- Email: [email protected]
Detection and enumeration of immunoglobulin secreting cells
Journal of Immunological Methods, 124 (1989) 35-42 Elsevier
35
JIM 05332
Detection and enumeration of immunoglobulin secreting cells Annette J. Schlueter, Mariangela Segre and Diego Segre Department of Veterinary Pathobiology, Universityof Illinois, 2001 S. Lincoln Avenue, Urbana, IL 61801, US.A. (Received 3 April 1989, revised received 9 June 1989, accepted 13 June 1989)
A method is described in which sheep red blood cells (SRBC) are coated with anti-immunoglobulin (Ig) antibodies (Ab) for use in reverse hemolytic plaque assays (RHPA) as follows. The non-complement fixing F(ab')2 fragments of rabbit anti-mouse Ig Ab are derivatized with the N-hydroxysuccinimide ester of palmitate. The hydrophobic palmitate tails spontaneously insert into the SRBC membranes, thus coating the cells with anti-Ig F(ab')2 molecules. The SRBC are lysed by successive additions of mouse Ig, rabbit anti-mouse Ig and complement. When this procedure is carried out in agar gel, Ig-secreting mouse cells produce localized hemolytic areas (plaques). The procedure is more reproducible and more sensitive than RHPA performed with protein A-coated SRBC. In principle, this procedure should be adaptable to the detection of cells secreting any molecule for which specific antibodies are available. Key words: Reverse plaque; Antibody derivatization; Antigen detection; Protein-coated sheep red blood cell
Introduction
Enumeration of immunoglobulin-secreting cells has commonly been done by the RHPA (Molinaro and Dray, 1974). For this procedure, red blood cells (RBC) have been sensitized in several ways, the most common of which is the use of chromic chloride to bind staphylococcal protein A to the cell surface (Gronowicz et al., 1976). This method does allow visualization of plaque-forming cells (PFC), but variation among preparations of sensitized RBC makes reproducibility of plaque counts difficult (Molinaro et al., 1981).
Correspondence to: D. Segre, Department of Veterinary Pathobiology, University of Illinois, 2001 S. Lincoln Avenue, Urbana, IL 61801, U.S.A. Abbreviations: Ab, antibody; CGG, chicken gamma globulin; DOC, deoxycholate; Ig, immunoglobulin; MGG, mouse gamma globulin; NHS-palmitate, N-(palmitoyloxy) succinimide; PBS, phosphate-buffered saline; PFC, plaque-forming cells; RHPA, reverse hemolytic plaque assay; RBC, red blood cells; SRBC, sheep red blood cells.
In order to overcome these disadvantages and provide a dearer picture of the total number of immunoglobulin-producing cells in a population, a new method of sensitizing RBC was developed. It is based on a protocol previously described for attaching new antigen receptors to the surface of B lymphocytes (Peacock et al., 1986). F(ab')2 fragments of anti-mouse immunoglobulin antibodies were derivatized with palmitate moieties. When the derivatized antibodies are mixed with RBC, they integrate into the cell membranes by virtue of the hydrophobic palmitate moieties. Thus the RBC become sensitized for the detection of lymphocytes secreting mouse immunoglobulin. When this study was initially undertaken, the aim was to develop a method for the detection of autoanti-idiotypic PFC. The potential antigen for these experiments was chicken gamma globulin (CGG), thus much of the preliminary work reported here was performed using anti-CGG-coated sheep RBC (SRBC). In order to validate the new sensitization method, however, RHPA were performed using anti-mouse IgG and IgM-coated SRBC.
0022-1759/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
36 Materials and methods
Animals Female BC3F1 mice were purchased from Harlan Sprague Dawley, Indianapolis, IN. They were from 2 to 3 months of age when used.
Antibodies and antigens Chicken gamma globulin (CGG, no. G-6516) was obtained from Sigma Chemical Co., St. Louis, MO. CGG was also prepared from chicken serum in our laboratory by chromatography on DEAEcellulose. Rabbit anti-mouse gamma globulin (MGG) antiserum and rabbit anti-CGG antiserum were produced in our laboratory. Rabbit anti-CGG F(ab')2 fragments were prepared by pepsin digestion (Nisonoff et al., 1960, 1961) and affinity purified by passage through a column of cyanogen bromide-activated Sepharose 4B (Pharmacia, Uppsala, Sweden) coupled with CGG. The CGGspecific F(ab')z fragments were eluted with 0.5 M glycine-HC1 buffer, pH 2.5, and immediately neutralized with 1 M NaOH. Mouse IgG (no. 015-000-003) and the F(ab')2 fragment of affinity purified goat anti-mouse IgM and IgG (no. 115-006-044) were obtained from Rockland, Gilbertsville, PA.
Preparation of palmitate-derivatized F(ab')2 fragments This method was modified from that of Peacock et al. (1986) as follows: N-(palmitoyloxy) suecinimide (NHS-palmitate, Sigma Chemical Co., St. Louis, MO) was dissolved at a concentration of 7 mg/ml in absolute ethanol by heating in a 55 ° C waterbath for approximately 10 min. 40/~1 of this solution were added to 1.0 ml of 3.75% deoxycholate (DOC, Difco Laboratories, Detroit, MI) in phosphate-buffered saline, pH 7.4 (PBS) preheated to 55 o C. This mixture was cooled to room temperature, and 2.8 ml of a 1.2 mg/ml solution of affinity purified F(ab')2 fragments were added. (These fragments were either rabbit anti-CGG or goat anti-mouse IgM and IgG, depending on the experiment.) The mixture was then incubated in a 37 °C waterbath for 24 h. Following incubation, the palmitate-derivatized F(ab')2 fragments were purified by passage over a Sephadex G-25 column
equilibrated with 0.15% DOC in PBS. The fractions containing the main protein peak from the column were pooled and concentrated to approximately 2 m g / m l by vacuum dialysis versus PBS containing 0.001% DOC and 0.02% sodium azide. After concentration, any precipitate that had formed was removed by centrifugation and the supernatant (containing the derivatized F(ab')2 [ragments) was stored at 4 ° C.
Coating of SRBC with palmitate-derivatized F(ab')e fragments Sheep red blood cells were washed three times in saline and resuspended at 2.5% in Alsever's solution containing 0.02% sodium azide. Varying amounts of palmitate-derivatized F(ab')2 fragments (e.g., 500/~g, 250/~g, 125/tg) were added to 1 ml of 2.5% SRBC, and the mixtures were incubated on a rotating drum at 37 °C for varying periods of time, as described in results section. The SRBC were then washed and resuspended at 2.5% in saline for use in hemolysis or hemagglutination testing, or 10% in saline for use in the reverse hemolytic plaque assay.
Binding of specific antigen by SRBC coated with palmitate-derivatized F(ab ') 2 antibody fragments For each preparation of SRBC coated with palmitate-derivatized rabbit anti-CGG F(ab')2 fragments to be tested, two series of two-fold dilutions of CGG were prepared in 0.2 ml volumes of saline. The initial dilution in each series was approximately 100/~g C G G / m l . 25/xl of a 2.5% suspension of coated SRBC in saline were added to each dilution and allowed to incubate in a 37 °C waterbath for 30 min. The cells were then washed once in saline containing 0.2% gelatin, and resuspended in 0.2 ml of rabbit anti-CGG antiserum diluted 1/1000 in complement buffer (isotonic veronal buffer, pH 7.4, as described by Hoffmann and Mayer (1977)). To determine the hemagglutinating titer, one dilution series was allowed to stand until the SRBC had settled. The other series was used to determine the hemolytic titer by adding to each dilution 25/~1 of guinea pig complement (Pel Freez Clinical Systems, Brown Deer, WI) diluted 1/10 in complement buffer, and incubating the series in a 37 °C waterbath for 30 min before reading by visual inspection. The
37 protocol was identical for SRBC coated with palmitate-derivatized goat anti-mouse IgM and IgG F(ab')2 fragments, except the antigen which was serially diluted was mouse IgG at a starting concentration of 20/zg/ml, and the second antibody was rabbit anti-MGG antiserum diluted 1/1500 in complement buffer.
Coating of SRBC with staphylococcal protein A A modified protocol of Henry and North (1980) was performed. A 6.6 mg/ml stock solution of CrC13 .6H20 (Allied Chemical, New York) in saline was prepared immediately before use. The stock solution was diluted 1/50, 1/100, and 1/200 in saline for incorporation into the coupling mixture. Three solutions of protein A (Sigma Chemical Co., St. Louis, MO) were prepared at final concentrations of 2.0, 1.0, and 0.5 mg/ml in saline. All of the possible permutations of protein A solutions and chromic chloride solutions were mixed with SRBC, which had been washed four times in saline, as follows: one part packed SRBC was resuspended in one part protein A solution, and then ten parts chromic chloride solution were added dropwise to this mixture while it was being vortexed. The final mixture was incubated at 37 o C on a rotating drum for 45 min. The treated cells were then washed four additional times with saline and resuspended at 2.5% in saline after the last wash in order to test their ability to bind immune complexes. Binding of MGG-rabbit anti-MGG immune complexes by protein A-coated SRBC One series of two-fold dilutions of rabbit antiMGG antiserum was prepared in 0.1 ml volumes of saline (starting dilution was 1/100). To each of these dilutions was added 0.1 ml of an 80/zg/ml solution of mouse IgG in saline. This mixture was incubated at room temperature for 15 rain, at which time 25/~1 of a 2.5% suspension of protein A-labeled SRBC in saline were added. The ceils were thoroughly resuspended and then allowed to settle to determine the hemagglutinating titer. Mouse immunization schedule and freezing of immune cells BC3F1 mice were immunized intraperitoneally with 100/zg CGG adsorbed on bentonite (Ritten-
berg and Pratt, 1969). 7 days later, the mice were killed by cervical dislocation and their spleens were removed to obtain lymphocytes for testing in the reverse hemolytic plaque assay. In some cases, the unused portion of spleen cells was resuspended in 1.25 ml RPMI 1640 (Gibco Laboratories, Grand Island, NY) containing 10% dimethyl sulfoxide (Sigma Chemical Co., St. Louis, MO) and 10% fetal calf serum (Hyclone Laboratories, Logan, UT), frozen in aliquots, and stored in liquid nitrogen for future use. Upon thawing, lymphocytes which have been frozen in this manner have been demonstrated to have a viability of approximately 70% and a plaque formation recovery rate of about 50% (Segre et al., 1969).
Reverse hemolytic plaque assay The reverse hemolytic plaque assay was run similarly to the conventional hemolytic plaque assay in use in our laboratory for several years (Miller and Segre, 1972; Segre, M. and Segre, D., 1972). Briefly, mouse spleen cells, either flesh or frozen and thawed at 37°C, were washed and resuspended at an appropriate concentration in Hanks' balanced salt solution (HBSS, Gibco Laboratories, Grand Island, NY). Live cell counts were obtained in all cases with eosin Y (Manufacturing Chemists, Norwood, OH) exclusion, according to the method of Mishell and Shiigi (1980). An agar-gelafin stock solution was prepared by dissolving 1.5 g of purified agar (Difco Laboratories, Detroit, MI) and 1.0 g gelatin (J.T. Baker Chemical Co., Phillipsburg, N J) in 100 ml deionized water by autoclaving at 10 lbs. for 10 rain. This solution was immediately filtered through glass wool and stored at 4 ° C. Just prior to plating out, the agar was liquefied in a boiling waterbath, and then placed in a 49 °C waterbath. Once this temperature was reached, the agar was mixed with an equal volume of 2 × HBSS containing 500 ~tg/ml DEAE-dextran (Pharmacia Fine Chemicals, Uppsala, Sweden), which was also at 49 ° C. All pipets, tubes, and petfi dishes used during this procedure were prewarmed at about 50 o C. Each sample to be tested was plated out in duplicate. 70 /tl of the spleen cell suspension, 70 /zl of the 10% coated SRBC suspension, and 1.4 ml of the agar-HBSS mixture were combined and
38 mixed well in a prewarmed 12 x 75 mm tube. This mixture was divided equally between two 60 × 15 mm petri dishes (no. 1007, Falcon Plastics, Oxnard, CA), which were quickly swirled and placed on a level surface chilled with ice to allow the mixture to harden quickly. Once the agar had solidified, 0.85 ml of rabbit anti-MGG antiserum diluted 1/1000 in HBSS was added to some of the plates. All plates were then incubated for 3 h at 37 °C in a humid atmosphere of 5% CO 2, 95% air. At the end of the incubation period, the antiMGG antiserum (if present) was removed by suction, and 0.85 ml of guinea pig complement which had been absorbed with SRBC and then diluted 1/30 in saline was added to each of the plates. The plates were then returned to the incubator for an additional 30 min. At the end of this period, complement was removed by suction, and the plates were fixed by the addition of 1.2 ml of a 0.3% tannic acid solution in saline. After 5 rain, the plates were washed once with saline. Excess saline was removed by suction, and the plates were then stained by the addition of 0.85 ml of freshly prepared tolidine stain as described by Segre, D. and Segre, M. (1976). Plaques were counted with the aid of a dissecting microscope.
Results
Derivatization of F(ab') 2 fragments using NHSpalmitate The major variables investigated in the preparation of the palmitate-derivatized F(ab')2 fragments were the relative quantities of the reactants and the final concentration of DOC. The process was monitored by running parallel hemagglutination and hemolysis tests on each preparation of reagent. The goal was to find conditions that allowed the preparation of reagents that did not cause SRBC to hemolyze spontaneously and that detected antigen by hemagglutination and by hemolysis with similar sensitivity. The conditions that were finally selected (70/~g NHS-palmitate/ mg F(ab')2 fragments; final DOC concentration of 0.001%) satisfied both requirements (Table I). The presence of DOC in the final preparation is necessary to keep the palmitate-derivatized protein in solution. However, high concentrations of
DOC tend to reduce the sensitivity with which F(ab')2-coated SRBC detect antigen. With 0.001% DOC, the concentration that was finally selected, there was some loss of derivatized protein, as shown by the appearance of a precipitate during the concentration step. The loss was especially severe with goat origin F(ab')2 fragments. Increasing the DOC concentration limited the loss of protein by precipitation, but also limited the amount of antigen detectable with sensitized SRBC. For example, with one preparation of sensitized SRBC, 0.3 /xg of antigen/ml were detected by hemolysis in the presence of 0.001% DOC, but only 5 /~g/ml were detected when the DOC concentration was 0.005%. Derivatized F(ab')2 fragment preparations were stored at 4 ° C rather than at - 2 0 °C to avoid the possibility of further precipitation upon thawing. The stability of a preparation was tested by its ability to bind specific antigen after being stored for 4 months in this fashion. The detectable antigen concentration was essentially identical between the two tests of the preparation, indicating that stability is not greatly affected by storage at 4 ° C, at least within this time frame.
Coating of SRBC with palmitate-derivatized F(ab')2 fragments Coated SRBC were tested for their ability to bind specific antigen and then undergo immunemediated hemolysis or hemagglutination. The resuits of typical tests are shown in Table I. The preparation made with 250 /~g derivatized antibody fragments/ml of SRBC was clearly quite sensitive, in that it was lysed and agglutinated by as little as 80 ng of antigen/ml. Sheep red blood cells coated with 125 #g F(ab')2 fragments/ml were able to detect antigen, but with 16 times less sensitivity than the cells sensitized with the 250 /~g/ml preparations. When the concentration of palmitate-derivatized F(ab')2 fragments was increased above 250 /tg/ml, the coated SRBC often had a tendency to lyse spontaneously, perhaps from overloading the erythrocyte membranes with palmitate residues. Thus 250 /~g was generally the concentration chosen for optimal antigen detection by coated SRBC. The same conditions proved to be optimal for SRBC coated with
39 TABLE I DETECTION OF ANTIGEN BY SRBC COATED WITH PALMITATE-DERIVATIZED RABBIT ANTI-CGG F(ab')2
F(ab')2 fragment
CGG (/~g/ml)
(/tg/ml)
20
10
5
2.5
1.25
0.63
0.31
+ + +
+ + +
+ + +
+ + +
+ + .
+ +
+ .
0.16
0.08
+
+ .
0.04
0
-
-
-
-
Hemagglutination 250 125 62.5 0
.
.
.
.
.
.
.
. .
.
.
.
.
.
.
.
.
.
. .
Hemolysis 250 125 62.5 0
4a 4 4 .
4 4 4 .
4 4 3 .
4 4 2 .
4 2 1 .
4 2 . .
.
4 1
4 .
. .
. .
3 . .
.
. .
. .
.
a The degree of hemolysis was assessed visually: 4 indicates complete hemolysis, 3 indicates 75%, 2 indicates 50%, 1 indicates 25% hemolysis.
palmitate-derivatized goat anti-mouse IgM and I g G F ( a b ' ) 2 f r a g m e n t s (results n o t shown). Initially, all a t t e m p t s to coat S R B C with p a l m i t a t e - d e r i v a t i z e d F ( a b ' ) 2 f r a g m e n t s utilized an overnight i n c u b a t i o n period. I n o r d e r to d e t e r m i n e the effect of shorter i n c u b a t i o n periods, S R B C were c o a t e d with a p r e p a r a t i o n of p a l m i t a t e - d e rivatized r a b b i t a n t i - C G G F ( a b ' ) 2 f r a g m e n t s for 1 h o r overnight. W h i l e a 1 h sensitization p e r i o d a p p e a r e d to b e as effective as overnight i n c u b a tion for c o a t i n g cells with 250 /~g F ( a b ' ) 2 fragm e n t s / m 1 of S R B C ( m i n i m a l d e t e c t a b l e a n t i g e n c o n c e n t r a t i o n was 0.3 / x g / m l vs. 0.15 / ~ g / m l for overnight i n c u b a t i o n ) , it was far less effective at c o a t i n g S R B C w h e n o n l y 125/~g of F ( a b ' ) 2 fragm e n t s were used ( 1 0 / ~ g / m l for 1 h i n c u b a t i o n vs. 1.25 / ~ g / m l for overnight i n c u b a t i o n ) . Since in m a n y instances it was m o r e c o n v e n i e n t to use overnight i n c u b a t i o n , this m e t h o d was r e t a i n e d for use in future e x p e r i m e n t s ; however, 1 h i n c u b a t i o n c o u l d b e s u b s t i t u t e d as long as the F ( a b ' ) 2 fragm e n t c o n c e n t r a t i o n was n o t d e c r e a s e d b e l o w the 2 5 0 / x g level.
Performance of SRBC coated with palmitate-derivatized F(ab')2 fragments in reverse hemolytic plaque assays I n o r d e r to test the a b i l i t y o f S R B C c o a t e d with goat anti-mouse IgM and IgG F(ab')2 fragments to i d e n t i f y i m m u n o g l o b u l i n secreting cells in a reverse h e m o l y t i c p l a q u e assay, a source of i m -
m u n e spleen cells was g e n e r a t e d b y i m m u n i z i n g mice w i t h C G G 7 d a y s b e f o r e the assay was p e r f o r m e d . W h e n these l y m p h o c y t e s were used, p l a q u e s were clearly d e t e c t a b l e as long as d i l u t e d a n t i - M G G a n t i s e r u m was p r e s e n t o n the plates d u r i n g the 3 h i n c u b a t i o n . I f this a n t i s e r u m was absent, however, n o p l a q u e s d e v e l o p e d . A l l further p l a q u e assays p e r f o r m e d with p r e p a r a t i o n s of S R B C c o a t e d with F ( a b ' ) 2 f r a g m e n t s were therefore p e r f o r m e d w i t h a n a n t i - M G G a n t i s e r u m present. T h e r e p r o d u c i b i l i t y of this reverse p l a q u e assay was tested with a single s u s p e n s i o n o f Ig-secreting cells b y using different p r e p a r a t i o n s of d e r i v a t i z e d F ( a b ' ) 2 f r a g m e n t s to l a b e l SRBC. T h e resulting n u m b e r s of p l a q u e s d e t e c t e d v a r i e d b y at m o s t 20%, as s h o w n in T a b l e II.
TABLE II REVERSE PLAQUES AND ANTIBODY PLAQUES OBTAINED FROM SPLEEN CELLS OF CGG-IMMUNIZED MICE SRBC coated with Indirect plaques/106 spleen cells anti-mouse IgM and IgG Prep. 1 Prep. 2 Prep. 3 Chicken anti-SRBC
1948 2 222 2 856 754
40 Plaque assays were also performed in which SRBC coated with chicken anti-SRBC antibodies were used to determine the number of lymphocytes secreting antibodies specific for CGG. Because the number of direct PFC was negligible, only indirect PFC were counted. When the same preparations of lymphocytes were used in the reverse plaque assay with F(ab')2 fragment-coated SRBC, approximately three times as many plaques were detected (Table II). Thus the F(ab')2 fragment-coated SRBC were sensitive enough to detect many more immunoglobulin-secreting cells than were SRBC which had been coated with C G G in order to detect only lymphocytes secreting antibodies specific for the immunogen. In order to compare the performance of F(ab')2 fragment-coated SRBC to the most widely used sensitization method for erythrocytes in the reverse plaque assay, several aliquots of SRBC were coated with protein A. Different concentrations of protein A and chromic chloride were used in the sensitization procedure in order to obtain an optimal preparation, and each batch was initially tested in a hemagglutination assay to determine if any coating had occurred. The results of these tests are shown in Table III. Many preparations did not show any evidence of coating at all, and others, such as batch G, showed too much coating, resulting in spontaneous agglutination of the SRBC. Preparations which showed behavior similar to that of batches D and H were selected for use in a
TABLE III HEMAGGLUTINATIONTEST OF SRBC COATEDWITH PROTEIN A Prep. A B C D E F G H I-
CrC13conc. (/~g/ml) 132 66 33 132 66 33 132 66 33
ProteinA conc. (mg/ml)
HA titer
0.1 0.1 0.1 0.5 0.5 0.5 2.0 2.0 2.0
1 : 1600 <1:400 <1:400 1 : 3200 <1:400 < 1:400 > 1 : 6400 1 : 3200 < 1 : 400
a The control (without immune complexes)was agglutinated as well.
TABLE IV COMPARISON OF REVERSE PLAQUE NUMBERS OBTAINED WITH PROTEIN A-COATED SRBC AND PALMITATE-DERIVATIZEDGOAT ANTI-MOUSE IgM AND IgG F(ab')2-COATED SRBC SRBC coated with protein A
Plaques/lO 6 spleen cells a
Prep. 1 Prep. 2 Prep. 3
1113 1295 521
Anti-mouse IgM and IgG
2 342
a
Spleen cells were obtained from mice immunized with CGG.
comparative reverse plaque assay with F(ab')2 fragment sensitized SRBC. The results of this study are shown in Table IV. In addition to detecting fewer antibody secreting cells, protein A-coated SRBC formed plaques that were fuzzier and more difficult to count.
Discussion Attempts to develop an improved procedure for coating SRBC with antibodies were modeled after the technique described by Peacock et al. (1986). When NHS-palmitate is reacted with antibodies under the appropriate conditions, the succinimide moiety on the palmitate is irreversibly acylated to primary amine groups, such as lysine residues, on the antibody fragments. A number of trials were performed to determine the optimal conditions under which this derivatization reaction occurs. The concentration of D O C during derivatization of the antibody fragments, subsequent removal of unreacted NHS-palmitate by Sephadex G-25 chromatography, and sensitization of SRBC is critical. The D O C concentration must be sufficient to maintain the NHS-palmitate, and later the derivatized antibody in solution, but not high enough to lyse the SRBC or interfere with their sensitization. In these trials, F(ab')2 fragments of antibodies were used rather than intact antibody molecules in order to avoid the possibility of activation of complement by the derivatized molecules coating the SRBC, which potentially would have resulted
41 in generalized lysis of the SRBC, thus negating plaque formation by the Ig-secreting cells. However, more recent trials have shown that SRBC coated with palmitate-derivatized whole antibody molecules do not lyse upon addition of complement. Therefore, either F(ab')2 fragments or intact antibody molecules can be used in the RHPA. Tests of SRBC sensitized with various preparations of derivatized goat anti-MGG indicate that the reverse plaque assay is quite reproducible. However, no plaques were detected unless a second antibody was present during the 3 h incubation. This is not unexpected, since essentially all reverse plaque assays described in the literature note this requirement. The second antibody may be necessary since many of the immunoglobulin molecules secreted by the plaque-forming cells are probably attached to the antibody-coated SRBC by their Fc regions rather than by their antigencombining sites, as is the case in the conventional plaque assay. This could make the Fc regions less accessible for complement fixation, decreasing the efficiency of lysis of the SRBC. A comparison of the number of plaques obtainable with the new reverse plaque procedure and a conventional hemolytic plaque procedure detecting antigen-specific plaques was performed using spleen cells from mice immunized with CGG, Approximately three times as many plaques were detected by the reverse plaque procedure, which may be explained if the affinities of the antibodies produced by many of the cells activated during immunization are too low to allow detection of plaque-forming cells in a conventional plaque assay. Alternatively, some of the excess plaques may result from lymphocytes secreting immunoglobulin of irrelevant specificity, as postulated by the idiotypic network theory (Jeme, 1974). Finally, a comparison was performed between F(ab')2 fragment-sensitized SRBC and protein A-sensitized SRBC, the latter being the most widely used detection system for reverse plaques. F(ab')2 fragment-sensitized cells detected almost twice as many lymphocytes as did the best preparation of protein A-sensitized SRBC. In addition, the variability in number of plaques detected was far greater among preparations of protein A-coated cells than among preparations of F(ab') 2 fragment-sensitized SRBC. A lack of con-
sistency has often been a criticism of the protein A reverse plaque method, and it appears that the F(ab')2 fragment sensitization method overcomes this drawback. The ability to coat SRBC with F(ab'): fragments of any antigenic specificity allows a potentially broad application of this methodology for the detection and enumeration of cells secreting almost any antigenic molecule. The only absolute requirement is that the molecule must have at least two exposed epitopes, one for binding to the coated SRBC and the other for attachment of the second antibody. Thus cells secreting interferons, interleukins, or viruses, for example, should be detectable by this technique. In addition, the method should be adaptable for the detection of plasma cells secreting anti-idiotypic antibodies, which was the original purpose of this study. Several of these applications are currently under development.
Acknowledgement A.J.S. is supported by a Lutheran Brotherhood fellowship.
References Gronowicz, E., Coutinho, A. and Melchers, F. (1976) A plaque assay for all cells secretingimmunoglobulinof a given type or class. Eur. J. Immunol. 6, 588. Hoffmann, L.G. and Mayer, M.M. (1977) Immune hemolysis and complement fixation. In: C.A. Williams and M.W. Chase (Eds.), Methods in Immunologyand Immunochemistry, Vol. IV. Academic Press, New York, pp. 137-166. Jerne, N.K. (1974) Toward a network theory of the immune system. Ann. Immunol. 125C, 373. Miller, G.W. and Segre, D. (1972) Determination of the relative affinity and heterogeneityof mouse anti-DNP antibodies by a plaque inhibition technique. J. Immunol. 109, 74. Mishell, B.B. and Shiigi, S.M. (1980) Preparation of mouse cell suspensions. In: B.B. Mishell and S.M. Shiigi (Eds.), Selected Methods in Cellular Immunology. Freeman, San Francisco, CA, p. 3. Molinaro, G.A. and Dray, S. (1974) Antibody coated erythrocytes as a manifold probe for antigens. Nature 248, 515. Molinaro, G.A., Eby, W.C. and Molinaro, C.A. (1981) The reverseplaque-forming cell assay. Methods Enzymol. 73, 326.
42 Nisonoff, A., Wissler, F.C., Lipman, L.N. and Woernley, D.L. (1960) Separation of univalent fragments from the bivalent rabbit antibody molecule by reduction of disulfide bonds. Arch. Biochem. Biophys. 89, 230. Nisonoff, A., Markus, G. and Wissler, F.C. (1961) Separation of univalent fragments of rabbit antibody by reduction of a single, labile disulfide bond. Nature 189, 293. Peacock, J.S., Londo, T.R., Roess, D.A. and Barisas, G. (1986) Biologic activity of antigen receptors artificially incorporated onto B lymphocytes. J. Immunol. 137, 1916. Rittenberg, M.B. and Pratt, K.L. (1969) Antitrinitrophenyl (TNP) plaque assay. Primary response of BALB/c mice to
soluble and particulate immunogen. Proc. Soc. Exp. Biol. Med. 132, 575. Segre, D. and Segre, M. (1976) Visualization of plaque forming cells in agar plates stained with o-tolidine. J. Immunol. Methods 12, 197. Segre, M. and Segre, D. (1972) Anti-DNP hemolytic plaques by mouse spleen cells in diffusion chambers. Immunol. Commun. 1, 143. Segre, M., Segre, D. and Inman, F.P. (1969) Comparison of Aal allotypic specificity carried by rabbit IgG and IgM. J. Immunol. 102, 1368.
35
JIM 05332
Detection and enumeration of immunoglobulin secreting cells Annette J. Schlueter, Mariangela Segre and Diego Segre Department of Veterinary Pathobiology, Universityof Illinois, 2001 S. Lincoln Avenue, Urbana, IL 61801, US.A. (Received 3 April 1989, revised received 9 June 1989, accepted 13 June 1989)
A method is described in which sheep red blood cells (SRBC) are coated with anti-immunoglobulin (Ig) antibodies (Ab) for use in reverse hemolytic plaque assays (RHPA) as follows. The non-complement fixing F(ab')2 fragments of rabbit anti-mouse Ig Ab are derivatized with the N-hydroxysuccinimide ester of palmitate. The hydrophobic palmitate tails spontaneously insert into the SRBC membranes, thus coating the cells with anti-Ig F(ab')2 molecules. The SRBC are lysed by successive additions of mouse Ig, rabbit anti-mouse Ig and complement. When this procedure is carried out in agar gel, Ig-secreting mouse cells produce localized hemolytic areas (plaques). The procedure is more reproducible and more sensitive than RHPA performed with protein A-coated SRBC. In principle, this procedure should be adaptable to the detection of cells secreting any molecule for which specific antibodies are available. Key words: Reverse plaque; Antibody derivatization; Antigen detection; Protein-coated sheep red blood cell
Introduction
Enumeration of immunoglobulin-secreting cells has commonly been done by the RHPA (Molinaro and Dray, 1974). For this procedure, red blood cells (RBC) have been sensitized in several ways, the most common of which is the use of chromic chloride to bind staphylococcal protein A to the cell surface (Gronowicz et al., 1976). This method does allow visualization of plaque-forming cells (PFC), but variation among preparations of sensitized RBC makes reproducibility of plaque counts difficult (Molinaro et al., 1981).
Correspondence to: D. Segre, Department of Veterinary Pathobiology, University of Illinois, 2001 S. Lincoln Avenue, Urbana, IL 61801, U.S.A. Abbreviations: Ab, antibody; CGG, chicken gamma globulin; DOC, deoxycholate; Ig, immunoglobulin; MGG, mouse gamma globulin; NHS-palmitate, N-(palmitoyloxy) succinimide; PBS, phosphate-buffered saline; PFC, plaque-forming cells; RHPA, reverse hemolytic plaque assay; RBC, red blood cells; SRBC, sheep red blood cells.
In order to overcome these disadvantages and provide a dearer picture of the total number of immunoglobulin-producing cells in a population, a new method of sensitizing RBC was developed. It is based on a protocol previously described for attaching new antigen receptors to the surface of B lymphocytes (Peacock et al., 1986). F(ab')2 fragments of anti-mouse immunoglobulin antibodies were derivatized with palmitate moieties. When the derivatized antibodies are mixed with RBC, they integrate into the cell membranes by virtue of the hydrophobic palmitate moieties. Thus the RBC become sensitized for the detection of lymphocytes secreting mouse immunoglobulin. When this study was initially undertaken, the aim was to develop a method for the detection of autoanti-idiotypic PFC. The potential antigen for these experiments was chicken gamma globulin (CGG), thus much of the preliminary work reported here was performed using anti-CGG-coated sheep RBC (SRBC). In order to validate the new sensitization method, however, RHPA were performed using anti-mouse IgG and IgM-coated SRBC.
0022-1759/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
36 Materials and methods
Animals Female BC3F1 mice were purchased from Harlan Sprague Dawley, Indianapolis, IN. They were from 2 to 3 months of age when used.
Antibodies and antigens Chicken gamma globulin (CGG, no. G-6516) was obtained from Sigma Chemical Co., St. Louis, MO. CGG was also prepared from chicken serum in our laboratory by chromatography on DEAEcellulose. Rabbit anti-mouse gamma globulin (MGG) antiserum and rabbit anti-CGG antiserum were produced in our laboratory. Rabbit anti-CGG F(ab')2 fragments were prepared by pepsin digestion (Nisonoff et al., 1960, 1961) and affinity purified by passage through a column of cyanogen bromide-activated Sepharose 4B (Pharmacia, Uppsala, Sweden) coupled with CGG. The CGGspecific F(ab')z fragments were eluted with 0.5 M glycine-HC1 buffer, pH 2.5, and immediately neutralized with 1 M NaOH. Mouse IgG (no. 015-000-003) and the F(ab')2 fragment of affinity purified goat anti-mouse IgM and IgG (no. 115-006-044) were obtained from Rockland, Gilbertsville, PA.
Preparation of palmitate-derivatized F(ab')2 fragments This method was modified from that of Peacock et al. (1986) as follows: N-(palmitoyloxy) suecinimide (NHS-palmitate, Sigma Chemical Co., St. Louis, MO) was dissolved at a concentration of 7 mg/ml in absolute ethanol by heating in a 55 ° C waterbath for approximately 10 min. 40/~1 of this solution were added to 1.0 ml of 3.75% deoxycholate (DOC, Difco Laboratories, Detroit, MI) in phosphate-buffered saline, pH 7.4 (PBS) preheated to 55 o C. This mixture was cooled to room temperature, and 2.8 ml of a 1.2 mg/ml solution of affinity purified F(ab')2 fragments were added. (These fragments were either rabbit anti-CGG or goat anti-mouse IgM and IgG, depending on the experiment.) The mixture was then incubated in a 37 °C waterbath for 24 h. Following incubation, the palmitate-derivatized F(ab')2 fragments were purified by passage over a Sephadex G-25 column
equilibrated with 0.15% DOC in PBS. The fractions containing the main protein peak from the column were pooled and concentrated to approximately 2 m g / m l by vacuum dialysis versus PBS containing 0.001% DOC and 0.02% sodium azide. After concentration, any precipitate that had formed was removed by centrifugation and the supernatant (containing the derivatized F(ab')2 [ragments) was stored at 4 ° C.
Coating of SRBC with palmitate-derivatized F(ab')e fragments Sheep red blood cells were washed three times in saline and resuspended at 2.5% in Alsever's solution containing 0.02% sodium azide. Varying amounts of palmitate-derivatized F(ab')2 fragments (e.g., 500/~g, 250/~g, 125/tg) were added to 1 ml of 2.5% SRBC, and the mixtures were incubated on a rotating drum at 37 °C for varying periods of time, as described in results section. The SRBC were then washed and resuspended at 2.5% in saline for use in hemolysis or hemagglutination testing, or 10% in saline for use in the reverse hemolytic plaque assay.
Binding of specific antigen by SRBC coated with palmitate-derivatized F(ab ') 2 antibody fragments For each preparation of SRBC coated with palmitate-derivatized rabbit anti-CGG F(ab')2 fragments to be tested, two series of two-fold dilutions of CGG were prepared in 0.2 ml volumes of saline. The initial dilution in each series was approximately 100/~g C G G / m l . 25/xl of a 2.5% suspension of coated SRBC in saline were added to each dilution and allowed to incubate in a 37 °C waterbath for 30 min. The cells were then washed once in saline containing 0.2% gelatin, and resuspended in 0.2 ml of rabbit anti-CGG antiserum diluted 1/1000 in complement buffer (isotonic veronal buffer, pH 7.4, as described by Hoffmann and Mayer (1977)). To determine the hemagglutinating titer, one dilution series was allowed to stand until the SRBC had settled. The other series was used to determine the hemolytic titer by adding to each dilution 25/~1 of guinea pig complement (Pel Freez Clinical Systems, Brown Deer, WI) diluted 1/10 in complement buffer, and incubating the series in a 37 °C waterbath for 30 min before reading by visual inspection. The
37 protocol was identical for SRBC coated with palmitate-derivatized goat anti-mouse IgM and IgG F(ab')2 fragments, except the antigen which was serially diluted was mouse IgG at a starting concentration of 20/zg/ml, and the second antibody was rabbit anti-MGG antiserum diluted 1/1500 in complement buffer.
Coating of SRBC with staphylococcal protein A A modified protocol of Henry and North (1980) was performed. A 6.6 mg/ml stock solution of CrC13 .6H20 (Allied Chemical, New York) in saline was prepared immediately before use. The stock solution was diluted 1/50, 1/100, and 1/200 in saline for incorporation into the coupling mixture. Three solutions of protein A (Sigma Chemical Co., St. Louis, MO) were prepared at final concentrations of 2.0, 1.0, and 0.5 mg/ml in saline. All of the possible permutations of protein A solutions and chromic chloride solutions were mixed with SRBC, which had been washed four times in saline, as follows: one part packed SRBC was resuspended in one part protein A solution, and then ten parts chromic chloride solution were added dropwise to this mixture while it was being vortexed. The final mixture was incubated at 37 o C on a rotating drum for 45 min. The treated cells were then washed four additional times with saline and resuspended at 2.5% in saline after the last wash in order to test their ability to bind immune complexes. Binding of MGG-rabbit anti-MGG immune complexes by protein A-coated SRBC One series of two-fold dilutions of rabbit antiMGG antiserum was prepared in 0.1 ml volumes of saline (starting dilution was 1/100). To each of these dilutions was added 0.1 ml of an 80/zg/ml solution of mouse IgG in saline. This mixture was incubated at room temperature for 15 rain, at which time 25/~1 of a 2.5% suspension of protein A-labeled SRBC in saline were added. The ceils were thoroughly resuspended and then allowed to settle to determine the hemagglutinating titer. Mouse immunization schedule and freezing of immune cells BC3F1 mice were immunized intraperitoneally with 100/zg CGG adsorbed on bentonite (Ritten-
berg and Pratt, 1969). 7 days later, the mice were killed by cervical dislocation and their spleens were removed to obtain lymphocytes for testing in the reverse hemolytic plaque assay. In some cases, the unused portion of spleen cells was resuspended in 1.25 ml RPMI 1640 (Gibco Laboratories, Grand Island, NY) containing 10% dimethyl sulfoxide (Sigma Chemical Co., St. Louis, MO) and 10% fetal calf serum (Hyclone Laboratories, Logan, UT), frozen in aliquots, and stored in liquid nitrogen for future use. Upon thawing, lymphocytes which have been frozen in this manner have been demonstrated to have a viability of approximately 70% and a plaque formation recovery rate of about 50% (Segre et al., 1969).
Reverse hemolytic plaque assay The reverse hemolytic plaque assay was run similarly to the conventional hemolytic plaque assay in use in our laboratory for several years (Miller and Segre, 1972; Segre, M. and Segre, D., 1972). Briefly, mouse spleen cells, either flesh or frozen and thawed at 37°C, were washed and resuspended at an appropriate concentration in Hanks' balanced salt solution (HBSS, Gibco Laboratories, Grand Island, NY). Live cell counts were obtained in all cases with eosin Y (Manufacturing Chemists, Norwood, OH) exclusion, according to the method of Mishell and Shiigi (1980). An agar-gelafin stock solution was prepared by dissolving 1.5 g of purified agar (Difco Laboratories, Detroit, MI) and 1.0 g gelatin (J.T. Baker Chemical Co., Phillipsburg, N J) in 100 ml deionized water by autoclaving at 10 lbs. for 10 rain. This solution was immediately filtered through glass wool and stored at 4 ° C. Just prior to plating out, the agar was liquefied in a boiling waterbath, and then placed in a 49 °C waterbath. Once this temperature was reached, the agar was mixed with an equal volume of 2 × HBSS containing 500 ~tg/ml DEAE-dextran (Pharmacia Fine Chemicals, Uppsala, Sweden), which was also at 49 ° C. All pipets, tubes, and petfi dishes used during this procedure were prewarmed at about 50 o C. Each sample to be tested was plated out in duplicate. 70 /tl of the spleen cell suspension, 70 /zl of the 10% coated SRBC suspension, and 1.4 ml of the agar-HBSS mixture were combined and
38 mixed well in a prewarmed 12 x 75 mm tube. This mixture was divided equally between two 60 × 15 mm petri dishes (no. 1007, Falcon Plastics, Oxnard, CA), which were quickly swirled and placed on a level surface chilled with ice to allow the mixture to harden quickly. Once the agar had solidified, 0.85 ml of rabbit anti-MGG antiserum diluted 1/1000 in HBSS was added to some of the plates. All plates were then incubated for 3 h at 37 °C in a humid atmosphere of 5% CO 2, 95% air. At the end of the incubation period, the antiMGG antiserum (if present) was removed by suction, and 0.85 ml of guinea pig complement which had been absorbed with SRBC and then diluted 1/30 in saline was added to each of the plates. The plates were then returned to the incubator for an additional 30 min. At the end of this period, complement was removed by suction, and the plates were fixed by the addition of 1.2 ml of a 0.3% tannic acid solution in saline. After 5 rain, the plates were washed once with saline. Excess saline was removed by suction, and the plates were then stained by the addition of 0.85 ml of freshly prepared tolidine stain as described by Segre, D. and Segre, M. (1976). Plaques were counted with the aid of a dissecting microscope.
Results
Derivatization of F(ab') 2 fragments using NHSpalmitate The major variables investigated in the preparation of the palmitate-derivatized F(ab')2 fragments were the relative quantities of the reactants and the final concentration of DOC. The process was monitored by running parallel hemagglutination and hemolysis tests on each preparation of reagent. The goal was to find conditions that allowed the preparation of reagents that did not cause SRBC to hemolyze spontaneously and that detected antigen by hemagglutination and by hemolysis with similar sensitivity. The conditions that were finally selected (70/~g NHS-palmitate/ mg F(ab')2 fragments; final DOC concentration of 0.001%) satisfied both requirements (Table I). The presence of DOC in the final preparation is necessary to keep the palmitate-derivatized protein in solution. However, high concentrations of
DOC tend to reduce the sensitivity with which F(ab')2-coated SRBC detect antigen. With 0.001% DOC, the concentration that was finally selected, there was some loss of derivatized protein, as shown by the appearance of a precipitate during the concentration step. The loss was especially severe with goat origin F(ab')2 fragments. Increasing the DOC concentration limited the loss of protein by precipitation, but also limited the amount of antigen detectable with sensitized SRBC. For example, with one preparation of sensitized SRBC, 0.3 /xg of antigen/ml were detected by hemolysis in the presence of 0.001% DOC, but only 5 /~g/ml were detected when the DOC concentration was 0.005%. Derivatized F(ab')2 fragment preparations were stored at 4 ° C rather than at - 2 0 °C to avoid the possibility of further precipitation upon thawing. The stability of a preparation was tested by its ability to bind specific antigen after being stored for 4 months in this fashion. The detectable antigen concentration was essentially identical between the two tests of the preparation, indicating that stability is not greatly affected by storage at 4 ° C, at least within this time frame.
Coating of SRBC with palmitate-derivatized F(ab')2 fragments Coated SRBC were tested for their ability to bind specific antigen and then undergo immunemediated hemolysis or hemagglutination. The resuits of typical tests are shown in Table I. The preparation made with 250 /~g derivatized antibody fragments/ml of SRBC was clearly quite sensitive, in that it was lysed and agglutinated by as little as 80 ng of antigen/ml. Sheep red blood cells coated with 125 #g F(ab')2 fragments/ml were able to detect antigen, but with 16 times less sensitivity than the cells sensitized with the 250 /~g/ml preparations. When the concentration of palmitate-derivatized F(ab')2 fragments was increased above 250 /tg/ml, the coated SRBC often had a tendency to lyse spontaneously, perhaps from overloading the erythrocyte membranes with palmitate residues. Thus 250 /~g was generally the concentration chosen for optimal antigen detection by coated SRBC. The same conditions proved to be optimal for SRBC coated with
39 TABLE I DETECTION OF ANTIGEN BY SRBC COATED WITH PALMITATE-DERIVATIZED RABBIT ANTI-CGG F(ab')2
F(ab')2 fragment
CGG (/~g/ml)
(/tg/ml)
20
10
5
2.5
1.25
0.63
0.31
+ + +
+ + +
+ + +
+ + +
+ + .
+ +
+ .
0.16
0.08
+
+ .
0.04
0
-
-
-
-
Hemagglutination 250 125 62.5 0
.
.
.
.
.
.
.
. .
.
.
.
.
.
.
.
.
.
. .
Hemolysis 250 125 62.5 0
4a 4 4 .
4 4 4 .
4 4 3 .
4 4 2 .
4 2 1 .
4 2 . .
.
4 1
4 .
. .
. .
3 . .
.
. .
. .
.
a The degree of hemolysis was assessed visually: 4 indicates complete hemolysis, 3 indicates 75%, 2 indicates 50%, 1 indicates 25% hemolysis.
palmitate-derivatized goat anti-mouse IgM and I g G F ( a b ' ) 2 f r a g m e n t s (results n o t shown). Initially, all a t t e m p t s to coat S R B C with p a l m i t a t e - d e r i v a t i z e d F ( a b ' ) 2 f r a g m e n t s utilized an overnight i n c u b a t i o n period. I n o r d e r to d e t e r m i n e the effect of shorter i n c u b a t i o n periods, S R B C were c o a t e d with a p r e p a r a t i o n of p a l m i t a t e - d e rivatized r a b b i t a n t i - C G G F ( a b ' ) 2 f r a g m e n t s for 1 h o r overnight. W h i l e a 1 h sensitization p e r i o d a p p e a r e d to b e as effective as overnight i n c u b a tion for c o a t i n g cells with 250 /~g F ( a b ' ) 2 fragm e n t s / m 1 of S R B C ( m i n i m a l d e t e c t a b l e a n t i g e n c o n c e n t r a t i o n was 0.3 / x g / m l vs. 0.15 / ~ g / m l for overnight i n c u b a t i o n ) , it was far less effective at c o a t i n g S R B C w h e n o n l y 125/~g of F ( a b ' ) 2 fragm e n t s were used ( 1 0 / ~ g / m l for 1 h i n c u b a t i o n vs. 1.25 / ~ g / m l for overnight i n c u b a t i o n ) . Since in m a n y instances it was m o r e c o n v e n i e n t to use overnight i n c u b a t i o n , this m e t h o d was r e t a i n e d for use in future e x p e r i m e n t s ; however, 1 h i n c u b a t i o n c o u l d b e s u b s t i t u t e d as long as the F ( a b ' ) 2 fragm e n t c o n c e n t r a t i o n was n o t d e c r e a s e d b e l o w the 2 5 0 / x g level.
Performance of SRBC coated with palmitate-derivatized F(ab')2 fragments in reverse hemolytic plaque assays I n o r d e r to test the a b i l i t y o f S R B C c o a t e d with goat anti-mouse IgM and IgG F(ab')2 fragments to i d e n t i f y i m m u n o g l o b u l i n secreting cells in a reverse h e m o l y t i c p l a q u e assay, a source of i m -
m u n e spleen cells was g e n e r a t e d b y i m m u n i z i n g mice w i t h C G G 7 d a y s b e f o r e the assay was p e r f o r m e d . W h e n these l y m p h o c y t e s were used, p l a q u e s were clearly d e t e c t a b l e as long as d i l u t e d a n t i - M G G a n t i s e r u m was p r e s e n t o n the plates d u r i n g the 3 h i n c u b a t i o n . I f this a n t i s e r u m was absent, however, n o p l a q u e s d e v e l o p e d . A l l further p l a q u e assays p e r f o r m e d with p r e p a r a t i o n s of S R B C c o a t e d with F ( a b ' ) 2 f r a g m e n t s were therefore p e r f o r m e d w i t h a n a n t i - M G G a n t i s e r u m present. T h e r e p r o d u c i b i l i t y of this reverse p l a q u e assay was tested with a single s u s p e n s i o n o f Ig-secreting cells b y using different p r e p a r a t i o n s of d e r i v a t i z e d F ( a b ' ) 2 f r a g m e n t s to l a b e l SRBC. T h e resulting n u m b e r s of p l a q u e s d e t e c t e d v a r i e d b y at m o s t 20%, as s h o w n in T a b l e II.
TABLE II REVERSE PLAQUES AND ANTIBODY PLAQUES OBTAINED FROM SPLEEN CELLS OF CGG-IMMUNIZED MICE SRBC coated with Indirect plaques/106 spleen cells anti-mouse IgM and IgG Prep. 1 Prep. 2 Prep. 3 Chicken anti-SRBC
1948 2 222 2 856 754
40 Plaque assays were also performed in which SRBC coated with chicken anti-SRBC antibodies were used to determine the number of lymphocytes secreting antibodies specific for CGG. Because the number of direct PFC was negligible, only indirect PFC were counted. When the same preparations of lymphocytes were used in the reverse plaque assay with F(ab')2 fragment-coated SRBC, approximately three times as many plaques were detected (Table II). Thus the F(ab')2 fragment-coated SRBC were sensitive enough to detect many more immunoglobulin-secreting cells than were SRBC which had been coated with C G G in order to detect only lymphocytes secreting antibodies specific for the immunogen. In order to compare the performance of F(ab')2 fragment-coated SRBC to the most widely used sensitization method for erythrocytes in the reverse plaque assay, several aliquots of SRBC were coated with protein A. Different concentrations of protein A and chromic chloride were used in the sensitization procedure in order to obtain an optimal preparation, and each batch was initially tested in a hemagglutination assay to determine if any coating had occurred. The results of these tests are shown in Table III. Many preparations did not show any evidence of coating at all, and others, such as batch G, showed too much coating, resulting in spontaneous agglutination of the SRBC. Preparations which showed behavior similar to that of batches D and H were selected for use in a
TABLE III HEMAGGLUTINATIONTEST OF SRBC COATEDWITH PROTEIN A Prep. A B C D E F G H I-
CrC13conc. (/~g/ml) 132 66 33 132 66 33 132 66 33
ProteinA conc. (mg/ml)
HA titer
0.1 0.1 0.1 0.5 0.5 0.5 2.0 2.0 2.0
1 : 1600 <1:400 <1:400 1 : 3200 <1:400 < 1:400 > 1 : 6400 1 : 3200 < 1 : 400
a The control (without immune complexes)was agglutinated as well.
TABLE IV COMPARISON OF REVERSE PLAQUE NUMBERS OBTAINED WITH PROTEIN A-COATED SRBC AND PALMITATE-DERIVATIZEDGOAT ANTI-MOUSE IgM AND IgG F(ab')2-COATED SRBC SRBC coated with protein A
Plaques/lO 6 spleen cells a
Prep. 1 Prep. 2 Prep. 3
1113 1295 521
Anti-mouse IgM and IgG
2 342
a
Spleen cells were obtained from mice immunized with CGG.
comparative reverse plaque assay with F(ab')2 fragment sensitized SRBC. The results of this study are shown in Table IV. In addition to detecting fewer antibody secreting cells, protein A-coated SRBC formed plaques that were fuzzier and more difficult to count.
Discussion Attempts to develop an improved procedure for coating SRBC with antibodies were modeled after the technique described by Peacock et al. (1986). When NHS-palmitate is reacted with antibodies under the appropriate conditions, the succinimide moiety on the palmitate is irreversibly acylated to primary amine groups, such as lysine residues, on the antibody fragments. A number of trials were performed to determine the optimal conditions under which this derivatization reaction occurs. The concentration of D O C during derivatization of the antibody fragments, subsequent removal of unreacted NHS-palmitate by Sephadex G-25 chromatography, and sensitization of SRBC is critical. The D O C concentration must be sufficient to maintain the NHS-palmitate, and later the derivatized antibody in solution, but not high enough to lyse the SRBC or interfere with their sensitization. In these trials, F(ab')2 fragments of antibodies were used rather than intact antibody molecules in order to avoid the possibility of activation of complement by the derivatized molecules coating the SRBC, which potentially would have resulted
41 in generalized lysis of the SRBC, thus negating plaque formation by the Ig-secreting cells. However, more recent trials have shown that SRBC coated with palmitate-derivatized whole antibody molecules do not lyse upon addition of complement. Therefore, either F(ab')2 fragments or intact antibody molecules can be used in the RHPA. Tests of SRBC sensitized with various preparations of derivatized goat anti-MGG indicate that the reverse plaque assay is quite reproducible. However, no plaques were detected unless a second antibody was present during the 3 h incubation. This is not unexpected, since essentially all reverse plaque assays described in the literature note this requirement. The second antibody may be necessary since many of the immunoglobulin molecules secreted by the plaque-forming cells are probably attached to the antibody-coated SRBC by their Fc regions rather than by their antigencombining sites, as is the case in the conventional plaque assay. This could make the Fc regions less accessible for complement fixation, decreasing the efficiency of lysis of the SRBC. A comparison of the number of plaques obtainable with the new reverse plaque procedure and a conventional hemolytic plaque procedure detecting antigen-specific plaques was performed using spleen cells from mice immunized with CGG, Approximately three times as many plaques were detected by the reverse plaque procedure, which may be explained if the affinities of the antibodies produced by many of the cells activated during immunization are too low to allow detection of plaque-forming cells in a conventional plaque assay. Alternatively, some of the excess plaques may result from lymphocytes secreting immunoglobulin of irrelevant specificity, as postulated by the idiotypic network theory (Jeme, 1974). Finally, a comparison was performed between F(ab')2 fragment-sensitized SRBC and protein A-sensitized SRBC, the latter being the most widely used detection system for reverse plaques. F(ab')2 fragment-sensitized cells detected almost twice as many lymphocytes as did the best preparation of protein A-sensitized SRBC. In addition, the variability in number of plaques detected was far greater among preparations of protein A-coated cells than among preparations of F(ab') 2 fragment-sensitized SRBC. A lack of con-
sistency has often been a criticism of the protein A reverse plaque method, and it appears that the F(ab')2 fragment sensitization method overcomes this drawback. The ability to coat SRBC with F(ab'): fragments of any antigenic specificity allows a potentially broad application of this methodology for the detection and enumeration of cells secreting almost any antigenic molecule. The only absolute requirement is that the molecule must have at least two exposed epitopes, one for binding to the coated SRBC and the other for attachment of the second antibody. Thus cells secreting interferons, interleukins, or viruses, for example, should be detectable by this technique. In addition, the method should be adaptable for the detection of plasma cells secreting anti-idiotypic antibodies, which was the original purpose of this study. Several of these applications are currently under development.
Acknowledgement A.J.S. is supported by a Lutheran Brotherhood fellowship.
References Gronowicz, E., Coutinho, A. and Melchers, F. (1976) A plaque assay for all cells secretingimmunoglobulinof a given type or class. Eur. J. Immunol. 6, 588. Hoffmann, L.G. and Mayer, M.M. (1977) Immune hemolysis and complement fixation. In: C.A. Williams and M.W. Chase (Eds.), Methods in Immunologyand Immunochemistry, Vol. IV. Academic Press, New York, pp. 137-166. Jerne, N.K. (1974) Toward a network theory of the immune system. Ann. Immunol. 125C, 373. Miller, G.W. and Segre, D. (1972) Determination of the relative affinity and heterogeneityof mouse anti-DNP antibodies by a plaque inhibition technique. J. Immunol. 109, 74. Mishell, B.B. and Shiigi, S.M. (1980) Preparation of mouse cell suspensions. In: B.B. Mishell and S.M. Shiigi (Eds.), Selected Methods in Cellular Immunology. Freeman, San Francisco, CA, p. 3. Molinaro, G.A. and Dray, S. (1974) Antibody coated erythrocytes as a manifold probe for antigens. Nature 248, 515. Molinaro, G.A., Eby, W.C. and Molinaro, C.A. (1981) The reverseplaque-forming cell assay. Methods Enzymol. 73, 326.
42 Nisonoff, A., Wissler, F.C., Lipman, L.N. and Woernley, D.L. (1960) Separation of univalent fragments from the bivalent rabbit antibody molecule by reduction of disulfide bonds. Arch. Biochem. Biophys. 89, 230. Nisonoff, A., Markus, G. and Wissler, F.C. (1961) Separation of univalent fragments of rabbit antibody by reduction of a single, labile disulfide bond. Nature 189, 293. Peacock, J.S., Londo, T.R., Roess, D.A. and Barisas, G. (1986) Biologic activity of antigen receptors artificially incorporated onto B lymphocytes. J. Immunol. 137, 1916. Rittenberg, M.B. and Pratt, K.L. (1969) Antitrinitrophenyl (TNP) plaque assay. Primary response of BALB/c mice to
soluble and particulate immunogen. Proc. Soc. Exp. Biol. Med. 132, 575. Segre, D. and Segre, M. (1976) Visualization of plaque forming cells in agar plates stained with o-tolidine. J. Immunol. Methods 12, 197. Segre, M. and Segre, D. (1972) Anti-DNP hemolytic plaques by mouse spleen cells in diffusion chambers. Immunol. Commun. 1, 143. Segre, M., Segre, D. and Inman, F.P. (1969) Comparison of Aal allotypic specificity carried by rabbit IgG and IgM. J. Immunol. 102, 1368.