Determination of Alternate Pathway Complement Kinetics by
Electron Spin Resonance Spectroscopy
DEAN A. HAWLEY, M.D., FREDERICK W. KLEINHANS, PH.D., AND JAMES L. BIESECKER, M.D., PH.D.
We describe a technic that measures the kinetics of the metabolic burst of peripheral blood neutrophils (PMN) by electron
spin resonance (ESR) spin trapping. Using this technic, a functional assay for the kinetics of the alternate pathway complement cascade is developed. PMN were stimulated by phagocytosis of opsonized zymosan (OpZym) to undergo the metabolic burst and the resulting free hydroxyl radicals generated
during the burst were trapped using 5,5-dimethyl-l-pyrrolineN-oxide (DMPO) to form a stable spin adduct. Rapid sequential measurements of the spin adduct concentration were
made using a computer-controlled ESR spectrometer to follow
the time course of the metabolic burst. Differences in the results obtained with OpZym and with PMN incubated with
unopsonized zymosan (Zym) and serum yields information
about the kinetics of opsonization.
The kinetics of the alternate complement pathway opsonization in sera from normal and C8-deficient patients were compared and found not to be affected significantly by the absence
of C8 complement protein. (Key words: Complement kinetics;
Electron spin resonance; ESR; Spin trap; DMPO; PMN) Am
J Clin Pathol 1983; 79: 673-677
THE PROTEINS of the alternate complement pathway
have been elucidated, but the kinetics of the cascade are
not well-characterized.19-20 During activation of the alternate pathway, opsonins are generated that are capable
of coating particles in the surrounding microenvironment. In the presence of neutrophils, the opsonized particles may be phagocytized, and the metabolic burst
within the neutrophils occurs. The purpose of this study
is to demonstrate a functional assay for the alternate
complement pathway and for the events involved in the
metabolic burst of peripheral blood neutrophils (PMN).
The alternate pathway can be activated using unopsonized zymosan (Zym), an extract of yeast cell wall
prepared from Saccharomyces cerevisiaeUM
PMN
phagocytize the opsonized zymosan (OpZym), the metabolic burst occurs, and free hydroxyl radicals (OH") are
produced. 9 The feasibility of trapping this OH* using 5,5dimethyl-1-pyrroline-N-oxide (DMPO) and detecting
the resulting free radical (DMPO/OH)' using electron
spin resonance (ESR) spectroscopy recently has been
demonstrated. 5 1 0 1 3 ' 5 We have developed a computerReceived August 13, 1982; accepted for publication September 9,
1982.
Address reprint requests to Dr. Hawley: Department of Medical
Research, Methodist Hospital of Indiana. Inc.. 1604 North Capitol
Avenue, Indianapolis, Indiana 46202.
Departments of Pathology and Medical Research, Methodist
Hospital Graduate Medical Center and Department of
Physics, Indiana University-Purdue University at
Indianapolis, Indianapolis, Indiana
automated ESR method that rapidly and sequentially
measures the amplitude of the spin adduct signal during
the metabolic burst.
The kinetics of the alternate pathway complement
cascade were measured using an in vitro system containing PMN, DMPO, and either OpZym or Zym and
serum. The time delay occurring when Zym and serum
are used in place of OpZym reflects the length of time
required for opsonization of the Zym by the serum. This
opsonization time was determined for normal serum
and for serum from a patient with a deficiency of the
eighth component of complement. The influence of various concentrations of complement was investigated by
measuring the metabolic activity of PMN activated by
Zym in serial dilutions of AB serum. The effect of temperature on opsonization, phagocytosis, and PMN metabolism also was investigated. To confirm that OH* was
the free radical trapped, inhibitors of the metabolism of
PMN were added to the reaction mixture. 515 The inhibitors included superoxide dismutase (SOD), catalase,
and nitroblue tetrazolium (NBT).
Materials and Methods
Whole blood with 100 U/mL sodium heparin (Panheprin, Abbott) was collected from normal, healthy
adults who gave informed consent for venepuncture.
PMN from whole blood were isolated by density gradient centrifugation with Ficoll-Hypaque and hypotonic
lysis of erythrocytes. PMN were resuspended at 4 X 107
PMN/mL in modified Krebs-Ringer's phosphate glucose buffer (KRP). 7 Zym (Sigma, St. Louis, MO) was
suspended at 8.28 mg/mL in KRP. OpZym was prepared by incubating the Zym suspension in 14% pooled
human AB serum for 10 minutes at 37°C. For several
experiments, Zym was suspended in KRP containing
SOD (Sigma) at 80 Mg/mL, catalase (Sigma) at 960 fig/
m L, or NBT (Sigma) at 165 Mg/mL. A 1.0 molar aqueous
0002-9173/83/0600/0673 $01.05 © American Society of Clinical Pathologists
673
HAWLEY, KLEINHANS, AND BIESECKER
674
FIG. 1. ESR spectrum of (DMPO/OH)- adduct produced by stimulated PMN. Signal obtained after 12 minutes incubation at 37°C of
50 yjL of PMN in KRP (4 X 7/mL). 50 nL zymosan (8.28 mg/mL).
25 nL aqueous DMPO (1 M). and 25 ML AB serum. Spectrometer
Settings: Field 3313 G, Freq 9.3 GHz. power 60 mW (dual cavity),
modulation
1.6 G @ 100 kHz. time constant 0.25 seconds, gain 2.5
X 104, computer controlled sweep 4 minutes total.
solution of DMPO (Aldrich, Milwaukee, WI) was used
for spin trapping.
Control serum was obtained by pooling normal human AB sera. C8-deficient serum was obtained from a
patient with known C8 deficiency demonstrated by absence of C8 on immunoelectrophoresis and absence of
detectable hemolytic complement activity. Sera were
frozen at - 7 0 ° C and, when necessary, heat-inactivated
at 56°C for 30 minutes.
ESR samples were prepared by combining 50 nL of
PMN, 50 ixL of the suspension of Zym, 25 pL of KRP
buffer, 25 fiL of the DMPO solution, and 25 ixh of
serum. For some experiments OpZym was substituted
by Zym and buffer was replaced with serum, and for
some experiments the serum was heat-inactivated. For
all assays the suspensions were mixed quickly in a glass
Table I. ESR Determined'PMN Metabolic Activity
Using Opsonized Zymosan and
Zymosan Plus Serum*
Time of Peakjij
(minutes)
Opsonized zymosanf
(N)
Zymosan plus serumf
(N)
3.7 ± 0.4
(7)
6.3 ± 0.4
(7)
Peak Amplitude
(mm)
247 ± 7
(7)
A.J.C.P. • June 1983
test tube and drawn into a 100-/iL capillary tube (Clay
Adams, 4625). The capillary tubes were sealed with Critoseal® (Sherwood, St. Louis, MO) and inserted into the
spectrometer cavity. The temperature of the cavity was
maintained at 37 ± 0.5°C.
We performed paired measurements comparing the
metabolism of PMN phagocytizing OpZym with the
metabolism of PMN phagocytizing Zym in the presence
of serum. Similarly, we compared the metabolism of
PMN that phagocytized Zym while incubated in pooled
AB serum with the metabolism of PMN phagocytizing
Zym while incubated in C8-deficient serum. The statistical significance of the results were evaluated with the
paired Mest.
Spectra were obtained using a Varian El09 X band
spectrometer with an E231 -2 dual sample cavity. Spectra
of the second line of the spin adduct quartet were obtained using settings of: frequency = 9.17 GHz, field
= 3254 gauss centered on the second line in the spectrum, sweep width = 3 gauss, sweep time = 60 seconds,
microwave power = 60 mW (into dual cavity), modulation amplitude = 2 gauss at 100 kHz, gain = 2.5 X 104,
and detection time constant of 1 second. Some spectra
were obtained by using a 90-second sweep with a 4second time constant. These fast sweep rates and long
time constants caused some spectral distortion, but they
provided maximum sensitivity and no loss of internal
consistency.
The spin adduct ESR signal decays in a period of
150
E
E
100
d)
"0
0
+)
50
QE
<
ca
s
in
ca
(M
T i me (m i n )
FIG. 2. PMN metabolic burst vs. time, stimulated with opsonized
zymosan (O) and with zymosan plus serum
(U). The reaction mixtures
208 ± 14
contained 50 ML PMN in KRP (4 X 107 PMN/mL), 25 ML aqueous
(7)
DMPO (1 M), and the following: (O) 50 nL opsonized zymosan in
KRP (8.28 mg/mL) and 50 ML KRP; (U) 50 ML zymosan. 25 nL KRP.
7
* The reaction mixtures contained 5011L PMN in KRP (4 X I0 PMN/mL). 25 nL aqueous
and 25 nL AB serum. The ordinate is the (DMPO/OH)'
signal amDMPO (I'M), and 25 j*L KRP plus 25 nL opsonized zymosan (8.28 mg/mL) or 25 tiLplitude
AB normalized to a spectrometer gain of 2.5 X 104. Spectrometer
serum plus 25 fiL zymosan.
settings: H0 = 3254 G, f = 9.17 GHz, AH = 3 G. P = 60 mW (dual
t Mean ± standard deviation. Number of measurements = 7.
cavity), t = 90 seconds, modulation = 2 G at 100 kHz. and time
X Measured from addition of serum or opsonized zymosan.
constant = 4 seconds.
§The peak time differences are significant at the P = 0.005 level.
Vol. 79 • No. 6
COMPLEMENT KINETICS BY ESR
13
minutes. In an experiment lasting 20 minutes or longer, the signal at any given time does not represent the
integrated spin adduct production from time zero.
Rather, it represents the average of spin adduct concentration over the previous few minutes with the spin adduct concentration from more recent times weighed
more heavily. Consequently, the underlying process of
OH* production is observed with a time resolution of
only a few minutes, despite the somewhat faster sampling rate of the experimental procedure.
The spectrometer was controlled by a microcomputer
(Hewlett-Packard 9825T®) programmed to sweep
through the second spectral line once every 60 or 90
seconds for a total period of 20 to 60 minutes. Spectra
were stored on a floppy disk for subsequent analysis and
graphing of spectral line amplitude versus time. A computer-controlled plotter (Hewlett-Packard 9872B) generated graphs directly from the data files.
675
150
FIG. 3. Effect of inhibitors on stimulated PMN metabolic burst.
Reaction mixtures consisted of 50 ML' PMN in KRP (4 X 107 PMN/
mL), 25 nL aqueous DMPO (1 M), 50 JIL AB serum and: (*) 50 ^L
zymosan in KRP (8.28 mg/mL); (C) zymosan + catalase (960 jig/mL);
(N) zymosan + NBT (165 Mg/mL); and (S) zymosan + SOD (80 \x%l
mL).
Results
A typical (DMPO/OH)* spin adduct spectrum obtained with stimulated PMN is shown in Figure 1. The
four-line spectrum had hyperfine splitting constants of
AN = AH = 14.8 gauss and a 1:2:2:1 intensity distribution, which is consistent with the reported ESR parameters of the (DMPO/OH)* adduct.4-5-615 Further experiments were conducted by measuring only the second
line of this signal and plotting its amplitude versus time.
A comparison of the metabolic burst of normal PMN
phagocytizing OpZym with the metabolic burst of normal PMN phagocytizing Zym plus serum is presented
in Table 1, and typical results are shown in Figure 2.
No signal was obtained using heat-inactivated AB serum
instead of AB serum. The peak of metabolic burst activity from PMN phagocytizing OpZym occurred 2.6
minutes before the peak metabolic activity from PMN
phagocytizing Zym in serum. The peak amplitude of
PMN phagocytizing OpZym was 19% higher than the
peak amplitude of PMN phagocytizing Zym in serum
(Table 1). C8-deficient serum used in place of normal
pooled AB serum for opsonization of zymosan produced
a metabolic peak 0.4 minutes later and the peak amplitude diminished by 12% (Table 2).
Results of experiments utilizing SOD, catalase, or
NBT in the incubation mixture are shown in Figure 3.
Catalase reduced the amplitude of the signal by 34%,
and SOD completely suppressed the signal. NBT decreased the signal amplitude by 40%, and a black precipitate was noted in the capillary tubes at the comple200
Table 2. ESR Determined Metabolic Burst Activity
Using Pooled AB Serum or C8-deficient Serum
for Opsonization of Zymosan*
Time of Peakt§
(minutes)
Peak Amplitude
(mm)
7.2 ± 0.6
Pooled ABf
170 ± 5
(5)
(N)
(5)
7.6 ± 0.4
150 ± 10
C8-deficientt
(5)
(5)
(N)
FIG. 4. Effect of serum dilution on PMN metabolic burst. Reaction
mixtures
consisted of 50 ML PMN in KRP (4 X 10' PMN/mL). 50
• The reaction mixtures contained 50 nL PMN in KRP (4 X 10' PMN/mL). 25 /iL aqueous
ML zymosan in KRP (8.28 mg/mL), 25 ML aqueous DMPO (1 M),
DMPO (1 M), 25 ML zymosan in KRP (8.28 mg/mL). and 25 JIL serum.
and: (D) 50 nh AB serum; (*) 25 ML AB serum and 25 iiL KRP;
t Mean ± standard deviation. Number of measurements = 5.
(A) 12.5 nL AB serum and 37.4 n\ KRP; (+) 6 ML AB serum and 44
$ Measured from addition of serum.
ML KRP.
§ Peak time and amplitude differences are significant at the P = 0.01 level.
Table 3. Quantitative Levels of C3, C3-activator, and
C4 in Pooled AB Serum and C8-deficient Serum
C3*
C3-activatorf
C4*
A.J.C.P. • June 1983
HAWLEY, KLEINHANS, AND BIESECKER
676
Normal A3 Serum
C8-deficient Serum
111 mg/dL
28.6 mg/dL
20.8 mg/dL
190 mg/dL
41 mg/dL
27.7 mg/dL
* Determined by radial immunodiffusion.
t Determined by laser nephelometry.
tion of the ESR measurement. The reaction mixture
with NBT was examjned by light microscopy and the
black precipitate was consistent with intracellular formazan. 12
The metabolism of the PMN phagocytizing Zym or
OpZym is related to the concentration of serum used
for opsonization. This is shown in Figure 4. The serum
concentration was varied from 28% to 3.5%. Peak response time changed from 5.6 minutes with 28% serum
to 26.6 minutes with 3.5% serum. The peak amplitude
dropped with both high (28%) and low (1%, not shown)
serum concentrations, but it remained fairly constant
over the range of 14% to 3.5% concentration. The amplitude of the signal with any one serum concentration
was also dependent upon the net concentrations of zymosan and PMN, and the zymosan:PMN ratio (results
hot shown).
The serum concentration results may be interpreted
to mean that complement concentration is an important
factor in determining the peak time and amplitude of
response. Therefore, complement levels were determined for the pooled AB and C8-deficient serum. Quantitative levels of C3, C3-activator (also known as B), and
C4 were determined by radial immunodiffusion and/or
laser nephelometry for both the patient and the pooled
serum (Table 3).2 C3, C3-activator, and C4 levejs averaged 49% higher in the C8-deficient serum than in the
AB serum.
To determine sensitivity to incubation temperature,
ESR measurements of the metabolism of PMN were
made at several temperatures. Although the time of peak
response was fairly constant between 35°C and 38°C,
there was a doubling of the response time by 39°C.
Above 37°C, the amplitude of the response diminished.
Discussion
The peak metabolic activity of PMN incubated with
OpZym occurred after 3 to 5 minutes of incubation. We
found that significant alternate pathway opsonization
of zymosan occurred within 2 to 3 minutes. Studies conducted by other investigators using different antigens
and methods have shown similar patterns of
results. 1-3,8,11.16,18.19,20 Deficiency of C8 in serum did not
have a significant effect on the kinetics of the opsonization of zymosan, and this is consistent with the known
noninvolvement of C8 in the alternate pathway." 1 8
Inhibitors of OH" production, including SOD and catalase, caused decreased signal amplitudes. 4 ' 515 We found
similar results with NBT. Formazan production and
decreased signal amplitude were observed when NBT
was present in the reaction mixture. The likely explanation is that NBT reacts with superoxide to produce
formazan, and this reaction decreases the amount of
superoxide available for OH' production. Physiologic
temperatures were optimal for OH" production. Increased temperatures caused diminished OH' production and others have reported similar findings.3-917
Serum dilution curves showed incremental increases
in opsonization time that corresponded to decremental
changes in serum concentration. These increases in opsonization time are likely a result of the decreased levels
of complement.
In summary, we have shown that the kinetics of the
alternate pathway of complement can be determined
using ESR spin trapping technics. No appreciable differences are found between pooled AB serum and C8deficient serum.
Acknowledgment. The authors wish to thank Brian G. Davies and
Steven T. Barefoot, DDS for assistance in acquiring the ESR data.
References
1. Allen RC: Evaluation of serum opsonic capacity by quantitating
the initial chemiluminescent response from phagocytizing
polymorphonuclear leukocytes. Infect Immun 1977; 15:828—
833
2. Biesecker JL, Triplett DA, Miller GA, Hawley DA: The pathologist as a consultant in the laboratory evaluation of the patient
with recurrent infections. American Society of Clinical Pathologists Workshop Manual. Chicago, American Society of
Clinical Pathologists, 1979
'
3. Cohen HJ, Newburger PE, Chovaniec ME, Whitin JC, Simons
ER: Opsonized zymosan-stimulated granulocytes—activation
and activity of the superoxide-generating system and membrane potential changes. Blood 1981; 58:975-982
4. Finkelstein E, Rosen GM, Rauckman EJ, Paxton J: Spintrapping
of superoxide. Mol Pharmacol l'979; 16:676-685
5. Green MR, Hill HAO, Okolow-Zubkowska MJ, Segal AW: The
production of hydroxyl and superoxide radicals by stimulated
human neutrophils—measurements by EPR spectroscopy.
FEBS Lett 1979; 100:23-26
6. Harbour JR, Chow V, Bolton JR: An electron spin resonance
kudy of the spin adducts of OH and H02 radicals with nitrones
in the ultraviolet photolysis of aqueous hydrogen peroxide solutions. Can J Chem 1974; 52:3549-3553
7. Hawley DA, MillerGA, Biesecker JL: Normal neutrophil function
in human renal allograft recipients receiving Minnesota antiiymphoblast globulin (MAG). Am J Clin Pathol 1981; 76:29^
33
Vol. 79 • No. 6
COMPLEMENT KINETICS BY ESR
8. Klebanoff SJ, Clark RA: Iodination by human polymorphonuclear leukocytes: a re-evaluation. J Lab Clin Med 1977; 89:675686
9. Klebanoff S}, Clark RA: The neutrophil: function and clinical
disorders, Chap 6; The metabolic burst, Chap 7, anti-microbial
systems. Amsterdam, North-Holiand, 1978, pp 409-466
10. Kleinhans FW, Hawley DA, Davies BG, Biesecker JL, Goodman
MR: Electron paramagnetic resonance assay of serum from
systemic lupus erythematosus patients. Bulletin of the American Physical Society 1981; 26:1223
11. Muller-Eberhardt HJ: The complement system^ The plasma proteins, 2nd Edition. Edited by FW Putnam. New York, Academic Press, 1975, pp 393-431
12. Nathan DG, Baehner RL, Weaver DK: Failure of nitroblue tetrazolium reduction in the phagocytic vacuoles of leukocytes
in chronic granulomatous disease. J Clin Invest 1969; 48:18951904
13. Okolow-Zubkowska MJ, Hill HAO: Spin trapping of superoxide
and hydroxyl radicals produced by stimulated human neutrophils. Developmental Biochemistry 1980; 11 B:201-210
14. Pillemer L, Blum L, Lepow IH, Ross OA, Todd EW, Wardlow
AC: The properdin system and immunity. Demonstration and
isolation of a new serum protein, properdin, and its role in
immune phenomena. Science 1954; 120:279
677
15. Rosen H, Klebanoff SJ: Hydroxyl radical generation by polymorphonuclear leukocytes measured by electron spin resonance spectroscopy. J Clin Invest 1979; 64:1725-1729
16. Rossi F, Romeo D, Patriarca P: Mechanism of phagocytosis-associated oxidative metabolism in polymorphonuclear leukocytes and macrophages. J Reticuloendothel Soc 1972; 12:127149
17. Rustagi PK, LoBuglio AF: Phagocytosis of technetium-99m sulfur
colloid by human polymorphonuclear leukocytes. J Immunol
Methods 1978; 23:69-78
18. Schreiber RD, Morrison DC, Podack ER, Muiler-Eberhard HJ:
Bactericidal activity of the alternative complement pathway
generated from 1 i isolated plasma proteins. J Exp Med 1979;
149:870-882
19. Verhoef J, Peterson PK, Quie PG: Kinetics of staphylococcal opsonization, attachment, ingestion, and killing by human polymorphonuclear leukocytes: a quantitative assay using ('H) thymidine labeled bacteria. J Immunol Methods 1977; 14:303311
20. Williams AJ, Hastings MJG, Easmon CSF, Cole PJ: Factors affecting the in vitro assessment of opsonization: a study of the
kinetics of opsonization using the technique of phagocytic chemiluminescence. Immunology 1980; 41:903-911
i