- Email: [email protected]
Nephelometric detection of circulating immune complexes using monoclonal rheumatoid factor
Journal oflmmunological Methods, 29 (1979) 311--321
311
© Elsevier/North-Holland Biomedical Press
NEPHELOMETRIC DETECTION OF CIRCULATING IMMUNE COMPLEXES USING MONOCLONAL RHEUMATOID FACTOR
HELEN WHITSED, W.H. McCARTHY and P. HERSEY 1
Kanematsu Memorial Institute, Sydney Hospital and (W.H. McC.) Melanoma Unit, Department of Surgery, University of Sydney, Sydney Hospital, Sydney, Australia (Received 5 February 1979, accepted 29 March 1979)
A nephelometric technique for the estimation of immune complexes (IC) in serum was developed using purified monoclonal rheumatoid factor from a human patient ( m R h F ) specific for complexed IgG. Standardisation of the assay was carried out with heat aggregated normal human IgG as a model complex and with IC composed in vitro from ovalbumin and rabbit antisera to ovalbumin. The nephelometric method was compared with [12sI]Clq radioimmunoassay ( C l q RIA). The lower limits of detection by the two methods were similar for both aggregated IgG and preformed ovalbumin/rabbit anti-ovalbumin IC. However, recognition of IC by the two methods differed with different ratios of antigen and antibody. When IC were formed at 10 times antigen excess the nephelometric technique was more sensitive than when IC were formed at equivalence or 10 times antibody excess. The C l q RIA method was most sensitive in detection of IC in antibody excess but failed to detect IC in antigen excess. Complexes formed in antigen excess also showed potentiated light scattering when 1.5% polyethylene glycol was used in the nephelometric system. The incidence of IC detected by the m R h F in sera from patients with rheumatoid arthritis and systemic lupus erythematosus was lower than with C l q RIA suggesting that the IC in these patients contain antibodies not detected by the m R h F used. IC in the sera of patients with melanoma were detected more frequently by the m R h F assay which may indicate the IC in these sera were in antigen excess. Detection of IC by m R h F nephelometry was rapid, technically simple and yielded results which complemented those of the established C l q RIA method. This assay system is a useful addition to methods currently available for detection of IC and the similar use of rheumatoid factors against different classes of antibody should extend its usefulness.
INTRODUCTION
The nephelometric measurement of serum proteins by detection of increased light scatter from an incident laser beam resulting from the formation of specific immune complexes (IC), may be extended to the measurement of pre-existing IC in serum. This has been achieved either by direct
1 Correspondence to: Dr. P. Hersey, Medical Research Department, Kanematsu Memorial Institute, Sydney Hospital, Macquarie Street, Sydney, N.S.W. 2000, Australia.
312 nephelometry (Virella et al., 1978) or by the use of a second reagent, e.g. human C l q , to potentiate the size of the existing IC (Lamberson et al., 1978). The technique described here uses a purified human monoclonal rheumatoid factor (mRhF) as the reagent which recognises complexed IgG antibody. The method compares favourably in sensitivity with the [12sI]Clq radioimmunoassay (Clq RIA) and is applicable to routine sample measurement. This report deals with the parameters optimal for measurement of IC in serum, using heat aggregated IgG (Hagg IgG) dilutions in pooled normal serum and complexes preformed from ovalbumin/rabbit anti-ovalbumin sera as models. Comparison of the results obtained on sera from patients with various diseases with those obtained by the C l q RIA suggests that use of the nephelometer and second reagents against immunoglobulins provides a simple rapid method for detecting IC. MATERIALS AND METHODS
Monoclonal IgM rheumatoid factor (mRhF) The mRhF reagent was purified from serum of a patient with nonHodgkin's lymphoma kindly supplied by Dr. P. Roberts-Thompson of the Flinders University, South Australia (Robert-Thomson and Bradley, 1979). IgM was precipitated from the serum at 4°C as part of a cryoglobulinaemic complex with IgG. The precipitate was washed several times with PBS (pH 7.4) by centrifugation at 4°C and re-dissolved in 0.05 M acetate buffer pH 5.0 to dissociate the IgM/IgG complex. RhF was purified by gel filtration through Sepharose 6B (Pharmacia) in a column 2.6 cm × 100 cm in the same buffer. Purity was checked by Ouchterlony double diffusion and immunoelectrophoresis (IEP) using rabbit antihuman Ig and anti-p chain antisera (Dako immunoglobulins, Copenhagen, Denmark). Protein concentration was measured by the method of Lowry et al. (1951). Normal human IgG and heat aggregated normal human IgG (Hagg IgG) for nephelometry Normal IgG was isolated over Whatman DE-52 anion exchanger from a pool of at least 6 normal sera using sodium phosphate buffer 0.007 M, pH 6.3. Purity of the IgG eluted was confirmed by immunoelectrophoresis against anti-human Ig and anti-7 chain antisera (Dako). Protein concentration was determined by the Lowry method. Aggregation of the IgG was achieved by heating to 63°C for 20 min. Large aggregates were removed by centrifugation (4°C, 2000 X g, 20 min) and the supernatant was applied to a Sepharose 6B column. Elution was with PBS, the first peak eluted being used subsequently as Hagg IgG. Normal human serum pool (NHS), used as background when reading Hagg IgG 'complexes' by nephelometry, comprised serum from at least 6 normal donors. The serum plus 0.02% sodium azide was centrifuged (4 ° C, 2000 X g,
313 30 min}, passed through a 0.45/~m Millipore filter, and stored at 4°C. Heat inactivated normal human serum (HI NHS), used in the [12sI]Clq RIA was the same pool as used for nephelometry, after subjection to 56°C for 30 min. Ovalbumin/rabbit anti-ovalbumin (OA/ROA) complexes. Rabbit antiovalbumin (ROA) antisera were raised by subcutaneous injection of New Zealand White rabbits with 100/~g ovalbumin (Sigma Grade V) in PBS, emulsified with complete Freund's adjuvant (Bacto, Detroit, MI). Two further injections of 100 pg were given at 2-week intervals in incomplete Freund's adjuvant and the animals were bled 2 weeks after the last injection. Equivalence curves for the whole antisera were obtained by nephelometry using 5 pl antiserum and from 0.25 #g to 50 pg ovalbumin in PBS. The antiserum with the highest titre was chosen for preparing IC in quantity at equivalence, 10 times antigen excess (w/v) and 10 times antibody excess.
Preparation of [l:sI]Clq Human Clq w~s isolated and iodinated by the method of Zubler and Lambert (1976), with the modification that before snap freezing for storage at --70°C, 0.02% Tween-20 was added to the [12sI]Clq preparation to minimise self-aggregation. Heat aggregated IgG used for the [12sI]Clq RIA was made from normal immunoglobulin (human) (Commonwealth Serum Laboratories, Melbourne, Australia) which was a 16% w/v gamma-globulin Cohn fraction II preparation. A 20 mg/ml dilution of this protein in PBS was heated at 63°C for 20 min. After centrifugation at 2000 × g for 20 min at 4 ° C, the supernatant was used as aggregated IgG. Serum samples tested for IC Normal serum samples were from 30 volunteer blood donors at the Sydney Central Blood Bank. Serum samples were obtained from patients with a variety of disorders grouped as follows: systemic lupus erythematosus (SLE) (49 patients), rheumatoid arthritis (10 patients) and a miscellaneous group of 35 patients with such disorders as glomerulonephritis and nephrotic syndrome (14), inflammatory eye disorders (3), inflammatory skin diseases (4), polyarteritis nodosa (1), scleroderma (2), Behqet's syndrome (1), myositis (1) and obscure infections (3). Sera were also tested from 43 patients with localized melanoma before and after surgical removal of the tumour. ASSAY PROCEDURES
Nephelometry The Hyland laser nephelometer PDQ was used for this study; specifications of and calibration procedure for the instrument have been described (Deaton et al., 1976). The upper limit of the per cent relative light scattering (% RLS) scale was set in the buffer blank mode at 100% RLS, using a standard reaction mixture of 10 pg RhF and 100 ~ug Hagg IgG in 20 pl NHS
314 made up to a final volume of 1040 pl with PBS. PBS was used as buffer blank and sample diluent, m R h F at the concentration used in each test run was the a n t i b o d y blank. All samples were tested in duplicate, at a final volume of 1040 pl. Hyland LAS-R disposable glass cuvettes were used; a 15 sec scan time was arbitrarily chosen. PBS, m R h F (250 pg/ml PBS) and Hagg IgG (500 pg/ml PBS) were filtered through 0.45 pm Millipore filters before use. Serum samples for IC determination were centrifuged at 2000 X g for 30 min at 4°C to lower background turbidity. Samples were read after incubation at room temperature for 60 min.
['2sI]Cl q radioimmunoassay RIA procedure was according to Zubler and Lambert (1976) except that on thawing an aliquot of the labelled protein for use in the RIA, dilution was carried o u t with 0.3 M NaC1/0.01 M EDTA, pH 7.5 containing 0.02% Tween20. The assay procedure was as follows. 100 pl of 0.2 M EDTA pH 7.5 was added to 50 pl test serum in 10 m m × 75 m m glass tubes and incubated at 37°C for 30 min. Positive controls comprised 50 pl Hagg IgG from 0.01 mg/ ml to 8 mg/ml, to which 50 pl HI NHS and 50 pl 0.2 M EDTA pH 7.5 were added, followed by the same incubation. The normal control w a s 5 0 pl HI NHS alone plus 100 pl EDTA as above. The tubes were transferred to ice and 50 pl of diluted ice cold ['2sI]Clq (previously centrifuged at 18,000 X g, 4°C, 40 min added. This was immediately followed by addition of ice-cold 3% PEG in borate buffer (0.025 M disodium tetraborate/0.1 M boric acid/0.075 M NaC1, pH 8.3). Tubes were left at 0°C w i t h o u t mixing for 60 min, after which they were centrifuged at 1500 × g for 20 min at 4°C. Supernatants were discarded and the precipitates washed with 1.0 ml 3% PEG. TCA precipitates were washed with 2.0 ml 20% TCA. Counts in the precipitates were recorded as percentage of total counts, as determined by precipitation with 1.0 ml 20% TCA of 50 pl ['2sI]Clq in 100 pl HI NHS. RESULTS
(1) Determination of optimal conditions for detection of Hagg IgG (a) Specificity of the mRhF reagent for complexed IgG. Fig. 1 shows the % RLS obtained using m R h F and amounts of IgG or Hagg IgG from 10 to 100 pg. The reagent b o u n d as little as 10 pg Hagg IgG, at which concentration binding with IgG was undetectable. Small positive light scattering shown with the IgG was probably due to the inevitable presence of a small number of spontaneous aggregates. (b ) Effect of RhF concentration on recognition of Hagg IgG. Fig. 1 shows representative results. The % RLS increased at a given concentration of Hagg IgG with increasing amounts of m R h F , e.g., at 100 pg of Hagg IgG the %RLS was 166% with 20 rig R h F compared to 42% RLS with 5 pg of m R h F . At low concentrations the reaction was limited by available m R h F as shown by the
315 200 RhF
ug 20 160
120 t/) ._1
/,'•_.•.
cc
1
~
O ~
~* - -
-~
•
~ SO
/. 0
0
,~0
60
80
100
ug IgG
Fig. 1. Nephelometric measurement of increasing amounts of Hagg IgG complexes using different amounts of RhF reagent (r, r,). Effect of 1.5% PEG on % RLS of Hagg IgG using 10/~g RhF is indicated by the dashed line. Reactions of 10 pg RhF with normal IgG (o o) is also shown.
plateau of % RLS at 20 #g of Hagg IgG when 5 pg of m R h F was added and at 40--60/~g of Hagg IgG when 10/Jg was added. No plateau of % RLS was evident with 20/Jg of m R h F . (c) Effect o f 1.5% PEG. A significant increase in % RLS was seen when 1.5% PEG was present in the m R h F (10/Jg), Hagg IgG reaction system, as shown by the dashed line in Fig. 1.
(2) Effect o f the ratio o f antigen and antibody in IC on the interaction with RhF The equivalence curve for the reaction of OA and ROA was determined by adding varying amounts of OA to 5/J1 of the ROA and recording the point of m a x i m u m % RLS in the nephelometer. This was 3.75/~g of OA for this antiserum. Complexes were then formed in 10 times antibody excess by addition of 50 #1 of ROA to 3.75/Jg OA and in 10 times antigen excess by addition of 37.5/~g OA to 5/J1 of ROA. The interaction of increasing amounts of these different IC with 10/~g of m R h F is shown in Fig. 2A where the % RLS after addition of 10/~g of R h F is plotted against the a m o u n t of a n t i b o d y in the 3 different groups of IC. At
316
60
/, / /
40
m
/
tt}
_J
20 j .° ..-o •"
..I •-'"D'*"
,/
2O
lID
/'
[email protected]'"
O""
;
;, ~ .... ~
io
3'o 4o so
ioo
200 3o0
ut antibody (linlO)
Fig. 2. Detection of ROA/OA complexes by nephelometry using 10 pg RhF (A) and by [12sI]Clq RIA (B). Complexes tested were made at equivalence ( ; ; ) , antibody excess (o . . . . . . $) and antigen excess ( e - - - - - - e ) . Reaction of each of the second reagents with native ROA (1 in 10) used to form the complexes is included (o o). % RLS and % [12sI]Clq precipitation are plotted against volume of antibody (1 in 10) present in the test system.
volumes of ar~.tibody up to 20/zl, IC formed at equivalence gave a higher % RLS than those formed in antibody or antigen excess. Above 20/al the % RLS curve of IC formed at equivalence showed a plateau, indicating mRhF was no longer in excess, while the % RLS curves of the other IC continued to rise. It is notable that IC in 'antigen excess' showed higher % RLS than the complex formed in antibody excess at all volumes in the range tested. Reaction of 10 #g mRhF with native ROA gave less than 1% RLS for volumes of 1 in 10 antiserum up to 250 #1, which was the largest volume of antiserum present in any of the OA/ROA tests.
317 6O
/ 1:10
.
z ~
1;1
/
O m
m
J
f
/
4O
~
O
]u:"
t/3 ,J Q¢
217
!
i
0.1
1.0
ug
ii=w '~
|
,
10
100
antigen
Fig. 3. Effeet of 1.5% PEG on m R h F d e t e c t i o n of R O A / O A IC f o r m e d at equivalence (1 : 1), a n t i b o d y excess (1 : 10) and antigen excess (10 : 1). ; r, c o m p l e x e s tested in PBS: • . . . . . . • c o m p l e x e s tested in presence of 1.5% PEG. % R L S is p l o t t e d against a m o u n t of antigen tested.
(3) Effect of antigen:antibody ratios at complex formation on detection of IC by Clq RIA The Clq precipitation curve with Hagg IgG was similar to that obtained by Zubler et al. (1976). The maximum precipitation of Clq was 65--70% with 400 #g of Hagg IgG in the assay. [12sI]Clq precipitation of 50 gl of pooled normal human sera in the assays was 5% + 1%. The percentage precipitation resulting for each of the 3 OA/ROA complexes
TABLE 1 E F F E C T O F I N C R E A S I N G A M O U N T S O F R h F R E A G E N T ON % R L S O F S E R U M S A M P L E S F R O M P A T I E N T S ' IC Patient
Bo Ri Si AI Ne O'Br Co Ry Bu
Diagnosis
RhA AML Erythema nodosum Myositis SLE SLE SLE RhA RhA
Mean % R L S 5 pg
10 pg
20/~g
10.6 17.6 51.7
40.4 17.2 58.5
56.8 11.4 42.7
0 65.9 5.2 0 0 19.2
6.8 62.3 10.4 27.6 6.3 26.5
32.7 54.7 18.3 44.4 46.7 28.0
318
is shown in Fig. 2B which also compares these values with the % RLS obtained after interaction of the complexes with mRhF shown in Fig. 2A. The complexes formed at equivalence and in antibody excess showed comparable precipitation but the complexes in antigen excess showed negligible C l q precipitation, indistinguishable from that obtained with the antiserum to OA alone. This contrasted with the results obtained after interaction with mRhF where complexes in antigen excess were readily detected in the nephelometer.
(4) Effect of 1.5% PEG The effect of 1.5% PEG on the detection of IC at different ratios of OA/ROA is illustrated in Fig. 3 where % RLS is plotted against the amount of antigen in the reaction mixture. A significant increase in % RLS was detected in solutions containing IC in antigen excess when 1.5% PEG was m
o
4o o
o
o
0_
u 20
o°
•
o Z L!J (--)
o o
:
• •
•
o
:
W
NORMAL
(30)
S LE
~h A
(49)
(lO)
MELAN
(84)
MISC,
(29)
Fig. 4. C o m p a r i s o n o f results o f assay for IC in sera from normal subjects and patients w i t h the disorders indicated, e , % R L S ; ©, % [ 1 2 s I ] C l q precipitation. Means -+ 2 S.D. o f the results for normal subjects are indicated by the solid and dashed lines respeetively.
319 TABLE 2 CORRELATION OF RESULTS OF Clq RIA AND mRhF ASSAYS FOR IC IN SERA FROM NORMAL SUBJECTS AND PATIENTS Results of assays
Groups of subjects studied
Clq RIA
Normal (30)
mRhF
SLE (49)
Melanoma a (84)
+
-
1
5
5
-
+
2
2
--
8
+
+
-
4
4
-
--
27
11
RhA (10)
32
1
71
Miscellaneous (30) 4 2 --
24
a 43 patients.
p r e s e n t . N o change in % I~LS was o b s e r v e d in s o l u t i o n s c o n t a i n i n g IC at e q u i v a l e n c e or a n t i b o d y excess in t h e p r e s e n c e o f 1.5% P E G .
(5) Detection o f IC in sera o f normal subjects and patients T o d e t e r m i n e t h e o p t i m a l m R h F c o n c e n t r a t i o n t o d e t e c t IC in p a t i e n t s ' sera, 5, 10 a n d 20 pg o f m R h F were a d d e d t o sera giving positive results w h e n 10 pg m R h F was used. T h e results s h o w n in T a b l e 1 i n d i c a t e t h a t 20 pg gave t h e highest % R L S w i t h m o s t o f t h e sera b u t in all instances t h e results w e r e clearly positive w i t h 10 #g o f m R h F . This a m o u n t was t h e r e f o r e used in s u b s e q u e n t assays o n p a t i e n t s ' sera. T h e results o f assays o n n o r m a l a n d p a t i e n t s ' sera are s h o w n in Fig. 4. T h e m e a n value f o r t h e C l q R I A o n t h e n o r m a l s u b j e c t s was 5.0%, S.D. 1.17. Sera w e r e c o n s i d e r e d positive f o r IC w h e n t h e values were a b o v e 2 S.D. f r o m t h e m e a n , i.e., g r e a t e r t h a n 7.3% p r e c i p i t a t i o n . T h e m e a n value b y t h e m R h F assay was 4.4, S.D. 3.2% R L S . Sera w e r e c o n s i d e r e d positive f o r IC w h e n t h e values w e r e greater t h a n 2 S.D. a b o v e t h e m e a n , i.e., g r e a t e r t h a n 10.8% RLS. T h e results o f these studies a n d t h e degree o f c o n c o r d a n c e b e t w e e n t h e 2 assays f o r t h e d i f f e r e n t p a t i e n t g r o u p s are s h o w n in T a b l e 2. T h e C l q R I A p r o v e d m o r e sensitive f o r d e t e c t i o n o f IC in t h e sera o f p a t i e n t s w i t h S L E , R h A a n d t h e m i s c e l l a n e o u s disorders. H o w e v e r , in sera f r o m m e l a n o m a p a t i e n t s t h e r e was a higher n u m b e r o f positive results w i t h t h e m R h F assay. T h e r e was s o m e c o n c o r d a n c e o f results on sera f r o m p a t i e n t s w i t h S L E a n d R h A in t h a t w h e n t h e m R h F assay was positive t h e C l q R I A was also positive { e x c e p t f o r results o n sera f r o m 2 S L E p a t i e n t s w h e r e t h e m R h F assay gave positive results w h e n t h e C l q R I A was negative). N o c o n c o r d a n c e was seen in t h e results o f t h e 2 assays o n p a t i e n t s w i t h m i s c e l l a n e o u s disorders or in m e l a n o m a p a t i e n t s .
320 DISCUSSION Interaction of R h F with immunoglobulins in IC has been used as the basis of several assays for IC (Delire and Masson, 1977; Levinsky and Soothill, 1977). Monoclonal RhF, as used in this study, has previously been shown to be more sensitive for the detection of IC and heat aggregates of IgG than polyclonal RhF as f o u n d in patients with rheumatoid arthritis (Winchester et al., 1971). The use of the nephelometer offers a direct m e t h o d of detecting the interaction of m R h F with IC which is rapid, simple to perform, and does n o t require extensive biochemical experience. In the evaluation of this assay it became evident that the ratio of antigen and a n t i b o d y in the complex may be of critical importance in the detection of IC by the various assays available. Hence although IC formed in antigen excess were detected optimally by the m R h F assay they were n o t detected at all by the C l q RIA method. Conversely IC in a n t i b o d y excess were detected optimally by the C l q RIA m e t h o d but less efficiently by the m R h F assay. These results presumably indicate that C l q is prevented by steric hindrance from binding to a n t i b o d y in the presence of excess antigen, but when a n t i b o d y is in excess, binding of C l q m a y proceed unhindered by the steric influence of the antigen. Why this does n o t also apply to binding of m R h F to antibody is unclear but it may be speculated t h a t the determinants for m R h F on the a n t i b o d y molecule remain exposed even in the presence of excess antigen. These results are relevant to understanding the discrepancy in results noted in assays on sera from patients with various diseases. It was notable t h a t IC in the sera of patients with melanoma were detected more frequently by the nephelometric m R h F assay than by the C l q RIA and that there was no concordance between the results of these two assays on melanoma sera. This would be consistent with the idea t h a t IC in melanoma patients are pred o m i n a n t l y small IC in antigen excess and are n o t detected by the C l q RIA. Conversely in patients with rheumatoid arthritis and SLE the C l q assay was positive more frequently than the nephelometric m R h F assay. This suggests that complexes in these sera were more frequently formed at equivalence or in a n t i b o d y excess. The m R h F assay was less frequently positive in tests on these latter sera which may indicate t h a t the antibody class in the IC was n o t detected by the m R h F used. These latter results suggest therefore t h a t the use of rheumatoid factors raised against different subclasses of a n t i b o d y as described by Levinsky et al. (1978) in these assays is needed to provide a more effective screen for IC in sera and to permit partial analysis of the IC. Rheumatoid factors produced by these methods are currently being evaluated for this purpose. The commercial m e t h o d for measurement of first order IC in the nephelometer incorporated 1.5% PEG in the buffer (Deaton et al., 1976}. Addition of PEG in this concentration had little effect on the detection of IC by m R h F except for complexes formed in antigen excess where an increase in
321
sensitivity was noted. This effect may have been due to a shift to the right in the equivalence point of the antigen antibody reaction resulting in the formation of more IC as reported by Liyana and Hellsing (1974). Addition of PEG to these assays therefore appeared useful in increasing the sensitivity of detection of certain complexes and its use in this way may provide a useful criterion to indicate when IC are in antigen excess. Our data suggest that nephelometric detection of IC after interaction with mRhF offers a simple and practical method of detecting IC in sera of patients. The methodology also appears suitable for analysis of IC in terms of antibody class and may provide information regarding the ratio of antibody and antigen in the complex. The use of this assay may be particularly valuable in patients with malignancy and other clinical conditions where IC are predominantly in antigen excess. ACKNOWLEDGEMENTS
This work was supported by the N.H.&M.R.C. of Australia. We wish to thank Dr. P. Roberts-Thomson of the Flinders Medical Centre, South Australia, for his gift of the mRhF containing serum and for suggesting the use of the nephelometer for this purpose. Dr. A. D'Apice, of the Royal Melbourne Hospital provided helpful assistance in the isolation of Clq. Miss A. Lind provided helpful technical assistance. We also thank A/Prof. R. Penney, St. Vincent's Hospital, Sydney and Dr. J. Webb, Royal North Shore Hospital, Sydney for supplying serum samples from patients. REFERENCES Deaton, C.D., K.W. Maxwell, R.S. Smith and R.L. Creveling, 1976, Clin. Chem. 22, 1465. Delire, M. and P.L. Masson, 1977, Clin. Exp. Immunol. 29, 385. Lamberson, H., N.J. Gilmore and C.K. Osterland, 1978, Fed. Proc. 37, 1668. Levinsky, R.J. and J.F. Soothill, 1977, Clin. Exp. Immunol. 29,428. Liyana, J. and K. Hellsing, 1974, Clin. Chem. 20, 1181. Lowry, O.H., N.J. Rosebrough, A.L. Farr and R.J. Randall, 1951, J. Biol. Chem. 193, 265. Roberts-Thomson, P.J. and Bradley, J. 1979, Clin. Exp. Immunol. In press. Virella, G., H.H. Fudenberg and C.C. Patrick, 1978, Fed. Proc. 37, 1668. Winchester, R.J., H.G. Kunkel and V. Agnello, 1971, J. Exp. Med. 134, 2865. Zubler, R.H. and P.H. Lambert, 1976, in: In Vitro Methods in Cell-mediated and Tumour Immunity, eds. B.R. Bloom and J.R. David (Academic Press, New York), p. 565. Zubler, R.H., G. Lange, P.H. Lambert and P.A. Miescher, 1976, J. Immunol. 116,232.
311
© Elsevier/North-Holland Biomedical Press
NEPHELOMETRIC DETECTION OF CIRCULATING IMMUNE COMPLEXES USING MONOCLONAL RHEUMATOID FACTOR
HELEN WHITSED, W.H. McCARTHY and P. HERSEY 1
Kanematsu Memorial Institute, Sydney Hospital and (W.H. McC.) Melanoma Unit, Department of Surgery, University of Sydney, Sydney Hospital, Sydney, Australia (Received 5 February 1979, accepted 29 March 1979)
A nephelometric technique for the estimation of immune complexes (IC) in serum was developed using purified monoclonal rheumatoid factor from a human patient ( m R h F ) specific for complexed IgG. Standardisation of the assay was carried out with heat aggregated normal human IgG as a model complex and with IC composed in vitro from ovalbumin and rabbit antisera to ovalbumin. The nephelometric method was compared with [12sI]Clq radioimmunoassay ( C l q RIA). The lower limits of detection by the two methods were similar for both aggregated IgG and preformed ovalbumin/rabbit anti-ovalbumin IC. However, recognition of IC by the two methods differed with different ratios of antigen and antibody. When IC were formed at 10 times antigen excess the nephelometric technique was more sensitive than when IC were formed at equivalence or 10 times antibody excess. The C l q RIA method was most sensitive in detection of IC in antibody excess but failed to detect IC in antigen excess. Complexes formed in antigen excess also showed potentiated light scattering when 1.5% polyethylene glycol was used in the nephelometric system. The incidence of IC detected by the m R h F in sera from patients with rheumatoid arthritis and systemic lupus erythematosus was lower than with C l q RIA suggesting that the IC in these patients contain antibodies not detected by the m R h F used. IC in the sera of patients with melanoma were detected more frequently by the m R h F assay which may indicate the IC in these sera were in antigen excess. Detection of IC by m R h F nephelometry was rapid, technically simple and yielded results which complemented those of the established C l q RIA method. This assay system is a useful addition to methods currently available for detection of IC and the similar use of rheumatoid factors against different classes of antibody should extend its usefulness.
INTRODUCTION
The nephelometric measurement of serum proteins by detection of increased light scatter from an incident laser beam resulting from the formation of specific immune complexes (IC), may be extended to the measurement of pre-existing IC in serum. This has been achieved either by direct
1 Correspondence to: Dr. P. Hersey, Medical Research Department, Kanematsu Memorial Institute, Sydney Hospital, Macquarie Street, Sydney, N.S.W. 2000, Australia.
312 nephelometry (Virella et al., 1978) or by the use of a second reagent, e.g. human C l q , to potentiate the size of the existing IC (Lamberson et al., 1978). The technique described here uses a purified human monoclonal rheumatoid factor (mRhF) as the reagent which recognises complexed IgG antibody. The method compares favourably in sensitivity with the [12sI]Clq radioimmunoassay (Clq RIA) and is applicable to routine sample measurement. This report deals with the parameters optimal for measurement of IC in serum, using heat aggregated IgG (Hagg IgG) dilutions in pooled normal serum and complexes preformed from ovalbumin/rabbit anti-ovalbumin sera as models. Comparison of the results obtained on sera from patients with various diseases with those obtained by the C l q RIA suggests that use of the nephelometer and second reagents against immunoglobulins provides a simple rapid method for detecting IC. MATERIALS AND METHODS
Monoclonal IgM rheumatoid factor (mRhF) The mRhF reagent was purified from serum of a patient with nonHodgkin's lymphoma kindly supplied by Dr. P. Roberts-Thompson of the Flinders University, South Australia (Robert-Thomson and Bradley, 1979). IgM was precipitated from the serum at 4°C as part of a cryoglobulinaemic complex with IgG. The precipitate was washed several times with PBS (pH 7.4) by centrifugation at 4°C and re-dissolved in 0.05 M acetate buffer pH 5.0 to dissociate the IgM/IgG complex. RhF was purified by gel filtration through Sepharose 6B (Pharmacia) in a column 2.6 cm × 100 cm in the same buffer. Purity was checked by Ouchterlony double diffusion and immunoelectrophoresis (IEP) using rabbit antihuman Ig and anti-p chain antisera (Dako immunoglobulins, Copenhagen, Denmark). Protein concentration was measured by the method of Lowry et al. (1951). Normal human IgG and heat aggregated normal human IgG (Hagg IgG) for nephelometry Normal IgG was isolated over Whatman DE-52 anion exchanger from a pool of at least 6 normal sera using sodium phosphate buffer 0.007 M, pH 6.3. Purity of the IgG eluted was confirmed by immunoelectrophoresis against anti-human Ig and anti-7 chain antisera (Dako). Protein concentration was determined by the Lowry method. Aggregation of the IgG was achieved by heating to 63°C for 20 min. Large aggregates were removed by centrifugation (4°C, 2000 X g, 20 min) and the supernatant was applied to a Sepharose 6B column. Elution was with PBS, the first peak eluted being used subsequently as Hagg IgG. Normal human serum pool (NHS), used as background when reading Hagg IgG 'complexes' by nephelometry, comprised serum from at least 6 normal donors. The serum plus 0.02% sodium azide was centrifuged (4 ° C, 2000 X g,
313 30 min}, passed through a 0.45/~m Millipore filter, and stored at 4°C. Heat inactivated normal human serum (HI NHS), used in the [12sI]Clq RIA was the same pool as used for nephelometry, after subjection to 56°C for 30 min. Ovalbumin/rabbit anti-ovalbumin (OA/ROA) complexes. Rabbit antiovalbumin (ROA) antisera were raised by subcutaneous injection of New Zealand White rabbits with 100/~g ovalbumin (Sigma Grade V) in PBS, emulsified with complete Freund's adjuvant (Bacto, Detroit, MI). Two further injections of 100 pg were given at 2-week intervals in incomplete Freund's adjuvant and the animals were bled 2 weeks after the last injection. Equivalence curves for the whole antisera were obtained by nephelometry using 5 pl antiserum and from 0.25 #g to 50 pg ovalbumin in PBS. The antiserum with the highest titre was chosen for preparing IC in quantity at equivalence, 10 times antigen excess (w/v) and 10 times antibody excess.
Preparation of [l:sI]Clq Human Clq w~s isolated and iodinated by the method of Zubler and Lambert (1976), with the modification that before snap freezing for storage at --70°C, 0.02% Tween-20 was added to the [12sI]Clq preparation to minimise self-aggregation. Heat aggregated IgG used for the [12sI]Clq RIA was made from normal immunoglobulin (human) (Commonwealth Serum Laboratories, Melbourne, Australia) which was a 16% w/v gamma-globulin Cohn fraction II preparation. A 20 mg/ml dilution of this protein in PBS was heated at 63°C for 20 min. After centrifugation at 2000 × g for 20 min at 4 ° C, the supernatant was used as aggregated IgG. Serum samples tested for IC Normal serum samples were from 30 volunteer blood donors at the Sydney Central Blood Bank. Serum samples were obtained from patients with a variety of disorders grouped as follows: systemic lupus erythematosus (SLE) (49 patients), rheumatoid arthritis (10 patients) and a miscellaneous group of 35 patients with such disorders as glomerulonephritis and nephrotic syndrome (14), inflammatory eye disorders (3), inflammatory skin diseases (4), polyarteritis nodosa (1), scleroderma (2), Behqet's syndrome (1), myositis (1) and obscure infections (3). Sera were also tested from 43 patients with localized melanoma before and after surgical removal of the tumour. ASSAY PROCEDURES
Nephelometry The Hyland laser nephelometer PDQ was used for this study; specifications of and calibration procedure for the instrument have been described (Deaton et al., 1976). The upper limit of the per cent relative light scattering (% RLS) scale was set in the buffer blank mode at 100% RLS, using a standard reaction mixture of 10 pg RhF and 100 ~ug Hagg IgG in 20 pl NHS
314 made up to a final volume of 1040 pl with PBS. PBS was used as buffer blank and sample diluent, m R h F at the concentration used in each test run was the a n t i b o d y blank. All samples were tested in duplicate, at a final volume of 1040 pl. Hyland LAS-R disposable glass cuvettes were used; a 15 sec scan time was arbitrarily chosen. PBS, m R h F (250 pg/ml PBS) and Hagg IgG (500 pg/ml PBS) were filtered through 0.45 pm Millipore filters before use. Serum samples for IC determination were centrifuged at 2000 X g for 30 min at 4°C to lower background turbidity. Samples were read after incubation at room temperature for 60 min.
['2sI]Cl q radioimmunoassay RIA procedure was according to Zubler and Lambert (1976) except that on thawing an aliquot of the labelled protein for use in the RIA, dilution was carried o u t with 0.3 M NaC1/0.01 M EDTA, pH 7.5 containing 0.02% Tween20. The assay procedure was as follows. 100 pl of 0.2 M EDTA pH 7.5 was added to 50 pl test serum in 10 m m × 75 m m glass tubes and incubated at 37°C for 30 min. Positive controls comprised 50 pl Hagg IgG from 0.01 mg/ ml to 8 mg/ml, to which 50 pl HI NHS and 50 pl 0.2 M EDTA pH 7.5 were added, followed by the same incubation. The normal control w a s 5 0 pl HI NHS alone plus 100 pl EDTA as above. The tubes were transferred to ice and 50 pl of diluted ice cold ['2sI]Clq (previously centrifuged at 18,000 X g, 4°C, 40 min added. This was immediately followed by addition of ice-cold 3% PEG in borate buffer (0.025 M disodium tetraborate/0.1 M boric acid/0.075 M NaC1, pH 8.3). Tubes were left at 0°C w i t h o u t mixing for 60 min, after which they were centrifuged at 1500 × g for 20 min at 4°C. Supernatants were discarded and the precipitates washed with 1.0 ml 3% PEG. TCA precipitates were washed with 2.0 ml 20% TCA. Counts in the precipitates were recorded as percentage of total counts, as determined by precipitation with 1.0 ml 20% TCA of 50 pl ['2sI]Clq in 100 pl HI NHS. RESULTS
(1) Determination of optimal conditions for detection of Hagg IgG (a) Specificity of the mRhF reagent for complexed IgG. Fig. 1 shows the % RLS obtained using m R h F and amounts of IgG or Hagg IgG from 10 to 100 pg. The reagent b o u n d as little as 10 pg Hagg IgG, at which concentration binding with IgG was undetectable. Small positive light scattering shown with the IgG was probably due to the inevitable presence of a small number of spontaneous aggregates. (b ) Effect of RhF concentration on recognition of Hagg IgG. Fig. 1 shows representative results. The % RLS increased at a given concentration of Hagg IgG with increasing amounts of m R h F , e.g., at 100 pg of Hagg IgG the %RLS was 166% with 20 rig R h F compared to 42% RLS with 5 pg of m R h F . At low concentrations the reaction was limited by available m R h F as shown by the
315 200 RhF
ug 20 160
120 t/) ._1
/,'•_.•.
cc
1
~
O ~
~* - -
-~
•
~ SO
/. 0
0
,~0
60
80
100
ug IgG
Fig. 1. Nephelometric measurement of increasing amounts of Hagg IgG complexes using different amounts of RhF reagent (r, r,). Effect of 1.5% PEG on % RLS of Hagg IgG using 10/~g RhF is indicated by the dashed line. Reactions of 10 pg RhF with normal IgG (o o) is also shown.
plateau of % RLS at 20 #g of Hagg IgG when 5 pg of m R h F was added and at 40--60/~g of Hagg IgG when 10/Jg was added. No plateau of % RLS was evident with 20/Jg of m R h F . (c) Effect o f 1.5% PEG. A significant increase in % RLS was seen when 1.5% PEG was present in the m R h F (10/Jg), Hagg IgG reaction system, as shown by the dashed line in Fig. 1.
(2) Effect o f the ratio o f antigen and antibody in IC on the interaction with RhF The equivalence curve for the reaction of OA and ROA was determined by adding varying amounts of OA to 5/J1 of the ROA and recording the point of m a x i m u m % RLS in the nephelometer. This was 3.75/~g of OA for this antiserum. Complexes were then formed in 10 times antibody excess by addition of 50 #1 of ROA to 3.75/Jg OA and in 10 times antigen excess by addition of 37.5/~g OA to 5/J1 of ROA. The interaction of increasing amounts of these different IC with 10/~g of m R h F is shown in Fig. 2A where the % RLS after addition of 10/~g of R h F is plotted against the a m o u n t of a n t i b o d y in the 3 different groups of IC. At
316
60
/, / /
40
m
/
tt}
_J
20 j .° ..-o •"
..I •-'"D'*"
,/
2O
lID
/'
[email protected]'"
O""
;
;, ~ .... ~
io
3'o 4o so
ioo
200 3o0
ut antibody (linlO)
Fig. 2. Detection of ROA/OA complexes by nephelometry using 10 pg RhF (A) and by [12sI]Clq RIA (B). Complexes tested were made at equivalence ( ; ; ) , antibody excess (o . . . . . . $) and antigen excess ( e - - - - - - e ) . Reaction of each of the second reagents with native ROA (1 in 10) used to form the complexes is included (o o). % RLS and % [12sI]Clq precipitation are plotted against volume of antibody (1 in 10) present in the test system.
volumes of ar~.tibody up to 20/zl, IC formed at equivalence gave a higher % RLS than those formed in antibody or antigen excess. Above 20/al the % RLS curve of IC formed at equivalence showed a plateau, indicating mRhF was no longer in excess, while the % RLS curves of the other IC continued to rise. It is notable that IC in 'antigen excess' showed higher % RLS than the complex formed in antibody excess at all volumes in the range tested. Reaction of 10 #g mRhF with native ROA gave less than 1% RLS for volumes of 1 in 10 antiserum up to 250 #1, which was the largest volume of antiserum present in any of the OA/ROA tests.
317 6O
/ 1:10
.
z ~
1;1
/
O m
m
J
f
/
4O
~
O
]u:"
t/3 ,J Q¢
217
!
i
0.1
1.0
ug
ii=w '~
|
,
10
100
antigen
Fig. 3. Effeet of 1.5% PEG on m R h F d e t e c t i o n of R O A / O A IC f o r m e d at equivalence (1 : 1), a n t i b o d y excess (1 : 10) and antigen excess (10 : 1). ; r, c o m p l e x e s tested in PBS: • . . . . . . • c o m p l e x e s tested in presence of 1.5% PEG. % R L S is p l o t t e d against a m o u n t of antigen tested.
(3) Effect of antigen:antibody ratios at complex formation on detection of IC by Clq RIA The Clq precipitation curve with Hagg IgG was similar to that obtained by Zubler et al. (1976). The maximum precipitation of Clq was 65--70% with 400 #g of Hagg IgG in the assay. [12sI]Clq precipitation of 50 gl of pooled normal human sera in the assays was 5% + 1%. The percentage precipitation resulting for each of the 3 OA/ROA complexes
TABLE 1 E F F E C T O F I N C R E A S I N G A M O U N T S O F R h F R E A G E N T ON % R L S O F S E R U M S A M P L E S F R O M P A T I E N T S ' IC Patient
Bo Ri Si AI Ne O'Br Co Ry Bu
Diagnosis
RhA AML Erythema nodosum Myositis SLE SLE SLE RhA RhA
Mean % R L S 5 pg
10 pg
20/~g
10.6 17.6 51.7
40.4 17.2 58.5
56.8 11.4 42.7
0 65.9 5.2 0 0 19.2
6.8 62.3 10.4 27.6 6.3 26.5
32.7 54.7 18.3 44.4 46.7 28.0
318
is shown in Fig. 2B which also compares these values with the % RLS obtained after interaction of the complexes with mRhF shown in Fig. 2A. The complexes formed at equivalence and in antibody excess showed comparable precipitation but the complexes in antigen excess showed negligible C l q precipitation, indistinguishable from that obtained with the antiserum to OA alone. This contrasted with the results obtained after interaction with mRhF where complexes in antigen excess were readily detected in the nephelometer.
(4) Effect of 1.5% PEG The effect of 1.5% PEG on the detection of IC at different ratios of OA/ROA is illustrated in Fig. 3 where % RLS is plotted against the amount of antigen in the reaction mixture. A significant increase in % RLS was detected in solutions containing IC in antigen excess when 1.5% PEG was m
o
4o o
o
o
0_
u 20
o°
•
o Z L!J (--)
o o
:
• •
•
o
:
W
NORMAL
(30)
S LE
~h A
(49)
(lO)
MELAN
(84)
MISC,
(29)
Fig. 4. C o m p a r i s o n o f results o f assay for IC in sera from normal subjects and patients w i t h the disorders indicated, e , % R L S ; ©, % [ 1 2 s I ] C l q precipitation. Means -+ 2 S.D. o f the results for normal subjects are indicated by the solid and dashed lines respeetively.
319 TABLE 2 CORRELATION OF RESULTS OF Clq RIA AND mRhF ASSAYS FOR IC IN SERA FROM NORMAL SUBJECTS AND PATIENTS Results of assays
Groups of subjects studied
Clq RIA
Normal (30)
mRhF
SLE (49)
Melanoma a (84)
+
-
1
5
5
-
+
2
2
--
8
+
+
-
4
4
-
--
27
11
RhA (10)
32
1
71
Miscellaneous (30) 4 2 --
24
a 43 patients.
p r e s e n t . N o change in % I~LS was o b s e r v e d in s o l u t i o n s c o n t a i n i n g IC at e q u i v a l e n c e or a n t i b o d y excess in t h e p r e s e n c e o f 1.5% P E G .
(5) Detection o f IC in sera o f normal subjects and patients T o d e t e r m i n e t h e o p t i m a l m R h F c o n c e n t r a t i o n t o d e t e c t IC in p a t i e n t s ' sera, 5, 10 a n d 20 pg o f m R h F were a d d e d t o sera giving positive results w h e n 10 pg m R h F was used. T h e results s h o w n in T a b l e 1 i n d i c a t e t h a t 20 pg gave t h e highest % R L S w i t h m o s t o f t h e sera b u t in all instances t h e results w e r e clearly positive w i t h 10 #g o f m R h F . This a m o u n t was t h e r e f o r e used in s u b s e q u e n t assays o n p a t i e n t s ' sera. T h e results o f assays o n n o r m a l a n d p a t i e n t s ' sera are s h o w n in Fig. 4. T h e m e a n value f o r t h e C l q R I A o n t h e n o r m a l s u b j e c t s was 5.0%, S.D. 1.17. Sera w e r e c o n s i d e r e d positive f o r IC w h e n t h e values were a b o v e 2 S.D. f r o m t h e m e a n , i.e., g r e a t e r t h a n 7.3% p r e c i p i t a t i o n . T h e m e a n value b y t h e m R h F assay was 4.4, S.D. 3.2% R L S . Sera w e r e c o n s i d e r e d positive f o r IC w h e n t h e values w e r e greater t h a n 2 S.D. a b o v e t h e m e a n , i.e., g r e a t e r t h a n 10.8% RLS. T h e results o f these studies a n d t h e degree o f c o n c o r d a n c e b e t w e e n t h e 2 assays f o r t h e d i f f e r e n t p a t i e n t g r o u p s are s h o w n in T a b l e 2. T h e C l q R I A p r o v e d m o r e sensitive f o r d e t e c t i o n o f IC in t h e sera o f p a t i e n t s w i t h S L E , R h A a n d t h e m i s c e l l a n e o u s disorders. H o w e v e r , in sera f r o m m e l a n o m a p a t i e n t s t h e r e was a higher n u m b e r o f positive results w i t h t h e m R h F assay. T h e r e was s o m e c o n c o r d a n c e o f results on sera f r o m p a t i e n t s w i t h S L E a n d R h A in t h a t w h e n t h e m R h F assay was positive t h e C l q R I A was also positive { e x c e p t f o r results o n sera f r o m 2 S L E p a t i e n t s w h e r e t h e m R h F assay gave positive results w h e n t h e C l q R I A was negative). N o c o n c o r d a n c e was seen in t h e results o f t h e 2 assays o n p a t i e n t s w i t h m i s c e l l a n e o u s disorders or in m e l a n o m a p a t i e n t s .
320 DISCUSSION Interaction of R h F with immunoglobulins in IC has been used as the basis of several assays for IC (Delire and Masson, 1977; Levinsky and Soothill, 1977). Monoclonal RhF, as used in this study, has previously been shown to be more sensitive for the detection of IC and heat aggregates of IgG than polyclonal RhF as f o u n d in patients with rheumatoid arthritis (Winchester et al., 1971). The use of the nephelometer offers a direct m e t h o d of detecting the interaction of m R h F with IC which is rapid, simple to perform, and does n o t require extensive biochemical experience. In the evaluation of this assay it became evident that the ratio of antigen and a n t i b o d y in the complex may be of critical importance in the detection of IC by the various assays available. Hence although IC formed in antigen excess were detected optimally by the m R h F assay they were n o t detected at all by the C l q RIA method. Conversely IC in a n t i b o d y excess were detected optimally by the C l q RIA m e t h o d but less efficiently by the m R h F assay. These results presumably indicate that C l q is prevented by steric hindrance from binding to a n t i b o d y in the presence of excess antigen, but when a n t i b o d y is in excess, binding of C l q m a y proceed unhindered by the steric influence of the antigen. Why this does n o t also apply to binding of m R h F to antibody is unclear but it may be speculated t h a t the determinants for m R h F on the a n t i b o d y molecule remain exposed even in the presence of excess antigen. These results are relevant to understanding the discrepancy in results noted in assays on sera from patients with various diseases. It was notable t h a t IC in the sera of patients with melanoma were detected more frequently by the nephelometric m R h F assay than by the C l q RIA and that there was no concordance between the results of these two assays on melanoma sera. This would be consistent with the idea t h a t IC in melanoma patients are pred o m i n a n t l y small IC in antigen excess and are n o t detected by the C l q RIA. Conversely in patients with rheumatoid arthritis and SLE the C l q assay was positive more frequently than the nephelometric m R h F assay. This suggests that complexes in these sera were more frequently formed at equivalence or in a n t i b o d y excess. The m R h F assay was less frequently positive in tests on these latter sera which may indicate t h a t the antibody class in the IC was n o t detected by the m R h F used. These latter results suggest therefore t h a t the use of rheumatoid factors raised against different subclasses of a n t i b o d y as described by Levinsky et al. (1978) in these assays is needed to provide a more effective screen for IC in sera and to permit partial analysis of the IC. Rheumatoid factors produced by these methods are currently being evaluated for this purpose. The commercial m e t h o d for measurement of first order IC in the nephelometer incorporated 1.5% PEG in the buffer (Deaton et al., 1976}. Addition of PEG in this concentration had little effect on the detection of IC by m R h F except for complexes formed in antigen excess where an increase in
321
sensitivity was noted. This effect may have been due to a shift to the right in the equivalence point of the antigen antibody reaction resulting in the formation of more IC as reported by Liyana and Hellsing (1974). Addition of PEG to these assays therefore appeared useful in increasing the sensitivity of detection of certain complexes and its use in this way may provide a useful criterion to indicate when IC are in antigen excess. Our data suggest that nephelometric detection of IC after interaction with mRhF offers a simple and practical method of detecting IC in sera of patients. The methodology also appears suitable for analysis of IC in terms of antibody class and may provide information regarding the ratio of antibody and antigen in the complex. The use of this assay may be particularly valuable in patients with malignancy and other clinical conditions where IC are predominantly in antigen excess. ACKNOWLEDGEMENTS
This work was supported by the N.H.&M.R.C. of Australia. We wish to thank Dr. P. Roberts-Thomson of the Flinders Medical Centre, South Australia, for his gift of the mRhF containing serum and for suggesting the use of the nephelometer for this purpose. Dr. A. D'Apice, of the Royal Melbourne Hospital provided helpful assistance in the isolation of Clq. Miss A. Lind provided helpful technical assistance. We also thank A/Prof. R. Penney, St. Vincent's Hospital, Sydney and Dr. J. Webb, Royal North Shore Hospital, Sydney for supplying serum samples from patients. REFERENCES Deaton, C.D., K.W. Maxwell, R.S. Smith and R.L. Creveling, 1976, Clin. Chem. 22, 1465. Delire, M. and P.L. Masson, 1977, Clin. Exp. Immunol. 29, 385. Lamberson, H., N.J. Gilmore and C.K. Osterland, 1978, Fed. Proc. 37, 1668. Levinsky, R.J. and J.F. Soothill, 1977, Clin. Exp. Immunol. 29,428. Liyana, J. and K. Hellsing, 1974, Clin. Chem. 20, 1181. Lowry, O.H., N.J. Rosebrough, A.L. Farr and R.J. Randall, 1951, J. Biol. Chem. 193, 265. Roberts-Thomson, P.J. and Bradley, J. 1979, Clin. Exp. Immunol. In press. Virella, G., H.H. Fudenberg and C.C. Patrick, 1978, Fed. Proc. 37, 1668. Winchester, R.J., H.G. Kunkel and V. Agnello, 1971, J. Exp. Med. 134, 2865. Zubler, R.H. and P.H. Lambert, 1976, in: In Vitro Methods in Cell-mediated and Tumour Immunity, eds. B.R. Bloom and J.R. David (Academic Press, New York), p. 565. Zubler, R.H., G. Lange, P.H. Lambert and P.A. Miescher, 1976, J. Immunol. 116,232.