5

Measurement of Reactive Oxygen Species

in Whole Blood and Mononuclear Cells
Using Chemiluminescence
Kuldip Thusu, Ehad Abdel-Rahman, and Paresh Dandona

1. Introduction
Free radicals generated by a wide variety of processes, such as ionizing
radiation (I), toxic xenobiotics (2), inflammation (3), and metabolites of mem-
brane hpld transformation (4), are involved in various diseases(5-9). A num-
ber of disorders m cellular and organ systems have been attributed to them.
Free radicals damage lipids (7), proteins (IO) and deoxyribonucleic acid (DNA)
(11,12) with consequent effects ranging from cell death to neoplasia (13). It is
therefore important to measure the amount of free radicals generated by varr-
ous cell and organ types. Traditionally, one would measure free-radical gen-
eration by electron spin resonance, which is costly or complicated, or measure
antioxidants that scavenge free radicals to get an estimate of the oxidative state.
We have developed a simple, sensitive, and reproducible method of measur-
ing free radicals m whole blood, mononuclear cells, neutrophils, and a number
of cell types using chemiluminescence. Our method also measures free-radical
generation in pure chemical systemsviz production of superoxide by xanthine/
hypoxanthine and xanthine oxidase (X0) in buffer or hydroxyl ion from
hydrogen peroxide, and so on.
2. Materials
2.1. Chronolog Aggregomefer/Luminometer
A two-channel or a four-channel chronolog aggregometer/ luminometer
along with computer interfacing is required. Chronolog (Havertown, PA)
also provides agglink software that displays and analyzes the data as the

From Methods m Molecular Biology, vol 108 Free Rackal and Ant/oxidant Protocols
Edtted by D Armstrong 0 Humana Press Inc , Totowa, NJ

57
58 Thusu, Abdel-Rahman, and Dandona

experiment is running; this 1sa great help. Alternately, a graph recorder can be
used as well.
2.2. Blood Samples
Ten milliliters of heparinized blood (14.3 USP/mL) is processed for the
preparation of mononuclear cells (MNC). The heparinized blood is used for
the assay of free radicals in whole blood and suspensions of MNC in buffer or
autologous plasma.
2.3. Reagents
1 Lummol. 10 mM stock in 0 1 h4 borate buffer.
a Borate buffer 0.1 M, pH 9.0 (9.5 g/250 mL is 0.1 M stock, adjust pH to 9.0.
b. Lummol Dissolve 0 1 77 g luminol in 100 mL of 0.1 M borate buffer
(0.177 g/100 mL is 10 mM).
2. Zymosan: 10 mg/mL suspension(Stock. 500 mg/SOmL).
a. 500 mg/SOmL m salme boiled for 15 min (double botler) m a small flask
placed m a beaker with distilled water and boiled on heating block.
b. Wash twice with salme by centrifuging at 235g and decant. Repeat the same
for second wash.
c. Suspend in salme to 500 mg/50 mL
3. Other agonists for the stimulation of free radicals that can be used are N-
formylmethionylleucmylphenylalanme (fMLP) (stock 10 mM in DMSO),
Phorbol ester (10 mM stock), Calcium tonophore (5 mM stock).

3. Methods
3.1. Preparation of MNC
1. MNC are Isolated (see Note 1) from heparmized blood (14.3 USP/mL) by a den-
sity gradient centrtfugation over Ficoll (Organon Tekmka Corp, Durham NC)
denstty of 1.077-l .080 g/mL at 20°C. Initially, the blood 1scentrifuged at 235g
for 10 min to obtain a buffy coat. The buffy coat is diluted to 3 mL m HBSS and
layered on top of 3 mL Atoll and respun at 235g for 30 mm at 22°C.
2. The mixed MNC band at the mterface is removed with a Pasteur pipet and the
cells are washed twice m Hank’s balanced salt solution (HBSS).
3. The remainder of the blood sample without buffy coat is centrifuged at 1200g m
order to obtain plasma.
4. MNC are washed m HBSS and resuspended m HBSS or medium Ml99 (Grbco,
Life Technologies Inc., Grand Island, NY) with 10% autologous plasma to make a
final concentration of 100,000 cells/ml for free-radical production (see Note 2)

3.2. Measurement of Oxygen-Free Radicals

1 Heparimzed venous blood samples is diluted m HBSS at a drlution of 1:9 and
kept at room temperature One mL sample of drluted blood is prpetted mto flat-
Measurement of Reactive Oxygen Species 59
bottom plastic tubes and Incubated at 37°C m a Chronolog lumi-aggregometer
for 3 min.
2. The chemiluminophore lummol (final cone 300 mM) is then added and free-
radical generation induced by fMLP (final cone: 20 n%t4) or any other agomst
like zymosan, calcium lonophore, phorbol mysterlc acid, etc. The dlamon of
lummol undergoes a reactlon with molecular oxygen to form a peroxide. This
peroxide is unstable and decomposes to an electronically excited state This
excited dlamon emits a photon that 1s detected by the chemilumeniscent detector.

r *Hz
0

LUMINOL
NH
20H-

-e 1

DIANION
0
x 02 A PEROXIDE

J-AMINOPTHALATE 3.AMINOPTHALATE 3-AMINOPTHAIATE

TRIPLET OIANION (tl) SINGLET DIANION (Sj) GROUND STATE DIANION S

Chemlluminescence 1srecorded for 15 mm (a protracted record after 15 min does

not alter the relative amounts of chemilummescence produced by various blood
samples). Our method, developed independently, 1s similar to that published by
Tosl and Hamedam (14).
3. We further established that, in our assay system, there 1s a dose-dependent mhi-
bition of chemilummescence by superoxide dismutase and catalase as well as
dlphenylene lodomum (DPI), a specific mhlbltor of nicotmamide ademne
dinucleotide phosphate (NADPH) oxidase, the enzyme responsible for the enzy-
matic production of superoxide ions. Our assay system 1s exquisitely sensitive
to DPI-induced inhibition at nanomolar concentrations (15) The specific mhibl-
tory effect of DPI on NADPH oxldase has been estabhshed by Hancock and
Jones (Id).

3.3. Free Radicals in Whole Blood and MNC

1. Heparmlzed whole blood diluted 1:9 with phenol-free HBSS or MNC suspen-
sions m HBSS or autologous plasma (see Note 2).
2. Stirrer speed: 1000 rpm, gain; start at 0 01 (to be increased or decreased as
required), unit temperature 37°C
3 Graph: 15 s/divlslon or as required
4. Test procedure* Run two channels simultaneously.
a Plpet 500 yL 1.9 diluted blood mto a cuvet contaming a stir bar
60 Thusu, Abdel-Rahman, and Dandona

b Place cuvet into the test well and cover the heater block.
c. Display graph on the screen and watt to ensure baseline has stabilized about
halfway across the first square
d Add 15 pL of 10 mit4 luminol stock solution into each of the cuvets and shut
the heating block (see Note 3)
e. Add 50 pL of (1’19 dilution in HBSS, stock 10 mg/mL), 1 pL of fMLP (stock
10 mA4), or any other agonist as required, after the baseline has advanced to
about half the length of the next square (see Note 3).
f. Keep the heater block closed and allow the curve to reach the maximum unttl
the end of the last square on the graph
g. Adjust the gain as required so that the luminescence curves are within the
range on the displayed graph.
h Once the gam is established, the test is performed at that parttcular gam settmg
1. The peak luminescence is recorded in mV from the pomt of zymosan or fMLP
addttton to the end of the graph.
J. The results are determined by computer settings or visual inspectton (see
Fig. 1).

3.4. Results
The characterizatton of active oxygen species using this technique IS sum-
marized m Table 1.
Lummol-dependent chemiluminescence offers a highly sensttrve means of
detecting active oxygen, although it is not specific towards any one oxygen
species. Superoxide-dismutase (SOD) (10 p,g) inhibited 81% of the chemilu-
minescence, whereas catalase (40 pg) inhibited only 47% of the activity. A
catalase concentratton (20 p,g) that by itself was essentially inactive comple-
mented SOD in eliminating all chemiluminesence. Total inhibition of the
response by DPI provides evidence for the detection of plasma membrane-
bound NADPH oxidase-mediated superoxrde production.

4. Notes
1. MNC should be prepared as soon as possible and the assaycompleted in less than 2 h.
2 The cell counts should be uniform for all experiments A differential count (com-
plex blood counts) should be performed on whole blood and the results normal-
ized against either mononuclear or neutrophil cell count. This 1s particularly
important for in vtvo experiments m which the admmistratton of a drug or any
other mterventton may change the cell counts For example, sterotds are known
to alter the differential cell counts m whole blood Therefore, the results should
be always expressed per neutrophil or mononuclear cell count.
3. Syringes or pipets should be designated for each reagent A carryover of an ago-
mst (like fMLP) into lummol will result in erratic baselines or the carryover of
DPI/SOD/catalase mto luminol may result in an unusual inhibitory response.
Measurement of Reactive Oxygen Species 61

ID 6261-1993 WSP
TEST mm < I I , i , ; ; / I 1 i I , l ,

coulltl 1 2 3 4
- - _-.-. -

fzzl5
IMIR
tmwnm m m l
SLOPE I,-rnL
STIRRER tom low 1m lew
MIW 0.s 0.5 __
-ntr
568 ul cell 6utpmmion
in RIIU IMIl
15 Ul Iwlml ,58 ul Qmean SC4LE XS M SECJDIU
58 ul FsntriPuged phuma ~l~~---~ CHJ - (#4-

Fig. 1. A representative tracing of zymosan induced free radical generatedby MNC

suspensiondiluted in autologous plasma.

Table 1
Characterization of Active Oxygen Species Using
Chemiluminescence Technique
Treatment Inhibition (%) f SE

10 ltg SOD” 81.5? 11.5

20 l,tg Catalaseb 5.5 + 13.2
40 yg Catalase 47.4 f 11.5
10 p.g SOD + 20 mg Catalase 100
40 nM DPI 100
a3100U/mgproteinof SOD.
b1.500
U/mg proteinof catalase

References
1 Roots, R. and Okada, S (1975) Estimatron of life times and diffusion dtstancesof
radicals mvolved m X-ray induced DNA strand breaks or killmg of mammalian
cells. Radiat. Res. 64,306-320
2. Oyanagi, Y. (1977) Strmulatory effect of platmum (IV) ion on the production of
superoxide radical from xanthme oxidase and macrophages.Biochem. Pharmacol.
26,473-476.
3 Tamura, M., Tamura, T , Tyagi, S. R., and Lambeth, J. D (1988) The superoxide-
generating respiratory burst oxtdase of human neutrophil plasma membrane. J.
Blol. Chem. 263, 17,621-17,626.
62 Thusu, Abdel-Rahman, and Dandona

4. Prasad, K. and Kalra, J. (1988) Oxygen free radicals and heart failure. Angiology
39,417-420.
5. Opre, L. (1989) Reperfuston inJury and tts pharmacologic modification. Circula-
tion 80,1049-1062
6 Floyd, R A (1990) Role of oxygen free radicals m carcmogenesrs and brain
rschemra. FASEB .I. 4,2587-2597.
7 Glaumann, B. and Trump, B F (1975) Studies on the pathogenesrs of rschemrc
cell mJury. III. morphologrcal changes of the proximal pars recta tubules (Ps) of
the rat kidney made ischemrc m VIVO. Vzrchows Arch [B] 19, 303-323.
8. Baker, G. L , Corry, R J., and Autor, A. P. (1995) Oxygen free radical mduced
damage in kidneys subJected to warm rshcemia and reperfusion Ann. Surg 202,
628-64 1
9. Parks, D. A., Bulkley, G. B., and Granger, D. N. (1983) Role of oxygen derived
free radicals in the digestive tract disease. Surgery 94,415-422.
10. Freeman, B. A., Rosen, G. M., and Barber, M. J. (1985) Superoxide perturbation
of the orgamzatron of vascular endothehal cell membranes. J. Biol. Chem 261,
6590-6593
11. Dandona, P., Thusu, K , Cook, S., Snyder, B., Makowskr, J., Armstrong, D., and
Nrcotera, T. (1996) Increased oxidatrve damage of deoxyribonucleic acid (DNA)
in dependent Diabetes mellitus. Lancet 347,444-445.
12. Cantoni, O., Sestrh, P., Cattabem, F., Bellomo, G., Pou, S., Cohen, M , and Ceruttr,
P. (1989) Calcmm chelator Qum 2 prevents hydrogen peroxide induced DNA
breakage and cytotoxrcrty. Eur. J. Bzochem. 82,209-212.
13. Ceruttr, P. A. (1985) Prooxidant states and tumor promotion. Sczence 227,375-38 1
14. Toss, M. F. and Hamedam, A. (1992) A rapid specific assay for superoxrde re-
lease from phagocytes m small volumes of whole blood. Am. J. Clzn. Puthol. 97,
566-573.
15 Nicotera, T., Thusu, K., Dandona, P (1993) Elevated productron of active oxygen
m Bloom’s syndrome cell lines Cancer Res. 53,5104-5107.
16. Hancock, J. T. and Jones, 0. T. G. (1987) Inhrbttron of drphenylene rodomum and
its analogues of superoxide generation by macrophages. Bzochem. J. 242, 103-107