Journal oflmmunologicalMethods, 87 (1986) 69-78
69
Elsevier JIM03741
Monoclonal anti-histamine antibody Preparation, characterization and application to enzyme immunoassay of histamine J.-L. G u e s d o n 1, D. Chevrier 1, J.-C. Mazi6
2 B.
D a v i d 3 and S. A v r a m e a s 4
I Laboratoire des Sondes Froides, " Hybridolab, ~ Unitb d'lmmuno-Allergie, and 4 Unitb d'lmmunocytochimie, lnstitut Pasteur, 75724 Paris Cedex 15, France
(Accepted 11 September 1985)
An enzyme immunoassay to measure histamine has been developed. A histamine-bovine serum albumin conjugate was prepared using 1,4-benzoquinone as the coupling agent and was employed to immunize mice for the preparation of monoclonal antibodies against histamine. After an initial screening to identify antigen-binding monoclonal antibodies the clones were isolated by limiting dilution cloning, grown in ascites and antibodies which had been secreted into the ascitic fluid were precipitated by ammonium sulphate at 50% saturation. A systematic approach for the determination of epitope specificities of monoclonal antibodies was performed. It was found that for the most specific antibody the main epitope encompassed the 2-histaminyl-l,4-benzoquinone moiety and that the K D value determined by indirect ELISA was 1.5 x 10 ~ M for the hapten part of the immunogen and 4.6 × 10 ~0 M for a histamine-Bqovalbumin conjugate. The selected monoclonal antibody could not recognize histidine or methyl-histamine. Using this antibody, we developed an enzyme immunoassay for histamine and pg amounts could be detected. The same assay was used to quantify the allergic release of histamine from guinea pig lung mast cells. Results obtained either by the present enzyme immunoassay or by a fluorometric assay were closely correlated (correlation coefficient r = 0.9702, n = 37). Key words: EL1SA," Histamine; 1,4-Benzoquinone; Monoclonal antibody
Introduction
Histamine, known to play a role in both gastric acid secretion and as the main mediator of allergic reactions, is suspected to have several other important functions (Beaven, 1978). In order to study
Abbreviations: BSA, bovine serum albumin; glycine-BqBSA, histamine-Bq-BSA, histidine-Bq-BSA, glycine, histamine or histidine bound to 1,4-benzoquinone-treated BSA; glycineBq-OVA, histamine-Bq-OVA, glycine or histamine bound to 1,4-benzoquinone-treated OVA; OVA, hen egg albumin; PBS, 0.15 M NaC1 containing 0.010 M phosphate buffer pH 7.4; TBS, 0.15 M NaC1 containing 0.010 M Tris-HCI buffer pH 7.5.
the relationship between histamine release and physiologic changes it is necessary to possess a reliable and sensitive assay. Several methods to quantify histamine in biological fluids have been reported over the years. Among them there are the fluorometric assay (Shore et al., 1959; Lorentz et al., 1972), high performance liquid chromatography coupled with fluorescence (Tsuruta et al., 1978; Skofitsche et al., 1981), the radioisotopic enzymatic assay (Snyder et al., 1966; Beaven et al., 1972), and gas chromatography mass spectrometry (Mita et al., 1980a,b). These methods are sensitive enough but do not possess the specificity a n d / o r the simplicity required for the quantitation of
0022-1759/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
70 histamine. Some of them are quite time-consuming and have the disadvantages of involving both complicated manipulations and expensive instrumentation. For the routine quantitation of histamine, an enzyme immunoassay would be more desirable because of the rapidity, simplicity and convenience needed if large numbers of samples are to be analyzed. The main requirement for the enzyme immunoassay is the preparation of an antibody specific for histamine. However, attempts to produce a specific antibody to histamine and histamine derivatives (Mira et al., 1984) with high affinity have not been successful to date. In the present study, we synthesized a histamine-protein conjugate using 1,4-benzoquinone as the coupling agent. This conjugate was injected into mice in order to prepare a monoclonal antibody against histamine. The specificity of the antibody obtained was studied and the epitope principally recognized was found to be a histaminebenzoquinone complex. This allowed us to develop a new method to quantify histamine.
purchased from CML (Nemours, France). Peroxidase-labelled antibodies specific for mouse IgG1, IgG2a, IgG2b, IgG3, IgM, lambda or kappa chains were purchased from Nordic Immunological Laboratories (Tilburg, The Netherlands).
Immunization Ten-week-old, female BALB/c, mice were immunized subcutaneously with histamine-Bq-BSA (100 #g) emulsified in Freund's complete adjuvant. On day zero complete adjuvant was used, on day 14 incomplete adjuvant was used. The mice were boosted intravenously with 50 /~g of histamine-Bq-BSA in 0.15 NaCI on day 21 and 25. Mice were bled on day 0, 14, 21, 25 and 28, for testing the amount of histamine specific antibodies.
Hybridoma fusion On day 28, 3 days after the last booster, spleen cells were prepared; and 1.2 × 108 spleen cells were fused with polyethylene glycol 1000 with 3 × 107 myeloma cells from B A L B / c mice, SP2/0 Ag8, in the logarithmic phase of growth according to K6hler and Miistein (1975).
Materials and methods
Preparation of histamine-protein conjugate Reagents The following reagents were used: histamine 2-HC1 (Serva, Heidelberg, F.R.G), L-histidine monohydrochloride, glycine, lysine, polyethylene glycol 1000, 30% H202 and Tween 20 (Merck, Darmstadt, F.R.G.). [2,5--~H]histamine dihydrochloride, 52 C i / m m o l and e[2,5-3H]histidine, 59 C i / m m o l (Amersham, U.K.), o-nitrophenyl-fl-D-galactopyranoside, ophenylenediamine, 1-methyl histamine, 1A-benzoquinone, ovalbumin (Sigma Chemical Co., St. Louis, U.S.A.), bovine serum albumin fraction V (Boehringer, Mannheim, F.R.G.), glutaraldehyde (TAAB Laboratories, Reading, U.K.), mouse IgG (Miles Laboratories, Elkhart, U.S.A.), Ultrogel AcA34 and Trisacryl G F 05 (Reactifs IBF, 92390 Villeneuve-la-Garenne, France), E. co# fl-Dgalactosidase (3.2.1.23) was a gift from Dr. A. Ullmann (Institut Pasteur, Paris). 5-methyl-cytidyl-histamine was kindly provided by Dr. Tam Huynh Dinh (Institut Pasteur); flat-bottomed polystyrene micro-ELISA plates (M29 LSE) were
1,4-benzoquinone is known to react with both thiol (Mason and Perterson, 1965) and amino groups of proteins (Mason and Lada, 1954; Byck and Dawson, 1968; Morrison et al., 1969) and has been successfully used to couple enzymes to antibody (Ternynck and Avrameas, 1977; Avrameas et al., 1978). In the present investigation advantage was taken of the fact that 1,4-benzoquinone reacts slowly at pH 4.5 with nucleophilic groups on proteins and that the mono-substituted quinone can then bind at alkaline pH to a hapten bearing an amino group. Histamine and histidine were bound to BSA or OVA using 1,4-benzoquinone as coupling agent, and employing different molar ratios and [3H]histamine or [3H]histidine as tracers. The most efficient hapten-protein conjugates were obtained as follows; 10 mg protein were dissolved in 0.1 M phosphate buffer pH 4.5 (1.7 ml) to which 0.3 ml of a solution containing 30 mg 1A-benzoquinone in 1 ml ethanol were added and the mixture allowed to react for 1 h at room temperature in the
71 dark. Protein with covalently bound benzoquinone molecules was separated from non-reacted benzoquinone by gel filtration (2.5 × 8 cm Trisacryl GF05 column in 0.15 M NaC1). The first brown fraction was collected and mixed with 0.5 ml of 0.1 M carbonate buffer pH 8.5 containing 100 mg of hapten (histamine or histidine). The mixture was allowed to stand for 20 h at room temperature in the dark and then chromatographed through a Trisacryl GF05 column as above to eliminate excess hapten. The degree of protein substitution was measured using 3H-hapten and counting the radioactivity of the preparation.
Preparation of amine-benzoquinone derivatives Amino acids or amines were bound to 1,4-benzoquinone to form addition products. These amine-benzoquinone adducts were used to test the specificity of monoclonal anti-histamine by performing ELISA-inhibition test, as described below. To prepare amine-benzoquinone derivatives, glycine, iysine, histidine, histamine or serotonin were each diluted to 2000 n m o l / m l in 0.2 M phosphate buffer p H 4.5 containing 3 mg 1,4-benzoquinone per ml. The amino acids or amines were allowed to react with benzoquinone for 1 h at room temperature in the dark. Then 2 M glycine solution and 2 M N a z C O 3 were added to the reaction to give a 0.1 M final concentration of glycine at p H 8.5. The mixture was incubated for 1 h at room temperature in the dark and used without further purification.
Antibody detection assays Culture supernatants were screened for antihistamine antibody by a solid-phase enzyme immunoassay. Polystyrene microtiter plates with 96 wells were coated either with OVA or histamineOVA at a concentration of 2 /~g/ml in TBS (0.15 M NaCI containing 0.01 M Tris HC1 buffer pH 7.5), 100/~1 per well. The plates were coated for 2 h at 37°C and then overnight at 4°C. After coating, the protein solutions were discarded and the plates washed twice with Tris-HCl-buffered saline p H 7.5 containing 0.1% Tween 20 (TBS-T). One hundred /~1 of 5-fold diluted supernatants were added to the respective wells (18 h at room temperature) followed by washing and by addition of a 1/1000 dilution of a sheep anti-mouse IgG
antibody conjugated to fl-o-galactosidase in TBS-T containing 1% BSA (TBS-T-BSA). After 2 h at room temperature the plates were washed and the galactosidase substrate solution added. The enzyme reaction was performed as described below. The positive wells were identified either visually or by measuring optical densities at 414 nm as described below.
Isotype determination Immunoglobulins were precipitated from ascitic fluids by a m m o n i u m sulphate at 50% saturation and their isotypes were determined using modifications of the antibody detection assay described above. The heavy chain subclasses were determined as follows. After addition of appropriate dilutions in TBS-T-BSA of the immunoglobulin to the histamine-OVA coated wells, 100 ~1 of a 1/1000 dilution in TBS-T-BSA of peroxidaselabelled antibodies specific for mouse IgG1, IgG2a, IgG2b, IgG3, and IgM was added (room temperature, 2 h) followed by the addition of peroxidase substrate and termination of the reaction as described below. For the determination of light chain isotypes the peroxidase-labelled antibodies specific for mouse IgG subclasses were replaced by peroxidase-labelled antibodies specific for mouse )~and K-chains.
Determination of epitope specificity In order to characterize the monoclonal antibodies, hybrids synthesizing antibody were cloned and grown in mice. The specificity of antibody present in ascitic fluids was analyzed by direct enzyme immunoassay and by an inhibition test. Direct enzyme immunoassay was performed following the method described above for the antibody detection assay. The plates were coated with BSA, OVA, histamine-Bq-BSA, histidine-Bq-BSA, glycine-Bq-BSA, histamine-Bq-OVA or glycineBq-OVA and the Ig isolated from ascitic fluid were examined on these coated plates using concentrations ranging from 1 /~g/ml to 1 n g / m l . Inhibition tests were performed by mixing isolated Ig (about 10 ng) with the various haptens at different concentrations depending on their nature. The mixtures were allowed to incubate for 18 h at room temperature, then an aliquot (100 ~1) of each mixture was added to a histamine-Bq-OVA
72 coated well and the antibody which remained free at equilibrium was determined.
Titertek multiskan MC photometer with a 414 nm filter.
Measurement of the dissociation constant of monoclonal anti-histamine antibody The determination of the dissociation constants (KD) of D22 antibody in equilibria in solution was performed using as antigens: histamine, histamine-benzoquinone-glycine, 5-methyl-deoxycytidyl-histamine and histamine-Bq-OVA, and following the ELISA procedure described by Friguet et al. (1985). Briefly monoclonal antibody was incubated with different concentrations of antigen until the equilibrium was reached, then the proportion of antibody which remained free at equilibrium was measured using a histamine-BqOVA-coated plate and galactosidase-labelled antimouse IgG antibody. The KD value was estimated by calculation of the slope of the regression line obtained using the simplification of the mathematical equation of Klotz (1953) proposed by Friguet et al. (1985).
Peroxidase determination Peroxidase activity was determined with 0.05 M citrate buffer containing 1 mg o-phenylenediamine/ml and 0.06% H202. The pH of the substrate solution was adjusted to 4.8. Each well received 0.2 ml substrate solution and plates were incubated in the dark for 10 min. The enzyme reaction was stopped by adding 50/~1 2 N H2SO 4 containing 0.5% Na2SO 3 and absorbance was read at 492 nm.
Preparation of enzyme-antibody conjugate Sheep serum anti-mouse IgG (H + L) was prepared as described elsewhere (Avrameas and Ternynck, 1969). Antibodies were isolated using affinity chromatography by passage of whole immune serum down an immunoadsorbent column prepared by coupling mouse IgG to Ultrogel beads activated with glutaraldehyde (Guesdon and Avrameas, 1976). Five mg of sheep anti-mouse IgG antibody was coupled with 10 mg of /3-galactosidase by the 1-step glutaraldehyde procedure (Avrameas et al., 1978). t~-D-Galactosidase determination /3-D-Galactosidase activity was measured by adding 0.2 ml of substrate solution per well. The substrate solution was composed of 0.1 M phosphate buffer, pH 7.0, containing 2.67 mM o - n i t r o p h e n y l g a l a c t o p y r a n o s i d e , 0.1 M /3mercaptoethanol, 1 mM magnesium Titriplex, 1 mM MgSO4, 0.2 mM MnSO 4. Plates were incubated in the dark at 37°C for 2 h, then the enzyme was inactivated by adding 0.1 ml of 1 M Na2CO 3 and absorbance was measured using a
Histamine liberation from lung guinea pig mastocytes The isolated lung of a guinea pig sensitized in vivo to ovalbumin was perfused with Krebs buffer. Before (control) and after in vitro challenge with the antigen, effluent samples were collected and analyzed for histamine. 0.1 ml aliquots were acidified by addition of 0.1 ml perchloric acid (4 N). The tubes were centrifuged and the histamine content in the supernatants was quantified by the automated fluorometric method described by Lebel (1983).
Results
A ntigen preparation Various cross-linking agents were tested for coupling histamine to proteins. 1,4-benzoquinone was found to give the best yield (data not shown). Different histamine/protein molar ratios were used to find the optimal binding of histamine on 1,4benzoquinone-treated BSA. Using [3H]histamine as tracer, it was found that a large excess of histamine was required to obtain a good binding yield (Table I). When 3694 mol histamine were added per mol of BSA, 17 molecules of histamine were bound per molecule of BSA. This conjugate was used to immunize mice. Histamine labelled ovalbumin with a molar ratio equal to 6.8 was used for coating the microtitration plate employed in the various enzyme immunoassays. Monoclonal antibody preparation Sera of B A L B / c mice immunized with hista-
73 TABLE i
TABLE II
RESULTS OF COUPLING HISTAMINE WITH BOVINE SERUM ALBUMIN (BSA) OR OVALBUMIN (OVA)
CONCENTRATION OF VARIOUS BENZOQUINONE ADDITION PRODUCTS NEEDED IN ORDER TO OBTAIN 50% INHIBITION OF THE D22 ANTIBODY BINDING ON HISTAMINE-Bq-OVA COATED PLATE
Protein carrier
BSA BSA BSA BSA OVA
Histamine/protein molar ratio during coupling
Mean number of histamine molecules bound per molecule of protein carrier
58 184 369 3694 2446
0.6 1.5 2.6 17 6.8
Substitution
Concentration (M) which reduced the antibody binding by 50%
O \ R2 0
m i n e - B q - B S A were t i t r a t e d by E L I S A using plates c o a t e d with O V A , BSA, h i s t a m i n e - B q - B S A or h i s t a m i n e - B q - O V A . All sera were f o u n d to c o n t a i n a n t i b o d i e s against h i s t a m i n e - B q - O V A , h i s t a m i n e Bq-BSA a n d BSA b u t were negative for O V A suggesting that i m m u n i z a t i o n h a d i n d u c e d antib o d i e s specific for histamine. Subsequently, one of these i m m u n i z e d mice was used for fusion 3 d a y s after the last i m m u n i z a t i o n . A f t e r the fusion 300 culture s u p e r n a t a n t s were tested b y E L I S A , 14 were f o u n d to have a n t i b o d i e s recognizing histam i n e - B q - O V A a n d a m o n g them o n l y 4 were negative for O V A . In o r d e r to o b t a i n larger a m o u n t s of a n t i b o d y these 4 h y b r i d o m a s were cloned.
Specificity of monoclonal antibodies In o r d e r to c o m p a r e the specificity of these m o n o c l o n a l antibodies, the Ig of ascitic fluids were p r e c i p i t a t e d b y a m m o n i u m s u l p h a t e a n d then tested b y E L I S A using plates c o a t e d with histam i n e - B q - O V A , h i s t a m i n e - B q - B S A , glycine-BqO V A , BSA, O V A or histidine-Bq-BSA. T w o antib o d i e s (D22 a n d I22) a m o n g the 4 o b t a i n e d d i d not b i n d to g l y c i n e - B q - O V A . Both specific antib o d i e s b e l o n g e d to the IgG1 subclass a n d possessed k a p p a light chain. It is n o t e w o r t h y that these m o n o c l o n a l a n t i - h i s t a m i n e a n t i b o d i e s d i d not b i n d to the plate c o a t e d with h i s t i d i n e - B q - p r o t e i n conjugate. W i t h the aim of analyzing which p a r t of the antigen was recognized b y the 2 antibodies, inhibition studies were p e r f o r m e d using various conc e n t r a t i o n s of inhibitors. The results in T a b l e II a n d T a b l e III d e m o n s t r a t e that the a n t i b o d i e s
RI
R2
Glycine Lysine Serotonin Histidine 1-methyl-histamine Histamine Histamine Histamine Histamine Histamine Histamine Histamine Ovalbumin
Glycine Glycine Glycine Glycine Glycine Glycine Glycyl glycine Glycyl glycyl glycine Tetraglycine Pentaglycine Hexaglycine N"-acetyl-L-lysine Histamine
> 1 x 10 3 > 1 x 10 - 3 > 1 x 10 3 > 1 x 10 - 3 > 1 x 10-3 2 X 10- 8 2 x 10 s 2 x 10 8 2 x 10 ~ 2 x 10- s 2 x 10 - 8 1 X 10 - s 7x 10 lo
p r o d u c e d h a d a high specificity for h i s t a m i n e linked to 1,4-benzoquinone. The a n t i b o d i e s were also shown to recognize free h i s t a m i n e and 5-methyld e o x y c y t i d y l h i s t a m i n e with a lower affinity b u t d i d not b i n d to histidine, and 1-methyl histamine b o u n d to 1,4-benzoquinone. Thus it can be d e d u c e d that the e p i t o p e recognized b y b o t h m o n o c l o n a l a n t i b o d i e s c o m p r i s e s 1,4-benzoq u i n o n e substituted in p o s i t i o n 2 b y histamine. T h e s u b s t i t u t i o n b y various amines or a m i n o acids in position 5 did n o t affect significantly the recognition of the h i s t a m i n e - b e n z o q u i n o n e a d d u c t b y these a n t i b o d i e s (Table II). M e t h y l a t i o n on N1 of the histamine molecule totally a b r o g a t e d a n t i b o d y binding.
Determination of the dissociation constant of D22 monoclonal antibody D i s s o c i a t i o n c o n s t a n t s ( K o ) of D22 a n t i b o d y
74 TABLE Ill
TABLE IV
CONCENTRATION OF MISCELLANEOUS COMPOUNDS N E E D E D IN O R D E R TO OBTAIN 50% INHIBITION OF THE D22 ANTIBODY B I N D I N G ON HISTAMINE-Bq-OVA-COATED PLATE
DISSOCIATION CONSTANTS OF MONOCLONAL D22 A N T I - H I S T A M I N E ANTIBODY
Compound (molecular weight)
Antigen concentration (M) giving 50% antibody inhibition
Bovine serum albumin (68000) Ovalbumin (45 000) Histidine (155) Histamine (111) 5-Methyl-deoxycytidyl-histamine (335)
>2x10 >2×10 >1×10 5x10 2.5x 10
5 5 2 3 4
NHCH2CH2~/f--N~ CH3~ ~
OH
in various antibody-antigen complexes were calculated for 4 different antigens: histamine, 5methyl deoxycytidyl-histamine, glycyl-benzoquinone-histamine and histamine-Bq-OVA. We used the method described by Friguet et al. (1985) which allows the measurement of the true dissociation constant in solution of unpurified antibodies. Using a Klotz representation (Fig. 1) the K D values were deduced from the slopes calculated from the linear regressions (Table IV). The affinity of the D22 antibody was 33-fold lower for glycylbenzoquinone-histamine and 8 x 106-fold lower for free histamine than for histamine-Bq-OVA.
Optimization of the enzyme immunoassay After determination of the specificity of the D22 monoclonal antibody for histamine we proceeded to test several different methods for the quantification of histamine. Best results were obtained with the procedure outlined in Fig. 2. We first determined the optimum conditions for BSA coating; investigating concentration, temperature and buffer composition. The best results were obtained by incubating BSA at 1 0 / ~ g / m l in
K D determined by the method of Friguet et al. (1985). Antigen
KI) (M)
Histamine 5-methyl-deoxycytidyl-histamine Glycyl-benzoquinone-histamine Histamine-Bq-OVA
3.7× 1.9 × 1.5 x 4.6×
10 10 10 10
3 4 s 1(~
PBS p H 7.4 for 2 h at 60°C and then for 18 h at 4°C. BSA was treated with 1,4-benzoquinone using various conditions of pH (0.1 M phosphate buffer p H 4.5, 6 or 8.3) concentrations (from 1 to 10 m g / m l ) , incubation times (from 0.25 to 6 h) and temperatures. It was found that the optimal concentration of benzoquinone differed with the p H used; the best results were obtained at pH 4.5 with 3 mg benzoquinone/ml (Fig. 3). The reaction was complete in 1 h. The influence of temperature on the sensitivity of the assay showed clearly that highest sensitivity was obtained by incubating benzoquinone at 4°C (Fig. 4). We then determined the conditions required for optimal binding of histamine to 1,4-benzoquinone activated BSA. Fig. 4 shows that incubation of histamine must be performed at 37°C and for at least 4 h. A preliminary experiment showed that no histamine binding occurred at pH 7.2. Therefore borate buffer pH 7.8, 8.5, 9.0, 9.2 and carbonate buffer p H 8.5, 9.4, 10.0, 10.8 were tested. The results obtained indicated that the best sensitivity was achieved when borate buffer pH 9.0 was used. Taking into account the above results, the following optimized enzyme immunoassay was finally selected and used. To each well of a microtitration plate, 0.1 ml PBS containing 1 /tg BSA was added. The plate was covered with an adhesive paper and incubated for 2 h at 60°C and overnight at 4°C. After this step the coated plate can be kept for several weeks at 4°C. Just before use the coating solution was discarded and the plate was washed twice with 0.1 M K H 2 P O 4 solution p H 4.5. Then each well was filled with 0.1 ml benzoquinone solution made by
75
A
B
10_
_
4 3_
1
1
|
o
i
~,
a
~i
o
xlO-'
i D
C
10_
xlO-"
1.4_ 1.3_ 1.2.
5_
1.1. 1.0
1
~
0
xlO -~'
1
1'o
o
xlO-'
Fig. 1. Klotz plots of the binding of histamine (A), 5-methyl-deoxycytidyl-histamine (B), histamine-Bq-glycine derivative (C) and histamine-Bq-OVA conjugate (D) to D22 lgG measured by ELISA. Abscissa: 1/total antigen concentration (M). Ordinate: 1/fraction of bound antibody.
m i x i n g 30 m g 1 , 4 - b e n z o q u i n o n e in 1 ml e t h a n o l w i t h 9 ml o f 0.1 M K H 2 P O 4 s o l u t i o n p H 4.5. T h e p l a t e was i n c u b a t e d for 1 h at 4 ° C , w a s h e d 3 t i m e s w i t h 0.1 M K H 2 P O 4 s o l u t i o n a n d the l i q u i d was r e m o v e d as c o m p l e t e l y as possible. T h e h i s t a m i n e
1.0support
1, conjugate
7
enzyme reaction
0.5. +',
antibod
6
~ primary amine
Fig. 2. Principle of histamine enzyme-immunoassay. Step 1: coating plate with BSA. Step 2: treatment of BSA coated plate by 1,4-benzoquinone. Step 3: histamine binding. Step 4: blocking of excess of benzoquinone by an amine (e.g., Tris buffer). Step 5: incubation with anti-histamine monoclonal antibody. Step 6: incubation with galactosidase/anti-mouse IgG conjugate. Step 7: enzyme reaction.
Fig. 3. Determination of optimal benzoquinone concentration for treating BSA at pH 4.5. Enzyme immunoassay was performed using 3 different concentrations of histamine: 10 ng/ml (solid line), 2.5 ng/ml (dotted line), 1 ng/ml (broken line). Optical density is plotted against benzoquinone concentration.
76
A
B
C
D
1.0
0.~
~,
22
37
i D
60 4
22
37
i I
60 4
2'2
37
I
6'() 4
22
37
~0
Fig. 4. Determination of optimal reaction temperatures. BSA-coated polystyrene plates were treated at pH 4.5 with 1,4-benzoquinone (3 mg/ml) at 4°C (A), 22°C (B), 37°C (C) or 60°C (D). Histamine binding to the benzoquinone-treated BSA was performed at temperatures indicated in abscissa. Enzyme immunoassay was performed using 3 different concentrations of histamine: 5 ng/ml (solid line), 2.5 n g / m l (dotted line) or 0.5 n g / m l (broken line). Optical density is given in ordinate.
sample was diluted in borate buffer pH 9.0, and 0.1 ml of this histamine solution was added to the well. The plate was incubated for 4 h at 37°C and washed with TBS-T. Each well was filled with monoclonal anti-histamine antibody diluted in TBS-T-BSA (0.1 ml), incubated overnight at room temperature and washed with TBS-T. 0.1 ml galactosidase-labelled anti-mouse IgG antibody diluted in TBS-T-BSA at 1 /~g/ml was then added. The plate was left for 2 h at room temperature, washed as above then placed upside down on a clean absorbent paper for 1 min. 0.2 ml substrate was added to determine galactosidase activity. Using this procedure and various concentrations of histamine it was possible to obtain a dose-response curve similar to that presented in Fig. 5. The detection limit, i.e., the smallest single result which is clearly detectable and different from the background noise, is defined as 3 standard deviations of the appropriate blank value. With this criterion, the present enzyme immunoas-
1.Q
0.5_
0.01
O~
"1
1~0
160 ng/ml
Fig. 5. Dose-response curve for histamine determined by the present enzyme immunoassay. Ordinate: absorbance at 414 nm. Abscissa: log histamine concentration (ng/ml).
77
ng/ml
10(
5C
10
1'0
5b
160
ng~ml
Fig. 6. Correlation between histamine concentration determined by the fluorometric method (ordinate) and by the enzyme immunoassay (abscissa). The histamine was released from guinea pig lung mast cells.
say can detect 45 pg in the sample; this value corresponded to an optical density of 0.0464_+ 0.0036 (8 points) and under the same conditions, the mean blank value was 0.0125 +0.0023 (8 points). The standard curve as illustrated in Fig. 5 is usable from 45 pg histamine to 2 ng per sample. The enzyme immunoassay and the fluorometric methods were compared for the determination of the histamine released from guinea pig lung mast cells. The results of the comparison are shown in Fig. 6. The histamine concentrations obtained with either the fluorometric method or by enzyme immunoassay were closely correlated (correlation coefficient r = 0.9702, n = 37), the equation of the regression straight line being y = 1.18x - 0.66.
Discussion Enzyme immunoassay is widely used in both the clinical and research fields to quantitate many biologically important small molecules. One of the critical aspects of the assay is the availability of specific antibody. To attempt to produce rabbit antibodies to histamine Mita et al. (1984) prepared 3 conjugates of different histamine derivatives to
bovine serum albumin: succinylhistamine-BSA, N~-trifluoroacetylhistamine-BSA and N%propionylhistamine-BSA. However, specific antibodies against histamine could not be produced by immunization with these 3 immunogens. Haydik (1983) reported the use of rabbit antibodies against histamine for histamine determination in a radioimmunoassay based on the competitive binding double antibody principle. However, she did not report the chemical structure of the immunogen used a]ad the,specificity of the antiserum was not determined. To obtain an efficient immunogen we tested several methods and various coupling agents to covalently bind histamine to BSA; 1,4-benzoquinone was found to give the highest yield of coupled histamine. In consequence we conjugated histamine to BSA via the N '~ primary amino group of the histamine molecule using the 2-step benzoquinone method. BSA bearing 17 molecules of histamine per molecule was injected into mice. The antisera showed reactivity against BSA, histamineBq-BSA, histamine-Bq-OVA but not against OVA indicating that the antisera could recognize the hapten moiety of the immunogen. On the basis of these preliminary experimental results it was considered possible to produce an antibody against histamine and attempts were made to produce the corresponding monoclonal antibodies in order to obtain large reproducible amounts of antibody. The study of the specificity of the most histamine-specific monoclonal antibody (D22) reveals that the combining site recognizes histaminyl-benzoquinone with a 2.5 x 10S-fold higher affinity than that observed with free histamine. Histamine was allowed to react on position 2 of 1,4-benzoquinone and then position 5 was blocked by different amines or amino acids. By blocking position 5 with different compounds the affinity of anti-histamine antibody was not affected, suggesting that the main epitope was composed of only 1 molecule of histamine linked to 1 molecule of benzoquinone. The D22 antibody showed a 20-fold higher affinity for 5 methyl-deoxycytidyl-histamine than for free histamine. This indicated that the antibody recognized the cytidyl ring as a structural analogue of benzoquinone. Moreover it was found that the affinity of the D22 antibody was 30-fold higher for histamine linked to OVA
78
than for histaminyl benzoquinone although the histamine was bound to OVA through benzoquinone. It could be argued that this difference was due to the amount of histamine linked to OVA, which was precisely determined using [3H]histamine as a tracer. Since no purification has been attempted after coupling, the precise amount of histaminyl benzoquinone in the benzoquinone histamine mixture was not known. An attempt was made to increase the affinity of the D22 antibody for binding histam,ine to benzoquinone but no significant effect on the K~ value was observed. The difference observed between the 2 Kt~ values may be explained by the multivalence of histamine-Bq-OVA and the stabilization effect of the protein carrier on the histamine-BqO V A / a n t i b o d y complexes. It is noteworthy that the monoclonal anti-histamine D22 antibody finally selected did not bind significantly either to histidine or to 1-methyl-histamine. Moreover it was found that the D22 monoclonal antibody possessed a strong affinity for the histamine-protein conjugate ( K n = 4.6 × 10 t~ M). These results prompted us to consider that the reaction between histamine and benzoquinone-treated protein could be applied in the determination of histamine. Using the optimum conditions defined and described in the results section, the enzyme immunoassay developed was able to detect specifically 4 × 10 ~3 mol histamine (45 pg) in a 0.1 ml sample. In order to increase sensitivity we used the fluorogenic substrate of galactosidase, methylumbeliferylgalactoside instead of its chromogenic one, o-nitrophenylgalactoside. In contrast to the results previously obtained with an immunoassay for the measurement of cyclic AMP (Joseph and Guesdon, 1982) we were not able to increase the sensitivity of the histamine immunoassay possibly because the affinity of the antibody for histamine is a limiting factor. A good correlation was observed when the commonly used fluorometric technique and the present enzyme immunoassay were used for the quantitation of the histamine released from guinea pig lung mast cells. In conclusion, it can be said that the method developed is flexible and sensitive enough to assay histamine in biological fluids, and it allows the
routine analysis of several hundred samples in 1 day. Work is now in progress to apply this method to the measurement of histamine in human plasma and other body fluids.
Acknowledgements We appreciate the gift of 5-methyl-cytidylhistamine by T a m Huynh Dinh. We thank J. Lefort for providing us with the lung perfusates and F. Gauthier for preparing the manuscript.
References Avrameas, S. and T. Ternynck, 1969, Immunochemistry 6. 53. Avrameas, S., T. Ternynck and J.L. Guesdon, 1978, Scand. J. lmmunol. 8 (suppl. 7), 7. Beaven, M.A., 1978, Monogr. Allergy 13, 1. Beaven, M.A., S. Jacobsen and Z. Horakova, 1972, Clin. Chim. Acta 37, 91. Byck, J.S. and C.R. Dawson. 1968, Anal. Biochem. 25, 123. Friguet, B., A.F. Chaffotte, L. Djavadi-Ohaniance and M. Goldberg, 1985, J. immunol. Methods 77, 305. Guesdon, J.L. and S. Avrameas. 1976, J. Immunol. Methods 11, 129. Haydik, 1.B., 1983, J. Allergy Clin. Immunol. 71, 152. Joseph, E. and J . L Guesdon, 1982, Anal. Biochem. 119, 335. Klotz, I.M., 1953, in: The Proteins, Vol. 1, eds. H. Neurath and K. Bailey (Academic Press, New York) p. 727. KOhler, G. and C. Milstein, 1975, Nature (London) 256, 495. Lebel, B., 1983, Anal. Biochem. 3, 16. Lorentz, W., H.J. Reimann, H. Barth, J. Kusche, R. Meyer, A. Doenicke and M. Hutzel, 1972, Hoppe-Seyler's Z. Physiol. Chem. 353, 911. Mason, H.S. and A. Lada, 1954, J. Invest. Dermatol. 22, 457. Mason, H.S. and E.W. Peterson, 1965, Biochim. Biophys. Acta 111, 134. Mita, H., H. Yasueda and T. Shida, 1980a, J. Chromat. 181, 153. Mita, H., H. Yasueda and T. Shida, 1980b, J. Chromatogr. 221, 1. Mita, H., H. Yasueda and T. Shida, 1984, Agents Actions 14, 574. Morrison, M., W. Steele and D.J. Danner, 1969, Arch. Biochem. Biophys. 134, 515. Shore, P.A., A. Burkhalter and V.H. Cohn, 1959, J. Pharmacol. Exp. Ther. 127, 182. Skofitsche, G., A. Saria, P. Holzer and F. Lembeck, 1981, J. Chromatogr. 226, 53. Snyder, S.H., R.J. Baldessarini and J. Axelrod, 1966, J. Pharmacol. Exp. Ther. 153, 544. Ternynck, T. and S. Avrameas, 1977, Immunochemistry 14, 767. Tsuruta, Y., K. Kohashi and Y. Ohkura, 1978, J. Chromatogr. 146, 490.