Elimination of soluble 123I-labeled aggregates of IgG in patients with systemic lupus erythematosus. Effect of serum IgG and numbers of erythrocyte complement receptor type 1

442
ELIMINATION OF SOLUBLE 1231-LABELED
AGGREGATES OF IgG IN PATIENTS WITH
SYSTEMIC LUPUS ERYTHEMATOSUS
Effect of Serum IgG and Numbers of
Erythrocyte Complement Receptor Type 1
CORNELIS HALMA, FERDINAND C. BREEDVELD, MOHAMED R. DAHA, DIK BLOK,
JEANNETTE H. EVERS-SCHOUTEN, JO HERMANS, ERNST K. J. PAUWELS,
and LEENDERT A. VAN ES
Using soluble '231-labeled aggregates of human
IgG ('231-AHIgG)as a probe, we examined the function
of the mononuclear phagocyte system in 22 patients with
systemic lupus erythematosus (SLE) and 12 healthy
controls. In SLE patients, a decreased number of erythrocyte complement receptor type 1 was associated with
less binding of '231-AHIgGto erythrocytes and a faster
initial rate of elimination of '231-AHIgG (mean f SEM
half-maximal clearance time 5.23 f 0.2 minutes, versus
6.58 f 0.2 minutes in the controls), with possible
spillover of the material outside the mononuclear phagocyte system of the liver and spleen. However, multiple
regression analysis showed that serum concentrations of
IgG were the most important factor predicting the rate
of '231-AHIgG elimination. IgG concentration may thus
reflect immune complex clearance, which in turn, would
influence the inflammatory reaction, in SLE.
Tissue deposition of circulating immune complexes (IC) is believed to be important in the pathogenesis of systemic lupus erythematosus (SLE) (1,2).
From the Departments of Nephrology, Rheumatology,
Clinical Pharmacy and Toxicology, and Medical Statistics, and the
Division of Nuclear Medicine, Department of Diagnostic Radiology,
University Hospital Leiden, Leiden, The Netherlands.
Cornelis Halma, MD: Department of Nephrology; Ferdinand C. Breedveld, MD: Department of Rheumatology; Mohamed
R. Daha, PhD: Department of Nephrology; Dik Blok, PharmD:
Department of Clinical Pharmacy and Toxicology; Jeannette H.
Evers-Schouten: Department of Nephrology; Jo Hermans, PhD:
Department of Medical Statistics; Ernst K. J. Pauwels, PhD:
Division of Nuclear Medicine; Leendert A. van Es, MD, PhD:
Department of Nephrology.
Address reprint requests to Cornelis Halma, MD, Department of Nephrology, University Hospital Leiden, PO Box 9600,
2300 RC Leiden, The Netherlands.
Submitted for publication March 5 , 1990; accepted in revised form November 6, 1990.
Arthritis and Rheumatism, Vol. 34, No. 4 (April 1991)
Because IC are removed from the circulation by the
mononuclear phagocyte system (MPS) (3), dysfunction of the MPS might result in delayed elimination of
IC, with subsequent deposition of IC in tissues. The
resulting inflammatory reaction could then cause the
manifestation of SLE symptoms. When MPS function
in SLE patients was measured using IgG-sensitized
erythrocytes (E-IgG) as a probe, markedly prolonged
half-maximal elimination times
were reported in
most studies (4-7). However, E-IgG are insoluble
corpuscular IC that are preferentially taken up by the
spleen (8,9), unlike soluble IC, which are cleared
mainly by the liver (10-12).
We have described a new method to measure
MPS function, using soluble radiolabeled aggregates of
human IgG ('231-AHIgG) as a probe (13,14). These
aggregates are preferentially removed from the circulation by the liver (13,15,16). In addition, they have
other immune complex-like activities, such as activating complement, with subsequent binding to human
erythrocytes through the erythrocyte C3b receptor
(complement receptor type 1).
The importance of erythrocyte complement receptor type 1 (E-CR1) in the elimination of IC has
recently been demonstrated in experiments on animals
(12,17,18). Complement-activating IC were infused
into the aortae of baboons. The IC rapidly bound to
erythrocytes, and were subsequently removed from
erythrocytes during their transit through the liver (12).
When these IC were injected into complementdepleted animals (17) or when IC that fixed complement poorly were used (18), less IC bound to erythrocytes, disappearance of IC from the circulation was
accelerated, and a greater proportion of IC was deposited outside the MPS (17,18).
443
IgG, E-CR1, AND IC ELIMINATION IN SLE
Based on the findings in these animal experiments, it was hypothesized that erythrocyte-bound IC
are taken up by the MPS in the liver and spleen only,
and therefore, several passages of the vascular system
are required before these IC are removed from the
circulation. On the other hand, freely circulating IC
(i.e., not erythrocyte-bound) are taken up not only by
the liver and spleen, but by other organs as well.
Therefore, unbound IC may be eliminated in one
passage through the vascular system (18).
Another mechanism that might impair the
erythrocyte-bound IC transport system would be a
deficiency of E-CR1. Decreased numbers of E-CR1
have been demonstrated in several diseases in humans; for instance, the majority of patients with SLE
have reduced levels of E-CR1 (19). Accordingly, SLE
patients with decreased E-CR1 should have an increased rate of removal of soluble complementactivating IC (as opposed to the delayed clearance of
insoluble IC, such as E-IgG).
We decided to study the relationship between
the clearance of '231-AHIgGand the number of E-CR1
in a group of SLE patients, selected by E-CR1 levels.
Second, we assessed the effect of E-CR1 number on
the uptake of '231-AHIgG by the liver and spleen.
Finally, we examined the effect of various factors
(levels of complement factors, circulating immune
complexes, IgG, and rheumatoid factor) on the clearance of '231-AHIgG, since these variables are thought
to influence the elimination of IC.
PATIENTS AND METHODS
Patients and controls. E-CRI numbers were determined in 77 patients with SLE. We then selected patients
with extremely high or low levels of C R l , trying to obtain a
group with differing degrees of disease activity as well. The
final study group comprised 22 patients (21 female and 1
male, age range 15-63 years). All patients met the American
Rheumatism Association criteria for the diagnosis of SLE
(20). Treatment for SLE consisted of nonsteroidal antiinflammatory drugs in 6 patients, antimalarial drugs in 10,
prednisone in 12 (mean SEM 11.6 k 1.8 mg/day; 3 patients
taking >20 mglday), daily azathioprine in 3, and pulse
cyclophosphamide every 3 months in 2. Splenectomy had
been performed in 1 patient. At the time of the study, all
patients were graded for clinical activity by one investigator
(FCB), using the Systemic Lupus Erythematosus Disease
Activity Index (SLEDAI) (21). The healthy control group
consisted of 6 men and 6 women, 20-35 years of age.
IgA deficiency was excluded in all subjects. All subjects were treated with oral iodide to prevent thyroid uptake of
Na'231. Blood pressure and pulse rate were monitored during
the first hour after 1231-AHIgGadministration.
*
The study protocol was approved by the Ethics
Committee of Leiden University Hospital. All subjects (and,
in the case of the 2 minors, their parents as well) gave
informed consent before participating.
Preparation of lz3I-AHIgG. AHIgG was prepared by
the Central Laboratory of The Netherlands Red Cross Blood
Transfusion Service, as previously described (13). This IgG
preparation contains -40% monomeric IgG and -60% IgG
of larger size, ranging between 300 kd and 20,000 kd (13).
AHIgG was subsequently radiolabeled with 12'1 using the
Iodogen method (22). Briefly, on the day of the study, 1 ml
of AHIgG was thawed and mixed with 15 MBq (1 MBq =
0.027 mCi) NaIz3I,in a sterile tube coated with Iodogen (100
pg). After 15 minutes, the radiolabeled solution was passed
over a 0.22-pm filter. The filter was flushed with saline, and
an aliquot of the final solution (-1 mg of 1231-AHIgG)was
administered to each subject.
The mean (5SEM) amount of radioactivity administered was 4.5 ? 0.3 MBq (122 & 8 pCi), of which 50.3
1.4% was associated with aggregates of IgG. This percentage
was calculated by counting the precipitate after treating a
sample of the labeled material with 3% (final concentration)
polyethylene glycol 6000 (PEG) for 30 minutes at 0°C. At this
concentration of PEG, at least 95% of IgG aggregates and
4%
of monomeric IgG is precipitated (13). All results
reported herein refer to the clearance of the PEGprecipitable and erythrocyte-bound radioactivity only.
Study of the clearance of lZ3I-AHIgG. Patients and
controls received - 1 mg of '231-AHIgG intravenously over
30 seconds. Serial blood samples were drawn from the
opposite forearm into chilled Vacutainer tubes containing
123 USP units of heparin. All samples were kept on ice and
processed simultaneously, 150 minutes after injection. In a
previous experiment, we had examined whether release of
I2'I-AHIgG occurs if a sample of whole blood is kept at 0°C
for 150 minutes. After injection of '231-AHIgG into 19
subjects, the mean +- SEM erythrocyte-associated radioactivity was 27.4 3% of whole blood when assayed immediately, versus 23.4
3% when assayed after 150 minutes.
This difference was not significant.
Residual radioactivity was counted in samples of
whole blood and plasma, and in precipitates of plasma after
treatment with 3% PEG as described above, to measure
aggregate-bound radioactivity. Erythrocyte-bound radioactivity (E-AHIgG) was measured after washing 0.5-ml samples of whole blood twice with 2 ml of phosphate buffered
saline at 0°C. The total amount of aggregates per milliliter of
whole blood (i.e., PEG-precipitable and erythrocyte-bound
aggregates) was calculated from the radioactivity of the
components of whole blood, in counts per minute, as follows:
*
*
*
Total aggregates = [(cpm in whole blood
-
cpm in E-AHIgG)
x (cpm in PEG-precipitated plasmdcpm in plasma)]
+ cpm in E-AHIgG
Radioactivity was measured in a Minaxi Auto-Gamma 5530
Counter (Packard, Downers Grove, IL).
CH5O and the individual complement components
C3, C4, and Clq were measured as described previously
HALMA ET AL
444
SLE
NORMAL
RANGE
I
0 5 10 15
30
45
120
60
min
Figure 1. Time curve for the elimination of 1231-labeledaggregated human IgG (polyethylene glycol
precipitable and erythrocyte bound) in 22 patients with systemic lupus erythematosus (SLE) and 12 normal
controls, expressed as the mean SEM percent of maximal level. Biexponential curves are seen, with an
initial phase of rapid elimination followed by a second, slower phase, starting 30-45 minutes after injection.
*
(23). To detect circulating immune complexes, the I2'1-Clq
binding assay (24) and the IgG-PEG assay (25) were used.
Anti-double-stranded DNA was measured by immunofluorescence, using Crithidia luciliae as the substrate (26). The
number of CRl on erythrocytes was calculated from a
radioligand binding assay, using the anti-CR1 monoclonal
antibody 2A8 (14). Because of varying degrees of anemia in
the SLE patients, the number of CRl was corrected for the
number of erythrocytes per liter, measured by counter
(Coulter, Hialeah, FL). Concentrations of IgG were measured by radial immunodiffusion. Rheumatoid factors were
measured in an isotype-specific enzyme-linked immunosorbent assay, as recently described (27).
Organ uptake of '"I-AHIgG. Radioactivity over hepatic and splenic areas was registered continuously for I
hour after the injection, with a Toshiba GCA 40A gamma
camera (Tokyo, Japan). All counts were first corrected for
background. Liver: spleen uptake ratios were calculated by
dividing the total levels of uptake of radioactivity in the 2
organs. However, uptake levels for each single organ are
given as organ: background ratio, after correction for surface
area. Thus, they are expressed as the ratio/pixel (pixel =
surface unit). From the organ uptake curves, maximal liver
and spleen uptake were determined, as well as the time
period between injection and the time of maximal uptake,
and the time it took for the 1iver:background ratio to
decrease by 50% (T,,,2-liver).
Statistical analysis. Residual radioactivity timecurves of the total amount of aggregates per milliliter of
whole blood were plotted. Because elimination of IC is a
receptor-mediated process that can be described mathematically, analogous to the elimination of drugs, the curves were
also analyzed according to standard pharmacokinetic methods. The curves appeared to be biexponential, suggesting a
2-compartment system, with a rapid distribution phase (first
half-life [first T,J) and a slower elimination phase (second
T1,J (28).
The half-lives were calculated, using linear regression analysis, as described previously (29). The TI,,, i.e., the
time course of a substance's remaining in the body, is
derived from 2 basic pharmacokinetic parameters: clearance
(i,e., the ability of the body to eliminate the substance) and
volume of distribution (i.e., the space in the body available
to contain the substance) (28). These 2 parameters were
calculated with a noncompartmental method, based on statistical moment theory (28). The advantage of this method is
that no assumption is made regarding a pharmacokinetic
model. Thereby, curve fitting, with its attendant inaccuracies, is obviated. The zero moment, the area under the curve
(AUC), was calculated by the trapezoidal rule, with extrapolation to infinity. The first moment, mean residence time
(MRT), is analogous to the half-life. Volume of distribution
at steady state (VDss) was calculated according to the
formula VDss = clearance X MRT (28). Clearance of
445
IgG, E-CR1, AND IC ELIMINATION IN SLE
radioactivity was calculated according t o the standard pharmacokinetic formula: clearance = doseiAUC (expressed as
mliminute). T h e d o s e substituted in the formula was the
number of PEG-precipitable c p m administered. Both clearance a n d V D s s were expressed per kg body weight.
G r o u p means were compared using Student's t-test
a n d analysis of variance (ANOVA). Correlations were studied using Pearson's product-moment correlation coefficient.
Because of the multiple testing done, we designated a
reduced (but arbitrary) level of P < 0.01 as significant for the
correlations. To assess t h e relative importance of some
selected variables in predicting variations in elimination and
Table 1. Comparison of laboratory, elimination, and organ uptake
variables after injection of 1231-labeledaggregated human IgG (Iz3[AHIgG) into 12 healthy controls and 22 patients with systemic lupus
erythematosus (SLE)*
Laboratory variable
E-CRI
(/erythrocyte)
E-CRI ( X 10'2iliter)
IgG (gmiliter)
CH5O (unitsiml)
C3 (mg%)
C4 (mg%)
C l q (mg%)
Elimination variable
Clearance
(ml/minute/kg)
VDss (ml/kg)
MRT (minutes)
First TI,, (minutes)
Second T,,,
(minutes)
E-AHIgG max (%)
Organ uptake variable
Liver area (pixels)
Spleen area
(pixels)
Liver max uptake
(ratioipixel)
Liver max time
(minutes)
Spleen max uptake
(ratioipixel)
Spleen max time
(minutes)
Liver:spleen max
ratio
T,,,-liver (minutes)
SLE
patients
Controls
Pt
479 2 57
748 2 83
0.010
2,189 f 280
12.3 f 1.3
253 40
66 f 3.7
13.5 2.0
8.5 f 0.7
3,636
9.9
335
55
19.3
9.4
0.6
4.82
*
*
5.20
?
685 f 62
165 f 24
5.23 f 0.2
147.7 18
*
*
305
0.9
f 33
f 3.6
2.1
f 0.5
5
f
*
f
0.4
645 f 75
146 2 23
6.58 f 0.2
154.3 20
*
0.003
NS
NS
NS
NS
NS
NS
NS
NS
0.011
NS
11.5 f 2.7
20.7
f
2.9
0.037
507
273
* 17
2.5 * 0.2
512
282
f
5
31
24
NS
NS
2.6
f
0.2
NS
14.2 2 0.9
16.3
f
0.9
NS
1.4 -+ 0.1
2.1
f
0.1
0.000
14.0 f 1.1
18.3
1.3
0.019
4.7 f 0.5
3.0
* 0.3
0.017
26.6 2 1.2
40.3
1.0
0.000
f 20
f
* Values are the mean SEM. E-CRI = erythrocyte complement
receptor type 1; VDss = volume of distribution at steady state; MRT
= mean residence time; T,,, = time to half-maximal elimination;
E-AHIgG max = maximal erythrocyte-bound radioactivity; max
uptake = maximal organ:background uDtake: max time = time of
maximal uptake after injection i f 12'I-AHIgG;liver:spleen rnax ratio
= ratio of maximal whole organ uptake of radioactivity by liver and
spleen; T,,,-liver = time for liver maximal uptake to decrease by
50%. See Patients and Methods for details.
t By Student's unpaired t-test. NS = not significant.
Table 2. Stepwise multiple linear regression analysis of the dependence of elimination variables on clinical and laboratory variables
and of the dependence of organ uptake variables on elimination
variables, in 22 patients with systemic lupus erythematosus*
Dependent
variable
Step
Independent
variable
1
2
IgG
SLEDAI
0.44
VDss/kg
1
IgG
0.55
First TI,,
1
2
IgG
CRl/liter
0.23
0.45
Second TI,,
1
SLEDAI
0.23
1
VDssikg
First TI,,
First T I n
0.26
0.20
0.39
Elimination
Clearance/kg
Organ uptake
Liver max uptake
Liver max time
Spleen max uptake
Spleen max time
1
1
1
R2t
0.62
* SLEDAI = systemic lupus erythematosus disease activity index.
See Table 1 for other definitions.
t R2 = proportion of the variance of the dependent variable that is
explained by the independent variable. The analysis was stopped
when P values of 0.05 were reached.
organ uptake of '231-AHIgG, forward stepwise multiple
linear regression analysis was performed. All values are
presented as the mean k SEM. T h e statistical software used
was SPSS-X a n d STATS.
RESULTS
Tolerance and elimination time of lZ3I-AHIgG.
Administration of '231-AHIgGwas well tolerated by all
subjects, with no significant symptoms or changes in
vital signs during the experiments. In both controls
and SLE patients, biexponential disappearance curves
were seen (Figure 1).
Comparison of SLE patients and controls. The
laboratory measurements and 1231-AHIgGelimination
and uptake variables were compared in SLE patients
and healthy controls (Table 1). Of the laboratory
variables, the only significant difference between the 2
groups
. was in the number of E-CR1 , both uncorrected
and corrected for the number of erythrocytes Per liter.
The E-CR1 level ranged from 68 to 992 Der ervthrocvte
in the SLE group and from 475 to 1,220 per erythrocyte in the controls. The initial elimination of 1231AHIgG was faster in the patient group (shorter first
T',~);the other elimination variables showed no differences between the 2 groups. The maximal amount of
'231-AHIgG bound to erythrocytes, expressed as perv
HALMA ET AL
446
Table 3. Relationships (Pearson's correlation coefficient) among clinical and laboratory, elimination, and organ uptake variables in 22
systemic lupus erythematosus patients who received 1231-labeledaggregated human IgG ('231-AHIgG)*
E-CRlI
SLEDAI liter CH50
SLEDAI
E-CRliliter
CH50
C3
C4
IgG
IgGRF
IgA-
RF
Clearance/
kg
Liver Liver Spleen
VDss/ First Second max max max
kg
T,,, T,,, uptake time uptake
0.41
-0.70t
-0.59t
-0.03
-0.41
0.12
-0.37
-0.52
0.31
0.51
-0.15
-0.05
-0.32
-
-0.41
-0.29
c3
-0.15
c4
-0.16
IgC
-0.16f
I&-RF
-0.03
IgA-RF
-0.01
Clearanceikg
0.53t
VDssikg
0.05
First TI/,
-0.47
Second T,,,
-0.48
Liver max uptake -0.06
Liver max time
-0.52
Spleen max uptake -0.31
Spleenmax time
-0.42
0.46
0.25
0.40
-0.03
0.01
-0.11
-0.31
-0.14
0.46t
0.30
0.07
0.36
0.27
0.31
0.63t
0.66t
0.25
-0.56t
-0.43
0.16
-0.02
-0.01
-0.21
0.08
0.04
0.43
-0.10
0.47
-0.35 -0.34
-0.52 -0.23
0.50t
-0.45 -0.49
0.45t
0.20
0.13 -0.66tf
-0.19
0.14 -0.74tt
-0.04
0.11 0.48
0.04
0.05 -0.05
0.41
0.20 -0.24
-0.03
0.34 0.06
0.61t 0.42 -0.28
0.04
0.01 0.09
0.48
-0.53t
-0.16
0.44
0.38
-0.12
0.19
-0.31
0.23
-0.53t
-0.42
0.24
0.32
-0.40
0.15
-0.37
0.39
0.24
0.03 -0.29
0.45
0.33
0.26 -0.15
0.63
0.25
0.09
0.38 -0.15 0.02
0.54 0.54
* SLEDAI = systemic lupus erythematosus disease activity index; E-CRl = erythrocyte complement receptor type 1; IgG-RF = IgG
rheumatoid factor; VDss = volume of distribution at steady state;
= time to half-maximal elimination; max uptake = maximal
0rgan:background uptake: max time = time of maximal uptake after injection of Iz3I-AHIgG.
t P < 0.01:
$ Discussed in the text.
cent of whole blood radioactivity (E-AHIgG max),
was higher in the controls.
To compare our results with those of a previous
study (1 4), we also analyzed the disappearance curves
of the PEG-precipitable material only (without erythrocyte-bound radioactivity). The curves were very
similar to the disappearance curves of the total amount
of L231-AHIgG
(erythrocyte-bound + PEG-precipitable
radioactivity). The first TlI2 of the PEG-associated
radioactivity was shorter in the patients (mean 2 SEM
5.34 ? 0.4 minutes, versus 6.88 k 0.2 minutes in the
did not
controls; P = 0.006), while the second
differ between the 2 groups (174.1 2 24 minutes in the
SLE patients versus 217.6 2 29 minutes in the controls).
Organ uptake data showed that both the maximal liver uptake and the time needed to reach this
maximum were similar in the SLE patients and the
controls. Maximal splenic uptake occurred earlier in
the SLE patients, but maximal amount of spleen
uptake was less than in the controls.
Variables determining elimination of '231-AHIgG
in SLE patients. To assess the relative importance of
several variables believed to play a role in the elimination of 1231-AHIgG,multiple regression analysis was
performed with the data obtained for the SLE patients
(Table 2). Analysis was first performed with clearance
of 1231-AHIgGas the dependent variable and selected
clinical or laboratory parameters as independent variables. The results showed that the concentration of
serum IgG was the best parameter for predicting the
clearance of '231-AHIgG. The SLEDAI also contributed to the variability of clearance rates. When the
other elimination variables were used as dependent
variables with the clinical and laboratory parameters
as independent variables, the serum IgG level was
found to also be the best predictor of first
and
distribution volume. The first T,, was partially predicted by the number of E-CRl as well. The first T I n
contributed weakly to organ uptake parameters.
Next, we investigated the correlations, and
especially the direction of the correlations (positive or
negative), between the clinical and laboratory parameters and the elimination and organ uptake data. These
calculations are presented in matrix form in Table 3.
They showed that higher disease activity scores
(SLEDAI) were weakly and nonsignificantly associated with greater clearance rates (0.01 < P < 0.05).
Lower numbers of E-CR1 were weakly and nonsignificantly associated with shorter first TI, (0.01 < P <
0.05). Complement levels showed no correlation with
elimination variables. There was an inverse correlation between the serum concentration of IgG and the
clearance and volume of distribution of '231-AHIgG
447
IgG, E-CR1, AND IC ELIMINATION IN SLE
r = - 0,66
L]
p < 0.01
0
P
0
UU
0
'
0
0
E
0
B
1
1
0
+I
B
U
3l
I
ClqBA +
0 G-PEG +
a aDNA +
Q RF
0
I
I
4
1
1
8
I
1
12
I
I
16
I
I
20
I
1
24
1
I
28
Figure 2. Correlation between clearance of 1231-labeledaggregated human IgG (polyethylene glycol [PEG]
precipitable and erythrocyte bound) per kg body weight (y-axis) and serum concentration of IgG (x-axis) in
22 patients with systemic lupus erythematosus. ClqBA = 12'I-Clq binding assay; G-PEG = IgG-PEG assay;
aDNA = anti-double-stranded DNA; R F = rheumatoid factor. There was no correlation between a positive
result on any of these 4 tests and the elimination of '231-labeled aggregated human IgG.
(Figures 2 and 3). A similar but weaker relation was
seen between rheumatoid factor levels and '231-AHIgG
clearance and first T,,, (Table 3).
Some laboratory parameters were not included
in the correlation matrix because the low number of
positive results precluded meaningful statistical analysis. These were circulating anti-DNA (6 patients with
positive results); Clq binding assay (4 patients with
positive results); IgM rheumatoid factor and IC measured with the IgG-PEG assay (3 patients with positive
results for each). No clear-cut relationship existed
between a positive result on one of these tests and the
elimination kinetics of AHIgG (Figures 2 and 3). No
correlation was found between the maximal amount of
'231-AHIgG bound to erythrocytes (E-AHIgG max)
and the number of E-CR1 (r = 0.26, P = 0.2).
Since E-CR1 numbers varied 2.5-fold among
the controls, we also looked for a correlation between
E-CR1 number and '231-AHIgG elimination parameters in this group. No correlation was found.
Effect of treatment on serum concentration of
IgG and '231-AHIgG elimination parameters. The serum concentration of IgG was the most important
variable determining the rate of elimination of 1231AHIgG. Levels of IgG differed widely in the patient
group (range 1-25.4 gm/liter), but they were not correlated with disease activity (Table 3). To investigate
whether this variation was related to treatment, we
divided the SLE patients into 3 groups. Group 1
consisted of 9 patients who did not receive any immunosuppressive therapy, group 2 consisted of 8 patients
who were treated with prednisone (mean _t SEM 10.2
? 1.9 mglday), and group 3 consisted of 5 patients who
were being treated with cytotoxic drugs. Four patients
in group 3 were also receiving prednisone (mean 5
SEM 14.4 ? 3.7 mg/day). The mean ? SEM serum IgG
HALMA ET AL
448
1.6 1
0
r = - 0.74
p
0
-= 0.01
1.2
Q
0
11
ClqBA +
G-PEG +
aDNA +
RF
i-
o
0
0
0
0
0.6 -
0
mE
0.4 0.2 -
0
0
0
-
01
I
I
4
I
1
8
1
I
I
12
I
16
I
I
1
20
24
28
g/L
Figure 3. Correlation between the volume of distribution in steady state of '231-labeled aggregated human
IgG (PEG precipitable and erythrocyte bound) per kg body weight (y-axis) and serum concentration of IgG
(x-axis) in 22 patients with systemic lupus erythematosus. See Figure 2 for definitions.
concentration was 6.3 t 2.2 gm/liter in group 3, versus
14.8 t 1.6 and 13.2 t 1.9 in groups 1 and 2, respectively ( P = 0.021 by ANOVA). The disease activity
index (SLEDAI) was 5.2 ? 2.1 in group 3, versus 1.4
t 0.4 and 3.4 L 1.2 in groups 1 and 2, respectively ( P
= 0.114). Clearance parameters did not differ between
the patients treated with steroids and those not receiving steroids.
DISCUSSION
Using '231-AHIgG as a probe of MPS function
in systemic lupus erythematosus patients and controls,
we found biexponential disappearance curves, with a
rapid distribution phase followed by a slower elimination phase. The first TI/, represents the disappearance
of the larger aggregates of IgG from the circulation
(13). In the SLE group, less '231-AHIgG bound to
erythrocytes, and the first TI/, was decreased. Except
for the difference in E-CRl numbers, for which the
patients were selected, other variables thought to
affect the clearance of IC showed no obvious differences between patients and controls (Table 1). The
number of E-CR1 partially predicted the first TIl2
(Table 2) and was positively, albeit not significantly,
correlated with the first TI/, (Table 3).
In recent experiments with model immune complexes in human subjects, Schifferli et a1 (30) also noted
an inverse correlation between E-CR1 number and elimination of IC in patients with a variety of IC-related
diseases. In their experiments, E-CR1 levels were correlated with the degree of binding of 1C to erythrocytes.
Such a correlation was not observed in the present
study. The inverse correlation between E-CR1 number and elimination of IC (30) and between E-CR1
number and distribution of 1231-AHIgG(this study) is
in accordance with the hypothesis described in the
Introduction of this report, i.e., that conditions leading
to decreased binding of IC to erythrocytes in vivo lead
to accelerated elimination of IC (18).
In the control subjects, no correlation was found
between E-CR1 and first
However, the distribution
IgG, E-CR1, AND IC ELIMINATION IN SLE
of E-CRl levels was more widespread in the SLE group
since the patients were selected for extremely high or
low levels; for instance, the coefficient of variation for
E-CR1 was 61% in the SLE group versus 30% in the
controls. This would tend to make the effect of E-CR1 on
first Tin more distinct in the patient group. An alternative explanation, of course, could be that the decreased
number of E-CR1 and the shortened TI,, are not causally
related and that both are manifestations of some underlying disease process.
The elimination of Iz3I-AHIgG in patients with
SLE has been analyzed in a previous study from our
institution (14). It was reported that the first TI,, of the
PEG-precipitable material, although somewhat shorter
in SLE patients, did not differ significantly between
patients and controls. Furthermore, the second TIl2
was prolonged in the SLE group (14). The results in
the current study (shorter first T,,, in the SLE group,
similar second T,,, in patients and controls) are obviously at variance with these earlier findings. However,
the 2 studies differ in several respects. The current
study population was selected by E-CR1 number. As a
result, the mean E-CR1 number in the patient group
was 64% of that in the controls, while in the previous
study this percentage was 81%. This may have caused
the first TI,, to be significantly shorter in the patients in
the current study. The 2 study populations differed in
other clinical parameters as well: The present group
had a smaller percentage of patients who were taking
prednisone, a smaller percentage who were positive
for IC, and a much wider range of IgG levels.
The discrepancy in the results of the second TI/,
can be explained by the longer followup period after
injection of Iz3I-AHIgG in the present study. Figure 1
shows that because of the shorter first TI,,, the second
part of the disappearance curve starts earlier in the
SLE group. Followup time in the previous study was
60 minutes (14). If this limit is applied to the data in
Figure 1 , it can be seen that in the controls, the very
short time interval between the end of the first phase
and the final time point (60 minutes) would cause a
seemingly steeper decline of the second part of the
curve (shorter second TI,,). By prolonging the followup to 120 minutes, this apparent difference between the 2 groups disappears.
Where is '231-AHIgG taken up? As found previously (1 4), liver: spleen uptake ratios were increased
in the SLE patients, suggesting splenic dysfunction.
We have shown that splenectomy by itself does not
impair the clearance of 1231-AHIgG,because increased
hepatic uptake compensates for the loss of splenic
449
function (31). In the SLE group no such compensation
was seen, maximal liver uptake and rate of uptake
being similar to the findings in controls. Since the
distribution volume of '231-AHIgGwas not different in
the 2 groups, the data suggest that in patients with
SLE, more Iz3I-AHIgGis eliminated outside the MPS
of the liver and spleen than in normal subjects. This
notion is compatible with the observations made in
studies of baboons, that impairment of the E-IC transport system leads to increased deposition of IC in the
lungs and kidneys, i.e., outside the MPS of the liver
and spleen (17,18).
However, a caveat is in order with regard to the
above statements. The organ uptake ratios represent
the sum of 3 dynamic processes: uptake, metabolism,
and excretion of radioactivity. Therefore, when, for
instance, all 3 processes are commensurately accelerated (as might occur after activation of the MPS),
organ uptake may not change. Thus, the seeming lack
of hepatic compensation for the loss of splenic function might theoretically be due to a changed equilibrium among the 3 dynamic processes mentioned
above. Evidence somewhat mitigating against this
possibility is the fact that both maximal liver uptake
and rate of uptake were very similar in the patients and
the controls.
The experiments reported here do seem to offer
some indirect evidence for the possibility that the
mononuclear phagocyte system is activated in patients
with SLE. For instance, hepatic radioactivity decreased faster in the patient group (shorter Tllz-liver,
Table I ) , which suggests enhanced catabolism or excretion of '231-AHIgG. Alternatively, the more rapid
decrease of hepatic radioactivity could just represent
decreased retention of the aggregates. To demonstrate
conclusively that more Iz3I-AHIgG is deposited outside the MPS of the liver and spleen in patients with
SLE would require total body scanning, which was not
done in this study.
In a regression model, 3 of 4 elimination variables tested were partially predicted by concentrations
of serum IgG. The strong correlation between VDss
and IgG suggests that the degree of Fc receptor
occupancy by IgG accounts for most of the variation in
distribution volume of Iz3I-AHIgG. It has been reported that increasing concentrations of IgG delay the
elimination of IgG-coated erythrocytes (32,33) and of
soluble IC (33). Other investigators could not confirm
these findings, however (7,34). In vitro studies have
shown that IgG blocks adherence of IC to macro-
450
phages (35,36) and that it inhibits Fc receptormediated phagocytosis (37).
The question arises as to whether the elimination of IC is determined by monomeric IgG or by IgG
complexes (contained in IC) or rheumatoid factor. The
results of testing for IC in this study do not suggest an
important role for IgG complexes. The clearance of
'231-AHIgGvaried inversely with the level of rheumatoid factor (Table 3 ) . However, since the levels of
rheumatoid factor were also correlated with the concentration of IgG, no conclusions can be drawn from
the weak association between rheumatoid factor level
and clearance of '231-AHIgG.
What determines serum concentrations of IgG
in patients with SLE? Because SLE is associated with
polyclonal B cell stimulation, one would expect a
relationship between disease activity and IgG concentration. However, patients with active disease generally receive more aggressive treatment, e.g., steroids
and cytotoxic drugs, both of which lead to a decrease
of serum IgG levels (38). Thus, concentrations of IgG
are theoretically determined by at least 2 conflicting
but interdependent factors, disease activity and drug
therapy. This is reflected by the differing results reported in the literature concerning the relationship
between IgG concentration and other disease variables
in patients with SLE (3941). Our results showed no
correlation between disease activity and IgG level
(Table 3 ) . More aggressive treatment with cytotoxic
drugs seemed to result in lower IgG concentrations,
which in turn were associated with faster rates of
elimination of '231-AHIgG. Steroid treatment did not
by itself depress serum IgG (or affect clearance of
'231-AHIgG),probably because of the low doses taken
by the patients.
The results of this study show that in patients
with systemic lupus erythematosus, the concentration
of IgG, and possibly the number of E-CR1 and the
activity of the disease, correlate with the rate of
elimination of IC. The faster initial elimination of IC in
patients with SLE may be the consequence of a
deficient erythrocyte transport mechanism, resulting
in deposition of IC outside the MPS. Although a causal
relationship cannot be inferred from the associations
observed in this study, they do suggest that certain
patient characteristics, such as IgG level and possibly
E-CR1 level, may influence the handling of IC and,
consequently, may have an effect on the extent and
site of the inflammatory reaction that causes the
manifestations of the disease.
HALMA ET AL
ACKNOWLEDGMENTS
The authors thank Drs. A. H. M. Heurkens, H. M.
Markusse, and J. D. Macfarlane for referring their patients.
The technical assistance of A. A. Voetman (Central Laboratory of The Netherlands Red Cross Blood Transfusion
Service), C. van der Keur, J. A. J . Camps, and R. I. J.
Feitsma is greatly appreciated.
REFERENCES
1. Koffler D, Agnello V, Thoburn R, Kunkel HG: Systemic
lupus erythematosus: prototype of immune complex
nephritis in man. J Exp Med 134:1698-1798, 1971
2. Theofilopoulos AN, Dixon FJ: Immune complexes in
human diseases. Am J Pathol 100536594, 1980
3. Mannik M: Pathophysiology of circulating immune complexes. Arthritis Rheum 25:783-787, 1982
4. Frank MM, Hamburger MI, Lawley TJ, Kimberly RP,
Plotz PH: Defective reticuloendothelial system Fcreceptor function in systemic lupus erythematosus. N
Engl J Med 300:518-523, 1979
5. Parris TM, Kimberly RP, Inman RD, McDougal JS,
Gibofsky A, Christian CL: Defective F c receptormediated function of the mononuclear phagocyte system
in lupus nephritis. Ann Intern Med 97526532, 1982
6. Bannister KM, Hay J, Clarkson AR, Woodroffe AJ:
Fc-specific reticulo-endothelial clearance in systemic
lupus erythematosus and glomerulonephritis. Am J Kidney Dis 3:287-292, 1984
7. Mantzouranis EC, Rosen FS, Colten HR: Reticuloendothelial clearance in cystic fibrosis and other inflammatory lung diseases. N Engl J Med 319:338-343, 1988
8. Frank MM: Pathophysiology of immune hemolytic anemia. Ann Intern Med 87:21&222, 1977
9. Van der Woude FJ, van der Giessen M, Kallenberg
CGM, Ouwehand W, Beekhuis H, Beelen JM, van Son
WJ, Hoedemaeker PhJ, van der Hem GK, The TH:
Reticuloendothelial Fc receptor function in SLE patients. I. Primary HLA linked defect or acquired dysfunction secondary to disease activity? Clin Exp Immuno1 55:473480, 1984
10. Arend WP, Mannik M: Studies on antigen-antibody
complexes. 11. Quantification of tissue uptake of soluble
complexes in normal and complement-depleted rabbits.
J Immunol 107:63-75, 1971
11. Kijlstra A, Knutson DW, Daha MR, van Es LA: Clearance and glomerular localization of preformed DNPanti-DNP immune complexes. Scand J Immunol10:421429, 1979
12. Cornacoff JB, Hebert LA, Smead WL, VanAman ME,
Birmingham DJ, Waxman FJ: Primate erythrocyteimmune complex-clearing mechanism. J Clin Invest
71:236-247, 1983
IgG, E-CR1, AND IC ELIMINATION IN SLE
13. Lobatto S, Daha MR, Voetman AA, Evers-Schouten
JH, van Es AA, Pauwels EKJ, van Es LA: Clearance of
soluble aggregates of human immunoglobulin G in
healthy volunteers and chimpanzees. Clin Exp Immunol
69:133-141, 1987
14. Lobatto S, Daha MR, Breedveld FC, Pauwels EKJ,
Evers-Schouten JH, Voetman AA, Cats A, van Es LA:
Abnormal clearance of soluble aggregates of human
immunoglobulin G in patients with systemic lupus
erythematosus. Clin Exp Immunol72:55-59, 1988
15. Jimenez RAH, Mannik M: Evaluation of aggregated IgG
in mice as an Fc receptor specific probe of the hepatic
mononuclear phagocyte system. Clin Exp Immunol 49:
200-208, 1982
16. Veerhuis R, Krol MC, van Es LA, Daha MR: Difference
in clearance kinetics of particulate immune complexes
and soluble aggregates of IgG in vivo. Clin Immunol
Immunopathol41:379-391, 1986
17. Waxman FJ, Hebert LA, Cornacoff JB, VanAman ME,
Smead WL, Kraut EH, Birmingham DJ, Taguiam JM:
Complement depletion accelerates the clearance of immune complexes from the circulation of primates. J Clin
Invest 74:1329-1340, 1984
18. Waxman FJ, Hebert LA, Cosio FG, Smead WL, VanAman ME, Taguiam JM, Birmingham DJ: Differential
binding of immunoglobulin A and immunoglobulin G1
immune complexes to primate erythrocytes in vivo. J
Clin Invest 77:82-89, 1986
19. Walport MJ, Lachmann PJ: Erythrocyte complement
receptor type 1 , immune complexes, and the rheumatic
diseases. Arthritis Rheum 31:153-158, 1988
20. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ,
Rothfield NF, Schaller JG, Tala1 N, Winchester RJ: The
1982 revised criteria for the classification of systemic
lupus erythematosus. Arthritis Rheum 25: 1271-1277,
1982
21. Chang C-H, Urowitz MB, Gladman D, Bombardier C:
Impact of SLE disease activity on survival with validation of a disease activity index (abstract). Arthritis
Rheum 31 (suppl 4):S56, 1988
22. Fraker PJ, Speck JC: Protein and cell membrane iodinations with a sparingly soluble chloroamide, 1,3,4,6tetrachloro-3a,6a-diphenylglycoluril.Biochem Biophys
Res Commun 80:849-857, 1978
23. Daha MR, Bertha RM, Thompson J, Kauffmann RH,
Nicholson-Weller A, Veltkamp JJ, Briet E: Combined
hereditary deficiency of the sixth component of complement and factor VIII coagulant activity in a Dutch
family. Clin Exp Immunol 48:733-738, 1982
24. Zubler RH, Lange G, Lambert PH, Miescher PA: Detection of immune complexes in unheated sera by a
modified '251-Clq binding test. J Immunol 116:232-235,
1976
25. Valentijn RM, van Es LA, Westedt ML, Daha MR: The
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
45 1
detection of circulating immune complexes containing
immunoglobulin G. J Clin Lab Immunol 14:73-79, 1984
Aarden LA, de Groot ER, Feltkamp TEW: Immunology
of DNA. 111. Crithidia luciliae, a simple substrate for the
determination of anti-dsDNA with the immunofluorescence technique. Ann NY Acad Sci 254505-515, 1975
Otten HG, Daha MR, de Rooij HH, Breedveld FC:
Quantitative detection of class-specific rheumatoid factors using mouse monoclonal antibodies and the biotin/
streptavidin enhancement system. Br J Rheumatol 28:
310-316, 1989
Gibaldi M, Perrier D: Pharmacokinetics. New York,
Marcel Dekker, 1982
Van der Woude FJ, Beekhuis H , Beelen JM, The TH:
Kinetic analysis of the blood clearance and organ uptake
curves of IgG-coated red cells in HLA-typed controls
and patients with Wegener's granulomatosis. Eur J Nucl
Med 12:130-136, 1986
Schifferli JA, Ng YC, Paccaud J-P, Walport MJ: The
role of hypocomplementaemia and low erythrocyte complement receptor type 1 numbers in determining abnormal immune complex clearance in humans. Clin Exp
Immunol75:329-335, 1989
Halma C, Daha MR, van Furth R, Camps JAJ, EversSchouten JH, Pauwels EKJ, Lobatto S, van Es LA:
Elimination of soluble '231-labelled aggregates of human
immunoglobulin G in humans: the effect of splenectomy.
Clin Exp Immunol 77:62-66, 1989
Kelton JG, Singer J, Rodger C, Gauldie J, Horsewood P,
Dent P: The concentration of IgG in the serum is a major
determinant of Fc-dependent reticuloendothelial function. Blood 66:490-495, 1985
Kurlander RJ, Hall J: Comparison of intravenous
gamma globulin and a monoclonal anti-Fc receptor
antibody as inhibitors of immune clearance in vivo in
mice. J Clin Invest 77:2010-2018, 1986
Van der Woude FJ, Kallenberg CGM, Limburg PC,
Beekhuis H , Ouwehand W, van der Giessen M, Hoedemaeker PhJ, van der Hem GK, The TH: Reticuloendothelial Fc receptor function in SLE patients. 11. Associations with humoral immune response parameters in
vivo and in vitro. Clin Exp Immunol 55:481-486, 1984
Arend WP, Mannik M: In vitro adherence of soluble
immune complexes to macrophages. J Exp Med 136:
514-531, 1972
Knutson DW, Kijlstra A, van Es LA: Association and
dissociation of aggregated IgG from rat peritoneal macrophages. J Exp Med 145:1368-1381, 1977
Segal DM, Dower SK, Titus JA: The FcR-mediated
endocytosis of model immune complexes by cells from
the P338d, mouse macrophage line. I. Internalization of
small, nonaggregating oligomers of IgG. J Immunol
130:130-137, 1983
Katz P, Fauci AS: Immunosuppression and immunoad-
452
juvants, Immunological Diseases. Fourth edition. Edited by M Samter. Boston, Little, Brown, 1989
39. Alarcon-Segovia D, Fishbein E: Serum immunoglobd i n s in systemic lupus erythematosus. Clin Sci 43:121131, 1972
40. Cass RM, Mongan ES, Jacox RF, Vaughan JH: Immu-
HALMA ET AL
noglobulins G, A, and M in systemic lupus erythematosus. Ann Intern Med 69:749-756, 1968
41. Blaese RM, Grayson J, Steinberg AD: Increased immunoglobulin-secreting cells in the blood of patients with
active systemic lupus erythematosus. Am J Med 69:345350, 1980