immune complex disorders of
skin, lung, and kidney, the inflammatory tissue damage
was modulated by catalase919; however, these were
short-term experiments that lasted less than 24 hr. The
authors' observations indicate that antioxidants may
be capable of providing extended protection against
inflammation. The severity of S antigen-induced ocular
inflammation could be significantly reduced over a
week-long period with the use of SOD and catalase.
This extended antiphlogistic activity may result from
the abundance and strategic distribution of antioxidants in the ocular structures, when compared with
distribution in lung and other organs.20'21 It is also well
established that enzyme antioxidants not only prevent
the initial direct tissue damage induced by oxygen metabolites but also reduce the severity of ongoing ocular
inflammation by preventing the generation of chemotactic factors and protecting the activity of antiproteases.19
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
Table 2. Ocular levels of superoxide dismutase
activity after intraperitoneal injections of PEGsuperoxide dismutase in guinea pigs
Experimental groups
Normal saline injected
(control)
Heat-inactivated PEG-SOD*
(inactive SOD)
PEG-SOD (active SOD)
units/mg protein *t
P value
1.62 ±0.13
1.61 ±0.12
2.73 ±0.26
/>>0.05
P< 0.001
* Values are the means of six samples ± SEM.
t One unit of activity is denned as the amount of SOD required to reduce
the rate of cytochrome c reduction (A optical density, 550 nM) from 0.025
absorbance unit per minute to 0.0125 absorbance unit per minute.
% PEG-SOD = polyethylene-glycol-modified superoxide dismutase.
Under the conditions of this experiment, catalase
proved equal to SOD in its effectiveness as an antiphlogistic agent. The authors reported high levels of
catalase activity in ocular tissue of rats with lens-induced uveitis following IP injections of polyethylenemodified catalase.22 Similarly, current studies of IP injections of catalase and SOD revealed that high levels
of this enzyme persist in ocular tissue for up to 24 hr
(Tables 2, 3). Most of these enzymic activities could
be in the uvea, which lacks a blood-ocular barrier. By
determining SOD and catalase levels in various ocular
structures following IP injection and dose-response
studies in EAU, we may be able to provide additional
data on the significance of the blood-ocular barrier
and distribution of these antioxidant enzymes in normal and inflamed eyes.
The heat-inactivated SOD treatment reduced choroidal inflammation (Table 1), but not to the degree
that was achieved with the active SOD treatment. This
finding suggests that the preservatives present in enzymes that function as antioxidants may suppress inflammation to some extent. Additional studies are in
progress that will serve to delineate the nature of these
preservatives and their role in reducing S antigen-induced uveitis. In contrast to the effects of heat-inacti-
Table 3. Ocular levels of catalase activity after
intraperitoneal injections of polyethylene-glycolmodified catalase (PEG-catalase) in guinea pigs
Experimental groups
Normal saline injected
(control)
Heat-inactivated PEG-catalase
(inactive catalase)
PEG-catalase (active catalase)
k/mg
protein*^
P value
2.20 ±0.07
2.29 ±0.12
3.28 ±0.24
P>0.05
P< 0.001
* Values are the means of six samples ± SEM.
t The content of catalase is based on the constant rate of a first-order reaction
as described by Aebi.15
892
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / May 1987
vated SOD on the inflammation, heat-inactivated
PEG-catalase had no effect on the uveal inflammation,
indicating the specificity of catalase in reducing the
severity of experimental uveitis.
Benzoic acid and its salt, sodium benzoate, have been
used as antioxidants in a variety of situations.23 Benzoic
acid is widely and safely used as a food preservative,
and as a preservative of numerous topical ocular drugs.
The amount of carbon dioxide released from benzoic
acid is a common assay for hydroxyl radical production
in both chemical and cellular-generating systems.24
Treatment with sodium benzoate resulted in reduced
inflammatory cell infiltration in the choroid and retina.
Similar to the effects of antioxidant enzymes, this reduction of inflammatory cells was associated with a
relatively well-preserved retina and retinal pigment epithelium. Of the various inflammatory cells, polymorphonuclear leukocytes were most markedly diminished
in all treatment groups.
In view of the long-term treatments applied in these
experiments, the antiphlogistic activities of SOD, catalase, and benzoic acid cannot be entirely attributed
to the prevention of direct cytotoxic effects brought on
by reactive oxygen metabolites. The authors observed
that the numbers of neutrophils were reduced by antioxidant treatment; thus, inflammatory amplification
mechanisms were clearly modulated. Presumably, the
cytolytic effects of leukocyte proteases also were reduced, possibly through the natural protection of antiprotease activity. The protective effects of antioxidants
in this severe form of experimental uveitis certainly
implicate reactive oxygen metabolites in the pathogenesis of tissue destruction—yet, the extended nature of
our studies does not allow us to separate clearly the
inhibition of direct cytotoxic processes from the inhibition of amplification effects.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Key words: oxygen radicals, free radical scavengers, experimental uveitis, superoxide dismutase, catalase, sodium benzoate, hydroxyl radical
20.
References
21.
1. Janoff A and Carp H: Proteases, antiproteases, and oxidants:
pathways of tissue injury during inflammation. In Current Topics
in Inflammation and Infection, Majno G, Cotran RS, and Kaufman N, editors. Baltimore, Williams & Wilkins, 1982, pp. 6282.
2. Weiss SJ and LoBuglio AF: Phagocyte-generated oxygen metabolites and cellular injury. Lab Invest 47:5, 1982.
3. Sery TW and Petrillo R: Superoxide anion radical as an indirect
mediator in ocular inflammatory disease. Curr Eye Res 3:243,
1984.
4. Badwey JA and Karnovsky ML: Active oxygen species and the
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
22.
23.
24.
Vol. 28
functions of phagocytic leukocytes. Ann Rev Biochem 49:695,
1980.
Roos D: The metabolic response to phagocytosis. In The Cell
Biology of Inflammation, Weissmam G, editor. Amsterdam, Elsevier/North-Holland Biomedical Press, 1980, pp. 337-385.
Babior BM: Oxygen-dependent microbial killing by phagocytes.
N Engl J Med 298:659, 1978.
Hoe S, Rowley DA, and Halliwell B: Reactions of ferrioxamine
and desferrioxamine with the hydroxyl radical. Chem Biol Interact 41:75, 1982.
Fridovich I: The biology of oxygen radicals: the superoxide radical
as an agent of oxygen toxicity; superoxide dismutases provide
an important defense. Science 201:875, 1978.
Johnson KJ and Ward PA: Role of oxygen metabolites in immune complex injury of lung. J Immunol 126:2365, 1981.
Rao NA, Calandra AJ, Tran T, and Marak GE Jn Role of oxygen
radicals in immune complex mediated intraocular inflammation.
ARVO Abstracts. Invest Ophthalmol Vis Sci 26(Suppl):98, 1985.
Marak GE Jr, Rao NA, Scott JM, Duque R, and Ward PA: Free
radicals and phacoanaphylaxis. In Advances in Immunology and
Immunopathology of the Eye, O'Connor GR and Chandler JW,
editors. New York, Masson, 1985, pp. 144-45.
Rao NA, Wacker WB, and Marak GE Jr: Experimental allergic
uveitis: clinicopathologic features associated with varying doses
of S antigen. Arch Ophthalmol 97:1954, 1979.
Dorey C, Cozette J, and Faure JP: A simple and rapid method
for isolation of retinal S antigen. Ophthalmic Res 14:249, 1982.
Crapo JD, McCord J, and Fridovich I: Preparation and assay of
superoxide dismutase. In Methods in Enzymology, Vol 53, Colowick SP and Kaplan NO, editors. New York, Academic Press,
1978, pp. 382-393.
Aebi H: Catalase in vitro. In Methods in Enzymology, Vol 105,
Colowick SP and Kaplan NO, editors. New York, Academic
Press, 1984, pp. 121-126.
Petrone WF, English DK, Wong K, and McCord JM: Free radicals and inflammation: Superoxide-dependent activation of a
neutrophil chemotactic factor in plasma. Proc Natl Acad Sci
USA 77:1159, 1980.
Mittag TW: Role of free radicals in ocular inflammation and
cellular damage. Exp Eye Res 39:759, 1984.
Mittag TW, Hammond BR, Eakins KE, and Bhattacherjee P:
Ocular response to superoxide generated by intraocular injection
of xanthine oxidase. Exp Eye Res 40:41, 1985.
Rehan A, Johnson KJ, Wiggins RC, Kunkel RG, and Ward PA:
Evidence for the role of oxygen radicals in acute nephrotoxic
nephritis. Lab Invest 51:396, 1984.
Rao NA, Thaete LG, Delmage M, and Sevanian A: Superoxide
dismutase in ocular structures. Invest Ophthalmol Vis Sci 26:
1778, 1985.
Calandra AJ, Dunn SA, Atalla LR, Bowe BE, and Rao NA:
Distribution of hydrogen peroxide and enzyme antioxidants in
ocular tissue. ARVO Abstracts. Invest Ophthalmol Vis Sci
26(Suppl):64, 1985.
Rao NA, Fernandez MS, Sevanian A, Till GO, and Marak GE
Jr: Antiphlogistic effect of catalase on experimental phacoananphylactic endophthalmitis. Ophthalmic Res 18:185, 1986.
Gilman AG, Goodman LS, and Gilman A: In Goodman and
Gilman's Pharmacological Basis of Therapeutics, Edition 6, New
York, Macmillan, 1980, p. 961.
Sagone AL Jr, Decker MA, Wells RM, and Democko C: A new
method for the detection of hydroxyl radical production by
phagocytic cells. Biochim Biophys Acta 628:90, 1980.